Welcome to Scribd, the world's digital library. Read, publish, and share books and documents. See more
Download
Standard view
Full view
of .
Save to My Library
Look up keyword
Like this
6Activity
0 of .
Results for:
No results containing your search query
P. 1
Time Domain Analysis Based Fault Diagnosis Methodology for Analog Circuits - A Comparative Study of Fuzzy and Neural Classifier Performance

Time Domain Analysis Based Fault Diagnosis Methodology for Analog Circuits - A Comparative Study of Fuzzy and Neural Classifier Performance

Ratings: (0)|Views: 142 |Likes:
Published by ijcsis
In this paper, we attempt to diagnose the occurrence of faults in analog electronic circuits based upon variations in time domain specifications corresponding to the circuit condition under consideration relative to the fault free circuit. To achieve this, both a fuzzy as well as a neural classifier have been utilized to operate with the fault dictionary data as base. Through this process, a general comparison is drawn out between the performance of either route in dealing with fault diagnosis of circuits. An illustrative example is considered, on which both the fuzzy and neural algorithms are tested, and their performance in fault diagnosis is compared. Further, the suitability of the fuzzy and neural techniques to various kinds of diagnosis problems depending upon the nature of data available is also discussed.
In this paper, we attempt to diagnose the occurrence of faults in analog electronic circuits based upon variations in time domain specifications corresponding to the circuit condition under consideration relative to the fault free circuit. To achieve this, both a fuzzy as well as a neural classifier have been utilized to operate with the fault dictionary data as base. Through this process, a general comparison is drawn out between the performance of either route in dealing with fault diagnosis of circuits. An illustrative example is considered, on which both the fuzzy and neural algorithms are tested, and their performance in fault diagnosis is compared. Further, the suitability of the fuzzy and neural techniques to various kinds of diagnosis problems depending upon the nature of data available is also discussed.

More info:

Published by: ijcsis on Jun 12, 2010
Copyright:Attribution Non-commercial

Availability:

Read on Scribd mobile: iPhone, iPad and Android.
download as PDF, TXT or read online from Scribd
See more
See less

08/25/2013

pdf

text

original

 
(IJCSIS) International Journal of Computer Science and Information Security,Vol. 8, No. 2, 2010
Time Domain Analysis based Fault DiagnosisMethodology for Analog Circuits-A ComparativeStudy of Fuzzy and Neural Classifier Performance
V. Prasannamoorthy
1
, R. Bharat Ram
2
, V. Manikandan
3
, N. Devarajan
4
 
1,2,4
 Department of Electrical Engineering, Government College of TechnologyCoimbatore, India
3
 Department of Electrical Engineering, Coimbatore Institute of TechnologyCoimbatore, India
 
1
prasanna_gct1995@yahoomail.com
 Abstract
 — In this paper, we attempt to diagnose the occurrenceof faults in analog electronic circuits based upon variations intime domain specifications corresponding to the circuit conditionunder consideration relative to the fault free circuit. To achievethis, both a fuzzy as well as a neural classifier have been utilizedto operate with the fault dictionary data as base. Through thisprocess, a general comparison is drawn out between theperformance of either route in dealing with fault diagnosis of circuits. An illustrative example is considered, on which both thefuzzy and neural algorithms are tested, and their performance infault diagnosis is compared. Further, the suitability of the fuzzyand neural techniques to various kinds of diagnosis problemsdepending upon the nature of data available is also discussed.
 Keywords—Fault diagnosis, fuzzy logic system, neural  networks,Sallen-key Bandpass filter.
I.
 
I
 NTRODUCTION
 The identification of faults in any analog circuit is veryuseful and, in a few instances, an inevitable measure inensuring competent performance of the circuit. In general, theanalog diagnosis approaches can be categorized into two [1],namely-simulation-after-test (SAT) and simulation-before-test(SBT). The simulation-after-test [2]-[4] approach involves thecomputation of various circuit parameters from the operationalcircuit and fault identification is carried out using these parameters, assuming that each measurement is independent of the other. This method is avoided due to the increase in processtime with increase in the size of the circuit, in addition to issuesconcerning non-linear circuits. On the other hand, a usefulalternative is found in the simulation-before-test approachwhich appreciably reduces the time taken for fault diagnosis.In the SBT approach [5]-[9], a predefined set of test stimuli areused to extract certain signatures from the Circuit-Under-Test(CUT) that are unique to each faulty condition. Thesesignatures can then be suitably systematized to create a “faultdictionary”, which is then checked for redundancies that mayresult in masking of certain faults. Evidently, the parameterschosen to pose as signatures must be quantities that areobservable for all conditions of the circuit.Both the above-mentioned approaches are fairly proceduralin nature and do not necessitate the prerequisite of anintuitional knowledge of the functioning of the CUT. Constantsupervision of the circuit is entailed to ensure stable performance over an extended period of time. Theidentification of faults in systems is often a combination of fault detection and isolation, necessarily in the same order,which is commonly known as FDI [10]-[11]. Early detection of faults in a circuit can greatly assist in maintenance of thesystem by avoiding possibly harmful damage borne out of thefault. Occasionally, a circuit may so damaged that it mightassume an unstable state, making it impossible to extractsignatures from it that might help in identifying the fault. Inother cases, a fault might just be too critical or dangerous to be provoked for the sake of obtaining a signature.Analog fault diagnosis is inherently complicated by poor mathematical models, component tolerances, nonlinear  behaviour of components, and limited accessibility to internalnodes of the circuit under test. In this paper, we state the resultsof a comparative study of the performance of fuzzy and neuralroutes in the detection and identification of faults in an analogelectronic circuit using the Simulation-Before-Test approach.This was achieved by taking into consideration the variations intime-domain response parameters pertaining to the transient of the CUT for a step input. A comprehensive fault dictionary was prepared from all the possible values of the parameterscorresponding to each state of the circuit, which was theneffectively utilized to construct a classifier capable of identifying the various faulty configurations of the CUT.II.
 
