PID Controller Parameter Optimization Based On Improved Glowworm Swarm Algorithm

In today's industrial process control,PID(Proportional Integral Differential)controller is widely used because of easy realization,simple structure,high reliability and high robust performance.According to relevant statistics show that more than 95%of the controller used the control thought in the actual industrial process.It is well known that PID controller performance is directly related to PID Controller Parameter.However,with the rapid development of modern industrial technology,the system control becomes more and more complex,the traditional PID controller parameter tuning algorithm has not be well adapted to these complex modern control system.But with the continuous development of artificial intelligence and computer technology to penetrate in the field of automation,have produced a lot of intelligent PID parameter tuning algorithm.The Glowworm Swarm Optimization algorithm are widely used because of its parallelism,self-organization,easy to implement,distributed and robustness characteristics as a kind of optimization calculation method.This thesis will use the improved glowworm swarm algorithm tuning PID controller parameters,The main contents of the study can be summarized as follows:(1)This thesis proposes an improved Glowworm Swarm Optimization Algorithm with the introduction of directed and adaptive step mechanism(Based on Directed mechanism and Adaptive-step mechanism GSO,D-AGSO).On the one hand,This thesis introduced the adaptive step strategy in the algorithm.Making the step length maintain a bigger value at the early stage of the iteration algorithm to search within the global optimal solution and prevent algorithm from being precocious and falling into local optimum.Maintaining the step length smaller at the late stage to keep the algorithm from jumping over the optimal solutions and improve the precision.On the other hand,if do not find other brighter glowworm in the dynamic decision space radius of the glowworm,the glowworm will randomly move.However,the glowworm failed to move justly at the cost of large amount of computation.This situation not only causes slow iteration speed,but also may lead the algorithm to local extremum,effect the overall performance of algorithm.In order to tackle the defect,This thesis takes the directed mechanism to speed up the iterative speed and improve the precision.The simulation experimental results show that D-AGSO features high efficiency and precision when compares to Glowworm Swarm Optimization Algorithm(GSO),Enhanced Glowworm Swarm Optimization Algorithm(EGSO)and Foraging-behavior Adaptive-step Glowworm Swarm Optimization Algorithm(FA-GSO).(2)This thesis constructs the simulation mode based on SIMULINK,a simulation tool of MATLAB.Based on Directed mechanism and Adaptive-Step GSO(D-AGSO),is applied to the PID parameters tuning.To verify proposed algorithm in this thesis,the thesis selects four kinds of different typical controlled object.By conducting simulation experiments based on SIMULINK,D-AGSO is evaluated in comparison with compares to Particle Swarm Optimization(PSO)algorithm,Ziegler-Nichols(Z-N)and the Differential Evolution(DE).The simulation experimental results show that the proposed algorithm features high efficiency and precision when tuning the PID controller parameters.(3)To verify the proposed algorithm in this thesis,the thesis selects an improved ITAE evaluation function.The simulation experimental results show that the proposed algorithm can effectively make the PID controller focus on the specific performance by changing the performance weighting coefficient.Therefore,this algorithm can be used to tuning the PID controller which is needed in the process of industrial control.