Power System Short-term Load Forecasting Based On Combined Model

Power sector as an important part of national economy, has a very important influence on maintain the healthy development of national economy and guarantee the stability of people’s life.Therefore,to ensure the sustained and effective operation of power grid is very important. With the continuous development of power industry, electric power system is also intelligent gradually, and under the new situation of green energy saving, the accurate load forecasting is more precious, load forecasting has become a job more and more important today.In the operation of the actual, load change process involves many unknown factors, so when doing the load forecasting is also need to consider the impact of these factors. 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.Firstly, this paper briefly introduced the electric power system short-term load forecasting and analyzes the significance of the research status at home and abroad; The second chapter presents the basic knowledge of the grey system theory and its modeling process in the short-term load forecasting.In the third chapter, the basic concept and the basic principle of SVM and the derivation of the improved S VM are introduced. The fourth chapter introduces the combination forecasting method based on gray and the improved support vector machine, calculates the weight of the two methods, built the optimal combination forecasting model for load forecasting, give the method to select the model parameters, and the correctness and effectiveness of the test algorithm are analyzed by the example simulation. The fifth chapter is introduced in detail the design of load forecasting software, which can allow users to realize in simple operation, can achieve real-time online on the load forecast,and get more accurate load forecasting results and other useful information. Finally, suggestions for further discussion are put forward.