Distributed fault‐tolerant model predictive control for intermittent faults

Abstract

Intermittent fault is a special kind of fault form, which often occurs in industry processes but has been less studied. This study focuses on the distributed fault-tolerant model predictive control (DFTMPC) problem for an intermittent fault in actuators or sensors. A novel state observer and DFTMPC controller integral framework are presented. The proposed distributed system state observer can both deal with the unknown inputs and compensate for the interconnected variables' estimation errors, achieving a more accurate estimation. A fault mode-switched model is constructed to describe the system with multi-intermittent faults in a hidden way. This allows us to design a distributed fault-tolerant controller without reconfiguration. For the proposed distributed model predictive controller, asymptotically stable conditions are given, meanwhile, the whole algorithm for multi-intermittent fault-tolerant control is concluded. A numerical example is simulated to verify the proposed method.