Fuzzy and Neuro-fuzzy Techniques for Modelling and Control

Abstract

This paper presents comparative evaluation of two fuzzy-derived techniques for modelling fuel spray penetration in the cylinder of a diesel internal combustion engine and a fuzzy control system for a small internal combustion engine. The first technique used a pure fuzzy paradigm, the parameters of this technique are a collection of intuitively comprehensible rules and fuzzy-set membership functions. While the visual nature of this system facilitates the optimisation of the parameters, the need for this to be accomplished manually is a disadvantage. The second technique used an adaptive neuro-fuzzy inference system (ANFIS), where automatic adjustment of the system parameters was effected by a neural network based on prior knowledge. The ANFIS exhibited improved accuracy compared to a pure fuzzy model. It also has the advantage over the pure fuzzy paradigm that the need for the human operator to tune the system by adjusting the bounds of the membership functions is removed. Future work is concentrating on the establishment of an improved neuro-fuzzy paradigm for adaptive, fast and accurate control of small internal combustion engines.