The electric power grid is a lot like the circulatory system in the human body – an intricate structure with millions of interconnected channels, large and small, through which energy must constantly flow – with a minimum amount of obstruction. It must be able to operate under a wide range of conditions, it must serve all extremities without exception, and it must be able to grow to meet new needs as the overall body grows.
Like a long distance runner challenged by new Olympic events in which it was never expected to compete, Ontario’s power grid has come across serious capacity limitations in certain areas. Although overall reliability can be maintained, certain kinds of power flow are likely to be increasingly constrained. By most accounts, congestion will increase in the near future, affecting more locations, more time frames, and greater amounts of energy. To a large degree, grid congestion is setting the upper limits for the location of new renewable generation projects in the short and medium term – projects that are the centerpiece of the Ontario government’s green energy program and according to some, one of the best hopes for improving the province’s environmental performance. As many generators are aware, competition for connection capacity in many parts of the provincial the grid has become intense. And even for those with already-installed plant, grid congestion can impose limits on when they can run, potentially reducing their output and revenue.
Grid congestion is not new of course. System planners and operators have been dealing with congestion since the grid was conceived. What’s new is the degree of congestion, its potential impact on a wider range of generators, and the range of tools being used to address it. Congestion management has been steadily moving up the list of priorities for managers of the system to address. What was important before is quickly becoming an essential step in planning – and accommodating – new connections. Key to managing congestion is recognition that it is a kind of mixed blessing: Although it is problematic for generators, the generally-desirable level of congestion is well above zero, and not all congestion is harmful from an economic perspective. (Some degree of congestion represents confirmation that the system is not over-built, as excess transmission would represent a waste of resources.) In fact, in a well-managed system, congestion can send important signals to market participants that help them make investments and adjust real-time behaviour to operate more efficiently.
What remedies are being applied to optimize and manage congestion, and what roles will each of the tonics likely play? In the months to come and in the pages that follow you can expect to hear more about conventional tools like dispatch, outage scheduling, settlement credits, congestion management payments, transmission expansion and distribution build-outs, as well as newer options such as strategically located distributed generation, demand response, FACTS technology, and IESO-initiated curtailment of generation.
Some of the specific issues that rise to the surface in current discussions of congestion management include the following:
• What specific tools does the IESO expect it will need to use to manage congestion as it occurs, and how can generators prepare themselves for the business implications of these tools?
• What will the OPA have to do to minimize the degree of congestion on the system, while keeping economic considerations in mind, if it is to integrate several thousand megawatts of new renewables into the grid, while properly assisting the IESO to manage its operational challenges, many of which are supply-related?
• Is a policy response necessary, or can congestion issues be adequately managed with the set of regulatory, planning and operational tools currently in place?
The scale of congestion in Ontario and why it’s important
The IESO publishes forecasts of “bottled generation” by region every three months as part of its regular 18-month Outlook. (Bottled generation is defined as generation that can’t run because transmission limits prevent it from being transferred outside of its immediate region.) These near-term forecasts are useful reading for anyone considering location decisions in Ontario. The OPA develops a range of projections like these for the longer term to assess the needs and plan for new transmission facilities. A quick glance at the tables shows just how volatile the situation is. For example, even in the IESO’s firm scenario, the amount of bottled generation forecasted in the Bruce and West region as of August 23 went from over 3000 MW in early September to only 187 MW in late October. The Bruce region seems to be the focus of considerable congestion. This situation will of course change dramatically when the new Bruce to Milton transmission line goes into service.
Kim Warren, Director of Planning and Assessments at the IESO, stresses that you don’t necessarily build transmission to deal with every possible example of congestion – there is always an economic balance to be struck between the costs of ongoing congestion and the costs of transmission expansion. “Congestion develops for a wide variety of reasons,” he notes. “It has a lot to do with ambient conditions, power flows, outages and a wide range of other real-time factors.”
As more intermittent renewable energy is added to the system, the likelihood of sudden unexpected jams increases. At the same time, as more and more distributed generation is added, there is potential to alleviate constraints in more ways than before. The OPA has signed contracts for new gas-fired generators many of which are intended to offset transmission constraints to a large degree. However, it will take time for all the benefits to be available. “As you’re bringing on gas that hasn’t matured, there will be temporary congestion between it and the not-yet-retired coal capacity,” Warren says. This is expected to be particularly noticeable in the region west of London in the Sarnia / Lambton area.
