Design of sensor location based on various fault diagnostic observability and reliability criteria

The success of any fault diagnosis technique depends critically on the sensors measuring the important process variables. To select the optimum number of sensors for a given process, qualitative (cause-effect), graph theoretic approaches, based on process digraph (DG) were proposed. Graph approaches that used the concepts of observability (detecting all faults) and resolution (identifying the exact fault) were discussed. In this article, besides briefly summarizing these approaches, a reliability formulation for selecting optimal sensors is presented. The formulation takes into account quantitative information such as fault occurrence probabilities, sensor failure probabilities, and sensor costs. The formulation can also handle various constraints. Heuristics to solve the posed problem are also discussed. Sensor location based on the proposed formulation is applied to a chemical process to illustrate the utility of the proposed approach. (C) 2000 Elsevier Science Ltd. All rights reserved.