Research On Attitude Detection And Control Of Inverted Pendulum Based On Particle Swarm Optimization Algorithm

Particle Swarm Optimization(PSO)is a kind of parallel and intelligent algorithm,which has attracted much attention in the field of intelligent optimization algorithm due to its advantages of less parameter setting,simple optimization principle and fast running speed,and is one of the research hotspots in the field of intelligent control.In this paper,PSO is selected as the controller parameter optimization algorithm to solve the problem of difficult parameter setting of the optimal controller,and on the basis of the traditional PSO algorithm introduces improved strategy according to the characteristics of algorithm to obtain better parameter values.Finally,the optimized controller is applied to the linear two-stage inverted pendulum system,and the simulation results show that the control quality of the system is higher.Firstly,this paper first describes the significance of the topic research,and then from the characteristics of intelligent optimization algorithm,summarizes the application of intelligent optimization algorithm in the field of control.It also introduces the origin,design ideas and development trend of PSO algorithm,and discusses the defects of PSO algorithm itself,pointing out the direction for the subsequent content of this paper.Secondly,in order to understand PSO more thoroughly and explore the influence of parameter information on algorithm optimization process,starting with the basic model of the algorithm,the influence of parameter setting on the convergence performance of PSO is discussed.According to the particle characteristics,the convergence of the algorithm is theoretically deduced and analyzed,and the parameter range to ensure the convergence of the particle swarm is determined.At the same time,the optimization direction of PSO algorithm is pointed out,which lays a theoretical foundation for the optimization strategy.Thirdly,based on the theoretical research and analysis of PSO in previous chapters,a Classification Sub-Population Differential Evolution PSO(CSP-PSO)algorithm is proposed.Firstly,nonlinear function expression is used to replace fixed weights and learning factors in PSO algorithm.Aiming at the problem that the algorithm is easy to fall into local extremum and the population diversity is poor,the Dynamic Population algorithm and Dynamic Multi-Swarm algorithm are introduced.On this basis,the interaction between particle populations is studied,and a classification subpopulation algorithm is proposed.In addition,the Differential Evolution(DE)strategy is incorporated in the late running period of the algorithm.In order to compare the performance of the improved algorithm,benchmark functions of various spatial dimensions are used for testing.The experimental results show that the CSP-DEPSO algorithm has the best performance.Fourthly,inverted pendulum is a classical teaching instrument to test the control algorithm.In this paper,the improved PSO algorithm is applied to optimize the parameters of the optimal controller,and then realize the control of inverted pendulum system.Above all,the line two-stage inverted pendulum is modeled and analyzed,then introduces the design principle of the optimal controller and parameter setting difficult problems of controller.On this basis,the advantages of PSO algorithm to optimize controller parameters are demonstrated.Finally,MATLAB/Simulink was used to write the program code and build the system model to realize the simulation and the algorithm is applied to the inverted pendulum.The simulation results show that the control performance of the system is greatly improved,which verifies the effectiveness of the improved PSO algorithm to optimize the controller parameters.Finally,the research process and experimental content of this paper are summarized,and the future research direction of PSO is prospected.