PERFORMANCE EVALUATION OF AN INDUSTRIAL MPC CONTROLLER
Jianping Gao1, K. Akamatsu2, Y. Hashimoto2, S. L. Shah3, B. Huang
Department of Chemical and Materials Engineering, University of Alberta, Edmonton, Canada T6G 2G6
This paper presents an industrial case study of the performance evaluation of a multivariate MPC-based controller as implemented on a 6-output, 6-input industrial process with 6 measured disturbance variables that are used for feed forward control. The industrial unit is a Para-Xylene (PX) production process at Mitsuibishis petrochemical complex in Mizushima, Japan. A generalized predictive controller-based MPC algorithm has been implemented on the PX process. Data from the PX unit before and after the MPC implementation is analyzed to obtain and compare several different measures of multivariate controller performance.
Session slot T-Mo-A11: New Directions in Control System Performance Monitoring/Area code 7a : Chemical Process Control

|