Abrupt and incipient fault isolation of nonlinear uncertain systems

This paper presents a robust fault diagnosis scheme for abrupt and incipient faults in nonlinear uncertain dynamical systems. A detection and approximation estimator is used for online health monitoring. Once a fault is detected, a bank of isolation estimators are activated for the purpose of fault isolation. A key design issue of the proposed fault isolation scheme is the derivation of adaptive residual thresholds associated with each isolation estimator. An occurred fault is isolated if the residual associated with the matched isolation estimator remains below its corresponding adaptive threshold, while at least one of the components of the residual associated with all other estimators exceeds its threshold at some finite time. Fault isolability conditions are derived for both the abrupt and the incipient fault cases, guaranteeing that the diagnosis scheme reaches the correct decision in finite time.