WECC Joint Synchronized Information Subcommittee Report to WECC
WECC Joint Synchronized Information Subcommittee Report to WECC PCC and OC October 2014 Presented by Dmitry Kosterev 1
Western Interconnection Synchrophasor Program (WISP) • Phase 1 WISP - $108 M investment completed – WECC JSIS supports reporting to DOE • Phase 2 WISP: – DOE awarded a new $6. 2 M grant to Peak RC, total of $12. 4 M with cost-share from BPA, CISO, SCE, SDG&E, IPC – DOE $2. 9 M are awarded to PG&E with the recipient cost-share of additional $3. 9 M 2
WECC JSIS Structure • Data Delivery and Management • Engineering Analysis Applications • Control Room Applications 3
Data Delivery and Management Dates: Entity 8/25/2014 -8/31/2014 Percent Availability PMU Count AESO APS BCHA BPA IPCO LDWP NVE NWE PAC PGAE PNM SCE SDGE SRP TEPC TSGT WAPA 94. 09% 94. 99% 81. 76% 99. 86% 83. 13% 99. 58% 69. 14% 62. 91% 73. 19% 0. 00% 75. 86% 38. 07% 99. 86% 99. 89% 100. 00% 98. 45% 99. 81% 17 17 11 55 20 15 6 3 4 2 6 7 4 1 2 2 5 Overall: 90. 13% 177 4
Data Delivery and Management • WISP 2. 0 will focus on improving data quality • Applications are developed for analysis of data availability and latency • Applications are being developed for analysis of data validity and data calibration 5
Engineering Analysis • Oscillation Analysis • Power Plant Model Validation • Frequency Response Analysis • System Model Validation 6
Oscillation Analysis • A paper is completed on modes of inter-area oscillations in the West • PDCI probing tests and brake tests: – 2013 report is completed – 2014 tests are done, report is under development • Analysis applications – “Prony Robot” is developed to calculate and tabulate oscillation frequency, damping and shape for 1, 000’s of simulation runs – Applications for ambient data analysis are available, training required 7
Oscillation Detection Application scans power and voltage at interties, power plants, PDCI, SVCs for growing or sustained high energy oscillations Events captured in 2014: Wind power plant control system oscillation at 14 Hz Generator PT failure resulting in oscillation (about 1 Hz) Hydro plant operating in rough zone for extended period of time (about 0. 33 Hz) 8
Oscillation Detection BEFORE AFTER 9
Power Plant Model Validation • PPMV plays in disturbance data into GE PSLF • A state-of-the art data management systems are developed • Technology transfer is in progress (BPA, IPC, PG&E, SRP) • Cost effective approach for compliance with MOD -026 and MOD-027 Reliability Standards 10
Frequency Response Analysis Tool • Frequency Response Analysis Tool (FRAT) works at – Interconnection (done) • Balancing Authority (done) – Power Plant / Generator (under development) • Methodology is consistent with NERC BAL-003 -1 • Adding capabilities to generate compliance reports • An interconnection frequency response baseline is developed from 2008 to today 11
Database of Events Add / View / Edit Events Performance Baseline 12
NERC-BAL-003 -1 Reliability Standard • FERC Approved BAL-003 -1 Frequency Response and Frequency Bias Setting, effective April 1, 2015 • BAL-003 -1 requires no UFLS for “design” event: – “Design Event” is 2 Palo Verde – WECC Coordinated UFLS starts at 59. 5 Hz • Western Interconnection i. FRO is about 950 MW per 0. 1 Hz – Recalculated annually • Frequency Response Measure (FRM), measured at Point B 13
NERC-BAL-003 -1 Reliability Standard FA FB FC FA = pre-disturbance frequency (average from -16 to 0 sec) FB = settling frequency (average from 20 to 52 sec) 14 FC = minimum (nadir) frequency
Western Interconnection Performance 15
