CN102520341A - Analog circuit fault diagnosis method based on Bayes-KFCM (Kernelized Fuzzy C-Means) algorithm - Google Patents
Analog circuit fault diagnosis method based on Bayes-KFCM (Kernelized Fuzzy C-Means) algorithm Download PDFInfo
- Publication number
- CN102520341A CN102520341A CN2011103965629A CN201110396562A CN102520341A CN 102520341 A CN102520341 A CN 102520341A CN 2011103965629 A CN2011103965629 A CN 2011103965629A CN 201110396562 A CN201110396562 A CN 201110396562A CN 102520341 A CN102520341 A CN 102520341A
- Authority
- CN
- China
- Prior art keywords
- test
- sample
- class
- fault
- bayes
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Landscapes
- Test And Diagnosis Of Digital Computers (AREA)
- Tests Of Electronic Circuits (AREA)
Abstract
The invention discloses an analog circuit fault diagnosis method based on a Bayes-KFCM (Kernelized Fuzzy C-Means) algorithm, which comprises the following steps of: carrying out fault diagnosis by adopting a kernelized fuzzy C-means clustering algorithm and firstly judging whether a new fault exits in a test sample, if YES, a diagnostic model of a new fault sample is trained to join a diagnosis system, or else, the fault positioning is carried out on the test sample according to a Bayes fault classification standard. In the invention, the wavelet transform pretreatment is carried out on the fault sample, and the multi-feature fusion is carried out on the wavelet coefficient energy value and the wavelet coefficient fractal dimension value of the sample to extract fault characteristics; and the frequency of an optimal measurable node and/or a test signal is selected through taking a maximum class inter-class distance as a basis. Compared with the prior art, the analog circuit fault diagnosis method realizes that the analog circuit fault diagnosis method, the new fault of an analog circuit can be effectively diagnosed, and the diagnosis accuracy can be improved.
Description
Technical field
The present invention relates to a kind of analog-circuit fault diagnosis method, more particularly to a kind of analog-circuit fault diagnosis method based on Bayes-KFCM algorithms.
Background technology
Functional test of the test and diagnosis of analog circuit mainly for circuit, divide to be divided into before survey in the order of test process according to emulation at present and emulate and survey post-simulation, shown according to Research Literature in recent years and Patent data, intelligent failure diagnosis method is the focus and emphasis of present research, this method belongs to emulation mode before survey, the problems such as intelligent diagnosing method can partly solve the ambiguity of analog circuit fault, uncertainty.Most common and wide variety of intelligent diagnosing method is neural network, and the analog-circuit fault diagnosis method based on SVM obtained large development in recent years, and an important branch is obtained as intelligent diagnosing method.However, both traditional intelligence methods belong to necessary known fault sample existing category class label during supervised learning algorithm, i.e. Training diagnosis model, but when a new failure occurs, both fault diagnosis algorithms can not effectively be diagnosed to be new failure.In addition, when new failure classes occur for circuit, the diagnostic system based on supervised learning algorithm needs all sample sets of re -training, and diagnosis efficiency is low.Therefore, it is a kind of suitable intelligent diagnosis algorithm based on the failure modes algorithm learnt without tutor while can also diagnose known fault class in order to effectively diagnose new failure.
In addition, the selection and extraction of fault signature are a key technologies in analog circuit fault diagnosing, because the component of analog circuit has tolerance feature, respond the output of circuit has randomness within the specific limits, and nonlinear characteristic is often presented in the output of circuit.Therefore the fault signature of the different faults of analog circuit there may be intersection or overlapping phenomenon, as fuzzy fault, it is impossible to be positioned by correct diagnosis.In order to solve the problems, such as in the identification of fuzzy fault feature, analog circuit fault diagnosing using signal transacting and dimension reduction method to the progress feature extraction of failure response signal and selection.Most common signal processing method is wavelet transformation, calculates the energy entropy of wavelet coefficient as fault signature, this method has been demonstrated to effectively extract fault signature.But the overlapping and crossover phenomenon of fuzzy fault feature can not be thoroughly solved only with single fault signature, if various features information never ipsilateral faults feature is selected, with certain message complementary sense.Therefore, realize that different characteristic fusion is the effective way of analog circuit fault feature extraction by information fusion technology.
The content of the invention
The technical problems to be solved by the invention are that the diagnosis efficiency for overcoming existing method for diagnosing faults to be brought using supervised learning algorithm is low, and the deficiency of efficient diagnosis can not be carried out to new failure, a kind of analog-circuit fault diagnosis method based on Bayes-KFCM algorithms is provided, the algorithm is without tutor's learning algorithm, with higher diagnosis efficiency, and efficient diagnosis can be carried out to new failure.
