GA type IMC control for pneumatic servo system

This paper presents a novel control scheme in which internal model control (IMC) is combined with a genetic algorithm (GA). IMC tuning rules have a number of advantages for enhancing control performance. IMC can minimize disturbance and use only one tuning parameter, so that it is attractive for industrial users. For more aggressive control, we decrease the tuning parameter (more rapid response); for more robust response, we increase the tuning parameter (slower response). Usually we design the tuning parameter by trial and error. But it is not so easy to design an optimal parameter for a specific system, especially for a nonlinear system. A neural network (NN) based IMC for a pneumatic servo system has been proposed. On the other hand, in the last twenty years, GA has been attractive as a method, which gives an optimal answer for global search, optimization and machine learning problems. In this paper, we propose an intelligent control scheme in which IMC is combined with GA. GA will get the optimal or "near optimal" tuning parameter on-line after some generations. The effectiveness of the proposed scheme is confirmed by experiments using the existent pneumatic servo system.