Ž.
Computers in Industry 43 2000 61–72
www.elsevier.nlrlocatercompind
Fault diagnosis using Rough Sets Theory
Lixiang Shen
a
, Francis E.H. Tay
a,)
, Liangsheng Qu
b
, Yudi Shen
b
a
Department of Mechanical and Production Engineering, National UniÕersity of Singapore, 10 Kent Ridge Crescent, 119260, Singapore
b
Institute of Diagnostics and Cybernetics, School of Mechanical Engineering, Xian Jiaotong UniÕersity,
Shaanxi, 710049, People’s Republic of China
Received 25 February 1999; accepted 29 January 2000
Abstract
The fault diagnosis on diesel engine is a difficult problem due to the complex structure of the engine and the presence of
multi-excite sources. Usually, one method of fault diagnosis can only inspect one corresponding fault category. In this paper,
a new method, Rough Sets Theory, is used to diagnose the valve fault for a multi-cylinder diesel engine. Through the
analysis of the final reducts generated using Rough Sets Theory, it is shown that this new method is effective for valve fault
diagnosis. Rough Sets analysis requires discretizing the fault condition attributes. However, in practice, some of the limits of
these attributes are unknown. A new discretization method has been created and the method is found to be suitable for
discretizing the attributes without a priori knowledge. q 2000 Elsevier Science B.V. All rights reserved.
Keywords: Fault diagnosis; Rough Sets Theory; Discretization
1. Introduction
Fault diagnosis on machinery has been well re-
wx
searched 1 . There are many effective methods that
are used to diagnose accurately and quickly a certain
wx
category of faults 2 . For instance, large-scale cen-
trifugal compressor can be diagnosed by holospec-
wx
trum technique 3 . However, up to now, it is diffi-
cult to diagnose more than one category of faults.
This is especially so in diagnosing the dynamic
characteristics of reciprocating machinery, such as
reciprocating compressor and diesel engine. This is
due to the complex structure of the reciprocating
machinery. Although many methods can be used to
determine specific fault category, such as broken
)
Corresponding author. Tel.: q65-874-6818; fax: q65-779-
1459.
Ž.
E-mail address: mpetayeh@nus.edu.sg F.E.H. Tay .
wx
valve and cracked crankshaft 4 , the results obtained
from such fault specific method are not easy to
interpret. There is a need to have a method that can
diagnose more than one category of faults in a
generic manner. In this paper, a method based on
Rough Sets Theory is proposed and implemented.
Ž.
Z. Pawlak Poland first proposed Rough Sets
Theory in 1982. This theory has been developed and
used in many domains, such as medical diagnosis
wx wx
5 , stock market forecast 6 , fault diagnosis in
wx
engineering domain 7 , decision making for bank
wx w x
manager 8 and some other uses 21 .
The advantage of Rough Sets Theory is that it
needs neither additional information about the data
nor is it necessary to correct the inconsistencies
manifested in data. Instead, rules generated are cate-
gorized into certain rules or possible rules.
wx
In this paper, the Rough Sets Theory 9 is used to
analyze the decision table composed of attributes
0166-3615r00r$ - see front matter q2000 Elsevier Science B.V. All rights reserved.
Ž.
PII: S0166-3615 00 00050-6