关键词: 短期风速预测, 关联规则, 粒子群优化算法, Elman神经网络

Abstract: In the short-term forecasting of wind speed, the traditional neural network is easy to fall into local extreme value and lack of dynamic performance, which leads to low accuracy of wind speed forecasting. In order to improve the prediction accuracy of wind speed, the wind speed forecasting model based on association rules and particle swarm optimization Elman neural network is proposed. PSO is used to optimize the parameters of Elman neural network model, which can improve the convergence rate of the algorithm and avoid falling into local extreme value, so as to get the best predictive value. And according to association rule, the meteorological factors are incorporated in the relevant algorithm.The Apriori algorithm is used to mine the association rules of wind speed and other meteorological factors, and the wind speed forecasting value is modified and compensated by the association rules. The experimental results show that the forecast effect of the proposed model is better than the traditional model, and it is proved that the combination of the association rules considering the meteorological factors can reduce the wind speed prediction error.

Key words: short-term wind speed forecasting, association rule, Particle Swarm Optimization（PSO）, Elman neural networks