for period 24 - 27 May 2010
Energy Systems Week
24 - 27 May 2010
|Monday 24 May|
|14:00-15:15||Hobbs, B; Ralph, D (Johns Hopkins/Cambridge)|
|What makes electricity different? Dumb grids, the ultimate just-in-time problem, and polar bears||Sem 1|
We argue that electricity is a fascinating source of problems for mathematicians because of the sector's economic and environmental importance, and because of its unique temporal and spatial structure.
The economy's most capital intensive sector must supply energy at the precise moment it is demanded -- there are no significant buffer stocks or storage, at least in most systems. Prices can, as a result, vary a hundred-fold in a day.
There are also no valves in this network; power flows follow all parallel paths, so one party's generation or consumption decisions affect everyone else, from the Isle of Wight to the Western Isles.
We review the classic optimization problems facing system planners and operators: static optimal dispatch on a constrained network, dynamic scheduling of generation unit operations, and long run investment capacity expansion. Although liberalisation has not changed the physics, the economics and decision responsibilities are drastically differen t now compared to the 1980s. Financial and risk considerations have never been more important, at the same time that society is demanded a rapid change to sustainable, intermittent, low emission generation technologies.
|15:15-15:45||Tea and coffee|
|15:45-17:00||Hobbs, B; Ralph, D (Johns Hopkins/Cambridge)|
|Things we don't know how to do: Huge nonconvex smart auctions; combining financial and structural models; and multilevel games||Sem 1|
Three increasingly important -- and challenging -- mathematical problems in power systems are considered in this talk.
The first is to obtain solutions to the mixed-integer, nonlinear, nonconvex, and stochastic short run operations problem that is the 'smart auction.' The primal solutions must be (near) optimal and feasible in order for the system to run and cost savings promised by liberalisation to be realized. Dual solutions should support the primal solution in order to encourage market parties to bid their costs and benefits honestly into the auction. Preventative control means that possible contingencies must be anticipated and prepared for.
Second, there is presently a gap between the deterministic, primal-solution oriented models that are used to evaluate alternative policies and the financial models and concerns that drive the decisions of risk-averse market actors. Policy models tend to be structural, in that their solutions can account for changes in technology, demand, and other market fundamentals. But they do a terrible job of representing price volatility, the conservatism of investors, or the effect of (non)availability of financial hedges upon investments. Financial methods, on the other hand, cannot represent the structural changes in the market that shift the ground beneath the feet of market participants, nor the reality of market power and price manipulation.
Finally, the interaction of large market players all along the supply chain -- from fuel to generation to transmission and finally to consumption -- represents a complex, multilevel game whose outcomes depend on the intensity of competition, learned collusive behavior, and regulatory threats. The consequences of our inadequate models of these interactions cost California ratepayers $20B in 2000-2001, and such models are needed to design markets that provide incentives for efficient investment and operations while restraining unproductive market manipulation.
|Tuesday 25 May|
|09:30-10:45||Meyn, S (Urbana-Champaign)|
|Dynamic models for electric power markets||Sem 1|
The goal of this tutorial is to lay a foundation for understanding the role of uncertainty and constraints in electric power networks and markets. While power generation, the transmission grid, and demand for power are complex beyond imagination, relatively simple models based on AC or DC power flow approximations can give a great deal of insight. If these models are augmented through the introduction of volatility in supply and demand, then the DC power flow model is essentially equivalent to a version of the inventory models considered in the 50 year old work of Clark and Scarf. Based on this analogy we can explain the high reserves demanded in typical electricity markets, and how reserves must grow with increasing uncertainty.
In highly deregulated markets, wholesale prices for power can oscillate in just a few minutes from zero, to ten thousand dollars - more than 100 times the average price for power. These prices may even become negative in some markets. It is argued that such exotic behavior can be explained, even in the absence of "market power", by considering the dynamic game of suppliers and consumers that includes the fundamental fact that power supply is subject to strict ramp constraints.
|10:45-11:15||Tea and coffee|
|11:15-12:30||Vinnecombe, G (Cambridge)|
|A scalable approach to the stability of power networks||Sem 1|
|14:00-15:15||Bialek, J (Durham)|
|Wide area blackouts: why do they happen and how can modelling help?||Sem 1|
Power systems in developed countries have been designed to operate with a very high level of reliability. Consequently, wide-area blackouts affecting millions of customers tended to happen very rarely, at least until recently. This situation seems to have changed over the last few years, as there has been a series of wide-area blackouts and disturbances affecting millions of customers in Italy, Sweden/Denmark and US/Canada (all three in 2003), continental Europe in 2006, and GB in May 2008. The talk will overview the common underlying reasons for the blackouts and discuss how mathematical modelling can help. Special attention will be paid to the increased penetration of Distributed Generation (DG), mostly wind-based.
Many believe that current operational procedures, designed to deal with a traditional power system characterised by a limited number of large centrally-controlled power plants, may not be adequate to deal with future power systems characterised by a very high penetration of distributed generators which are smaller in size, uncontrollable and stochastic in nature. The new operational paradigm will require a new approach to power system modelling and control.
|15:15-15:45||Tea and coffee|
|15:45-17:00||Kirschen, D (Manchester)|
|New formulations of the Optimal Power Flow (OPF) problem||Sem 1|
Transmission system operators are responsible for maintaining the security of the power system ("keeping the lights on") even when the system is affected by unpredictable but unavoidable contingencies such as the sudden loss of a transmission line or the sudden outage of a generating plant. To achieve this, they typically use a combination of corrective actions (e.g. redispacthing the generating units after a contingency has occurred) or preventive actions (i.e. keeping the system in a state such that it will not collapse if a contingency does occur - a "secure" state). This can be formulated as a very large non-linear optimisation problem because the state of a network with thousands of nodes and branches must be checked from thousands of credible contingencies.
