The URL can be used to link to this page

Home My WebLink AboutEMTP Studies for the Railbelt Transmission System Final Report 1995INSTITUTE OF NORTHERN cr EMTP STUDIES FOR THE RAILBEL isecatenis TRANSMISSION SYSTEM by James W. Cote, Jr. John D. Aspnes Douglas R. Ritter December 1995 FINAL REPORT ALASKA FAIRBANKS FAIRBANKS, ALASKA ! { | | UNIVERSITY OF | 99775-5906 Report No. INE/TRC 96.05 907-474-7137 Electrical Engineering Department FAX 907-474-6087 Unwersity of ALaska Fairsanks School of Engineering * PO Box 755900 « Fairbanks, Alaska 99775-5900 Ae October 18, 1996 Mr. Thomas A. Lovas Chairman, Reliability Criteria Committee Alaska Systems Coordinating Council 5601 Minnesota Drive P.O. Box 196300 Anchorage, AK 99519-6300 Dear Mr. Lovas: As we have discussed, please find enclosed a final report to the Alaska Systems Coordinating Council (ASCC) titled EMTP Studies for the Railbelt Transmission System and dated December 31, 1995. The authors are Professors James W. Cote, Jr. and John Aspnes, and Mr. Douglas R. Ritter. In addition to the final report, a thesis by Mr. Ritter titled Electromagnetic Transients Study of the Alaskan Railbelt Power System and dated August 1995 is also attached. A 3.5” floppy disk containing railbelt Alternative Transients Program (ATP) data files accompanies the thesis. This thesis project was supported by the project described in this final report. The Alternative Transients Program was the version of the Electromagnetic Transients Program (EMTP) used in this work. The summary final report was provided to members of the Reliability Criteria Committee last winter, but Mr. Ritter’s thesis has not previously been distributed. Twenty-five copies of these documents are being provided to Ms. Penny Haldane, ASCC coordinator, to be sent to your distribution list. Professor Cote, Mr. Ritter, and | are grateful to the ASCC for its support and for the opportunity to be of service to the electric utilities of Alaska. Sincerely, Get. Capree Dr. John Aspnes, P.E. Professor of Electrical Engineering University of Alaska Fairbanks EMTP STUDIES FOR THE RAILBELT TRANSMISSION SYSTEM By James W. Cote, Jr. John D. Aspnes Douglas R. Ritter Department of Electrical Engineering University of Alaska Fairbanks Fairbanks, Alaska 99775-5900 A research report sponsored by the Alaska Systems Coordinating Council December, 1995 Report No. INE96.05 Institute of Northern Engineering University of Alaska Fairbanks Fairbanks, Alaska 99775-5910 1. Introduction The Alaska Systems Coordinating Council (ASCC) contracted with the University of Alaska Fairbanks (UAF) to perform Electromagnetic Transients Program (EMTP) studies for the Railbelt transmission system. Work began in May 1994 and was completed in Fall, 1995, under UAF proposal number INE94.25. This report summarizes the results of these studies. This work was primarily performed by a UAF graduate student, Mr. Douglas Ritter, who detailed his work in a Master of Science thesis entitled "Electromagnetic Transients Study of the Alaskan Railbelt Power System". We are submitting with this report Mr. Ritter's thesis as documentation of the work performed under this contract. This report will outline the scope of work performed and highlight the results. UAF was asked to perform the following work : A) Create a Railbelt EMTP database for electromagnetic transient studies B) Evaluate the Out-of-Step relays located along the Anchorage - Fairbanks Intertie C) Evaluate Single Pole Reclosing (SPR) options along the Anchorage - Fairbanks Intertie D) Determine acceptable amounts of, and the impact of, total harmonic distortion (THD) on customer loads Items A, B, and C above were completed. The evaluation of the Out-of-Step (OOS) relays along the Intertie was expanded as our studies evolved, leaving no time for item D above (THD studies). An overview of these results follows. 2. Railbelt EMTP Database A generalized EMTP database was assembled for the Railbelt transmission system. EMTP models include a large amount of detail for all transmission system components. The impact of electromagnetic phenomena associated with any event on the power system is EMTP Studies - Final Report Page 1 known to be limited to local equipment, and is not expected to propagate long distances. Lightning strikes and switching surges propagate quickly, yet rapidly become attenuated as the surge propagates away from the inception point. The localized impact and the large amount of data required by the EMTP program to accurately model these phenomena allow the engineer to build localized EMTP models and discourage the building of large, complex or complete system models. As such, six separate EMTP databases were built, one each for the six control areas modeled in the PSS/E power flow and transient stability database. These six models roughly break down as follows : 1. Anchorage Municipal Light and Power (AMLP) Chugach Electric Association (CEA) Fairbanks Municipal Utilities System (FMUS) Golden Valley Electric Association (GVEA) Homer Electric Association (HEA) uc a Matanuska Electric Association (MEA) These six EMTP databases were built from the existing PSS/E power flow database and system one-line diagrams as provided by the Railbelt utilities. The steady state response of the EMTP model was checked against the PSS/E power flow solution for accuracy. We know of no existing electromagnetic response with which we can further verify the accuracy of the database. The EMTP provides the engineer with several different models for standard electrical equipment, each with different levels of detail. For example, transmission lines can be modeled with extensive detail which incorporates traveling wave phenomena and phase sequence unbalance (non-transposed conductors). This type of model is required when investigating lightning strikes or other rapidly changing transient conditions. At the opposite end of the detail spectrum, a line model which uses a simple positive sequence lumped RLC parameter, PI-equivalent model is available. We used a lumped resistance, distributed parameter line model which can support traveling wave phenomena for EMTP Studies - Final Report Page 2 modeling all transmission lines. This model assumes equal phase transposition along the length of the line. Transformers are modeled as three single phase transformer banks, with saturation and magnetizing current ignored. Generators are modeled as ideal voltage sources behind the subtransient reactance. All the models chosen are the best detailed models available without employing extensive modeling effort for all equipment. The models are as accurate as possible given the available power flow data and system one-line diagrams. The various EMTP tutorials, rulebooks, and workbooks suggest using these types of models whenever additional data is unavailable, and suggests using additional detail only near the location where an electromagnetic event will be introduced. Extensive modeling detail is believed to provide no additional accuracy in the studies while greatly increasing the amount of computational time required. EMTP and ATP versions of the Program The original EMTP program was developed at Bonneville Power Administration (BPA) back in the 1970's. The original BPA version was freely distributed to utilities and academia without charge. The Electric Power Research Institute (EPRI) undertook a project to revamp the EMTP program in the mid to late 1980's. This version is now available to EPRI member utilities and universities at a small distribution fee. However, the fee to non-EPRI members is large. This EPRI version continues to be known as "EMTP". BPA resisted the so-called commercialization of the EMTP. As a result, several BPA past and present employees continue to support and develop the EMTP code, completely separate from EPRI. This program has become known as the "Alternative Transients Program" (ATP). At present, both programs, EMTP and ATP, are completely data file compatible. The EPRI version of the software now incorporates a "user friendly" Windows interface for processing the input data and the simulation results. However, the construction of the input database and simulation instructions remain antiquated. This is also true of ATP. The ATP group has developed a Windows-based graphical input program called ATPDRAW. This program is still evolving but offers promise as a convenient tool for constructing a database. The ATP group licenses use of ATP for a $10 fee, and supports an active Internet mailing list. This mailing list exchanges E-mail EMTP Studies - Final Report Page 3 between ATP users, providing an excellent forum for questions and answers to all sorts of ATP modeling and simulation issues. This excellent support and access to the world-wide group of ATP experts makes the ATP a better choice than EMTP (EPRI version) for the engineer who needs to perform electromagnetic transients studies. This report will continue to use the term EMTP to refer to either of these versions of EMTP, given that the database format and models are the same. Documentation Several EMTP rulebooks exist which document all the EMTP models and simulation instructions and features. The EPRI versions of the rulebooks and other documentation is useful for ATP and has been rewritten in a cleaner format. An updated ATP rulebook is under construction. Several workbooks exist from EPRI and from various short courses. These workbooks provide tutorials on common EMTP simulations and analyses, and should be used by any engineer getting started in performing EMTP work. A complete tutorial on electromagnetic phenomena usually requires at least a one semester graduate course in electric power engineering, or experience working under an industry expert in transient phenomena. The standard textbook on this subject is "Electrical Transients in Power Systems", Second Edition, by Allan Greenwood, published by Wiley- Interscience in 1991, ISBN 0-471-62058-0. We refer anyone who is beginning to perform EMTP analyses to this book and to the various EMTP rulebooks and workbooks. There are also frequent offerings of short courses for the EMTP and ATP, usually around the same time as the summer power meeting of the IEEE. The University of Minnesota, and the ATP developers, offer these courses approximately once a year. Using the Railbelt Databases The Railbelt databases (six in all) are included on diskette with this report. These databases will also be available via the Internet if so desired. Each database will need to be tailored to each particular study. Regrettably, this is the nature of electromagnetic studies. The engineer will need to choose more detailed models of some equipment for EMTP Studies - Final Report Page 4 each study. For example, when investigating a lightning strike, the affected transmission line should be modeled accurately to include conductor spacing and actual conductor transposition. The ground characteristics (resistivity, etc.) should also be included. Tower parameters and insulators may also need to be modeled. Clearly, modeling each and every insulator for every conceivable study is impractical. The database constructed by UAF should be sufficiently accurate one or two lines away from the disturbance point, greatly reducing the time required for constructing the database and running the simulation. 3. Out-of-Step Relays along the Intertie The ASCC requested an analysis of the Out-of-Step (OOS) relays which exist along the Anchorage-Fairbanks Intertie. Of primary concern were the OOS relays at Healy which look both North and South out of Healy along the Intertie. These relays respond to out- of-step conditions which are detected based on apparent impedance, (i.e., voltage divided by current), and the rate with which the impedance moves through impedance lenses. Out-of-step conditions occur as a result of electromechanical phenomena, not electromagnetic phenomena. As such, the EMTP was inappropriate for these studies. The PSS/E transient stability program was more appropriate for this work and was used. While investigating the OOS relays, we were made aware that GVEA had some concerns about two disturbances which appeared to operate the OOS relays. The operation of the OOS relays was thought to be incorrect. GVEA suggested that we model these disturbances and verify our PSS/E model against actual dynamic system monitor (DSM) data. We could then check the operation of the OOS relays, assuming our model accurately tracked the actual disturbance. First Disturbance - North Pole Unit Trip On January 10, 1994, North Pole Unit 1 tripped due to a flame out condition (turbine trip), while generating 50 MW. The generator underwent a soft trip, with the generator EMTP Studies - Final Report Page 5 breaker opening approximately 0.23 seconds after the turbine tripped. Eventually, GVEA separated from the rest of the Railbelt at Gold Hill Substation, along the Intertie south. The GVEA system went black shortly thereafter. The PSS/E simulation of this event was difficult because of the severity of this disturbance in the GVEA service area. The PSS/E simulation did however closely track the first 0.75 seconds or so of the disturbance, as compared to the available DSM data from Gold Hill. Of particular interest were the frequency, voltage, and Intertie flow values at Gold Hill. These values greatly impact the PSS/E simulation's calculation of impedance and OSS conditions. Second Disturbance - Musk OX 69 kV Line to Line Fault On November 20, 1994, a prolonged line-to-line fault occurred on the 69 kV tap to Musk Ox Substation. The Intertie at Healy was eventually opened north and south due to operation of the OOS relays at Healy. This fault scenario was difficult to model using PSS/E because the PSS/E model is a balanced, single phase representation of the positive sequence only. However, an approximation to the correct positive sequence response can be made. The PSS/E simulation again accurately modeled the actual response seen in the DSM data. However, the simulation did not seem to show a true OOS condition occurring at Healy. Out-of-Step Relay Settings A detailed investigation of the OOS relays at Healy was done to understand their operation and the system conditions necessary to cause operation. The relays include a supervisory overcurrent relay which must pick up to begin the OOS operation. In addition, several impedance lenses, blinders, and timers are used track the impedance and its rate of change during a disturbance. Our analysis indicated that the supervisory overcurrent setting appeared to be too low. Normal operating current of 240 Amperes, equal to about 57 MW on the Intertie, picks up the overcurrent element. This setting EMTP Studies - Final Report Page 6 appeared to be much too low. As a result, the OOS relay appears to be armed much of the time under normal operating conditions. We understand that the OOS settings have been changed. Additional Studies to Verify OOS Settings In response to questions raised by GVEA, UAF performed additional PSS/E studies to investigate possible OOS conditions at Healy, and alternative control action. Specifically, we attempted to open the Intertie only to the south of Healy for OOS conditions, and also looked at blocking all OOS action. A series of simulations were performed to evaluate these options. The studies are based on the 1994 winter base case produced by Power Technologies, Inc. (PTI). The details of these studies appear in Mr. Ritter's thesis, and are summarized below. Case 3A : soft trip of North Pole Unit 1, Intertie opened north and south at Healy due to OOS relay action, severe load shedding and collapse occurs in Fairbanks Case 3B : soft trip of North Pole Unit 1, Intertie south at Healy opened when North Pole unit breaker opens, under-frequency load shedding (UFLS) occurs in Fairbanks (82 MW), severe under-frequency occurs at Healy and Chena generators Case 3C : soft trip of North Pole Unit 1, Intertie remains closed at Healy, 18.1 MW of load in Fairbanks is directly tripped 10 cycles after the North Pole unit is tripped, no UFLS occurs, system response is stable Case 3D : soft trip of North Pole Unit 1, Intertie remains closed at Healy, 6.2 MW of load in Fairbanks is directly tripped 10 cycles after the North Pole unit is tripped, no UFLS occurs, system response is stable, voltage sags are somewhat low (0.70 pu) EMTP Studies - Final Report Page 7 Case 3E : same as case C, except that a fault is introduced after 5.0 seconds to check overall system stability after the unit trip at North Pole; response is stable These simulations were all conducted from one system base case, without investigation of alternate dispatches and operating points. We understand that some modifications have subsequently been made to the OOS relays at Healy, and that these studies helped to verify the expected system response at Healy to loss of generation in Fairbanks. 4. Single Pole Reclosing (SPR) on the Intertie The ASCC requested that UAF investigate SPR options along the Anchorage-Fairbanks Intertie. SPR is an automated protective scheme for opening of one phase of a transmission line when a single line to ground (SLG) fault is detected. Approximately 80- 90% of all faults appear to be temporary SLG faults, which could be self clearing if the faulted conductor is de-energized momentarily. The conductor must typically be de- energized for 0.5 to 1.0 seconds for the fault to clear. The faulted conductor can then be automatically re-energized and should remain in service. If the fault persists, all three phases of the line are typically opened and not reclosed. Along weak transmission corridors (large line impedances and / or very few lines between utilities), system stability problems do not allow three phases of a transmission line to be opened for much more than a few cycles, without causing a system collapse or large amounts of under-frequency load shedding. The primary feature of SPR is the ability to open only one phase for many typical faults, while continuing to carry power along the remaining two phases. If the fault self clears, the third phase is re-energized and system stability may be maintained. Whereas three phase opening may cause a system instability, SPR may keep the systems in synchronism. Three basic problems exist with SPR. First, the arc associated with the fault must be extinguished. Second, impact torques at nearby generators must not be large enough to EMTP Studies - Final Report Page 8 cause turbine blade or shaft damage. Third, the unbalance created by the open phase causes harmonics and negative sequence currents and voltages, which must not be large enough to cause damage to equipment. Secondary Arcs When a SLG fault occurs along a transmission line, and the faulted phase is opened, the faulted phase voltage does not go to zero. Capacitive coupling between the phases can be large enough to maintain a voltage on the open phase which sustains some arc current. This remaining arc current is called a "secondary arc". The magnitude of the voltage induced, and the magnitude of the secondary arc, are dependent on the location of the fault, the geometry of the transmission line, and the characteristics of the object causing the fault. Practical experience seems to indicate that secondary arc currents of less than 20 A tend to self extinguish, while larger arc currents tend to maintain the fault. Impact Torques Impact torques are created when switching surges and fault currents cause a sudden change in electrical torque on generator / turbine systems. These impact torques exist whether single pole or three pole opening and closing of breakers is implemented. SPR can cause additional problems because the reclose operation can expose the power system to a fault a second time, a short time after the first exposure. Harmonics and Negative Sequence The unbalanced operation which occurs when one phase is open creates negative sequence voltages and currents. Generators exposed to these negative sequence currents undergo additional heating and see a double frequency (120 Hz) component in the stator currents. The unbalanced conditions also generate increased harmonics. These harmonics can also increase generator heating. Standard SPR practice usually recloses the open phase or trips open all three phases within a short time after the first phase is opened, in order to minimize the negative sequence and harmonic effects. EMTP Studies - Final Report Page 9 SPR Results Mr. Ritter's thesis details several SPR options along the Intertie. Many different SLG fault locations along the Intertie were studied, with SPR implemented. Secondary arcs, impact torques, and harmonics were quantified. Impact torques, harmonics, and negative sequence currents did not appear to be a problem. This concurred with utility experience as documented in the power system literature. The secondary arc levels did however create some concern. It is common industry practice to employ some sort of remedial measures to reduce the secondary arc currents below 20 A. Long lines such as the Intertie commonly experience large secondary arc currents because the capacitive coupling effects accumulate over the long line distances. Common secondary arc neutralization methods include the use of wye connected 3-phase reactor banks to compensate the line capacitance, and the use of high speed grounding switches to ground each end of the line immediately after the breakers open. The reactor bank scheme has the drawback that line capacitance in the steady state is affected, reducing the transfer capability of the line. High speed grounding switches ground the line at each end when opened, reducing the cumulative effect of capacitive coupling. This does not however ground the whole line or force the secondary arc current to zero. UAF found that neither of these approaches solved all secondary arc problems. UAF did find the use of line sectionalizers helpful in decreasing the secondary arc currents. By sectionalizing the Intertie, only along the open phase, and only between the two breakers necessary to clear the faulted phase, the capacitive coupling of the faulted phase was reduced. This reduces the secondary arc currents to acceptable levels. Mr. Ritter's thesis includes the results of PSS/E studies used to determine the maximum allowable open time of the single phase along the Intertie, such that the Railbelt remains stable for the fault scenario. Also included are numerous simulations used to determine the harmonic and negative sequence exposure to the Healy generators caused by SPR. EMTP Studies - Final Report Page 10 5. Conclusions UAF has developed an EMTP database which should be used as a starting point for transmission system EMTP studies. Additional modeling effort is always required at those locations where transient events are initiated. UAF studied the Out-of-Step relays at Healy in conjunction with two actual disturbances which occurred in 1994. The PSS/E simulations reasonably tracked the DSM data, yielding confidence in the simulation results and the PSS/E calculations of impedance (OOS relay inputs) at Healy. Additionally, several PSS/E studies were run to analyze alternatives to opening the Intertie both north and south of Healy for OOS conditions. The studies provided some information to support changing the OOS relay settings at Healy. UAF investigated SPR options along the Intertie. Preliminary results show that SPR is feasible. Detailed dynamic simulations and equipment cost estimates would still be required before a benefit to cost evaluation is performed. The technical feasibility of SPR has been demonstrated. Attachments Mr. Ritter's Master of Science thesis is attached, along with a diskette (DOS format) containing the EMTP databases for the Railbelt. EMTP Studies - Final Report Page 11 ie - ALASKA SYSTEMS