NERC-BAL-003 -1 Reliability Standard • i. FRO is prorated among Balancing Authorities based on annual generation and served load energy • Annual median BA FRM needs to exceed BA FRO 16
BA Frequency Response Calculation FA FB FC Pint_B Pint_A FRM = (Pint_B – Pint_A + BA Gen Loss) / ( FA – FB) 17
Balancing Authority Performance Interchange response is measured for compliance with NERC BAL-003 -1 Generation response is calculated to determine how much frequency response to acquire 18
NERC-BAL-003 -1 Reliability Standard • BAL-003 -1 applies to Balancing Authorities • BAs need to acquire frequency response • FERC held a conference on frequency response acquisition in April 2014 • Formation of Reserve Sharing Groups is allowed • Trading of FRO is not supported by NERC Staff • Transmission reservations and impact on SOLs ? 19
May 16, 2014 RAS Event 20
May 16, 2014 AC RAS Event Frequency 60. 05 60 59. 95 NW RAS = 2, 563 MW John Day – Grizzly #1 back in service Kemano = 110 MW 59. 9 59. 85 Springerville = 445 MW 59. 8 59. 75 59. 7 23: 18: 00 23: 18: 15 23: 18: 30 23: 18: 45 23: 19: 00 23: 19: 15 23: 19: 30 23: 19: 45 23: 20: 00 23: 20: 15 23: 20: 30 23: 20: 45 21
May 16, 2014 Sequence of Events • Prior to event: – John Day-Grizzly #2 and Grizzly-Malin #2 500 -k. V lines were out of service • 23: 18: 15: – John Day-Grillzy #1 500 -k. V lines tripped to lock-out – AC RAS gen. drop is initiated of about 2, 500 MW at Grand Coulee, Wells, John Day, Biglow Canyon Wind and Dooley Wind – Chief Joseph braking resistor is operated • Few seconds later: – Kemano RAS in BC operated on power swing and tripped unit # 1 with 110 MW • 40 seconds later: – Springerville unit#4 in Arizona tripped with ~425 MW due to boiler instability • 23: 20: 30: – John Day-Grizzly #1 500 -k. V line is restored back to service 22
Model Validation Studies • Validation Base Case Development – WECC Planning base case is modified to match system conditions – Generation MWs and statuses are taken from Peak RC WSM – Area loads are adjusted to match major paths flows • Wind generation – May 16 event had significant amount of wind generation – EMS wind generation models are inadequate for most of WECC system – Many wind generators in California have no dynamic models • Sequence of events is simulated in GE PSLF • WISP data from BPA, SCE, LADWP and SRP is used 23
Model Validation Studies Remarkable agreement of frequency response 24
Chief Joseph Brake • Chief Joseph brake operated as designed • In May 16 event, Chief Joseph brake had little impact on COI stability • However, the brake operation aggravated the Kemano power swing • There were no dynamic models for Chief Joseph brake – BPA developed one since, will be seeking approval at next MVWG • BPA Planning to initiate review of security of Chief Joseph brake control algorithm 25
Kemano RAS • There is no dynamic model for Kemano RAS – Kemano owners to provide dynamic RAS models • There is no visibility of Kemano dynamic performance – Kemano owners to install and network PMUs at Kemano and Kittimat • Kemano owners to review the security of RAS algorithm 26
Springerville • Springerville unit #4 tripped due to boiler instability, although the frequency was well within PRC-024 Reliability Standards • Springerville would have tripped for 2 Palo Verde outage • 2 PV+Springerille would aggravate the COI voltages and power pick-up • Springerville owners are looking into solutions 27
May 26, 2014 RAS Event 28
May 26, 2014 RAS Event PDCI Converter 1 failure triggered PDCI RAS gen. drop of 2, 806 MW Frequency drop to 59. 66 Hz, well above UFLS levels of 59. 5 H 29
Baselining Frequency Deviations PDCI RAS Events 30
- Slides: 30