Thomas J. Overbye, University of Illinois, & James D. Weber
PowerWorld Corp. and Energy Visuals, Inc.
The electric power system is now one of the largest and most complex of man-made objects, with billions of components, tens of millions of kilometers of transmission line, thousands of generators, and power outputs ranging from under 100 kW to more than 1000 MW.
In recent years, the increasing integration of regional electric power systems has been further complicated by the purchase of electricity from distant utilities and independent suppliers, exploiting price differentials to reduce costs. At the Tennessee Valley Authority, the number of transaction requests through its service territory grew over tenfold in just three years.
Thus power grid data, once relevant to regional groups of utilities, is now communicated to a range of new entities being established to manage the restructured grids, e.g. Independent System Operators (ISOs) and Regional Transmission Organizations (RTOs). In order to be effective, these institutions must be able to understand fast-changing situations almost instantaneously, and choose corrective strategies nearly as quickly.
Figure 1: Transmission line/transformer PTDFs for a transfer from Illinois Power to PJM. Click here to enlarge image
The needs of power managers and marketers have become more urgent as access to the power grid has opened and competition has grown. They must now be able to see how much existing and proposed transactions will cost and to understand the availability of electricity at any time and place in the system.
The impact of dynamics like power flow, loop flow, and reactive power, which once mattered only to operational power engineers, now must be made clear to every player in the system—marketers, ISOs, RTOs, public service commissions and perhaps even consumer and voters, to whom such boards are answerable.
Visualization software for the electric power grid can display a large amount of information in each computer-generated image, enabling viewers to interpret this data more quickly and more accurately than previously possible. This capability will become indispensable, as electricity transactions grow in number and complexity.
How flows are managed
When electric power is transferred on the grid, few mechanisms control its route from generator to distant load. That route is often indirect, dictated by the impedances of the lines, other transactions that may be underway, and its points of entry and exit. In effect, a single transaction between a generator and a utility can spread over a significant portion of the grid—a phenomenon known as loop flow.
The percentage of a specific transfer that flows through any component in the grid—for example a transformer—is known as the power transfer distribution factor (PTDF). A transaction that would send power through an overloaded component may not be allowed, or may be curtailed if the transfer is already under way. The US term for such curtailments is transmission-line loading relief (TLR). A grid component owner that detects actual or potential overloading alerts the relevant authority—an ISO or RTO, for example—and asks for relief. The ISO thereupon orders loading relief measures. Any transaction involving a distribution factor higher than a predetermined level—set by the North American Electric Reliability Council (NERC) at five percent of the transaction—is a candidate for curtailment. Thus, if more than five percent of the power transferred as part of a transaction will go over a specific grid component that has become subject to a TLR, the transaction may be scaled back or even canceled.
TLR measures may in turn affect other existing and proposed transactions, creating a cascading effect throughout the system. This requires near-instantaneous analysis by a wide range of utilities, power producers, grid supervisors, and power marketers. The need for state-of-the-art visualization tools at every level is clear, as bottlenecks in this complex system can quickly cause nasty price spikes, and if left unaddressed, could result in brownouts or blackouts.
In the past, marketers or operators would need to scan long numerical tables of distribution factors to understand how line loading relief measures might affect a market or reliability area—an increasingly challenging task as the list of entries grows. In any large grid system, there are huge numbers of distribution factor sets, one for each pair of buyers and sellers. Visualization tools can assist a manager or marketer in considering the affect of a loading relief on a transmission line, especially on those not directly involved in a transaction. Other complicating factors can include planned or unplanned outages and a myriad of others contingencies. Visualization software assists the viewer in understanding the answers to these questions.
Visualizing power flow
Tracking electric power through a transmission network requires the calculation of both the real and the reactive power flow on each and every transmission line, transformer, and bus (that is, the voltages at each distribution node). With tens of thousands of buses and branches, such numbers were traditionally presented in either tabular output or else as data in a static ‘one-line’ diagram. (One-line diagrams display the underlying three-phase system as a single line.)
Figure 2: Color contour of hypothetical Northeast bus Locational Marginal Prices. Click here to enlarge image
The challenge for visualization is to make these concepts intuitive. One simple yet effective technique is the use of animated line flow [http://powerworld.com/spectrum/]. Here, the size, orientation and speed of the animated arrows indicate the direction of power flow on the line, bringing the system almost to life.
