Fault Diagnosis Based on Causal Computations

Abstract

Abstract-This work focuses on residual generation for modelbased fault diagnosis. Specifically, a methodology to derive residual generators when non-linear equations are present in the model is developed. A main result is the characterization of computational sequences that are particularly easy to implement as residual generators and that take causal information into account. An efficient algorithm, based on the model structure only, that finds all such computational sequences, is derived. Further, fault detectability and fault isolability performance depend on the sensor configuration. Therefore, another contribution is an algorithm, also based on model structure, that places sensors with respect to the class of residual generators that take causal information into account. The algorithms are evaluated on a complex, highly non-linear, model of a fuel cell stack system. A number of residual generators are computed that are, by construction, easy to implement and provide full diagnosability performance predicted by the model. Index Terms-Fault diagnosis, causal computations, sensor placement, fuel cell stack system.

Sensor placement for fault diagnosis performance maximization in Distribution Networks

“…In this paper, this issue has been ignored. However, functions F D (S) and I(S) could be adapted to take into account this constraint in the sensor placement analysis phase, by following the causality framework introduced in [14]. Then, the solution obtained from the sensor placement analysis would guarantee a set of easily computable residual generators.…”

Model-based Optimal Sensor Placement Approaches to Fuel Cell Stack System Fault Diagnosis*

“…In this paper, this issue has been ignored. However, any of the three approaches could be adapted to take into account this issue in the sensor placement analysis phase, by following the causality framework introduced in Rosich et al (2012). Then, the solution obtained from the sensor placement analysis would guarantee a set of particularly easy computable residual generators.…”

“…Then, the optimal search on the candidate sensor set is performed. Finally, in (Rosich et al, 2012), a method that takes into account the causal computability of the unknown variables in the residual generation is developed. This paper presents a comparative study of three modelbased optimal sensor placement approaches: a heuristic search , an incremental algorithm and a Binary Integer Linear Programming formulation approach (Rosich et al, 2009) .…”

Fault Detection and Isolation in Critical Infrastructure Systems

“…In (Sarrate et al, 2014), a structural model of a water distribution network is obtained for FDI system design. See (Rosich et al, 2012) and (Travé-Massuyés et al, 2006) for a comprenhensive description of ARR design methodologies based on structural analysis.…”