Sensor selection in neuro-fuzzy modelling for fault diagnosis

Zhou, Yimin and Zolotas, A (2010) Sensor selection in neuro-fuzzy modelling for fault diagnosis. In: Industrial Electronics (ISIE), 2010 IEEE International Symposium on, 2010, Italy.

Full text not available from this repository.

Abstract

In this paper, sensor selection relating to neuro-fuzzy modeling for the purpose of fault diagnosis is discussed. The input/output selection in fuzzy modelling plays an important role in the performance of the derived model. In addition, with respect to fault tolerant issues, the impact of the faults on the system, i.e. possible incipient and abrupt faults, should be detected in the earliest possible instance. The paper first presents a brief introduction to neuro-fuzzy modelling, and proceeds to sensor selection with the aim of considerably improving the quality and reliability of the system. We study faults, both of abrupt and incipient nature, that can be diagnosed in an immediate sense. A two-tank system is used as an example to illustrate the studied concepts

Item Type: Conference or Workshop Item (Paper)
Keywords: fault diagnosis;fault tolerant issues;input selection;neuro fuzzy modelling;output selection;sensor selection;two tank system;fault diagnosis;fuzzy control;neurocontrollers;sensors;
Schools and Departments: School of Engineering and Informatics > Engineering and Design
Subjects: T Technology
T Technology > TJ Mechanical engineering and machinery > TJ0212 Control engineering systems. Automatic machinery (General)
Related URLs:
Depositing User: Argyrios Zolotas
Date Deposited: 20 Nov 2012 20:52
Last Modified: 20 Nov 2012 20:52
URI: http://sro.sussex.ac.uk/id/eprint/42756
📧 Request an update