A kind of analog-circuit fault diagnosis method based on Bayes-KFCM algorithms, comprises the following steps:
Step A, selection treat that the optimal of lateral circuit surveys node and frequency test signal;
Step B, to circuit under test input test signal, simulate various typical malfunctions, gather the optimal voltage output value for surveying node, obtain fault data as training data;Identical test signal and it can survey under node, data of the collecting test circuit under actual working state are used as test data;
Step C, the feature for extracting fault data and test data respectively, and carry out denoising, generation training sample set and test sample collection;
Step D, it is trained using training sample set pair fault diagnosis model, and fault diagnosis is carried out to test sample collection using the fault diagnosis model trained;
The step D specifically includes following steps:
Step D1, using KFCM clustering algorithms fault diagnosis model is trained, is specially:Training sample set is mapped to by higher dimensional space by kernel function;Then by fuzzyCMean Method is clustered, when the ratio between the sample number that correctly clusters and all cluster sample numbers are more than or equal to a default threshold value, then algorithm stops, training terminates, it regard the Clustering Model trained as diagnostic model, the distance value of training sample maximum with such cluster centre distance in the cluster centre of all kinds of training samples, and every class training sample is obtained simultaneously, wherein,nFor the class number of training sample;
Step D2, test sample is mapped to higher dimensional space by kernel function, test sample is calculated in higher dimensional space to the distance of all kinds of cluster centres, wherein;WhenWhen, then test sample is new failure classes sample, and new failure classes sample is clustered using KFCM clustering algorithms, and new Clustering Model is added into diagnostic system;Otherwise fault location, wherein Bayes failure modes criterion such as following formula are carried out to test sample using Bayes failure modes criterion:
In formula,It isiThe number of training of class,It is all number of training,It is test samplexFromiThe distance of class training sample cluster centre,It is alliClass training sample and the sum average value of its cluster centre distance, Bayes failure modes represent test samplexBelong to maximumThe failure classes of value.
Further, the present invention in feature extraction using multiple faults Fusion Features and selection method, it is specific as follows to overcome the problem of single failure feature can not thoroughly solve the overlapping and crossover phenomenon of fuzzy fault feature:
Feature extraction described in step C is specific in accordance with the following methods:
Step C1, the magnitude of voltage progress multilevel wavelet decomposition by collection, resolve into detail coefficients and approximation coefficient;
Step C2, the energy entropy for calculating every layer of detail coefficients and approximation coefficient, vectorial first character representation as voltage signal being made up of multi-level Wavelet Transform coefficient energy entropy;
Step C3, the values of fractal dimension for calculating every layer of detail coefficients and approximation coefficient, vectorial second character representation as voltage signal being made up of multi-level Wavelet Transform coefficient values of fractal dimension;
Step C4, two kinds of features are merged using linear summation method, wherein linearly summation fusion formula is expressed as below:
In formula,The characteristic vector of fusion is represented,WithRepresent respectively and obtained energy entropy and values of fractal dimension calculated by one layer of coefficient of wavelet decomposition D of signal,WithWavelet coefficient energy value and wavelet coefficient values of fractal dimension weight shared in fusion are represented respectively, and;
Step C5, calculating fusion feature vectorIn each feature and remaining feature overall relevance size, and be ranked up from high to low, selection feature coherent integration contribution margin is more thanKBefore %hIndividual feature, carries out Feature Dimension Reduction,KSpan be(0,100);If sampleIn containPIndividual sample, the intrinsic dimensionality of each sample isW, then sampleCharacteristic correlation coefficient vector obtain in accordance with the following methods:
Then, sample is calculatedCovariance matrix, wherein by characterRepresent each element in covariance matrix:
Then, according to matrixCalculate correlation matrix:
Finally, each feature is calculated respectively to remainingWThe coefficient correlation sum of -1 feature:
Further, node is surveyed in order that optimal and/or the selection of frequency test signal is more representative, so as to improve examining property of failure, the present invention is also using the alternatively foundation of between class distance in maximum kind, node is surveyed to optimal and/or frequency test signal is selected, it is specific as follows:
The frequency test signal is selected by the following method:
Step A1, the amplitude-frequency response for obtaining circuit under test;
Flex point and its neighbouring frequency on step A2, selection amplitude-frequency response, are used as frequency sets to be selected;
Step A3, manual simulation some typical test failures, fault sample value is used as in output end collection circuit under test response magnitude of voltage of circuit under all frequency excitations to be selected of circuit, calculate between class distance in the class of different test failure class samples, and select between class distance in the class of test failure class it is maximum treat selected frequency as test frequency.
The optimal node of surveying is selected by the following method:
Step A4, it regard all measurable test nodes in circuit under test as test node to be selected, some typical test failures of manual simulation, circuit under test is loaded into using the test signal of selection as driving source, the magnitude of voltage for gathering all test failures on all test nodes to be selected is used as fault sample value;
Between class distance in the class of test failure class sample, is selected in the class of all failure classes before between class distance maximum in step A5, each test node to be selected of calculatingMIndividual test node,MFor the integer set in advance less than test node to be selected sum.
