Research On Parameter Identification Of Power System Load Model

With the introduction of large-scale new energy generation and a series of related power electronic equipment,the scale and complexity of China’s power grid is expanding day by day.In order to design and dispatch the power grid more reasonably and promote the safe and stable operation of the power system,it is of great significance to establish the load model which is consistent with the actual power grid.Overall measurement method is a common method in load modeling field,which has the advantages of simplicity,high efficiency and practicability,but the reliability of identification results will also be affected by load model and optimization algorithm.Therefore,this paper mainly focuses on the optimization algorithm,model parameter sensitivity analysis and model parameter identification in load modeling.The load modeling research focuses on the selection of model structure and identification algorithm.This paper introduces the classification and structure of commonly used load models in power system simulation calculation,expounds the principle and method of load model parameter identification,and points out that the simulation evolution method is of great significance to improve the identification accuracy and efficiency.According to the mathematical description of the CLM model,the solution of the initial steady state and dynamic process is derived in detail,which provides a theoretical basis for the parameter identification of the load model.Grey Wolf algorithm（GWO）has the defect that it is easy to fall into local optimum in the late iteration.In order to improve the convergence speed and global optimization ability of GWO algorithm,an improved strategy is proposed to introduce dynamic weight factor and hybrid DE algorithm.The effectiveness of the improved GWO algorithm is verified by four test functions.According to the structure of the CLM load model,the independent minimum parameter set to be identified and the adaptive value function of parameter identification are given,and a recursive identification strategy of the CLM model based on the improved GWO algorithm is proposed,which provides an algorithm support for parameter identification of the load model.In order to improve the efficiency of load modeling and promote the practicability of CLM load model,this paper studies the active and reactive power sensitivity of parameters under two different faults by perturbation method,and then determines the dominant parameters of CLM load model.Then two schemes of partial parameter identification and full parameter identification are proposed,and the validity of the dominant parameters k_pm and K_Lof equivalent motor are verified by an example,which provides a basis for identifying only the sensitivity parameters and selecting the typical values of the other parameters in load modeling.In order to verify that improved GWO algorithm can improve the accuracy of parameter identification of load model,this paper builds a 3-machine and 9-node example system on PSD-BPA platform to conduct transient simulation of power system,and the sensitivity parameter set in CLM model is identified according to two kinds of fault data and three kinds of optimization algorithms,and the advantages of the improved GWO algorithm are verified.Then fourier transform,symmetric component method and power calculation are used to preprocess the transient data of 220k V Chaoyang to 110k V Chaomeng line collected by fault recording device,and load modeling is carried out based on the measured data.The validity of the improved GWO algorithm for load model parameter identification is further verified,which has a certain engineering significance.