NONLINEAR MODEL DECOMPOSITION FOR ROBUST FAULT DETECTION AND ISOLATION USING ALGEBRAIC TOOLS

Denis Berdjag (1) , Cyrille Christophe (2) , Vincent Cocquempot (2) , Bin Jiang

1 LAGIS - Laboratoire d'Automatique, Génie Informatique et Signal
2 STF - Systèmes Tolérants aux Fautes

This paper is concerned with fault detection and isolation in nonlinear dynamic systems. A structural decomposition of the system is made, using the criterion of robustness to unknown inputs. To achieve this decomposition, special mathematical tools are used. The algebraic nature of these tools permits a more general formulation of the residual generation problem. These mathematical tools are presented with illustrations and the unknown input-free decomposition method is described. Some aspects of the method are discussed, and solutions are proposed to improve the existing algorithm. An application on an academic example is also provided.

Mots clés: system decomposition; Residual generation; Decoupling; Nonlinear systems; Algebra of functions

Domaines: Automatique

Denis Berdjag, Cyrille Christophe, Vincent Cocquempot, Bin Jiang. NONLINEAR MODEL DECOMPOSITION FOR ROBUST FAULT DETECTION AND ISOLATION USING ALGEBRAIC TOOLS. International Journal of Innovative Computing, Information and Control, 2006, 2 (6), pp.1337-1354. ⟨hal-00198435⟩