Improvement Of Particle Swarm Optimization And Its Application In PID Parameters Tuning

Optimization problems are widely found in scientific research and engineering application.Practice shows that research and solve the optimization problem can improve the efficiency of the system,reduce the consumption of resources and promote economic development,with great theoretical and practical value.Particle Swarm Optimization(PSO)is a heuristic optimization algorithm for solving optimization problems,by simulating birds' migration and foraging behavior to make particle search in the complex space.It has the characteristics of easy implementation,robustness,wide applicability and has become the focus of research in the field of optimization theory and optimization algorithms.However,PSO is also easy to fall into the local optimal,the latter part of the convergence speed is slow,the solution accuracy is poor.In order to improve the performance of the algorithm and widen its application range,we still need to further improve and optimize the PSO.In order to overcome the shortcomings of the PSO,this paper propose a novel cloud particle swarm optimization algorithm based on pattern search method(PCPSO)and use it to solve the multimodal function optimization problem.This algorithm use the mutation strategy based on normal cloud model to make PSO maintains a high population diversity and effectively avoids falling into local optimum;PCPSO employs cloud model particle swarm optimization algorithm(CPSO)to do global searching in the feasible zone,and then use pattern search method(PSM)to improve the accuracy of the sub-optimal solution which CPSO has found.The simulation tests demonstrate that PCPSO ensures convergence speed,meanwhile the convergence accuracy,as well as the number of extreme points,is strikingly improved.Proportional-Integral-Derivative(PID)controller is the most commonly used control technology in industrial production,it has the characteristics of simple structure,high control precision and good stability.For the problem of the traditional PID parameter tuning method is ineffective in complex environment,this paper combines the proposed algorithm with PID control theory to propose a PCPSO-PID parameter tuning method.This method uses PCPSO to optimize the parameters of the controller,in order to find out the most suitable combination and make the controller work in the best condition.The simulation of typical controlled object and beer fermentation temperature control system shows that:this method does not rely on the functional state of the controlled object,and has the advantages of wide application,strong robustness,fast response and good control performance.