基于优化聚类的组合风速短期预测  

Short-term wind speed forecasting of combined ELM based on optimal clustering

作　　者：陈记牢 栗惠惠 李富强 郝飞 张圆美 Chen Jilao;Li Huihui;Li Fuqiang;Hao Fei;Zhang Yuanmei(School of Mechanical and Electrical Engineering, Hohhot Vocational College, Hohhot 010060, China;StateGrid Jilin Electric Power Co., Ltd. Economic and Technical Research Institute, Changchun 130062, China)

机构地区：[1]呼和浩特职业学院机电学院,内蒙古呼和浩特010060 [2]国网吉林省电力有限公司经济技术研究院,吉林长春130062

出　　处：《可再生能源》2017年第12期1841-1846,共6页Renewable Energy Resources

基　　金：国家高技术研究发展计划"863"资助项目(SS2014AA052502)

摘　　要：准确的风功率预测对电力系统安全、稳定运行具有重要意义,而风速预测是风功率预测的关键。文章提出一种基于优化模糊C均值(Optimal Fuzzy C means,OFCM)聚类的组合风速短期预测方法。首先,采用模拟退火遗传算法优化模糊C均值聚类算法的初始聚类中心;其次,基于优化模糊C均值聚类算法将初始风速属性样本数据进行分组;再根据不同风速样本组,运用极限学习机(Extremely Learning Machine,ELM)构建组合风速预测模型;最后,通过风速实测值与预测值的对比,验证了该方法的可行性。Accurate wind speed prediction is the key to wind power forecastingand very important to the safe and stable operation of power system. This paper presents a method of combined short-term wind speed forecasting based on optimal fuzzy C means （OFCM）clustering.First of all,the initial clustering center of fuzzy C means clustering algorithm is optimizedby using simulated annealing genetic algorithm. Then,the initial wind speed attribute data is classified. According to the differentwind speed samples, combined wind speed forecasting model is builtby using extreme learning machine （ELM）.Finally, the feasibility of the method is verified by comparing the measured data with predicted value.

关 键 词：风速预测 模拟退化遗传算法 FCM聚类 极限学习机 

分 类 号：TK81[动力工程及工程热物理—流体机械及工程]