Adaptive fuzzy control of a water bath process with neural networks

Abstract

Recently, the emergence of artificial neural networks has made it feasible to integrate fuzzy logic controllers and neural models for the development of adaptive fuzzy control systems. This paper proposes an adaptive fuzzy-neural control scheme by integrating two neural models with a basic fuzzy logic controller. Using the backpropagation algorithm, the first neural network is trained as a plant emulator and the second neural network is used as a compensator for the basic fuzzy controller to improve its performance on-line. The function of the neural network plant emulator is to provide the correct error signal at the output of the neural fuzzy compensator without the need for any mathematical modeling of the plant. The difficulty of fine-tuning the scale factors and formulating the correct control rules in a basic fuzzy controller may be reduced using the proposed scheme. The scheme is applied to the temperature control of a water bath process. Simulations show how the neural networks improve the performance of a poorly tuned basic fuzzy controller. The performance of the adaptive fuzzy-neural controller is compared to the basic fuzzy logic controller and a conventional digital-PI controller under identical conditions of varying complexities in the process. The experimental results show that the adaptive fuzzy-neural control scheme is superior in performance to the other two controllers.