Abstract:A goal in many applications is to combine a priori knowledge of the physical system with experimental data to detect faults in a system at an early enough stage as to conduct preventive maintenance. The mathematical model of the physical system is the information available before hand. One of the key issues in the design of model-based fault detection schemes is the effect of the model uncertainties such as severe parametric uncertainty and unmodeled dynamics on their performance. This paper presents the application of a nonlinear model-based adaptive robust observer design to help in the detection of faults in sensors for a class of nonlinear systems. The observer is designed by explicitly considering the nonlinear dynamics of the system under consideration. Robust filter structures are used to attenuate the effect of the model uncertainty which combined with controlled parameter adaptation helps in reducing the extent of model uncertainty and in increasing the sensitivity of the fault detection scheme to help in the detection of incipient failure. This continuous monitoring of the state estimates enables us to detect any off- nominal system behavior and detect faults even in the presence of model uncertainties. I. INTRODUCTION Model-based fault detection and diagnosis systems have found extensive use because of the fast response to abrupt failure and the ease of implementation of the model-based FDD systems in real-time algorithms. The most common methods for model-based fault detection are either based on state estimation or parameter estimation. A comprehensive review of the different methods for fault detection and their applicability to a given physical system has been presented in (1). A detailed review of the different methods for observer based fault detection is given in (2), (3). The major challenge in the design of model-based FDD systems is that they rely on an idealized assumption that a perfect mathematical model of the physical process is available and this model is utilized in the design of state- observers to detect sensor faults. In practice however, model uncertainty is always present and hence, the fault detection algorithm should be robust to the presence of model un- certainty. Another conflicting design requirement for a FDD scheme is to be able to detect incipient failure as early as possible to help in preventive maintenance. This would require an increase in the sensitivity of the FDD scheme. Hence, the robustness of the fault detection algorithm to modeling errors without losing sensitivity to faults is the key problem in the application of model based fault detection algorithms.

Fault-Structure-Based Active Fault Diagnosis: A Geometric Observer Approach

Active Fault Diagnosis for Linear Systems: Within a Signal Processing Framework

Active fault detection based on set‐membership approach for uncertain discrete‐time systems

Exclusion Tendency-Based Observer Design Framework for Active Fault Diagnosis

Robust Fault Diagnosis for Uncertain Linear Systems: A Set-Based Geometric Approach

A combined passive-active method for diagnosing multiplicative fault

A Geometric Approach to Fault Detection and Isolation of Linear Discrete-time Systems.

Joint Observer Gain and Input Design for Asymptotic Active Fault Diagnosis

An Optimal Fault Detection Approach for Piecewise Affine Systems Via Diagnostic Observers.

Observer‐based fault diagnosis for autonomous systems

Pressure Sensor Fault Diagnosis by Using Adaptive Robust Observers

Fault estimation for nonlinear systems: an observer structure design criterion technique

Concurrent Fault Detection and Anomaly Location in Closed-Loop Dynamic Systems with Measured Disturbances.

Integrated Design of Input and Observer Gain for Active Fault Diagnosis Based on Hybrid Stochastic-Deterministic Approach.

DISCUSSION OF SOME IMPORTANT ISSUES OF OBSERVER BASED FAULT DIAGNOSIS TECHNIQUE

Active Fault Diagnosis for Uncertain LPV Systems: A Zonotopic Set-Membership Approach

Robust Model-Based Fault Detection Using Adaptive Robust Observers

A Novel Geometry Optimization Strategy to Online Active Fault Diagnosis of LPV Systems.

Fuzzy Observer-Based Fault Detection Design Approach for Nonlinear Processes

Active Fault Detection Based on Residual Ellipsoid

An alternative paradigm of fault diagnosis in dynamic systems: orthogonal projection-based methods