Research On PID Parameters Tuning Based On Improved PSO Algorithm

Abstract：In the industrial process control, PID control is the most commonly usedcontrol method. The PID controller has the advantages of simple structure, easyrealization and strong robustness, is widely used. The PID parameter tuning is thecore of the PID controller. But with the development of modern industry, more andmore complicated control processes have appeared, the traditional PID parametertuning method has been unable to fully meet the development. PID parameter tuningis being scientific research workers is an important research task. Researching of anovel, efficient parameter tuning is necessary.Particle swarm optimization (PSO) algorithm is American social psychologistJames Kennedy and electrical engineer Russell Eberhart putted forward on1995.Because of its simple structure, involves fewer parameters, easy realization.Once put forward, cause the great attention of scientific community.In this paper, using particle swarm optimization algorithm to optimize PIDparameters, doing the following several main aspects of the work: First, elaborated thePID parameter tuning of the development, analysis of the effect of the PID threecontrol parameters; Second, Introduced the particle swarm optimization algorithm theorigin, algorithm flow and the influence of parameters on the algorithm. Deepunderstanding of the connotation of the particle swarm optimization algorithm; Third,analysis the role and influence of learning factors in particle swarm optimizationalgorithm, and to explore the learning factor range design two experiments. Throughthe experimental research, improved particle swarm optimization algorithm learningfactor selection scheme. Through the experimental research, the particle swarmoptimization algorithm learning factors are improved, obtained relatively apparentimprovement effect; Fourth, research of particle swarm optimization algorithm tooptimize the PID parameters tuning principle, and use improved particle swarmoptimization algorithm for tuning PID parameters in Matlab software: Fifth, analysisof the dual water model, and use the Matlab/Simulink established the double-holding water tank simulation model, and use the particle swarm optimization algorithm forwater tank control system of PID parameters optimization, and obtained betterresult.This paper studies the improved particle swarm optimization algorithm in PIDparameter tuning use. A large number of simulation experiments to prove, particleswarm optimization algorithm based on PID parameter tuning method is a practical,efficient, accurate parameter optimization method.