Research Optimizing Neural Network in Short-Term Electric Load Prediction

Abstract

In order to improve the prediction accuracy of short-term load, this paper puts forward a model of short-term load prediction (VPQPSO-BPNN) based on the optimized BP neural network. Firstly, by phase-space reconstruction, it conducts reconstitution for the historical data of short-term load, and then through VPQPSO algorithm, it finds the optimal parameters of BP neural network, and finally establishes the optimal short-term load prediction model, and test the model performance by simulation experiment.