Artificial hydrocarbon networks fuzzy inference systems for CNC machines position controller

Abstract

This paper proposes a novel position controller for computer numerical control (CNC) machines based on a hybrid fuzzy inference system that uses artificial hydrocarbon networks in its defuzzification step, so-called fuzzy-molecular inference system. The fuzzy-molecular-based position controller is characterized to improve the accuracy in position and the time machining. In order to prove these characteristics, a case study was run over a reconfigurable micromachine tool (RmMT) assembly in lathe configuration. In addition, a workpiece machining in the RmMT assembly serves to realize a comparative analysis between the proposed controller and three other controllers: a classical PID controller manually tuned, a PID controller auto-tuned, and a fuzzy Mamdani controller. Experimental results validate the performance and the implementability of the proposed fuzzy-molecular position controller against the others.