Tuning of a Neuro-fuzzy Controller by Genetic Algorithms with an Application to a Coupled-tank Liquid-level Control System

Fuzzy logic, neural networks and genetic algorithms are three popular artificial intelligence techniques that are widely used in many applications. Due to their distinct properties and advantages, they are currently being investigated and integrated to form new models or strategies in the areas of system control. In this paper, a neuro-fuzzy controller (referred to as NFCGA) based on the radial basis function neural network is tuned automatically using genetic algorithms (GA). A linear mapping method is used to encode the GA chromosome, which consists of the width and centre of the membership functions, and also the weights of the controller. Dynamic crossover and mutation probabilistic rates are also applied for faster convergence of the GA evolution. Application of the NFCGA to a coupledtank liquid-level laboratory process is investigated in this paper. Compared to a manually-tuned conventional fuzzy logic controller and a PID controller which are applied to the same process, the NFCGA shows considerable robustness and advantages.