Research and Implementation of SVM Optimized by Group Genetic Algorithm in Micro-grid Load Forecasting

Abstract

SVM algorithm in load forecasting needs artificial experience to set the parameters of C and kernel parameters of  , which will have a certain impact on the adaptability of the model. Therefore, the SVM has some shortcomings in the selection of model parameters: when meeting the large sample data modeling, parameter adjustment range will increase. Meanwhile, the number of model adjustments is too much and the modeling efficiency will be reduced. In this paper, we use the adaptive ability of group genetic algorithm to optimize the parameters in SVM so as to improve the optimization of the model.