G
ENERALIZED
A
LGORITHM
 The fault diagnosis methodology, involving either a fuzzyor a neural system, may be divided into five distinct steps asfollows:Step I: Formulation of transfer function of the circuit under test assuming nominal values of all components in the circuit.Step II: Simulation of time-domain response of the circuitwhen a unit step signal is given as input, for possiblecombinations of component faults.Step III: Determination of time-domain response parameters,namely, settling time and peak amplitude for each timeresponse plot.
306http://sites.google.com/site/ijcsis/ISSN 1947-5500
 
(IJCSIS) International Journal of Computer Science and Information Security,Vol. 8, No. 2, 2010
Step IV: Suitable pre-processing of available data and designof classifier.Step V: Isolation of faults by using the signatures extractedfrom a suspicious circuit by feeding its time response parameters to the classifier obtained in step IV.III.
 
T
EST
C
IRCUIT
 To test the performance of the fuzzy and neural techniques,we chose the Sallen-Key bandpass filter circuit [12] shown inFig. 1. The Sallen-Key bandpass filter is a second order active filter, which is greatly appreciated for its simplicityof design. The filter section shown in Fig. 1 can also becascaded to form second order filter stages, resulting in larger order filters. The op-amp provides buffering between filter stages, so that each stage can be designed independently. Thiscircuit is suitable for filters which have complex conjugate poles. When implementing a particular transfer function, adesigner will typically find all the poles, and group them intoreal poles and complex conjugate pairs. Each of the complexconjugate pole pairs are then implemented with a Sallen-Keyfilter, and the circuits are cascaded together to form thecomplete filter.
Fig. 1 Sallen-Key Bandpass Filter circuit
The transfer function of the Sallen-Key bandpass filter circuitiswhereThe circuit shown in Fig. 1 has the component values thatcorrespond to the nominal frequency of operation of 25 kHz.IV.
 
S
IGNATURE
E
XTRACTION
 In order to obtain the fault signatures for each faultcondition, the values of components are varied to +50% or -50% of their nominal values shown in Fig. 1 and the circuit isexcited using a unit step signal as input, thus enabling toconstruct the fault dictionary. While the values of thesupposedly faulty components are manipulated, it is ensuredthat the remaining components maintain their nominal valuesas in the original circuit. Distinct variations in response may be seen for every faulty configuration, as shown in Fig. 2, 3and 4. The two required time-domain response parameters,settling time and peak amplitude are noted for all the responsecurves. These, being characteristic to a particular configuration, are used as the fault signatures.
Fig. 2 Step response curves for single faultsFig. 3 Step response curves for double faultsFig. 4 Step response curves for multiple faults
0123
 
4
 
5
 
67x
-5
00.511.5
 
STEP RESPONSE
 
TIME (s)
 
F35F36F37F38F39F4000.511.52
 
2.5
 
3
 
3.54x
-5
00.511.5
 
STEP RESPONSE
 
TIME (s)
 
F1F2F3F4F5F6F7F8F9F10F11F12F13F1400.511.5
 
2
 
2.5
 
3
 
3.544.55x 10
-
00.20.40.60.811.2
 
STEP RESPONSE
 
TIME (s)
 
F15F16F17F18F19F20F21F22AMPLITUDEAMPLITUDEAMPLITUDE
307http://sites.google.com/site/ijcsis/ISSN 1947-5500
 
(IJCSIS) International Journal of Computer Science and Information Security,Vol. 8, No. 2, 2010
There are a total 2n number of single faults for which thestep response plot in shown in Fig. 2, whereas there are n(n-1)/2 number of double faults, n(n-1)/3 triple faults and n(n-1)/4 quadruple faults for which the step response curvesare shown in Fig. 3 and Fig. 4. Other combinations of component faults get eliminated automatically due torepetitions of the same in the fault dictionary, the formulaestated above holding good in determining the final count of fault conditions post elimination.
TABLE ITIME DOMAIN SPECIFICATIONS FOR SINGLE FAULTS
Fault IDFaultycomponentsSettling time*10^-5 secPeak Amplitude
F1R1
 4.3834 0.7211F2R1
 4.2557 1.1709F3R2
 5.2859 0.8600F4R2
 3.8230 0.9788F5R3
 5.1273 1.1024F6R3
 3.6507 0.5770F7R4
 4.7622 0.7089F8R4
 3.0360 1.5381F9R5
 3.6338 1.1908F10R5
 4.9734 0.6228F11C1
 6.1093
 
