EMRG 1998-2002 Final Activity Report

Caveat

This draft report is incomplete, and may be inaccurate in some details 

Introduction

This report covers four years of activity by the Energy Modelling Research Group, within the Department of Management at the University of Canterbury.  In our previous report, we commented that:

“The external environment has changed significantly in this period…  with the establishment of the New Zealand spot market marking a major milestone in local developments. Thus we are shifting from a period of intense research into how theoretical markets might behave, into an era where interested parties must cope with the realities of an actual market. We believe that the research carried out here has had an important, and beneficial, influence on the development of this new environment, but also expect that "research" activities will now be increasingly performed by commercial parties and consultancies in a real word context. In fact, we note with pleasure the contribution which our own graduates are already making in this regard. Conversely, we are comfortable with the reduced size of our own group in this new environment.”

Change has continued apace during the succeeding years, with the breakup of the Electricity Corporation of New Zealand (ECNZ), and the introduction of “Full Retail Competition” in 1999, creating a radically different market situation.  EMRG activities have been deliberately wound down over this period, and although ECNZ continued its support for part of this period, the group has not pursued sponsorship from any of its successor organisations, being in a position to complete several projects using the funds already provided for that purpose by ECNZ, and earlier by Trans Power (NZ) Ltd.  Supplementary funding was also provided by way of a contract with Putnam Hayes and Bartlett (PHB), prior to its takeover by PA Consulting in 2000.

During this period, three Ph.D. candidates, Tristram Scott, Gavin Bell and Olly McCahon have completed, while two more, Andrew Kerr and Stephen Batstone currently have theses under examination.  The group no longer employs any research assistants, and Bhujanga Chakrabarti continues as the sole remaining (part time) Ph.D. candidate.  There have been major changes on the academic staffing front, too:

  • ohn George resigned from the university in 1998, and took up a position with PHB (now PA) in Wellington
  • Grant Read, who for most of this period was a part time Associate Professor, retired at the end of 2001, although he continues to be associated with the group, in the role of Adjunct Professor.
  • Deb Chattopadhyay, who joined the group as a research fellow in 1997, took up a staff appointment, but resigned in 2000, subsequently  joining Charles River Associates (CRA) in Wellington.

Thus it would be fair to say that the group has now been wound down almost to the point of non-existence, and this report has largely been written to provide a final summary of, and conclusion to, our activities.  On the other hand, it seems likely that 2003 will see at least two new PhD students commencing work in the area, thus continuing the group’s work, albeit without the formal structure provided by EMRG in the past.  Indeed it is worth noting that two theses on electricity sector organisation and management have already been completed within the Department, outside of the EMRG structure.

While we are bringing this phase of the Group’s activities to a close, it is pleasing to note that market solutions of the type researched, and to some extent pioneered, here continue to gain international acceptance.  For example the US Federal Energy Regulatory Commission in issuing a “standard market design”, intended to set the pattern for future developments in that country, recently made the following comment:

Building on Best Practices

The Standard Market Design proposal builds on the wealth of experience the Commission has accumulated in its nearly decade-long experience with competitive wholesale power markets. It draws on lessons learned from the catastrophic failure of California's electricity market restructuring, as well as elements of highly successful wholesale power markets elsewhere in the country and overseas in the United Kingdom and New Zealand. The commission has looked to these and other markets to assure it adopts best practices available for a sound competitive market operation, to assure the prevention of market dysfunctions with adverse effects for customers, the electric industry and the nation as a whole.

While we are bringing this phase of the Group’s formal activities to a close, it should be recognised that many past and present group members, and indeed a considerable number of Management Science graduates from this department who were never formally part of EMRG, have played significant roles in such developments in many countries, and continue to do so.

The remainder of this report provides a brief overview of recent research in specific areas, followed by a listing of people involved in the group over recent years.  Relevant papers published during this period are listed in each section [1] , and a list of working papers is appended.  As in previous years, some of the group's work is contained in contract reports, which have not been listed for reasons of confidentiality.

