Adaptive Threshold Generation for Robust Fault Detection Using Interval LPV Observers

Abstract In this paper, adaptive threshold generation for robust fault detection of nonlinear system described by means of a Linear Parameter Varying (LPV) model is analyzed. Uncertainties due to parameter variations are considered unknown but bounded by intervals. Their effect is addressed using an interval LPV observer based on zonotopes. The design procedure of this observer is implemented via pole placement using LMIs. The optimal threshold is generated by evaluating the worst-case residual's energy evolution in the time domain. When this adaptive threshold is used, the minimum detectable fault is characterized. Finally, an application example based on a two-degree freedom helicopter will be used to assess the validity of the proposed approach.