A Reasoning Architecture for Expert Troubleshooting of Complex Processes

Abdul Naveed, Jiaming Li, Bhaskar Saha, Abhinav Saxena, and George J. Vachtsevanos
Submission Type: 
Full Paper
phmc_12_049.pdf3.44 MBSeptember 24, 2012 - 11:35am

Complex military and industrial systems (machines, aircraft, etc.) experience fault/failure modes that must be diagnosed accurately and rapidly in order to sustain the operational availability of these systems as high as possible. Considerable downtime translates into loss of productivity and increased maintenance costs. In a manufacturing or industrial setting, problems reported by the machine’s internal checks (fault indicator) or an operator alert the maintainer of possible problem areas that must be addressed. Unfortunately, in these situations, the alert or advisory does not offer sufficient information to the maintainer that may permit the accurate and rapid diagnosis of the problem.
This paper introduces a novel reasoning methodology, in combination with appropriate models and measurements (data) to perform accurately and expeditiously expert troubleshooting for complex military and industrial processes. This automated troubleshooting tool is designed to support the maintainer/ repairman by identifying and locating faulty system components. The enabling technologies build upon a Model Based Reasoning paradigm and a Dynamic Case Based Reasoning method acting as the intelligent database. A case study employs a helicopter Intermediate Gearbox as the application domain to illustrate the efficacy of the approach.

Publication Year: 
Publication Volume: 
Publication Control Number: 
Page Count: 
Submission Keywords: 
model based reasoning
case based reasoning
dynamic case based reasoning
Submission Topic Areas: 
CBM and informed logistics
Health management system design and engineering
Model-based methods for fault detection, diagnostics, and prognosis
Submitted by: 

follow us

PHM Society on Facebook Follow PHM Society on Twitter PHM Society on LinkedIn PHM Society RSS News Feed