A dynamic diagnostic method for consecutive faults in nonlinear uncertain systems

This paper proposes a robust fault diagnosis method for nonlinear uncertain systems with multiple faults, addressing the possible occurrence of two consecutive faults in each state equation. A model-based monitoring module with two submodules is developed, enabling the diagnosis of both faults. The first submodule incorporates decision-making schemes for detecting and isolating the primary fault, enabling its partial or full isolation. The second submodule is introduced to detect the secondary fault while simultaneously determining the partial or full isolation of both the primary and secondary faults. A key design aspect of the proposed fault diagnosis method is that it effectively uses the information from the primary fault isolation process, particularly when partial isolation occurs, to detect and isolate a secondary fault in each system state equation. The boundedness of the system variables and the robustness of the proposed fault diagnosis scheme are analytically shown. Finally, the effectiveness and applicability of the developed framework is demonstrated through numerical simulations.