Robust fault detection for the dynamics of high-speed train with multi-source finite frequency interference

Area of Science:

• Control Systems Engineering
• Transportation Engineering
• Signal Processing

Background:

• High-speed trains (HST) are susceptible to faults caused by environmental factors like wind gusts and operational noise.
• Accurate fault detection is crucial for ensuring the safety and reliability of HST operations.
• Existing fault detection methods may struggle to effectively distinguish between faults and external disturbances.

Purpose of the Study:

• To develop a composite fault detection scheme for HST dynamics.
• To design an unknown input observer-like (UIO-like) fault detection filter robust to disturbances.
• To ensure prompt and reliable fault detection by decoupling disturbances and enhancing fault sensitivity.

Main Methods:

• Utilizing system input and output measurements to design a fault detection filter generating residual signals.
• Proposing a disturbance partitioning method to decouple external influences from residual dynamics.
• Defining detection subspaces to assign specific faults, ensuring diagonal influence on residual signals.
• Applying H_∞ performance index within a finite frequency domain to constrain residual dynamics.

Main Results:

• The proposed UIO-like filter effectively generates residual signals for fault detection.
• Disturbance decoupling is achieved by partitioning into non-influential and H_∞-constrained subsets.
• Faults are assigned to distinct detection subspaces, enabling prompt and isolated detection.
• Simulations confirm the effectiveness and advantages of the developed composite fault detection scheme.

Conclusions:

• The composite fault detection scheme provides a robust and effective solution for HST.
• The method successfully mitigates the impact of wind gusts and operating noises.
• The proposed approach enhances the reliability and safety of high-speed train operations through precise fault identification.