CN108563874A - A kind of analog circuit intermittent fault diagnostic method - Google Patents
A kind of analog circuit intermittent fault diagnostic method Download PDFInfo
- Publication number
- CN108563874A CN108563874A CN201810343669.9A CN201810343669A CN108563874A CN 108563874 A CN108563874 A CN 108563874A CN 201810343669 A CN201810343669 A CN 201810343669A CN 108563874 A CN108563874 A CN 108563874A
- Authority
- CN
- China
- Prior art keywords
- state
- intermittent fault
- circuit
- fault
- intermittent
- 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.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/30—Circuit design
- G06F30/36—Circuit design at the analogue level
- G06F30/367—Design verification, e.g. using simulation, simulation program with integrated circuit emphasis [SPICE], direct methods or relaxation methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/30—Circuit design
- G06F30/36—Circuit design at the analogue level
Landscapes
- Engineering & Computer Science (AREA)
- Computer Hardware Design (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Evolutionary Computation (AREA)
- Geometry (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Microelectronics & Electronic Packaging (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The invention discloses a kind of analog circuit intermittent fault diagnostic methods, without studying the dynamic characteristic of intermittent fault system complex, only obtain enough intermittent fault data and are analyzed and handled to realize the diagnosis of intermittent fault.Include the following steps：Off-line modeling is carried out to circuit under test by circuit simulating software first, sensitivity analysis is carried out to circuit under test, chooses generation node of the element of high sensitivity as intermittent fault and permanent fault；Circuit under test is applied and is encouraged, the K output signal that K MonteCarlo is analyzed under different conditions is obtained；J layers of WAVELET PACKET DECOMPOSITION are carried out to the output signal of each state, extract the energy feature of each inband signal as feature vector；K feature vector of each state is formed into observation sequence；It trains to obtain the corresponding GHMM models of different conditions by the observation sequence of each state, completes off-line modeling；Actual measurement diagnosis is finally carried out, the observation sequence of analog circuit measured signal is inputted in trained GHMM model libraries, obtains diagnostic result.
Description
Technical field
The invention belongs to analog circuit fault diagnosing fields, are related to a kind of analog circuit intermittent fault diagnostic method.
Background technology
Analog circuit has a very wide range of applications in industrial production, daily life and high-tech area.In electronic circuit
In system, intermittent fault seriously affects system performance.In hybrid circuit, intermittent fault occurrence frequency be permanent fault 10~
30 times, be the main reason for causing thrashing.Although digital circuit accounts for about 80% in electronic equipment, analog circuit accounts for about
20%, but in its life cycle, about 80% failure comes from analog circuit.Obviously, the reliability and maintainability of analog circuit
It has been largely fixed the reliability and maintainability of electronic equipment.
Intermittent fault, which is one kind, to be occurred repeatedly, the unprocessed impermanent failure died away, and have randomness,
Having a rest property and repeatability.In electronic equipment especially large scale integrated circuit, since manufacturing process is bad and lack of standardization uses
Caused component rosin joint, chip pin and line loosening etc. can lead to the generation of circuit intermittent fault.
Currently, there are many method of extraction fault signature, such as Fast Fourier Transform (FFT), wavelet transformation, wavelet package transforms, cepstrum
With Wigner distribution etc..Fault-signal is often non-stationary and nonlinear in analog circuit, some conventional methods cannot be ideal
Extract fault signature in ground.Wavelet transformation is a kind of Time-Frequency Analysis Method suitable for non-stationary signal, but it is lost signal height
The relevant information of frequency band.Wavelet package transforms overcome the shortcomings that wavelet transformation, can be carried out to entire signal frequency range uniform
It decomposes.After WAVELET PACKET DECOMPOSITION, signal can be decomposed into multiple independent frequency bands without occurring to omit and be overlapped.
Due to intermittent fault complicated mechanism, the method condition for establishing analytic modell analytical model more just carves, is not easy to realize.And Gauss is mixed
It is a kind of method for diagnosing faults based on statistical model to close Hidden Markov Model (GHMM), without to intermittent fault characteristic and machine
Reason carries out deep analysis, it is only necessary to the GHMM models of probability statistics can be established by the data of each state operation of system,
It is assured that system current state by the distortion of measured signal and each state, to reach intermittent fault diagnosis
Purpose.
Invention content
In view of this, the purpose of the present invention is to provide a kind of analog circuit intermittent fault diagnostic methods.This method passes through
Wavelet package transforms carry out feature extraction to measured signal, using wavelet-packet energy as feature vector, the sight that feature vector is constituted
It surveys in the trained GHMM model libraries of sequence inputting and carries out match cognization, to achieve the purpose that diagnose intermittent fault.
