APPLICATION OF A FUZZY LEARNING ALGORITHM TO NUCLEAR STEAM-GENERATOR LEVEL CONTROL

In order to reduce the load of tuning works by trial-and-error for obt aining the best control performance of conventional fuzzy control algo rithm, a fuzzy control algorithm with learning function is investigate d in this work. This fuzzy control algorithm can make its rule base an d tune the membership functions automatically by use of learning funct ion which needs the data from the control actions of the plant operato r or other controllers such as PI controller. Learning process in fuzz y control algorithm is to find the optimal values of parameters, which consist of the membership functions and the rule base, by gradient de scent method. Learning speed of gradient descent is significantly impr oved in this work with the addition of modified momentum term. This co ntrol algorithm is applied to the steam generator level control by com puter simulations. The simulation results confirm the good performance of this control algorithm for level control and show that the fuzzy l earning algorithm has the generalization capability for the relation o f inputs and outputs and it also has the excellent capability of distu rbance rejection.