FUZZY CONTROLLER-DESIGN BY USING NEURAL-NETWORK TECHNIQUES

This paper investigates the relationship between the piecewise linear fuzzy controller (PLFC), in which the membership functions for fuzzy variables and the associated inference rules are all in piecewise linear forms, and a Gaussian potential function network based controller (GPFNC), in which the network output is a weighted summation of hidden responses from a series of Gaussian potential function units (GPFU's). Systematic procedures are proposed for transformation from a PLFC to its GPFNC counterpart, and vice versa. Based on these transformation principles, a series of systematic and feasible steps is presented for the design of an optimized PLFC (PLFC*) by using neural network techniques. In the design procedures, the simplified PLFC is used as the initial controller structure, then a GPFNC, which gives the approximate control response to the initially given PLFC, is found for further optimization. The optimized GPFNC (GPFNC*) can be implemented directly to actual systems, and the GPFNC* could further be converted into Its fuzzy counterpart (PLFC*) if more structural interpretation of the intelligent control strategy is required. Several numerical examples are supplied to illustrate the fact that the the proposed design procedures could result in an optimal neural or fuzzy controller with superior servo control performance. A neutralization process is also used to demonstrate the feasibility and the potential applicability of these intelligent controllers on the regulation of highly nonlinear chemical processes.