Compared with prior art, analog-circuit fault diagnosis method of the invention has the advantages that:
(1)Two kinds of features of Energy-Entropy and fractal dimension after wavelet transformation are extracted simultaneously, and Fusion Features are carried out by method of linearly summing, and the feature of fault-signal is extracted from a variety of sides, with certain complementarity, the one-sidedness of single features is compensate for.
(2)By calculating the correlation size of every kind of characteristic component in characteristic vector and being ranked up, before selection feature coherent integration contribution margin is more than 90%hIndividual feature, can effectively remove redundancy in characteristic vector and incoherent feature, the dimension of fault feature vector is reduced, the efficiency of diagnosis algorithm is lifted.
(3)Using based on the Fuzzy c-means cluster learnt without tutor(KFCM)Algorithm is as fault diagnosis algorithm, and overcoming traditional supervised learning algorithm needs the limitation of failure classes label, and energy efficient diagnosis goes out new failure classes.In addition, in face of there are new failure classes, traditional intelligent diagnosis system needs all sample sets of re -training, and diagnosis efficiency is low, need to only train the diagnostic model of new failure classes using without tutor's learning algorithm, and add in diagnostic system.
(4)A kind of failure modes criterion is constructed according to the characteristics of KFCM algorithms and according to Bayes decision theories, the criterion can judge fast and accurately and positioning is out of order.
Brief description of the drawings
Fig. 1 is the flow chart of the analog-circuit fault diagnosis method of the present invention;
Fig. 2 is multiple faults Fusion Features and the flow chart of system of selection in the inventive method.
Embodiment
Technical scheme is described in detail below in conjunction with the accompanying drawings:
The analog-circuit fault diagnosis method of the present invention, as shown in figure 1, comprising the following steps:
Step A, selection treat that the optimal of lateral circuit surveys node and frequency test signal.
Node is surveyed in order that optimal and/or the selection of frequency test signal is more representative, so as to improve examining property of failure, the present invention surveys node and/or frequency test signal is selected using the alternatively foundation of between class distance in maximum kind to optimal.Specifically, step A is specifically included:
Step A1, the amplitude-frequency response for obtaining circuit under test;
Flex point and its neighbouring frequency on step A2, selection amplitude-frequency response, are used as frequency sets to be selected;
Step A3, manual simulation some typical test failures, fault sample value is used as in output end collection circuit under test response magnitude of voltage of circuit under all frequency excitations to be selected of circuit, calculate between class distance in the class of different test failure class samples, and select between class distance in the class of test failure class it is maximum treat selected frequency as test frequency;Wherein, between class distance in classBe calculated as prior art, its computing formula is as follows:
Wherein
In formula, whereincFor classification number,n i ForiThe sample number of class,P i It isiThe prior probability of class sample,RespectivelyiThe characteristic vector of class,Represent theiThe mean vector of class sample set,mAll all kinds of sample set grand mean vectors are represented,Referred to as inter _ class relationship matrix,For within class scatter matrix, it is considered that inter _ class relationship is tried one's best greatly, and within-cluster variance is as far as possible small, be conducive to classification;
Step A4, it regard all measurable test nodes in circuit under test as test node to be selected, some typical test failures of manual simulation, circuit under test is loaded into using the test signal of selection as driving source, the magnitude of voltage for gathering all test failures on all test nodes to be selected is used as fault sample value;
Between class distance in the class of test failure class sample, is selected in the class of all failure classes before between class distance maximum in step A5, each test node to be selected of calculatingMIndividual test node,MFor the integer set in advance less than test node to be selected sum.
Step B, to circuit under test input test signal, simulate various typical malfunctions, gather the optimal voltage output value for surveying node, obtain fault data as training data;Identical test signal and it can survey under node, data of the collecting test circuit under actual working state are used as test data.
Step C, the feature for extracting fault data and test data respectively, and carry out denoising, generation training sample set and test sample collection.
