A Knowledge Transfer Platform for Fault Diagnosis of Industrial Gas Turbines

Zhang, Yu and Jombo, Gbanaibolou and Latimer, Anthony (2018) A Knowledge Transfer Platform for Fault Diagnosis of Industrial Gas Turbines. In: 22nd IEEE International Conference on Intelligent Engineering Systems, 21-23 June, 2018, Gran Canaria, Spain.

Documents
A Knowledge Transfer Platform for Fault Diagnosis of Industrial Gas Turbines
[img]
[Download]
[img]
Preview
PDF
PID5379491.pdf - Whole Document

2MB
Item Type:Conference or Workshop contribution (Presentation)
Item Status:Live Archive

Abstract

The aim of this paper is to introduce the bases of an intelligent fault diagnostic platform, which assists in detecting mechanical failures of Industrial Gas Turbines (IGTs). This comprises an integration of an expert system and its complementary signal processing techniques. The essential characteristic here is not to exclude humans (experts) from the diagnostic process, but rather to transfer their knowledge and experience to a computerized platform. The automated process executed by the computerized platform is to ensure the scalability and consistency in fault diagnosis; while the humans are required to corroborate the transparency and liability of the outcomes. In this paper, a Knowledge Transfer Platform (KTP) is proposed for fault diagnosis of industrial systems. It is then designed and tested for combustion fault diagnosis using field data of IGTs. The preliminary results have revealed the feasibility and efficacy of the proposed scheme, which has the potential to be further extended to a large industrial scale and to different engineering diagnostic applications.

Keywords:knowledge transfer, fault diagnosis, industrial gas turbine, expert system, signal processing
Subjects:G Mathematical and Computer Sciences > G700 Artificial Intelligence
J Technologies > J941 Office Machinery Maintenance
Divisions:College of Science > School of Engineering
ID Code:32432
Deposited On:20 Oct 2018 19:53

Repository Staff Only: item control page