Research On Power Load Forecasting Based On Grey System Theory

Power load forecasting is the basis for the development of power grid development planning, the determination of the target grid structure, the development of investment plans and project construction plans. Accurate load forecasting, to reasonably determine the construction schedule and the investment plan of power grid project, improve investment efficiency and sustainable development ability of the enterprise, to meet the social and economic development of urban and rural residents with electricity demand and ensure the production and life of electricity demand has important significance.To plan the work itself in as the starting point, in view of the current planning work in the load forecasting accuracy is not high, the lack of theoretical support for power grid planning problems such as science, this paper summarizes and analyzes the current common power load forecast method, such as elastic coefficient method, single consumption method, load density method and so on, the application of various prediction methods and their advantages and disadvantages are summarized. Secondly, combing the theory of grey system theory and the GM(1,1) model based on the grey system theory, the process and method of the model construction are studied. With the powerful noise reduction ability of Fourier series, the residual correction method based on Fourier series is studied, and the GM(1,1) model is improved and optimized, which greatly improves the accuracy of the model prediction. Finally, this paper takes the Laiwu 2006-2015 annual electricity consumption data as an example, using the residual error correction GM(1,1) model based on Fourier series, the 2016-2020 years of electricity consumption is predicted, and the feasibility of the model is verified.Based on the residual Fourier series of modified GM(1,1) model is to further optimize the original GM(1,1) model, the new model corrected, precision accuracy is obviously improved compared with the original model, it has certain reference value and significance for power load forecasting and power grid planning work.