Hxd1 Locomotive Braking System Fault Diagnostic Techniques And Applications

The brake system is the core device of HXD1 electric locomotive, to ensure the safe operation of locomotive, real-time accurate monitoring and effective online diagnosis of the braking system is an urgent and important issue. According to the difficulty of transducer fault diagnosis and the distributed modular construction feature of the braking system, in this paper, with the purpose of fault online diagnosed precisely, two key technologies containing the fault diagnosis of important sensors in brake system and the flexible reconfigurable implementation of online dia-gnostic system are researched.Firstly, the fault characteristics of HXD1 locomotive brake system are studied. Fault is divided into hard and soft fault to classified research, and a set program of fault diagnosis system is proposed in which hard fault is diagnosed based on knowledge and soft fault is diagnosed based on PCA. Focus on the difficulty in soft fault diagnosis of sensors, such as the accurate analytical models are difficult to obtain and the knowledge diagnostic which rules are inapplicability, principal component analysis algorithm (PCA) based on data-driven is introduced to online diagnose. Because of the multiple working conditions during brake system's running, Online fault diagnosis method based on multiple working conditions and multiple principal component models for locomotive brake system is developed, in which a steady factor is defined to judge the steady state, and the PCA model changes according to the change in working conditions.Through experiment and simulation results the ability of PCA method in soft-fault diagnosis is verified.Secondly, for the inaccuracies of model precision and determine parameters of the traditional static PCA method when the brake system running in a fixed condition for long time, a PCA diagnosis method based on the adaptive step mechanism is proposed in this paper. The principle component model of the braking system is updated by a suitable step to follow the drift of operating parameters of the braking system instantaneously. The simulation experiments shows by means of this method the fault can be diagnosed more rapidly and accurately than the traditional static PCA, so the accuracy and real-time characteristics of the soft fault online diagnosis are improved.Finally, according to the variable configuration and complex inter-actions between modules, the hardware and software implementations are designed based on the flexible reconfigurable method. The online diagnosis system is designed by modular method, when the working module of braking system changes or in an unusual circumstances, the diagnosis system can realize the following response in a flexible reconfigurable way. The practical operation results of diagnsis system running on Datong-Qinhuangdao line provided, which prove that the online diagnosis system designed in this paper is totally feasible and effective in the protection of the safe operation of locomotive.