Power Load Forecasting Based on BP Neural Network Optimized by Genetic Algorithm

Power load forecasting has important research value for power grid coordination scheduling, power distribution load, regional production planning and so on. In order to improve the accuracy of power load forecasting, based on the BP neural network model, this paper optimizes the BP neural network parameters by genetic algorithm, organically combines their respective global search ability and local search ability, and updates and improves the weights and thresholds of the neural network. In the analysis of power load forecasting in a city, the prediction error MAPE value of the model is 2.0102 %, which is significantly higher than that of the single BP neural network model, and has reliable reliability and generalization ability.