Abstract: Most of the process systems exhibits non-linear behaviour, so conventional controllers are not able to provide accurate data. At present, various soft computing techniques are used to overcome imprecision and uncertainty effects of conventional controllers. Various soft computing techniques like fuzzy logic, genetic algorithm and particle swarm optimization have been suggested for optimum setting of PID controller parameters. In this paper, the performance of all the three soft computing techniques is compared for three tank level process control system. This comparative study is carried out for set point tracking of three tank level process using MATLAB/ SIMULINK. The simulation results shows that tuning the PID controller using PSO provides fast and stable system with low overshoot.

Keywords: Three tank process, Soft computing, Fuzzy logic, Genetic algorithm, Particle swarm optimization.