Robust Integration of Model-Based Fault Estimation and Fault-Tolerant Control

When Fault Diagnosis and Fault-Tolerant Control Must Work Together

In industrial, aerospace, and networked systems, a fault is rarely just a single isolated event. It can degrade performance, push a subsystem toward instability, or propagate through interconnected plants. Fault estimation can provide a direct reconstruction of the fault, while fault-tolerant control can compensate for its effects. The hard part is making those functions work together robustly when uncertainties are present. This monograph addresses that integration problem.

From Fault Estimation to Compensating Control

The book centers on model-based fault estimation rather than fault detection and isolation alone. That choice matters because estimated fault shapes can be used directly in the control law, opening a route to what the authors call true integration of fault diagnosis and fault-tolerant control. The discussion covers basic concepts, integration strategies, and practical demonstrations, with attention to the bi-directional robustness interactions that arise between estimation and control.

Five Integration Strategies for Linear Systems

• Sequential integration — fault estimation and fault-tolerant control are designed in a defined order, with feasibility analyzed under ideal and uncertain conditions.
• Iterative integration — baseline and enhanced control gains are refined through an iterative algorithm.
• Simultaneous integration — reduced-order observer-based fault estimation and sliding mode fault-tolerant control are addressed together.
• Robust decoupling — design conditions are developed to reduce unwanted interactions between estimation and control.
• Adaptive decoupling — adaptation is used to handle uncertainty while maintaining integration performance.

Each strategy is accompanied by tutorial examples that show how the design conditions translate into observer and controller synthesis.

Beyond Linear Models: Nonlinear and Large-Scale Systems

The later chapters extend the framework beyond linear systems. Nonlinear and large-scale systems are considered, along with applications in renewable energy, robotics, and network systems. These chapters connect the theoretical development to industrial areas where reliability and safety are central concerns.

Who Will Benefit from This Research Monograph

This book is intended for researchers working on fault diagnosis and fault-tolerant control in universities and industry. It can also serve as supplementary reading for graduate or postgraduate courses on fault diagnosis and FTC. Readers who already have a foundation in control theory and observer design will find a systematic treatment of an open research problem, together with design strategies and application examples that point toward further work.