Research on Short-term Load Forecasting Model Based on iForest-GA-BPNN

The stable operation of the power system plays a vital role in the safety of electricity consumption and the production of electricity. In order to ensure the stability of electricity and adjust the power arrangement in time, it is very important to forecast the electricity load in the short time in the future. Without solving the problem of short term load forecasting, a short term power load forecasting model based on iForest-GA-BPNN is proposed in this paper. The model first uses the isolated forest algorithm to screen the collected power load data and supplement the data vacancy value. Then, based on the filtered data, the BP algorithm with improved weight and threshold value by genetic algorithm is used as the prediction model to forecast the short-term power load. Then, the performance of the proposed prediction model is verified by an example. The numerical results show that the proposed iForest-GA-BP algorithm has good prediction performance. It can be applied to the load forecasting of the power system, which can effectively improve the accuracy of the load forecasting, so as to provide theoretical support for the power consumption planning and power production of the power grid department, and improve the overall power consumption efficiency of the country.