基于数值天气预报及模糊聚类的风电功率智能组合预测  

INTELLIGENT COMBINED PREDICTION OF WIND POWER BASED ON NUMERICAL WEATHER PREDICTION AND FUZZY CLUSTERING

作　　者：杨家然[1] 王兴成[1] 罗晓芬[2] 蒋程[3] 

机构地区：[1]大连海事大学信息科学技术学院,大连116026 [2]华能威海发电有限责任公司,威海264205 [3]国网北京市电力公司,北京100031

出　　处：《太阳能学报》2017年第3期669-675,共7页Acta Energiae Solaris Sinica

基　　金：国家自然科学基金(60574018)

摘　　要：提出一种基于数值天气预报及模糊聚类的风电功率智能组合预测方法。以数值天气预报(NWP)数据为基础,利用模糊减法聚类的方法将原始数值天气预报(NWP)数据分成若干典型天气类型;针对不同的天气类型分别建立T-S模糊模型、时间序列模型、多元线性回归模型、灰色模型;利用智能优化算法进行多模型的优化组合,得到最优组合预测模型。对国内某风电场的风电功率预测结果表明,所提出的预测方法可行、有效,具有较好的预测精度。Wind power prediction accuracy has important implications for the scheduling and stable operation of the power system. An intelligent combined prediction algorithm of wind power based on numerical weather prediction and fuzzy clustering was proposed in the paper. Based on numerical weather prediction （NWP） data and using the method of fuzzy clustering subtraction, the original NWP data is divided into several typical weather patterns; T-S fuzzy model, time-series model, multiple linear regression model and gray model are established respectively according to different wheather types; the combination of multi-model is optimized using intelligent optimization algorithms and the optimal combination prediction model is obtained. Prediction results of a domestic wind farm indicated that the proposed combination prediction method is valid and effective in shortterm wind power prediction with better prediction accuracy.

关 键 词：智能优化 模糊聚类 组合预测 数值天气预报 风电功率 

分 类 号：TM614[电气工程—电力系统及自动化]