Research On Power System Short-Term Load Forecasting Based On Data Mining

The short-term load forecasting is an important routine for power dispatch department. Especially with the foundation and the development of power market, short-term load forecasting will bring into play a more and more important role. Its precision directly influences power system's security, profit and quality. Therefore, how to improve the forecasting precision is the emphasis on the study of short-term load forecasting.The short-term load forecasting is constructs the mathematic model at the basis of analyzing the speciality of the real load and applying the knowledge of maths and computer to real course, finally obtains accurate forecasting results. Before analyzing the speciality of the real load and constructing the mathematic model, its necessary to processing the data to wipe off the bad-data which does not accord with the fact.Data mining is a rising domain of data-base and data-warehouse, it's a technology of data intelligent analysis which rised at the end of 20th century. Its strong suit is its powerful capacity of processing data and then discovering useful rule and relation. As a rising technology, it has broad foreground at applying to power load forecasting system. In this paper, we make great efforts to combine data mining knowledge with power system, and then obtain the short-term load forecasting method based on data mining.In data processing, this paper firstly picks up the character curve which is embodiment of natural load curve by method of Cluster analysis, then checks and modifies the bad-data according as the character curve. The results are well. In load forecasting, firstly obtains load hefts at different frequencies by wavelet transforming, then forecasts every heft by data classification and regression multianalysis, in the end fits every forecasting result together to get the final result. The method in this paper has been applied to the real load of Guizhou province, the example shows that this method can obtain good results in the electric load short-term forecasting.