In recent years, this problem has become even more complex because of the added uncertainty introduced by the stochastic nature of wind generation. Operators would very much like to know what the worst pattern of credible uncertainty is and whether they have enough resources to cope with any contingency using corrective actions if this pattern of uncertainty does materialise.
|Wednesday 26 May|
|11:00-12:15||O'Malley, M (Dublin)|
|Mathematical modelling for wind energy integration studies||Sem 1|
Driven by concerns over climate chnge, there is a drive to connect increasing volumes of renewable generation in electrical power systems worldwide. This creates a major change in the nature of power systems, as the availability of renewable generating capacity is a primarily matter of resource availability (e.g. how windy it is); by contrast, the availability of conventional generation capacity (coal, gas, hydro, nuclear etc) is primarily a matter of mechanical availability as long as an adequate fuel supply is available.
This tutorial will review existing applications of mathematical modelling in wind integration studies worldwide. This will include: - statistical characterisation of the wind resource - stochastic optimisation of power systems with wind - probabilistic models to determine network upgrade requirements for connecting wind generation - quantification of wind generation's contribution to supporting demand It will further describe open modelling challenges looking forward to ever higher volumes of wind generation.
|Chair: Frank Kelly|
|14:25-14:30||Open for Business: Introduction and welcome - Sir David Wallace|
|14:30-15:00||Meyn, S (Urbana-Champaign)|
|Is there a deregulated electricity market operating in the world today?||Sem 1|
While markets for electric power are operational all over the world, regulation of these markets has increased significantly over the past decade. In particular, electricity markets operating in the U.S. today are subject to price caps, and severe penalties for not following the "rules of the game". Indeed, electricity markets in the U.S. are no more "deregulated" than the U.S. highway system: Drivers can drive wherever they want, and whenever they want, but their range of behavior is tightly constrained. Given the complexity of power systems, combined with the complexity and uncertainty of markets and the key role of electricity in society, I argue that similar constraints must be placed on the participants in any electric power market. However, a fuller understanding of such complexities and the use of appropriate models to represent them are mandatory to succeed in the design of the ’electricity highways’ of the future.
If we can ensure that the overall system is more predictable this will facilitate the management of the power grid, and thereby enhance its reliability. In addition, a better understanding of dynamic markets will allow appropriate market designs that achieve the benefits from competition, while managing risks, and enhancing incentives.
This opens important research questions. In particular, how can we create market environments that take into account these critical issues?
Achieve an appropriate level of reliability in the face of significant variability of supply and demand in a market environment.
The cost of reliability in terms of efficiency, and other metrics.
Sufficient profits for generators so that they will remain in the game (for both the short- and long-term).
Incentives for creating new energy resources, or making better use of current resources such as wind and sun.
|15:00-15:30||Smith, S (Ofgem)|
|Ensuring the adequacy of future energy systems||Sem 1|
Since liberalisation in the early 1990s, Great Britain's energy market structure has been successful in delivering an appropriate level of security of supply. However, we will need to move in the future to alternative sources of supply and to reduce carbon emissions. It is far less certain that market mechanisms alone will be sufficient to effect the necessary step changes, in particular to deliver the required new capacity. This presentation will discuss how the market structure might evolve to meet this key challenge.
|15:30-16:00||Bialek, J (Durham)|
|Mathematical modelling of future energy systems||Sem 1|
Future energy networks will feature greatly increased interaction between heat, transport and electricity networks, as well as much greater storage, active demand, and stochastic-output renewable generation in electricity systems; moreover, it is likely that inter-continental scale networks will be developed to link from renewable sources (e.g. Saharan solar and hydro from Europe's mountain areas) with demand centres. This presentation will discuss the modelling challenges posed by the changing nature of electrical power systems, in terms of real-time control, market structures, and supply adequacy risk modelling.
|16:00-16:30||Murray, C (National Grid)|
|A national grid fit for the future||Sem 1|
The transition to a low carbon economy presents significant challenges and opportunities for National Grid. Chris will explore the changing nature of our energy supplies and National Grid’s role in the nation’s drive to balance energy security, sustainability and affordability.
|16:30-17:00||Tea and coffee|
|17:00-18:00||(Chair) Pollitt, M|
|Panel discussion||Sem 1|
Open for Business panel discussion with Michael Pollitt (Chair), Sean Meyn, Steve Smith, Janusz Bialek, Chris Murray, David MacKay (if available)
|19:00-21:00||Dinner at Churchill College|
|Thursday 27 May|
|09:45-10:15||Riches, S (EPSRC)|
|UK Research Councils Energy Programme||Sem 1|
|10:15-11:00||Meah, N (Dept of Energy and Climate Change)|
|Including future uncertainty in economic projection models||Sem 1|
|11:00-11:30||Tea and coffee|
|11:30-12:15||Richards, A (National Grid Control Centre)|
|Demand forecasting||Sem 1|
|12:15-13:00||Tritschler, M (KEMA consulting)|
|Smartgrid control challenges||Sem 1|
|15:00-16:00||Plenary session and report back|
|16:00-16:30||Tea and coffee|