| Power system simulation is an important basis for power grid planning,dispatching and stable operation,and load modeling,as the basis for power sy stem simulation,has always been the focus of scholars’ attention as a difficulty and focus in the field of power systems.In recent years,fault data after large disturbances are mostly used in load modeling to complete parameter identification.In actual power grids,there are few large disturbance faults.The measured data are mostly small disturbance data,and there are noise signals in the relevant data.There are certain difficulties in small disturbance data such as signals,and the load model parame ters cannot be accurately identified.Therefore,relying on the scientific and technological project of Hunan Electric Power Research Institute,this paper conducts research on the identification of classical load 、 two-stage parameter identification and the identification based on PMU small disturbance data.In terms of parameter identification of classical comp osite load model,the established classical comprehensive load mathematical model and parameter identification method are verified based on simulatio n data and measured data.According to the actual situation of the power grid,a com posite load model and parameter identification method suitable for the power grid are established,in which the load model parameter initialization method and differential evolution algorithm are analyzed;PSASP simulation data and Hunan Langli substation measured data are selected for example analysis.The composite load model proposed in this chapter has good description ability,but the identification method cannot identi fy the measured small disturbance data,but can only identify PSASP simulation data and large disturbance measured data.In the aspect of two-stage parameter identification method,a two-stage parameter identification method suitable for small disturbance identification is proposed to solve the problem that the traditional identification method of differential evolution algorithm cannot identify the measured small disturbance data.The identifiability of the small disturbance data is analyzed,and the overa ll framework of the two-stage parameter identification method and the parameters identified in each stage are described.Based on the simulation data and simulation noise data,an example analysis is carried out to verify the effectiveness of the two-stage parameter identification method proposed in this chapter.The result shows that the two-stage parameter identification method proposed in this paper has an ideal identification ability.Based on the parameter identification of PMU measured data,the propo sed two-stage parameter identification method is verified by the PMU measured data of Hunan Power Grid.In order to make the measured data meet the requirements of parameter identification,the feasibility analysis and spectrum analysis of the measured data are carried out;the wavelet denoising method is further used to reduce the influence of Gaussian noise on parameter identification;the two-stage parameters proposed in this paper are analyzed by the PMU measured data of Hunan Power Grid.The simulation data to verify the identification method is from the electrical station of Tong Xin and Quan Tang.The two-stage parameter identification method proposed in this paper can effectively identify the measured ambient data,which is greatly improved compared with the traditional identification method. |
