The Modeling And Parameter Identification Of Excitation System Of Synchronous Generators

In power system excitation system of synchronous generator, mainly through excitation regulation, makes the role of generator fully played to improve power system reliability. With expansion of the network, the problem of network security has been increasingly prominent, the importance role of excitation system of generator in excitation regulation has been more and more concerned. One of the key factors in affecting the security and stability of power system is the accurate mathematical models and measured parameters. Therefore, to identify the parameters of excitation system according to characteristics of the structure of excitation system is quite necessary so that we can get the exact parameters of excitation system. Through that we can accurately monitor operating status and forecast operating performance of systems, then achieving improvement of security and stability of the network so that to meet the planning, operation and management of power grid.Based on the analysis of structural characteristics of generator excitation system, This paper identifies the parameter of the linear and nonlinear part of the excitation system, and simulating and identifying the accuracy and validation of various algorithms through MATLAB/SIMULINK, the total content as following:1) Making detailed analysis of various types of excitation system models;2) Using the method of RGLS to identify the parameters of the linear part of excitation system;3) Using MATLAB/SIMULINK to build generator running model with no load and infinite bus system model with single machine including excitation system, Collecting input and output data of excitation system operation;4) Using the method of improved genetic algorithm (GA) to identify the parameters of the nonlinear part of excitation system;5) As there are some problems in traditional BP neural network, such as the slow convergence, easily trapped in local minimum solutions and so on, I put forward improved BP neural network to identify the parameters of the nonlinear part of excitation system;...