COORDINATING COUNCIL ue i: = Q \ An association of Alaska's electric power systems promoting improved reliability a \ through systems coordination. X& [EPG baa OY we 333 W. 4th Avenue, Suite 220 Council Chairman: Charles Walls Vice Chairman: Eugene Bjornstad Secretary Treasurer: Michael Irwin Executive Committee: Thomas Stahr Michael P. Kelly Steve Bushong Wayne Carmony Brad Reeve Robert Martin Jr. Anchorage, Alaska 99501-2341 Phone: (907) 269-4629 Fax: (907) ae December 17, 1996 Mr. Percy Frisby Director Division of Energy P.O. Box 190869 Anchorage, Alaska 99519-0869 Dear Mr. Exisby: The final report to the Alaska Systems Coordinating Council, titled EMTP Studies for the Railbelt Transmission System and dated December 31, 1995, is enclosed. This report includes the thesis by Mr. Douglas Ritter titled Electromagnetic Transients Study of the Alaskan Railbelt Power System and dated August 1995. A 3.5 inch diskette containing railbelt Alternative Transients Program (ATP) data files is in a pocket at the back of the report. A few additional copies are available upon request. The distribution list is attached. Please call if you have any questions. Sincerely, Gry Mel Penny L. Haldane ASCC Coordinator Enclosure as stated. Mr. Randy Simmons Executive Director Alaska Energy Authority 480 W. Tudor Road Anchorage, Alaska 99503 Mr. Sam Mathews Major Projects Engineer Homer Electric Association, Inc. 3977 Lake Street Homer, Alaska 99603 Mr. Charles Walls General Manager Alaska Village Electric Cooperative 4831 Eagle Street Anchorage, Alaska 99503 Mr. David Gerdes Acting Electric Department Superintendent Fairbanks Municipal Utilities System City of Fairbanks P.O. Box 2215 Fairbanks, Alaska 99707 Mr. Thomas Stahr Genera! Manager Anchorage Municipal Light & Power 1200 East First Avenue Anchorage, Alaska 99501 Mr. Wayne Carmony General Manager Matanuska Electric Association, Inc. P.O. Box 2929 Palmer, Alaska 99645 Mr. Eugene Bjornstad - General Manager Chugach Electric Association P.O. Box 196300 Anchorage, Alaska 99519-6300 ASCC EMTP report distribution 12/17/96 Mr. Norman L. Story General Manager Homer Electric Association, Inc. 3977 Lake Street Homer, Alaska 99603 Mr. Robert Wilkinson Interim General Manager Copper Valley Electric Association, Inc. P.O. Box 45 Glennallen, Alaska 99588 Mr. Mike Kelly General Manager Golden Valley Electric Association, Inc. P.O. Box 1249 Fairbanks, Alaska 99707 Mr. Dave Calvert Utilities Manager Seward Electric System P.O. Box 167 Seward, Alaska 99664 Mr. Nicki French Assistant Administrator Alaska Power Administration U.S. Department of Energy 2770 Sherwood Lane, Suite 2B Juneau, Alaska 99801 Mr. Percy Frisby Director Division of Energy P.O. Box 190869 Anchorage, Alaska 99519-0869 Mr. Tom Lovas Director Production Chugach Electric Association P.O. Box 196300 Anchorage, Alaska 99519-6300 Mr. Mike Massin Director Engineering Chugach Electric Association P.O. Box 196300 Anchorage, Alaska 99519-6300 Mr. Brian Hickey Power Control Manager Chugach Electric Association P.O. Box 196300 Anchorage, Alaska 99519-6300 ASCC EMTP report distribution 12/17/96 Mr. John Cooley System Planning Manager Chugach Electric Association P.O. Box 196300 Anchorage, Alaska 99519-6300 ELECTROMAGNETIC TRANSIENTS STUDY OF THE ALASKAN RAILBELT POWER SYSTEM THESIS Presented to the Faculty of the University of Alaska Fairbanks in Partial Fulfillment of the Requirements for the Degree of MASTER OF SCIENCE By Douglas R. Ritter, B.S. Fairbanks, Alaska August 1995 ABSTRACT A generalized Electromagnetic Transient Program (EMTP) database was assembled for the Alaskan Railbelt power system. Until this time, no Alaskan Railbelt power system database existed for the EMTP. This database will allow the study of electromagnetic transient phenomena in the interconnected Railbelt power system. Alaska System Coordinating Council (ASCC) members expressed concern that protective relays along the Anchorage to Fairbanks Intertie were set incorrectly. A series of computer simulations were undertaken to determine if this was true. One simulation indicates that a relay may have operated improperly. Suggestions are made as to how the relay setting may be modified. A technique to improve system stability is also presented. ASCC members demonstrated interest in studying the applicability of single-pole reclosing (SPR) on the Intertie. This subject was studied using computer simulations. This work indicates that SPR on the Intertie is technically feasible. Parameters for SPR implementation and operation are suggested. TABLE OF CONTENTS SA Ceca ter ee wares oe ec oege csrs coe sce renee ea iii LIST OF FIGURES © csscsscsccssscescateses suc cusss sees evs sceseonssovsavseas casacsstavds ces ceveneasescanceaveseavaces vii | AS 2a Gd 0 59) Dh eer ner ea peer pe cecre rarer eae rrrerey rece errer reenter x ACKNOWLEDGMENTS wsscc sees estecceacesersatescvsssesasavesascvsstsvereves eae ravnmeevianseresvenerseevtors xi CHAPTER 1 ALASKAN RAILBELT POWER SYSTEM EMTP DATABASE..........2...::05 l De USSU GUCtI OM So cscs ccc cs cece cc cess S en enceeccacecece es fan eneceeg tes cote tecestcceseses tes seessces 1 1:2 Deseription, of the EM UP i ssseecssesescces cocaine cosesesteesstusssvsnsstsienmerancmessestesmsareerssseores 1 Se MVS iA Pass etcc ose cetecctcdct oes tcsesascptedetcedsetceo>tcreanect os ide rsottaonstesssiteetts-ceccssbreoseacs 2 154 General Database DeScrip ttm xc sescse ee seca ec ete tse eee ee 3 1.5 Individual Utility Database Descriptions. asc sssscssssscaze st cas ccsexeseé save coseseseeasccssdseceves sveies 5 1.5.1 Anchorage Municipal Light and Power (AMPL.DAT) .........eceeeeseeeseeteseeerees 5 1.5.2 Chugach-Electrical Association (CEA.DAT) ...........s5...055.-...csscossecensonssntesotessees 5 1.5.6 Fairbanks Municipal Utilities System (FMUS.DAT).......00..e cece 6 1.5.3 Golden Valley Electric Association (GVEA.DAT) ............cceesessesseesseeeeseeneeees 6 1.5.4 Homer Electric Association (HEA.DAT) .0.....eieceeseeeeeeeccesceeeceeeeeeeseeeeeseeaee 7 1.5.5 Matanuska Electric Association VIA A ea sce ee ee 7 M50 OMe MUNSON aa aca acca occa ccs cca sneered oo scacuava stan eermacssemumeas aetraee coe naar 7 CHAPTER 2 HEALY OUT-OF-STEP PROTECTIVE RELAYS. se: scscs. ccczevsrerseenscsaversasarssess cores 9 Dea Lith OC Ct OM tetseeae sets cere cee nase wrists cece e av iecg: tues oetemssuaeeeatenessaereessreeenstesentaes 9 2.2 Case 1: North Pole Gas Turbine #1 Unit Trip...........ccscesscsssssssesseeeesenseeseeeeseseneees ll - Vv 2-3 \@ase 1/Simulation, Results ...).-1.:--.-c-scc-s-s-ceoconsescetecseoncnssog snaeateactous sss ssecssusnarraceseerss 12 2.4 Case 2: Line to Line Fault at Musk Ox 69 kV Substation .....0..0...ccececceceeeeeeeseeeeee 19 2.5)1 Gases Simulation Results). 2.2.2.2. s.-a-csceszesacsaess tase saseesanstosesnaseessaeess sess otsseesestespsssoes 20 2.6 Case 3: Stability Improvement Metho ...................-:.--ccccssecesecceseeseccecsceceesoceeseesecees Si 2:6:1 Case 3A Simulation, Description.....:....c...s20csc«csocseosoeccesecsacceeescon-seceeaceeetereareee 38 216.2 (Case SA /Simulation Results x: secsc: sacexesde anc stance scsaucnseacssensessoossusesronstsesssestneuewese 38 2.6.3 Case 3B Simulation Description ...........c.cceeseescesceeeseceeceseeeeeeeesseaceaeeeeseeeesees 39 2:64 [Case SB Simulation; RESUS a stsee teeta toestatonetec aaesns tena see tee tae seen ewe settee 39 2:6:5 [(Gase 5G Simulation: Description 2 ct. tee-ceeseroneseceeecectet cat teense eee 50 2.616) Case © Simmlation) Resultsit. 1... secesssscctaccecses-evecsncestoecsseceouearsnceorsnssaroice ssi fies 50 26577: CaseSD) Simmlation| Description .....sx-ac.ces-cesaseessasecosvessectea= dasasveesnesseseseseoe-95ee4 56 2.6.8: Case 3D’ Simnullation\ Results 1.2 0.5 o-t.ca.ccec--ncaae cece sas at oncsesrescocctesntetesesctates 56 2.6.9 Case SE Simulation Description a..secsssecvescercseseseseseccevceceeretenessecseee eneseererotes 56 2:6! LO (Case 3E, Simulation) Results: isrcccscsvcerectesreveretrscreeosectanssarsecter seston tsetenie ess 62 DT) (CASe BISUMIMATICS. | ssc rercteseceeeensecrenetececeoneneetec neta ene eed OTe OEE eee eee eee 62 2.8 Discussion of Stability Improvement Method .00.........ccceccececeseeeseceseceeeeeeeseeeeeeeaees 70 DD COMCIUSION. 2 -:--2.5-ccescocegnasessessusstes nt csusnt aps asusasese SPs tosses xen ave yes swsssssassseseteassxcsssewtevone 70 CHAPTER 3 SINGLE-POLE RECLOSING ON THE INTERTIE 000... cs cseseseeeeees 72 SMM) Mntroduction\..0.).2.t.cs-ccs-sas-ncccssoccnssnovatses'ssesusassoaesuseusnevsrsrstarsgssouedursorsersucsnecessvesonsan 72 32 Description of Single-Pole!Reclosing)....-1t.t-s---.1-2-c-4--saceucecvossesasantsnrsvosnsanseaverorsesee 72 325 COmputer Simiulatiom DCSCrIPtlONS taccavewogeconssnaercoctstcscsacsnsserseseotesecsuesceoeesescusaees stan 78 3,341) (Description of the EMTP'System Models ortett-t cess ansesseescssaseonesnstssesrenge sees 79 3.4 Transient Stability Results: Dead-Time Determimation..................:cccessesessesseseseeeenes 82 = vi 3.5 EMTP Results: Secondary Arc Currents, Transient Recovery Voltages................... 96 316 Overvoltages/ During: Smgle-Pole,Reclosin g<-...c.. .sc<sscescscis-sovasensesescssnasswes asses seavosses 97 5:7 Secondary Arc) Neutralization Methods i vcccresescsscco ste esa ten ceca mmo eanwscacesunesnbacssaxeahs 101 3.8 Heating of the Healy Unit due to Negative Sequence Currents ..............::c:ceeseees 111 3.9 Electrical Torque of the Healy Unit During Single-Pole Reclosing...................05 2) Sel OPAdaptive Reclosin geecries tases osccnesceastace sere espns saree celeste tr adeiaadonssccsnssanoeesenrsaes 112 3.11 Implementing Single-Pole Reclosing on the Intertie.............eeeeeeeeeeseseeseeeeeeeees aS Be T2RC@omclhusio neta eae ac nrc an oes co ntese eaves cerste teats cceoe tse cobseaton compen Sesses 116 RRERE REN CES ose see rata rtaesesestunhe lien leceneseccocentecocbadus ont siyussasonsoutcorsussisucwsusseeress 118 APPENDIX I ELECTROMAGNETIC TRANSIENTS PROGRAM DATA FILE.................. 121 APPENDIX IL DIFFERENCES BEWTEEN TRANSIENT STABILITY PROGRAMS AND PE CVE De resists sossveresscsceceorsecutaassascessece pipe sstcencscets csooaes-crccecsrantoscousesmsasssososasens@ 138 Figure 2-1 Figure 2-2 Figure 2-3 Figure 2-4 Figure 2-5 Figure 2-6 Figure 2-7 Figure 2-8 Figure 2-9 Figure 2-10 Figure 2-11 Figure 2-12 Figure 2-13 Figure 2-14 Figure 2-15 Figure 2-16 Figure 2-17 Figure 2-18 Figure 2-19 Figure 2-20 Figure 2-21 Figure 2-22 LIST OF FIGURES Conditions at Gold Hill Substation, Case 1 Disturbance (1/10/94) ............. 13 Conditions at Gold Hill Substation, Case 1 Disturbance (1/10/94)............. 14 Conditions at Gold Hill Substation, Case 1 Disturbance (1/10/94)............. 16 Conditions at Gold Hill Substation, Case 1 Disturbance (1/10/94)............. 17 Conditions at Gold Hill Substation, Case 1 Disturbance (1/10/94)............. 18 Conditions at Gold Hill Substation, Case 2 Disturbance (11/20/94) ........... 21 Conditions at Gold Hill Substation, Case 2 Disturbance (11/20/94) ........... 22 Conditions at Gold Hill Substation, Case 2 Disturbance (11/20/94) ........... 24 Conditions at Gold Hill Substation, Case 2 Disturbance (11/20/94)........... 25 Conditions at Gold Hill Substation, Case 2 Disturbance (11/20/94) ........... 26 (Conditions: at Gold) FRM yors5scccccccnconsocszsscscseccsaceasexcssssssnsarests sos suctsnseatssstenkes 27 Machine (& Bus FrequencieS..<.:..<.<....0c.csocsusexesasesene sessessasesooseseustesensersscssqree 28 GVEAUSystem! Voltages iecescecccccesseccsecsasessscxcovesssessesses300stésesssesessseceonccaroeees 29 Rotor Angle Differene:...........s<c.+sensseseucssensencesesosesosoesoutessorsuressqusnrooreaseso 30 MBIT O WET FIO WS ccacrecrscasresesessacnazsecesscnesserucsusencecsaterecasecestccesucececesacsessessee 31 Healy: North RX: Blots cesosoxcaqewcon sasecconsvsaes susscscvsassosstdsssossevs SeotecsosserseSs0 10a oe 33 Healy South RXORIOG incr. srencscecessstscssecsusuvarssouscesnsesessesisscnssesstersesneceseeets 34 Westinghouse SDBU-2 Out-of-Step Relay Block Diagram............:s:00+ 35 Machine Sc: Bus Frequencies ..-c.ccscerscsasecsessscsscersocessnscccescssesetversencessestsses? 40 GVEAVSYStemM VOltAgeS.<.....:.<:...0s.s0-cnscnssovsnsacesessesssesusessoosseroncescessesssseess 41 RotorAnigle Differene vireseccscccteccecsssusnsecssscsoecnecsusexsorcesessesecrivevesisseses=tsee9 42 Healy North RX: PlOb .s.s.csscessseceecevensususosarcasssesvasesssoccoseseoseseronsossosesessssessed 43 Figure 2-23 - Line Power FLOWS ssssss.scsscssscsussnsessosoossvevcoos swasnsresvssessucssessrshvassevevtssusesvasases 44 Bigure 2-24 --Machirie & Bus) Frequencies i2gc. cscs cos. oeee ets seseseerss opp sensei ccns ee cscctns 45 Figure 2-25 GVEA System Voltages. sssssscesssssavesanssazsacussvessessecsavesetsosazesetecsesserteaeais 46 Figure 2-26 Rotor Angle Difference 00.0... cee esecseesesseesccsecseessceseeeseesseeeaeeaeseeceaeeenees 47 Figure 2:27 — Healy Nortlt RX Blo tess iccscccseccsscsssacsscensssacovcsssaacsssvasssvicostoetessrsaatsasseeconneud 48 Figure 2-28 Line Power PLOWS ssecccceasssscestsasavsconsansvenssesvecesssassenvavseatvsnvunseossansesaseaccsueced 49 Figure 2-29 Machine & Bus Frequencies ...........cceesssecsecsesscsseseeseeseeseesecseeseseeeseseeaees 51 Figure 2-30 GVEA System Voltages iis ssciccs.asscscessvessossaus ove soveissovsssusvssessuastivesestsvenenes 32 Figure 2-31 Rotor Angle Difference 0.0.0.0... eeeceseeseesceecsseeseesecsseeeeessseseneesssenseessens 53 Figure 2-32 “Healy North RX: Plotesccsscc. <ssccascccvscssnsecspcssaseceossonseas sosavsnssosovseussesssaveonssss 54 Figure 2-33 Lime Power FIOWS .........cccssssssssssssesescscsessssssssesceessessssssesesscsessceesesecssseeesees 55 Figure 2-34 —- Machine & Bus Frequencies sssssasscccavesassesssssacovesesssscnsssenseveasossasossseraessvesve 57 Figure 2-35. GVEA ‘System: Voltages iicaicccss.cccoccessssssesssvassensoossudsveusanenseosnsunnsrsssseososess 58 Figure 2-36 Rotor Angle Difference............::.-sesccrescosssonsessorssceacooversuisissssnsssssssausessoseess 59 Figure 2:37: Healy North RX. Plot ccssscccsesccssvacesversssucacecsuscveats crsvavsses cons toeocsaveutssasscoceses 60 Figure 2-38 Line Power FIOWS...........sssessscsssesssrersesesscscssersnsscsceneacensacsssacecencacaseecereees 61 Figure 2-39 = Machiné:& Bus Erequenici¢s)......s<.0scessseasccssetatececescotsisesevsssaseusvtsssteesecesesess 63 Figure 2-40 GVEA System: Voltages os. .csssssssscassisssanssessesssessesessassnconsonsessenssasessssssesouss 64 Figure 2-41 Machine Rotor Angle Difference ..........ccscsssesesesesesesesesseeesesenesesesssesenees 65 Figure 2-42 Healy North RX Plote.......ccccecccssesssseescescesceeessessceseeseesseeeeeeeseeeeseseseessengs 66 Figure 2-43 — Line Power HlOwWS sicsesscizsics sesscoveassensarsoneoatessassonaenssusssasesssoneversossosevoasisesis 67 Figure:3-1==System as'Modeled:in: AUP. << .so.sussssssosewsssuscreasvssess once cesses oven ssssevessasavensasees 83 Figure 3-2 Rotor Angle Difference .........cccsscsssssssssesessesessesessesscssseeeesesseseesseeeeensenees 84 Figure 3-3. Rotor Angle: Difference isc.<.0:<ccssersvavsusescavsususssesensssvssvessvssvssevsossosevensenexaszises 85 Figure 3-4 Figure 3-5 Figure 3-6 Figure 3-7 Figure 3-8 Figure 3-9 Figure 3-10 Figure 3-11 Figure 3-12 Figure 3-13 Figure 3-14 Figure 3-15 Figure 3-16 Figure 3-17 Figure 3-18 Figure 3-19 Figure 3-20 RotomAngle Difference ence cesctstsssccsrsrsecrenstaseccetsrsscaeeesrsesesceeesererees 86 Rotor Angle Difference strc cesseccresersseorsecsessusesssesesecrsscuesassoasersucscsorearsvessesore 87 RotomAnple Differences serecteseeccrercetesnceersectesesectotstcrcentsceseetstcesecesceteey 88 Rotom Angle Difference -ceccsccensassrocsoncscusstseseseseasetecsascrsusarerosescsusatenteess 89 IROtOTAN PIE Difference ncrercstasscesestessscseessceotsteassucectesctsnacsurscteccserteesececes 90 Machine & Bus ErequenCyictss-tccc-cesensaccoscacsessusscsncscsorscesscuscsnesecstsenswessesee® 91 Rotor Angle Ditterenceserretce terete creesecteccsrsttestareseecereecueestessestscesesets 93 Rotor Angle Difference ct stsrocccncossonsssnsccasvorseasestonsveresscsacesesosnearesnsstatsitses 94 Rotor Angle Difference exctsccstcacterstcrescccesteseseucercectserscescestenseacvacctecssaceats 95 Secondary Arc Current at Nenana, No Neutralization............::cccsesesseees 102 Secondary Arc Current at Cantwell, No Neutralization.............:sessssee 103 Secondary Arc Current at Cantwell, With Neutralization - AIREACCOMISCRE MO tee rctctsteniecseaterstctrseeatrstctcceststgeeacessarectcterstosteceatorete 105 Secondary Arc Current at Cantwell, With Neutralization - Grounding Switcheserrcctettcocercctescensrscessassesercescecsterseesececvareersesatecesnesees 107 Instantaneous Current Flow in Open Phase With No Fault, Instantaneous Current Flow in Open Phase With Fault .............::cscscesees 108 Secondary Arc Current at Cantwell, With Neutralization - Sectionalizer. ssecssseccncesesesscecessesvenssvsueasses sossusensooevsvesstsvasesosssoscavecsorsetsutves 110 Healy Generator Electrical Torque, SLG at Cantwell, Reclose TEE-HLS, Sectionalizer at Cantwell ..........:cccccssceseeeerereereees 113 Healy Generator Electrical Torque, SLG at Nenana, ReClOse HES GO erm merctsstrtctesrcsstenerssecesteraercesscrssstsertererenesesuectstcsees 114 Table 2-1 Table 3-1 Table 3-2 Table 3-3 Table 3-4 Table 3-5 LIST OF TABLES Case Simulation Conditions and Results ..........:c:cscssesessesesseeseesesseseesseseesees 10 Transient Stability Simulation Results... .5.<50<<.sss<s0ssenssessssvasssecsesesssosccsovessse 92 Secondary Arc Current, TRV, Induced Voltage; w/ 3-0 SVC Voltage Regulations sscccscccsessessssscsscessessocessucssevssssscesessecsosaess 96 Voltages at 138 kV Buses Along the Intertie During SPR (per unit) .....98, 99 Voltages at 138 kV Buses Along the Intertie During SPR (per unit); Healy exciter/voltage regulator monitors Healy 138 kV Bus................ 100 Secondary Arc Current, TRV, Induced Voltage; Teeland-Healy Line Section, With Neutralization ............:cccseseseeeeeeeee 104 ACKNOWLEDGMENTS I would like to thank Dr. James Cote for all his help on this project as my graduate advisor. This project would not have been possible without his input. I would also like to thank Dr. John Aspnes and Dr. Joseph Hawkins for their participation as members of my graduate committee. I would also like to thank the Alaska System Coordinating Council for funding this project. I also wish to thank Mr. Steve Haagenson, Golden Valley Electric Association, for his input on the material in chapter 2, and Mr. Kevin Jones and Mr. Tim Devries, Golden Valley Electric Association, were very helpful in obtaining information on portions of the Railbelt power system. I would also like to thank Mr. Gabor Furst, Furst Consultants Inc., B.C., Canada, for his assistance with the ATP and for his input on the material in chapter 3. CHAPTER 1 ALASKAN RAILBELT POWER SYSTEM EMTP DATABASE 1.1 Introduction The Alaskan Railbelt Utilities requested that an Electromagnetic Transients Program (EMTP)/Alternative Transients Program (ATP) database be created to represent a generalized model of the Alaskan Railbelt power system. This database was created and verified with results obtained from Power Technologies Inc.'s (PTI) PSS/E power flow program. The database is intended to be used as a resource for building transient program data cases. 1.2 Description of the EMTP The EMTP was designed to be used to study high speed electromagnetic transients in electrical power systems. All three phases of a power system can be represented in an EMTP model. The EMTP is typically used to study switching transients, lightning strikes, and unbalanced three-phase conditions. The EMTP has a number of built-in models to represent electrical power system equipment. Control systems can be modeled through the use of the built-in Transient Analysis of Control Systems (TACS) or the newer MODELS routines. See Appendix II for an expanded description of the EMTP. 1.3 EMTP vs. ATP Currently there are two versions of the "EMTP". One version is the Electromagnetic Transients Program and is distributed by the Electric Power Research Institute (EPRI). The other version is the Alternative Transients Program (ATP) and is distributed by the Canadian/American (CAN/AM) EMTP User Group. Both the EPRI EMTP and the ATP versions were obtained for comparison. The most common platform for both versions is the IBM PC. The EPRI EMTP version runs under the Windows operating system. The ATP version runs under DOS using the DBOS DOS extender/memory manager. DBOS is also used with the non-Windows PTI PSS/E software. The EPRI EMTP is extremely expensive for non-EPRI member utilities. The ATP version (with DBOS and utility programs) is available to almost anyone for $10. The ATP version appears to be maintained more regularly, allows much larger data input files, and runs approximately twice as fast as the EPRI EMTP (version 2.1 tested). The EPRI version does have a better graphical output processor used to create plots of simulation data. The ATP plotting programs are adequate for production work. The EPRI EMTP version would be easier to integrate into a networked computer system as it runs under Windows. The ATP version is shipped with a utility called ATPDRAW which is used to create data input files graphically. This program gives a first time user a tremendous advantage. Files created with ATPDRAW can also be used with the EPRI EMTP. The ATP version would be the logical choice for a majority of prospective EMTP/ATP users. The CAN/AM EMTP User Group supports an exchange of EMTP/ATP related information through an electronic mail list-server. An individual subscribes to this free 3 service by simply sending the list-server an e-mail address, name, and an affiliation. Once subscribed, a user can submit questions to the list-server which are then repeated to all the other list-server subscribers. Answers are provided by actual users either through the list- server or directly to your e-mail address. Many of the list-server subscribers are engineers applying ATP to real world problems. This is an excellent means of obtaining advice as to the operation and implementation of the EMTP/ATP. 1.4 General Database Description It would be very difficult to produce a database that would accurately model every piece of equipment in an electrical power system for all types of transients. In general, the equipment data required to assemble a transient data case is not supplied with the equipment from the manufacturer. Therefore, it is up to the end user to apply correct engineering judgment to produce an accurate model and interpret the results obtained. The user should take a portion of the database and augment it to accurately model the section of the system under study. This may require contacting the manufacturer of the equipment under study for information. As an example, if a person were interested in studying the transient recovery voltages produced after a breaker cleared a fault, an improved high frequency transmission line model would probably be used. Other connected equipment would probably be represented with generalized models or equivalent impedances and ideal voltage sources. It would be up to the individual to determine the impact of the simplified/generalized models on the results. The Railbelt database was assembled as six sections representing each Railbelt utility's service area as specified by the utility's one-line diagram. The following six utility systems are represented: Anchorage Municipal Light and Power (AML&P), Chugach Electric Association (CEA), Fairbanks Municipal Utilities System (FMUS), Golden Valley Electric Association (GVEA), Homer Electric Association (HEA), and the Matanuska Electric Association (MEA). The data contained in the database represent the portions of the system energized at voltages greater than 34.5 kV. All database sections have been verified by comparing steady-state results to the PTI PSS/E case "CASE4.SAV" power flow results. All sections have been configured to produce the "CASE4.SAV" results. Therefore all loads, transformer turns ratios, machine rotor angles, static var systems, and capacitor banks reflect the data necessary to produce the correct results for verification. If there was no agreement between the data represented on the utility one-line and the data in the PSS/E database, the PTI data were used. All machines are represented as an ideal voltage source behind a subtransient reactance. All static var systems are represented as fixed inductors/capacitors. Most transformers are represented as ideal single-phase transformer banks (3-phase banks) with winding impedances only. No saturation, magnetizing resistance, or internal capacitance was included. Other transformers that connect smaller loads are ignored. In this case, the loads are reflected to the transformer high side. The power flow into the load from the high side was used to calculate the load element values. 