The truth is, congestion is hard to measure and even harder to predict. It is inherently location- and time-specific. The real-time remedies and the longer-term prevention strategies operate at opposite ends of the system. The IESO’s models are helpful for generators in planning their outage schedules, but in real time, the dispatch orders from the IESO, and other real-world conditions, can result in quickly changing congestion conditions. In fact, predicting congestion is a highly specialized skill, and forecasts are undertaken on an hourly basis by the IESO’s control room operators. Generators in certain regions of the province are quite likely hiring their own congestion specialists to help with their risk assessments.
Given the long lead times required to prepare congestion forecasts, the IESO’s predictions are not intended to be the final word on the subject. It publishes a lot of additional information that is relevant to transmission and congestion, such as historical inter-tie flows, schedules, transfer limits, internal shadow prices and constraints that are binding the market dispatch. Although the IESO releases a great deal of relevant data, predicting congestion remains challenging, and market participants are to a large extent on their own if they want to figure out with any precision how congestion is likely to impact their operation in future years, or whether a particular location decision will be heavily compromised by future congestion.
In fact, congestion comes in two distinct varieties: global (transmission-related) and local (distribution-related). The IESO is responsible for managing the province-wide grid and focuses its efforts on transmission congestion – issues that impact the ability to transfer large volumes of power from one part of the province to another. Ontario’s market rules include an intricate system to compensate generators if they are forced off by the IESO for reliability reasons. The rules provide for some generators who offer in at or below the clearing price to receive Congestion Management Settlement Credits (CMSCs) if the IESO could not schedule them for reliability or congestion reasons. Much of the discussion around potential congestion management initiatives will understandably focus on the impacts on the system of CMSCs.
Genivar’s Don Krause notes that not all transmission restrictions that could restrict a generator facility’s ability to deliver power, would be eligible for congestion management payments. One such situation that would not be eligible for these payments is transmission system restrictions in radial line connections. Intermittent generators are not eligible, nor are non-quick-start generators eligible unless they are already synchronized. The IESO’s System Impact Assessments (SIA) studies for generation applications identify the required upgrades to provide adequate capacity in the transmission facilities including the radial lines, however some system events can still cause the need for restricting generation deliveries.
The latter part of 2010 promises to be a particularly difficult period for generation in Ontario, as significant amounts of inter-tie transmission capacity with New York are expected to be out of service, effectively bottling up even more generation than would normally be the case. In examples of this nature it becomes apparent just how closely related congestion is to another major problem on the grid: Surplus Baseload Generation or SBG, which occurs when baseload generation is greater than the demand for power. It can lead to very low or even negative market prices for power.
The IESO does not manage local congestion within distribution systems, and there are no settlement credits available to compensate generators who are constrained off for local congestion reasons. Yet local congestion could well prove to be as great a challenge as global congestion. Distributors in Ontario, including Hydro One, are likely paying more attention than ever to management of congestion, given the historically significant increase in generation volumes they are being asked to handle, and the relatively limited set of tools they have for addressing it. Distributors will generally not make connections that trigger congestion on their systems. In most cases they will have to expand their distribution capability first.
The current techniques for management of congestion
Kim Warren explains that if congestion affects reliability, the IESO can transfer load, recall outages, and decline to approve scheduled outages, along with taking any number of other control actions. “In the 18-Month Outlook we forecast a broad range of things, including congestion. Experienced Market Participants will use that information to schedule outages.”
System balancing techniques used routinely by the IESO take congestion into account in the normal course of operations. These include the IESO’s routine dispatch scheduling, a range of control actions such as utilizing operating reserve, the use of the Congestion Management Settlement Credits that apply in some cases, Automatic Generation Control, publication of data to help market participants anticipate congestion and adjust their operations accordingly, and in some cases, curtailment of generation. Curtailment has become a particularly sensitive issue as it directly affects generator revenues, and the contractual provisions for curtailment have evolved as the OPA designs each successive generation of contracts.
Congestion management is part of a bigger grid operation picture, of course. It is rarely dealt with in isolation from other technical procedures for grid balancing. The IESO, the OPA and Hydro One are continually assessing a range of system management issues aside from congestion, and they are all managed with the same set of operational controls.
