Frequency Domain Design and H∞ Optimization Methods For Robust Fault Diagnosis
As described in previous chapters, there are many ways, such as the unknown input observer, eigenstructure assignment, optimally robust parity relations, for eliminating or minimizing disturbance and modeling error effects on residual and hence for achieving robustness in FDI. Whilst these techniques are different, one feature is common among them, the original frameworks of these methods were developed for ideal systems or with a special uncertainty structure and then efforts have been made to include non-ideal or more general uncertainties. In contrast, H∞ optimization is a robust design method with the original motivation firmly rooted in the consideration of various uncertainties, especially the modeling errors. H∞ optimization has been developed from the very beginning with the understanding that no design goal of a system can be perfectly achieved without being compromised by an optimization in the presence of uncertainty. Hence this technique is very suitable for tackling uncertainty issues. After two decades of development, it is now playing a leading role in tackling the robustness problem in control systems. It is reasonable to seek the application of these examines in other areas, including the robust design of FDI systems. This chapter studies the frequency domain approaches, including H∞ optimization, for robust fault diagnosis.
KeywordsTransfer Matrix Fault Diagnosis Linear Matrix Inequality Disturbance Attenuation Fault Estimation
Unable to display preview. Download preview PDF.