Short-term Wind Speed Forecasting Model Based on Spiking Neural Network

Abstract

Short-term wind speed forecasting plays an important role in the daily power system operation. Therefore, this paper presents a novel model based on spiking neural network (SNN) used spike response model (SRM). Further, to achieve both smaller training errors and higher precision forecasting, the basic SpikeProp learning algorithm is improved by adaptively adjusting the learning rate and adding momentum items. Then, this paper selects the actual sampling data from a wind farm to verify the effectiveness and advantages of the proposed model.