Sensor selection and placement using low complexity dynamic programming
Date of Issue2012
IEEE Conference on Prognostics and Health Management (2012 : Denver, Colorado, US)
School of Electrical and Electronic Engineering
In this paper, a novel approach is proposed for sensor selection and placement in systems for the purpose of fault detection and isolation (FDI). This new approach benefits from the basic fact that faults are embedded in the analytical redundancy relations (ARRs) and that the occurrence of a fault will cause the corresponding ARRs to change. For FDI purposes, each ARR is connected to a set of sensors that represent the measurable variables. New concepts of fault associated sets and fault distinguishable sets are introduced to develop a low complexity dynamic programming algorithm to minimize the number of sensors needed and simultaneously to guarantee all possible faults being detectable and isolable. A case study of a fuel-cell system shows that the proposed method performs well when the numbers of faults and sensors are moderate.