THE ADVISOR, A FUZZY KNOWLEDGE-BASED SYSTEM FOR RAPID INSTRUMENTATION AND CONTROL USING A TAGUCHI-BASED QUALITY FUNCTION: PART Ⅰ-THEORY

摘要

A fuzzy knowledge-based software system, called The Advisor, is developed for mechanical designers to rapidly select instrumentation and control system solutions. Its knowledge-based system uses application constraints to select feasible solutions, and its fuzzy logic algorithms rank those solutions based upon user preferences. A new Quality Function, inspired by the Taguchi Quality Function, is proposed to combine quantitative and linguistic information in the fuzzy decision-making inference engine to better capture human tradeoff behavior in the equipment selection process. To date, a Motor Advisor and a Temperature Sensor Advisor have been developed and successfully tested. Part Ⅰ of this paper presents the theory and methodology of The Advisor, its component algorithms to support the selection of motor systems, and the implementation of the new Quality Function in the Motor Advisor. Part Ⅱ tests The Advisor in the selection of a load motor, motor driver, and motor controller in a motor failure design project.