The present invention in feature extraction using multiple faults Fusion Features and selection method, to overcome the problem of single failure feature can not thoroughly solve the overlapping and crossover phenomenon of fuzzy fault feature, specifically as shown in Fig. 2 including:
Step C1, the magnitude of voltage progress multilevel wavelet decomposition by collection, resolve into detail coefficients and approximation coefficient;
Step C2, the energy entropy for calculating every layer of detail coefficients and approximation coefficient, vectorial first character representation as voltage signal being made up of multi-level Wavelet Transform coefficient energy entropy;
Step C3, the values of fractal dimension for calculating every layer of detail coefficients and approximation coefficient, vectorial second character representation as voltage signal being made up of multi-level Wavelet Transform coefficient values of fractal dimension;
Step C4, two kinds of features are merged using linear summation method, wherein linearly summation fusion formula is expressed as below:
In formula,The characteristic vector of fusion is represented,WithRepresent respectively and obtained energy entropy and values of fractal dimension calculated by one layer of coefficient of wavelet decomposition D of signal,WithWavelet coefficient energy value and wavelet coefficient values of fractal dimension weight shared in fusion are represented respectively, and;In present embodiment, weightWithValue be 0.5;
Step C5, calculating fusion feature vectorIn each feature and remaining feature overall relevance size, and be ranked up from high to low, selection feature coherent integration contribution margin is more thanKBefore %hIndividual feature, carries out Feature Dimension Reduction,KSpan be(0,100), in present embodiment, before selection feature related progressive contribution margin is more than 90%hIndividual feature;If sampleIn containPIndividual sample, the intrinsic dimensionality of each sample isW, then sampleCharacteristic correlation coefficient vector obtain in accordance with the following methods:
Then, sample is calculatedCovariance matrix, wherein by characterRepresent each element in covariance matrix:
Finally, each feature is calculated respectively to remainingWThe coefficient correlation sum of -1 feature:
In present embodiment, denoising is carried out using traditional wavelet transformation soft threshold method.
Step D, it is trained using training sample set pair fault diagnosis model, and fault diagnosis is carried out to test sample collection using the fault diagnosis model trained.
The present invention proposes in the new algorithm that a kind of KFCM algorithms are combined with Bayes sorting criterions, the present invention and is called Bayes-KFCM algorithms.
KFCM is a kind of without tutor's learning algorithm, and original feature space data are mapped into higher dimensional space by kernel function, then clustered by fuzzy C-mean algorithm method.Assuming that having in training samplekClass failure, then cluster numbers bekCondition that is individual, being stopped using a certain default accuracy rate as KFCM clustering algorithms, wherein accuracy rate are calculated by below equation and obtained:
The object function of KFCM algorithms is:
In formulaIt is Subject Matrix,Represent degree of membership,It is Weighted Index,v j It is the cluster centre in the input space,cIt is the classification number of cluster,It is Nonlinear Mapping,KIt is kernel function.
According to Bayes optimizing decision principles, a new sample should be assigned in the classification with optimal posterior probability:
Wherein prior probability, classification correlation densityCan be by pseudo- density functionObtain,, therefore posterior probabilityIt is equal to:
WhereinTypically one constant, whereinIt isiThe number of training of class,It is all number of training,It is test samplexFromiThe distance at class fault cluster center,It is alliClass training sample and the sum average value of its cluster centre distance.So the above-mentioned sorting criterion based on Bayes optimizing decision principles, can be expressed as following form:
Specifically, step B is specifically included:
Step D1, using KFCM clustering algorithms fault diagnosis model is trained, is specially:Training sample set is mapped to by higher dimensional space by kernel function;Then by fuzzyCMean Method is clustered, when the ratio between the sample number that correctly clusters and all cluster sample numbers are more than or equal to a default threshold value, then algorithm stops, training terminates, it regard the Clustering Model trained as diagnostic model, the distance value of training sample maximum with such cluster centre distance in the cluster centre of all kinds of training samples, and every class training sample is obtained simultaneously, wherein,nFor the class number of training sample;
Step D2, test sample is mapped to higher dimensional space by kernel function, test sample is calculated in higher dimensional space to the distance of all kinds of cluster centres, wherein;WhenWhen, then test sample is new failure classes sample, and new failure classes sample is clustered using KFCM clustering algorithms, and new Clustering Model is added into diagnostic system;Otherwise fault location, wherein Bayes failure modes criterion such as following formula are carried out to test sample using Bayes failure modes criterion:
In formula,It isiThe number of training of class,It is all number of training,It is test samplexFromiThe distance of class training sample cluster centre,It is alliClass training sample and the sum average value of its cluster centre distance, Bayes failure modes represent test samplexBelong to maximumThe failure classes of value.
