Data-Driven Correction Approach to Refine Power Curve of Wind Farm Under Wind Curtailment

Abstract

Regardless of the rapid development of wind power capacity installation around the world, wind curtailment is a severe problem to be solved. Wind curtailment can cause abundant outliers and change the original characteristics of operation data in wind farms. Power curve cannot be accurately modeled with these outliers and consequently wind power forecasting as well as other applications in power system will be negatively affected. In this paper, the characteristics of the outliers caused by wind curtailment are analyzed. Then, a data-driven outlier elimination approach combining quartile method and density-based clustering method is proposed. First, the quartile method is used twice for eliminating sparse outliers. Then density-based spatial clustering of applications with noise method is applied to eliminate stacked outliers. A case study is carried out by modeling the power curves of a wind farm and 20 wind turbines in this wind farm. The accuracy of power curve modeling is significantly improved and the elimination procedure can be completed in a very short time, indicating that the proposed methods are effective and efficient for eliminating outliers. The performance of the methods is insensitive to their parameters and can be directly used in different cases without tuning parameters, both for wind turbines and wind farms.