Tabu Search Algorithm Optimized ANN Model for Wind Power Prediction with NWP

Abstract

Wind power prediction is of great importance for the safety and stabilization of grids. The most important and difficult problem now is to enhance the prediction precision. BP (Backward Propagation) neural network has been used extensively in wind power prediction. But BP network is apt to getting into local minima and its convergence rate is slow. Tabu search is a kind of intelligent algorithm, which can achieve the global optimizations. This paper put forward a wind power prediction model of BP neural network optimized by tabu search algorithm with memory function. The result shows that with appropriate input parameters, the wind power prediction model of neural network based on tabu search algorithm can improve the prediction precision as well as the convergence rate.