Applying hybrid genetic–PSO technique for tuning an adaptive PID controller used in a chemical process

The conventional PID controller has static parameters that cannot be changed at different operating conditions. As a result, the term ‘adaptive PID controller’ has appeared to solve this problem. This controller can be tuned using intelligent techniques such as Fuzzy Logic Control, Neural Network Control, or Adaptive Neuro-Fuzzy Inference Systems. However, the choice of the suitable parameters for these intelligent controllers has a direct effect on their performance. Metaheuristics algorithms—with their powerful performance, speed, and optimal parameter selection—can be applied for choosing controller parameters efficiently. In this paper, a hybrid of genetic algorithm and particle swarm optimization is proposed to tune the parameters of different adaptive PID controllers. To evaluate the performance of the proposed hybrid optimization method on the different adaptive PID controllers, these controllers are applied to control the operation of one of the most difficult chemical processes, the divided wall distillation column. The proposed column used in this work separates a ternary mixture of ethanol, propanol, and n-butanol. Our proposed hybrid optimization technique is compared with the genetic algorithm, and simulation results show that our proposed hybrid genetic-particle swarm technique outperforms genetic algorithm for different disturbances.