Jay Giri
THE OBJECTIVES OF REAL-TIME GRID MANAGEment (RTGM) are to prevent widespread blackouts, ensure the grid is operating within safe limits, and to basically “keep the lights on” for all customers.
Today, there is an increasing growth of the following:
Furthermore, the grid infrastructure is aging and more prone to failure. All these factors contribute to greater operational uncertainties that make it more difficult to “keep the lights on.”
This article describes the evolution of RTGM and how it is adapting to meet today’s challenges. It also describes other recent, complementary functions deployed across the grid that also aid in blackout prevention.
The electric power grid consists of millions of equipment components—it is one of the largest and most complex realtime engineering machines in existence. Customer demand for electricity changes second by second. Whenever demand changes, generation must react immediately to maintain system frequency. If demand increases, generation must increase; if demand decreases, generation must decrease. Changes in load and generation result in power flows on transmission lines that are unpredictable because electricity follows the path of least electrical resistance, as dictated by the laws of physics. So, power flows in the grid are always in a state of flux.
The electric power grid is one of the largest and most complex real-time engineering machines in existence.
The objective of RTGM is to ensure that the grid is operating within safe limits, system frequency stays close to normal, customer demand is being met, and, of course, to help prevent cascading blackouts. If a blackout does occur, the objective also becomes restoration of power to customers as soon as possible. Customers are tolerant of outages of less than a few hours. Implementing these operational objectives requires a vast infrastructure of hardware and software solutions.
Since the 1950s, utilities have deployed a single, centralized control center for RTGM. The primary control center function is the Energy Management System (EMS). The devastating 1965 U.S. northeast blackout prioritized the need for an EMS at every utility. Since then, EMS capabilities have continually grown and evolved to meet nascent challenges.
Figure 1 provides an overview of an EMS control center. It consists of many big screen wall displays, operator stations, display monitors, trend charts, audible alarms, and communications consoles. The operating station at the bottom of the figure is the supervisor; the others are for monitoring the transmission system, communicating with generating stations, coordinating switching operations with distribution system operators, and negotiating interchange scheduling contracts with neighbors.
The primary function of the control center is the EMS. Measurements from substations are received and displayed at the EMS every few seconds. These include frequency, circuit breaker statuses, voltages, and power flows in transmission lines and transformers. These measurements are asynchronous, have no time tags, variable latencies, and unknown inaccuracies. The EMS applications and analytics are specifically designed to handle these not-perfect, asynchronous measurements.
The grid operating paradigm has traditionally been “reactive.” Decisions are made in reaction to observing current grid conditions. It is like driving a car by looking just at the dashboard. “Driving by dashboard” becomes particularly challenging when the road ahead is not necessarily straight, flat, or clearly visible, which is the state of grid operations today.
The EMS operator’s primary focus is situational awareness (SA). SA is defined as perception of elements in the environment, comprehension of their meaning, and projection of their status in the near future. If a problem needs to be fixed, the operator can remotely implement changes to grid equipment, communicate with generation and substation operators, and coordinate with neighboring EMS operators.
EMS operator tasks include
Figure 2 shows subsystems and applications of the EMS. Applications on the left are EMS operator functions, while applications on the right are EMS engineer functions.
Contingencies are outages of transmission lines, transformers, generators, loads, entire substations, or communication channels.
Supervisory control and data acquisition (SCADA) was the first EMS application. SCADA retrieves measurements from the field every 2–4 s and is our “eyes into the grid.”
Without SCADA, the Rest of the EMS Functions Cannot Function
SCADA also allows operators to implement remote control actions, such as change circuit breaker status, change the tap changers on transformers, and so on. SCADA issues alarms if limit violations are detected. The SCADA Load Shed function can immediately curtail predefined customer loads during emergencies to arrest a cascading blackout. SCADA also records and stores historical real-time data for future retrieval and analysis. SCADA helps with blackout prevention by identifying problems in the grid and shedding load if required.