References

  1. E.G. Read and D. Chattopadhyay, “Electricity Market Models: Lessons from Australasian Experience”  Interfaces (accepted subject to minor revision)

Top

Research Areas

Transmission Pricing

Bhujanga Chakrabarti is writing up his Ph.D thesis on  "Reactive Power Management and Pricing".

The first part of the thesis deals with reactive power pricing. The present work extends the full AC model by incorporating voltage stability margin constraints, so that the system is constrained to operate at a certain distance (margin) from the point of instability for security reasons. The generators are required to hold both active and reactive reserve power in order to maintain the margin. The model optimally allocates real and reactive reserves among the generators to meet a pre-specified voltage stability margin. This is a non-linear programming problem and is solved using an NLP solver.  Consideration is given to the introduction of markets for MVAr production and MVAr reserve in addition to those presently dealing with MW energy and MW reserve. In these markets, the suppliers are assumed to offer these commodities as they do at present for the MW-energy and MW reserve markets. These ideas are demonstrated using a reduced North Island system of New Zealand.

  1. E.G. Read: "Electricity Sector Reform in New Zealand: Lessons from the Last Decade”  Pacific Asia Journal of Energy Vol 7, No 2, 1997, p. 175-191
    The second part of this thesis deals with long-run reactive power (VAR) planning, incorporating the voltage stability margin constraint into a contingency constrained VAR planning model. The main objective of long term reactive power planning is to plan for installation of the optimal amount of reactive power capacity at key locations. This needs to assure energy supply during normal conditions, and also for a set of contingency conditions, while maintaining a pre-defined distance from the point of voltage collapse i.e. maintaining voltage stability margin. 
    This is a mixed integer non-linear programming (MINLP) problem and is solved using a GAMS solver, DICOPT. A new model for voltage stability constrained VAR planning is developed by simplifying the MINLP model using statistical approximation methods. A new technique is also developed to preserve voltage stability margins by optimal load shedding incorporating dynamic load shed characteristic.

References

E.G. Read: "Transmission Pricing in New Zealand", Utilities Policy,  Vol. 6, No. 3, 1997, p. 227-236.

  1. Chakrabarti, B B, E. Grant Read, D. Chattopadhyay, “Optimal Reactive Power Management in Electric Power System” proc. 35 th Annual Conference, Operational Research Society of New Zealand, Wellington, December 1-2, 2000. pp 75-6 (with slides).
  2. Chattopadhyay D, B.B. Chakrabarti, and E.G. Read; “Pricing for Voltage Stability”- proc. IEEE / PICA conference, pp. 235-40, Sydney, May 20-24, 2001. Proceedings published on the CD.
  3. Chakrabarti, B.B, D. Chattopadhyay  and C. Kumble; “Voltage Stability Constrained VAR Planning- A case study for New Zealand”- Proc. IEEE/LESCOPE 2001, pp 86-91, Halifax, Canada, 11-13 July,2001.
  4. ]Chattopadhyay, D and B.B. Chakrabarti, “Voltage stability constrained VAR planning: Model simplification using statistical approximation” International Journal of Electrical Power and Energy Systems, pp. 349-358 vol 23, June 2001.
  5. Chattopadhyay, D,  and B.B. Chakrabarti; “Reactive Power planning incorporating Voltage Stability” –Int. Journal of Electrical Power and Energy Systems, pp 185-200, vol 24, March 2002.
  6. Chattopadhyay, D,  and B.B. Chakrabarti; “A preventive/corrective Model for voltage Stability incorporating dynamic load shedding” –to appear in  Int. Journal of Electrical Power and Energy Systems.
  7. Chattopadhyay D, B.B. Chakrabarti and E.G. Read; “A Spot Pricing Mechanism for Voltage Stability”- submitted in Int. Journal of Electrical Power and Energy Systems.
  8. Chattopadhyay, D. and D. Gan, “Market Dispatch Incorporating Stability Constraints” International Journal of Electrical Power and Energy Systems. 2001.

Integrated Energy/Reserve Market Modelling

Aspects of this topic are now incorporated into Bhujanga Chakrabarti’s thesis, and are discussed above under “Transmission Pricing”.

