Power System Short-term Load Forecasting Based On Combined Model

The power load forecasting has a very important significance, not only for the management of the power system scheduling also affect the interests of the power plants, power supply and transmission companies. Power plays a very important role in China’s economic development, people’s living standards and social stability, with the continuous development of the electric power industry, the power system is also gradually intelligent, energy saving environment at the same time, the accuracy of the load forecasting is particularly important, load forecasting has also become an increasingly important task.Load forecasting may be affected by many uncertain factors, such as weather, economic policy impact. There are no kinds of methods which can ensure the results of the prediction can be satisfied in any case. In the concrete prediction problem, we should choose appropriate model combined with the actual situation.This paper briefly describes the significance of the power system load forecasting, the research status at home and abroad, and reviews load forecasting methods, describes several existing load forecasting methods, analyzes their strengths and weaknesses, and then deeply studies the grey system theory, time series method, and combined forecasting method. This paper mainly summarized the basic idea and process of the grey system theory which proposed by professor Deng Julong, the basic theory and modeling steps of times-series approaches which proposed by Box-Jenkins. On this basis, the paper puts forward a method that is the optimal combination of the grey system theory and the times-series approaches, and then uses the combined forecasting model for load forecasting. Forecast simulation for a specific forecast instance, according to the historical load data grid, and to verify the correctness and validity of the algorithm. Drawn combination forecasting method with single forecasting method through the analysis of simulation results, compared with a higher accuracy. The author Designed load forecasting software, and users can simply use the load forecasting software, in the process, users can see the entire load forecasting course, also can get load forecasting results and other useful information.