There is a complex interplay between congestion issues at the transmission and distribution levels. The causes of congestion, and many of the solutions to it, will frequently reach across the boundary between transmission and distribution, whereas the human systems for managing congestion are quite distinct between the two levels. The IESO’s work, and to a large extent the OPA’s work, focus on the transmission system, leaving local distribution companies to manage congestion at the distribution level. At the local level of course, there are dozens of different distributors who may take different approaches to congestion management. While local congestion is currently a relatively infrequent problem, it could become much more evident in the near future.
Working without the benefit of province-wide market rules that can be used to reward those who alleviate congestion, local grid managers may face the most difficult challenges of all. Yet a single solution, whether at the distribution or transmission level, will quite often produce congestion relief at both ends.
Dave Jutla, Manager of Capacity Planning at Toronto Hydro, notes that a critical related issue is the need to improve the reliability of many older transmission lines in Ontario’s large cities that serve the primary consumers of power in the province. As an example, he cites Toronto, which consumes something like 18% of the province’s power by itself. “Enhancement of reliability through refurbishment of old infrastructure and addition of new transmission, preferably underground cables, should form part of this big picture,” he says.
Dave Jutla contends that some of the planning criteria currently used by the IESO, OPA and Hydro One’s transmission planners “will not deliver the higher degree of reliability required by present day consumers of electricity. Reliability of power delivery through transmission lines in Ontario’s large cities does not match the reliability of cities like New York, Paris, London and Tokyo.” He suggests that transmission and distribution planners may have to include more underground systems, which are less prone to bad weather, in the capital investment plans they bring to the OEB. Mr. Jutla bases some of his concerns on a close look at the official outage restoration criteria in the province, which may be hard to meet in some cases, he believes. The OPA observes however that investments to reduce congestion may increase reliability but investments to improve reliability may not necessarily reduce congestion.
The question often arises, who should initiate proposals for new infrastructure to relieve congestion, and who should own and pay for them? There are no conclusive answers to these questions. The OPA, in its role as planner of the system, identifies areas of weakness and suggests approaches to shoring up the system. But there are a number of alternative procedures that may come into play for actually building the grid capacity. Frequently, issues of this nature are identified through System Impact Assessments instigated by a generator’s connection application. The choice of development process can depend on whether the distributor, transmitter or some other party steps forward with a proposal to build congestion-relieving infrastructure. Under the most recent set of proposals, the OPA’s Economic Connection Test is intended to be the screen for what projects will be accepted by the OEB as reasonable for development work, and (importantly) eligible for cost recovery.
Longer term solutions
Managing congestion is not just a real time concern. A good deal of analysis goes into congestion issues when the IESO performs System Impact Assessments (SIAs) for proposed new generation connections. These reports, and other studies like them, are significant considerations when proposals are drawn up for new transmission lines and electrical infrastructure. The OPA notes however that an SIA primarily focuses on the feasibility of connecting a certain amount of capacity to the system and what enhancements to the system would be necessary to accommodate this capacity. Whether or not this capacity would be capable of being delivered at all hours of the year is a congestion issue and is not examined in the SIA directly.
In the medium term, it is expected that new congestion will be mitigated by the restriction of new generation connection, now principally administered through the OPA’s TAT and the associated distributor DATs.
Ultimately the most definite and long-lasting solution to congestion is more grid capacity. Increasing attention will likely be paid to determining which grid expansions are most economic. Although adding grid capacity can be a slow and expensive process, it’s an essential aspect in the integration of new generation. All over North America there are indications that grid expansion, previously seen as environmentally questionable, is being treated as an important enabler for renewable energy. From California to New England, transmission lines are being approved as part of long term proposals for greening the supply mix. The OPA is of course developing plans for the Ontario grid that identify resource and transmission needs in an integrated manner.