5 All transmission lines are represented using the distributed parameters, lumped resistance, traveling wave line model. No circuit-to-circuit coupling is included in any of the database sections. The end user is responsible for upgrading the model to include circuit-to-circuit (zero sequence) coupling. 1.5 Individual Utility Database Descriptions 1.5.1 Anchorage Municipal Light and Power (AMLP.DAT) In this database section AMLP units 3, 6, and 7 are on-line. Positive and zero sequence coupled impedances are connected between the generator voltage sources and the generator bus. These impedances represent the subtransient impedances. All of the 34.5 kV load appears at the 34.5 kV bus in plant #1. The tertiary windings in all of the substation transformers (115:12.5) are ignored. An ideal voltage source is connected at the East Terminal junction as an equivalent for the CEA system. MEA appears as a load connected at the Briggs Tap junction in this case. 1.5.2 Chugach Electric Association (CEA.DAT) In this database section Beluga units 1, 6, 7, and 8 are on-line. Positive and zero sequence coupled impedances are connected between the generator voltage sources and the generator bus. These impedances represent the subtransient impedances. Capacitors have been added at Pt. Woronzof to simulate PTT's sectioned line model. The Tyonek load is connected directly to the bus of Beluga unit #4. A load at Teeland (230 kV) represents the MEA/GVEA system. A load at AML&P (230 kV) represents the AMLP/MEA system. A load at Lawing (115 kV) represents the City of Seward. An ideal voltage source is connected at the Quartz Creek junction as an equivalent for the Homer Electric Association system. The out-of-service under-sea cable between Pt. Woronzof and Pt. MacKenzie is in service in this database section. 1.5.3 Fairbanks Municipal Utilities System (FMUS.DAT) In this database section the Chena 5 unit is on-line. Positive and zero sequence coupled impedances are connected between the generator voltage sources and the generator bus. These impedances represent the subtransient impedances. The GVEA system is represented by an inductance in this case. 1.5.4 Golden Valley Electric Association (GVEA.DAT) In this database section Healy and Zender G.T. #1 are on-line. Positive and zero sequence coupled impedances are connected between the generator voltage sources and the generator bus. These impedances represent the subtransient impedances. An ideal voltage source at Teeland represents the CEA/MEA system. An ideal voltage source at the Ft. Wainwright power plant represents the Ft. Wainwright system. University of Alaska (UAF) and Eielson Air Force Base are represented as loads. 1.5.5 Homer Electric Association (HEA.DAT) In this database section Bradley Lake units 1 and 2 and Cooper Lake units 1 and 2 are on- line. Positive and zero sequence coupled impedances are connected between the generator voltage sources and the generator bus. These impedances represent the subtransient impedances. The CEA system is represented by a load at Dave's Creek. The City of Seward is represented by a load at the Lawing junction. 1.5.6 Matanuska Electric Association (MEA.DAT) In this database section Eklutna units 1 and 2 are on-line. An ideal voltage source connected at AML&P plant #2 represents the AML&P system. An ideal voltage source at Pt. MacKenzie (230 kV) represents the CEA system. A load at Douglas represents the GVEA system. 1.6 Conclusion A generalized EMTP/ATP database of the Alaskan Railbelt transmission system was assembled. The database should be used as a starting point for transient studies in the Alaskan Railbelt power system. As each transient phenomenon has different data requirements, it would be very difficult to assemble a database that would represent every transient condition. Therefore, the end user is required to augment the database with an appropriate level of detail for each study. Validation of the database sections with PTI power flow results has eliminated all known errors. Two versions of the EMTP were obtained and compared. Relevant strengths and weaknesses of the EPRI EMTP and ATP versions were mentioned. The ATP version has many features in its favor. It is substantially less expensive for non-EPRI members, it is maintained (updated) more regularly, and it can run twice as fast as the EPRI EMTP. The EPRI EMTP version should be considered if the EMTP is to be used in a networked environment. CHAPTER 2 HEALY OUT-OF-STEP PROTECTIVE RELAYS 2.1 Introduction The Alaska Systems Coordinating Council (ASCC) provided funding to the University of Alaska Fairbanks Electrical Engineering department to support study of the Alaskan Railbelt power system. University of Alaska Fairbanks faculty and ASCC members identified potential areas of study. One of these areas included investigating the out-of- step relays along the Intertie between Fairbanks and Anchorage. Concerns had been raised that the out-of-step relays along the Intertie may need to be reset. While investigating these concerns, Golden Valley Electric Association (GVEA) was approached about two recent disturbances on its system. GVEA had determined that these disturbances appeared to have caused improper actuation of the out-of-step protective relays at Healy. It was suggested that studying these disturbances might provide some insight into why the relays tripped incorrectly. It was decided that the Electromagnetic Transients Program (EMTP) was not adequate for a study of this nature (Appendix II). Therefore, transient stability work was performed using Power Technologies Incorporated's PSS/E (version 22.0a) software. A series of transient stability cases were simulated with the PSS/E software. The first of Table 2-1. Case simulation conditions and results Healy North Pole Under Freq. Direct Trip Generation Generation Intertie LoadShed Load Shed Case Simulation (MW) (MW) Trip (MW) (MW) Out-of-Step? Stable ? 1 North Pole Trip, 1/10/94 0 50 = ze a a a 2 Line-Line Fault, 11/20/94 26 0 Yes 3.4 MW n/a No No 3A N.P. Trip 26 50 Yes 81 0 Yes ese** 3B N.P. Trip 26 50 HLS** 82 0 No Yes 3C N.P. Trip 26 50 No 0 18.1 No Yes 3D N.P. Trip 26 50 No 0 6.2 No Yes 3E ~=N.P. Trip & Fault Buses 26 50 No 0 18.1 No; All Cases Yes * Poor simulation results due to a lack of detailed information ** The Intertie was opened to the south of Healy. *** See text for further explanation. 11 these cases was an attempt to compare simulation results with actual Dynamic System Monitoring (DSM) data. The second case was a simulation of a line-to-line fault in the GVEA system that caused the out-of-step relays at Healy to operate. The third case and its associated subcases are simulations demonstrating a method that could possibly be used to improve system stability in the Fairbanks area in the event of a loss of local generation. See Table 2-1. 2.2 Case 1: North Pole Gas Turbine #1 Unit Trip Case 1 models an actual disturbance and was run to compare simulation results with GVEA dynamic system monitoring (DSM) data. On 1/10/94, gas turbine #1 at North Pole was producing 50 MW. The 26 MW Healy coal plant was off-line for maintenance. Intertie flow into the Gold Hill substation was 45 MW. Fairbanks Municipal Utilities System (FMUS) operation was unknown. The FMUS system was assumed to be tied to GVEA with no power transfer as is the normal operating procedure. University of Alaska Fairbanks (UVP), Ft. Wainwright (FWP), and Eielson (ELP) actual system operations were unknown. The following conditions were assumed from the DSM plots supplied by GVEA: UAF was tied to GVEA but the power transfer was unknown, Ft. Wainwright was tied to GVEA but power transfer was unknown; the tie between Eielson and GVEA was open. GVEA supplied information indicated that Ft. Greely (FGP) was tied to GVEA and generating. It was assumed that Ft. Greely was not a significant factor in this simulation and therefore was treated as a load. The following reasons support this assumption: Total generation capability at Ft. Greely is 5.5 MW if all diesel units are on- line. Ft. Greely is on the end of a weak transmission line. 12 Based on the GVEA DSM data, the following events occurred during this disturbance. A sensor tripped, thereby shutting fuel off to North Pole unit #1 at t=0 seconds. The unit output ramped down over approximately 0.23 s before the unit breaker was opened (soft trip), disconnecting the unit from the GVEA system. This disturbance caused the impedance trajectory, as seen by the distance relay at Gold Hill, to enter the relay's protective zone 2 causing the GHS1B1 breaker to open at t=0.88 s, disconnecting the GVEA system from the Intertie at Gold Hill. GVEA under-frequency load shedding began at t=1.08 s. The breaker tying GVEA to the UAF power plant opened at t=1.12 s. The tie between Ft. Wainwright and GVEA was opened at t=1.25 s. The tie between FMUS and GVEA was assumed to have opened at t=1.35 s. Without any remaining local generation the GVEA system collapsed by t=1.92 s. The above information was used to create a PSS/E simulation. 2.3 Case 1 Simulation Results Attempts to model this disturbance proved difficult. Many problems were encountered when trying to make this simulation run. It was necessary to modify default network solution parameters. The integration time step was reduced to 0.002 s, the acceleration factor was reduced to 0.1, and the maximum number of power flow iterations per time step was increased to 1000. Comparing GVEA supplied DSM data (figure 2-1) collected at the Gold Hill substation to simulation results shows the simulation voltage at the Gold Hill 138 kV bus falling off too rapidly (figure 2-2). The simulation voltage recovers to values exceeding 1.2 pu (96 kV) around t=.85 s. The simulation Static VAR 1-7 on3iy Conditiensiat ColdeHill Substation Cdee iavuppameer( ly da aay) 60.0 59.5 ——Intertie Power Flow (MW) - — SVC Output (MVAR) Pete ----Va (kV, L-G) Frequency (Hz.) 58.5 58.0 57.5 57.0 56.5 kV, MW, MVAR 56.0 55.5 55.0 O10 OL Zee OLA Os6 te Ore eOmmabe2rale Are ed Gunn leGui 2-0 Frequency (Hz.) £1 7-7 OUNS14 kV Conditions at Gold Hill Substation Case 1 Disturbance (1/10/94) 90 80 70 60 50 40 —DSM, Va (kV, L-G) 30 - — Simulation, Va (kV, L—G) 20 10 vl 15 Compensator (SVC) output (figure 2-3) and Intertie flow into Gold Hill (figure 2-4) both fall off rapidly due to the low voltage. The simulation frequency waveform looks similar to the DSM data (figure 2-5) except for a large dip at approximately 0.72 s. The simulation network solution began to diverge at approximately 1.35 s and the simulation was stopped at this point. One reason for the dramatic voltage decay was thought to be a lack of accurate load modeling in the PTI supplied base case. All loads are modeled as constant current and constant susceptance loads. The MAPCO refinery uses a number of large motors in their operation. It was thought that these motors might provide some form of voltage support. In an attempt to improve simulation results, the PSS/E Complex Load (CLOAD) model was applied to the MAPCO 13.8 kV bus in the database. The CLOAD model was selected because it only requires the user to supply a percentage of load type. The entire load at the MAPCO bus was modeled as large motor load. Unfortunately, adding the CLOAD model at the MAPCO bus caused this case's network solution to diverge early in the simulation. Every effort was made to make the case converge, without satisfactory results. Due to a lack of detailed information and more DSM data it was decided that nothing more could be gained from further simulation of this case. More information concerning basic GVEA system operation must be established before any additional work can be completed. This case did provide an opportunity to learn the operation of the PSS/E software. €-7 ONSLY MVAR Conditions at Gold Hill Substation Case 1 Disturbance (1/10/94) 90 80 4 - — Simulation SVC Output —DSM SVC Output 70 4 60 50 40 30 20 10 91 p-7 oun31y Conditions at Gold Hill Substation Case 1 Disturbance (1/10/94) ——DSM Intertie Power Flow - — Simulation Intertie Power Flow 0.0 O02 04 06 08 10 1.2 1.4 1.6 1.8 2.0 Time (s) LI S-7 aunt Conditions at Gold Hill Substation Case 1 Disturbance (1/10/94) Se —,- 60.0 PS L 59.5 ; Ne LY i SS + 59.0 \ 3 \ / F 58.5 \ I \ + 58.0 a | \y L 57.5 b L 57.0 wy L 56.5 seeeey DSM Frequency (Hz.) Vin --—Simulation Frequency (Hz.) if F 56.0 t 55.5 — } 44+ 55.0 0.0 0.2 O04 06 08 $1.0 1.2 1.4 1.6 1.8 2.0 Time (s) Frequency (Hz.) 81 19 2.4 Case 2: Line to Line Fault at Musk Ox 69 kV Substation On November 20th, 1994 a line-to-line fault occurred on the 69 kV line near the Musk Ox 69 kV substation. Operating conditions are known for this disturbance. The Healy plant was base loaded at 26 MW. Intertie flow into Gold Hill was approximately 80 MW. North Pole generation was off-line. FMUS was tied to GVEA but was transferring no power. Ft. Wainwright had 10 MW of generation on-line and was exporting 2 MW to GVEA. The ties to both UAF and Eielson were assumed open. The following outlines the system events that occurred during the disturbance. This information is based on the GVEA DSM data and the GVEA disturbance report generated for this disturbance. The L-L fault near the Musk Ox 69 kV substation occurred at t=.09 s. At t=.42 s the Gold Hill SVC tripped off-line on an under-voltage condition. At t=.57 s the GHS1B4 breaker at Gold Hill opened on distance relaying. At t=.64 s the ZNS1B3 breaker at Zehnder opened on distance relay REJO!. These two breaker operations removed the Gold Hill to Zehnder 69 kV line from service. At t=.69 s the HLS1B6 (north) and HLS1B7 (south) breakers opened due to out-of-step conditions, interrupting Intertie flow to Fairbanks and islanding Healy. Ft. Wainwright and FMUS trips are assumed to occur at t=.73 s and t=.95 s respectively. This was determined from the DSM 'REJO- Remote-End-Just-Opened is a Schweitzer distance relay option that allows the relay to provide accelerated tripping times. When the relay "sees" a sudden increase in its fault current contribution it assumes that the remote breaker has opened and that the fault remains on the line. REJO operates on unbalanced faults and senses when at least one phase current exceeds the REJO current detector. This method allows differentiation between the remote breaker opening due to an internal unbalanced fault and normal opening to interrupt load current. 20 data describing the January 10th, 1994, disturbance and this disturbance. Without local generation the system collapsed after 75 cycles had elapsed. A bus (#105) representing the Musk Ox 69 kV bus at the Musk Ox substation was added to the base case. As supplied, the base case lumped the line and transformer impedances together to represent the system between the Musk Ox 69 kV tap and Musk Ox 12.5 kV bus. Modifications were made to base case generation and load to match the conditions.at the time of the disturbance. A dynamics run simulating the disturbance was executed. The PSS/E transient stability software cannot accurately model unsymmetrical faults because it uses a single phase, positive sequence only representation of the system under study. A L-L fault was simulated by using the PSS/E power flow and fault analysis program to calculate the L-L fault impedance at the Musk Ox 69 kV bus. The negative sequence portion of the calculated fault impedance was applied as an admittance and susceptance at the Musk Ox 69 kV bus during the dynamics simulation run. This method provides correct transient stability results for the positive sequence only. 2.5 Case 2 Simulation Results The simulation and DSM data (figure 2-6) compare well considering the fault calculation method and assuming a fault impedance based on an unknown faulting element (unknown fault impedance). Simulation frequency and DSM frequency look similar (figure 2-7) when one ignores the large, inaccurate frequency spikes in the DSM data. The simulation frequency decay rate is somewhat slower than the DSM decay rate. 3.4 MW of load is 9-7 auns1 kV, MW, MVAR Conditions at Gold Hill Substation 90 80 70 60 50 40 30 20 10 Case 2 Disturbance (11/20/94) 65 a | aii WA I vi F 64 1 \ ' x ' ‘ ! \ r 63 t ‘ 1 \ i ts ~---7, 4 p re - 62 ‘ , ‘ AU, ‘ Hs \ N bil 61 x XN Newewoooonn. il t+ 60 —Intertie Pwr FLow (MW) |. 59 - — SVC Output (MVAR) ----Va (kV, L-G) sosees Frequency (Hz.) r 58 - 57 56 Frequency (Hz.) 17 LT Guns1y Conditions at Gold Hill Substation Case 2 Disturbance (11/20/94) 65 serene DSM Frequency (Hz.) --—SlImulation Frequency (Hz.) -L 64 air r 63 N i ees i PB || | ees : © a op Ge i = i Oo i { rg + 58 a i F - : + 57 | ‘ ad aaa a aT la aT a Aaa ee 0:0) | (0227 || 074) |) 026 |) (0-8) |) 10) |) 1.2) | | 154 | |) 11561) 428) | 220 Time (s) ct 23 shed via under-frequency load shedding in the simulation. Simulation Intertie power flow compares favorably. Simulation voltage compares fairly well with the DSM data (figure 2-8) until approximately t=.73 s where the simulation voltage falls off to zero. DSM voltage drops off and then recovers to a peak at t=1.3 s. While the DSM voltage is recovering, DSM frequency is quickly decreasing to less than 56 Hz. It appears from the DSM data that there is still an energy source somewhere in the GVEA system. It is unknown what element in the GVEA system is providing this energy. According to the GVEA disturbance report all generation in the area had tripped off-line by t=.95 s. Simulation SVC output (figure 2-9) and Intertie power flow into Gold Hill (figure 2-10) are related to the voltage and agrees with the DSM data considering the lower voltages experienced in the simulation. A second plot was created with a smaller time scale to show the first one second of data only (figure 2-11). A series of plots were generated describing conditions along the Intertie to see how this disturbance affected other portions of the Railbelt system. Figure 2-12 shows the frequency at the Healy unit, the FMUS Chena 5 unit, and Chugach Electric Association's (CEA) Beluga 138 kV bus. Healy accelerates after being disconnected from the system. Figure 2-13 shows voltages at various locations in the GVEA system. Figure 2-14 plots the difference in rotor angles between Healy and CEA's Beluga 6 unit and Healy and the Chena 5 unit. This plot shows the rotor angle difference oscillating, but does not indicate an out-of-step condition before the Intertie is opened. Figure 2-15 shows the line power transfers at three different points on the Intertie and on the 69 kV tie between FMUS and GVEA. An impedance plot at Healy, looking north along the Intertie, is shown in figure 8-7 ans kV Conditions at Gold Hill Substation Case 2 Disturbance (11/20/94) 90 80 70 60 50 40 30 4 20 4 se | 1 ——DSM Va (kV, L-G) 10 4 I - — Simulation Va (kV, L—G) I | 6 ty T T T T T T T T TT T T T — {7 TT 0.0 0.2 0.4 0.6 08 1.0 1.2 1.4 1.6 1.8 2.0 Time (s) ve 6-7 OANSLY MVAR Conditions at Gold Hill Substation Case 2 Disturbance (11/20/94) ——DSM SVC Output (MVAR) - — Simulation SVC Output (MVAR) 0.0 0.2 O04 0.66 08 1.0 1.2 1.4 1.6 1.8 2.0 Time (s) St OI-7 auns1y MW Conditions at Gold Hill Substation Case 2 Disturbance (11/20/94) 90 ——DSM Intertie Power Flow (MW) 80 - — Simulation Intertie Power Flow (MW) 0.0 0.2 0.4 0.6 08 1.0 1.2 1.4 1.6 1.8 2.0 Time (s) 97 a TIIH G109 LY SNOILIONOS (SONOI3S) 3WIL LEt2t Sool 12 HOW *3NL eoNdo ae 00006°0 00002 °0 0000S ‘0 0000€*O Ht 00001 °0 Hl 0000! 00008°0 00009°0 0000h‘0 00002°0 o'o tea ea innih 7 » jo | ie (EL ] |||) || |e | b ‘ | ' ' 2)! LW ' ; | ‘ | ' | wo ' | o ‘ 5) ‘ 4 | + o 4 fs a o mo i ea ° ie ey) J) RO || co ME ||| -e5I))| || heal | " ALN a TEST Mg _ ; ais As = = esas] oO — =| ~) 1 =o o SH a eines a So cn a 2 =| x Pl Neetesics xf) || gst) o or 5 ai) eH |||) hy | ° pee ti = a x oO Seis fy) 22] || G2 | 3 se & ||'S1 oy S| uw su Pal ou S| FE [omy SU) see ed a o 3B oO) os = can). || fr ||) ca tof) |) I ac =o o Cs) me (8 | ra} 3 eae et SW iil = = i ie a a oa x r = wo vo | c —z oe | z= az an a = 8 w a = ey atta S,|}| eaiees = ar uo x w o se Ne ||| ens Te iS )|||\|o 2 5 a) | I ce 8 Meliss se. eM eis: By oe |e |e + 0 “ S3JIINSNOAYS SNA ¥ ANIHIUW QI'St S661 S2 Hab *3NL ISONOJAS) aH 00006°0 00002 0 0000S‘0 O000E 0 00001 "0 0000°T 00008'0 00009°0 0000h‘0 00002°0 0°O 3 le |e | | so |o lo i ‘ win fan : ' ‘ ‘ e ie te ‘ : ‘ ' Ce | mee lets ' ' ‘ ' " ‘ 1 ' ° y ' : | ' o f oC : | 2 : | ot ee a o =z S| S w oo 6S + =o + 0 4 Fo a) q Qo N u cl — . ew fon 2 > 1 = = a of % a =H Orv 2 oO tral o Ss “m7 | oa TF TwWZ 2 | =| or 5S > a oF SI exe Bg} ca) >-~ Oo . u SI x Ww sw x S =) Sr J S| “ | ao Hr i «| gc > u + al = c zm © 3 ao © a ® i ru J) oI 2 / aT Z| = sl 2 x I z ac oa | zs) x wo ac = oO andl Zaz ao c od WwW we ee 2 aw = Se =z Ge wo x WwW QD oe ao tz = go | o lo fo 2 o |o lo 3 wo [wo lw sis] a | os Ha SB |S |S Qa “SN S3H9ULIOA W3LSAS USAID 91?S1 Sb Se H4b *3Nk nsONGoeS) Baul 00006°0 00002 ‘0 0000S‘0 OO00E 0 000010 aa 7 000080 00009°0 0000h*O 00002°0 0°O © jo |e [eo o o o So s |S |s {8 s |s {8 [8 x 2 |& 18 s |o fo |e ‘ t ° v ‘ t : | ' 4 | ' ! fo ij feo * ‘ w ' ' oc ' l , a c ' ; | t 4 . + fe db f ' « \Omlccomict iat ici bet a ar qooee> + wo ' . Sz i ' Ww ; : =o ot or oa oo ‘ : co dy = | =a ' 5 " ss} S} Ss} LL. | yoo _| sa . ew ~ pot ' 1 = rs a 5 BOB ' ' a = e@ruvu | =| -| ' o cm i re = i ' 2 oc Cw2Z uy * SHS a4 4 2 3, 2 za ul = oO wi Slassas -— se soc} S| in SRST gy o 8 ag * w sl | -,| oof «| ' | SE ad wy -| a | ' a KEa eee ea : cient Joy oye ; o )® Zz} ad 2 yoo. eu x ZI x x ‘ . ae ay a a4 2 3k ' a or ' = wo : ac Me RR agente comity ouaee mttaer berm 6. Bt a Beedle eileen Rewer eee ell: © sees See ces pms ies meme enn eens eonemenn ome sdoccce oO ad ' z & = ' ao i" c qui ' Ww u ' = ae { ee Fame ccs mre tres lien eee R rece Eocene a eee Mis eee een ee ee | p z= su ‘ =z ae ; ue ' sft ws i Oak , ; ao ‘Zz , ' == et teed eee Mb eee dees eee eee ne = ' i rf f k f ' e ais (|snie i f ' : q i ! | i i SSeS ‘ ; : ; ; : : ' ' ay nl ee | 0 a SINSYISSIO JIONY YOLOY 8O?hI So66l BI YBW “LHS (SONOI3S) JWIL 00006°0 00002 ‘0 0000S‘0 O000E 0 00001°O 0000°! 