Automatic generation control (AGC) is the only closed-loop application at the EMS. It consists of
LFC changes generat ion in response to load. ED determines the optimal generation schedule to minimize overall utility cost. Some utilities have replaced ED with an interface to the Market Management System (MMS). The MMS processes bids from generation providers and selects the most economical bid schedules.
AGC also ensures that contractual interchange agreements are met. A load forecast function maintains a demand forecast,
hourly for the next week. AGC runs every 4–10 s, while ED and MMS updates are received every minute or so. AGC helps with blackout prevention by ensuring that system frequency is stable and kept close to normal.
These are off-line versions of the generation applications and are used for studies by operators and engineers.
State Estimation State estimation (SE) provides a comprehensive, holistic view of grid conditions. It provides a “best guess” of current grid conditions using imperfect SCADA measurements with a network model of the grid. The model represents a single utility plus portions of immediate neighbors.
SE calculates the “best guess” of system conditions at every substation, including substations not measured by SCADA. SE also provides overload alerts and identifies bad SCADA measurements.
SE is a complex, large-matrix solution and SE typically runs every 10–30 s. SE helps with blackout prevention by providing a holistic view of grid conditions and identifies problem areas and parts of the grid that are islanded.
A contingency analysis (CA) uses the SE solution to perform a series of “what-if” studies to determine the grid’s resilience to potential vulnerabilities. Contingencies are outages of transmission lines, transformers, generators, loads, entire substations, or communication channels.
Managing the growing uncertainties of the future grid will require creative, innovative solutions. A CA processes a list of predefined contingencies and calculates their postcontingency steady-state conditions. The contingencies are prioritized based on severity of impact, and the operators decide whether immediate action is required.
One monitor is typically always dedicated to display CA results. A CA is a computationally intensive function as it involves many successive power flow solutions (one for each contingency). A CA runs at some multiple of the SE run cycle. A CA helps with blackout prevention by proactively identifying future contingencies that will cause problems.
The optimization applications include
OPF determines generation basepoints to minimize transmission losses. It is used during stable grid conditions to minimize cost. SENH processes potentially harmful contingencies to determine corrective and preventive actions that will alleviate their impacts. OPF and SENH are advanced analytics and run at some multiple of the SE run cycle. SENH helps with blackout prevention by providing corrective actions for potential future contingencies.
Study network functions are offline versions of EMS network applications and used for studies by operators and engineers.
Figure 3 shows the real-time sequence of the EMS applications and their typical execution cycle rates in seconds.
The EMS applications consist of millions of lines of software code.
EMS Applications Run 24/7, With a Remarkable 99+% Reliability
SCADA displays of substations are used by the operator to monitor grid conditions and perform supervisory control actions. Similarly, other EMS applications have displays from which the operator can monitor and manage the grid. The EMS consists of hundreds of display pages. Some are customer specific, and some are standard vendor displays.
Activities On a typical day, operator actions are rout ine, procedural, and uneventful, and 90% of the time this is the case. When an emergency is detected, the operator’s focus zooms in on identifying the problem and taking corrective actions. Most emergencies are not major and can be handled locally by the utility. Some are major system-wide emergencies and will need coordination with neighbors as they could lead to a widespread blackout. These “blackout-threatening situations” happen relatively infrequently, but when they do, they have severe consequences, with many customers losing electricity.
The grid operator is like an airport’s air traffic controller: routine tasks most of the time, occasional small emergencies, and less frequent major emergencies. It is imperative that operators are continually trained to handle these emergency situations as time is of the absolute essence to prevent a cascading blackout.
On a normal day, the operator’s tasks are routine:
During emergencies, the operator’s tasks become more urgent:
During major system-wide emergencies, all control center personnel are mobilized:
• Coordinate with neighbors on a joint corrective plan
• Coordinate with substation and generation operators
• Engage training simulator (TS) engineers to determine corrective plans
Implement actions to prevent a cascading blackout
The gr id equipment components modeled in the EMS are
The EMS network assumes that all three phases are all equally loaded and is represented as a single-phase, positivesequence model.
