Investigating the use of Echo State Networks for prediction of wind power generation

Abstract

This paper presents the results of models created for prediction of wind power generation using Echo State Networks (ESN). An echo state network consist of a large, randomly connected neural network, the reservoir, which is driven by an input signal and projects to output units. ESN offer an intuitive methodology for using the temporal processing power of recurrent neural networks without the hassle of training them. The models perform forecasting of wind power generation with 6 hours ahead, discretized by 10 minutes and with 5 days ahead, discretized by 30 minutes. These models use ESNs with spectral radius greater than 1 and even then they can make predictions with good results. The forecast horizons presented here fall in medium-term forecasts, up to five days ahead, which is an appropriate horizon to subsidize the operation planning of power systems. Models that directly predict the wind power generation with ESNs showed promising results.