00008 °0 00009°0 0000h"0 00002°0 0°0 co jo fo s |o fo @ |o |o ° v i | | wo | o | A al) on < ali al! = . Ve lb > . a Ao wn | 4] = = gq OQ) a) ia SS} a -—a y | a co Ss} al " co oo) 6G + = oa od | S ou a al a x oe Tw2a g ora | -| | 2g za x= >{ >} > : Ges a al 2 te = 2 Oe ad bs x =| a wor— ul ul ul a = we wd . a o 7 @ ot mh 4 cy . ar ; { a| of a a oz at Ma = wo ac —-Zz wn Ww w a " "4 = 86 7 3 2 eo ay zr x x e z a S| OS es = Jl 2 32 =z ce wa as Ww o je eo 1-2 ee “€ co jo fo 5 Ss |8 |s . cs |o |o yet S| 8 1 om BRADLEY © SOMW. 1994 WINTER NORMAL. SMO14 Y3MOd JNI1 LIST Sol S2 yd *3NL 00006°0 0000'! tT r 00008 '0 00002 0 00009°0 (SONOI43S) 00005 "0 SWI L 0000h"0 0000E°0 00002°0 000010 eo lo |o So > |S © [o ° ° 3 + e v o ‘ o ' | o ' : | + o 4 e « wo zz w = a. ‘ og co = Fa = " = Se & Zt = es Ss St og ero os yy cm 2 ; z| cTwZ fe) Z| a ors wo wy za vo os wl x x} of x Oo - oO} uy su uo he - xzZ4 = o} ore c al z Dou > zou . ‘ oO - 3] z = «@ Z| Z| x Zz x ar | oO = BE | on wo —£ I az ao ai u. « sw 5— ce wo w ae 1z I- 2 o lo fo = o fo |o 3 eo jo fo see) s lo |o ¥e¥ oc jn |e CCANT-HEAL (P) J CHNL#® 217: -10.00 90.900 Figure 2-15 32 2-16. The rightmost parallel lines represent the right inner and outer blinders of the Westinghouse SDBU-2 out-of-step relay at Healy, looking north. The left parallel line is the transmission line impedance line. The actual relay also has another set of inner and outer blinders to the left of the transmission line impedance line. The large arc is the impedance circle of the supervisory over-current relay. In this simulation, the north- looking SDBU-2 would not have tripped on an out-of-step condition. An impedance plot at Healy, looking south along the Intertie, is shown in figure 2-17. One set of blinders are plotted as in figure 2-16. The relay settings were assumed to be the same for the south- looking SDBU-2. This plot indicates that the south-looking SDBU-2 relay will trip on out-of-step. The impedance trajectory is seen passing outside the supervisory over- current relay impedance circle. Before this occurs however, all requirements to initiate an out-of-step trip have been fullfilled. The trip would occur 20 ms after the impedance trajectory passes outside the outer blinder. The following conditions must be satisfied to create a SDBU-2 out-of-step trip: 1) Either the instantaneous over-current (50) type or 3-@ distance (21-2, 3) type supervisory relay must pick up. This relay must remain set until the 20 ms constraint of step 4 is met. See figure 2-18. 2) Assuming a right to left impedance trajectory; the B4 outer blinder picks up. 3) The B3 inner blinder must pick up no sooner than 50 ms after the B4 outer blinder picks up. It must take the impedance trajectory 50 ms or greater to traverse the area between the B4 and B3 blinders. fea) “107d XY HIYON ATH3H h2thl S661 @l WEN ‘LYS hh‘O6I™MC (¥) UN3N-TH3HI #261 * INHD 00°Ohh 00 02€ 00°002e 000°08 00 °Oh- 00°00S _ 00°08E 00°092 00'Oh! 000°02 0*00I- e + ° S ° s so - $ i 3 = of wo Si s a ° wo Lo} . s or (is ero On = = Se wo = =o w rr as Ez s n t 203 = = o-s ¢F a VY Oo oe ae 8 See 2 CT ow 5 Sew 94 oc o “> eu 78s ad . «Cg 8s i fe = wc —f Ee 3 a7 Oo w= id is C5 o> uw wo 2 ac =n oe se o © em ¢ hu nor & . ar eo J E c oa ® x wo ot eg c w= -Zz * o oz § Zz az os c cu - a we ac a « 2 Z sw _ 2 = 3S— w s =z Ge - 3 woe a Ww oC se ae 3 o : z ¢ $ i «, ge3| s f=2) a — “101d XH HLNOS ATN3H hethl Sb66l I YbW -500.0 -440.0 = wo ; n oS o : oO o ws - er 8 a zs o Wo s zz =o : Ww Le) ng a x« co on " Pa . ew a & = _ wn ~ = io Boe = S iv c2z Zc TF ow 5 =ee Or Oo ° = co hss aa o «FS 8 ta o wy 'aa : =n i see = Cc = QoQ n= oe o> 08 a = iO a o ee a no . ar ew a -_ qc oa 8 = wo ee: c -Zz is Oo s om (2 Zz oz Oo Ss 25 ra Ge as - Ww we ac ee « 2 aw - = 5-6 a) =z Ge i= wo = Ww oC sr ° oC 12 s ee A 3 3 S 8 sie Es “LHS 00001 000'0h hh*Ob6I™C (¥) LNUI- 1H3HI +902 # INH 0"002- 0 00°08- O*OhI- 0*092- “O2E- Figure 2-17 81-7 ONS wesEIg Yor” Avjoy de3¢-Jo-3NG 7-NAAS esnoysuyse Ay Optional Close to Block SKDU Block Override 3 Unit Tap on O-S%,, | o— 62 | AND A » Blocks SKDU Required Only based 2 - 3D Unit Trip for O-S Tripping S AND Outer At 1 1] O-S Trip o Blinder 50/0 AND } e—| 0/500 Switch 3 AND Pulse “J 4 20/0 [-~ Creuit Inner Binder -| - Trip 50 ——}_ AND 20/0 BFI Optional -y-—- 18 | 2 f cd 2)-2'(38) ' Indicate Optional — 4 95 ih Use Either Ly Ly Recioss 7 ( ) T T Blocking Non Operate te Device Number Chart Options: Legend: 21-2 - SKDU a) Restrict Trip and Reclosing Block OS - Out-of-Step 2-2 21BI 21B-0 - SDBU-2 Relay b) Trip Blocking BFI - Breaker Failure 50 - SIU 2-84 c) O-S Tripping When 21B-O Resets Initiate 95 - SRU Output Relay R 62 - Timer (SRU) Note: ¥ ~ Telephone Relay 21BI Set to Exclude Stable Swing Area. B, Bp 83 Bg \ A Symbols Example of Notation: \ det i, - NOT 20/0= 20 Ms. Operate Time 218-0 x - AND O Ms. Reset Time Se 36 4) The impedance trajectory must remain between the inner blinders for 20 ms or more after the B3 inner blinder first picks up. 5) The impedance trajectory must now leave the inner and outer blinders within 500 ms after the 20 ms constraint in 4 is satisfied. The impedance trajectory can either continue its direction of travel and pass through the B2 inner and B1 outer blinders or reverse direction and pass back through the B3 inner and B4 outer blinders. 6) The SDBU-2 will then trip 20 ms after the impedance trajectory travels outside either the B1 or B4 outer blinder. This simulation provided acceptable results for a portion of the simulation run. The first one second of data compares favorably with the DSM data. The remaining portion of the simulation (1-2 s) does not. As the out-of step trip occurred at t=.69 s I feel confident that this simulation is a reasonable representation of the system during this disturbance. It should be noted that Healy did not go out-of-step until it was separated from the system. The original cause for the apparent out-of-step trip was slow clearing of a L-L fault located near the Musk Ox 69 kV substation. The L-L fault fell into the zone 2 circles of both distance relays protecting this area of the system. Therefore a 25 cycle delay was experienced before the relays tripped the breakers. It is understood that the distance relaying has been adjusted to provide faster clearing times for similar faults. This should eliminate events like this from occurring in the future. The setting for the supervisory instantaneous over-current (50) relay on the SDBU-2 falls below the normal operating current experienced on the Intertie. This current setting (240 37 A/phase) corresponds to a three phase Intertie flow of 57 MW. This power flow is exceeded during normal operations. It is not obvious why the over-current setting is so low. It would appear that one level of the SDBU-2's trip scheme is constantly armed during normal operation. It should be determined whether the instantaneous over-current relay was originally set correctly. If the over-current relay setting for the south-looking SDBU-2 was increased to 260A/o (impedance circle radius=305 Q2) the relay would not trip in this simulation. In this case, the impedance trajectory would pass outside the over-current relay before the 20 mS constraint described previously in step 4 was satisfied. Since the over-current relay would drop-out at this point, an out-of-step trip would be aborted. If a more conservative setting was desired an over-current relay setting of 290A/o would place the impedance : trajectory outside the over-current relay impedance circle before it passed through the outer blinders. 2.6 Case 3: Stability Improvement Method The following simulations attempt to shed some light on a question raised by Mr. Steve Haagenson, GVEA Engineering Manager, regarding a situation similar to Case 1. Mr. Haagenson wondered whether it would be possible to maintain any part of the GVEA system if a North Pole generation outage occurred while Healy was on-line. Mr. Haagenson thought it might be possible to open the Intertie to the south of Healy after a North Pole trip and maintain the GVEA system to the north. A series of simulations were completed in order to begin to study this scenario. 38 2.6.1 Case 3A Simulation Description Case 3A is the base case for this series of simulations. It was used to determine if the system will go out-of-step with the Intertie connected south of Healy. The PTI-produced winter '94 base case was modified slightly to provide what was intended to be a representation of realistic winter operating conditions. The modification performed to the base case increased the power flow to FMUS. Healy is base loaded at 26 MW. Intertie flow from Healy to the north is 52 MW. North Pole GT#1 is producing 50 MW. Ft. Wainwright is exporting 3 MW to GVEA. FMUS is producing 20 MW and is tied to and importing 8 MW from GVEA. At t=0 s the prime mover for the North Pole gas turbine #1 was shut off by setting TRATE in the turbine model to zero. The unit electrical output was allowed to ramp down until t=.25 s when the generator breaker was tripped. The Intertie is opened north and south of Healy to simulate an out-of-step trip at t=1.648 s. Under-frequency load shedding is permitted. The simulation is run for 5 seconds. 2.6.2 Case 3A Simulation Results By leaving the Intertie in service, Healy goes out-of-step with both Chena 5 and Beluga 6. It appears that Healy accelerates away from the rest of the system due to the very low voltage at Healy. It is understood that when an out-of-step trip occurs, the Intertie is opened north and south of Healy. Therefore, due to the fact that Healy's rotor angle is diverging from the rotor angles of the units to the north and south, the Intertie should trip north and south of Healy (as modeled in the simulation). Under-frequency load shedding begins at t=1.804 s. 81 MW is shed by t=1.856 s. The system does not stay together and 39 in reality FMUS and Ft. Wainwright would probably trip off-line about halfway through the simulation due to under-frequency conditions caused by the loss of 52 MW. The out- of-step condition occurs primarily because of the collapsing voltage at Healy. The GVEA system does not survive. See figures 2-19 through 2-23. 2.6.3 Case 3B Simulation Description The previous base case was used. The Intertie south of Healy is tripped immediately after the North Pole unit is taken off-line at t=.25 s. The simulation is run for 5 seconds. 2.6.4 Case 3B Simulation Results By tripping the Intertie south of Healy immediately, Healy stays in step with FMUS's Chena 5 unit. Under-frequency load shedding begins at t=.48 s and has shed approximately 82 MW by t=.56 s. Voltages throughout the GVEA system drop to .9 pu and then rise to levels greater then 1.2 pu as a result of load shedding. The capacitor bank at the Dawson Road substation switches out at t=4.28 s. System voltages stabilize between 1.0 and 1.05 pu. Both the Healy unit and the Chena 5 unit experience operating speeds below 57.5 Hz. These units may be tripped by under-frequency relays. The Healy unit and the Chena 5 unit do swing together. Therefore the system does stay in step in this simulation but a majority of GVEA customers are out of service. If Healy and Chena 5 stay on-line system restoration appears to be fairly straightforward. See figures 2-24 through 2-28. 0 oP BRADLEY © 90MW 1994 WINTER NORMAL. SSIINANDAYS SNE ¥ AINIHIUW Ee'sl S66l S2 ddd ‘3NL 000S*h (SON093S) 000s" . 000s "2 SWIL 000s"! 0000S"0 0000°2 o o co co os |o |o |s6 o jw fo [oe rn ec n rc wo [lw [wo lw + ° - e ‘ : | " ! " ‘ | re ‘ : | ° + |e fb o s ° z rc) = =| oO] = 7 3 2 & aoe oA i a A oo a} 3 a | Z| o = _ wo} eS a = < om = | ce 7 oO J, ¢ - wo 2 o od a 2 & w> = S| eo el o| i) 3 3 oF a c i} a rz ao fF Y og co uy Sa i x eee 7 | So SI row S| -| 0 Zits yw a E —r ol oe sat w “ -~ oO a al 3 S i * -a “| a se zz 27 6H CG OS z] 3} Zz 2 os = o Ss w> av oz aw S ra mw co ou 2 « aw -O cz eS 2 ° 2 °o So £ Ss |e |S |s 3 wn w wo wo if: pia|eie |e 3] o {jo |o Jo Is==) cs =F Figure 2-19 = SABI TOA WALSAS Y3AD : a ( ) 3W he'Sl S661 S2 Ydb *3NL SONOI3S) 3WIL 000S*h 000s*€ 000S‘2 000s‘t 0000S°0 . 0000'S 0000°h 0000°E 00002 0000°I o'o eo lo jo fo s |s |e |{s es |s {Ss |{s s |S |s |[s a Ia [a a so lo fo fo + e v o ‘ : | oO ' | ©) ‘ = : | " ' | E i | e + o 4 o ad z 2 | a Se ee Ss 7 8 c| a af o q = oS = S| om om nN or 9 ce | x 7 o oC ™ wl z > wr = oO} - a =) Ss 2] al S| @ 2 -Je uJ =| oO wW _ Kz Npou} of £ oy - £o yo oo oS} a ww | | «| a a -o wy | fat) -| ao zu al | | | a —= uw x wn |S} SS o : Zz Z| Z Zz i -O ~*~ = a xr ' - ne o a a oO ‘ 3 FO : a Ou ' a ' = w> : oO a O = oz ' aw . « > ‘ Ww zo ‘ Ke eu : = acc : - Ou > =z = ' a o =~ aw ; cr -O ‘ oc az = eS “TTé eo jo fo fo 2 6 |S |e |S 3 o ° ° ° je} uy fw ju ta Fie =— te — lool nN * -39N3H34410 3T9NY YOLOY ISthl S66l Bl YYW “1S See eee 000S*h 000S°E 000S°2 000s"! 0000S °0 42000'S ¢ 0000°2 0000"! So 19 iN Pesan es 2 |o |o ° v wo ; I he | ie w | i SF lh G it aly | ° a | o Ser alll ipo ad q a = (1) ike) 2 HE | i) o aw a] w a oo ag a ea et ey | ay oa oa Sa een es = ac | 4 a co eas Bs =| = o o = i ag es oe e -LYEe be < > > >a |||| at di) ee ||| ea uw ww Ww x uJ ual a! eo oO} x) xr |) tot Pea acs | Un te < ed, + | a wn a A] Cages Se -a ses| a ae | wl ol Be pa || | 6 = c w> pl | eee | ssa tes is A . et ee Paste mien ou =| 3] 2 ca eS a a gq oo eu Sa ccezm9 001000000 UTTER TVET emg TT epee eee er TIT TR eer eure eee LU MM AeA Oa Lea get = Ou ae IN ee a z= aw OF || =o) yl) a Se em NIM TAS TI ARE Pees ee URES UML LeU Ue ee Oeste LU SUSU ae ee UL Da ee ie oe ae ae f eerie z. esi leet. Pied o o o | i | EES -—-- eo +a (I -@--— ood Figure 2-21 on * 107d XH HLYON ATU3H ISth! S66l SI YOW “LYS hh O6I*C (¥) UNIN- 1H3HI feb *# INHD 00°Ohh 00 °02E 00°002e 000°08 00°Oh- . 00°00S 00'08E 00092 00°OhI 000°02 0"00I- S 7 3 | | | | | Ss 1 ' ‘ ‘ ' ' ' > ' ' ‘ ‘ ‘ ' ' e ‘ ' ‘ ‘ ‘ ‘ ' 7 ee woe : a ene Ngee ee foe eee | oa eed aceasta sé pe lee specie ep eR ep ices — oP > ‘ ‘ ‘ o s ‘ , o Le ‘ ' 0 o | foneeeece dewi me meo si lis eas saat ect mee den ie PMR, ome ben ff, meh a Pom Neotomn nnn A = ° ‘ : 4 = ns ‘ ‘ " : ‘ ‘ - o ‘ ' 2 ‘ ‘ . 27 8 I ena Nr) vere eee © eee es fer wo z.efu0T7 . ee ee eye ~~ wo = ' ' 2 =o ' ; a ' ' a ori es ‘ ‘ w or ' ‘ a zs ts ‘ Se Se ee ey eee a pee eee, ae hee _ a — x ° , a ad z= x 2 £ ‘ ' a By ‘ ‘ = a GO ' ' Sor o =e ‘ 7 os zCrw ‘ i tu > a Gs = ' ' e = . © oe les Nee Sees =e ope eee era ee ae ee ee ee | eee ez ou eg ’ + > co a 7? = , : wow w Fw : ‘ = -o a x ; a Oo 7zw — + wits ' ‘ = a uo ga 3 ; ! ac o fo. _———- fencocees Sh coe ees ces aye tee eee ef eh a ee ae o - Hem 8g ‘ : -o noo 8 1 ' + #@ (san a ' ' a Gy ' ' a wa 8 ‘ : = osg ' ' ac lu > Zz _ = oe Peace Re GG 8 es 5 ek Ws Aa dt ee oO aU or ge ' ' z2 os on ' 1 aw ” f s oc > - i 1 Ww ro ac ' ' er FW « S : : Zz aa - alo some se a bees eet’ «ry © Bins eae Sienna irs ears mueller cocses a realm = mete meer = Ou w S ‘ ' = 2% . m ' ‘ « : : sc =o ‘ c az S ‘ ‘ = FD a5 ‘ shes eee cei ey emt c twee ye ae 3 ‘ ‘ Syne > : : 2 + : ‘ i 8 { f ic Fe, 5 ge3| 8 | e 2-22 Fi a * §MO74 YaMOd 3NI7I TO?ST Sbbt 8f YW *1HS oa epee 000s"! 000080 aes 0Q00°2 00001 0°0 o jo Jo flo S 4 : © . oO 2 ' BS ' 7 ‘ ‘ ‘ ‘ ni ‘ ‘ t e v i wo ‘ : | © ‘ oO ‘ | " ' | ‘ K ‘ | ‘ ° ioe tt " © 4 ear ee en ea eee one ie a ee ieee TT acai ie s ' z ' 3 yogoq 2 a ye a & wee fet Spf a sn a a = a rhe IN : / ' = x e = 2 & Vb - Ss. 8 woo : (A 2 s 2 a 2 yo ‘ w> uo Gt oa ' @ te 2 i =| S| a ‘ ase g of WY Gl iz - £0 5 4 Tl a wow tay ol | uw “ 2a bo o o - QQ z2u> ce] oa} 7] oa = > a wo ® “| . - . all -— e a =| Fd =| Z| a 2 xq 7 = / 8 x oO o| oO 3 FO oO a oO. = ttt ete to aes meee re ie Oo au Z o2 aw ac > WJ _—o e Fw a ee ltl lata Sc todealctcl ll ont ow = = ac ad aw o =O moor £ eo |o fo fo S 5 |e |e [s 3 > |e q <8, Oo * o ° sie a lo IR IA ers ers et + S3IINSNOIJY4 SNE ¥ INIHIUW 82'SI S6b6l S2 Hd *3NL pSONGO SS) Saul 000S"h 000S"E 000S ‘2 000s"! 0000S ‘°0 0000'S 0000°€ 0000°2 a o lo fo fo os |s |s Js a no [wo [wo lw “a ete fe fe ra w fw fw lw w oS x + ° v e ‘ 3 | 5 ' | o ‘ a : | w ‘ | = ' E | V7 + ° 4 = Zi = a 6 a3 og ff 8 + a “| 4 «a = ma fa + a E| <2 a > ae a a el a zs | a a = > = oy a 5 & o ow 7 wo 2 a Cw 2 oO} a a = e>+Ss => oF Fo e@ aqic a UW wy a = =z ow yg iS ~ a0~. uo oo 7 Www w x ot 6 te Clk $I | Sf tg EEG H et ew ewepe epee ewe ao Fu wo 2 al cg —-xru “|- a 2] - c oO © Fo) et eer pe eet peerless o . ul aj = ey 3 2 . fe - 8a Zz 27 2 ~3 ay FO x ma z S ou = Oo SS ee ee ele en er acc ore a ca te oO a z oz aw a _ lu zo e Fu er i cl cere ccc secre eer ose meen meow - Ou = 2 a = aw oc -~—-o Cc a2 = FD = 2 lo |o fo : eo |s |s fo 3 B IR IR IB $2 sn lu [au tw 223] © fo jo fo | | ‘Oo vt SADYLIOA WALSAS Y5AD 62'SI Sob! S2 Hdd *3NL (SQN0I3S) 000S*h 000S"E 000S°2 0000'S : AW lt 000s‘! 0000S"0 0000°2 0000°1 co |o fo fo eo |6 Je |S eo |e |s {8 wn So So oS o =j s |x fa iw x 3 lo |o fo we ° z _ir - ' : | = ' | 9 ' “0 ' | a ' | = ' : | b ' ; oc + é © + = Zz a oI a. ian) ia) = s} s} S| Ss = i as c| oa @ = S| om} mom = al a | -| =| So a m oc = wz > Cw 2 ol = 2 I i) z} al| os} @ 2 4 J] =| oO ire) —2 Np oa} Gof) OF Ben oy oF of wow uw | ot a 2 J zou wl ~| co) ua oO Fu- mm ost ial & = ® « ® ® =f a . Z| Z| z z =o rl x (FOU ie oy og a 3 FO a ou = a lu > Oo av z oz aw oc = ww z= Ee Fw =z acc _ Ou = 2A ac a aw oa ~—oO oe cZz = - 2S = co lo lo fo = s |e jo |s 3 Se |o jo |o eg, Ss Ie [a fw WES 5 Figure 2-25 6 . SINSYIS4IG JIINY YOLOY (SONOI3S) SWI 60?SI Sb6l BI YbUW “1US SD 000S*h 000S°2 000s*I . 000050 . : 0000°E z pnoo"s Sieus cs |o Is t ay o © © t a et) ied ies rz . . A we =) x e ly z eal a | oO a | w q oid ze aq: a, ¢q « 75 5 — + | re « « = o al oO 2 z) 2 a wy a wy z i \o La a YO ao} a ul a ul NQ 7 = 7 7 7 7 E aA oA a = = & ¢q| « < o So =i ~m _ 7 — cw 2 a) |e e aque = a5 3 3] 3 im = Qo 2 ~| wo ww i oy ss Sn i Oo rt Es oO zw ut ul ul < —Dw | + | ac a uy oO} uy o - a -a =| el) el . se - 7 7 5 Bio eeta g qos Pl eke rics e w> eS A So ae ‘i %e . Zz om 3 a a aw Z| a 2 fe mal Fal w x - = e ew Z aa = Su = a7 a a aw cs =O oC “Zz - 5 Z 2 |e ls 3 ° 3 is i so |o Io sie! s |3 |s 0 vt 1OId X¥ HLYON AIU3H hh*O6I™C (y) UNJN- 1H3HJ +261 # INHD 00 02E 00°002 000°08 00‘Oht 000°02 3 | so s ‘ o = ' ft : ae Ss w es ° 7 x & ° = ~ S ° wo = a = s = s w . - a b © a -s ¢ Ww ear FS Lee... e Zz: ¢ z wo — zo rr a ar as = on iva we eo 3 gaz s = fe See oO J z a os w 53 aru ys > 2 e au = oss -Zz , — = eo wus .c & uw Wa J aww — =o ea Ts Oo =w i ee el a —=x we Wg °o ra 8 c o oO. ( oO ° -m 3 -a noon n . aa - J FO Gq oe ® = eS iS a > Zz ee oO yu or 8 z= OZ On aw ,. va = H w ra « e FW a = Z aa e ea = ou wo s = a) - g iva = aw a —O oc cz iS = lets R e = z 2 + 8 8g fs 3 teed . — —. Figure 2-27 a * smo14 Y3MOd 3NI7 60SI Sool Bl YUH “1yS (SOND) BS 1S Hitt 000s "2 0000S"0 0900°2 e fo fo | | I > sy : ' ' ' wo =, es a ‘ ' ' = ~ ‘ ' ‘ oa , 7 t uw 7 ' S ' “ fi : : ‘ i ‘ | ‘ =} ' | es ‘ a : I w ‘ | = ‘ : | be : : t a + CF 6 (I zs a yoo aoa a ey al os} oe 0 c 2) =| a =| Q = 7 a) of | x = 2 | « XN Pre uy Ss 8 oA gE 0 2 D 7 oo. w= a 2 a a) wl ol al 2 eo a3e y eS x om a0 « o a ul = wow w og = a zs a2 ZO el Oe oO fw a Z| rd q =ru ee a wo | «| "| oe | | aes 3 ZZ 2 Zi al + ad ro OG a a -OoO = Zo = c lu > Oo 416 2 Oz aw a 2 la re e rw Zz aa = ou =z 27 « Sa aw cs -a oO iz =_ 2 é o lo fo = Ss |S |s 3 : |S é s 2 Io <3. a |o |S 323 = fo [= 50 2.6.5 Case 3C Simulation Description The simulation conditions for this case were the same as case B except: the Intertie south of Healy remains connected and 18.1 MW+? of load is shed 10 cycles (at t=.42 s) after the North Pole unit is tripped at t=.25 s. The simulation is run for 5 seconds. 2.6.6 Case 3C Simulation Results By leaving the Intertie in service and waiting 10 cycles to shed 18.1 MW after the North Pole unit trip, Healy stays in step with Chena 5 and the Beluga 6 unit. No additional load shedding takes place. Voltages throughout the GVEA system drop to about .9 pu, oscillate slightly and then stabilize approximately .025 pu below the initial voltage in the worst case. The GVEA system does stay together and only 18.1 MW of load are shed. This method of immediately shedding load appears to be superior to tripping the Intertie south of Healy and allowing the system to ride out the disturbance. See figures 2-29 through 2-33. 2 The load was shed at three substations. 6.2 MW/1.9 MVAR of load were shed at the University Avenue substation. 5.8 MW/1.9 MVAR of load were shed at the Peger Rd. substation. 6.1 MW/2.0 MVAR of load were shed at the Hamilton Acres substation. iil SSIININOIYS SNG ¥ JNIHIUW 2€'SI S661 S2 Hdd “INL (SONOI39S) 3WIL 000S*h 000S "Ee 000s "2 000s"I 0000S°0 0000°S 0000°h 0000°€ 0000°2 0000°1 oo ce lo |o fo os |o |e |so ow fo iw fa cote de dre w fw fo jo * ? 7 ‘ : | o mie wo ‘ i | Or 1 : | > : : | oe + o 4 no 2. ) ‘oI aa Sig - of . + Oo} 2 nm wZz FM)||||eu ay wg | ow Ss a Hl eee vier ok: gl Q = cw oF |\o5 J nN = Weuw bh ao ul « So F§Lm nt | z a o TF Pea Z 2g a = =| a) Tilia ca 4 2 wo aa | wo . or) wu a x uw ou || oe at ae a = x I Ww —Ouw x 3 2 a | tt ae So «| o Qo z20— 7 ot sl a ww we =| all >" | c ° o al oe ||| = = : + 1 fet gat Se 3 Z| 2 i || es as lll =) tz ||| oO a| = = = o ac li Oo Jo z o=- a iv a ae sta —~ Fa 2 || ce = 93 =z Zr vo = aw o —c Cc as a eye 2 o jo fo lo 2 co |o |s lo 3 on jo [wo lw ie us lu iu tw 353] o jo |o fo nN w BRADLEY © 390MW. 1994 WINTER NORMAL. SSIBLIOA WSLSAS U3A9 ce-Sl £ IN SERVICE CLES AFTER N.P. GT#1 TRIP NTTRE INTERTI Cy us TRIP 18.1 MW 10 { TRIP NORTH POLE GTe#l. ' OTRECT VE cowac 06165. ince Famer RUN3F HUE: Figure 2-30 S661 S2 Had ‘nL (SONOI3S) 3WIL 000S*h 000s "€ 0005 ‘2 000s"! 0000S°0 0000°S 0000°h 0000°€ 0000°2 0000°I 0°0 ° ° ° ° ] Seidl itic ! ‘ SH Sil siits So o So o . § Pope HEE ILE a em ites |i tes ' i i ' + e v i : A H | ‘ ‘ ‘ [—------ Nae weer © mee Scene min = Daa ' : | i ‘ SE | ‘ ‘ an Sav t H dada oa : >| >| >| S| AEE ' ' SE BAS Le of wo @) ©) ' ' Se NP | Ne ‘ ‘ dH) eta || ' ' || Nee a ' ! Sle eS ' t Zz q | a , ' ve =| oO} ive) mes So. ane! fn oe om ame all mae oe Sai eel ete ' i i) ul ul to f i ‘ eat ||) || | ges ' ' ' Se Meet IPA |e : : ' aI J aj | }— ------ Non eee eee dee ee eee de------- . Z| Zz Zz z c : . x zy x a ‘ ‘ ' jean) | | eo) ||| coal) || |e ‘ 7 ‘ MN ean TTT TIN TIT . ' ‘ ‘ c>al|| [ext tle) Ne f t : f ; SH Slesiiiis | ( ' H ' Sa Sse ; j ; , ‘ |e AL ae i f ' i i oO i JINAYASIIO JTINY YOLOY 61'S S66l BI YYW “1uS AS ONO IS ant 000S*h 000S "2 000s"1 0000S °0 0000'S 0000°h 0000°2 0000°1 00 cor ail PS a je ic a |o |o 7 : : 9 y i | a : wo ; | ‘ Sn ol: q 4 ts ’ > qi a, a ° ' ze Se Suu ced Tc ; ' re (uuliee tedeas no ee , = a a ' ; ~ 7 G3 a & ‘ ‘ , roe Srl ‘ ‘ ah : ps NE a ' ' wz a oo ' ' = = AnH outs Lee cnn Ma cA Tiling Ay oA oA t : ' S| aes Ein a si 1 ' N = weu a a a ‘ 1 ‘3 S fun = = ‘ ‘ Chie ice: ' ' r= ale ‘ ‘ ° oe ef ‘ ‘! wu a > = i . ix bh z ‘ a fsa ie: ¢ Se AS ‘ mS ell es ad o r I ‘ |e es Fool ea ' ee Oe ' s ei SE Ti ' o - ica Mies ' i po ee ee ' PTD ea ee a : Ne oe of of a ‘ co ST ' = - ‘ ac les wo wo a) f |) oo fe ‘" ‘w ‘ z o- a | | ' ire pe | ler ' Diets Ea l= ln ' — eo ‘ z ar =o = z2e oO = aus o ” . | i estes 3 Feet eel e3, s |o |o Fe] > |o |o ee 3 ee eet TTT wT y ||| ork BLOW Aa BttSI S661 Gt BUN “LYS hh‘ O6(™C (4) UNIN- 1U3HI #261 * INHO 00 "Ohh 00 ‘02€ 00002 000°08 00 °Oh- 00°00S 00*0eE 00°092 00'Oh! 000°02 o'00I- o - s s 2 3 3 z s 6 a o bs ea ad o wo ° * OF 3 = °o >= ° aes ~s 8 we 9) Sle Sia =: 8 zal 0% Sal wo) = : rane we Be yl | ee Pes lilie Se =|) Gos mites = = N = | ee eae g > e|| ice Seay M 2 a Zz ° wo et The = eae a oe ao 25 aa || a E> eS le = | ae eel c ° ey ellie o -— em s = oo 2 al aie fe es a i 2 ao 8 ew: is @ 2300 oo 3 Zz Oo- Gs a - ac a - Ww -— ac - eo « 2 Zz a - rs z ha = i te 8 vo zs Qqaw Cc =a Cc ae = a) 2 Ta S Z ; . s See g ees a nw “ §MO14 Y3MOd INIT 61:Sl Sb6l Bl YYW “LYS (SONOI3S) 3WIL 000S*h 000S'E 000S°2 000s‘T 0000S°0 0000°S 0000°h 0000°E 00001 0°0 ° le |e |e 7 «|e 2 é 3 © lo jo ‘ ' o ' ' Nu ' t ° v ‘ : | a ‘ PN | cc a ee ce ee ee eee wo ‘ j | OF ' ; | ns ' | ean ‘ : cn i wre + o 4 nO ze Sa, pl ° , ay oa oo w2 5 a =| a a = my S| a) - see a 2 al | ' =z mw A a 2| c N = Weuw S bum i a oe o C*1.2eaZz wn z ‘ . Se wg] 2 2 wr wl . yg Oo o Wu x > Gia a 4 4 Y I CS — seo eh o 4 a al Jory x o 7 Oo 2U-- e—_ 21S Pe aqw iu OM y ae i ” wz 47 2 3 2 ; 8 = o| | oO is o qo = - c ws Oo jo 2 o- a a a w == —_ ea z cu =o : =z Ze ‘ : te] . ' zs aw f Cc a ‘ : c «oc ‘ ‘ - Fa ‘ “tf z eee leeelare le ' 3 % lo — |S ' ' : : ‘ ' bri s |e Js |s ' ' ' : : ' : Pe = lee _ ee eee ee LL 56 2.6.7 Case 3D Simulation Description The simulation conditions for this case were the same as case B except: the Intertie south of Healy remains connected and 6.2 MW3 of load is shed 10 cycles (at t=.42 s) after the North Pole unit is tripped at t=.25 s._ The simulation is run for 5 seconds. 