Active Fault Diagnosis and Fault-Tolerant Control in Dynamic Systems

Active fault diagnosis and fault-tolerant control represent an integrated approach to ensuring the safe and uninterrupted operation of dynamic systems in the presence of component failures or unexpected disturbances. Active diagnosis involves the deliberate injection of auxiliary signals or excitation inputs to provoke characteristic system responses that reveal incipient or subtle faults, thereby enhancing detectability and isolation. Fault-tolerant control builds on diagnostic information by reconfiguring control laws or adapting parameters in real time to mitigate the impact of faults on performance. Together, these techniques form a closed-loop strategy in which diagnosis informs control decisions, and control adaptations facilitate more effective diagnosis. Applications span aerospace, industrial process control, automotive systems, energy networks and robotics, where safety, reliability and continuous operation are paramount. Contemporary research emphasises computational efficiency, robustness to model uncertainty and bounded disturbances, and the integration of model predictive control with active diagnosis to achieve fast response, minimal performance degradation and automated decision making under constraints.

Research from Nature Portfolio

No recent Nature Portfolio content available.

Active Fault Diagnosis and Fault-Tolerant Control in Dynamic Systems publication trend

The graph below shows the total number of articles in active fault diagnosis and fault-tolerant control in dynamic systems across all publications each year (not limited to Nature Index journals).

Technical terms

Active fault diagnosis: A method that injects designed auxiliary signals into a system to amplify fault-related signatures and improve detection and isolation.

Fault-tolerant control: A control strategy that adapts or reconfigures control actions in response to diagnosed faults to preserve system performance and safety.

Model predictive control (MPC): An optimisation-based control technique that predicts future behaviour over a finite horizon and solves a constrained optimisation problem at each time step.

Residual generation: The process of computing a discrepancy signal between measured outputs and model-based estimates, used to detect deviations indicative of faults.

Linear parameter-varying (LPV) system: A dynamic system model whose parameters change over time within known bounds, enabling representation of nonlinearity and uncertainty.

Separation hyperplane: A geometric boundary that separates data or reachable sets corresponding to different system configurations or fault conditions.

Robust positively invariant (RPI) set: A set of states from which, under bounded disturbances, the system trajectories remain inside indefinitely, ensuring constraint satisfaction and stability.

References

1. Constrained Active Fault Tolerant Control Based on Active Fault Diagnosis and Interpolation Optimization. Entropy (2021).
2. Active Fault Isolation for Multimode Fault Systems Based on a Set Separation Indicator. Entropy (2023).
3. Model-based sensor fault detection, isolation and tolerant control for a mine hoist. Measurement and Control (2022).