Robust tuning for machine-directional predictive control of MIMO paper-making processes. (October 2016)

• Record Type: Journal Article
• Authors: He, Ning; Shi, Dawei; Forbes, Michael; Backström, Johan; Chen, Tongwen
• Is Part Of: Control engineering practice. Volume 55(2016)
• Page Start: 1
• Page End: 12
• Publication Date: 2016-10
• Subjects: MIMO systems -- Model predictive control -- Controller tuning -- Time-domain specifications -- Parametric uncertainty -- Paper machines; Automatic control -- Periodicals; 629.89
• DOI: 10.1016/j.conengprac.2016.06.008
• Languages: English
• ISSNs: 0967-0661

Abstract

Abstract: This paper solves the controller tuning problem of machine-directional predictive control for multiple-input–multiple-output (MIMO) paper-making processes represented as superposition of first-order-plus-dead-time (FOPDT) components with uncertain model parameters. A user-friendly multi-variable tuning problem is formulated based on user-specified time domain specifications and then simplified based on the structure of the closed-loop system. Based on the simplified tuning problem and a proposed performance evaluation technique, a fast multi-variable tuning technique is developed by ignoring the constraints of the MPC. In addition, a technique to predict the computation time of the tuning algorithm is proposed. The efficiency of the proposed method is verified through Honeywell real time simulator platform with a MIMO paper-making process obtained from real data from an industrial site. Abstract : Highlights: A user-friendly MPC tuning problem for MIMO paper-making processes is studied. A fast multi-variable MPC tuning technique is developed. A method to predict the computation time of the MPC tuning algorithm is proposed.