Reservoir Management in a Competitive Environment

Apart from publication of some earlier work with Yang Miao, Stephen Batstone built on the earlier work of Tristram Scott (see EMRG 1995-1997 Annual Report), which had looked at the effect that a fixed level of long-term financial contracts has on a dominant firm's reservoir management and spot market behaviour. 

Stephen used Scott’s stochastic dual dynamic program for the duopoly in the New Zealand hydro-thermal market to examine the extent to which firms’ faced profit variability when they acted in a profit maximising, risk neutral fashion. The model, which assumed a fixed level of forward contracts had been sold by the generators, was run for a range of contract levels, and the reservoir management strategies were tested via simulation over a range of inflow scenarios [14] .  The results suggests that in the particular scenario modelled, profit risk is quite low for low levels of contracts.  This conclusion was important for the research Stephen undertook for the remainder of his thesis ([19] , [20] , see “Contract Optimisation” below), as it implied that modelling the generators as risk averse would not significantly alter their behaviour.

Aspects of this Scott’s earlier work were also incorporated into Andrew Kerr’s PhD thesis, which are discussed in “Utility Maximising Reservoir Management” below.

  1. S. R. J. Batstone and T. J. Scott, Long-term Contracting in a Deregulated Electricity Industry: Simulation results from a Hydro Management Model , In Proceedings of the ORSNZ 33rd Annual Conference (1998),  p. 147-156.
  2. E.G. Read and M. Yang  "Analytical Dual DP for Reservoir Management with Correlation" Water Resources Research, Vol 35, no 7, 1999, p. 2247-2257

Top

Utility Maximising Reservoir Management

Andrew Kerr’s PhD thesis, entitled “ Stochastic Utility Maximising Dynamic Programming Applied to Medium-Term Reservoir Management” [16] , applies a technique initially developed by Grant Read, John Kaye and Shantha Ranatunga (University of New South Wales, Australia), to the problem of medium-term reservoir management with inflow uncertainty.  To quote from Andrew’s abstract:

Medium-term reservoir management is a classic planning problem to which stochastic dynamic programming has been applied.  An aspect of reservoir management modelling often neglected is 'risk', although it has been identified as being of prime importance.  A utility function can imply an attitude to risk, and in this thesis, a modified stochastic dynamic programming model (SUMDP) is presented which can maximise expected utility, where utility is defined over the range of terminal storage and 'wealth' outcomes and hence is dependent on all decisions made over the planning horizon. 

SUMDP is applied to reservoir management in regulated and deregulated representations of the New Zealand electricity system.  Experimental results showed that increasing the relative risk aversion to low terminal wealth values reduced the mean and variability of wealth and was achieved by conserving water and hence increasing storage.  This effect was amplified by the contract level of the hydro firm in a deregulated case where the reservoir firm was a price setter with financial contracts and the remaining players were price takers. 

SUMDP can be applied to other problem classes, one of which is stochastic route choice in acyclic networks [17] SUMDP is discussed in this context and applied to some example problems.  Rather than a single (static) route choice decision being optimal at each node of the network, SUMDP produces optimal non-static decisions which are dependent on the accumulated time taken to reach the node and take into account the utility associated with the time taken to travel the route.  There are few approaches discussed in the literature which produce non-static solutions, consider uncertainty, and consider risk, so SUMDP also contributes to this literature.” 

Top

Steve Batstone’s work also considers the effect of long term contracts and risk neutral, profit maximising reservoir management methods on profit risk.  See “Contract Optimisation” below for a fuller description.

References

  1. A. L. Kerr, Stochastic Utility Maximising Dynamic Programming Applied to Medium-Term Reservoir Management, Ph.D. Thesis, Department of Management, University of Canterbury, 2001.
  2. A. L. Kerr, Utility Maximising Dynamic Route Selection in Cyclic Stochastic Networks in, Proc. of The First Western Pacific and Third Australia-Japan Workshop on Stochastic Models in Engineering, Technology, and Management, eds R J Wilson, S Oaski, M J Faddy, 23-25 Sept 1999.

 

Scenario Development

Although many Mathematical Programming techniques have been developed for application to decision making under uncertainty, these techniques are based on three implicit assumptions:

  1. Probabilities can be determined for the outcomes of the uncertain parameters, and
  2. The decision maker is risk neutral, and
  3. All of the decision maker's concerns can be included in the formulation. 

