Study On The Model Of Mid-long Term Load Forecasting For The Distribution Network

Load forecasting plays an important role in the power system planning and running. It is the basis of the power system planning and construction, and also is the premise of reliable supplying and economic running. The accuracy of load forecasting directly affects the rationality of investment, network layout and network running.Not only the forecasting of the load gross in the district, but also the spatial load forecasting (SLF) should be done for the mid-long term load forecasting of the distribution network planning. Accordingly the standards of the new equipments, the time and the location of installation could be determined as well as the reasonable planning could be made. This paper discusses the methods of load forecasting fitting for the mid-long term load forecasting in detail. On the basis of the analysis of examples, all the methods are compared, the merits and demerits of which are also summarized. Different methods offer the dissimilar information and accuracy of forecasting, and then the combined load forecasting is presented. In this paper one combined load forecasting method based on fuzzy synthetic evaluation is advanced, in which the experience of the forecasters and the uncertainness of load forecasting are considered adequately. Traditionally the method of fiizzy logical reasoning is adopted for the small area land-use analysis, but the calculation process of this method is very complex and its expansibility is too limited. So a fuzzy inference model based on artificial neural network (ANN) is put forward to analyze the properties of the small area. This method has the merits of improving the accuracy of land-use analysis as well as simplifying and mending the arithmetic of SLF.