Some Theories And Their Applications Research Of Local Power Network Load Forecast And Optimization Mode

Along with the power industry has gradually entered the electricity market, the property right of power plants is becoming diversified, the contradictions between electricity supply and demand continuously highlight, how does the power control sector develop power generation dispatch plan and realize the economic dispatch of electricity production to ensure maximum economic benefits, for the electricity enterprises, load forecast has become an important part of the power system managemen modernization. The study for power network load forecast has become an important area in modern power system science.Load forecast will be affected by population growth, economic development rate, industrial structure change, weather conditions and other factors. In addition to the selection of reliable analytical methods, forecast accuracy also depends on judgment ability of the tipster. Load forecasting has a history of several decades, and therefore there are many basic theories and practice methods to solve the problem of load forecast. But with the constant development of power systems, increasing complexity of power network and the establishment of the electricity market, it is needed to study load forecast continuously and to study new forecasting methods by using new theories and new mathematical tools. Therefore the core of this paper is to establish a new higher accuracy methodology of power network load forecasting by using a new mathematical means in information processing area----artificial immunization model.After carrying on analysis on electric power system constitution, electric power industry characteristic and the present application situation of power network load forecast, this paper summarizes the changing characteristics of current power supply situation. And it conducts systemic and comprehensive research, analysis and appraisal to each method of existing power network load forecast. Based on the characteristics of local power network load forecast, new research direction of power network load forecasting based on artificial immunization system is established. Field research and specific practice are also conducted. Using physiology principle of immunity system and information processing characteristics, it distinguishes the immune system with ANN and GA by using contract method. The model of power network load forecast based on artificial immunization algorithm is established for the first time. And the reasons which cause the errors are analyzed. According to the actual situation of power network operating in Fuxin region and based on historical data of power network load, this paper carries on forecast for power network load situation in Fuxin electricity market, including the demand in the next five years and social consumption and maximum load the next 10 years. Selecting load forecast as a starting point, it carries on optimization design for urban power network equipment distribution, and creatively presents ETA(Equipment-target assignment)issues and provides a mathematical description and solving method. In this process some useful improvements are added to artificial...