Claims (7)
1. a kind of analog-circuit fault diagnosis method based on Bayes-KFCM algorithms, comprises the following steps:
Step A, selection treat that the optimal of lateral circuit surveys node and frequency test signal;
Step B, to circuit under test input test signal, simulate various typical malfunctions, gather the optimal voltage output value for surveying node, obtain fault data as training data;Identical test signal and it can survey under node, data of the collecting test circuit under actual working state are used as test data;
Step C, the feature for extracting fault data and test data respectively, and carry out denoising, generation training sample set and test sample collection;
Step D, it is trained using training sample set pair fault diagnosis model, and fault diagnosis is carried out to test sample collection using the fault diagnosis model trained;
Characterized in that, the step D specifically includes following steps:
Step D1, using KFCM clustering algorithms fault diagnosis model is trained, is specially:Training sample set is mapped to by higher dimensional space by kernel function;Then by fuzzyCMean Method is clustered, when the ratio between the sample number that correctly clusters and all cluster sample numbers are more than or equal to a default threshold value, then algorithm stops, training terminates, it regard the Clustering Model trained as diagnostic model, the distance value of training sample maximum with such cluster centre distance in the cluster centre of all kinds of training samples, and every class training sample is obtained simultaneously, wherein,nFor the class number of training sample;
Step D2, test sample is mapped to higher dimensional space by kernel function, test sample is calculated in higher dimensional space to the distance of all kinds of cluster centres, wherein;WhenWhen, then test sample is new failure classes sample, and new failure classes sample is clustered using KFCM clustering algorithms, and new Clustering Model is added into diagnostic system;Otherwise fault location, wherein Bayes failure modes criterion such as following formula are carried out to test sample using Bayes failure modes criterion:
In formula,It isiThe number of training of class,It is all number of training,It is test samplexFromiThe distance of class training sample cluster centre,It is alliClass training sample and the sum average value of its cluster centre distance, Bayes failure modes represent test samplexBelong to maximumThe failure classes of value.
2. the analog-circuit fault diagnosis method as claimed in claim 1 based on Bayes-KFCM algorithms, it is characterised in that the value of threshold value described in step D1 is 90%.
3. the analog-circuit fault diagnosis method as claimed in claim 1 based on Bayes-KFCM algorithms, it is characterised in that feature extraction described in step C is specific in accordance with the following methods:
Step C1, the magnitude of voltage progress multilevel wavelet decomposition by collection, resolve into detail coefficients and approximation coefficient;
Step C2, the energy entropy for calculating every layer of detail coefficients and approximation coefficient, vectorial first character representation as voltage signal being made up of multi-level Wavelet Transform coefficient energy entropy;
Step C3, the values of fractal dimension for calculating every layer of detail coefficients and approximation coefficient, vectorial second character representation as voltage signal being made up of multi-level Wavelet Transform coefficient values of fractal dimension;
Step C4, two kinds of features are merged using linear summation method, wherein linearly summation fusion formula is expressed as below:
In formula,The characteristic vector of fusion is represented,WithRepresent respectively and obtained energy entropy and values of fractal dimension calculated by one layer of coefficient of wavelet decomposition D of signal,WithWavelet coefficient energy value and wavelet coefficient values of fractal dimension weight shared in fusion are represented respectively, and;
Step C5, calculating fusion feature vectorIn each feature and remaining feature overall relevance size, and be ranked up from high to low, selection feature coherent integration contribution margin is more thanKBefore %hIndividual feature, carries out Feature Dimension Reduction,KSpan be(0,100);If sampleIn containPIndividual sample, the intrinsic dimensionality of each sample isW, then sampleCharacteristic correlation coefficient vector obtain in accordance with the following methods:
Then, sample is calculatedCovariance matrix, wherein by characterRepresent each element in covariance matrix:
Finally, each feature is calculated respectively to remainingWThe coefficient correlation sum of -1 feature:
Wherein feature coherent integration contribution marginCalculated according to below equation:
5. the analog-circuit fault diagnosis method as claimed in claim 3 based on Bayes-KFCM algorithms, it is characterised in thatKValue be 90.
6. the analog-circuit fault diagnosis method based on Bayes-KFCM algorithms as described in claim any one of 1-5, it is characterised in that the frequency test signal is selected by the following method:
Step A1, the amplitude-frequency response for obtaining circuit under test;
Flex point and its neighbouring frequency on step A2, selection amplitude-frequency response, are used as frequency sets to be selected;
Step A3, manual simulation some typical test failures, fault sample value is used as in output end collection circuit under test response magnitude of voltage of circuit under all frequency excitations to be selected of circuit, calculate between class distance in the class of different test failure class samples, and select between class distance in the class of test failure class it is maximum treat selected frequency as test frequency.
