基于混合Copula函数的风电机组异常识别方法

风电机组的功率曲线是衡量风电机组运行性能的重要指标,其中存在大量异常数据直接影响风电机组运行维护,研究功率曲线异常识别方法对提高风电机组运行稳定性具有重大意义.本文通过对功率曲线的特性分析,分三种工况建立了基于概率分布函数(Copula函数)的功率曲线异常数据识别模型及影响因素Kendall秩相关分析,从而确定了与功率...

Abnormal Identification Method of Wind Power Turbine Based on Copula Function

The power curve of a wind turbine is an important index to measure the operation performance of a wind turbine,but there are a lot of abnormal data which directly affect the operation and maintenance of the wind turbine.It is of great significance to study the effective identification method of the abnormal data of the power curve for improving the operation stability of the wind turbine.In this paper,through the analysis of the characteristics of the power curve,the identification model of the abnormal data of the power curve based on the probability distribution function(copula function)and Kendall rank correlation analysis of the influencing factors are established in three working conditions,so as to determine the characteristic parameters highly related to the power curve;The upper and lower boundaries of the power curve are obtained by copula function,and the time series analysis and Euclidean distance calculation of the suspicious points outside the upper and lower boundaries are used to judge whether the doubtful points are abnormal points;the SVM linear regression method is used to establish the prediction model,and the SCADA data of a wind farm is used to verify the proposed method.The results show that under three different conditions,the method can accurately identify the abnormal data of the power curve and accurately locate the fault,which provides a new auxiliary method for the operation and maintenance of the wind turbines.