Electric power industry | EPFL Graph Search

The electric power industry covers the generation, transmission, distribution and sale of electric power to the general public and industry. The commodity sold is actually energy, not power, e.g. consumers pay for kilowatt-hours, power multiplied by time, which is energy. The commercial distribution of electricity started in 1882 when electricity was produced for electric lighting. In the 1880s and 1890s, growing economic and safety concerns lead to the regulation of the industry. What was once an expensive novelty limited to the most densely populated areas, reliable and economical electric power has become an essential aspect for normal operation of all elements of developed economies. By the middle of the 20th century, electricity was seen as a "natural monopoly", only efficient if a restricted number of organizations participated in the market; in some areas, vertically integrated companies provide all stages from generation to retail, and only governmental supervision regulated the rate of return and cost structure. Since the 1990s, many regions have broken up the generation and distribution of electric power. While such markets can be abusively manipulated with consequent adverse price and reliability impact to consumers, generally competitive production of electrical energy leads to worthwhile improvements in efficiency. However, transmission and distribution are harder problems since returns on investment are not as easy to find. Although electricity had been known to be produced as a result of the chemical reactions that take place in an electrolytic cell since Alessandro Volta developed the voltaic pile in 1800, its production by this means was, and still is, expensive. In 1831, Michael Faraday devised a machine that generated electricity from rotary motion, but it took almost 50 years for the technology to reach a commercially viable stage. In 1878, in the United States, Thomas Edison developed and sold a commercially viable replacement for gas lighting and heating using locally generated and distributed direct current electricity.

Investment decisions with flexibility under uncertainty in the competitive electricity supply industry

Jun Dong

The restructuring of the electricity supply industry (ESI) has introduced new actors and market mechanisms. Electricity prices are more market oriented and fluctuate greatly. These changes bring about more uncertainties and risks to the market participants. When there are high risks in the future, the actors tend towards short-term decisions in which the risks are relatively easier to be coped with than in the long-term decisions. Much research has been carried out on short and medium-term risk management. The contracts of financial derivatives are managed for hedging risks. However, the development of long-term planning and risk management methods in the liberalised ESI has so far not attracted much attention. Long-term anticipations and planning have always been a significant issue given the specific characteristics of the ESI and the importance of security of supply. In liberalised electricity markets, the traditional centralised least-cost generation capacity planning is less relevant. The planning activities have become decentralised, i.e., individual actor makes planning for achieving its goals under certain risks. The ESI is in a transition from monopoly to competitive market and the liberalised electricity market has a short history. It is dominated by structural changes and actors' game behaviour, which are usually not of a random nature. Therefore, it is not always able to model the long-term uncertainties with the widely used probability methods in risk management. Under the large uncertainties and with the difficulty of uncertainty modelling, how would the actors make long-term investment decisions and manage risks? This thesis is an attempt to answer this question. The problem under study in this research is generalised as "investment planning with flexibility in the competitive electricity market". We argue that to select a flexible alternative strategy is one of the ways for managing long-term risks. The flexibility means that the actor is capable of adapting the investment decisions under different market situations in the future. For example, a project can be deferred to wait for more information, be expanded if the market situation becomes favourable or be abandoned if the market situation becomes unfavourable. For an electricity generator, the problem is capacity planning, for example, to construct a new power plant. For a consumer, it is energy portfolio planning. The alternatives may include different purchasing contracts and investment for self-generation. The purpose of this research is to develop a new approach for dealing with these kinds of problems. The requirements of the new approach are proposed in this thesis as: The flexibility of the investment projects or strategies should be accounted in the decision-making; The uncertainty modelling should reflect the long-term trends and characteristics of the liberalised electricity markets; Electricity price forecasting should reflect the interactions among the market players; The approach is an integration of scenario building, market simulation, strategy evaluation and decision-making. The new approach is developed within a conceptual framework of an intelligent decision support system proposed by Dr E. Gnansounou. The framework consists of four modules: Scenario builder, Problem formulator and attributes evaluator, Electricity market multi-agent system and Decision making aid. The principal researches focus on the methods of attributes evaluation taking into account flexibility actions and agent-based market simulation. The attributes evaluation of investment alternatives emphasizes the incorporation of uncertainty modelling and flexibility actions. Two types of attributes evaluation methods are developed using real options theory and fuzzy set theory respectively. Real options theory, in which uncertainty is modelled with probability methods, is introduced for computing and incorporating the flexibility value. In addition, a new method is developed for attributes evaluation taking into account the flexibility actions, using fuzzy set to represent imprecise or vague knowledge in electricity markets. For a long-term perspective, the main factors influencing investment decisions, such as electricity prices, are affected mainly by the structural changes and strategic behaviours of the actors in the market. It is not pertinent to forecast the values of these variables from the historical data using statistic models. The agent-based market simulation system is developed to simulate the interactions among the market participants and to provide electricity price scenarios. Three applications are dealt with by applying the proposed approach. Firstly, agent-based market simulation is applied to an illustrative power system. Secondly, a generation capacity planning and project evaluation for an electricity generator is performed. Thirdly the energy portfolios of a consumer is evaluated and compared. The research found that the new method to take into account flexibility actions is practical. It incorporates the flexibility actions into the cash flows. The flexibility cost is given exogenously. The value of a base strategy with flexibility actions is calculated. The alternatives are compared with these values. It is different from the real options method. The real options method assumes that the future value of the strategy is known with a predefined probability distribution and the value of the option is calculated. It deals with the pricing of a contract or a project. Differently from that, the new method deals with the comparison of portfolio given the prices and the costs are assumed as known and included into the cash flows, and how to consider the flexibility in the portfolio comparison. The real options method is operative for incorporating flexibilities in investment decisions if the uncertain variables follow the Wiener stochastic process. The advantages of this method are that it provides flexibility values and optimal timing of a project; it models the future value of the project with the stochastic process therefore it does not require the estimations on the values of driving variables for the lifetime of the project; it uses risk-free rate that is easier to estimate than the discount rate used in the DCF method. The limitations are that the assumptions of probabilistic distributions of the future value of the project are not always satisfied. The determination of the stochastic process and the estimation of the volatility of the value of the project are not straightforward. As a complement to the real options method and dealing with a different problem, the fuzzy method is proved in this thesis to give a "softer" approach when modelling the imprecision with fuzzy numbers. This method is suitable to electricity markets with incomplete information, lack of sufficient data and non-random movement. It is found that the agent-based simulation model is effective for giving "what if" scenarios on electricity market states in the future. The model is flexible in simulating imperfect electricity markets where game play exists. The main contributions of the research are summarised as follows: A new approach of investment decision support is developed fitting to electricity markets with regard to uncertainty and flexibility modelling. Incorporation of flexibility into the investment decisions is developed as a way for managing long-term risks. Uncertainty modelling with the real options method and imprecision data modelling with the fuzzy method can be applied according to the pertinence to the situation of electricity markets. The attributes evaluation of the investment strategies is based on agent-based market simulation. Instead of traditional least-cost centralised planning, the generation expansion planning is considered as decentralised planning where the individual agent makes expansion planning, taking into account other agents' plans.

EPFL2004

Investment decisions with flexibility under uncertainty in the competitive electricity supply industry

Jun Dong

EPFL2004