2.6.8 Case 3D Simulation Results By leaving the Intertie in service and shedding 6.2 MW 10 cycles after the North Pole unit trip, Healy stays in step with Chena 5 and Beluga 6. No additional load shedding takes place. The voltage plot shows two large voltage sags to .65 and .77 pu respectively. The system does stay together and only 6.2 MW is shed. See figures 2-34 through 2-38. 2.6.9 Case 3E Simulation Description This simulation was an attempt to determine system stability after the North Pole unit trip and load shed events of case C. The simulation conditions for this case were the same as case D except: a three phase fault is applied at the Musk Ox 69 kV tap at 5 s. The fault is cleared 5 cycles later by opening the Gold Hill/Musk Ox tap/Zehnder 69 kV line. The simulation is run for 10 seconds. 3 The load (6.2 MW/1.9 MVAR) was shed at the University Avenue substation. i wn SSIININOIYS SNG ¥ INIHIYW (SONOI3S) 3WIL BE SI 4° S661 Se Hd ant nancen o00se ov0s-2 0005"! 0000S‘0 0000'S e 0000°€ 0000°2 e |e |e le | cs |o |e |s 5 ‘ wo io jw |e ; ' ie le |e fe ' : wo lw fw |i + ° 7 : i | ' | a ' | we ‘ | oa a” Se ‘ ' | = ‘ : es : z - i oie it or re] z J 4 3 3 acti fo ol Od Ww: 1 oe ul So wt —Zz re « | "A sk = 2 | > ' =z aa go YF © — ‘ N = wwr a J = ‘ o o e-em w Z| r ; Tc Zuz a 9 a & ' z, =a5 | 4 a oF tos ° « aga Ho = wo a x uy YQ ' Be > Cw. ul = a x ' w =34wW x rs) ' 43 FOU oi + 5 ' QO 2z>= a + a ‘ qG wou + al 2 S t c @ a ee ee ee ee a a | — o -o ‘ _ J ‘ — J 3 * . ' + 8 Zz] 27 ZW 3 ' jwor=z St = Zl ' aq oF a o a ; = ' c ip i ee ee ee en a ns a ee ee ee ee ee — oO as n Zz -o.0 : a ' c a ‘ w xr ' — Fa ' ee he celts cre Snel cree 3 Mie ge cima mets ane Sees Pe meee were matin wan ofr am ome aoe ml = 0 ' = 2 ‘ 2 ' 7 ow ' o —c ' co cn ' = FQ | ff pe eee bee eb te ee ee ee eee eho eee eee = == ‘Ts 2 jo fo fo ' = Ss |s |e |6 ' 3 wo wo wo wo t sie) a fas dae fos REY > [8 |S |S __| 0 wo BRADLEY © SO0MW. INTER NORMAL. 1994 W na | rower TECHmOL GIES. | ince SAD¥110A WALSAS USAID 6EFsl a Ly or > a- ws Or oO z a we —Zz ‘= oo Www i Zu -a wo cw =o ko z> Wo “2 8 r= Oz we ae oo a ao me eo « 3 7+ oO aw =o c— FO RUN3H TRG: : (SONOI39S) 3WIL Sb6l S2 HdH ‘INL 000S‘*h 000s "Ee 000s "2 000s‘! 0000Ss‘0 0000'S 0000°h 0000°€ 0000°2 0000°! 0°O eo lo fo |o ] s |s |8 |§ ! s js {8 |s ' s le |s {s ' Ss If Is If so jo Jo fo + ? v d | ' | ‘ | ‘ \ ye TT + fe |b alooAlo ala ‘ on] nid “ = ' wy Ss : on ' : S om OF A ‘ nN e| =) =| 7 ‘ g | | 7 | a 2] o 3 ' = | =a aq ‘ a BBL BP penned netfee z ao od 8 o ‘ os wo oa oat ‘ a oe ! . o a o ' aj 3 Ss} S| EL .----- fe ceeds Zz z Zz z ' =| a bs =r . ao 2 a 3 ‘ 2 le le le i sis |s {s : s |e |s {8 i a [8 |& 8 i a w JINSY3IIIG JIINY YOLOY T€?St S66 Bl YYW ‘1uS SONOS AS! SAU 000S*h 000s ‘Ee 000S°2 0000S"0 . 0000'S 0000°h 0000°€ 0000°2 000°! 0°O fe |e |e | so |o |o o |o |o . . . e v | | a | we ou qt ie S =|) « Ws nor _ wl a oO « | = oF a oOo Ss, >s 2, 5 a! Je a we ao of of ‘© —Zz aI y RO teeter on. = ca 3 2 By a = Ww qa ¢ o So ERM “| =| = CS 2uz -a5 = e c 2 Pte ee ete eee eee . ww q >| >| ’ in > Cw. Zz a a ' : WwW =sW 4 sg S teawt_a~ : eh oO = x _ , Oo 2z>=— o oo Go : \LaA aq wow 4 ; Zz - cs ae Oo -o i ee ‘ 1 < & ; a ' ‘ z@ bE a8 S 3 ' ‘ ec owes al i iteeee mast —— “ o as . ' Zz oo =! 3 3 ' ' 2 ' ‘ « oe = =| =| ‘ ' fe eas ao of ST ' ' — Fa : : Zz ar _ Pore 3 cts cas s = 90 : =z Ze ‘ Oo : > aw ; sc —c ' Cc c= ' _ ro — . 2 eo jo jo ' : oe |o |s ‘ Z. So So So ' is 2 | I ] So \o LO1d X¥ HLYON ATYSH I€'ST S661 Ql YUN ‘LYS hh O6I™*C (Y) BNIN- 13H] fel # INHI 00°Ohh 00"0z€ 00 "002 000 "08 00Oh- 0000S 00"08e 00°092 00°Oht 000 "02 0"001- ° os Ss ' = ‘ e ' s z ~ 8 o : a _ o be —— ae So = 8 = or Ws ear 8 Or zs Oo wo z =o Sara id= a a x we: on iS) cc Zz za . om = ca nx a x= Oo = Ss ee om ofS o CF Zu Ss crew _ epee es) 7 2 mo ow «PDs oy on wore * 4 KO 2 xr Oo 2> a oe LD qc wo ee ios 3 a iO! | [eeerergereorsr ao “oO - mM 3 —= oo & a ie po | = = ize iS = on cows ir Oo 4. ors Zz 30 os a 7 oc a - wo xcs Cs He Fae a 2 Z2 cr he ; = 6 Oo S =z Ze B = vo z= aw Cc =a cr a; S —- Fa : 3 anu F & 7 ; s ic ze. 5 sey s ee a ~ \o BRADLEY © SOMW. 1994 WINTER NORMAL. 1 frown TEcwmaL oci€s.| ine SMO14 YSMOd JNI1 Test a we on =F > = We Or oO = a Li os —z = oo tu Er Zu —a wo cw —4 HO z> wo “oO ® =z OF we ay oo a a xr eo ar So Ze vo aw =o cn Ka RUN3H FIEe: SbbI 8! p (SONOI3S) 3WIL YBW LYS ONO é 000S‘h o00s'e 000s*2 000s‘t 0000S"0 0000°2 0000°1 BVI AE | | o . o : BS ' iS ' ' ‘ t ° v ‘ : i ‘ | ' ; | ' : | oe bole oft y Joga of y a a e a 2 a | a neal uo GZ S| | e = & 4 of wt Gg yz oo OG oO ees . 4) Sl) el ball i ia ~ . | J 2 F z} zy 6 CF x vt} S oO Go 2 |o fo & {8 |8 s ae | s |% |o R Io |8 Sas Figure 2-38 62 2.6.10 Case 3E Simulation Results By leaving the Intertie in service, waiting 10 cycles to shed 18.1 MW after the North Pole unit trip, and then applying and clearing a three-phase fault at the Musk Ox 69 kV tap, Healy stays in step with Chena 5 and the Beluga 6 unit. No additional load shedding takes place. This simulation would seem to point out that the GVEA system is somewhat robust after its operating point has changed due to the North Pole trip/18.1 MW load shed. See figures 2-39 through 2-43. A similar case was run with a 3 fault applied at the South Fairbanks 69 kV substation. The fault is cleared by opening the Ft. Wainwright to South Fairbanks and Peger Road to South Fairbanks 69 kV lines. The GVEA system recovered from this disturbance without difficulty. 2.7 Case 3 Summaries Case 3A determines that the GVEA system will go out-of-step due to the sudden loss of 50 MW of generation in the Fairbanks area. Healy goes out-of-step but not in the classical sense. The voltage is so low at the generator bus that the Healy plant can not deliver any power and simply accelerates away from the systems connected to the north and south. Figure 2-22 shows impedance trajectory information at the Healy 138 kV bus looking north. While the information on this plot is difficult to decipher, close examination shows that the SDBU-2 at Healy should trip on out-of-step (as modeled in the simulation). The trip should occur on the first swing through the right set of blinders. The impedance trajectory starts at Z=365-j65 Q. oa © — S3IININDAYS SN@ ¥ INIHIUW ZhtSI So6l S2 Hdd *3NL PSONG 25) =ait el 0000°6 0000 0000°S 0000°€ 000071 000°O! 0000°@ 0000°9 0000°h 0000°2 0°0 ec lo lo s ls Is |g | | | | ae |e ' ' ' ‘ ' ‘ SS ee ee : i { ‘ i t oC |e |e ‘ ' ' : : ‘ : ‘ ‘ : . ° v ' i | ' f | : : | ww > : ow ‘ : | =" ||| Se ee « ae + + wo o aes o| | a =a 3a 0 e} of ot} w> ay oO ire) S| a =x c gq =I =| om . - cs Z| Oo} > t x ao | a = =| N x wmy om a ; q| oO F¥-m 7 wy z| z o he ee 2 Py = =| i) => & a 2 za = UI wv - = we a x uo Yg ix = ac - ul 2 x ae Ww —ew x 2 ol ° = a | oc «| oO} Oo 25-— 1 wy - zl a waw +] oo * —| « vas of Y a o oO - a 5 a ar a 3 2 3 4 —| ic z oa z co oO} a x = = s ac we Oo 30 z o-= a c a. i — Fa z ar = sa Zz Ze vo Zz aw o -—c o ce - = Fa ae ee ee ee “TT 2 jo fo feo : ' = so |o |e fo ' 3 ow lo [wo jw ' ° : ‘ ' sie) ola fa ew Ss 2 o jo © | | % SSDBLIOA WSLSAS BSIAI Eh?Sl S6b6l S2@ Yd' “INL (SONOI3S) 3WIL 0000°6 0000°% 0000'S 0000°€ 0000°1 90000! 0000's 0000°9 0000°h 0000°2 0'0 e |e T s |s js |s oe [so |e [5s os |o |o {56 a ia [uo fa 3 lo fo lo ' ' a7 v { : | ‘ : 1 ' : I — Ss ae Se * ww : u ow ; | ‘ = : | ‘ Si ; ‘ oc iS di ' RN eee 4 a ' ‘ Zz 4 —Om 5 ‘ ool al on ‘ wu> = es = = ' So , St St eee eee eee 7 vt = £2 S| om om om ‘ N SoS H-—m « a x} 7] ‘ wy > ‘ 7 205 gS a FS a : ° =o co Ss a Gl uh ¢ wi Ni uf oof & + I ~ « o oo oo OG : Woe uw a +4 =| a ' ‘ a taal wv oof af ' ‘ Qo 25- om} om} om om ' ' a wrew ' ' cu Pe ee ee deneeeeee 1 oO - Zz CUZ] : , — ee.) xr x =x =x a 2 . we o oF of Go ' i J rs ‘ ‘ a Oo ' ' ss - ' ‘ i rs Oe i ee een cere er quae seme 5 : Oo 30 : ‘ Zz o- : ‘ a A : iat a ‘ wu cs a> y H— FO Swot Z2 cr = o . = a ' 15 ‘ + aw f co =a ‘ Cc a; ‘ as mR 838 fF ob fb Kemnciates eres ee eumcsaie aeceemuneres { i : ‘ é eo lo |o fo ‘ 3 oe |e |s fo ‘ 3 ° o ° ° 1 ce N N Nu Nn ‘ by level al w \o SINSY¥3I4IG JT9NY YHOLOY ehist ww ow =u >F « we wo w ZS) =—o w> = oor x «ow = wm So F- oo = = ° = wu >- « —-= ls Oo 22 a wa a w o - + ez J rE qo = = a ls o ajo z o- a iv a w xr — Fa Zz ae = 0 xX Zr “vu a aw Cc =a Cc a- = §Oo Uw eee! x | TEOMmOL OG IES. ince RUN3J FILEs Figure 2-41 ceeimo mun osT be (SONOI3S) 3WI1 0000°6 0000°¢ 0000°S 0000°E 0000°1 000°Or 0000°8 0000°9 0000°h 0000°2 o'o © lo |e ] s ls |s \ a |o |e ' te ie te : ely ; ; I ‘ I ‘ } ‘ : Fo) 4 \ a a) al] P_% so a ' f 0} Bae eee ieee Se Se a } ! | a} ' ' SJ a ' ' > z a ' ' yl wi ) ' ' a ¥ 2 ‘ { wo Uv! oo) ‘ ' uy uo ee oe Ay oAl od ‘ 1 eg a ¢ | : Ae ; i ce >| >| : waclec ee eeee a al a ' > =x ire) ww) rr nm ' Fs ‘ ; o 3 3 : er ‘ a 94S ' ; | ol ol ' ‘ a oa ' ‘ go a iL Nee cee eee ee oeee sees ‘ po i ‘ ‘ Zz z Z| s ‘ x x x ‘ ‘ a 4 3 ‘ : e |e |o : s |s {|s ; s ls |e : = |3 |8 ‘Oo ‘© 101d Xu HLYON A183H hh O6[™*C (Y) HNJN- 103H) Eh?SI Sb66l Bl YUW “LYS 00°OhK 00‘oze 00002 00°00S 00 “0ge 00 ‘092 0"00I- -100.0 -40.00 =5S 0.0167 20.000 80.000 INCREMENT: NENA (X) 7x190.44 140.00 INTERTIE IN SERVICE 138 KV BUS © T ENTIRE FAULT FWS FILE CIRCOS TIC RUN3J Figure 2-42 0 CHEAL 200.00 RELAY: 10. TSTOP: 00 BRADLEY © SO0MW. MW, 1933s 260. 1 0 18. 0. CHNL& 320.00 TSTART: 1994 WINTER NORMAL TRIP NORTH POLE GTel, 380.00 } DIRECT TRIP 440.00 rower TECHNOL OG ES, ince $00.00 6~ \o SMO14 Y3MOd JNI1 EhFSt Sbbl Bl HUW “1S Nee 0000°6 0000°¢ 0000°S 0000°€ 0000! 0000°8 0000°9 0000°h © |e fe fe Ss |e |e fe a ’ + |e |v : : | ‘ ; | wo ‘ i | ow ‘ : | =n : : | — i ; i c i i we ha o wo zo So = yy} q q x) = cs me y a al :& | el | z £o dj ai a See | Q-2 2 a 2 ye -ns | wy a z ° ex) a oF w $I wi oS) | -~ cc « oI o eh lon EA lw ad ot | of of & Oo 25> sa aq wou yO a c w . = . oO 3 . 2 z a] zs ga 53 + ex =| o 2 oO J rE 5 iz 16 = = c We oOo jo Zz o- a ac a w xc — ea Z ae = 9 = 2e vo > aw Co —oc toa m-- = a 2 eo lo fo fo ; s js js js i s 2 |o lo Ise, au jo [a low ih sf EE Figure 2-43 68 Case 3B may answer the question posed by Mr. Haagenson. The GVEA system will stay together if the Intertie south of Healy is disconnected. That is, it will not go out-of-step. However, the frequency at both the Healy and FMUS units drops below 57.5 Hz (figure 2-24). Both units may trip on under-frequency protective relaying. Approximately 80 MW of load shedding occurs by the end of the simulation. Healy and Chena 5 do swing together (figure 2-26). If both survived the under-frequency excursion some generation would remain on-line. Figure 2-27 shows that the impedance trajectory at Healy stays outside the outer blinders. Case 3C provides an alternative to the undesirable load shed and possible under-frequency unit tripping results of case 3B. This method of immediately (within 10 cycles) shedding load could be implemented through some sort of remedial control system with system monitoring. The 10 cycle delay before load shedding allows a reasonable amount of time for logic and breaker operation. Particular system conditions would allow the remedial control system to arm and monitor the system for out-of-range conditions. Based on the severity of the generation loss or disturbance, the appropriate amount of load would then be shed as fast as possible. The result is that a minimum amount of load would be shed and the GVEA system would stay together. A similar logic control scheme would be required to employ the immediate Intertie trip of case 3B. Figure 2-32 shows that the impedance trajectory stays well outside the outer blinders of the SDBU-2 at Healy. Case 3D was one of many attempts to see how much load shedding was required to keep the GVEA system together. The results presented in case 3D point out that a shed of 6.2 MW is probably very close to the actual system conditions that would still allow recovery. 69 Figure 2-37 describes the impedance trajectory at Healy approaching and penetrating the outer blinder and then reversing direction. Even with only 3.7 MW shed the system still stays together in the simulation but voltage profiles look very poor. Case 3E was run to try to determine GVEA system stability after the North Pole unit tripped off, 18.1 MW of load was shed and the system began to stabilize. A 3 fault was applied at the Musk Ox 69 kV tap at t=5 s. Clearing this fault required the opening of the 69 kV line between the Gold Hill and Zehnder substations. The system recovered from this disturbance without difficulty. See figures 2-39 through 2-43. Two similar cases were run with a 3q fault applied at the South Fairbanks 69 kV substation and the Ft Wainwright 138 kV bus. The fault at South Fairbanks was cleared by opening the Ft. Wainwright to South Fairbanks and Peger Road to South Fairbanks 69 kV lines. The GVEA system recovered from this disturbance without problem. The 30 fault at the Ft. Wainwright 138 kV bus was cleared by opening the Gold Hill to Ft. Wainwright 138 kV line, the Ft. Wainwright to North Pole 138 kV line, and opening the breaker on the 138 kV side of the three winding transformer at the Ft. Wainwright substation. The system recovered from this disturbance satisfactorily. Two simulation runs were completed with a hard trip* of the North Pole unit. With a hard trip of the North Pole unit and a 3.7 MW load shed 10 cycles after the North Pole trip, Healy goes out-of-step with Chena 5 and Beluga 6. With a hard trip of the North Pole unit and a 6.2 MW load shed 10 cycles after the North Pole trip, Healy stays in-step with 4 A hard trip is defined as a unit breaker opening with no prime mover tripping. 70 Chena 5 and Beluga 6. Voltage profiles for the 6.2 MW shed case look poor with two sags to .58 and .76 pu. While a soft trip of the North Pole unit is not a worst case scenario it was considered a more likely occurrence. 2.8 Discussion of Stability Improvement Method System stability is directly related to the amount of additional Intertie flow required during a disturbance. Using case 3C as an example, when 50 MW of generation is lost and 18.1 MW is shed, the additional Intertie flow required is 31.9 MW (results in a stable system). When 3.7 MW is shed the additional flow required is 46.3 MW (very close to unstable). It seems unusual that there appears to be such a well defined upper limit on the amount of Intertie flow that will still allow a system recovery. The SVC units may allow the system to remain stable right up to a system stability breakpoint. If this breakpoint could be well defined based on various system conditions, this information could be used to employ the immediate load shed scheme described in case 3C. 2.9 Conclusion Two recent disturbances in the Golden Valley Electric Association (GVEA) system were simulated using the PTI software. Of these two cases, one provided acceptable results as compared to dynamic system monitoring (DSM) data. The successful simulation was used to study an apparent out-of-step relay trip at Healy. The simulation revealed that the necessary conditions occurred to initiate an out-of-step trip of the Westinghouse SDBU-2 relays. The simulation shows that the south-looking SDBU-2 relay trips and may have 71 tripped for the actual disturbance. Based on the differences between rotor angles, an out- of-step condition was not observed before the Intertie was opened. This study indicates that the supervisory over-current relay setting may be too low. A third case was constructed to aid in the development of a method to maintain system stability after a large loss of local generation occurred. A method is presented in case 3C which requires the immediate (within 10 cycles), direct tripping of a small amount (less than 20%) of load after the loss of the local generation. This method eliminates the severe system disturbances produced by waiting for an out-of-step condition to occur. System stability was tested after the loss of generation and load shed. The system stayed stable for three different fault conditions. It should be noted that this method was developed without regard to cost or future system upgrades. CHAPTER 3 SINGLE-POLE RECLOSING ON THE INTERTIE 3.1 Introduction The Alaska Systems Coordinating Council expressed an interest in determining the applicability of single-pole reclosing on the Anchorage to Fairbanks Intertie. While the Electromagnetic Transients Program database for the Alaskan Railbelt power system was being completed, research was conducted to determine how other utilities worldwide had implemented single-pole reclosing. Engineers with single-pole reclosing experience were contacted through the EMTP/ATP electronic mail list-server for advice. A series of computer simulations were then run to determine the effects of single-pole reclosing on the Intertie and surrounding systems. The results of the research and computer simulations are presented. 3.2 Description of Single-Pole Reclosing Approximately 90% of all faults on a transmission line involve only one of the three phases and ground [1]. Many of these single-line to ground (SLG) faults are transient in nature and can be self-clearing if the affected line is de-energized momentarily. Single-pole reclosing (SPR) is an automated technique used to help remove self-clearing SLG faults. When an SLG fault is detected, both ends of the faulted phase are opened and the line is 72 73 de-energized for a short period of time. The time the line is de-energized is referred to as dead-time. It is hoped that the main power fault arc created by the faulting element will extinguish itself during the dead-time. The line is re-energized (reclosed) after the dead- time has expired. If the fault was successfully cleared during the dead-time the phase will be back in operation. If the fault did not clear during the dead-time the fault is assumed to be permanent. It is general practice to then lock open all three phases. The fault is then cleared manually. Currently whenever a fault develops on the Anchorage to Fairbanks Intertie, the faulted section is isolated by locking open all three-phases to the north and south of the fault. Therefore, no form of reclosing (SPR or three-phase) is employed on the Intertie. The system operator could either attempt a manual reclose or dispatch a line crew to check and clear the fault. Depending on the fault location and the Intertie operating conditions, the opening of all three phases can result in the loss of all Intertie flow into Fairbanks. As there may be no spinning reserve or local generation on in Fairbanks, a system collapse in Fairbanks may ensue. The single-pole reclosing technique can be advantageous for important single lines, such as the Intertie, that tie two systems together. As described previously, many of the faults that occur on a transmission line involve a single phase and ground. Therefore, it is normally not necessary to open all three phases to clear many of the faults experienced on a transmission line. By only opening one phase the other two phases remain in service and continue to transfer power!. This can maintain the system until the fault is cleared and the ! Power transfer is reduced to approximately 54% of the pre-fault power flow [2]. 74 open phase is reclosed. It is not possible to maintain two systems connected via a single line with three-phase reclosing. The dead-times required are so long (usually 0.5 s minimum) that the systems will lose synchronization during the dead-time. Single-pole reclosing can therefore improve system stability dramatically for systems tied via single lines. There are three basic problems associated with single-pole reclosing that must be dealt with to ensure effective SPR operation. The first problem is ensuring that the arc has been properly extinguished before the line is reclosed. The open phase remains energized due to coupling with the in-service phases. This coupled energy can sometimes support a secondary arc. Secondly, the act of opening and closing a phase during the fault clearing/reclosing process can produce impact torques that may damage turbine-generator shafts. The reclosing action may also produce harmonics that can resonate with turbine blades. This can cause fatigue and perhaps failure of the turbine blades. And third, by opening a single phase, an unbalanced condition is produced. The negative sequence currents produced by this unbalanced operation can cause heating of the armature windings of nearby generators and motors. A voltage unbalance of 3.5% can produce a 25% or larger increase in motor temperature [3]. Immediately after an SLG fault is detected on a line employing SPR, the circuit breakers at each end of the phase open to clear the main power fault arc. After the power arc clears, the arc, now termed a secondary arc, can continue due to the mutual coupling from the in- service phases. The open phase is energized due mainly to the line-line capacitive 75 coupling. If a secondary arc is formed, it must be extinguished before the open phase is reclosed. If the secondary arc is not eliminated the system will reclose into a fault. The transient recovery voltage is also an important parameter to consider. The transient recovery voltage (TRV) is the voltage produced immediately after the secondary arc extinguishes. As the fault current is almost purely inductive, the voltage will be at a maximum when the arc extinguishes at a current zero. The TRV can approach twice the maximum open phase voltage due to resonance between line inductance and shunt capacitance at the fault [4]. If the TRV is high enough, a restrike can occur through the original ionized arc path. Other factors such as humidity and wind are variables that are uncontrollable but do have an effect on arc extinction. The best determining factors for secondary arc extinction are the secondary arc current and transient recovery voltage. The amount of voltage, and therefore the amount of secondary arc current, induced into an open phase is proportional to the amount of capacitive coupling between the three phases. For a fixed amount of capacitive coupling, the secondary arc current is proportional to the length of the line and the line energizing voltage. The pre-fault power flow on the line can also determine the amount of voltage induced into the open phase. A heavily loaded line will have a decreased voltage profile across the line and a lower induced voltage in the open phase [5]. 76 Kimbark [1] reported a figure of 18 amps rms or less that defined the maximum, unneutralized secondary arc current that would allow fast, self-extinction of a secondary arc. Secondary arc self-extinction should occur within a dead-time of 0.5 s or less with a arc current of this level. Kimbark later reported in [2] a secondary arc current of 20 amps rms or less that would provide fast extinction without neutralization. These figures have been borne out by years of successful single-pole reclosing experience [6,7]. Systems that produce a secondary arc current higher than 30 amps rms employ some form of secondary arc neutralization. This neutralization generally takes the form of the wye connected 4- reactor scheme described in Kimbark's papers [1,2]. This technique develops a parallel resonance between the distributed shunt capacitance and the external inductance (neutralizing reactors). If the neutralizing system is perfectly tuned the secondary arc current can be reduced to zero. However due to practical construction, losses, and harmonics, the secondary arc current is not reduced to zero. The secondary arc current can be reduced to 10-20% of the unneutralized value in practical installations. Another technique used by the Bonneville Power Authority [5] eliminates the shunt reactors and uses high speed grounding switches to short both ends of the open phase to ground. Lines longer than 50 miles and energized at voltages of 500 kV and above generally require neutralization. The Europeans have been using single-pole reclosing extensively since the 1950's. Many of these lines use no neutralization, although some would be considered medium length lines by American standards. Reference [6] describes systems in France and Germany that use no neutralization. These are 400 kV, 147 mile long and 380 kV, 73 mile long lines respectively. 