Another visualization approach that assists in quickly understanding network overloads is the dynamically-sized pie chart. The percentage filled on the pie chart of each one-line diagram indicates how near that transmission line is to its thermal limit. Color change and size can be added to flag critical points on the system as part of the animated graphical interfaces, but there are still challenges. Thousands of lines must often be considered, and continually checking each value is impractical. In the alternative, the use of tabular displays to sort the values by loading percentage results in a loss of geographical relevance. One solution is setting the graphic display so that pie charts only become visible in when the line load rises above a specified threshold [again see http://powerworld.com/spectrum/]. As well, complex dynamics such as the Power Transfer Distribution Factors (PTDF) for a power transfer from Illinois Power to PJM (Figure 1) can be much more quickly understood with the aid of visualization software.
What is clear is that visualization is a technology whose time has come. The Department of Energy’s National Grid Study, released in mid 2002, says that a major technological challenge facing the future of electric power in the US is operational uncertainty—concerning both reliable power and economical power. And a major technology fix DOE recommends is improved visualization, giving a bird’s-eye view of the power system [DOE National Grid Study, p.F-21].
Contouring the grid
For decades, one-line diagrams have been embellished with digital numerical displays of the nearest bus’ values. However, when working with a large number of buses, patterns emerge slowly. One method of displaying these data patterns is through the use of contours. Continuous, spatially distributed data such as equal-temperature in weather forecasting is familiar to most, but in power systems, the data is not spatially continuous. Bus voltage magnitudes are only measured at buses, and power only as flows on the lines, yet the spaces between buses and lines appear in contour displays as continuous gradients.
In practice, the main purpose of a contour is to show trends in data. Between the buses or the lines, colors span the two-dimensional contour region and create a graduated change in those colors between valid data-points. These colors can be set as the weighted averages of nearby data-points with specific averaging functions providing specific results. What this color gradation technique does do is to clearly illustrate the spatial relationships within the data. For example, Figure 2 shows the Locational Marginal Prices (LMP) for the area in question quickly and effectively.
Adding a third dimension
When visualizing power system data, there is usually no corresponding “physical” representation for the variables. For example, there is no physical correlate to the reactive power output of a generator, or for the marginal cost of enforcing a transmission line constraint. These are abstract calculated values, to be added as numerical indicators to diagrams in which power flows are represented in two dimensions. The data of interest in a power system can also include bus voltage magnitudes and prices, transmission line loadings, generator reserves and bids, and scheduled flows between areas. One method for managing these additional layers of information is to move from two-dimensional to interactive, three-dimensional or 3-D visualization.
In power systems, there is too much data for everything of interest to be displayed in a 3-D system. One approach is to allow rotation about more than axis. If the point of view or ‘camera’ rotates about the axis passing through its sides, it can also change its angle with respect to the horizon (elevation). The 3-D environment makes important information harder to overlook and otherwise hidden relationships easier to see. For example, the relationship between present generation output and reserve capacity is evident at a glance in Figure 3.
The 3-D display can also help address the reactive power output of generators in terms voltage problems. Voltage security analysis requires a simultaneous awareness of both the bus voltage magnitudes and the generator characteristics, including the generator reactive reserves. One very productive approach has been the combination of both techniques, with animated graphs or contouring displayed on the 2-D plane, and additional information provided by the third dimension. While the illustrations for this article have been kept relatively simple to illustrate the display options, a great deal of complexity is easily understandable with this combination approach.
Building the data map
In the past, the most common use of visualization software in electric power flow has been to better understand one’s own local or regional system. However, as power grid integration and energy markets continue to expand, so does the need for data at the national level. This national data allows more effective assessment of boundary effects between local control areas and longer range energy transactions. There are approximately 140 different utility control areas in the US, each trying to manage its own portion of the grid and navigate the ‘seams’ between themselves and their adjacent areas.
Location and analysis of this data is not an easy task, and it is not getting any easier. As the aftermath of the ‘9-11’ attacks on U.S. commercial and military targets continues to unfold, one challenge within power management and marketing has involved access to information. National security has become a critical issue at all levels and the collection and assembly of power grid information has brought with it the need to assure its responsible use by certified users.