In Ontario the government can move initiatives of this nature forward by issuing instructions or directives to the central agencies like OPA and Hydro One. In September 2009, a letter of instruction from the then Minister of Energy and Infrastructure to Hydro One identified a set of high priority transmission infrastructure projects as part of the government’s green energy program. Subsequently, the current Minister of Energy sent a letter to the OPA asking for a “transmission expansion plan updating the September 2009 instruction to Hydro One” and the OEB has promulgated its “Framework for Transmission Project Development Plans” in which it establishes its policy on designating work to be done by competitive transmitters. As such, Hydro One has currently put its development work with respect to major transmission projects to connect new renewable supply on hold. A smaller set of projects identified in September 2009, those listed as part of “Schedule B,” are proceeding. These investments, which are focused on reducing a number of localized distribution constraints, will undoubtedly alleviate congestion in these areas when they are completed. At the same time there are infrastructure build-out proposals going through the normal development process, and a range of propositions coming from distributors in the context of their Green Energy Plans. It should be noted however that these Green Energy Plans relate to the distribution system, not to bulk transmission. Distributors are required to have the OPA review their Green Energy Plans to assess if there are any upstream bulk system issues that would impede those distribution plans.
Curtailment provisions in generator contracts
There are provisions in many of the generation contracts in Ontario that specify financial obligations which can assist the IESO in its efforts to manage a similar problem, Surplus Baseload Generation (SBG). For example, under the FIT contracts, generators can be eligible for special payments if the IESO requires them to curtail output for particular reasons, as long as the necessary contractual conditions are met.
Generally speaking, most kinds of generation contracts in Ontario require the generator to be a market participant. Market rules therefore take effect, and they can require the generator to comply with a range of instructions from the IESO, including curtailing output for congestion reasons. See the sidebar below for more information on how contractual provisions related to congestion management operate.
Future options that may help with congestion management
In the longer term, a range of solutions are likely to come forward with implications for congestion management. The IESO is monitoring whether generators with storage abilities start coming on line. Such systems could be of great value if the IESO is able to influence their operation to alleviate congestion. As smart grid technologies find their way onto the grid, distributors could do much the same, perhaps acting in part as partners with the IESO in alleviating congestion.
In addition to technical solutions, policy decisions can impact congestion and the options for managing it. One of the key examples of this is the degree to which public policy is used to determine supply mix choices and the operating strategies for various fuel types. For example, in Ontario there will likely be many hours in which the supply of carbon-free generation outstrips the level of consumption and exports. At such times, the IESO will likely have to choose between whether to reduce output from nuclear, hydro-electric, and wind facilities – all of which have relatively low marginal cost of operation, and minimal operating emissions. This is another example where the IESO will have to manage SBG, and may choose to use techniques similar to those it would use for congestion management. Clearly, the choices the IESO makes in terms of such supply curtailment will have major impacts on the affected generation businesses, and quite likely on the state of congestion in various parts of the province.
As Kim Warren puts it, “If you can’t back off an intermediate fuel such as natural gas, you have to choose between baseload facilities such as nuclear reductions, spilling water or backing off someone else.”
Looking further into the future, there is a long-term policy question about the appropriate level of congestion to work towards, and if it’s even realistic to set standards in the normal sense of the word. Ideally, from the perspective of achieving the most competitive and efficient market for wholesale power, economic signals rather than administrative instructions would be relied upon to cause generators to gear up or back off their plants in response to hourly prices, making congestion management less critical. However, if the system is so congested that dispatch is driven largely by reliability and policy-related concerns, the effectiveness of the hourly price as a motivator for generators to enter and leave the market will decline. Although a perfectly balanced system may rely heavily on price signals to minimize congestion, there will almost certainly be a need for non-price measures to be used as well. Just how much reliance should be placed on market forces vs. administrative tools is one of the perpetual design questions faced by market participants and grid managers. One touchstone is, of course, the fact that the OPA is expected to remain the entity that will be primarily responsible for planning the future system, including the resource procurement, degree of congestion and transmission build-out.
Ultimately, many of the answers probably lie in the technologies that are still to come as part of the growth of smart grids. Such technologies can make it possible for extremely granular decisions to be made by operators of generation and load, enabling them to respond to relieve congestion at or near real time. One of the challenges is, of course, devising systems that will convert the benefits of such finely-tuned operational adjustments into revenue streams reliable enough to help finance the smart grid investments. As has been said in many places, it will take some time to see how and when these increasingly intelligent control systems take a major role in controlling power flows. No doubt many of Ontario’s local power distributors are working on developing and implementing such options. It is certainly a story with many local variations and in which new developments may change the picture yet again.
