Paper Title
DESIGN OF NEURO FUZZY SYSTEMS
Abstract
Classical control theory is based on the mathematical models that describe the physical plant un-der consideration.
The essence of fuzzy control is to build a model of human expert who is capa-ble of controlling the plant without thinking in
terms of mathematical model. The transformation of expert's knowledge in terms of control rules to fuzzy frame work has not
been formalized and arbitrary choices concerning, for example, the shape of membership functions have to be made. The
quality of fuzzy controller can be drastically affected by the choice of membership functions. Thus, methods for tuning the
fuzzy logic controllers are needed. In this paper, neural networks are used in a novel way to solve the problem of tuning a fuzzy
logic controller.
The neuro fuzzy controller uses the neural network learning techniques to tune the member-ship functions while keeping the
semantics of the fuzzy logic controller intact. Both the architec-ture and the learning algorithm are presented for a general
neuro fuzzy controller. From this gen-eral neuro fuzzy controller, a proportional neuro fuzzy controllers is derived. A step by
step algo-rithm for off-line training is given along with numerical examples.