Control allocation based fault-tolerant strategy for a bio-ethanol processor system integrated to a PEM fuel cell. (September 2019)

• Record Type: Journal Article
• Authors: Luppi, P.A.; Braccia, L.; Patrone, M.; Zumoffen, D.
• Is Part Of: Journal of process control. Volume 81(2019)
• Page Start: 40
• Page End: 53
• Publication Date: 2019-09
• Subjects: Bio-ethanol processor -- Hydrogen production -- Proton exchange membrane fuel cell -- Fault-tolerant control -- Control allocation; Process control -- Periodicals; Fabrication -- Contrôle -- Périodiques; Process control; Periodicals; Electronic journals; 660.281
• DOI: 10.1016/j.jprocont.2019.05.021
• Languages: English
• ISSNs: 0959-1524

Abstract

Highlights: Control structure design is exclusively based on a steady-state model of the process. Partial/total actuator faults are efficiently handled without reconfiguring the high-level controller. Control and optimization tasks are efficiently integrated demanding a reduced computation time. On-line controller re-allocation allows to manage not predefined actuator faults. System references tracking and disturbance rejection denote remarkable performance. Abstract: This paper presents a control approach to obtain actuator fault tolerance in a bio-ethanol processor system coupled with a proton exchange membrane fuel cell. The proposed strategy is based on the development of two cascaded modules. First, a high-level controller consisting on single-input single-output conventional proportional-integral control loops is designed for computing virtual actions which represent the overall control effort. Second, a control allocation module is proposed for the on-line redistribution of the virtual commands onto the available healthy actuators. This scheme is able to compensate from actuator position/rate saturations, to severe abnormal events such as loss of effectiveness and lock-in-place actuator faults. Some attractive features of the proposal are: (i) the control structure design is exclusively based on a steady-state model of the process, (ii) different actuator faults can be efficiently handled without the need of reconfiguring the high-level controller, (iii) the control and optimization tasks are efficiently integrated demanding a reduced computation time, (iv) the on-line controller re-allocation allows to manage not predefined actuator faults, (v) a considerable set of disturbances can be rejected with remarkable performance. Dynamic simulations performed on a rigorous nonlinear model of the process show the benefits of the proposed strategy for both fault-free as well as different fault situations comprising partial and total actuator faults.