1.0152F12C1
 2.6625
 
0.6645F13C2
 4.7794
 
0.7566F14C2
 3.9499
 
1.1012
 TABLE IITIME DOMAIN SPECIFICATIONS FOR DOUBLE FAULTS
Fault IDFaultycomponentsSettling time*10^-5 secPeak Amplitude
F15R1
,R2
 5.7921 0.6968F16R1
,R2
 3.0089 1.2654F17R1
,R3
 4.6712 0.9312F18R1
,R3
 2.9020 0.8483F19R1
,C1
 6.0179 0.8436F20R1
,C1
 2.3671 0.9403F21R2
,R5
 4.8269 1.1242F22R2
,R5
 3.2166 0.6416F23R3
,R5
 4.1115 1.5028F24R3
,R5
 3.9499 0.413F25R3
,C1
 7.4602 1.2456F26R3
,C1
 2.4867 0.4152F27R3
,C2
 5.4507 0.9526F28R3
,C2
 3.0668 0.7383F29R5
,C1
 5.1273 1.3780F30R5
,C1
 3.0089 0.4745F31R5
,C2
 4.1115 1.0019F32R5
,C2
 4.5108 0.7619F33R2
,C2
 5.8528 0.7379F34R2
,C2
 1.993 1.2744
From the above shown curves, the values of settling timeand peak value for each fault condition are noted down andaccordingly, fault dictionaries are created. A general rule of thumb is that once the fault signatures have been collected andorganized into a fault dictionary, the data must be optimized by eliminating signatures that bring about masking of faultswhose signatures match. However, in the case of a faultdictionary made up of time-domain response parameters, thedata is found to be free of redundancies hence making itfeasible to identify each faulty configuration on the basis of variations in two parameters alone.
TABLE IIITIME DOMAIN SPECIFICATIONS FOR TRIPLE FAULTS
Fault IDFaultycomponentsSettlingtime*10^-5 secPeak Amplitude
F35R1
,R2
,C1
 7.9128 0.8071F36R1
,R2
,C1
 1.6563 0.9866F37R3
,R4
,C2
 6.1429 0.7571F38R3
,R4
,C2
 2.4867 1.2456F39R1
,R2
,C2
 6.6527 0.586F40R1
,R2
,C2
 2.9279 1.5191F41R1
,R3
,C1
 6.5752 1.0816
F42R1
,R3
,C1
 1.7798 0.6463F43R2
,R3
,C1
 9.2037 1.1879F44R2
,R3
,C1
 1.6083 0.4277F45R1
,C1
,C2
 6.5752 0.7211F46R1
,C1
,C2
 2.1278 1.1709F47R2
,R5
,C1
 6.7579 1.2896F48R2
,R5
,C1
 1.9938 0.4779TABLE IVTIME DOMAIN SPECIFICATIONS FOR QUADRUPLE FAULTS
Fault IDFaultycomponentsSettlingtime*10^-5 secPeak Amplitude
F49R1
,R2
,R3
,R5
 5.4507 1.1908F50R1
,R2
,R3
,R5
 2.4867 0.6228F51R1
,R2
,R3
,C1
 9.1640 1.0152F52R1
,R2
,R3
,C1
 1.3313 0.6645F53R1
,R4
,C1
,C2
 7.4992 0.5693F54R1
,R4
,C1
,C2
 1.7798 1.9389F55R2
,R4
,C1
,C2
 8.362 0.6978F56R2
,R4
,C1
,C2
 3.3142 2.0385F57R4
,R5
,C1
,C2
 6.487 0.888F58R4
,R5
,C1
,C2
 2.1623 0.888F59 R1R2R3R4R5C1C9.7305 0.888F60 R1R2R3R4R5C1C1.0812 0.888F61 Fault Free 4.3247 0.888
The fault dictionaries for single component fault,double component fault, triple component fault and other multiple component faults are presented in the tables I, II, IIIand IV respectively. An upward arrow indicates a deviation of 50% above the nominal value, whereas a downward arrowindicates a 50% decrement. The fault free condition is alsotabulated and is given as ID F61.
308http://sites.google.com/site/ijcsis/ISSN 1947-5500

Activity (6)

You've already reviewed this. Edit your review.
1 thousand reads
1 hundred reads
dareg79 liked this
dareg79 liked this
Chaitanya Knv liked this

You're Reading a Free Preview

Download
/*********** DO NOT ALTER ANYTHING BELOW THIS LINE ! ************/ var s_code=s.t();if(s_code)document.write(s_code)//-->