Figure 4 depicts the “node-breaker model” used to represent substations. “Nodes” are junctions of components, shown by numbers. “Breakers” are circuit breakers that connect equipment, shown as a rectangular box named breaker.
The complete EMS network model is the aggregation of all substations’ node-breaker models.
The EMS network model is customer specific and built by engineers using tools provided by the EMS vendor. The nodes and breakers are then linked to corresponding SCADA measurements.
Models are also built for applications in the other EMS subsystems: generation, SCADA, TS, and system engineering. Building all these models is a huge task and entails many engineer months of effort.
The components in the node-breaker models are defined with specific parameters, such as resistance, reactance, capacitance, and so on. These components include transmission lines, transformers, shunt capacitors and reactors, phase-shifting transformers, high-voltage dc lines, flexible ac transmission systems (FACTS) devices, and so forth. Each component has a steady-state model for stable grid conditions and a dynamic model for disturbed grid conditions.
A dynamic security assessment (DSA) simulates the dynamic behavior of the grid when a fault or unplanned event occurs. Dynamic models of equipment are used in DSA.
A DSA charts the specific path taken, with a subsecond resolution, from a pre-event steady-state condition to the postevent steady-state condition. The transition path may traverse a region that results in instability or that triggers subsecond relays. This will change operating conditions prior to reaching the eventual postevent state. A DSA identifies the vulnerabilities caused by transient and voltage instabilities.
Voltage stability is typically performed every 5 min and transient stability typically every 60 min.
The function of the TS is to train operators on typical daily activities. The TS is also called the dispatcher TS or operator TS. It is an offline function and is not a part of the EMS’ real-time sequence. The primary function of the TS is to
• Normal daily operations
• Emergencies that lead to a blackout
• Restoration after a blackout.
Additionally, the TS is used to
The TS consists of the following subsystems:
• Power flow simulation, which provides “SCADA measurements” every 4 s
• Dynamic simulation, which assumes uniform system frequency every second
• To schedule events and disturbances for a training scenario
• To build emergency blackout scenarios.
Today’s TSs are replacing PSS with a transient stability dynamic simulation with millisecond resolution. This is essential for today’s advanced EMS control centers, which are described later.
Large electrical interconnections consist of many individual electric utilities. It is important to coordinate one utility’s actions with neighbors as they are electrically connected and could be affected by your actions.
The benefits of interutility cooperation and information exchange include
• better transparency across utility borders
• better wide area decision making
Managing the growing uncertainties of the future grid will require creative, innovative solutions. The road ahead is less predictable, less visible, and much more uncertain. We cannot just drive “by looking at the dashboard every few seconds” anymore.
Some practical, promising solutions to manage these uncertainties are described next.
Look-ahead simulations will help operators “see” what is coming up so that they can act proactively to avert problems.
The TS can be used to quickly run multiple future scenarios, and to provide best- and worst-case projections.
These projections provide the foresight needed to develop corrective actions ahead of time to bypass situations that lead to a blackout.
Since 2009, thanks to U.S. Department of Energy funding from the American Recovery and Reinvestment Act, there has been a sudden spurt in the deployment of a new type of grid synchrophasor measurement technology across U.S. utilities.
Synchrophasor measurements are subsecond grid measurements from phasor measurement units (PMUs). They include three-phase magnitude and phase angles of voltages and currents, frequency, and rate of change of frequency. They have precise timestamps and are measured at rates of 30–120 measurements per second. They track the subsecond dynamic behavior of the grid.
Wide area monitoring systems (WAMS) have been implemented and use these synchrophasor measurements to provide valuable additional insights to the operator. WAMS implementations have grown in the United States and also worldwide. WAMS dramatically enhance traditional EMS capabilities.
WAMS Are the Most Seismic, Beneficial Step Change in RTGM Capabilities in Control Center History
Synchrophasors: EMS-WAMS Today, many EMS control centers receive PMU data. An immediate benefit is that grid stress can be immediately identified by simply monitoring PMU voltage phaseangle differences between two different locations.
