Posts Tagged ‘Blackout’

Phasor Measurement Units (PMU)

Real time system monitoring is a relatively new tool available to power system operators, allowing them to analyze all aspects of a large power system continuously.  Phasor measurement units are the leading technology behind the newfound ability to provide instant analysis for problems in a geographically enormous power system.  Using a synchronous clock based off GPS timing signals, phasor measurement units (PMU) can very accurately measure current and voltage phasors with almost no time difference between meters [1].  This allows for real time information regarding power angles and power flow, system status, and possible problems.  With phasor measurement refreshing frequencies as high as 60Hz, the synchronized clock time of all PMUs is an essential requirement for providing accurate information to data centers, where a delay of several microseconds could lead to corrupt results.  Because of the timing constraints required by PMUs to perform properly, PMUs now abide by the IEEE standard C37.118, which defines standardized measurement methods, timing tolerances and communication channels.

Wide Area Monitoring Systems (WAMS)

The widespread implementation of PMUs has led to wide area monitoring systems (WAMS) allowing for situation reports for large parts of the transmission system.  PMUs transmit this system information continuously to data centers where computers can record and monitor the state of the entire power system, performing actions to maximize power flow while maintaining system stability [2].

Fault detection and proper relay functioning is one of the most important tasks in transmission systems, accounting for approximately 70% of all major disturbances  [3].  By using PMUs and having snapshots of the entire system updated up to 60 times per second, the detection of faults is very quick.  By having all of this data available at a centralized location, coordination of relays during faults can be optimized for the situation, resulting in the best fault clearing schemes.

2003 North-eastern Blackout

Wide area monitoring systems have been integrated into many transmission systems globally, allowing transmission operators to have continuous real time information about the state of the transmission system.  WAMS technology plays an important role in generation and protection, allowing generating facilities to observe system conditions continuously and maximize their output for different loading scenarios.  The eastern North American WAMS recorded all major transmission system information during the 2003 blackout, and provided critical data for the reconstruction of the sequence of events leading up to the blackout [4].  However, the lack of intelligent control schemes left the system incapable to react quickly enough to maintain stability, resulting in the loss of power to millions of people.

The 2003 North American blackout showed the world and the grid operators that the conventional power system ideas put in place over 100 years ago are not sufficient for the complex and continually growing power system of modern times.  The blackout left over 50 million people without power, and was caused by the incorrect tripping of transmission lines and generation facilities [5].  Investigations conducted revealed that the cause of the cascading blackout was due to distance relays operating within Zone 2 and Zone 3, with preset calculations.

The problem originated when the Midwest Independent System Operator (MISO) had a problem with its state estimator [6], and the information gathered from the eastern WAMS due to offset in PMU sampling times [4].  The state estimator is responsible for indicating potential problems with system parameters and operations, without this tool operators were unaware of the initial problems leading up to the blackout.  After initial transmission trips to which operators were not aware, overloading of remaining lines caused them to reach thermal limits, ground faulting through trees.  After losing several large transmission lines, distance relays started operating incorrectly in zone 3, seeing low impedance due to high load current and low voltage from tripped generation capacity.  Had the WAMS been operating correctly, the initial problematic events that lead to the eventual blackout would have been identified, and problems could have been corrected before the situation elevated to such severe levels.

Open Phasor Data Concentrator (OpenPDC)

Phasor Data Concentrators (PDC) are devices distributed throughout the transmission system designed to collect data from the many phasor measurement units.  Due to the high volume of data collected, each node typically collects data from only five or six individual PMUs and forwards the data to concentrator devices.

In October 2009, the Tennessee Valley Authority (TVA) released data collection software for industry use called SuperPDC (Super Phasor Data Concentrator) [7], which is responsible for aggregating measurements from multiple PDCs and archiving measurements for subsequent event analysis.  It is now available under an open source license under the name openPDC.

This software allows the TVA to collect data from its 120 online PMUs (see Phasor Measurement Unit Map) that together measure almost two thousand parameters several times per second.  In all, the TVA archives 150 million measurements per hour (requiring 36 GB of storage space per day) [8].

Tennessee Valley Authority

In conjunction with graduate students from Washington State University (WSU), the Tennessee Valley Authority collected data from its PMUs to observe local area oscillations within the system during a major switching event.  During a planned switching of 500kV transmission lines at the Cumberland Fossil Plant (CUF), the system experienced a dangerous undamped local-mode power oscillation at 1.2Hz, which continued until operators detected the problem and reversed the switching three minutes later.  At its peak, the oscillations escalated up to a 700MW variation in transmitted power (see Cumberland Fossil Plant Oscillation Event).

Without the phasor measurement units in place, detection of this nearly catastrophic event would not have been possible and the system could have suffered a total collapse.  It remains unknown whether the power system stabilization (PSS) equipment was not yet installed or otherwise out of service during the fault [9].  Fortunately, but both local- and inter-area oscillations can be detected using this method and the software is available for immediate use by any utility [10].

Electric Reliability Council of Texas

The amount of power transfer over transmission lines is limited by the thermal limit of the line, putting constraints on profits and maximum generation capacity.  Lines ratings are typically set to conservative constant values for the sake of safety and reliability, but newer technologies are enabling utilities to vary equipment ratings based on environmental factors including humidity and ambient temperature.

When the Electric Reliability Council of Texas (ERCOT) implemented dynamic rating of their transmission lines, they were able to maximize utilization of existing infrastructure, which had direct financial benefit for bulk generation facilities exporting power.  A control system used data including current atmospheric conditions, forecasted temperatures and system loading to determine the maximum power transfer limits, which usually exceeds the constant ratings given by the manufacturer [11].

