Power Plant Decision Support System Research And Development

With the reform of the power industry, China has established the power generation market. The power plants were separated from the traditional power industry system in which the generation, transmission and distribution belong to a vertical integration. Power companies are facing not only the risk of fuel cost but also the price competition in the market. Under this environment, it is urgent for the generation companies to strengthen the cost management and make rational bidding strategy.Based on the cost management theory, this dissertation analyses the consumed cost of the power plant during production and business process. According to the correlation with output, the cost is classified as fix cost and variable cost. With the application of regress theory and probability theory, the forecasting model of the cost is established, and then the output and profit are known. This is the foundation of the cost management and control.Before the trade-off the managers need to know clearing the running cost. This dissertation particularly describes a method, which can form the characteristic curve of cost. The application of game theory and probability theory is presented. The strategy model is established via only adjusting the slope of bidding curve, and then the bidding of power market could be treated as two-person gaming by introducing the concept of "virtual rival". Furthermore, the problem of multi-person gaming and incomplete information, which remain very difficult in the game theory, could be efficiently worked out. The approach to solving the optimal strategy with game theory is developed. The optimal result on probability can be given by solving the simple model which limit of own maximum generation is not considered, and then the optimal bidding strategy based on the market load and the virtual rival bidding strategy changing, can be obtained under the consideration of limit of own maximum generation by using the parameter estimation to estimate the parameter of the virtual rival bidding curve to settle.Based on the theoretical study and real application, the framework of the decision support system is constructed in this paper. This system based on SCADA, DCS and MIS. It is closed loop which includes real cost analysis, bidding strategy and system management.