While there are many decision making situations for which these assumptions are appropriate, there are many other situations for which they are not.  In particular, these assumptions are seldom supportable for strategic decision making problems. Strategic decision making must consider possible future events that have seldom, if ever, occurred before, and for which probabilities cannot be determined.  Because the situation will occur only once, and the decision will have a large impact, the decision maker is unlikely to be risk neutral.  Finally, the decision makers will often have concerns that cannot be represented in a mathematical programming formulation.

Olly McCahon’s PhD thesis presents an approach to decision making under uncertainty that relaxes the three assumptions listed above.  He assumes that the uncertain future can be described as a small set of scenarios.  These scenarios can be considered to have separate, competing objectives, because decisions that prepare well for one scenario generally prepare poorly for the others.  The problem is formulated as a multiobjective optimisation problem, and a set of non-dominated decisions is found.  The decision maker can choose a decision from this set according to his/her attitudes to risk, and to account for other requirements that cannot be represented in a mathematical programming formulation.  This approach is developed for problems with continuous variables, and then extended to problems that include binary variables.

Olly’s thesis was successfully examined in May 2001. 

  1. O. C. McCahon, Noninferior Set Scenario Analysis, Ph.D. Thesis, Department of Management, University of Canterbury, 2001.

Top

Contract Optimisation

Several analyses of imperfect electricity markets, such as the work of Scott described above, have considered the effect of a fixed level of forward contracts on the gaming behaviour of dominant firms in electricity markets, and other commodity markets.  Stephen Batstone’s thesis extended this perspective to consider the re-negotiation of these contracts, especially given the effect that past spot market behaviour might have on contract negotiations.  An equilibrium model of joint spot and contract market equilibria is developed, under the assumption that firms behave according to Cournot conjectures in both markets.  While a supply function analysis would have been more realistic, especially for the spot market, the explicit consideration of the effect generators’ output decisions have on spot price variance rendered such an analysis analytically complicated.  Thus the Cournot results are interpreted as bounds on the likely behaviour of the firms.

A key result of the thesis is that when generators consider the effect their output decisions have on the distribution of spot prices, used by consumers in the formation of their demand for contracts, they face incentives to “destabilise” the market.  While generators will naturally vary their output in response to the cost “state” they observe, they amplify this variation, by departing from the short-run profit maximising output level, in order to increase the risk premiums they receive on their contract sales.  In particular, since we assume that consumers only consider high price periods to be “risky”, the generators will push prices even higher in high-cost states, than they would in the absence of consumer risk-aversion or a contract market.  This behaviour yields increases in profit, for the generators, of around 5% [19] .

References

  1. S. R. J. Batstone, An Equilibrium Model of an Imperfect Electricity Market, In Proceedings of the ORSNZ 35th Annual Conference (2000),  p. 169-178.
  2. S. R. J. Batstone, Aspects of Risk Management in Deregulated Electricity Markets:  Market Power, Storage and Long-term Contracts, Ph.D. Thesis, Department of Management, University of Canterbury, 2002.

Top

People

Academic Staff

  • E. Grant Read
    Dr Read founded and lead the Energy Modelling Research Group.  He has been closely involved with the electricity sector for nearly 30 years, both as a researcher and a consultant.  After graduating with an Operations Research PhD on “Optimal Operation of Power Systems”, he spent several years at the New Zealand Ministry of Energy, and at various US universities, before joining the academic staff in 1985.  For many years he held a part time appointment, while also acting as a part time consultant, working independently, and also with leading national and international consultancies, including Ernst and Young, CORE, Putnam Hayes and Bartlett (PHB), PA, and Charles River Associates.  He has now resigned from the university staff, but still holds the honorary position of Adjunct Professor in Management Science.  He is a Past President of the Operations Research Society of New Zealand.  

