Improvement Of Intelligent Optimization Algorithm And Its Application In System Identification

Intelligent optimization algorithms have become the focus of research due to their strong optimal search ability.Scholars from all walks of life have proposed a variety of intelligent optimization algorithms to solve optimization problems in various fields,such as finance,chemical industry,control science,material engineering,electronic information science and technology,aerospace and other fields.Due to the introduction of intelligent optimization algorithms,the research progress of the subject has been greatly improved.In recent years,in control science,intelligent optimization algorithm has also made some achievements in the field of system identification,and has made good application in solving the problem of modular nonlinear system identification.At the same time,with the increase in process complexity and rich process requirements,non-uniformly sampled systems appear more and more frequently in industrial production,and how to solve the identification problem of non-uniformly sampled systems has become a research hotspot.The main research contents of this paper include four points:(1)In order to solve the problem that Grey Wolf Optimization algorithm is easy to fall into local optimum and low convergence accuracy,logistic chaotic mapping strategy is introduced to initialize the population,so that the initial population distribution is more uniform;hierarchical system is proposed to highlight the rank mechanism of wolves in GWO algorithm;velocity vector idea is introduced to improve convergence accuracy and speed;elimination mechanism is proposed to eliminate wolves who are Black sheep.In addition,according to the hierarchical system of the wolf society,let the higher-level wolf have the right to reproduce,improve the accuracy of the algorithm,improve the convergence speed of the algorithm.(2)In order to verify the effect of the intelligent optimization algorithm,18 test functions are used to test the optimization performance of the improved grey wolf algorithm.The test results show that the improved algorithm has stronger optimization effect than other algorithms.At the same time,in order to improve the application of intelligent optimization algorithm in identification,the improved grey wolf Optimization algorithm(GWO_PSO)is used to identify the Wiener system parameter model of modular nonlinear system.The simulation results show that the identification effect of GWO PSO algorithm is better than other intelligent optimization algorithms.(3)Aiming at the problem that the derivation of traditional identification methods such as least square method is complex and identification is not universal,particle swarm optimization algorithm is used to identify the linear system with non-uniformly sampled,Hammerstein system model with nonuniformly sampled and non-uniformly sampled Hammerstein system model with noise.The model of non-uniformly sampled system is solved by lifting technique,and the optimal solution is obtained by particle swarm optimization,which is the parameter estimation of the system to be identified.The optimal solution is obtained by particle swarm optimization,that is,the estimated parameters of the system with identification.From the experimental results,it can be seen that the intelligent optimization algorithm has a good application in the identification of the model parameters of the system with non-uniformly sampled.(4)An Improved particle swarm optimization(IMPSO)algorithm is proposed to improve the low accuracy and slow convergence of particle swarm optimization algorithm in identifying the model parameters of non-uniformly sampled system.The simulation experiment is carried out with the non-uniformly sampled Hammerstein model as the simulation object.Through the experimental results,it can be concluded that the IMPSO algorithm effectively improves the identification ability of PSO algorithm.