Backstepping-based distributed abnormality localization for linear parabolic distributed parameter systems (Q2059321)

This paper focus attention on the abnormality localization issue of the distributed parameter systems with related to the spatial domain, which is different from the fault detection in time domain. A systematic framework is presented to settle this issue for a class of linear parabolic PDEs by additional a limited number of in-domain measurements except one boundary measurement rather than the full-state measurement. The proposed methodology involves two designs: one is an abnormality detection filter (ADF) design and the other is an abnormality localization filter (ALF) design. Employing the backstepping transformation and eigen-spectrum, the residual evaluation functions are directly constructed. As an verification of the effectiveness of the proposed method, numerical simulations on a heat transfer rod are conducted.