[1] Yilu Liu, Lamine Mili, Jaime De La Ree, and Reynaldo Francisco Nuqui, “State Estimation and Voltage Security Monitoring Using Synchronized Phasor Measurement,” Virginia Polytechnic Institute and State University, Blacksburg, VA, PhD Dissertation 2001.
[2] Charles Proteus Steinmetz, “Complex Quantities and Their Use in Electrical Engineering,” in Proceedings of the International Electrical Congress, Chicago, Illinois, 1893, pp. 33-74.
[3] Yi Zhang, M. Prica, M.D. Ilic, and O.K. Tonguz, “Toward Smarter Current Relays for Power Grids,” in IEEE Power Engineering Society General Meeting, Montreal, QC, 2006, p. 8.
[4] J.F. Hauer, N.B. Bhatt, K. Shah, and S. Kolluri, “Performance of “WAMS East1″ in Providing Dynamic Information for the North East Blackout of August 14, 2003,” in IEEE Power Engineering Society General Meeting, Denver, CO, 2004, pp. 1685-1690.
[5] A.P. Apostolov, “Distance Relays Operation During the August 2003 North American Blackout and Methods for Improvement,” in IEEE Russia Power Technology, St. Petersburg, Russia, 2005, pp. 1-6.
[6] Pacific Northwest National Laboratory, Electricity Infrastructure Operations Center (EIOC). (2010, March) Looking back at the August 2003 blackout. [Online].   http://eioc.pnl.gov/research/2003blackout.stm
[7] Tennessee Valley Authority. (2009, October) TVA Opens Data Collection Software for Industry Use. [Online].   http://www.tva.gov/news/releases/octdec09/data_collection_software.htm
[8] Tennessee Valley Authority. (2010) openPDC Introduction. [Online].   http://openpdc.codeplex.com/
[9] Gary Kobet, Ritchie Carroll, Ryan Zuo, and Mani V. VEnkatasubramanian. (2009, October) Oscillation Monitoring System at TVA. [Online].  http://www.naspi.org/meetings/workgroup/2009_october/presentations/kobet_tva_oscillation_monitoring_tools_20091008.pdf
[10] openPDC Extensions. (2010, March) Extensions to the openPDC software, including WSU’s Oscillation Monitoring System (OMS). [Online].   http://openpdc.codeplex.com/wikipage?title=Extensions&referringTitle=Home
[11] Kyeon Hur et al., “High Wire Act: ERCOT Balances Transmission Flows for Texas-Size Savings Using Its Dynamic Thermal Ratings Application,” IEEE Power and Energy Magazine, vol. 8, no. 1, pp. 37-45, January-February 2010.

One of my partners wrote the majority of this article for a report submitted to ECE4439: Conventional, Renewable and Nuclear Energy, taught by Professor Amirnaser Yazdani at the University of Western Ontario. It is included here for completeness with the rest of the articles. I edited the article and wrote the sections entitled: Open Phasor Data Concentrator (OpenPDC) and Tennessee Valley Authority.

Read Full Post »

Historical events provide the greatest indication of our need for a more flexible, more intelligent and more reliable power system.  In the Western world, the Tennessee Valley Authority’s bulk transmission system has achieved five nines of availability for ten years (ended 2009) [1], which corresponds to under 5.26 minutes of outage annually.  However, while the grid is generally robust to disturbances, catastrophic events like the 2003 North-eastern Blackout serve as a solemn reminder of the fragility of our system, susceptible to cascading outages originating from a handful of preventable failures in key parts of the system.  More concerning is the increasing incidence of widespread outages: in the US, 58 outages affected over 50,000 customers from 1996 to 2000 (an average of 409,854 customers per incident), compared to 41 occurrences for the same number of customers between 1991 and 1995 [2].

The essence of smart grid technology is the provision of sensors and computational intelligence to power systems, enabling monitoring and control well beyond our current capabilities.  A vital component of our smart grid future is the wherewithal to detect a precarious situation and avert crisis, either by performing preventative maintenance or by reducing the time needed to locate failing equipment.  Moreover, remotely monitoring the infrastructure provides the possibility of improvements to the operational efficiency of the power system, perhaps through better routing of electric power or by dynamically determining equipment ratings based on external conditions such as ambient temperature or weather.

In the face of changing requirements due to environmental concerns as well as external threats, it is becoming extraordinarily difficult for the utility to continue to maintain the status quo.  As the adoption of plug-in [hybrid] electric vehicles intensifies, the utility must be prepared for a corresponding increase in power consumption.  The transition to a more intelligent grid is an inevitable consequence of our ever-increasing appetite for electricity as well as our continued commitment to encouraging environmental sustainability.

The deregulation of the electric power system also presents new and unique challenges, since an unprecedented number of participants need to coordinate grid operations using more information than ever before.  If we are to maintain the level of reliability that customers have come to expect from the power system, we must be able to predict problems effectively, rather than simply react to them as an eventuality.

As the grid expands to serve growing customer demands as well as a changing society, we must proceed cautiously to ensure the system preserves its reputation of reliability.  It is incumbent upon us to carefully analyze past events and implement appropriate protection and control schemes using modern technologies.  It is clear that the power system of tomorrow will depend upon the design and preparation we conduct today.

[1] Tennessee Valley Authority. (2010, March) TVA Transmission System. [Online].   http://www.tva.gov/power/xmission.htm
[2] M. Amin, “North America’s electricity infrastructure: are we ready for more perfect storms? ,” Security and Privacy, IEEE, vol. 1, no. 5, pp. 19-25, September-October 2003.

I originally wrote this article for a report submitted to ECE4439: Conventional, Renewable and Nuclear Energy, taught by Professor Amirnaser Yazdani
at the University of Western Ontario.

Read Full Post »