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

Abstract

Abstract:The problem of optimal sensor placement for FDI consists in determining the set of sensors that minimizes a pre-defined cost function satisfying at the same time a pre-established set of FDI specifications for a given set of faults. This paper recalls three model-based optimal sensor location approaches: an Incremental search, a Heuristic search and a Binary Integer Linear Programming (BILP) formulation. The main contribution of this paper is a comparative study that addresses efficiency, flexibility and other issues. The performance of the approaches is demonstrated by an application to a fuel cell stack system.

Robustness Analysis of Sensor Placement for Leak Detection and Location under Uncertain Operating Conditions

“…Usually, the sensor placement problem is presented as an optimization problem where the cheaper sensor configuration fulfilling some given diagnosis specifications is sought [9,10]. Nevertheless, a baseline budget is usually assigned to instrumentation by water distribution companies which constraints the cost of the sought sensor configuration.…”

Sensor development and optimization for a proton exchange membrane fuel cell system in automotive applications

“…The model-based approaches require a good understanding of the system and it is really difficult to model the system with very high accuracy. With respect to PEMFC applications, Sarrate et al [132] reviewed three different model-based methods to optimize sensor placement for fault detection and isolation (FDI) purpose. 8 candidate sensors were used and the heuristic approach was the most efficient as the solution was found in 0.12 s. The two optimal sensors were cathode and anode pressure sensors.…”

“…In Ref. [132], cost of sensor was the optimizing index and was assigned according to the ease of installation and the price of corresponding sensor.…”

Clustering techniques applied to sensor placement for leak detection and location in water distribution networks

“…Usually, the sensor placement problem is presented as an optimization problem where the cheaper sensor configuration fulfilling some given diagnosis specifications is sought [12], [13]. Nevertheless, a baseline budget is usually assigned to instrumentation by water distribution companies which constraints the cost of the sought sensor configuration.…”