Short-term wind power prediction based on extreme learning machine

Abstract

Large-scale grid connection of green and clean energy has become a development trend in the energy field. Accurate prediction of wind power output power can reduce the negative impact of wind power on the power system. In this paper, we use extreme learning machine method to achieve short-term prediction of wind power, and determine the optimal number of hidden layer neurons by lattice method. In order to verify the effectiveness of extreme learning machine neural network, we compare the simulation results of extreme learning neural network method and BP neural network method. Simulation results show that the prediction accuracy of extreme learning machine method is similar to that of BP neural network. At the same, considering that extreme learning machine only need to calculate the weight matrix from the hidden layer to the output layer, so extreme learning machine has certain computing advantage compared with BP neural networks.