Research On Parameter Identification Of Excitation System

At first, least squares algorithm is put into the parameter identification of linear tache in excitation system and gain good effect. Secondly, genetic algorithm, simulated annealing algorithm and particle swarm algorithm are improved aim at their defects. These improved algorithms are put into the parameter identification of nonlinear excitation system. These artificial intelligence algorithms can identify the parameters of the nonlinear excitation proved by the examination of MATLAB. At last neural network is put into the system identification of the excitation system whose structure is unknown. The neural network is trained to carry out the simulation of the excitation system whose structure is unknown. The excitation whose structure is unknown can be simulated well by the neural network which have been trained and this conclusion can be proved by the comparison between the output of the neural network and the output of the practical system.