The use of neuro-fuzzy networks in the control of nonlinear systems

The recent success of the application of fuzzy logic in industry automation has motivated its use in the control of nonlinear systems. Its simplicity and the fact that is not a time consuming method, make it a very promising approach for this kind of control problems. Some difficulties arise, such as the adjustment of the rule base and the choice of the membership functions. A new approach that combines the learning capability of the neural networks with the simplicity of fuzzy logic has been identified as neuro-fuzzy methods. This paper presents an overview of various neuro-fuzzy approaches, including their special features. A DC motor with a nonlinear load is controlled using the fuzzy-neural method. The paper is concluded with an analysis of the simulation results