Research On Typical Parameters Analysis And Identification Methods Of Power System Load Model Parameters

With the rapid development of the power grid,the power supply structure is diversified,the grid structure is complicated,and the load components are diversified.The safe and stable operating pressure of the power grid is gradually increasing,and the accuracy of the simulation model is becoming higher and higher.The load model is especially dynamic The choice of load model has an important influence on the reliability of the system stability calculation results.Induction motor load is the most important component of dynamic load,and there are a large number of induction motor loads with different parameters in the power system.However,the type of induction motor is not clear,and the distribution of its model parameters is unclear.In this regard,this paper collected and sorted 224 motor load model parameters through literature research,and based on this,deeply analyzed the types and typical parameters of induction motors and used them for load model parameter identification.This article mainly researches from three aspects: induction motor model parameter investigation,induction motor typical parameter analysis and comprehensive load model parameter identification.In the investigation of induction motor model parameters,the load model parameters of 224 motors were obtained based on literature investigation.Aiming at the problem of a large number of missing mechanical torque coefficients and initial slips in the parameter set,the torque coefficient filling method of the K nearest neighbor algorithm and the initial slip estimation method based on load rate or critical slip are proposed respectively,and good filling effects are obtained,Laid a solid and reliable data foundation for the subsequent analysis of typical parameters.In the analysis of typical parameters of induction motors,based on 224 sets of filled-in motor parameters,taking into account the differences in the transient response of different motors,using the simulated motor transient power curve as a clustering basis,using the K-means algorithm,7 types of For the motors with different state characteristics,the parameter distribution characteristics of 7 types of motors are analyzed,and 7 types of effective typical parameter values are extracted.The 7 typical parameter values obtained complement the motor parameter sets recommended by IEEE and State Grid Co.,Ltd.,and provide a reference for the typical parameter values of the induction motor load model.In terms of parameter identification of the integrated load model,the sensitivity of the dynamic response of the integrated load model to the model parameters is analyzed based on the trajectory sensitivity method,and the key identification parameters are determined based on this;based on the measured fault recorder data,the genetic algorithm and the ant colony are used.Based on the advantages of the algorithm,the genetic ant colony hybrid optimization algorithm uses seven typical parameter values as the initial values of the parameters to be sought,and carries out load model parameter identification,which improves the accuracy of parameter identification and provides support for more accurate grid stability analysis and calculation.