77 Impact torques can be developed from rapid changes in air gap torque. Disturbances in the electrical system cause these air gap torque transients. Therefore, single-pole reclosing imposes repeated disturbances to the turbine-generator shaft unit. The fault provides the initial disturbance. The opening of the faulted phase applies another disturbance. The reclosing of the faulted (but now clear) phase provides another disturbance. A fourth disturbance is experienced if the reclosure was unsuccessful (fault still exists, .. re-clearing). A number of single-pole reclosures over a period of time could cause a loss of shaft life. There have been many papers written on the subject of turbine- generator shaft fatigue duty [8,9,10,11]. It would appear from reviewing these papers that single-pole reclosing has a very small impact on the life of the generator-turbine shaft. The European contributing author in [6] reports that they have not had any shaft failures that could be attributed to reclosing. It would be in the best interest of the individual considering single-pole reclosing to contact the generator manufacturer and discuss the effects of single-pole reclosing on the particular unit. This should definitely be considered if single-pole reclosing is going to be applied near a generating unit. A problem similar to turbine-generator shaft fatigue is that of turbine blade fatigue. "Single-pole switching (SPS) and high-speed reclosing (HSR) can excite 120 Hz torque oscillations that may resonate with the coupled blade-rotor torsional natural frequency and inflict damage on the blades" [12]. The units studied in [12] did not appear to be adversely effected by this type of torque oscillation. As in the case of shaft fatigue, it would again be in the best interest of the person contemplating single-pole reclosing to contact the manufacturer of the generator/turbine unit for recommendations. 78 The unbalanced condition imposed by opening one phase for single-pole reclosing can produce negative sequence currents in the armature windings of nearby machines. The negative sequence armature currents can induce 120 Hz currents in the rotor which cause additional heating of the machine windings. Kundur [13] lists general figures for permissible short-time negative sequence current capabilities of some generic machines. These figures are: salient pole generator, I,2t=40; round rotor generator, indirectly cooled, Ip2t=30. 3.3 Computer Simulation Descriptions A general Electromagnetic Transient Program (EMTP) database of the Alaskan Railbelt power system was created and described in chapter one. The EMTP and a portion of the Alaskan Railbelt power system EMTP database was used, with modifications and enhancements, to study the applicability of implementing single-pole reclosing on the Anchorage to Fairbanks Intertie. The EMTP was used specifically to determine secondary arc current levels, transient recovery voltages, and harmonic generation. The EMTP program used was the latest Alternative Transients Program (ATP) version. Power Technologies Incorporated's PSS/E software (version 22.0a) was used to determine the transient stability of the power system with single-pole reclosing applied to two different sections of the Intertie. The WN94.SAV base case, which models a winter 1994 peak load, was used. It was assumed that single-pole reclosing would be applied on the Teeland-Healy and Healy-Gold Hill line sections of the Intertie. The PTI software has an option (Single Pole Circuit Breaker) that allows the user to calculate the equivalent line 79 section impedance with one phase open. The one phase open impedance was calculated for these line sections. Various dead-times were simulated to determine the maximum single-phase open time that would allow a stable system recovery. 3.3.1 Description of the EMTP System Model The ATP input data case used for the SPR simulations was assembled in blocks. The ATP uses FORTRAN formatted text files for data input files. It is very easy to make a mistake when assembling these data files. Therefore, a step-by-step approach was taken in assembling the final data input file. A basic case was first assembled from the larger Alaskan Railbelt database. All transformers were modeled using ATP's TRANSFORMER model without saturation; these data were not available. All transmission lines were modeled with the ATP's distributed parameters line model. Initially all machines and equivalents were modeled as ideal voltage sources. After this basic case was verified additional features were added. ATP steady-state operation matched the PSS/E WN94.SAV power flow case. The generator at Healy was represented using ATP's built-in Synchronous Machine 59 (SM-59) model and was the first addition to the basic case. The SM-59 model provides an excellent representation of a three-phase synchronous machine. It allows some external controlling inputs such as exciter and governor controls. Without external inputs, the model initializes the exciter and governor values at steady-state values and assumes they, are constants throughout the simulation. The SM-59 also allows the modeling of multiple masses on the turbine-generator shaft [14]. Only basic machine information was available 80 for the Healy unit. The Healy unit was modeled with the available data and with a single mass on the turbine-generator shaft. The Healy exciter control was the second addition to the basic case. The Healy exciter was modeled using ATP's built-in Transient Analysis of Control Systems (TACS) routines. The Healy exciter/regulator control model used was the IEEE AVR type 1 [15]. Saturation was not modeled in the exciter. The exciter/regulator monitors the machine terminal voltage as does the actual exciter. The TACS exciter control was tested with the SM-S9 by switching a 18 MW test load in and out of service at the Healy 138 kV bus. The correct exciter and SM-59 response to the change in load was observed. The governor at Healy was not modeled as its dominant time constant is 0.55s. It was felt that the governor could not respond fast enough to make a difference in this simulation. Prime mover input power was held at the ATP calculated initialization value (Pmech=Petec) throughout the simulation. The Static VAR Compensator (SVC) at Healy was the next piece of equipment modeled. The model used for the Healy SVC unit was taken directly from the EPRI EMTP Workbook IV [15]. TACS was used to model the SVC. This model varies from the actual unit in several ways. The input filters added by General Electric after initial construction are not modeled [16]. The remainder of the control system is similar to the actual unit. Feed-forward gains are different than the actual SVC gain because a rms input voltage transducer was used in the SVC model. The actual unit uses a voltage squared transducer. Both the actual and simulated SVC units regulate VAR output based on a aggregate voltage input from all three phases. The General Electric SVC unit does not 81 allow full thyristor conduction. The model of the SVC unit at Healy does allow for full thyristor conduction. Therefore the values of the inductors used in the actual installation are smaller than the values used in the simulation (Q;=V,,2/X;). It appears as though the reactors used at the Healy SVC are not rated for full available current [16]. The harmonic filters at the Healy SVC were modeled. Correct operation of the Healy SVC model was verified by switching a 18 MW test load in and out of service at the Healy 138 kV bus. If the object of the study had been to determine the impact of SPR on SVC operation then the SVCs could have been modeled more accurately. The object was to model the effect of the SVCs on the system during SPR. Therefore, it was felt that the differences between the actual and simulated SVCs did not have a negative impact on this study. Models of the Gold Hill and Teeland SVC units were added. The SVC units at Healy, Gold Hill, and Teeland are the same model made by the same manufacturer. Each particular unit has a different VAR rating. Each SVC model was modified to reflect its actual VAR ratings. The harmonic filters at Gold Hill and Teeland were included at each site. The transformers at the Teeland substation were modeled. The system connected to the Teeland 115 kV bus was represented with an equivalent. A 230 kV equivalent was placed at Point MacKenzie at the end of the 230 kV line connecting Point MacKenzie to Teeland. All of the equipment between Teeland and Gold Hill is individually modeled in ATP. The transformer at Gold Hill and the SVC unit are included. An equivalent representing the Golden Valley Electric Association (GVEA) 138 kV system was connected at the Ft. 82 Wainwright end of the Gold Hill - Ft. Wainwright transmission line. An equivalent representing the 69 kV system was connected at the Gold Hill 69 kV bus. See figure 3-1. The Thevenin impedance for all of the equivalents was found by using the PSS/E power flow and fault analysis software and applying a single-line to ground fault at each equivalent location. The equivalent was modeled as a three-phase coupled R-L branch. An ideal voltage source with the correct magnitude and phase angle was connected to the Thevenin impedance. 3.4 Transient Stability Results: Dead-Time Determination A series of transient stability runs were first completed with the PSS/E software to determine a realistic dead-time for each line section. Three different cases, each with dead-times varying from 0.4 to 1.0 s were run. Case A simulates an SLG fault at the Teeland 138 kV bus and a single-pole reclose of the Teeland-Healy line with varying dead- times. Case B simulates an SLG fault at the Healy 138 kV bus and a single-pole reclose of the Healy-Gold Hill line with varying dead-times. Case C simulates an SLG fault at the Healy 138 kV bus and a single-pole reclose of the Teeland-Healy line with varying dead- times. See Table 3-1. In every case, the system stays in-step until the dead-time exceeds 0.75 s (Figures 3-2 through 3-7). In case B-2, with a dead-time of 0.75 s, the system goes out-of-step (Figure 3-8). Examining the frequency plot (Figure 3-9) for case B-2 shows that the SLG fault/reclosing action at Healy causes Healy to first decelerate and then accelerate. Figure 3-8 shows Healy's rotor angle quickly diverging from the system to the north during this I-¢ emMsLy Pt. Mackenzie Wo Or Teeland pole Ses r 1 Douglas Cantwell 7 Nenana 1 ~ 3* ! rq ' 1 tf L230 ww 115 0 [mom! 3 {ane mi | 129.7 mi B03 mit | | 882 mi 435 mi ae ' BrP oe cit mer er ahi ' 1 r 1 if sewed Aa™% |! baneeeesennee . ! a i i] Ye Ye | Teelond 115 kv > 4 ' t ' ' ' ' ' ' ' System as Modeled in ATP m cor Gold Hill Ft. Wainwright 138 kV Be) £8 4, oS SJIN3AY34ISIO JTINY YOLOY €2thl S66l S2 ddb ‘*3NL (SONOI3S) WIL 0000°6 0000°¢ 0000°S 0000°€ 0000°1 000"ot 0000°@ 0000°9 0000°h 0000°2 00 eo jo lo s |o lo o |o lo = lee zw ore -—oO ee a we : | ne | = on : | sey i | ae | a a al ws a all z ue » 3) Ps a rm q q om Oo} q oO 8 Ss} 2] 5 oes =} a} =x" a oO a Or ul oO ul N i 7 —— a od ol a = ode qa ia o z a ~ = cr aqua ‘al SO, wl] =a) + ec uw F bs a 3) 8 oe Ww zw ¥ ital a ew os oO = x a qa uy ul ul x ou oe! +] *) oS = [et ee ee i 7 nw 4 = - - 1wW 5] ;| | sO) @ wo} fas s ul “ “i = p= Tt) = —n yy wo oO nr . . . z 2 2 a Fr #2 = = ¢& os dS = oO r 4 wo _ uw =z - ac> a wo Co -a Cc za eae ! € eo jo fo ‘ z Ss |s |s ‘ os |o fo ‘ Ea o oO o %® JINIY3I4IO JTONY HOLOY (SON ) W h2thl S66l S2 dd ‘3NL SONOI3S) 3WIL 0000°6 0000°¢ 0000°S 0000°€ 0000°! bs 000"0! 4 0000°9 0000°h 0000°2 0'0 co jo fo s |o |o o o o a rT 1: T mm TL) IL eet ss zo on ei eunly o- | em OO) aes ee or I @ I og : rw ee oo qj: « a By als ba aw al MT lia Im q a o | | | cl! | 16 : Bo | |S eo ae } o 6 SET Te eee ve tet ates tale a te ‘ 9) = || o i 7 ' =X Or | eh) |) ‘ co rs ia a] al a : o i =) ic = =| = ‘ Cc yw ‘ ee wl mi 4 1 e c=H Sele ae = wo ¢q| > > ' i =, || oo iS ea aA a, w zw Fa SI ta a ew Oo = x QO aqa-— a jo G a ou | | “ c Oo ot 0) “| oa) fe - a nw =i =| =| «| fia ; 7 3 +N rol ot a ae | os = a a = po =) am en gq ot oO HF * * * = z =z = ce iG |) Guice I iB aI ww - oO z OND te So oe age ee ee ees es Si ge eee ose pas ee ge ee ct eee ese tee ty eeeseeas = =n Zor a> = wel o -a on | 2a ma | ee 1 Tis est fest lg 2 Ss |e fo 2 a d 3 ei & |e |~ is 2 | 6 sa SINSYS4ISIG JTONY YOLOY 92th! S661 S2 YdH ‘3NL (SQNOIIS) BWIL 0000°6 . 0000°S . 0000°€ . 0000°1 000"0r o lo fo cs |s |o @o oO oO wi : : : wo zo Si ° v r= ! oe oe | on | o oa 4 || 4 =o ql: a oie a} Wo =| | Rm a “| a o s q Oo x Z| fie a} OG} ru ao of a ere ooo 6G _ = H 7 7 + Z Hr o = a “ acre a al 2 oc ~wo 3Za vo -| + 2 qa-wm —_ wo q > > in > § me 2| = = ‘ a w zw rod 6G ‘ ' a rw oO x x ' ' GO aa- o og ' ; « Ou ' ' © oO ay ol ol CF theese eee 4 © ae a ‘ ‘ =| a3 ' ' tw | 7 7 ' ' Oo O @ : = 235% a, a ' ' = 50 ' ' am fu of ww Lo. a : Oo nF i. a fe ' 1 Zz a ' ' ; 2 ' ' oc ad =| Tv = ’ 5 w Ww cs) oo | ‘ | - oO ' ‘ =z i fics (ee ee ce = en : ' = - ' ‘ c> ‘ 7 > ug ‘ ' oC Fao ‘ : cs za ' ‘ _ —a }~~ Ho aie wien dee oewie oe § oes 5 ce ecbhen ese : e eo jo fo ' ‘ ‘ ' ' ‘ ' : Ss |S |s ‘ ' ‘ ' ' ' ' ' ' eg, os |o |o ‘ ‘ ‘ : ' ' ' ‘ ' ses 2 |o |e L | 7 * SINSYSSSIG JTINY YOLOY Ihi2l S661 S2 Ydd *3Nk Sceeoee ites 0000°6 0000 0000'S 0000°€ 000071 00°01 0000°8 0000°9 0000°h 0000°2 oo eo fo fo s |s lo o |o |o tH : ’ 7 zo ©» -—o Gus ° v WE | oo \ s or | pe = | i | a Fotos wo qi: « « Cy | ea ae | 0 o rm a “| «= om Qo) q oO} fail >] 2 35 as =f a. St xu a Oo no eT eit yo an eae Ay oA A aN = oce aq oaa || (ae ae PML ten AIL ee 2 c Gua +z a wy | =| e Uw = wo a} >| > f iz || (pe es Z| =| a ra | a ga & : J oi nt Oo} =x x : oO aa- o oo ‘ BS ou ede ‘ a # a3 aw 4 a 5 : / 1W | ; i 2 35% a} oto ' = ||p>o ‘ ce en oy of of \ oO MF A : 2 ayia ae fe | ne ce zx =| Oz ' eslllitar sg dg ' - o ' = (a a | (ei ’ Ze ! o> ' SU f oC ro ! oS Za ‘ | | lee ’ 2 o lo fo ‘ z Ss |e |e ' z. os |o lo ' see a |o |o | 8 2 JINSY34SI0 JIINY YOLOY N ) W hhi2l S66l S2 Hdd ‘3NL (SONOI3S) awd 0000°6 0000°¢ 0000°S 0000°E . 0000°t . 000 °O1 0000'@ 0000°9 0000°h 00002 Ox) eo fo fo os |o |o o |o |o wi 1 . : wo zo on ls ? me rE ; | <p : | we : wo : | °o i | oa : rw a: oy oI 6a ai al, = os “|e =lh 2 Iw ol nt | Im a q| oO Q| cq Oo} = « 2 Z| 3 Go ao of a eu oa of o aT uo oo - = i 7 i Z Or e cI oy = ax qa gq oS + « = SS Cc +-we 320 vo -| - eo qaqa-w wo gz > ~ > = - a a a vey zw x ww ua] J rus o or Ur QO aa-— o of} GS < ow | | | c£ oO a of ow o 3- jy a 4 aw a =| o . is | | " a > sa Ss coa- a S| Bt = Do cen Fy vy 4 Oo nor © . . z 4 so} ne 2 c aa o of GS - oO Zz wo = Aa ' =z - : o> ' +r wo ‘ x £. ‘ Cc 2a : ~ oe i ’ ’ ’ i ' ' é ° ° ° ' ' : ‘ ' ‘ ' ' ' = Ss |S js ' ' ' : ' 3 * ° ' ' 1 ' ' ' ‘ ‘ : fe s |s |s ‘ : ‘ : ' ' 3S = |F Lt | | 9 oP SIN3Y34510 JTINY YOLOY Olthl S66l S2 dd¥ *3NL SSO eae 0000" 0000°¢ 0000's 0000"e 0000"! 000"! 0000°8 0000°h 3 aoe | s |o lo i 1 : 2 je |2 ' ' ‘ wn ’ 5 : ' ' ' o ' ‘ ‘ : t 6a oof o |v ' R= : | ' wo ' wr | : oO | t a ' og | ‘ ae Soo 2 ; ra qi al, « " ci 2 e tu o| 7 oO ° i OF a a ' o o al oO ' ee Ss} 2] 5S : Be a} a} ' mu a o| a ' be ul ul . se a} ' > =z Me S| i om S a gg) ' o Tc >wa ‘ 2 ji = a. | -| ’ eo an a ‘ wo q > > v 7 i > xr ae z a a ' ~“! Ww zw ry ic a i : J orus o| ot OF ' ' ' GO aa-= gy od SG ' ‘ ' a Ou Hi \ 5 Co. ft | . ' o 35 eo te me qe hee eee : i Sal ' ' ata 1 d 1 ‘ ' q o| ' ‘ ‘ 2 3% ay Mn ' ' ' = Do ' ‘ ‘ crn ree cree ene ns ee ie ere : Oo mF . | ‘ ‘ : zo 2 ‘ ' : = ‘ ' : «© aa =| z| = : ‘ : . 2 fer (er es ‘ ' ‘ e oo ' ‘ ‘ Z wo eee fae : = =a ‘ ’ : ze i ' : + ' ' ' a id : : : oe Fao f ‘ t oe za ‘ ‘ ‘ = ent : odes : é eo jo fo ‘ ‘ ' 3 Ss |s {|s ‘ ' : ; os |o fo : ‘ ' SF] > |o |e = ee Sk de eee & 3IN3Y34SIO 3TONY HOLOY { ) 3WI hEthI S66l S2 Hdd ‘3Nk SE OE a 0000°6 0000°% 0000°S 0000°€ 7 0000"! . 90000! 0000°8 i. 0000°9 0000°h 0000°2 ; 0°0 a T os lo |o ' o © oO } a iy = aT 1 in + : . ' ~ ‘ zo on as ely rE il u~ | wr aa) | @ | w rw ea ee Sf i o 4 aw mle Im « a mo 2) « oO} Zo 418 2 “uo wh x wy ow i) Oo} | Fr ul oO ul eo - = n 7 7 c ZX Or a7 ao4 ° = ar a| al a o +. a o cS >we iz in =| 2 aw * wo Cr ee iz > =r ° = a a ww Zw x wl wi] a Rew oo oO} x x Oo aqa-— oy of Go a ou a co | of oot. |< an =| / Iw E a) mn aes aqa- =| sles = 20 c - a “ wy oO oF J ‘a i z os) a} on zz 2 eae 5 7 i a og - o Zz wo =< ed zoe o> = wo Cc -a. Or zo: - =a 7 2 eo lo fo = Se fo |s 8 e os |o |o i, eo: _ lhe - b--e...g- —-L S \AIN3NDIY4 SNE ¥ ANIHIUW Lhthl S661 S2 dé *3nk IsqNoaae) ult 0000°6 0000°¢ 0000°S 0000°E 0000°I 000'Or 0000°8 0000°9 0000°h 0000°2 0°0 Sir ean siees So foo ie: te o lo io [a i ' wo Lona r i & bs 7 w wo fw [we lw ‘ rs 1 Zo on eh t ? v rE ' | o- ' | Wwe ‘ we ‘ | a ‘ og ' | Tw | oo t | 7 : A + 6 sw Im m Sl 2 2 ci S| 2 O| | So ee tel CL eu im Oo wi QO ae =| a = | ise Ret Rene a ZX Gr 0} = sor a on = a 2 a SG eee Ss eee ne wo z 2 o CF ~we oy 8 eS JZ20 5 coll ere | e aqa—-H a ul wo o om wo a ete pre my os Ge ” uo SI eB x w Zw x 4 2 —_ ew = ss 2 So Gace 1a wl + am a Ou pel a (mE) o ali Cn eae i ® x nw =I . a 1w Z| Zz ~| | sO SSH IN et ley eo. oO iG x = — ce ren oO nr Zs a ac ad w wn - Oo =z lus = =H = - > a iad of oC Fa Oren os = =a é 2 lo |o fo = so |o |o |o 3 ow ilo [wo jo sie a ta fa daw Ha mol) ha [0 || [0 Table 3-1: Transient Stability Simulation Results Case A-1 B-1 A-3 B-3 C-3 Description SLG at Teeland, reclose Teeland-Healy section SLG at Healy, reclose Healy-Gold Hill section SLG at Healy, reclose Teeland-Healy section SLG at Teeland, reclose Teeland-Healy section SLG at Healy, reclose Healy-Gold Hill section SLG at Healy, reclose Teeland-Healy section SLG at Teeland, reclose Teeland-Healy section SLG at Healy, reclose Healy-Gold Hill section SLG at Healy, reclose Teeland-Healy section Dead-Time 0.4 - 0.6 0.4 - 0.6 0.4 - 0.6 0.75 0.75 0.75 1.0 1.0 1.0 92 Qut-of-Step ? No No No No Yes No No Yes Yes time period. Healy goes out-of-step immediately after the line is reclosed. The reclosing action actually appears to force Healy out-of-step. The Beluga units’ frequency increases. Chena 5 (FMUS) and North Pole GT#1 unit frequencies decrease. Cases A-2 and C-2 stay in-step with a dead-time of 0.75 s (Figures 3-10, 3-11). 0.75 s appears to be the upper limit for a stable system recovery for case C-2. With a dead-time of 1.0 s, case A-3 is the only case that remains in-step after the line reclosure (Figure 3-12). 1.0 s appears to be the upper limit for a stable system oy - SINSYS4SS1IO JTINY YyOLOY €€thl Sb6l S2 Ydd *3NL aoe 0000°6 0000" 0000'S 0000" 0000°1 00°01 0000°9 0000°h y Se pleas oe eas ‘ ‘ ‘ o |o |o ' : ‘ o eaten : \ > ae EE S snsmunube @ hou eleetir ataul 1 oO - T r v i ' ' ' rou ' ' ' ow en ' ' ' we ; | : ! ' aN : | ' ' ' : ek cee gl Boole! ‘ ow. ' | ' t ' wa : ' ' i ra al: || Fi ' wo ai a) « i f ' uu o “lt uu ' wo 1 “a i ¢ Yr Yt eet re zm q a 4 2 : om QI a Oo yh ' 4 = s SI z 5 ' ' i =o a ele ‘ ' ‘ zo wy a OB ' ‘ ' ° Or ul uo ul Nie Sere wala emesis SN ae Gans ceara Sipe imam Cant aed fe emttese _ . i 7 7 i oy oy nn , =z or 2 eo ' on eq} a) « Cc Gwa ‘ ale (OL wl el al ' 2 DAK {is = re) a ~ > r ix -~ -_ Zaid s ated a 2 Yl ‘ : a aqaqa- uy ul ul 4 ‘ = on “ r| a ‘ ‘ “J o ov bene e-- ee L oO soe =| al = f : nw = ae ae ‘ ' . hw 5| 5| | ’ z@ 3% 7 8 & ' ‘ = So ' t crn alee te asa 4 cecewese eee ee Oo HF e} ele : ' ' Zz =) ~| Sj ' ' ' or Zz} Z| z ' ' ' © aad lear ' ‘ ' val ' ' ' So ' : ' = pececceee pesceeeee ‘ So5 ' ‘ ' zr : ‘ ' o> t ' : - Wo ‘ ‘ ‘ oa = 6. ' ' ' a za ' ' ' - —a t wececces Lbewscce ‘. rs oe jo fo ‘ Z s |S |8 : i os |o lo ‘ fe o |o |e ey = |= |[- Ss a JINSY34I410 JINN YOLOY : . (SON ) JW Ohthl S66! S2 dd ‘3NL SONOJ3S) 3WIL 0000°6 0000°S 0000°€ 000°0! 0000°9 0000°h 2 |= |e | s |s |o ; i o |o |e ‘ w . 5 : ' Re ee eee, ‘ zo Sis e iy rE : | Or $ | Wwe 3 or ; I o : oe mf | aw a Sy Se >=) qi a, < i ey Ses Wn el : em « « ' mo Oj q Oo ‘ — & =) 4 a ' oo a) wW I . cai oz & ! =e qj oo 6G ' ' . 7 3 terceee = oe yo oA os : ' = Gc = a] a ¢@ ' ' oi Sa = fe] s : : cr >we i" , fra i ‘ e a-4 2 al es — t wa? sg} >} > pages + 7 : ‘ > tLe z a ad os ' ui x wl ‘ ‘ aS rw a ¢} OY ' ¥ a qa = o of UG ‘ ‘ a o . uw A ‘ : c o o 2. iss * ns - Iw ce 4 ; a 3% a} af 8 ' : = 20° ' ' om ee y oF oF e ean oO mF “| : : z ' ' wr zz} OC ‘ i x x =x s i 5 ae 3} 2} 5 - oO ' ' ci ey 4 pow Sere _ SD 7 i Zz + ‘ we» ' x tad ' o a ; o za ' - -a2 = eee et eee Lt TV oe jo lo ' ‘ : s |s |s8 : i ‘ ef s |o lo ‘ : ‘ Nes 22. |S | aie B tod Lk _ — e 3-11 Fi w a SINSYS4ISIO JIINY YOLOY ISthl So6l S2 Hdd *3nL Ne 0000°6 0000°¢ 0000's 0000°€ 0000"! 000°0! : 0000°9 0000°h Selene |S | | s |o |o t ‘ oO oO @o ' ' wn 1. 7. it : ' Pott ttt {ttt Etna nt Relat eb artenlmetmto rob) storstmsbochentonts Sivabahete lsvluiediieiens dante es t ' an : ‘ ow iv h i ' ws : | ‘ ' n— | ' ' fo a i pe cere es Paks Dl ee he ce he. ©, J----- ne I : ‘ ja | , a a al eagal nies t ‘ rw Ee ene 4 { wo qi: a q t : us fail saee| Hela j ! & | | ol 4 oie of oa 6S ' rau Ss fF OS ' =° om a ' zu ao oO | ' N or ut ul ul ¥ al ‘ i 7 i 4 ri Zor te ‘ on = oar qa a | ; Some an ty ia 7 77 v c aqua \ Iza (patted } 2 ww _ wo a ~ ~ iz > - ” z st | ww zw soo) a ew oO x x oO qa ut ul ul « ou | os] os] eo A oT tO) alee (ale feat Ve aes a eet eho phe a =a nw + =| | . ow Pa | oO a a aI a cl: ea = 20 crn a 2 7 bamctonenes a mags ease ees SOp gs alata SSE: § eels Sheen: admin Rin Patechsecmes ocean a Oo wre . . . = zz} oe cr m aad a a A w ow - o z eee eae celeseesiil welts tem bestegetiamhanteiesteaenseniishensestetestestaal _ = = - ac> = Wo ov -_-a eo za - Neen eee ee ee ec al aa aula 3 es |s |s s |o |o il s |g |g 96 recovery for case A-3. As noted earlier, case B-2 goes out-of-step with a dead-time of 0.75 s and case C-2 appeared to be very close to going out-of-step. 3.5 EMTP Results: Secondary Arc Currents, Transient Recovery Voltages The transient stability simulations defined the dead-times from a system stability point of view. To ensure effective single-pole reclosing, the secondary arc must be extinguished within the dead-time. A series of ATP simulations were run to determine secondary arc current levels and associated values. Table 3-2: Secondary Arc Current, TRV, Induced Voltage; w/ 3-@ SVC voltage regulation Secondary Max. Open Transient Secondary Fault Line Arc Current Line Voltage Recovery Voltage Arc Location Section (Amps,rms) (kV, peak) (kV, peak) Harmonics? Healy | HLS-GHS 12.2 18.2 @ GHS -30.7 3rd Nenana HLS-GHS 7.5 10.0 @ HLS -16.4 Very Little Gold Hill _HLS-GHS 1D 17.3 @ HLS -15.0 Very Little Healy TEE-HLS 22.9 18.4 @ TEE -31.4 Very Little Cantwell TEE-HLS 23.4 14.5 @ TEE -32.5 High freq. Teeland TEE-HLS 19.5 14.3 @ HLS -38.0 High freq. A dead-time of 0.5 s was selected for all ATP runs. In every case the secondary arc current reaches a steady-state value in approximately 0.3-0.4 s. The location of the fault 97 was moved along each line section. The results of the initial simulation are shown in Table 3-2. Secondary arc current values for the Healy-Gold Hill line section were found to be well below the recommended 20 amps rms . The secondary arc current values for the Teeland- Healy section were found to be higher and in two cases exceed 20 amps. The higher secondary arc current experienced on this section of line is due to the increased length of the Teeland-Healy section and subsequent higher induced voltage. The length of the Teeland-Healy section is 195 miles versus 103 miles for the Healy-Gold Hill section. This section of line is also of a different construction and has higher capacitive coupling than the Healy-Gold Hill section. The maximum open line voltage is the maximum voltage measured on the open phase. 