Like the human body, the system needs to be monitored carefully, exercised regularly, and encouraged to grow to meet new challenges.
Abbreviations used:
ACES: Accelerated Clean Energy Supply
CES: Clean Energy Supply (typically high efficiency gas-fired)
CHP: Combined Heat and Power
EMCES: Early Movers Clean Energy Supply
FACTS: Flexible AC Transmission Systems
FIT: Feed-in Tariff
RES: Renewable Energy Supply
RESOP: Renewable Energy Standard Offer Program
* * *
The ECT and OPA's process for planning system expansion
The Ontario Power Authority, by virtue of its role as the primary planner for the provincial electric system, carries a major responsibility for ensuring that uneconomic congestion in the future is kept within reasonable limits. Like the IESO, it has a suite of tools for planning and managing the development of the grid, and congestion management is never far from their minds. The OPA notes however that although many see its role as generally related to minimizing uneconomic congestion, with respect to the FIT program the OPA actually does its congestion planning on a percent of time basis, and makes no claims about how much congestion is economic or uneconomic for those investments.
For the purpose of determining transmission build out for enabling renewable resources, the OPA uses two defined metrics in evaluating proposals for major new transmission facilities: The expected congestion levels and unitized cost. In terms of the allowable level of congestion, the OPA is planning for a “generally uncongested” system. The measure they have proposed for this is to have congestion during no more than 5% of the hours in a year on any major transmission path on the Ontario transmission grid. While there is no industry standard for an acceptable level of congestion, the 5% value is in line with other jurisdictions in North America. Any such standard is of course a tradeoff between the cost of a more robust grid versus more efficiently producing either lower-cost or environmentally-preferred generation on the system.
Under the Economic Connection Test (ECT) established as part of the Feed-in Tariff (FIT) program, a maximum cost for new ratepayer-funded transmission network capacity has been established for FIT projects. The general rule is that to initiate a transmission project’s development work, the total cost for all required ratepayer-funded transmission network capacity should be no greater than $500 per kilowatt of generation enabled ($500/kW). It should be noted that all network transmission costs are combined when this comparison is made, meaning that if a generator triggers multiple network transmission upgrades to get to market the combined cost of these network upgrades is evaluated and compared to the threshold. The decision to build the facility ultimately will be determined by the OEB in a Leave-to-Construct application once the early phase development work is complete
The OPA’s Director of Transmission Integration Bob Chow explains: Because the FIT is a standard offer program, the power cost is predetermined, and the offer to connect has already been made, in principle. A procedure that frequently exists in other kinds of economic evaluations, an examination of the cost of the resource a proposed project relies on, is absent here. The sole question from a planning perspective then becomes what system expansion, where and when, is economic. And the metric applied there is the cost of the additional line against the amount of generation that will be enabled as a result of the investment. If the expansion comes in under $500/kW, it passes the test.
If a given expansion project meets these two tests, there is a great deal of flexibility in how to bundle projects together for attachment to a new facility, and in developing the expansion plan. Every case is different, Chow explains, and each requires detailed analysis of the options – which projects to connect in a given area, whether to build one line or more, and where, what voltage the lines should be built to handle, the use of other technologies, like FACTS or Static-Var compensators that might be deployed, and so on. Inevitably, a key part of the ECT process is the development of options. It’s a rigorous process, but with a lot of flexibility, he says.
Having established what the OPA considers an adequate expansion to serve a given project or group of projects, the next step is to find a transmitter to build the line. Selection of a transmitter is a process that is regulated by the OEB, which has recently posted its proposed framework on how a transmission developer is to be selected for a given expansion project. (See related article elsewhere in this issue of IPPSO FACTO.) The OEB document “Framework for Transmission Project Development Plans,” released August 26, is available at www.oeb.gov.on.ca/. There are also choices on the part of the project proponent, such as the possibility of paying for construction of enabler lines themselves. Consideration of such options may involve a dialogue with the generator.
Bob Chow notes that the Bruce to Milton line, the need for which was initially identified by the OPA and which is presently the only major work under way by Hydro One, is in many respects the progenitor of the present system. “There wasn’t any ECT process at the time, but if it were to be proposed today, the Bruce to Milton line would probably be subject to it. We didn’t realize it at the time, but it formed the pattern for future transmission expansion – it considered contracted and prospective needs; it enabled renewable resources; it assessed congestion versus cost of expanding infrastructure; it clarified the roles and responsibilities of the transmitter, OPA and the IESO in such projects; and it provided a rational process for advancing transmission projects,” Chow said.
