Parameter Tuning of a PID Controller Based on the Cellular Genetic Algorithm.

The PID controller is the most widely used controller, and the performance of each controller depends on its three gain parameters. The genetic algorithm (GA) is an optimization process that involves three types of operators: selection, crossover, and mutation. Inspired by natural evolution, a GA searches for the optimum solution automatically from a collection of potential solutions. The GA was employed for the optimization of PID controller parameters in this paper; this algorithm evolves from generation to generation until the stopping criterion is met or the optimal solution is found. The traditional GA has several shortcomings, such as the great influence of the best individual on the whole population, poor global search ability, and premature convergence; therefore, an improved cellular genetic algorithm (CGA) was proposed whose genetic operators occur between the individual and its neighbors. CGA avoids the adverse influence of the best individual and the possibility of falling into a local optimum. The simulation results show that CGA outperforms traditional GA. CGA is more efficient than GA in global search activities, and its optimum solution is better than that of GA. [ABSTRACT FROM AUTHOR]