| The deregulation to power industries around the world has been undergoing morethan twenty years with large amount of experience achieved. Due to the importance andparticularity of power industries, keeping power systems efficient, economic, safe andreliable real-time operation, and long-term resource adequacy and sustainabledevelopment are the objectives of successful power markets operation. While it isemphasized in generation competition and transmission open access, the deregulation toelectricity demand is essential to power markets' further development.From the technical and economical points of view, the dissertation focuses ondeeply researching some of these technical and economical problems facing electricitydemand sides in power markets.Electricity prices are the focus of power markets. Accurate forecasting electricityprice has large effects on all participants' benefits. After an analysis to the formationmechanism of electricity prices, especial market clearing price (MCP) in power markets,and to the conventional forecasting MCP approaches, the thesis proposes a novelapproach for the first time that the residual error of the conventional forecasting modelis also modeled for forecasting, and the forecasted model errors are then used to modifythe forecasted prices in order to improve the accuracy of forecasting. The forecastederrors are also applied to estimate the confidence interval of the real prices. For thoseprice spikes that are difficult to be dealt with by the conventional approaches, waveletanalysis with good localized characteristic is introduced to forecast these verychangeable prices integrated with the conventional approach. The numerical tests showthe proposed approach is easy for modeling with merits in improving the accuracy offorecasting, and providing sufficient information.The thesis systematically discusses the newly emerging decision-making problemsof the load service entities (LSEs) procuring electric energy in power markets. Based onModern Portfolio Theory, the procurement decision-making framework consisted ofthree subproblems: market analysis, portfolio analysis and portfolio selection, isproposed, and the general procurement model for portfolio selection is set up. Thequalitative analysis is discussed on both spot markets and bilateral contracts to revealthe correlation among bilateral contract prices and procurement cost and differentallocation schemes, some instructive conclusions are derived out. The numerical testsare based on the electric energy procurement selection of the LSE with small-sizedgeneration capacity and reserve obligation, and the results show the proposed approachreasonable and effective.Risk in power markets, especial one faced by LSEs is surveyed. Risk managementmethods: diversification and hedging, commonly used in financial businesses, are thenintroduced for the LSEs to manage energy procurement risk. Risk assessment approach:Value at Risk (VaR) is applied to assess the risk of energy procurement portfolioselection between long-term contracts and spot markets. The ratio of profit over VaR ispresented to compare different procurement portfolio selections. When the LSEs'attitudes toward risk are considered, the combination of utility function and VaR is usedto determine the optimal procurement portfolio, and to assess risk level of procurementschemes. Numerical tests illustrate that the presented approach provides the LSEs amethodology solving long-term electric energy procurement selection and riskmanagement.The operation mechanism and evaluation methods to demand side management(DSM) in power markets are discussed. Multi-auction mechanism is designed to manageinterruptible load for imcomplete competitive power markets, and demand side biddingis discussed in detail for competitive power markets. The evaluation methods to DSM'seffects on the generation and transmission system reliability are analyzed by numericalexamples. Finally, the DSM realization mechanism suitable for the Chinese PowerMarket is suggested.
|
PDF Full Text Request