Collaborative Diagnosis of Spatiotemporal Faults and Sensor Anomalies in Parabolic Distributed Parameter Systems

Area of Science:

• Control Systems Engineering
• Industrial Process Monitoring
• Fault Diagnosis

Background:

• Industrial processes like heat transfer and chemical diffusion are complex distributed parameter systems (DPSs).
• These systems exhibit strong spatiotemporal (S-T) coupling, where component malfunctions pose significant safety risks.
• Existing diagnostic methods may not adequately address the simultaneous occurrence of process faults and sensor anomalies.

Purpose of the Study:

• To propose a model-based framework for the collaborative diagnosis of S-T faults and sensor anomalies in DPSs.
• To develop effective fault detection and isolation (FDI) algorithms for these complex systems.
• To design a cooperative fault estimation algorithm that handles the coexistence of multiple fault types.

Main Methods:

• Utilizing a reduced-order model derived from the spectral method.
• Establishing two distinct observer sets for process faults and sensor anomalies.
• Developing FDI algorithms based on fault characteristics and employing an unknown input observer (UIO) for cooperative estimation.
• Ensuring stability and convergence using the Lyapunov direct method.

Main Results:

• The proposed FDI algorithm effectively detects and isolates S-T faults and sensor anomalies in a heat-transfer rod model.
• The cooperative fault estimation algorithm demonstrates high accuracy, with a root-mean-square error (RMSE) below 0.31.
• Numerical simulations validate the robustness and effectiveness of the collaborative diagnosis framework.

Conclusions:

• The developed model-based framework provides a reliable solution for collaborative fault diagnosis in DPSs.
• The proposed methods enhance safety and operational integrity in industrial processes with S-T coupling.
• This approach offers a significant advancement in diagnosing complex faults and sensor anomalies simultaneously.