The Research And Application Of Power System Load Foresting Method

With the development of our national electric industry, the management of electrified wire netting is more and more modernized. People pay more attention to the study on electric system load forecasting than ever.Firstly, the paper studies the characters and sorts of load. Based on the analysis of five domestic power utilization segment markets, the relations among electricity consumption of different industries and its influencing factors are built up and respectively modeled to perform comprehensive forecasting. On the basis of the deep analysis to load characters, the paper puts forward a method of corporation forecasting which selects one or a few effective load forecasting methods from many methods according to the factual influence elements. When the load of agriculture and illume is forecasted, the load is decomposed into stationary terms that is not related to meteorologic factors and random term that is related to meteorologic factors by use of wavelet analysis. Because the forecasting accuracy of stationary term is higher and the amplitude of random term, although it is difficult to forecast, is smaller, so the impact of forecasting error bringing about by random factors on final forecasting result is weakened.Considering the bad prediction of weather and economic data and the difficulty to obtain the data, it's not feasible to contact the load with its influencing factors some times. In fact, the weather information of each month is correlative and be comprised load series. The paper attempt to applied sequential cluster method to set up the classification standard, then it regards the correlation coefficients of record values as weights and predicts the future loads by using Markov chain model with weights. This method make the best of the information comprised in load series and solved the problem of obtaining weather information. Not only the concrete value of the monthly sales electric energy but its range in the future is gained. So the prediction is valuable.The method which was proposed in this article was tested by using the load data of Hunan province.At last, its application is introduced.