ABSTRACT
This chapter proposes a hybrid learning system based on Artificial Immune Network Fuzzy System and Artificial Immune Network-Fuzzy Neural Network System for the optimal control of a nonlinear system. A significant challenge faced by engineers is the control of the dynamic characteristics in highly nonlinear industrial systems. It is well known that most industrial control systems exhibits nonlinearities from such processes as dynamics of fluids, electronic circuits, processes, or the mechanics of complex systems. The optimal shape of the fuzzy membership function is provided by the concentration of antibody, depending on error between the set point and the nonlinear system output. The hybrid system tunes the shape of fuzzy membership function to optimally control and reflect the characteristics of the nonlinear system. The fuzzy-neural network system learning represents one of the most effective existing algorithms used to build the linguistic models for control systems or decision making.
