Improved Particle Swarm Algorithm And Its Application In PID Neural Network Control

With the rapid development of scientific research and the expansion of application requirements,the traditional optimization algorithms often have limitations in solving complex nonlinear problems,and the application and development of electronic computing technology make the research of various intelligent optimization algorithms become popular.Particle Swarm Optimization(PSO)algorithm is a typical community-based optimization algorithm.Compared with the traditional optimization algorithm,the concept of PSO algorithm is simple and easy to understand,adjustable parameters and strong search ability.However,the PSO algorithm still has the advantages of premature convergence and easy to fall into the local optimum.Because of its theoretical basis,it needs to be improved and applied to the actual optimization problem.By analyzing and studying the basic principle of PSO algorithm,this paper proposes a PSO improved algorithm for multi-strategy fusion,which is used to optimize the PID neural network controller.Finally,the optimized controller is applied to the hot water boiler combustion system.The main contents of this paper are as follows:First of all,due to the classical linear decreasing strategy,the change of inertia weight does not match the change of algorithm search process,this paper proposes a PSO improved algorithm with multi-strategy fusion.The algorithm uses the initial initial population position of random initialization,and then introduces the stochastic factor to the inertia weight,and theoretically analyzes the stability and convergence.The simulation results show that the multi-strategy PSO algorithm proposed in this paper has a significant improvement in search performance compared with the standard PSO algorithm and LWPSO algorithm.Then,PSO algorithm is used to optimize the connection weight of PID neural network controller.Three different PSO algorithms are used to optimize the PID neural network controller by using standard PSO algorithm,LWPSO algorithm and multi-strategy fusion PSO algorithm.The variable is strongly coupled to the controlled object.The results show that the PID neural network controller optimized by multi-strategy fusion PSO algorithm can effectively control the system,and not only the control quantity is close to the target and the control output error is smaller than the other two control methods.Finally,the hot water boiler combustion system is used as the control object of the PID neural network,and the PID neural network controller is optimized by the standard PSO algorithm,the LWPSO algorithm and the PSO algorithm of multi-strategy fusion respectively.Control effect of comparative experiments.