4.7 Article

Wind Power Curve Modeling With Hybrid Copula and Grey Wolf Optimization

期刊

IEEE TRANSACTIONS ON SUSTAINABLE ENERGY
卷 13, 期 1, 页码 265-276

出版社

IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/TSTE.2021.3109044

关键词

Wind power generation; Wind turbines; Wind speed; Optimization; Hybrid power systems; Biological system modeling; Splines (mathematics); Grey wolf optimization; hybrid copula; mixture of asymmetric gauss; power curve modeling; Weibull distribution

资金

  1. National Natural Science Foundation of China [71671029]

向作者/读者索取更多资源

This paper proposes a hybrid copula-based wind power curve model (HCCM) that takes into account the relationship between wind speed and errors. Experimental results show that the proposed model significantly improves the accuracy of wind turbine forecasting.
Wind power curve modeling has important applications in many fields, such as wind turbine condition monitoring and wind power forecasting. There are various studies on wind power curve construction though, the relation between wind speed and errors is not taken into account when constructing wind power curves. To fill this gap, a hybrid copula-based wind power curve model (HCCM) is proposed in this paper. The independent distribution of wind speed is modeled by a Weibull distribution, and the independent distribution of errors is modeled by a mixture of asymmetric gaussian models. Then a hybrid copula model, including four copula functions, is designed to construct the joint distribution. Finally, a grey wolf optimization algorithm is applied to optimize the regression parameters of wind power curves by taking the log-likelihood function of the joint distribution as the fitness function. The proposed model is compared with ten benchmark models on four wind farms and two wind turbines. Experiments show that in terms of RMSE the HCCM achieves 99.3131%, 99.1197%, 99.1715% and 98.9339% on four wind farms, respectively, and around 80% improvement in RMSE on two wind turbines.

作者

我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。

评论

主要评分

4.7
评分不足

次要评分

新颖性
-
重要性
-
科学严谨性
-
评价这篇论文

推荐

暂无数据
暂无数据