Simultaneous fault detection and isolation and fault-tolerant control using supervisory control technique: asynchronous switching approach

Abstract

This article focuses on the design of a novel active fault-tolerant control scheme based on supervisory control technique for a class of nonlinear systems. This framework relies on a supervisory switching among a finite family of predesigned candidate controllers, which simultaneously performs isolation and accommodation of intermittent faults. This method does not require any additional model or filter bank for fault isolation. Two controller switching algorithms are introduced based on the dwell time and state which are designed especially for this purpose. There is often some time delay between fault occurrence and accommodation. This delay, which is called as the fault detection and isolation delay, causes the asynchronous switching between the system mode and the candidate controller. For the investigation of the stability of the faulty system under asynchronous switching, we explicitly construct piecewise Lyapunov function based on the knowledge of the known Lyapunov function for each operating mode. Then, by using this piecewise Lyapunov function, a new average dwell-time condition is provided on the maximum admissible fault occurrence rate. This condition guarantees the input-to-state stability of the system with respect to the reference signal. The behavior and performance of the proposed fault-tolerant control/fault detection and isolation scheme are demonstrated on the pitch-axis air vehicle model. The hardware-in-the-loop simulation is an important test for the evaluation of the air vehicle autopilot system before flight test. Therefore, the hardware-in-the-loop simulation results are presented to illustrate the effectiveness of the proposed method in the autopilot control loop.

Using extended Kalman filter for failure detection and prognostic of degradation process in feedback control system

“…For a given level of degradation in an actuator in a closed-loop system, the control effort could be insufficient, so as to maintain a specific performance. In current researches, 14,15 the authors have developed the fault-tolerant control (FTC) approach to overcome this difficulty. It can generally mitigate a few effects of system components failures.…”