7. the analog-circuit fault diagnosis method based on Bayes-KFCM algorithms as described in claim any one of 1-5, it is characterised in that the optimal node of surveying is selected by the following method:
Step A4, it regard all measurable test nodes in circuit under test as test node to be selected, some typical test failures of manual simulation, circuit under test is loaded into using the test signal of selection as driving source, the magnitude of voltage for gathering all test failures on all test nodes to be selected is used as fault sample value;
Between class distance in the class of test failure class sample, is selected in the class of all failure classes before between class distance maximum in step A5, each test node to be selected of calculatingMIndividual test node,MFor the integer set in advance less than test node to be selected sum.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN 201110396562 CN102520341B (en) | 2011-12-05 | 2011-12-05 | Analog circuit fault diagnosis method based on Bayes-KFCM (Kernelized Fuzzy C-Means) algorithm |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN 201110396562 CN102520341B (en) | 2011-12-05 | 2011-12-05 | Analog circuit fault diagnosis method based on Bayes-KFCM (Kernelized Fuzzy C-Means) algorithm |
Publications (2)
Publication Number | Publication Date |
---|---|
CN102520341A true CN102520341A (en) | 2012-06-27 |
CN102520341B CN102520341B (en) | 2013-12-18 |
Family
ID=46291327
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN 201110396562 Expired - Fee Related CN102520341B (en) | 2011-12-05 | 2011-12-05 | Analog circuit fault diagnosis method based on Bayes-KFCM (Kernelized Fuzzy C-Means) algorithm |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN102520341B (en) |
Cited By (22)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103198217A (en) * | 2013-03-26 | 2013-07-10 | X·Q·李 | Fault detection method and system |
CN104090227A (en) * | 2014-07-16 | 2014-10-08 | 电子科技大学 | Measuring point selection method in analog circuit fault diagnosis |
CN104502834A (en) * | 2014-12-30 | 2015-04-08 | 南京农业大学 | OFRF (output frequency response function) modeling based nonlinear analog circuit fault parameter identification method |
CN104573740A (en) * | 2014-12-22 | 2015-04-29 | 山东鲁能软件技术有限公司 | SVM classification model-based equipment fault diagnosing method |
CN104635146A (en) * | 2015-02-06 | 2015-05-20 | 南京农业大学 | Analog circuit fault diagnosis method based on random sinusoidal signal test and HMM (Hidden Markov Model) |
CN104849650A (en) * | 2015-05-19 | 2015-08-19 | 重庆大学 | Analog circuit fault diagnosis method based on improvement |
CN105044562A (en) * | 2015-08-26 | 2015-11-11 | 云南电网有限责任公司电力科学研究院 | Distribution network fault positioning method based on Bayes formula algorithm |
CN105652182A (en) * | 2015-12-28 | 2016-06-08 | 北京航天测控技术有限公司 | Circuit board fault positioning system and circuit board fault positioning method based on circuit network and graph search |
WO2017024692A1 (en) * | 2015-08-07 | 2017-02-16 | 合肥工业大学 | Method of simulating analog circuit diagnostic fault using single measurement node |
CN106444468A (en) * | 2016-05-03 | 2017-02-22 | 重庆青年职业技术学院 | Information machine adapter detection system and method |
CN106896306A (en) * | 2017-04-26 | 2017-06-27 | 国网上海市电力公司 | A kind of GIS oscillatory surges pressure test signal antinoise method |
CN107220670A (en) * | 2017-05-27 | 2017-09-29 | 重庆大学 | Supervised Artifical Taste system features extracting method is had based on wavelet transform |
CN107843779A (en) * | 2017-09-26 | 2018-03-27 | 国网河南省电力公司民权县供电公司 | A kind of Power System Fault Record classifying and analyzing method and system based on fuzzy clustering |
CN105699882B (en) * | 2016-01-22 | 2018-05-29 | 电子科技大学 | Analog-circuit fault diagnosis method based on vibration test technology |
CN108470462A (en) * | 2018-05-18 | 2018-08-31 | 上海会为智能技术有限公司 | A kind of signal lamp failure detection method, device, equipment and medium |
CN109726770A (en) * | 2019-02-19 | 2019-05-07 | 金陵科技学院 | A kind of analog circuit fault testing and diagnosing method |
CN109885018A (en) * | 2019-01-18 | 2019-06-14 | 广东寰球智能科技有限公司 | A kind of management method of the expert think tank system towards mental package equipment |
CN110579709A (en) * | 2019-08-30 | 2019-12-17 | 西南交通大学 | fault diagnosis method for proton exchange membrane fuel cell for tramcar |
CN111307453A (en) * | 2020-03-20 | 2020-06-19 | 朗斯顿科技(北京)有限公司 | Transmission system fault diagnosis method based on multi-information fusion |
CN111523594A (en) * | 2020-04-23 | 2020-08-11 | 湖州师范学院 | Improved KNN fault classification method based on LDA-KMEDOIDS |
CN113159088A (en) * | 2021-01-07 | 2021-07-23 | 中国地质大学(武汉) | Fault monitoring and diagnosis method based on multi-feature fusion and width learning |