Figure 5 illustrates the modern EMS (EMS-WAMS) control center functions. They include the new subsecond PMU measurements and PMU analytics, which augment traditional EMS capabilities.
The functions on the left are of a traditional EMS. The functions on the right are WAMS applications. WAMS do not need a model of the power grid; the actual power grid is the model.
WAMS subsecond analytics include
The benefits of an EMS-WAMS include
Figure 6 shows a depiction of how Pacific Gas & Electric (PG&E) enhanced its EMS with WAMS-PMU solutions.
This EMS-WAMS has been successfully deployed at the PG&E testbed facility in San Ramon, California. It includes an advanced TS with a transient stability engine to t rain operators on the new subsecond PMU applications.
The Indian national grid is one of the largest interconnections in the world and has an installed capacity of more than
400 GW.
Figure 7 shows the four levels of the Indian EMS hierarchy: national (entire country), regional (groups of neighboring states), state, and substation (within a state). They consist of approximately 500 computers that gather, analyze, display, and exchange data across the country.
In 2013, India began deploying more than 1,800 PMUs. These PMUs provide unprecedented subsecond visibility of real-time conditions across the entire country. It truly is mind-boggling to see these updates across the nation serving more than a billion customers.
Figure 8 presents an overview of India’s PMU deployments across the country.
WAMS are transitioning to become wide area management systems. The EMS has traditionally been the heartbeat of the grid.
Figure 9 shows how WAMS can become the new heartbeat as the primary source of data to many other utility functions. WAMS promises to dramatically improve our abilities to prevent blackouts.
Many other solutions have also been deployed to enhance grid management and help blackout prevention. Most are not at the high-voltage transmission grid level, but at the lower-voltage distribution level. They include
They are briefly described next. Remedial Action Schemes RAS, also known as special protection systems, have been widely used to automatically trigger predesigned protection schemes when certain adverse transmission grid conditions are detected.
Each RAS is a custom heuristic scheme and is developed as follows:
Customized RAS have been in service for decades and have helped avert cascading blackouts.
The prospect of building an online, real-time RAS is truly exciting. Here, subsecond PMU measurements would be used to detect a problem, and smart intelligence would trigger appropriate subsecond controls to protect the grid. The controls include FACTS and other similar thyristor-based controls. This is depicted as the “Wide Area Controls” function in Figure 9.
Millions of protective relays have been deployed across the grid for almost a hundred years, with the objective of immediately isolating and protecting grid equipment during emergencies. It is finally time to deploy a real-time protective relay for the “most expensive grid equipment of all,” the grid itself.
Real-Time RAS is the “Protective
Relay for the Grid”
This is not a trivial development. Nevertheless, it is an unprecedented gigantic step toward vastly improving prevention of blackouts.
The DMS manages the lower-voltage distribution grid in real time by providing monitoring, analytics, and visualization capabilities.
In North America, distribution systems are unbalanced and require a three-phase analysis. So, the DMS consists of a suite of three-phase monitoring and analytics. The analytics include fault location, fault isolation, and intelligent switching schemes to quickly restore power to affected customers.
The DMS improves management of the lower-voltage grid, which in turn helps improve reliability of the higher-level transmission grid.
OMS are used after a blackout to identify affected customers and restore power more quickly to them. The OMS expedites restoration of power to customers after a blackout.
GMS solutions manage a portfolio of fleets of generation.
GMS are also being developed to manage the uncertainties associated with renewable energy and distributed energy resources. Better uncertainty management at the generation level will improve grid reliability and the ability to avert blackouts.
Substation automation solutions (SAS) are advanced monitoring, intelligence, and automation solutions implemented at substations. When problems are detected locally, controls at the substation can be deployed to contain the problems and mitigate their propagation to the rest of the grid.
