Current approaches to plant fault detection require an overall process model. This paper argues that a hierarchical scheme is more efficient in which the lowest level concentrates on validating the signals transmitted by individual sensors. Signals are described in terms of standard time-series models. A fault is defined by the signal's deviation from its expected behaviour and various signal processing techniques are described which detect these aberrations.
Get full access to this article
View all access options for this article.
References
1.
AnyakoraS NLeesF P (1972). ‘Detection of instrument malfunction by the process operator’, Chemical Engineer264, 304–309.
2.
AnyakoraS NLeesF P (1973). ‘The detection of malfunction using a process control computer: simple techniques for instrument malfunction’, The Use of Digital Computers in Measurement, IEE Conference, York, September, 35–42.
3.
ÅströmK JWittenmarkB (1984). Computer controlled system: theory and designPrentice-Hall.
4.
BassevilleM (1988). ‘Detecting changes in signals and systems — a survey’, Automatica24(3), 309–326.
5.
BoxGJenkinsG (1976). Time series analysis: forcasting and controlHolden-Day.
6.
ChatfieldC (1984). The analysis of time series — an introduction, 3rd edition, Chapman and Hall.
7.
ChowE YWillskyA S (1984). ‘Analytical redundancy and the design of robust failure detection systems’, IEEE Transactions on Automatic ControlAC-29 (7), 603–614.
8.
ClarkR N (1978). ‘A simplified instrument failure detection scheme’, IEEE Transactions on Aerospace and Electronics SystemAES-14 (4), 558–568.
9.
ClarkR NSetzerW (1980). ‘Sensor fault detection in a system with random disturbances’, IEEE Transactions on Aerospace and Electronics SystemAES-16 (4), 468–473.
10.
ClarkR NFosthD CWaltonV M (1975). ‘Detecting instrument malfunctions in control systems’, IEEE Transactions on Automatic ControlAC-11 (4), 465–473.
11.
de SáD. (1988). ‘The evolution of the intelligent measurement’Measurement and Control21(5), 142–144.
12.
DeckertJ CDesaiM NDeystJ JWillskyA J (1977). ‘F8-DFBW sensor failure identification using analytical redundancy’, IEEE Transactions on Automatic ControlAC-22 (5), 795–803.
13.
DeckertJ CFisherJ LLaningD BRayA (1982). ‘A signal validation methodology for nuclear power plants’, Proceedings ACC, Arlington, Virginia, June, 113–120.
14.
FriedlanderB (1982). ‘A modified prefilter for some recursive parameter estimation algorithms’, IEEE Transactions on Automatic Control, AC-27 (1), 232–235.
15.
El MadboulyEFrankP M (1983). ‘Robust instrument failure detection via Luenberger observers in nuclear power plants’, IFAC Symposium, Florence.
16.
FaveneeJ M (1987). ‘Smart sensors in industry’, Journal of Physics E: Science Instruments20(10), 1087–1090.
17.
GertlerJChangH-S (1986). ‘An instability indicator for expert control’, IEEE Control Systems Magazine August, 14–17.
18.
HalmeASelkäinaho (1984). ‘Instrument fault detection using an adaptive filtering method,’IFAC 9th World Congress Reprints, 3, Budapest, Hungary, 167–172.
19.
IsermannR (1984). ‘Process fault detection based on modelling and estimation methods’, Automatica20(4), 387–404.
20.
LeesF P (1973). ‘Some data on the failure modes of instruments in the chemical plant environment’, Chemical Engineer277, 418–421.
21.
LjungLSöderströmT (1983). Theory and practice of recursive identificationMIT Press.
22.
MassoumniaM-A (1986). ‘A geometric approach to the synthesis of failure detection filters’, IEEE Transactions on Automatic ControlAC-31 (9), 839–846.
23.
MehraR KPeschonJ (1971). ‘An innovations approach to fault detection and diagnosis’, Automatica7(5), 637–640.
24.
MulhollandHJonesC R (1983). Fundamentals of statisticsELBS and Butterworths.
25.
RayADesaiMDeystJ J (1983). ‘Fault detection and isolation in a nuclear reactor’, Journal of Energy, 7 (1), 79–85.
26.
UpadhyayaB R (1985). ‘Sensor failure detection and estimation’, Nuclear Safety26(1), 32–43.
27.
UpadhyayaB RKerlinT W (1978). ‘Estimation of response time characteristics of platinum resistance thermometers by noise analysis technique’, ISA Transactions, 17 (4), 21–38.
28.
UpadhyayaB RKitamuraM (1981). ‘Stability monitoring of boiling water reactors by time series analysis of neutron noise’, Nuclear Science and Engineering77, 480–492.
29.
UpadhyayaB RSkorskaM (1984). ‘Sensor fault analysis using decision theory and data-driven modeling of pressurized water reactor subsystems’, Nuclear Technology64(1), 70–77.
30.
VostrýZ (1975). ‘New algorithm for polynomial spectral factorization with quadratic convergence I’, Kybernetika11(6), 415–422.
31.
WillskyA S (1976). ‘A survey of design methods for fault detection in dynamics systems’, Automatica12(6), 601–611.
32.
WillskyA SJonesH L (1976). ‘A generalised likelihood ratio approach to detection and estimation of jumps in linear systems’, IEEE Transactions on Automatic ControlAC-21, February, 108–112.
33.
UsoroP BSchickI CNegahdaripourS (1985). ‘An innovation-based methodology for HVAC system fault detection’, Journal for Dynamic Systems, Measurement and Control, Transactions ASME, 107 (4), 284–289.
34.
WoodG G (1988) ‘International standards emerging for Fieldbus’, Control Engineering2, October, 22–25.
