Particle Swarm Optimization (PSO) Controller Tuning of Underwater Thruster

Abstract

A proportional-integral-derivative (PID) controller is very popular in applications that require an easy and comprehensive control. In this project, a Particle Swarm Optimization (PSO) technique is integrated with PID controller. The algorithm is used to optimize the controller, which applied at underwater thruster system. There are two strategies of tuning PID gains for the application of underwater thruster system. The first strategy is developing thruster model based on Hammerstein-Weiner system identification. Then the second strategy is the proposed algorithm which is PSO techniques for PID gain optimization. The objective of this project is to control the performance of the thruster in order to minimize the energy consume by thruster in Autonomous Underwater Vehicle (AUV). The PSO-PID controller tuning with the PSO algorithm is used to search the optimum value for the controller parameters automatically. The algorithm will find the pBest which is personal best position for the particles and then will automatically search for gBest (global best position for the particles). All controllers gain Kp, Kd, and Ki will be calculated and used in PID controller to control thruster system. The performance was measured based on the rise time, settling time, overshoot and steady state error. Based on the simulation result the PSO-PID controller give a better response in all categories compared to conventional tuning. PSO has a good computational efficiency, easy implementation and stable convergence characteristics. The results of 0.6056(Kp), 0(Ki) and 0.5520(Kd) show that PSO-PID algorithm can perform well in nonlinear control system design. The power consume by thruster is minimized and see by the decreasing time of searching but zero value of overshoot so it increase the searching efficiency of thruster as more longer performance.