基于遗传BP神经网络的短期风速预测模型

Short-term wind speed forecast model for wind farms based on genetic BP neural network

为了提高风电场短期风速预测精度,提出将遗传算法和反向传播(BP)神经网络相结合的预测模型.采用自相关性分析找出对预测值影响最大的几个历史时刻风速,以历史时刻的风速、温度、湿度和气压作为BP神经网络预测模型的输入变量;利用遗传算法的全局搜索能力获得BP神经网络优化的初始权值和阈值;采用优化后的BP神经网络分别建立1、2、3 h的短期风速预测模型.实验结果表明,该方法较BP神经网络具有预测精度高、收敛速度快的优点.

To improve the short-term wind speed forecasting accuracy for wind farm, a prediction model based on back propagation（BP） neural network combining genetic algorithm was proposed. Autocorrelation analysis was used to discover historical wind speeds which have significant influence on predicted wind speed. The input variables of BP neural network predictive model were historical wind speeds, temperature, humidity and air pressure. Genetic algorithm was used to optimize the weights and bias of BP neural network. Optimized BP neural network was applied to predict wind speed an hour before, two hours before and three hours before individually. The simulation results show that the proposed method offers the advantages of high precision and fast convergence in contrast with BP neural network.