The Neural Network Model Based on PSO for Short-Term Load Forecasting

A new algorithm for load forecasting - the neural network model based on particle swarm optimization (PSO-NN) for short-term load forecasting is proposed in this paper. The method is simple, easy to realize and its convergence rate is quick. The overall optimal solution of the problem can be found in great probability, and the intrinsic defects of artificial neural network, such as slow training speed and the existence of local minimum points, can be effectively overcome. Simulation results show that forecasting precision and speed can be improved by this method, and its forecasting capability is obviously better than the neural network model based on BP algorithm (BP-NN)