* * *
Contractual provisions for managing oversupply and congestion
The Ontario Power Authority, by virtue of its role as the primary planner for the provincial electric system, carries a major responsibility for ensuring that uneconomic congestion in the future is kept within reasonable limits. Like the IESO, it has a suite of tools for planning and managing the development of the grid, and congestion management is never far from their minds. The OPA notes however that although many see its role as generally related to minimizing uneconomic congestion, with respect to the FIT program the OPA actually does its congestion planning on a percent of time basis, and makes no claims about how much congestion is economic or uneconomic for those investments.
For the purpose of determining transmission build out for enabling renewable resources, the OPA uses two defined metrics in evaluating proposals for major new transmission facilities: The expected congestion levels and unitized cost. In terms of the allowable level of congestion, the OPA is planning for a “generally uncongested” system. The measure they have proposed for this is to have congestion during no more than 5% of the hours in a year on any major transmission path on the Ontario transmission grid. While there is no industry standard for an acceptable level of congestion, the 5% value is in line with other jurisdictions in North America. Any such standard is of course a tradeoff between the cost of a more robust grid versus more efficiently producing either lower-cost or environmentally-preferred generation on the system.
Under the Economic Connection Test (ECT) established as part of the Feed-in Tariff (FIT) program, a maximum cost for new ratepayer-funded transmission network capacity has been established for FIT projects. The general rule is that to initiate a transmission project’s development work, the total cost for all required ratepayer-funded transmission network capacity should be no greater than $500 per kilowatt of generation enabled ($500/kW). It should be noted that all network transmission costs are combined when this comparison is made, meaning that if a generator triggers multiple network transmission upgrades to get to market the combined cost of these network upgrades is evaluated and compared to the threshold. The decision to build the facility ultimately will be determined by the OEB in a Leave-to-Construct application once the early phase development work is complete
The OPA’s Director of Transmission Integration Bob Chow explains: Because the FIT is a standard offer program, the power cost is predetermined, and the offer to connect has already been made, in principle. A procedure that frequently exists in other kinds of economic evaluations, an examination of the cost of the resource a proposed project relies on, is absent here. The sole question from a planning perspective then becomes what system expansion, where and when, is economic. And the metric applied there is the cost of the additional line against the amount of generation that will be enabled as a result of the investment. If the expansion comes in under $500/kW, it passes the test.
If a given expansion project meets these two tests, there is a great deal of flexibility in how to bundle projects together for attachment to a new facility, and in developing the expansion plan. Every case is different, Chow explains, and each requires detailed analysis of the options – which projects to connect in a given area, whether to build one line or more, and where, what voltage the lines should be built to handle, the use of other technologies, like FACTS or Static-Var compensators that might be deployed, and so on. Inevitably, a key part of the ECT process is the development of options. It’s a rigorous process, but with a lot of flexibility, he says.
Having established what the OPA considers an adequate expansion to serve a given project or group of projects, the next step is to find a transmitter to build the line. Selection of a transmitter is a process that is regulated by the OEB, which has recently posted its proposed framework on how a transmission developer is to be selected for a given expansion project. (See related article elsewhere in this issue of IPPSO FACTO.) The OEB document “Framework for Transmission Project Development Plans,” released August 26, is available at www.oeb.gov.on.ca/. There are also choices on the part of the project proponent, such as the possibility of paying for construction of enabler lines themselves. Consideration of such options may involve a dialogue with the generator.
Bob Chow notes that the Bruce to Milton line, the need for which was initially identified by the OPA and which is presently the only major work under way by Hydro One, is in many respects the progenitor of the present system. “There wasn’t any ECT process at the time, but if it were to be proposed today, the Bruce to Milton line would probably be subject to it. We didn’t realize it at the time, but it formed the pattern for future transmission expansion – it considered contracted and prospective needs; it enabled renewable resources; it assessed congestion versus cost of expanding infrastructure; it clarified the roles and responsibilities of the transmitter, OPA and the IESO in such projects; and it provided a rational process for advancing transmission projects,” Chow said.