Title: Model based fault diagnosis, fault tolerant control and reconfiguration of hydraulic and thermo-fluid processes using analytical redundancy

Authors: S.K. Ghoshal, A.K. Samantaray, S. Samanta

Addresses: Department of Mechanical Engineering and Mining Machinery Engineering, Indian School of Mines University, Dhanbad 826004, India. ' Department of Mechanical Engineering, Indian Institute of Technology, Kharagpur 721302, India. ' Department of Mechanical Engineering, Asansol Engineering College, Kanyapur, Sen Raleigh Road, Asansol 713304, India

Abstract: Analytical redundancy relation (ARR) based approach for fault detection and isolation (FDI) is presented in this article with applications to a hydraulic and a thermo-fluid process. Bond graph modelling, which allows unified representation of multi-energy domain system dynamics, is used to develop model based quantitative FDI schemes. Also, some issues of fault tolerant control and system reconfiguration have been taken up with model simulation. It is shown that ARRs can be systematically derived by exploring causality in bond graph models and fault indicators can be interactively designed. The sensitivity of fault indicators has been validated through simulation.

Keywords: analytical redundancy; bond graphs; fault accommodation; fault isolation; fault signature; residual; model based fault diagnosis; fault tolerant control; fault detection; system reconfiguration; hydraulics; thermofluids; modelling.

DOI: 10.1504/IJAAC.2009.026782

International Journal of Automation and Control, 2009 Vol.3 No.4, pp.363 - 384

Received: 12 Jul 2008
Accepted: 18 Nov 2008
Published online: 26 Jun 2009 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article