CN113409284A (en) * | 2021-06-28 | 2021-09-17 | 北京百度网讯科技有限公司 | Circuit board fault detection method, device, equipment and storage medium |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101614787A (en) * | 2009-07-07 | 2009-12-30 | 南京航空航天大学 | Analogical Electronics method for diagnosing faults based on M-ary textural classification device |
CN101871994A (en) * | 2010-06-11 | 2010-10-27 | 南京航空航天大学 | Method for diagnosing faults of analog circuit of multi-fractional order information fusion |
CN101907681A (en) * | 2010-07-15 | 2010-12-08 | 南京航空航天大学 | Analog circuit dynamic online failure diagnosing method based on GSD-SVDD |
-
2011
- 2011-12-05 CN CN 201110396562 patent/CN102520341B/en not_active Expired - Fee Related
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101614787A (en) * | 2009-07-07 | 2009-12-30 | 南京航空航天大学 | Analogical Electronics method for diagnosing faults based on M-ary textural classification device |
CN101871994A (en) * | 2010-06-11 | 2010-10-27 | 南京航空航天大学 | Method for diagnosing faults of analog circuit of multi-fractional order information fusion |
CN101907681A (en) * | 2010-07-15 | 2010-12-08 | 南京航空航天大学 | Analog circuit dynamic online failure diagnosing method based on GSD-SVDD |
Cited By (37)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103198217B (en) * | 2013-03-26 | 2016-06-22 | X·Q·李 | A kind of fault detection method and system |
CN103198217A (en) * | 2013-03-26 | 2013-07-10 | X·Q·李 | Fault detection method and system |
CN104090227A (en) * | 2014-07-16 | 2014-10-08 | 电子科技大学 | Measuring point selection method in analog circuit fault diagnosis |
CN104090227B (en) * | 2014-07-16 | 2016-10-26 | 电子科技大学 | A kind of measuring point system of selection in analog circuit fault diagnosing |
CN104573740B (en) * | 2014-12-22 | 2018-07-03 | 山东鲁能软件技术有限公司 | A kind of equipment fault diagnosis method based on svm classifier model |
CN104573740A (en) * | 2014-12-22 | 2015-04-29 | 山东鲁能软件技术有限公司 | SVM classification model-based equipment fault diagnosing method |
CN104502834A (en) * | 2014-12-30 | 2015-04-08 | 南京农业大学 | OFRF (output frequency response function) modeling based nonlinear analog circuit fault parameter identification method |
CN104635146B (en) * | 2015-02-06 | 2017-04-19 | 南京农业大学 | Analog circuit fault diagnosis method based on random sinusoidal signal test and HMM (Hidden Markov Model) |
CN104635146A (en) * | 2015-02-06 | 2015-05-20 | 南京农业大学 | Analog circuit fault diagnosis method based on random sinusoidal signal test and HMM (Hidden Markov Model) |
CN104849650A (en) * | 2015-05-19 | 2015-08-19 | 重庆大学 | Analog circuit fault diagnosis method based on improvement |
CN104849650B (en) * | 2015-05-19 | 2018-03-02 | 重庆大学 | One kind is based on improved analog-circuit fault diagnosis method |
US10539613B2 (en) | 2015-08-07 | 2020-01-21 | Hefei University Of Technology | Analog circuit fault diagnosis method using single testable node |
WO2017024692A1 (en) * | 2015-08-07 | 2017-02-16 | 合肥工业大学 | Method of simulating analog circuit diagnostic fault using single measurement node |
CN105044562A (en) * | 2015-08-26 | 2015-11-11 | 云南电网有限责任公司电力科学研究院 | Distribution network fault positioning method based on Bayes formula algorithm |
CN105652182B (en) * | 2015-12-28 | 2018-10-02 | 北京航天测控技术有限公司 | A kind of board failure positioning system and method based on circuit network and graph search |
CN105652182A (en) * | 2015-12-28 | 2016-06-08 | 北京航天测控技术有限公司 | Circuit board fault positioning system and circuit board fault positioning method based on circuit network and graph search |
CN105699882B (en) * | 2016-01-22 | 2018-05-29 | 电子科技大学 | Analog-circuit fault diagnosis method based on vibration test technology |
CN106444468A (en) * | 2016-05-03 | 2017-02-22 | 重庆青年职业技术学院 | Information machine adapter detection system and method |
CN106444468B (en) * | 2016-05-03 | 2023-09-05 | 重庆青年职业技术学院 | Information machine adapter detection system and method |
CN106896306B (en) * | 2017-04-26 | 2019-04-30 | 国网上海市电力公司 | A kind of GIS oscillatory surge pressure test signal antinoise method |
CN106896306A (en) * | 2017-04-26 | 2017-06-27 | 国网上海市电力公司 | A kind of GIS oscillatory surges pressure test signal antinoise method |
CN107220670A (en) * | 2017-05-27 | 2017-09-29 | 重庆大学 | Supervised Artifical Taste system features extracting method is had based on wavelet transform |
CN107220670B (en) * | 2017-05-27 | 2020-07-14 | 重庆大学 | Method for extracting characteristics of supervised artificial taste system based on discrete wavelet transform |
CN107843779A (en) * | 2017-09-26 | 2018-03-27 | 国网河南省电力公司民权县供电公司 | A kind of Power System Fault Record classifying and analyzing method and system based on fuzzy clustering |
CN108470462A (en) * | 2018-05-18 | 2018-08-31 | 上海会为智能技术有限公司 | A kind of signal lamp failure detection method, device, equipment and medium |
