Research On The Identification Of Synchronous Generator Parameters Based On Measured System Disturbance

The accuracy of power system dynamic simulation has important influences on the safe operation of electric power system,economic scheduling,stability analysis and decision support.Several accidents and tests at home and abroad have shown that the inaccuracy of the component model parameters is the main reason for the deviation of the simulation results and the actual dynamic behavior of the system.Regarding the issue above,this paper conducts a research on generator model verification and parameter identification based on measured system disturbance based on measuring data of phase measurement unit(PMU).This paper at first builds the generator simulation model,introduces the concept of hybrid dynamic simulation,inputs the measured information to boundary bus to realize the equivalence of the external network,and combines the geometric correlation coefficient with residual numerical analysis to establish the effective evaluation strategy of the generator model.On this basis,the parameters of the model are identified by rough set attribute reduction and particle swarm optimization algorithm,and the parameter identification theory is applied to the practical case verification analysis.The difference analysis between simulation and measured data is the theoretical basis of parameter identification.Regarding the hidden information of traditional power software,the poor object interaction and difficult identification to call the internal package parameters,this paper chooses the MATLAB platform with good open source compatibility,constructing the parameter identification model.Refer to PSD-BPA stable description control block diagram to build generator model of a Yunan thermal Power Unit.Meanwhile,call measured data by programming realizes the automatic import of BPA parameters and the preprocessing of PMU data.And the paper conducts further researches on the adjustment calculation of voltage phase angle and the zero point interpolation instability problem so as to establish the parameter identification analysis simulation module and provide the basis for the study below.The goal of the equivalent network technology is to simplify the system model equation and improve the research efficiency under the premise that the system equation is consistent or basically consistent before and after the equivalence guarantee.The traditional equivalent method relies on the whole network data,while it is extremely difficult and cumbersome to obtain accurate data under the market environment.This paper uses the hybrid dynamic simulation method,keeping the research subsystem and injecting the measured voltage modulus and phase angle signal into the boundary bus of the simulation subsystem to realize the mixed dynamic equivalence of the external system.The theory of model validity evaluation is also introduced,and the accuracy of the model is evaluated by comparing the fitting degree of the measured and simulated trajectory,and establish the generator model verification and evaluation method based on the numerical and shape similarity calculation.Finally identify and correct the parameter of generators with low modelling verification evaluating indicator.Make attribute reduction and dependency calculation of perturbation matrix based on rough set theory to reduce the number of parameters that are to be identified.Identify the reduction models through the particle swarm algorithm.Combined with the actual power plant model identification results,it proves the feasibility and correctness of the proposed method in this paper.