3.6 Overvoltages During Single-Pole Reclosing It was noticed that some of the bus voltages were either far below (0.77 pu) or far above (1.35 pu) nominal during the dead-time (Table 3-3). The abnormal voltages are caused by the severe unbalance that occurs with one phase open. It was thought that the SVC three- phase voltage regulator/thyristor control may have been ineffective due to the unbalanced conditions. The SVC models were modified to include phase independent regulation/thyristor control and another series of runs were completed. This modification affected the voltages along the line sections and therefore the secondary arc current values. Table 3-3: Voltages at 138 kV buses along the Intertie during SPR (per unit) Fault Location Healy Healy Nenana Nenana Gold Hill Gold Hill Healy Healy Line Section HLS-GHS HLS-GHS HLS-GHS HLS-GHS HLS-GHS HLS-GHS HLS-TEE HLS-TEE Voltage @ Teeland Va-1.13 Vb-1.04 Vc-0.99 1515 0.99 0.91 1.10 1.06 0.99 0.98 1.02 0.95 1.13 1.02 0.99 1.14 1.01 0.92 0.98 1.05 0.98 1.02 1.06 1.05 Voltage @ Healy Va-1.29 Vb-1.09 Vc-1.03 L335 1.09 1.03 1.23 Lal 1.00 0.86 0.94 0.77 1.23 iM 1.00 1.31 1.11 0.87 0.95 0.98 1.11 1.00 1.01 1.10 Voltage @ Gold Hill Va-1.02 Vb-1.02 Vc-0.96 1.02 1.01 1.01 1.02 1.01 0.94 1.00 0.94 0.98 1.03 1.02 0.96 1.02 1.00 1.02 0.98 0.99 1.06 1.00 1.00 1.04 SVC Control 19 30 19 30 Io 36 1o 36 99 Table 3-3, cont.: Voltages at 138 kV buses along the Intertie during SPR (per unit) Fault Line Voltage@ Voltage@ Voltage @ SVC Location Section Teeland Healy Gold Hill Control 0.97 0.95 0.98 Cantwell HLS-TEE 1.05 0.98 0.99 1o 0.98 Pt 1.06 1.02 1.00 1.01 Cantwell HLS-TEE 1.05 1.00 1.00 30 0.99 1.09 1.03 1.05 0.96 0.99 Teeland HLS-TEE 1.05 0.99 0.98 1o 1.02 Pal, 1.06 1.05 1.00 1.01 Teeland HLS-TEE 1.05 1.01 0.99 36 0.98 1.06 1.02 The modification to implement per phase SVC regulation/control did improve some of the under-voltage conditions (Table 3-3). The over-voltage conditions were improved but only marginally. The inductive VAR output of the SVCs may not be high enough to clamp some of the voltages seen during single-pole reclosing. Due to the additional voltage support provided by the modified SVC models, the amount of induced voltage increased and, correspondingly, the amount of secondary arc current increased in most cases. Overall, modifying the SVCs for single-phase regulation did not improve the voltage regulation during single-pole reclosing. 100 Table 3-4: Voltages at 138 kV buses along the Intertie during SPR (per unit); Fault Location Healy Nenana Gold Hill Healy Cantwell Teeland Line Section HLS-GHS HLS-GHS HLS-GHS HLS-TEE HLS-TEE HLS-TEE Voltage @ Voltage @ Teeland Healy Va-1.02 Va-1.02 Vb-1.05 Vb-1.11 Vc-0.97 Vc-0.90 1.02 1.02 1.06 1.14 0.97 0.91 1.12 1.29 1.06 1.08 1.01 1.06 0.98 0.93 1.05 0.97 0.99 1.14 0.97 0.96 1.05 0.98 0.98 ila 1.01 0.96 1.05 0.97 0.96 1.10 Voltage @ Gold Hill Va-1.02 Vb-1.01 Vc-0.98 1.02 1.02 0.98 1.02 1.02 0.94 0.99 0.99 1.06 0.98 0.99 1.06 0.99 0.99 1.06 Healy exciter/voltage regulator monitors Healy 138 kV bus; w/ 16 SVC voltage regulation SVC Control 19 19 19 19 19 A second attempt was made to improve voltage regulation during single-pole reclosing. The exciter/voltage regulator at Healy was modified to regulate the 138 kV bus voltage at Healy. An initial run was completed with the SVCs modeled for three-phase regulation/control and the exciter/regulator monitoring the Healy 138 kV bus. No improvements were seen for this configuration. The exciter/regulator modification was 101 then combined with single-phase SVC regulation/control. Voltage regulation improved somewhat and the data are presented in Table 3-4. The modifications performed to try to improve voltage regulation during single-pole reclosing provided marginal results. Each modification was chosen so that it might prove beneficial to overall system operating conditions and not just single-pole reclosing operation. The over-voltages produced during single-pole reclosing are a major concern. The duration of the over-voltage is the duration of the dead-time. It may be possible to reduce these voltages by increasing the inductive VAR output of the SVCs at the buses that experience over-voltages (i.e. Healy). Another possible modification would be to switch in a fixed inductor during single-pole reclosing. 3.7 Secondary Arc Neutralization Methods The secondary arc current produced on the Healy-Teeland line section exceeds the recommended 20 amps rms and this section may require neutralization. The modifications to improve voltage regulation during single-pole reclosing produced poor results. Therefore, the data produced in the initial simulation (Table 3-2, 3-0 SVC regulation/control, no neutralization) are considered representative of SPR operating conditions and are used from this point forward (Figures 3-13, 3-14). Examples of secondary arc neutralization techniques were mentioned earlier. Simulations were run with Kimbark's 4-reactor scheme and BPA's high speed grounding switches on the Healy- Teeland line section. A third option was also explored. This involved sectionalizing the open phase of the Healy-Teeland line section. The results are presented in Table 3-5. €I-€ sNsIy Secondary Arc Current at Nenana No Neutralization 100 80 60 40 20 +1 Bi} 0 Current (Amps) -100 }——_-—_—_——— T T T 1 0.18 0.24 0.30 0.36 0.42 0.48 0.54 0.60 0.66 0.72 Time (s) col Secondary Arc Current at Cantwell No Neutralization 100 | > oa oO oO oO 2 N oO | UVLO Whit mii Hv VEY Current (Amps) oOo fF N o Oo Oo —80 -—100 T T 0.18 0.24 0.30 ee 0.42 0.48 0.54 0.60 0.66 0.72 Time (s) 104 Table 3-5: Secondary Arc Current, TRV, Induced Voltage; Healy-Teeland Line Section, With Neutralization Secondary Max. Open Transient Fault Arc Current Line Voltage Recovery Voltage Neutralization Location (Amps.rms) (kV, peak) (kV, peak) Type Healy 30.2* 14.3 None 4-reactor Scheme Cantwell 14.0* 11.6 -5.0 4-reactor Scheme Teeland 14.1* 10.9 None 4-reactor Scheme Healy 27.5 1.2 21.4 Grounding Switches Cantwell 10.1 1.1 -35.2 Grounding Switches Douglas 13.0 2.3 15:9 Grounding Switches Teeland 51.6 1.2 25.2 Grounding Switches Healy 71 24.5 29.3 Sectionalizer, Cantwell Cantwell 17.4 56.6 33.6 Sectionalizer, Cantwell Teeland 17.2 84.8 30.2 Sectionalizer, Cantwell * The secondary arc current reported is the rms value of the sinusoidal component. These values are all offset by a d.c. component. The offset is 17, 8, and 64 amps for Teeland, Cantwell, and Healy respectively. Two reactor banks were modeled in the 4-reactor scheme. The line section from Healy to Teeland was compensated at each end. The amount of inductive compensation at each end was 50% of the total line section charging VARs. The 4-reactor scheme produced poor results (Table 3-5, Figure 3-15). A d.c. offset is produced in the 4-reactor simulations due to the trapped charge on the line. The line conductance is very small as modeled by ATP so the charge remains on the line throughout the simulation. Also, the CI-€ ony With Neutralization — 4 Reactor Scheme 100 80 60 40 20 0 | N Oo Current (Amps) | | a oO —80 -—100 Secondary Arc Current at Cantwell , M MN YY YY VU oY 0.18 0.24 0.30 0.36 0.42 0.48 Time (s) T coo 0.54 0.60 0.66 1 0.72 Sol 106 compensating reactors are very large (phase L=4.0 H, neutral L=.67 H) and produce a large RL time constant with the series resistance of the line. If the d.c. offset is ignored, the secondary arc currents are still high even with this neutralization. This section of line may be too long to allow effective neutralization of the secondary arc current using the 4- reactor scheme. The method of using reactors to reduce the secondary arc current has a side effect of introducing inductive VARs (25 MVAR inductive in this case). A power flow run of this system with the same pre-fault operating conditions reported that SVC compensation at Teeland and Healy was 7.6 MVAR and 13.1 MVAR inductive. Installing an additional 12.5 MVAR (inductive) at each substation would upset the ratings of the SVCs at these buses as the capacitive VAR ratings would be decreased 12.5 MVAR each. Neutralization by using reactor banks would appear to be the most expensive option [4]. Based on the poor results obtained through modeling, unnecessary additional inductive VARs, and assumed high cost, the 4-reactor scheme does not appear feasible. The neutralization of secondary arc current by the use of high speed grounding switches produced better results (Table 3-5, Figure 3-16). Neutralization via grounding switches does not appear to reduce the secondary arc current below the recommended 20 amps rms for faults that occur close to a grounding switch at each end. Applying the grounding switches at each end of the line section after the phase is de-energized produces a closed loop between the open phase and the earth. Assuming a flat voltage profile and no fault on the line, a current i, would flow from one end of the line to the other due to the induced voltage from the in-service phases (Figure 3-17). Placing a line-ground fault on this grounded line produces currents that flow into the fault (i) and out of the fault (i), where i)+i7 =i. Moving the fault towards one end causes either i, or i to approach i, OI-€ an31y Secondary Arc Current at Cantwell With Neutralization — Grounding Switches 100 80 60 40 20 —20 —40 Current (Amps) —60 —80 -100 + ert it 0.18 0.24 0.30 0.36 0.42 0.48 0.54 0.60 0.66 0.72 Time (s) LOL LI-¢ eIns1y Circuit Breaker a = Circuit Breaker oN oo | HSGS { t J HSGS Instantaneous Current Flow in Open Phase With No Fault Circuit Breaker , 4 HSGS (5 2 / #scs _ = Breaker Instantaneous Current Flow in Open Phase With Fault 801 109 and the secondary arc current rises. The data reported in Table 3-5 show that with the SLG fault at Teeland, i,=51.6 A, and with the fault at Douglas, i,=13.0 A. The Teeland- Douglas line was segmented into four sections, each 6.5 miles long, and secondary arc current was measured at each node. This was done to try to determine how close in a fault could be and still achieve the recommended secondary arc current of 20 A or less. The secondary arc current 6.5 miles from Teeland was found to be 14.4 A. It is not understood why the secondary arc current is so high at the exact ends of the line yet is relatively small (14.4 A) only 6.5 miles from the end. This phenomenon may be caused by the line model used. High speed grounding switches do appear to be a feasible and less expensive option [4] for secondary arc neutralization on this line section. They have proven to be a very reliable means of secondary arc extinction in practice. A third alternative, using a sectionalizing switch at Cantwell, was investigated. After the line is de-energized at each end by the circuit breakers, a sectionalizing switch at Cantwell opens to segment the de-energized phase. It was thought that the secondary arc current could be reduced by shortening the length of the open phase through sectionalizing. This technique appears to work well as the data in Table 3-5 indicate. The highest secondary arc current on this line section occurs at Cantwell and is measured on the long (156 mile, Teeland-Cantwell) section of line (worst case, Figure 3-18). This method of sectionalizing could possibly be the least expensive of all neutralization methods to implement. With or without neutralization, the highest transient recovery voltage measured was 38.0 kV peak. This is a result of the relatively low line energizing voltage. As the maximum 8I-€ BINSIy Current (Amps) Secondary Arc Current at Cantwell With Neutralization — Sectionalizer 100 | bh oO 1 -—80 4 —100 i. 1 T 1 1 : i MAUL UU UMA AAA myveriiyvavivervireniien 0.18 0.24 0.30 0.36 0.42 0.48 0.54 0.60 0.66 0.72 Time (s) 1 Ol 111 TRV is less than one-third of the peak line-neutral voltage (112.7 kV), an arc restrike should not be a problem based on these simulations. 3.8 Heating of the Healy Unit Due to Negative Sequence Currents Single-pole reclosing would produce a large unbalance at the Healy 138 kV bus while a faulted phase is open. The unbalance would produce negative sequence currents that ~ could cause additional heating of the Healy generation unit. The effects of disconnecting the generator step-up transformer's phase A from the 138 kV bus were calculated. It was assumed that opening phase A is this manner would be a worst case condition. The unbalance produced when one line section phase is open should be less as the Healy 138 kV bus would remain connected either to the north or south. One per unit current was i assumed to be flowing in phases B and C of the transformer high side. The integrated product of machine negative sequence phase current (I) and time (t), 1,2t [13], was used to determine the heating effects. Using symmetrical components, the negative sequence current magnitude was calculated to be 0.5774 pu in the generator for this unbalance. The duration of the unbalance (dead-time, worst case) was 1.0 s. These figures produced 1,2t=.333. [13] lists a short-time capability figure of 1>2t=30 for a round-rotor, indirectly cooled machine. Heating due to negative sequence currents as a result of single-pole reclosing does not appear to be a problem for the Healy unit. 112 3.9 Electrical Torque of the Healy Unit During Single-Pole Reclosing Plots of the Healy unit electrodynamic torque (electrical torque) for different single-pole reclosing scenarios are shown in Figures 3-19 and 3-20. Figure 3-19 shows the resulting torque for a SLG fault at Cantwell with single-pole reclosing of the Teeland-Healy line section using sectionalizing. Figure 3-20 shows the electrical torque produced during a SLG fault at Nenana and a single-pole reclosing of the Healy-Gold Hill line section. The overall variation in torque during SPR is due to exciter control. These figures should be viewed only as estimates of the shaft torques experienced. The different shaft masses and shaft elasticity were not modeled in these simulations as these data were not available. Therefore, the shaft and its components were not accurately modeled. The large torque variations are due to the SLG fault and last for the duration of the fault (5 cycles). The figures also show a 120 Hz torque oscillation during the dead- time. The 120 Hz oscillation is produced by negative sequence currents flowing in the armature windings caused by the unbalanced condition. 3.10 Adaptive Reclosing Adaptive reclosing is a technique that uses a logical system to monitor a line section during the reclosing process. When adaptive reclosing is applied to single-pole reclosing the chances of reclosing into a fault are reduced substantially. Before a phase can be re- energized, the voltage on the three phases is monitored for uncleared faults. As an example, if |Va | is below a pre-determined open phase maximum voltage, it is assumed 6I-€ ANSI Healy Generator Electrical Torque SLG at Cantwell, Reclose TEE—HLS 0.20 Sectionalizer at Cantwell 0.15 0.10 0.05 Torque (Nm*1 0°) 0.3 0.4 0.5 Time (s) 0.6 O27, CUT O7-€ BNSLy Healy Generator Electrical Torque SLG at Nenana, Reclose HLS—GHS 0.20 FS nN) 0.15 O = ae & 0.10 —_ NY } @ 0.05 = or hens O 0.00 +H Ke —9)05 T aT T T T T T T T T T 7 0.1 0.2 0.3 0.4 0.5 0.6 0:7 vil 115 that the fault did not clear during the dead-time. Reclosing for phase A would be blocked, and phases B and C would be opened. If |val=|vb| or |Val=| Vc] then a line-line fault exists between phase A and another phase. Phase A reclosing is blocked and phases B and C are opened. Since this system prevents reclosing into a fault, the torque spike caused by re-energizing a faulted line is eliminated [17]. Adaptive reclosing could also be used to reduce the required dead-time. Instead of using a conservative fixed dead-time, the system could monitor the open phase and determine exactly when the secondary arc extinguished. The open phase could then be reclosed immediately. System stability would be improved because of the shorter dead-time. 3.11 Implementing Single-Pole Reclosing on the Intertie Applying single-pole reclosing to the Intertie would require the modification and replacement of existing equipment and the installation of new equipment. The circuit breakers currently used at Teeland, Healy, and Gold Hill would have to be replaced with independent-pole circuit breakers. Employing the sectionalizer secondary arc neutralization scheme at Cantwell would require the addition of a sectionalizer at Cantwell and relaying to control the sectionalizer. The sectionalizer relaying could probably be local to Cantwell and could monitor fault and phase opening events. New relays to control and supervise the single-pole reclosing of the line sections would be required at Teeland, Healy, and Gold Hill. It is understood that the SVC units will trip off-line if one phase is lost. The SVC units would have to be modified to prevent this from occurring during a single-pole reclosing operation. The SVC units are important to the stability of this system and must stay on-line during the single-pole reclosing process. Additional 116 work should be performed to determine the benefits of converting the SVCs to phase independent operation. 3.12 Conclusion The effects of single-pole reclosing on the Anchorage-Fairbanks Intertie have been investigated. Single-pole reclosing allows systems connected by a single transmission line to stay in synchronization while a single-line to ground fault is cleared. Currently all three-phases of the Intertie are tripped for all faults. Single-pole reclosing of the Intertie should provide improved Alaskan Railbelt power system security. No substantial considerations were given to the cost of implementing single-pole reclosing on the Intertie or the impact of future system upgrades. Secondary arc neutralization may be required on the Teeland-Healy line section. The Healy-Gold Hill line section produced secondary arc current values well below the recommended 20 A rms. Three methods to reduce the secondary arc current on the Teeland-Healy line section were explored. The method of sectionalizing the Teeland- Healy line section appears to provide the most benefits. The dead-time for the Teeland- Healy line section could be as long as 0.75 s. The dead-time for the Healy-Gold Hill line section should not be any longer than 0.6 s. Adaptive reclosing type control relays should be given serious consideration. This type of relay could sense when the secondary arc extinguished and could provide shorter dead-times and improved system stability. 117 The manufacturer of the Healy generation unit should be contacted regarding the implementation of single-pole reclosing near the unit. Recommendations for this particular unit should be obtained in the following areas: effects of impact torque due to reclosing, effects of harmonics due to reclosing on turbine blades, and negative sequence heating limits. REFERENCES [1] Kimbark, E.W., "Suppression of Ground-Fault Arcs on Single-Pole Switched EHV Lines by Shunt Reactors," JEEE Transactions on Power Apparatus and Systems, vol. PAS-83, pp. 285-290, March 1964. [2] Kimbark, E. W., "Selective-Pole Switching of Long Double-Circuit EHV Line," IEEE Transactions on Power Apparatus and Systems, vol. PAS-95, pp. 219-230, January/February 1976. (3] Blackburn, J. L., Protective Relaying: Principles and Applications, New York: Marcel Dekker, Inc., 1987, pg. 373. [4] Greenwood, A., Electrical Transients in Power Systems, New York: John Wiley and Sons, Inc., 1991, pg. 49. [5] Hasibar, R.M. et. al., "The Application of High-Speed Grounding Switches for Single-Pole Reclosing on 500 kV Power Systems," JEEE Transactions on Power Apparatus and Systems, vol. PAS-100, pp. 1512-1515, April 1981. [6] IEEE Working Group, "Single-Pole Switching for Stability and Reliability," Report of a panel discussion held at the 1984 PES Summer Meeting, JEEE Transactions on Power Systems, vol. PWRS-1, pp. 25-36, May 1986. {7] Haun, R.K., "13 Years' Experience with Single-Phase Reclosing at 345 kV," [EEE Transactions on Power Apparatus and Systems, vol. PAS-97, pp. 520-528, March/April 1978. 118 119 [8] Joyce, John S., T. Kulig, and D. Lambrecht, "The Impact of High-Speed Reclosure of Single and Multi-Phase System Faults on Turbine-Generator Shaft Torsional Fatigue," JEEE Transactions on Power Apparatus and Systems, vol. PAS-99, pp. 279-291, January/February 1980. [9] IEEE Working Group, "Effects of Switching Network Disturbances on Turbine- Generator Shaft Systems," Interim Report, JEEE Transactions on Power Apparatus and Systems, vol. PAS-101, pp. 3151-3157, September 1982. [10] Bowler, C.E.J., P.G. Brown, and D.N. Walker, "Evaluation of the Effect of Power Circuit Breaker Reclosing Practices on Turbine-Generator Shafts," JEEE Transactions on Power Apparatus and Systems, vol. PAS-99, pp. 1764-1779, September/October 1980. [11] IRBEE Working Group, "IEEE Screening Guide for Planned Steady-State Switching Operations to Minimize Harmful Effects on Steam Turbine-Generators," JEEE Transactions on Power Apparatus and Systems, vol. PAS-99, pp. 1519-1521, July/August 1980. [12] Gonzalez, A. J. et. al., "Effects of Single- and Three-Pole Switching and High- Speed Reclosing on Turbine-Generator Shafts and Blades," JEEE Transactions on Power Apparatus and Systems, vol. PAS-103, pp. 3218-3228, November 1984. [13] Kundur, P., Power System Stability and Control, New York: McGraw-Hill, Inc., 1994, pg. 1110. [14] Leuven EMTP Center, Alternative Transients Program Rule Book, Meverlee: LEC, 1987. [15] Electric Power Research Institute, Electromagnetic Transients Program (EMTP) Volume 4: Workbook IV (TACS), Palo Alto: EPRI, 1989, pp. 6.1-6.27. 120 [16] General Electric Industry Sales and Service, Healy SVC Starting on South System and Teeland SVC Starting on North System, Report to the Alaska Power Authority, Schenectady: General Electric, 1988. [17] El-Serafi, A.M., and S.O. Faried. "Effect of Adaptive Reclosing on Turbine- Generator Shaft Torsional Torques" JEEE Transactions on Power Systems, vol. 9, pp. 1730-1736, November 1994. APPENDIX I ELECTROMAGNETIC TRANSIENTS PROGRAM DATA FILE Cc C THIS IS A DATA FILE USED TO SIMULATE SPR ON THE TEELAND-HEALY LINE SECTION. C HIGH SPEED GROUNDING SWITCHES ARE USED FOR NEUTRALIZATION. C A SINGLE-LINE TO GROUND FAULT IS APPLIED AT CANTWELL. C IMPORTING THIS FILE INTO THE WORD PROCESSOR HAS CORRUPTED THE VITAL COLUMN C SPACING REQUIRED BY ATP. HOWEVER, ALL DATA CAN BE OBTAINED THROUGH THE USE OF THE C ATP RULEBOOK AND CLOSE EXAMINATION OF THIS FILE. Cc C SPR22.DAT - DATA FILE WITH S.M.59 AT HEALY WITH TACS EXCITER CONTROL AND C TACS MODELED SVC AT HEALY, GOLD HILL AND TEELAND. SVC MODEL IS A GENERIC C MODEL WHICH PROVIDES VERY GOOD RESULTS. THE "NEW" INPUT FILTERS AND C VOLTAGE REGULATOR AS DESCRIBED IN THE G.E. SVC STARTING STUDY WERE NOT C MODELED. DUE TO THE RMS INTEGRATORS THAT ARE USED TO MEASURE THE MONITORED C SOURCES WERE SET-UP IN SPR10.DAT WITH FIXED SVC VALUES FROM THE PTI "WN94" C CASE. HIGH SPEED GROUNDING SWITCHES, BPA SCHEME. 3-PHASE SVC REGULATION. C HEALY REGULATES TO THE GENERATOR TERMINALS. C INTERTIE SINGLE POLE RECLOSING STUDY C SWITCHING CONDUCTED AT TEELAND AND HEALY Cc BEGIN NEW DATA CASE Cc C *FIRST MISC. DATA CARD* CDELTAT TSTOP XOPT COPT EPSILN TOLMAT * * * #. * 1.0E-5 .75 0 0 C *SECOND MISC. DATA CARD* C IOUT IPLOT IDOUBL KSSOUT MAXOUT IPUN MEMSAV_ ICAT NENERG IPRSUP naan anne eM Ree oceee 100000 ‘5 0 0 0 0 0 1 c TACS HYBRID ¢€ C MODEL OF HEALY SVC - BASED ON EXAMPLE IN EMTP WORKBOOK IV, PGS. 6-21,29 Cc C *FIRING PULSE DETECTION* Cc C GET THYRISTOR (SWITCH) STATUS C PHASE A THYRISTORS 93THABIA 93THAB2A C PHASE B THYRISTORS 93THBCIA 93THBC2A C PHASE C THYRISTORS 93THCAILA 93THCA2A C NCxx IS 1 IF EITHER PARALLEL CONNECTED THYRISTOR IS OPEN 88NCAB =.NOT.(THAB1A.OR.THAB2A) 88NCBC =.NOT.