Optimization-Based Incipient Fault Isolation for the High-Speed Train Air Brake System

The high-speed train (HST) air brake system (ABS) is crucial to guarantee a safe and reliable operation environment for passengers on board. Recently, a novel combined index has been proposed for incipient fault detection of the ABS. Following detection, fault isolation is quite desired so as to find out the root causes of the occurred fault and provide valuable information for preventive maintenance and repairs. In this work, the combined index is briefly revisited and its detectability analysis is first formally developed, which can reveal certain detection properties theoretically. Then, a fault-recovery-based optimization strategy is presented for fault isolation purpose. Through theoretical reasoning it is indicated that, with accurately assumed fault direction information, the proposed methodology is able to accomplish correct isolation. Extensive tests on a practical ABS platform demonstrate the effectiveness of the proposed fault detection and isolation approaches.