Research on Power Load Prediction Based on Elman Network with Improved Particle Swarm Optimization

Abstract Because electricity is so difficult to store, the electric power department needs store enough in advance, whether customers use it or not. To lower cost of this aspect, essential realize high-precision load forecasting. The Elman neural network (ElmanNN) prediction model based on improved Particle Swarm Optimization (IPSO) algorithm constructed. (PSO) brought in, and learning factor optimized for limitation that PSO can easily obtain local optimal solution. Finally, applied ElmanNN. ElmanNN solves problems slow training speed effortless minimum. simulation experiments verify has a smaller error prediction.