In this book, attempts are made to design a intelligent Proportional-Integral-Derivative(PID)controller for processes. A simple Genetic Algorithm(GA)method has been applied to various single-input- single-output(SISO) & Multi-Input-Multi-Output(MIMO)process models for obtaining optimum PID parameters for minimization of error performance indices. The proposed method can choose the best PID parameters for each generation. Then, using the reproduction, crossover and mutation to create the new population for other PID parameters, the optimum PID parameter are obtained.The proposed method is compared with Ho et al., Ziegler-Nichols and Wang et al. It is observed that the results using proposed method are better in terms of IAE, ISE and ITAE. A simple and effective design methodology of decentralized PID controllers is developed based on extension of the GA based SISO controllers. Initially the system is decoupled using decoupler matrix and the first-order-plus-delay-time(FOPDT) model of each subsystems is obtained. FOPDT model obtained from decoupler, retains the unit characteristics of systems. A genetic approach is applied to obtain the parameters of the Decentralized PID controller.
