Application of mutual information and improved support vector machine in power load forecasting

Abstract

So as to improve the accuracy of power load-forecast, the author first uses mutual information to determine the correlation between power load and other factors, and then uses Skyhawk optimization algorithm to optimize the kernel function and penalty coefficient in the support vector machine, which is verified by experiments. The results show that the power load has a strong correlation with temperature and atmospheric pressure, and the mutual information values are 1.5638 and 1.1042 respectively; Taking these two variables as the input characteristics of the forecasting model, the R2 value of power load-forecast is above 0.9, and good results are achieved. This research method provides a new idea for power load-forecast.