    For many years Dr Read’s research focussed on economic dispatch, reservoir management, expansion planning and pricing issues, and on the development of techniques for such application to such problems, and to related areas, including oil, coal and gas stockpiling.  More recently he has focussed on electricity sector restructuring and market issues, being a key adviser to the Electricity Sector Task Force and the Wholesale Electricity Market Study in New Zealand, while undertaking a variety of international consultancies, for both corporate and governmental clients.  In particular, he has taken a lead role in the study of generator gaming issues, and in theoretical development and practical implementation of mechanisms for energy/reserve market co-ordination, nodal spot pricing, energy contracting, inter-regional trading, and transmission infrastructure cost recovery. 

    Much of this work was incorporated into the design of the New Zealand electricity market.  He has subsequently played key advisory roles with respect to the Victorian gas market, and the ongoing development of both the New Zealand and Australian national electricity markets, chaired a national electricity industry committee in New Zealand, and been heavily involved in designing and implementing new electricity market arrangements in Ontario, Malaysia, Singapore, and the Philippines.   
  • John A. George
    In 1998, John left his position as Associate Professor of Management Science and Head of Department of Management, to join a leading international business consulting firm, Putnam Hayes and Bartlett, in Wellington.  That practice has since been taken over by PA Consulting, and John is now responsible for team of OR/MS consultants working mainly on electricity sector problems.  Major projects have included development of electricity market rules for the Philippines, and implementation of electricity market systems for Singapore.
  • Deb Chattopadhyay
    Deb Chattopadhyay has worked extensively on development of optimization models to address power system planning and operations problems. He was a research fellowe with EMRG, and taught Managemnent Science in the Department of Management, University of Canterbury between 1997-2000.  He is now is a Principal in the New Zealand office of Charles River Associates (CRA), a leading international business consulting firm.. His area of consulting and research include a wide range of activities in the electricity industry including market design, market modeling, pricing, environmental issues and power system engineering aspects. 

Doctoral Students

  • Stephen R. J. Batstone
    Stephen submitted his PhD thesis, entitled “Aspects of Risk Management in Deregulated Electricity Markets:  Market Power, Storage and Long-term Contracts” for examination in May 2002.  He was briefly involved in some private consulting work with Grant Read, and Deb Chattopadhyay/Andrew Kerr at Charles River Associates, but is now based in Auckland as a Strategy Analyst with Mighty River Power Ltd.
  • Gavin J. Bell
    Gavin’s PhD thesis, entitled “Non-convex Network Models and Applications to Energy Modelling” was submitted and examined successfully in 1998.  He is now co-founder and director of a risk management consultancy firm, Open-Vision GMbH, based in Germany.
  • Andrew L. Kerr
    Andrew submitted his PhD thesis, entitled “ Stochastic Utility Maximising Dynamic Programming Applied to Medium-Term Reservoir Management”, for examination in January 2002.  He left EMRG in late 2000 to join PA Consulting in Wellington.  Six months later he joined Charles River Associates where he is a consultant working primarily on energy-related projects.
  • Olly C. McCahon
    Olly’s PhD thesis, entitled “Noninferior Set Scenario Analysis” was examined successfully in May 2000.  Olly continues to work with large-scale math programs in his position as Manager of Business Optimisation for Fletcher Challenge Forests Ltd, based in Rotorua.
  • Tristram J. Scott
    Since submitting his PhD thesis entitled "Hydro reservoir management for an electricity market with long term contracts" in 1998, Tristram has been working as a consultant to the energy sector.  Tristram worked first for Caminus, in the UK, taking on the role of manager of the Cambridge based Model Development Group. 
  • During that time he (among other things) authored the GasOptions suite of software, which received the Energy and Power Risk Management award of Innovation of the Year, 2001. 
  • Tristram left Caminus in April 2002, and now works as an independent energy consultant based in Cambridge, England.  Tristram's primary focus is on building optimisation and simulation based models for the energy sector.  He is an expert in stochastic dynamic  programming, a technique which he has applied to a wide range of problems in the energy sector, especially in the areas of hydro reservoir management, gas storage optimisation, and short term power station optimal dispatch.
  • Bhujanga Chakrabarti
    Bhujanga Chakrabarti, an engineer working for Transpower in Wellington, joined the EMRG group during this period, and plans to submit his PhD, entitled “Reactive Power Management and Pricing”, in 2002.