(THBC1A.OR.THBC2A) 88NCCA =.NOT(THCAIA.OR.THCA2A) 121 ce C *VOLTAGE REGULATOR* (e C GET SVC PRIMARY TRANSFORMER VOLTAGES 90HLS13A 60.0 90HLS13B 60.0 90HLS13C 60.0 C CALCULATE PRIMARY RMS VOLTAGES 88VRMSA 66+HLS13A 60.0 88VRMSB 66+HLS13B 60.0 88VRMSC 66+HLS13C 60.0 C CALCULATE AVERAGE 88VAVG =(VRMSA + VRMSB + VRMSC)/239023.0 Gc 88VREFSV =1.02 88DELTAV =VAVG-VREFSV C EXPERIMENT WITH V1 BLOCK GAIN (1.5) Vl +DELTAV 1:5 1V2. +DELTAV 1.0 0.0 ©8601 1.0 001 c V3 ««+V1~ +2 1.0 -.2 .2 C EXPERIMENT WITH INTEGRATOR GAIN (300.0) 1INTDV +V3 300.0 0.0 1.0 1.0 00 1.0 C RELATE INTDV TO SIGMA THROUGH NON-LINEAR RELATIONSHIP BELOW 88SIGMA 56+INTDV 0.0 0.05 0129021432 0.1 0972693085 0.2 -2972693085 0.3 -6129021432 0.4 1.0 05 9999.0 Cc C *THREE PHASE FIRING PULSE GENERATION* 88FIRSAB58+UNITY +NCAB 60.0 1.0RESTAB 88FIRSBC58+UNITY +NCBC 60.0 1.0RESTBC 88FIRSCAS8+UNITY +NCCA 60.0 1.0RESTCA GC SECOAB -UNITY +SIGMA +FIRSAB SECOBC -UNITY +SIGMA +FIRSBC SECOCA -UNITY +SIGMA +FIRSCA (oe C THE -.0036 THRESHOLD CORRECTS FOR TIMING ERRORS IN TACS C CORRECTION=3*TIMESTEP*2*60 88THIRAB52+UNITY -.0036 SECOAB 88THIRBC52+UNITY -.0036 SECOBC 88THIRCAS2+UNITY -.0036 SECOCA C FIRING PULSES 11FPAB1 1.0 0013387 .0017387 11FPBC1 1.0 0041167 .0045167 122 11FPCA1 1.0 0068944 .0072944 Cc TOPSAB +FPAB1 +THIRAB TOPSBC +FPBC1 +THIRBC TOPSCA +FPCA1 +THIRCA C DEVELOPE RESET PULSES RESTAB -TOPSAB +UNITY RESTBC -TOPSBC +UNITY RESTCA -TOPSCA +UNITY C *PULSE SHAPING* C INTEGRATOR GAIN IS THE INVERSE OF THE TIME STEP (100000) 88FIRAB 158+TOPSAB 100.E3 1.0RTPAB 88FIRBC158+TOPSBC 100.E3 1.0RTPBC 88FIRCA158+TOPSCA 100.E3 1.0RTPCA C INTEGRATOR RESET DELAY 88HSMAB 54-FIRAB1 +UNITY .0003 88HSMBC 54-FIRBC1 +UNITY 0003 88HSMCA 54-FIRCA1 +UNITY 0003 ic RTPAB +HSMAB RTPBC +HSMBC RTPCA +HSMCA Cc C 33SIGMA VRMSA VRMSB VRMSC DELTAV Cc C NAME INITIAL Cee Pee TIVAVG 1.0 TIDELTAV 0.0 TISIGMA .25 Cc C *MODEL OF GOLD HILL SVc* Cc C *FIRING PULSE DETECTION* € C GET THYRISTOR (SWITCH) STATUS C PHASE A THYRISTORS 93THABIB 93THAB2B C PHASE B THYRISTORS 93THBC1B 93THBC2B C PHASE C THYRISTORS 93THCAIB 93THCA2B C.NCxx IS 1 IF EITHER PARALLEL CONNECTED THYRISTOR IS OPEN 88NCABGH =.NOT.(THAB1B.OR.THAB2B) 88NCBCGH =.NOT.(THBCIB.OR.THBC2B) 88NCCAGH =.NOT.(THCAIB.OR.THCA2B) ¢ C *VOLTAGE REGULATOR* c C GET 69kV BUS VOLTAGES 90GHS69A 60.0 123 90GHS69B 60.0 90GHS69C 60.0 C CALCULATE PRIMARY RMS VOLTAGES 88VRMSGA66+GHS69A 60.0 88VRMSGB66+GHS69B 60.0 88VRMSGC66+GHS69C 60.0 C CALCULATE AVERAGE 88VAVGGH =(VRMSGA + VRMSGB + VRMSGC)/119511.5 Cc 88VREFGH =1.02 88DELTGH =VAVGGH-VREFGH C EXPERIMENT WITH V1 BLOCK GAIN (1.5) VIGH +DELTGH 1.5 1V2GH +DELTGH 1.0 0.0 01 1.0 .001 Cc V3GH +V1IGH +V2GH 10 -2 2 C EXPERIMENT WITH INTEGRATOR GAIN (300.0) 1INTGH +V3GH 300.0 0.0 1.0 1.0 0.0 1.0 C RELATE INTGH TO SIGMAG THROUGH NON-LINEAR RELATIONSHIP BELOW 88SIGMAG56+INTGH 0.0 0.05 0129021432 0.1 0972693085 0.2 -2972693085 0.3 -6129021432 0.4 1.0 0.5 9999.0 Cc C *THREE PHASE FIRING PULSE GENERATION* 88FIGHAB58+UNITY +NCABGH 60.0 1.0REGHAB 88FIGHBC58+UNITY +NCBCGH 60.0 1.0REGHBC 88FIGHCAS8+UNITY +NCCAGH 60.0 1.0REGHCA Cc SEGHAB -UNITY +SIGMAG +FIGHAB SEGHBC -UNITY +SIGMAG +FIGHBC SEGHCA -UNITY +SIGMAG +FIGHCA Cc C THE -.0036 THRESHOLD CORRECTS FOR TIMING ERRORS IN TACS C CORRECTION=3*TIMESTEP*2*60 88THGHAB52+UNITY -.0036 88THGHBCS2+UNITY -.0036 88THGHCAS2+UNITY -.0036 C FIRING PULSES 11FPGAB1 1.0 0013387 .0017387 11FPGBC1 1.0 0041167 .0045167 11FPGCAL 1.0 0068944 .0072944 c GHPSAB +FPGABI! +THGHAB GHPSBC +FPGBC1 +THGHBC GHPSCA +FPGCA1 +THGHCA SEGHAB SEGHBC SEGHCA 124 C DEVELOP RESET PULSES REGHAB -GHPSAB +UNITY REGHBC -GHPSBC +UNITY REGHCA -GHPSCA +UNITY C *PULSE SHAPING* C INTEGRATOR GAIN IS THE INVERSE OF THE TIME STEP (100000) 88FIRAB258+GHPSAB 100.E3 1.0RTPGAB 88FIRBC258+GHPSBC 100.E3 1.0RTPGBC 88FIRCA258+GHPSCA 100.E3 1.0RTPGCA C INTEGRATOR RESET DELAY 88HSGMABS54-FIRAB2 +UNITY .0003 88HSGMBCS54-FIRBC2 +UNITY 0003 88HSGMCAS54-FIRCA2 +UNITY 0003 c RTPGAB +HSGMAB RTPGBC +HSGMBC RTPGCA +HSGMCA 77VAVGGH 1.0 77DELTGH 0.0 71SIGMAG -25 Cc Cc C *MODEL OF TEELAND SVC* Cc C *FIRING PULSE DETECTION* Cc C GET THYRISTOR (SWITCH) STATUS C PHASE A THYRISTORS 93THABIC 93THAB2C C PHASE B THYRISTORS 93THBC1C 93THBC2C C PHASE C THYRISTORS 93THCAIC 93THCA2C C NCxx IS 1 IF EITHER PARALLEL CONNECTED THYRISTOR IS OPEN 88NCABTL =.NOT.(THABIC.OR.THAB2C) 88NCBCTL =.NOT.(THBC1C.OR.THBC2C) 88NCCATL =.NOT.(THCAIC.OR.THCA2C) Cc C *VOLTAGE REGULATOR* c C GET 138kV BUS VOLTAGES 9OTEE13A 60.0 9OTEE13B 60.0 90TEE13C 60.0 C CALCULATE PRIMARY RMS VOLTAGES 88VRMSTA66+TEE13A 60.0 88VRMSTB66+TEE13B 60.0 125 88VRMSTC66+TEE13C 60.0 C CALCULATE AVERAGE 88VAVGTL =(VRMSTA + VRMSTB + VRMSTC)/239023.0 ¢ 88VREFTL =1.005 88DELTTL =VAVGTL-VREFTL C EXPERIMENT WITH V1 BLOCK GAIN (1.5) V1ITL +DELTTL 1.5 1V2TL +DELTTL 1.0 0.0 ©=—Ol 10 001 c V3TL +VITL +V2TL 10 -2 2 C EXPERIMENT WITH INTEGRATOR GAIN (300.0) 1INTTL +V3TL . 300.0 0.0 1.0 1.0 00 861.0 C RELATE INTTL TO SIGMAT THROUGH NON-LINEAR RELATIONSHIP BELOW 88SIGMATS6+INTTL 0.0 0.05 0129021432 0.1 0972693085 0.2 .2972693085 0.3 6129021432 0.4 1.0 0.5 9999.0 : Cc C *THREE PHASE FIRING PULSE GENERATION* 88FITLABS8+UNITY +NCABTL 60.0 1.0RETLAB 88FITLBC58+UNITY +NCBCTL 60.0 1.0RETLBC 88FITLCAS8+UNITY +NCCATL 60.0 1.0RETLCA Gc SETLAB -UNITY +SIGMAT +FITLAB SETLBC -UNITY +SIGMAT +FITLBC SETLCA -UNITY +SIGMAT +FITLCA c C THE -.0036 THRESHOLD CORRECTS FOR TIMING ERRORS IN TACS C CORRECTION=3*TIMESTEP*2*60 88THTLAB52+UNITY -.0036 SETLAB 88THTLBC52+UNITY -.0036 SETLBC 88THTLCAS2+UNITY -.0036 SETLCA C FIRING PULSES 11FPTAB1 1.0 0013387 .0017387 11FPTBC1 1.0 0041167 .0045167 11FPTCA1 1.0 0068944 .0072944 ic TLPSAB +FPTAB1 +THTLAB TLPSBC +FPTBC1 +THTLBC TLPSCA +FPTCA1 +THTLCA C DEVELOPE RESET PULSES RETLAB -TLPSAB +UNITY RETLBC -TLPSBC +UNITY RETLCA -TLPSCA +UNITY C *PULSE SHAPING* 126 C INTEGRATOR GAIN IS THE INVERSE OF THE TIME STEP (100000) 88FIRAB358+TLPSAB 100.E3 1.0RTPTAB 88FIRBC358+TLPSBC 100.E3 1.0RTPTBC 88FIRCA358+TLPSCA 100.E3 1.0RTPTCA C INTEGRATOR RESET DELAY 88HSTMABS54-FIRAB3 +UNITY 0003 88HSTMBCS54-FIRBC3 +UNITY .0003 88HSTMCAS4-FIRCA3 +UNITY .0003 Cc RTPTAB +HSTMAB RTPTBC +HSTMBC RTPTCA +HSTMCA TIVAVGTL 1.0 77DELTTL 0.0 TISIGMAT 2 c C *HEALY EXCITER MODEL* Cc C TRANSFER FUNCTION BLOCKS: INPUT FILTER BLOCK C NAME + IN1+ IN2+ IN3+ IN4+ INS GAIN F-LO F-HI N-LO N-HI Cc a eS ee ee eee ee ee en oe, a ee | 1 VF + VT 1.0 C COEFFICIENTS: AMPLIFIER BLOCK Cc NO N1 N2 N3 N4 NS N6 N7 c OD D1 D2 D3 D4 DS D6 D7 * C TRANSFER FUNCTION BLOCKS: AMPLIFIER BLOCK C NAME + IN 1+ IN2+ IN3+ IN4+ INS GAIN F-LO F-HI N-LO N-HI C -----#--* -----# 4 _---# #_--# ¥_-* # R Pscainsil Bamuel IO cst Bscsal * 1 VA - VF- VFD+ VREF 20.0 3 1.0 C COEFFICIENTS: AMPLIFIER BLOCK Cc NO Nl N2 N3 N4 NS N6 N7 Cc bo D1 D2 D3 D4 DS D6 D7 1 15 C TRANSFER FUNCTION BLOCKS: EXCITER BLOCK C NAME + IN1+ IN2+ IN3+ IN4+ INS GAIN F-LO F-HI N-LO N-HI Ca a NP Wc WN Bc Wisc Bcc W® 1 DVE+ VA 1.0 C COEFFICIENTS: EXCITER BLOCK Cc NO N1 N2 N3 N4 NS N6 N7 Cc bo D1 D2 D3 D4 DS D6 D7 127 C TRANSFER FUNCTION BLOCKS: FEEDBACK BLOCK C NAME + IN 1+ IN2+ IN3+ IN4+ INS GAIN F-LO F-HI N-LO N-HI C W--- 2 2 nnn * *. * * * 1 VFD + DVE 03 C COEFFICIENTS: FEEDBACK BLOCK Cc NO NI N2 N3 N4 NS N6 N7 Cc bO D1 D2 D3 D4 DS D6 D7 1 65 C TRANSFER FUNCTION BLOCKS: SUM DVF AND UNITY C NAME + IN1+ IN2+ IN3+ IN4+ IN5 GAIN F-LO F-HI N-LO N-HI Cie c Se, VE + DVE+UNITY ¢ C ACQUIRE TERMINAL VOLTAGES FROM ATP USING TACS DEVICES C NAME A B c T-START T-STOP Cc — Miccrnnmnemsnenal Msccamsnssl a, I incon scnsciitaecnsnnsemcncaiaammenil Pecneneiyed * 9O0HLGENA 60.0 90HLGENB 60.0 90HLGENC 60.0 Cc C CALCULATE P.U. RMS AVERAGE TERMINAL VOLTAGE Cc C NAME = FREE FORMAT FORTRAN EXPRESSION (oo a a a ce 88 VT =(((ABS(HLGENA)+ABS(HLGENB)+ABS(HLGENC))*SQRT(3/2))/13800)*.523598776 99 VREF =1.0 c C NAMEI NAME2 NAME3 NAME4 NAMES cS anne C33 VE ec C SET TACS INITIAL CONDITIONS C SOME INITIAL CONDITIONS FOUND BY RUNNING SIMULATION WITHOUT DISTURBANCE C TO STEADY STATE. C NAME INITIAL C= Mciasaitaull * 77 VE 1.3 TIHLGENA = 11267.7 TTHLGENB ~ -5633.9 TTHLGENC = -5633.9 77 VT 1.0 77 VA 65 77 DVE 30 77 VFD — .0031 71 VF 1.0 Cc C *POWER METER* C 90TEE13A 60.0 C91TMS13A 60.0 C91HLSSIA 60.0 C 88PINSTT =TEE13A*TMS13A 128 C 88PINSTH =HLS13A*HMS13A C 1PTEE +PINSTT 120.0 Cc 1.0 Cc 0.0 1.0 C 1PHLS +PINSTH 120.0 Cc 1.0 Cc 0.0 1.0 C 88PDTEE 53+PTEE -00833 C 88PDHLS 53+PHLS -00833 C PAVGT +PTEE -PDTEE C PAVGH +PHLS -PDHLS C 33PAVGT PAVGH BLANK CARD ENDING TACS CONTROL Cc C NOW START WITH ATP DATA CASE Cc C *BRANCH DATA CARD* Cc C TRANSMISSION LINES: C MODELED AS A DISTRIBUTED PARAMETER, C LUMPED RESISTANCE, FREQENCY INDEPENDENT LINE. C BUS ABUSBBUSABUSBR L C LENG * * * *. *. * * C 4-6 C FOR LINES WITH NO ZERO SEQUENCE CAP. DATA Zco=.5Zc+ C 230 kV LINE: C LINE DATA: PT. MAC 230 - TEELAND 230 -1IMAC23ATEE23A 0.505 8.75 .004 20.00 -2MAC23BTEE23B 0.153 2.68 .018 20.00 -3MAC23CTEE23C C 138 kV LINES: C LINE DATA: TEELAND 138 - DOUGLAS 138 -1TESSIADOU13A 447 6.93 .007 26.0 -2TEE13BDOU 13B -169 2.05 .015 26.0 -3TEE1 3CDOU13C C LINE DATA: DOUGLAS 138 - CANTWELL 138 -1DOU13ACWS13A -386 5.80 .009 129.7 -2DOU13BCWS13B 050 1.62 .018 129.7 -3DOU13CCWS13C C LINE DATA: CANTWELL 138 - HEALY 138 -1CWS13AHLSS1A -386 5.80 .009 39.32 -2CWS13BHLS13B 050 1.61 .019 39.32 -3CWS13CHLS13C C LINE DATA: HEALY 138 - NENENA 138 -1HLS13ANNS13A 454 6.79 .007 56.23 -2HLS13BNNS13B 168 2.11 .014 56.23 -3HLS13CNNS13C C LINE DATA: NENANA 138 - ESTER 138 -INNS13AETS13A 454 6.79 .007 43.47 -2NNS13BETS13B 1682.11 .014 43.47 -3NNS13CETS13C C LINE DATA: ESTER 138 - GOLD HILL 138 -1ETS13AGHS13A 456 6.79 .007 3.63 -2ETS13BGHS13B -168 2.12 .014 3.63 -3ETS13CGHS13C 129 130 C LINE DATA: GOLD HILL 138 - FORT WAINWRIGHT 138 -1GHS13AFWS13A 403 6.92 .007 10.34 -2GHS13BFWS13B -118 1.96 .015 10.34 -3GHS13CFWS13C Cc C LOADS REPRESENTED AS A INDUCTANCE REFLECTED HIGH. ¢c REFERENCE C BUS1 BUS2 BUSI BUS2 R L C Ca Wrsccssssal PRiscmud Pccomal Picasa! Biccasmed a * C TEELAND 34.5: 2.4 MW/1.1 MVAR TEE34A 486.8 TEE34B 486.8 TEE34C 486.8 TEE34A 2817.2 TEE34B 2817.2 TEE34C 2817.2 C DOUGLAS: 3.5 MW/.6 MVAR DOU13A 5560.1 DOU13B 5560.1 DOU13C 5560.1 DOU13A 8.60E4 DOU13B 8.60E4 DOU13C 8.60E4 C HEALY: 5.1/1.8 HLS13A 3887.2 HLS13B 3887.2 HLS13C 3887.2 HLS13A 2.92E4 HLS13B 2.92E4 HLS13C 2.92E4 C NENENA: 1.6/.6 NNS13A 1.23E4 NNS13B 1.23E4 NNS13C 1.23E4 NNS13A 8.69E4 NNS13B 8.69E4 NNS13C 8.69E4 C ESTER: 1.8/.5 ETS13A 1.09E4 ETS13B 1.09E4 ETS13C 1.094 ETS13A 1.04ES ETS13B 1.04ES ETS13C 1.04E5 Cc Cc C HARMONIC FILTERS/SHUNT CAPACITORS AT TEELAND CSTH TLSVSA 35.7 7.89 TLSVSB 35.7 7.89 TLSVSC 35.7 7.89 C7TH TLSVSA 32.0 4.50 TLSVSB 32.0 4.50 TLSVSC 32.0 4.50 C 11TH TLSVSA 32.3 1.80 TLSVSB 32.3 1.80 TLSVSC 32.3 1.80 Cc C HARMONIC FILTERS/SHUNT CAPACITORS AT GOLD HILL C 5TH - 1OMVAR GHSVSA 0.0 6.33 44.7 GHSVSB 0.0 6.33 44.7 GHSVSC 0.0 6.33 44.7 C7TH - 23 MVAR GHSVSA 0.0 1.38 104.7 GHSVSB 0.0 1.38 104.7 GHSVSC 0.0 1.38 104.7 e€ C HARMONIC FILTERS/SHUNT CAPACITORS AT HEALY CSTH-9MVAR HLSVSA 06 5.31 53.0 HLSVSB 06 5.31 53.0 HLSVSC 06 5.31 53.0 C7TH - 13 MVAR HLSVSA 04 1.83 78.3 HLSVSB 04 1.83 78.3 HLSVSC 04 1.83 78.3 Cc C THYRISTOR CONTROLLED REACTORS AT GOLD HILL RGHCABGHSVSB 39.9 RGHCBCGHSVSC 39.9 RGHCCAGHSVSA 39.9 C THYRISTOR CONTROLLED REACTORS AT HEALY REACABHLSVSB 20.8 REACBCHLSVSC 20.8 REACCAHLSVSA 20.8 C THYRISTOR CONTROLLED REACTORS AT TEELAND RTLCABTLSVSB 34.4 RTLCBCTLSVSC 34.4 RTLCCATLSVSA 34.4 Cc C SNUBBERS ACROSS REACTORS AT GOLD HILL RGHCABGHSVSB 26600. RGHCBCGHSVSA 26600. RGHCCAGHSVSC 26600. C SNUBBERS ACROSS REACTORS AT HEALY REACABHLSVSB 13867. REACBCHLSVSC 13867. REACCAHLSVSA 13867. C SNUBBERS ACROSS REACTORS AT TEELAND REACABTLSVSB 22933. REACBCTLSVSC 22933. REACCATLSVSA 22933. C RESISTORS TO PREVENT CURRENT HOGGING AT GOLD HILL GHSVSATHABIB 001 RGHCABTHAB2B -001 131 GHSVSBTHBC1B .001 RGHCBCTHBC2B 001 GHSVSCTHCAIB -001 RGHCCATHCA2B .001 C RESISTORS TO PREVENT CURRENT HOGGING AT HEALY HLSVSATHABIA -001 REACABTHAB2A 001 HLSVSBTHBCIA .001 REACBCTHBC2A .001 HLSVSCTHCA1A -001 REACCATHCA2A -001 C RESISTORS TO PREVENT CURRENT HOGGING AT TEELAND TLSVSATHABIC 001 RTLCABTHAB2C 001 TLSVSBTHBCIC 001 RTLCBCTHBC2C 001 TLSVSCTHCAIC 001 RTLCCATHCA2C -001 C SNUBBERS ACROSS THYRISTORS - HEALY THAB1ATHAB2A 500.E3 4.E-5 THBC1ATHBC2A 500.E3 4.E-5 THCA1ATHCA2A 500.E3 4.E-5 C SNUBBERS ACROSS THYRISTORS - GOLD HILL THABIBTHAB2B 500.E3 4.E-5 THBC1IBTHBC2B 500.E3 4.E-5 THCAIBTHCA2B 500.E3 4.E-5S C SNUBBERS ACROSS THYRISTORS - TEELAND THAB1CTHAB2C 500.E3 4.E-5 THBCICTHBC2C 500.E3 4.E-5 THCAICTHCA2C 500.E3 4.E-5S c C TRANSFORMERS MODELED AS IDEAL, NO SATURATION OR LOSS AS THESE DATA WERE C NOT AVAILABLE. C 3-PHASE TRANSFORMERS MODELED AS SINGLE PHASE UNITS CONNECTED AS A C BANK. ZERO SEQUENCE FLUX IS AFFORDED A RETURN PATH. MAY NOT BE AN C ACCURATE MODEL OF A 3 LEGGED CORE TYPE TRANSFORMER. ce C TEELAND 230:138:13.8, GY-GY-DE 60 MVA TRANSFORMER Cx TRANSFORMER TL23PA 9999 Cc owned a a Bonen! cua a * 1TEE23A 3.015 263.2 132.8 2TEE13A -762 -3.86 79.85 3TLSVSATLS VSB 117 1.56 13.8 C PHASE B ---------------------------------- TRANSFORMER TL23PA TL23PB 1TEE23B 2TEE13B 3TLSVSBTLSVSC C PHASE C --------------------------------—~ TRANSFORMER TL23PA TL23PC 1TEE23C 2TEE13C 132 3TLSVSCTLSVSA Cc c C TEELAND 230:115:34.5, GY-GY-DE 250 MVA AUTO 9999 1TEE23A -11 65.7 132.8 2TEE11A -08 -1.23 66.35 3TEE34A TEE34B 93 12.0 34.5 C PHASE B ---------------------------------- TRANSFORMER TLAUPA TLAUPB 1TEE23B 2TEE11B 3TEE34BTEE34C C PHASE C ---------------------------------- TRANSFORMER TLAUPA TLAUPC 1TEE23C 2TEE11C 3TEE34CTEE34A Cc C GOLDHILL 134.55:69:13.8, GY-GY-DE 60 MVA AUTO TRANSFORMER GHSPHA 9999 Cc ell ieadl anteineetacnedamiadasnall 1GHS13A 362 67.8 77.7 2GHS69A -157 -1.41 39.8 3GHSVSAGHSVSB 057 1.40 13.8 C PHASE B ---------------------------------- TRANSFORMER GHSPHA GHSPHB 1GHS13B 2GHS69B 3GHSVSBGHSVSC C PHASE C ---------------------------------- TRANSFORMER GHSPHA GHSPHC 1GHS13C 2GHS69C 3GHSVSCGHSVSA Cc c C HEALY GENERATION 134.55:13.2, GY:DE, 30 MVA One tS Ie TRANSFORMER HLSPHA 9999 1HLS13A 1.51 99.8 77.7 2HLGENAHLGENB 044 2.88 13.2 C PHASE B -- TRANSFORMER HLSPHA HLSPHB 1HLS13B 2HLGENBHLGENC C PHASE C -- a= TRANSFORMER HLSPHA HLSPHC 133 1HLS13C 2HLGENCHLGENA Cc C HEALY SVS 138:12,GY:DE, 20 MVA TRANSFORMER HSVSPA 9999 1HLS13A 1.47 97.2 79.7 2HLSVSAHLSVSB 033 .832 12.0 C PHASE B ---------------------------------- TRANSFORMER HSVSPA HSVSPB 1HLS13B 2HLSVSBHLSVSC C PHASE C ---------------------------------- TRANSFORMER HSVSPA HSVSPC 1HLS13C 2HLSVSCHLSVSA (e C EQUIVALENT SYSTEM IMPEDANCES: C FT. WW 138 kV SIF38RXAFWS13A 5.14 162.0 52F38RXBFWS13B 22.49 200.8 53F38RXCFWS13C C GOLD HILL 69 kV 51G69RXAGHS69A 2.37 37.0 52G69RXBGHS69B 4.30 34.9 S3G69RXCGHS69C C PT. MAC 230 kV 51P23RXAMAC23A 4.88 108.1 52P23RXBMAC23B 9.25 97.2 53P23RXCMAC23C C TEELAND 115 kV SITISRXATEE11A 1.30 49.6 52T1SRXBTEE11B 4.20 43.4 53TISRXCTEE11C iC BLANK CARD ENDING BRANCH DATA ( C RECLOSING SWITCHES FOR TEELAND - HEALY LINE Cc C BUSABUSB TCLOSE TOPEN IE V FLASH SPECIAL eee Runa ommend enn Een ——— # C HEALY HLSI3AHLSSIA_— -1.0 183333 HLSI3AHLSSIA 68333 —:10.0 CNENANA TEEI3ATESSIA = -1.0 .183333 TEE13ATESS1A_ .68333 10.0 Cc C HIGH SPEED GROUNDING SWITCHES HLSS1A 185 67 TESSIA 185 67 (e C FAULTING SWITCH - SINGLE LINE TO GROUND AT CANTWELL 134 CWS13A 10.67 Cc C TACS CONTROLLED SWITCHES (THYRISTORS) C ANODECATHO Vig Ihold = Tdeion C PHASE A THYRISTORS - HEALY CLOSED SAM GRID O/C EO 11THAB1 AREACAB FIRAB1 11THAB2AHLSVSA FIRAB1 C PHASE B THYRISTORS - HEALY 11THBC1AREACBC FIRBC1 11THBC2AHLSVSB FIRBC1 C PHASE C THYRISTORS - HEALY 11THCA1AREACCA FIRCA1 11THCA2AHLSVSC FIRCA1 C PHASE A THYRISTORS - GOLD HILL 11THABIBRGHCAB FIRAB2 11THAB2BGHSVSA FIRAB2 C PHASE B THYRISTORS - GOLD HILL 11THBC1IBRGHCBC FIRBC2 11THBC2BGHSVSB FIRBC2 C PHASE C THYRISTORS - GOLD HILL 11THCAIBRGHCCA FIRCA2 11THCA2BGHSVSC FIRCA2 C PHASE A THYRISTORS - TEELAND 11THABICRTLCAB FIRAB3 11THAB2CTLSVSA FIRAB3 C PHASE B THYRISTORS - TEELAND 11THBCICRTLCBC FIRBC3 11THBC2CTLSVSB FIRBC3 C PHASE C THYRISTORS - TEELAND 11THCAICRTLCCA FIRCA3 11THCA2CTLSVSC FIRCA3 C MEASURING SWITCHES TEE13ATMS13A MEASURING BLANK CARD ENDING SWITCH DATA Cc C IDEAL VOLTAGE SOURCES C NAME VI AMPLITUDE FREQUENCY TIME-0 Al C -----# 6 oF +. ooo ene eB nanan +... C FT. WAINWRIGHT 138 kV EQUIVALENT: (-41.9) 14F38RXA 0 115380.8 60.0 -43.3 -l. 10.0 14F38RXB 0 115380.8 60.0 -163.3 -l. 10.0 14F38RXC 0 115380.8 60.0 -283.3 -1. 10.0 C GOLD HILL 69 kV EQUIVALENT: (-45.4) 14G69RXA0 549298 60.0 -44.6 -1. 10.0 14G69RXB 0 549298 60.0 -164.6 -l. 10.0 14G69RXC 0 54929.8 60.0 -284.6 -1. 10.0 C PT. MAC 230 kV EQUIVALENT: (-7.0) 14P23RXA 0 190798.9 60.0 -68 -1. 10.0 14P23RXB 0 190798.9 60.0 -126.8 -1. 10.0 14P23RXC 0 190798.9 60.0 -246.8 -1. 10.0 C TEELAND 115 kV EQUIVALENT: (-15.2) 14TISRXAO 91361.9 60.0 -15.1 -l. 10.0 14T1SRXBO 91361.9 60.0 -135.1 -l. 10.0 TSTOP 136 14TISRXCO 91361.9 60.0 -255.1 “1. 10.0 c C HEALY GENERATION CS.M.59 - VALUES ENTERED ARE FOR THE 26 MW HEALY PLANT C ALL DATA IS FACTORY SUPPLIED EXCEPT THE FOLLOWING WHICH COMES FROM THE C PTI MODEL FOR HEALY: Xq, Xq’, Xdi", Xq", Tqo’, Tqo" C XI WAS CALCULATED BY MULTIPLYING .95 TIMES Xo, AS RECOMMENDED IN THE ATP C RULEBOOK. C REACTANCES MEASURED AT RATED CURRENT WERE USED. C THE NEUTRAL GROUNDING TRANSFORMER IMPEDANCE WAS SET TO ZERO AS THESE DATA WERE CNOT AVAILABLE. ic C CLASS 1 CARD: ONE CARD FOR EACH PHASE (3 TOTAL) G C BUS TERMVOLT FREQ ANGLE (-50.7) * S9HLGENA 114065 60.0 -50.7 S9HLGENB) 11406.5 60.0 -170.7 S9HLGENC 114065 60.0 -290.7 Cc C CLASS 2 CARDS: DEFINES MACHINE ITERATION TOLERANCES AND PARAMETER FITTING c G EPSUBA EPOMEG EPOGEL NIOMAX C——— I c FM C= PARAMETER FITTING 1.0 ¢c C CLASS 3 CARDS: DEFINES MACHINE PARAMETERS Cc CKMKE NP SMOUTP SMOUTQ RMVA_ RKV AGLINE _ SI S2 C -*-*_--*--------— iciirtinll eee Dimond lame Oe * 11 2 29.412 13.8 152.5 .1470 .5200 Cc Cc ADI AD2) AQIi AQ2 AGLQ_ SIQ_ S82Q C-------=' — el ial C—O Diiecunal Gael _——— * 10 12 10 12 -1 Cc iC Ra Xl Xd Xq Xd’ Xqi Xd" Xq' C— = Wiccccsncmeasll Pccumummall Wi cicusniananl Picncmonl ee, Di ccrnmmuiaaal a * 00147) 0475, «1.79 1.74 8.26 ©7025 20 ic C Tdo' Tqo' Tdo" Tgo" Xo Rn Xn —— Msmenpuan Mcntenmnnsansd DP ccnamsamnestl Pinned Pcmmvnd a * 3.6 30.033 .065 05 00 0.0 ic C CLASS 4 CARD: DEFINE SHAFT MASS PARAMETERS Cc EXTRS HICO DSR DSM_ HSP_ DSD Cc ae ee — eo ! * aa ceed * 1 1.0 .029404 BLANK CARD ENDING S.M. 59 MASS DATA Cc CGA Nl N2 N3 N4 N5 N6 N7 N8 NO NI1O Nil N12 137 1 14 BLANK CARD ENDING S.M. 59 OUTPUT REQUESTS Cc C CLASS 6 CARD: DEFINE CONTROL FUNCTIONS FOR TACS. ENTER 71 INCOLUMN C 1-2 FOR EXCITER, 72 FOR GOVENOR (MECHANICAL POWER CONTROL), 73 FOR OTHER C VARIABLES LISTED ON PAGE 8-22,23,24 OF ATP RULEBOOK. Cc BUS KI C -----*-----¥__* 71 VE FINISH BLANK CARD ENDING SOURCES C *SELECTIVE NODE VOLTAGE OUTPUT CARD* C NAMEI NAME2 NAME3 NAME4 NAMES NAME6 NAME7 NAME8 NAME9 ECT.=> CWS13ADOU13AHLSS1ATESS1A BLANK CARD ENDING NODE VOLTAGE OUTPUT BLANK CARD ENDING THE CASE APPENDIX II DIFFERENCES BETWEEN TRANSIENT STABILITY PROGRAMS AND THE ELECTROMAGNETIC TRANSIENTS PROGRAM Transient stability is the ability of a power system to maintain synchronism during a severe disturbance and then return to a steady-state operating point. Severe disturbances that can affect transient stability include line-switching operations, loss of generation, or a sudden loss of load. Power system equipment electromechanical phenomena are the primary interest of a transient stability study. These phenomena generally occur in the time frame of 0.01-10.0s. Transient stability programs use simplified models to represent portions of the simulated power system. Balanced three-phase systems are represented. Therefore, just the positive sequence network is modeled. Only synchronous frequency voltages and currents are assumed in the simulated system. For this reason, d.c. offset currents and harmonic components are ignored. Transmission lines and most other branch elements are represented with linear models (i.e. t-equivalent transmission line models). Generating machines are represented with a classical model. However, to ease simulation, the Thevenin equivalent and voltage source of the classical model is replaced by an Norton equivalent with a controlled current source. Most transient stability programs provide multiple generating machine models of varying complexity. Machine dynamics (electromagnetic and electromechanical properties) are modeled as external control elements and are used to regulate the controlled current source to provide an accurate 138 139 machine representation. Differential equations that represent system behavior are solved through numerical integration. Typical transient stability simulations provide the user with many system values. These values are generally displayed graphically. The Electromagnetic Transients Program (EMTP) and Alternative Transient Program (ATP) are used to study the electromagnetic phenomena of a power system. Power system electromagnetic phenomena generally occur in the time frame of 1 1s to 500 ms. While the study of electromagnetic behavior of a power system is the principal application of the EMTP, the current versions allow for the modeling of many electromechanical properties as well. The EMTP provides a three-phase representation of a power system. It is therefore ideal for studying unbalanced conditions. Many of the modeling simplifications made in transient stability programs can not be applied to the study of power system electromagnetic behavior. Therefore, the EMTP models generally tend to be more complex than transient stability models. One area were the EMTP and transient stability program models differ greatly is in the modeling of transmission lines. Various line models exist in the EMTP to match the requirements of the study being conducted and the data available. It was mentioned earlier that linear models are used to represent transmission lines in transient stability programs. While a m-equivalent model is available in the EMTP, it is generally only used to represent short lines. Transmission lines are represented in most EMTP studies with the traveling wave models. The more complex traveling wave models are of the frequency dependent type. The assumption of only a synchronous frequency existing in a power system is not valid in electromagnetic transient studies. The frequency dependent transmission line models 140 acknowledge the fact that the resistance and reactance of a transmission line vary with frequency. This phenomenon is particularly important in studies were harmonics may be generated or are of interest. Transformer modeling also varies significantly between transient stability programs and the EMTP. Transformers are generally modeled as a lumped impedance branch in a transient stability program. In the EMTP transformers are most commonly represented with an ideal transformer model that includes saturation and magnetizing resistance. The impedance is specified for each winding. A three-phase, three-leg, core type transformer model is also available. Electrical machines are represented either by ideal voltage sources or one of two machine models in the EMTP. The EMTP machine models are very similar to the models used in transient stability programs (with dynamics). The two EMTP models differ in that one model is used to represent three-phase synchronous machines (S.M.59) and the other is used to represent other rotating equipment (Universal Machine). The S.M. 59 model is generally used to represent generating machines. The Universal Machine is used to represent various types of motors. Both models allow some external controlling inputs. Control systems can be modeled with the Transient Analysis of Control Systems (TACS) in the EMTP. With TACS the user can model almost any control system used in electrical power systems. Typical uses include the modeling of a exciter/voltage regulator to control an §.M. 59 model. TACS can be used to control switches in the EMTP to simulate thyristors used in high voltage d.c. converters or static var compensators. 141 The EMTP can produce both a text summary file of the simulation and/or a graphical file. A separate post processor/graphics package is used to display and manipulate the data and produce hard copy output.