Short-term load forecasting model based on ridgelet neural network optimized by particle swarm optimization algorithm

Abstract

In this paper, the short-term load forecasting model based on ridgelet neural network optimized by the particle swarm optimization algorithm is proposed. The ridgelet neural network is simulated based on the visual cortex of the human brain. Compared with the traditional neural network, the neurons of the ridgelet neural network have directional characteristics, which can receive more dimensional information and have the ability to process higher dimensional data, and can better approximate nonlinear high dimensional functions. The particle swarm optimization algorithm is used to train the ridgelet neural network in this paper. The learning algorithm can not only speed up the convergence of the network, but also greatly reduce the probability of getting into the local minimum in the learning process. Through the simulation using the actual load data of power grid, simulation results show that the proposed model can effectively realize load forecasting and achieve the engineering accuracy requirements.