In addition, the ever-evolving landscape of the North American power grid has mandated moving beyond initial base of FERC 715 filings and the NERC system to tracking the evolution of the ISO and RTO configurations as they develop. One approach to meeting these challenges has involved the creation of a Transmission Atlas, www.energyvisuals.com, which marries public data to visualization software.
While the Eastern and Western Interconnects may one day operate as coherent wholes, the NERC regions have long been the geographic base for subdividing these larger entities into manageable analytical units. But as the Midwest Independent System Operator (MISO) continues to grow—for example through its upcoming merger with the Southwest Power Pool—mapping structures must track these changes in order to remain relevant. The upcoming extension of the PJM (Pennsylvania, Jersey and Maryland) and PJM West systems into the proposed PJM South network will geographically interface with MISO, and further complicate the picture. And so as the information needs of appropriate and certified users continue to grow, the range of data displayed through visualization software must continue to grow as well.
There are two additional factors which expand the visualization challenge. First, the rate of change is increasing. The FERC 715 and NERC filings were quarterly, and thus responded to the very real differences in power demand between the four seasons of the year. The ISO pattern of system updates is still less standardized than FERC has been, but it is very likely to be much more often. Second, the in-house resources that are available to those in the electric power sector have continued to shrink, in terms of man-hours or resources available for in-house analysis of data. To quote the National Transmission Grid Study, “Unfortunately, most of the utility staff with access to these data are too burdened by day-to-day tasks to use the data or the tools required to analyze the data. Repeated staff reductions have meant that this complex task has almost vanished from utility organizational charts.”
This is because the building, analyzing and testing of power cases for accuracy is labor-intensive work. It requires experience and knowledge of how power systems respond, both in daily operation and in crisis. Accordingly, the combination of visualization software with pre-analyzed grid data profiles such as Transmission Atlas can address this engineering capability shortfall for both utilities and power marketers.
The evolution of market power
Power flows matter not only to operations engineers and power traders, but also to the authorities charged with deciding whether two utilities should be allowed to merge, whether a new combustion turbine is needed in a developing suburb, or whether absence of a single transmission line could send electricity prices soaring. What’s more, overloads on just a few transmission lines can affect even the largest power markets, whereupon prices spike, some players reap huge profits, and political firestorms arise about utility restructuring and deregulation.
One regulatory concern is that benefits from breaking up the old vertically integrated utilities will be minimized if unbundled generation and transmission companies exercise quasi-monopolistic power over local and regional markets. Either collusion among players or “gaming” the system—taking advantage of legal loopholes and operational quirks to create or exploit bottlenecks and chokepoints—are strongly resisted by regulatory entities.
Regulatory groups have dubbed this activity ‘market power,’ or the ability of one seller or a group of sellers to maintain prices above competitive levels for a significant period of time. In most jurisdictions with market competition, spot prices can be determined at every node (or bus) in the system. The U.S. name for this is the locational marginal price (LMP). Under truly competitive conditions, it equals the marginal cost to any generator bidding into the system of providing electricity to that point in the transmission system to any generator bidding into the system (again see Figure 2).
In the absence of overloads, spot marginal prices are relatively equal across a power market (depending on how resistive line losses are accounted for). But when overloads occur, spot prices can rapidly diverge. Because LMPs are bus-based values, contouring is again extremely useful for showing market-wide patterns. This ability will become even more useful as FERC’s proposed Standard Market Design (SMD) structure is implemented.
To be truly effective, computer visualization must be interactive and it must be fast. Today, a standard desktop computer has the capacity and speed to create a high resolution contour map within a few seconds. Fast contouring, coupled with easy zooming and panning, can equip the grid analyst with an interactive tool with which to quickly explore a power system data set. For instance, zooming could be used to provide more details about pricing, while dynamically sized pie-charts could be used to show lines that are close to but not yet exceeding their limits.
For the power industry, visualization techniques could not have come at a better time. Power markets are increasingly competitive and the transmission system is growing ever more complex, but visualization solutions offer a strong tool for cutting through the clutter.