The deregulation of energy markets around the world, including power markets has changed the way operating assets in these markets are managed. Independent power asset owners and even utilities operating in these markets no longer operate their assets based on the cost of service approach that prevailed under regulation. Just as in other competitive markets, the objectives of asset owners in power markets revolve around maximizing profit for their shareholders. To this end, financial valuation of physical assets in power markets should incorporate different strategies that are used by asset operators to maximize profit. A lot of observed strategies in power markets are driven by a number of factors, the key among which are:
• asset operators are no longer obligated to supply service or manage their assets in certain prescribed ways, rather they have rights to operate, within applicable market rules, using techniques that maximize their profits,
• revenues are driven by uncertain market factors, including power price, cost and/or availability of fuel stock and technical uncertainties, and
• power assets have physical operating and equipment constraints and limits.
Having flexibilties (“options”) to optimize their assets (inline with shareholders’ objectives), rational asset managers react strategically to gradual arrival of information , given applicable equipment constraints, by revising previous decisions in such a way that only optimal (or near optimal) decisions are implemented. As a result, the appropriate approach to valuing power assets in competitive markets must account for managerial flexibilities or “real options” in the presence of uncertainties and technical constraints.
The focus of this work is to develop a robust valuation framework for physical power assets operating in competitive markets such as peaking or mid-merit thermal power plants and baseload power plants. The goal is to develop a modeling framework that can be adapted to different energy assets with different types of operating flexibilities and technical constraints and which can be employed for various purposes such as capital budgeting, business planning, risk management and strategic bidding planning among others. The valuation framework must also be able to capture the reality of power market rules and opportunities, as well as technical constraints of different assets.
The modeling framework developed conceptualizes operating flexibilities of power assets as “switching options’ whereby the asset operator decides at every decision point whether to switch from one operating mode to another mutually exclusive mode, within the limits of the equipment constraints of the asset. As a current decision to switch operating modes (in the face of current realization of relevant uncertainty factors) may affect future operating flexibilities of the asset and hence cash flows , a dynamic optimization framework is employed. The developed framework accounts for the uncertain nature of key value drivers by representing them with appropriate stochastic processes. Specifically, the framework developed conceptualizes the operation of a power asset as a multi-stage decision making problem where the operator has to make a decision at every stage to alter operating mode given currently available information about key value drivers. The problem is then solved dynamically by decomposing it into a series of two-stage sub-problems according to Bellman’s optimality principle. The solution algorithm employed is the Least Squares Monte Carlo (LSM) method.
The developed valuation framework was adapted for a gas-fired thermal power plant, a peaking hydroelectric power plant and a baseload power plant. This work built on previously published real options valuation methodologies for gas-fired thermal power plants by factoring in uncertainty from gas supply/consumption imbalance which is usually faced by gas-fired power generators. This source of uncertainty which has yet to be addressed in the literature, in the context of real options valuation, arises because of mismatch between natural gas and electricity wholesale markets. Natural gas markets in North America operate on a day-ahead basis while power plants are dispatched in real time. Inability of a power generator to match its gas supply and consumption in real time, leading to unauthorized gas over-run or under-run, attracts penalty charges from the gas supplier to the extent that the generator can not manage the imbalance through other means. A savvy gas-fired power plant operator will factor in the potential costs of gas imbalance into its operating strategies resulting in optimal operating decisions that may be different from when gas-imbalance is not considered. By considering an illustrative power plant operating in Ontario, we show effects of gas-imbalance on dispatch strategies on a daily cycling operation basis and the resulting impact on net revenue. Results show that a gas-fired power plant is over-valued by ignoring the impacts of gas imbalance on valuation.
Similarly, we employ the developed valuation framework to value a peaking hydroelectric power plant. This application also builds on previous real options valuation work for peaking hydroelectric power plants by considering their operations in a joint energy and ancillary services market. Specifically, the valuation model is developed to capture the value of a peaking power plant whose owner has the flexibility to participate in a joint operating reserve market and an energy market, which is currently the case in the Ontario wholesale power market. The model factors in water inflow uncertainty into the reservoir forebay of a hydroelectric facility and also considers uncertain energy and operating reserve prices. The switching options considered include (i) a joint energy and operating reserve bid (ii) an energy only bid and (iii) a do nothing (idle) strategy. Being an energy limited power plant, by doing nothing at a decision interval, the power asset operator is able to time-shift scarce water for use at a future period when market situations are expected to be better. An illustrative example considered shows the impact of the different value drivers on the plant’s value and dispatch strategies. Results show that by ignoring the flexibility of the asset owner to participate in an operating reserve market, a peaking hydroelectric power plant is undervalued.
Finally, the developed valuation framework was employed to optimize life-cycle management decisions of a baseload power plant, such as a nuclear power plant. The applicability of real-options framework to the operations of baseload power plants has not attracted much attention in the literature given their inflexibility with respect to short-term operation. However, owners of baseload power plants, such as nuclear plants, have the right to optimize scheduling and spending of life cycle management projects such as preventative maintenance and equipment inspection. Given uncertainty of long-term value drivers, including power prices, equipment performance and the relationship between current life cycle spending and future equipment degradation, optimization is carried out with the objective of minimizing overall life-cycle related costs. These life-cycle costs include (i) lost revenue during planned and unplanned outages (ii) potential costs of future equipment degradation due to inadequate preventative maintenance and (iii) the direct costs of implementing the life-cycle projects. The switching options in this context include the option to shutdown the power plant in order to execute a given preventative maintenance and inspection project and the option to keep the option “alive” by choosing to delay a planned life-cycle activity. Results of an illustrative example analyzed show that the flexibility of the asset owner to delay spending or to suspend it entirely affects the asset’s value accordingly and should be factored into valuation.
Applications can be found for the developed framework and models in different areas important to firms operating in competitive energy markets. These areas include capital budgeting, trading, risk management, business planning and strategic/tactitcal bidding among others.
| Identifer | oai:union.ndltd.org:WATERLOO/oai:uwspace.uwaterloo.ca:10012/3436 |
| Date | January 2007 |
| Creators | Oduntan, Adekunle Richard |
| Source Sets | University of Waterloo Electronic Theses Repository |
| Language | English |
| Detected Language | English |
| Type | Thesis or Dissertation |
Page generated in 0.0033 seconds