DESIGN AND IMPLEMENTATION OF SOFT COMPUTING TECHNIQUE ON BALL BALANCER SYSTEM

Abstract

This project presents design and control of a 2D Ball Balancer Arrangement using PID controller tuned with Particle Swarm Optimization (PSO), Simulated Annealing (SA) and Genetic algorithm (GA). The project also compares the results of proposed control techniques with Classical PID controller and Fuzzy logic controller (FLC). The Ball Balancer Arrangement is a non-linear system with complex plant transfer function. Classical control methods such as Classical PID and Fuzzy logic controller are also able to control the Ball Balancer Arrangement but manual tuning and rule base optimization is slow and inefficient. PSO, SA and GA are heuristic algorithms which can find the best solution without the need of human experience. Heuristic algorithms improve the solution in every iteration based on cost function minimization. The comparative analysis of PSO, SA and GA tuned PID and their comparison with Classical PID and FLC is shown in terms of delay time, rise time, and settling time and the study demonstrates that PSO tuned PID controller performs the best. The designing and simulation have been successfully performed in MATLAB/ Simulink environment.