ISSA-SVR Based Power Forecasting for Proxy Power Purchase Customers

With the steady progress of the power purchase business of the grid company, the power purchase business system is gradually improved, and the accurate prediction of the power consumption of the proxy power purchase users lays the foundation for the division to guarantee the safe and stable supply of power. Therefore, this paper proposes a power consumption prediction model for agency power purchase users based on ISSA optimized support vector regression (SVR) and differential autoregressive moving average (ARIMA), taking into account the feedback from the prediction result calibration. Firstly, the adaptive weighting factor is introduced to improve the performance of SSA algorithm, and the kernel function coefficient g and penalty coefficient C are solved in the SVR model to establish the ISSA-SVR prediction model; then the residuals of the prediction results are used to establish the ARIMA error prediction model to predict the error of the test set; and finally, the initial prediction results are revised by the error prediction results calibration feedback. The test results show that after the residual correction, the prediction accuracy is significantly improved compared with the traditional prediction model and the ISSA-SVR model without calibration feedback.