Fault detection for a class of strict-feedback systems via deterministic learning

A fault detection approach is proposed for a class of strict-feedback nonlinear systems with dynamic uncertainties in this paper. We present the approach based on deterministic learning consists of three phases: Firstly, by combining adaptive neural with backstepping method, a neural networks (NN) controller is designed for controlling a class of nonlinear closed-loop systems with dynamic uncertainties in strict-feedback form so as to achieve guaranteed the convergence of tracking errors in a finite time. Secondly, in the learning phase, the overall closed-loop system dynamics underlying normal and fault modes are locally accurately approximated through deterministic learning. The obtained knowledge of system dynamics is stored in constant RBF networks. Finally, in the detecting phase, a bank of estimators are constructed using the constant RBF networks to represent the learning normal and fault modes. By comparing the set of estimators with the monitored system, a set of residuals are generated, and the average L₁ norms of the residuals are taken as the measure of the differences between the dynamics of the monitored system and the dynamics of the learning normal and fault modes. The occurrence of a fault can be rapidly detected according to the smallest residual principle. Simulation studies are given to demonstrate the effectiveness of the proposed approach.