Back to search

University of British Columbia

Distinguishing sensor and system faults for diagnostics and monitoring

Abstract

dc:description

Automated FDD systems depend entirely on reliability of sensor readings, since they are the monitoring interface of the system. With an unexpected variation in a sensor’s reading from its anticipated values, the challenge is to determine if it is symptom of a fault in the sensor or the monitored system. The ability to identify the source of faults is crucial in the monitoring of a system, as different corrective actions are required in case of sensor or system faults. To address this issue, first, it is clarified that by strict duplication of sensor elements, it is feasible to differentiate between sensor and system faults. However, duplication is not always practical. Hence, by aiming to identify the minimum degree of sensor redundancy, a priori knowledge of physical relationships (functional redundancy) between monitored variables is used to check the credibility of existing sensor observations via Analytical Computational Substitutions (ACS). In the proposed methodology for a certain class of systems, the system variables are modeled with serially connected causal network. Then the concept of Moving Monitoring Window (MMW) is introduced, which covers three nodes at the same time, as it traverses through the nodes of the system in the direction of causality. The Logic Set Unit consists of all system/sensor state possibilities called System Behavioral Modes, which allows decision-making on the health status of sensor or system or a combination. The generalization by deduction reveals that if the number of sensors is greater than 1.5 times of the number of monitored variables, the task of distinguishing between sensor and system faults can be done, as long as serial causality is valid between the monitored variables. Removing any more sensors from this configuration leads to inability to locate the faults, due to the lack of adequate behavioral modes for diagnosis decision. The effectiveness of the approach is verified on a system of interconnected multi reservoirs and control valves.

Degree

thesis:*
Name thesis:degree_name
Master of Applied Science - MASc
Level thesis:degree_level
master's
Discipline thesis:degree_discipline
Mechanical Engineering
Grantor dc:publisher
University of British Columbia
Year dc:date
2015

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • Taiebat, Morteza

Rights

dc:rights
Statement dc:rights
  • Attribution-NonCommercial-NoDerivs 2.5 Canada
Language dc:language
eng

Identifiers

dc:identifier.*
Handle dc:identifier
http://hdl.handle.net/2429/54763

Chain of custody

source
Harvested from
University of British Columbia
Base URL
circle.library.ubc.ca/oai/request
Last updated
2026-11-19
Source record
OAI-PMH GetRecord
related terms
citation

Taiebat, Morteza. Distinguishing sensor and system faults for diagnostics and monitoring. master's thesis, University of British Columbia, 2015. http://hdl.handle.net/2429/54763