Also, by applying basic Kirchhoff current and voltage laws (a DMS SE), a valid, coherent set of substation measurements can be created for transmission to the EMS. This “sanitized” consistent set of data will significantly improve EMS SE performance, which in turn helps the rest of the EMS network applications run reliably. SAS result in improved grid management and help mitigate blackouts.
DR programs provide incentives to customers to encourage them to participate in voluntary short-term load reduction during peak demand periods. Shaving the peak demand avoids startup of expensive generators like gas turbines.
It is not too farfetched to imagine that during a blackoutimminent emergency, DR could be called upon to shed load to prevent a blackout.
DERMS manage smaller, closer-to-the-load, generation resources. These include renewables, storage devices, electric vehicle energy, and distributed and modular resources. They are connected to the low-voltage distribution network.
DERMS manage and optimize a portfolio of distributed energy resources, which will reduce propagation of their variabilities and uncertainties to the transmission grid. This in turn will help assist with blackout prevention.
Microgrids are growing at the lower-voltage distribution level. Microgrids consist of local generation and storage, serving local customers. For the most part, they are selfsustaining. By and large, they do not rely on the transmission grid and hence are protected against transmission grid problems. Nevertheless, during microgrid emergencies, they draw upon the transmission grid for support. So instead of the transmission grid being their “lifeblood,” it becomes just a “lifeline.”
Microgrids include smart cities, smart villages, smart campuses, smart buildings, smart communities, and so on.
Well-managed, self-sustaining microgrids benefit the EMS because they are predictable most of the time. For transmission grid emergencies, microgrids could potentially be called upon to reverse roles and provide support back to the transmission grid. Microgrids can be beneficial toward avoiding blackouts.
Grid interconnection consists of millions of components. It is managed in real time by many independent electric utilities. Each utility’s EMS is responsible for its own portion of the grid and also for helping neighbors during emergencies to prevent wide area blackouts.
Uncertainties in grid operations have been growing at an unprecedented rate. The growth of renewable energy resources, new stakeholders, and new programs poses increasing grid management challenges. These new deployments are not designed with an objective to help prevent blackouts, so the threat of a blackout increases.
Hence, the EMS operator is gradually losing control and the ability to “keep the lights on.” Traditional EMS capabilities will need to be enhanced significantly to manage these new challenges.
Fortunately, new PMU measurements and analytics have recently been integrated at control centers. PMU analytics dramatically enhance EMS capabilities to manage these new challenges. They provide unprecedented subsecond visibility to augment SA, and to help the operator make more informed quick decisions to prevent blackouts.
New monitoring, intelligence, and control solutions are also being implemented at substations and at the distribution system. These decentralized solutions are a groundswell of valuable additional support to aid in blackout prevention.
To summarize, we have a plethora of solutions deployed to ensure grid reliability. These solutions continue to evolve to meet future challenges. Unfortunately, it is not easy to quantify how many blackouts have actually been averted in real life.
To be successful in the future, we need to continue to attract smart minds, work cooperatively, leverage technology advances, promote research and innovation, and learn from successes in other similar industries.
Considering the immense complexity and size of the electricity supply chain, our track record for “keeping the lights on” has been very remarkable indeed! Kudos to electric utilities worldwide!
“Advancement of synchrophasor technology in ARRA projects,” U.S. Department of Energy, Washington, DC, USA, DOE ARRA Rep., Mar. 2016. [Online]. Available: https://www.smartgrid.gov/files/documents/20160320_Synchrophasor_Report.pdf
G. Bakke, The Grid. Bloomsbury, London, U.K., 2016. [Online]. Available: https://www.bloomsbury.com/us/grid-9781632865687/
Wiley, Hoboken, NJ, USA. Smart Grid Handbook. (2016). [Online]. Available: https://www.wiley.com/en-us/Smart+Grid+Handbook%2C+3+Volume+Set-p-9781118755488
Jay Giri retired as director of GE Grid Software Solutions and is now an independent consultant based in Redmond, WA 98053 USA.
Digital Object Identifier 10.1109/MPE.2023.3247090
Date of current version: 19 April 2023
1540-7977/23©2023IEEE