CN109885018A (en) * | 2019-01-18 | 2019-06-14 | 广东寰球智能科技有限公司 | A kind of management method of the expert think tank system towards mental package equipment |
CN109885018B (en) * | 2019-01-18 | 2020-10-09 | 广东寰球智能科技有限公司 | Management method of expert intelligent library system for metal packaging equipment |
CN109726770A (en) * | 2019-02-19 | 2019-05-07 | 金陵科技学院 | A kind of analog circuit fault testing and diagnosing method |
CN110579709A (en) * | 2019-08-30 | 2019-12-17 | 西南交通大学 | fault diagnosis method for proton exchange membrane fuel cell for tramcar |
CN110579709B (en) * | 2019-08-30 | 2021-04-13 | 西南交通大学 | Fault diagnosis method for proton exchange membrane fuel cell for tramcar |
CN111307453B (en) * | 2020-03-20 | 2021-11-12 | 朗斯顿科技(北京)有限公司 | Transmission system fault diagnosis method based on multi-information fusion |
CN111307453A (en) * | 2020-03-20 | 2020-06-19 | 朗斯顿科技(北京)有限公司 | Transmission system fault diagnosis method based on multi-information fusion |
CN111523594A (en) * | 2020-04-23 | 2020-08-11 | 湖州师范学院 | Improved KNN fault classification method based on LDA-KMEDOIDS |
CN113159088B (en) * | 2021-01-07 | 2022-07-15 | 中国地质大学(武汉) | Fault monitoring and diagnosis method based on multi-feature fusion and width learning |
CN113159088A (en) * | 2021-01-07 | 2021-07-23 | 中国地质大学(武汉) | Fault monitoring and diagnosis method based on multi-feature fusion and width learning |
CN113409284A (en) * | 2021-06-28 | 2021-09-17 | 北京百度网讯科技有限公司 | Circuit board fault detection method, device, equipment and storage medium |
CN113409284B (en) * | 2021-06-28 | 2023-10-20 | 北京百度网讯科技有限公司 | Circuit board fault detection method, device, equipment and storage medium |
Also Published As
Publication number | Publication date |
---|---|
CN102520341B (en) | 2013-12-18 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN102520341B (en) | Analog circuit fault diagnosis method based on Bayes-KFCM (Kernelized Fuzzy C-Means) algorithm | |
Kim et al. | Optimization of cluster-based evolutionary undersampling for the artificial neural networks in corporate bankruptcy prediction | |
Hruschka et al. | Extracting rules from multilayer perceptrons in classification problems: A clustering-based approach | |
He et al. | A two-stage genetic algorithm for automatic clustering | |
Gheyas et al. | Feature subset selection in large dimensionality domains | |
CN108062572A (en) | A kind of Fault Diagnosis Method of Hydro-generating Unit and system based on DdAE deep learning models | |
CN104809722A (en) | Electrical device fault diagnosis method based on infrared thermography | |
CN103926526A (en) | Analog circuit fault diagnosis method based on improved RBF neural network | |
CN112906770A (en) | Cross-modal fusion-based deep clustering method and system | |
Wang et al. | A dendritic neuron model with adaptive synapses trained by differential evolution algorithm | |
CN1656472A (en) | Plausible neural network with supervised and unsupervised cluster analysis | |
Shen et al. | Mining entity synonyms with efficient neural set generation | |
CN105095836A (en) | Skin texture detecting method and apparatus based on Gabor features | |
CN114118138A (en) | Bearing composite fault diagnosis method based on multi-label field self-adaptive model | |
CN113222035A (en) | Multi-class imbalance fault classification method based on reinforcement learning and knowledge distillation | |
Yan et al. | Trustworthiness evaluation and retrieval-based revision method for case-based reasoning classifiers | |
CN115310837A (en) | Complex electromechanical system fault detection method based on causal graph attention neural network | |
CN113343123B (en) | Training method and detection method for generating confrontation multiple relation graph network | |
CN115310589A (en) | Group identification method and system based on depth map self-supervision learning | |
CN104537383A (en) | Massive organizational structure data classification method and system based on particle swarm | |
Li et al. | Many-objective coevolutionary learning algorithm with extreme learning machine auto-encoder for ensemble classifier of feedforward neural networks | |
CN112668633A (en) | Adaptive graph migration learning method based on fine granularity field | |
Ren et al. | Multiple sparse detection-based evolutionary algorithm for large-scale sparse multiobjective optimization problems | |
Liu et al. | A weight-incorporated similarity-based clustering ensemble method | |
Stoean et al. | Elitist generational genetic chromodynamics-a new radii-based evolutionary algorithm for multimodal optimization |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
C14 | Grant of patent or utility model | ||
GR01 | Patent grant | ||
CF01 | Termination of patent right due to non-payment of annual fee | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20131218 Termination date: 20161205 |