Study On Fault Equivalent Modeling Method For Wind Farm Fault Characteristics Analysis

The accuracy of the power system model is the theoretical basis for analyzing the fault characteristics,and it is crucial for the theory of relay protection in the system.With the centralized grid connection of large-scale wind power,the fault characteristics of power electronic converter equipment in the system are gradually becoming prominent in areas with a high proportion of renewable energy.The grid connected converter equipment has poor current capacity and the fault traversal control changes with the voltage drop of the power grid,which is significantly different from the fault characteristics of synchronous generator power supply and affects the performance of system relay protection action.Station modeling is the main means of fault characteristic analysis.However,wind farms usually contain a large number of wind turbines,which leads to a significant increase in the number and complexity of network nodes,a significant increase in the calculation of network short-circuit current,and even non convergence of calculation results;The single machine equivalent model of the farm also has the problem of high model simplification and low accuracy.Therefore,it is urgent to study an accurate equivalent modeling method for multi machine faults in wind farm stations.In response to the above issues,the paper starts from the fault characteristics of wind turbines and conducts research on multi machine equivalent modeling methods for wind farms faults.The main work of the paper is as follows:Combining practical engineering fault recording data and theoretical analysis,based on data mining fitting method,the single machine fault characteristics of permanent magnet synchronous generators and doubly fed induction generators under different control modes were analyzed,which served as the basis for selecting clustering indicators using clustering equivalent modeling method.The accuracy of the analytical model is verified by simulation and engineering fault recording data validation.Based on the influencing factors of fault characteristics of doubly fed induction generators,a two-step clustering method is adopted.Firstly,the initial clustering is carried out according to the control strategy.Secondly,the degree of voltage drop,wind speed,and voltage phase at the grid point are used as clustering indicators for further clustering.An improved K-means algorithm is proposed,which takes the fault current equivalent error of the unit as the sample spacing and the equivalent electrical quantity of the same unit as the center coordinate of the algorithm;By combining the equivalent error of doubly fed induction generators with the contour coefficients of clustering algorithms,the clustering number of the algorithm was adjusted,and a fault multi-machine equivalent model suitable for fault characteristic analysis of doubly fed wind farms was constructed.Based on the main factors that affect the fault characteristics of permanent magnet synchronous generators,PMSG are first grouped according to control strategies,and then further divided according to the degree of voltage drop,wind speed,and voltage phase at the grid connection point.For permanent magnet synchronous generators under the same control method,the d and q axis currents of permanent magnet synchronous generators before and after equivalence are reduced,and the analytical relationship between the shortcircuit current error generated by equivalence and the electrical quantity boundary of the unit is derived.Based on the required accuracy,the grouping boundary for establishing a multi machine fault equivalence model is adjusted,and a fault multi machine equivalence model suitable for fault characteristic analysis of permanent magnet wind farms is constructed.