Identification and Control of Stable and Unstable Systems by Using Genetic Algorithm Method

Genetic algorithms can be applied for the identification and control of both stable and unstable systems. This paper brings systematically the development of the identification and the control for four processes of different behaviors. The contribution is focused on establishing optimal identified models by using two different approaches, least squares procedures and genetic algorithms optimization. In addition, a based PID controller for each process is synthesized applying genetic algorithms. Then, the processes' control based on reference model is also established and both control methods were compared. Concerning identification, simulation results showed that genetic algorithm reveals a better agreement in modeling the four types of considered models than least squares methods. On the other hand, it is shown also that the designed genetic based controllers are more reliable to stabilize the behaviors of the different studied processes than reference model controllers.