| With the advent of the era of digital large data,information is gradually diversified,exploded and difficult to deal with.How to use these data efficiently for daily life and industrial production service has become the focus of various industries.As a kind of clean,efficient and cheap energy,wind power is being paid attention to and exploited by various countries.Various data monitoring and recording systems are installed in wind farms.How to make full use of these historical data and play the advantage of historical operation data is helpful for the management,monitoring and maintenance of wind farms.In this paper,the following work is done to evaluate the low voltage crossing capability of wind farms based on running data.1.The working principle of the doubly fed wind turbine is analyzed.The FCM clustering algorithm is applied to the clustering problem of wind farms.2.The fault crossing principle of the doubly fed generator,the relationship between the fault characteristics of the equipment in the wind farm and the mutual influence,the influence of the fault behavior of the wind turbine to the power grid and the analysis of the performance index based on the power grid are introduced.3.On the basis of investigating the data recording equipment of wind farm,this paper analyzes the problems existing in the running data,and puts forward the corresponding solutions to the problems of different data sources,such as the multi time scale,the data anomaly,the missing data and so on.By analyzing and analyzing the characteristics of 5 kinds of data monitoring and recording system in wind farm,the correlation between running data and low voltage crossing ability is summarized and summarized,and the method of voltage drop feature extraction based on generalized transformation is proposed.Based on the grid connection guidelines,a method for evaluating the ride through capability of wind farms is presented. |