Design of state space linear quadratic tracking control using GA optimization for batch processes with partial actuator failure. (February 2015)

• Record Type: Journal Article
• Authors: Zhang, Ridong; Jin, Qibing; Gao, Furong
• Is Part Of: Journal of process control. Volume 26(2015:Feb.)
• Page Start: 102
• Page End: 114
• Publication Date: 2015-02
• Subjects: Batch processes -- Linear quadratic control -- Genetic algorithm -- Actuator failure; Process control -- Periodicals; Fabrication -- Contrôle -- Périodiques; Process control; Periodicals; Electronic journals; 660.281
• DOI: 10.1016/j.jprocont.2014.11.008
• Languages: English
• ISSNs: 0959-1524

Abstract

Highlights: A state space model based LQ control scheme using GA optimization for batch processes with unknown disturbances and partial actuator faults is proposed. The design is applied to injection velocity control. Improved closed-loop control performance is achieved even with unknown disturbances and actuator faults. Abstract: This paper presents a new state space model design of linear quadratic (LQ) tracking control scheme using genetic algorithm (GA) optimization for batch processes under partial actuator faults. To develop the LQ control, the process model is first transformed into a new state space formulation that includes both the process state and the output tracking error dynamics. Then the subsequent LQ control is formulated. As analytical tuning rules of the elements in the weighting matrices are not known, a GA is introduced to optimize these elements to achieve the desired closed-loop process responses. Simulation on several processes demonstrates that the proposed LQ scheme obtains the design objective well with improved control performance.