基于遗传算法的支持向量机短期风速预测

对风电场风速实现较准确的预测，可以有效减轻并网后风电场对电网的影响。支持向量机模型的预测精度在很大程度上依赖于模型参数的选择，为提高预测模型的泛化能力和预测精度，应用遗传算法选择支持向量机的模型参数，再根据选择的参数对小时风速进行预测。实验结果表明本文方法能够获得较高的风速预测精度。

Forecasting of Short-term Wind Speed with Support Vector Machine Based on Genetic Algorithms

Giving a high precise wind speed forecast for wind farms can effectively relieve disadvantageous impact of wind power plants on power systems. The selection of parameters for Support Vector Machine （SVM） has a large impact on the forecasting accuracy. For enhancing the generalization performance and prediction accuracy,genetic algorithms （GA） are applied to select parameters for SVM model in this study, and then hourly wind speed are forecasted according the selected parameters. The simulation results show that using the model proposed on the paper to predict wind speed can achieve a high accuracy.