Fuzzy supervisor control system based on human's experience and know-how and its application to hydrogen purity controls in petroleum plant

摘要

This paper presents a practical control system that is designed based on human's experience and control know-how, and plays the role of supervisors of conventional PID controllers. Each supervisory controller consists of two control blocks, estimation block and compensation block. We use a statistical model using multi-regression analysis for the estimation block to estimate parameters of plant operation, and fuzzy logic for the compensation block to correct operating condition. The supervisor and PID control systems have been applied to hydrogen purity control in a large scale petroleum refining plant. The systems have shown good performances and therefore have advantages of (1) reducing operator's interaction, (2) reducing energy consumption, and (3) reducing CO{sub}2 generation.