考虑主环境因素的GWO-SVR风电功率超短期预测  

Ultra-short-term wind power prediction considering main environmental factors based on GWO-SVR

作　　者：徐炜君[1] XU Weijun(Department of Electrical Information Engineering,Northeast Petroleum University at Qinhuangdao,Qinhuangdao 066004,China)

机构地区：[1]东北石油大学秦皇岛校区电气信息工程系,河北秦皇岛066004

出　　处：《电子设计工程》2023年第15期150-156,共7页Electronic Design Engineering

基　　金：黑龙江省属本科高校引导性创新基金项目(2018YDQ-02)。

摘　　要：随着我国风电产业的高速发展,风电功率预测的作用也愈显突出。提高风电功率超短期预测的稳定度、速度和精度,是更加合理地利用风电的关键技术之一。在深入分析影响风机出力主要环境因素的基础上,对真实风电场的环境监测历史数据进行了降维处理,以风速、风向、温度及湿度四个主环境因素作为GWO-SVR预测模型的训练和测试集,进行预测分析。不同预测模型的对比表明,降维处理可有效降低模型的复杂程度,降低无用数据对预测结果的影响,预测结果的稳定性、速度及精度均有提高。With the rapid development of China’s wind power industry,the role of wind power prediction is becoming more and more prominent.Improving the stability,speed and accuracy of ultra-short-term prediction of wind power is one of the key technologies to make more rational use of wind power.Based on the in-depth analysis of the main environmental factors affecting the output of the wind power generator,the dimension reduction processing is carried out on the historical environmental monitoring data of the real wind farm,taking the four main environmental factors of wind speed,wind direction,temperature and humidity as the training and test set of GWO-SVR prediction model,the prediction analysis is carried out.The comparison of different prediction models shows that the dimension reduction processing can effectively reduce the complexity of the model,reduce the impact of useless data on the prediction results.The stability,speed and accuracy of the prediction results are improved.

关 键 词：环境因素 风电功率 支持向量回归机 灰狼优化算法 超短期预测 

分 类 号：TM614[电气工程—电力系统及自动化]