In order to achieve the above objectives, technical scheme of the present invention provides a kind of analog circuit intermittent fault diagnostic method, described
Method includes the following steps：
1) it uses circuit simulating software to carry out sensitivity analysis to circuit under test, chooses the element of high sensitivity as interval
The tolerance of the happening part of failure and permanent fault, each element of given circuit is 5%.
2) circuit under test is applied and is encouraged, in the off-line modeling stage, by circuit simulating software, just for different elements
Normal state, intermittent fault state and permanent fault state execute K Monte respectively according to the variation range of component parameters
Carlo is analyzed, and finally respectively obtain circuit under test under normal condition, intermittent fault state and permanent fault state K exports
Signal.
3) j layers are carried out to the output signal under normal condition, intermittent fault state and the permanent fault state in step 2)
WAVELET PACKET DECOMPOSITION extracts the energy feature of each inband signal as feature vector, and each state has K feature vector.
4) K feature vector under each state in step 3) is formed into observation sequence.Pass through the observation sequence of each state
Row training obtains different elements different conditions corresponding GHMM models (normal condition, intermittent fault state and permanent fault shape
State), complete off-line modeling.The wavelet packet character that the signal for surveying circuit is carried out to step 3) simultaneously is extracted, construction feature vector,
It is used with carrying out test identification.
5) observation sequence of the measured signal of circuit is inputted in trained GHMM model libraries, by measured signal and respectively
The similarity of model comes the normal condition of identification circuit, intermittent fault state and permanent fault state, obtains diagnostic result, completes
The diagnosis of intermittent fault.
What the present invention reached has the beneficial effect that：The present invention is not necessarily to study the dynamic characteristic of intermittent fault system complex, only needs
Enough intermittent fault data are obtained to be analyzed and handled to realize the diagnosis of intermittent fault.It will be believed by wavelet package transforms
Number high and low frequency component in the energy feature that is extracted as fault signature, to improve the accuracy of identification of intermittent fault.In interval
In terms of fault diagnosis, compared with conventional method, Hidden Markov Model is a kind of dual random process, it not only turns between state
Shifting is random, and the observation of each of which state is also random, is suitble to processing dynamic stochastic process, true state in model
It is hiding, presence and the feature of state can be perceived by observed value.For intermittent fault randomness, intermittence and repeatedly
This dynamic random failure of property, Hidden Markov Model can preferably handle the identification of this kind of failure.Compared to discrete hidden Ma Erke
Husband's model, effective information loses problem caused by GHMM overcomes this process of observation quantification treatment, can preferably locate
Continuous signal is managed, to improve fault recognition rate.
Description of the drawings
In order to keep the purpose of the present invention, technical solution and advantageous effect clearer, the present invention provides following attached drawing and carries out
Explanation：
Fig. 1 is the flow diagram of analog circuit intermittent fault diagnostic method of the present invention.
Fig. 2 is three layers of WAVELET PACKET DECOMPOSITION structure chart of analog circuit intermittent fault diagnostic method of the present invention.
Fig. 3 is GHMM training and the diagnostic flow chart of analog circuit intermittent fault diagnostic method of the present invention.
Specific implementation mode
For a clearer understanding of the present invention, being described in detail below in conjunction with attached drawing：
Referring to Fig.1, the present embodiment includes the following steps：
1) it uses circuit simulating software to carry out sensitivity analysis to circuit under test, chooses the element of high sensitivity as interval
The tolerance of the happening part of failure and permanent fault, each element of given circuit is 5%.
2) circuit under test is applied and is encouraged, in the off-line modeling stage, by circuit simulating software, just for different elements
Normal state, intermittent fault state and permanent fault state execute K Monte respectively according to the variation range of component parameters
Carlo is analyzed, and finally respectively obtain circuit under test under normal condition, intermittent fault state and permanent fault state K exports
Signal.
Though dynamic characteristic of the present invention without studying intermittent fault system complex, needs to obtain enough intermittent fault numbers
The diagnosis of intermittent fault is realized according to being analyzed and being handled.In view of being difficult to obtain large number of intermittently number of faults in actual circuit
According to, and be difficult to ensure the generation of intermittent fault in measurement process, therefore consider to obtain analog circuit offline by simulation software
Intermittent fault data, to carry out the training of intermittent fault GHMM models.
The acquisition of data is emulated under its interval malfunction, between being used as by the key components obtained in step 1)
Node occurs for failure of having a rest, and the switch elements such as relay are added at key node, by switch elements such as control relays
Intermittent break-make come simulate rosin joint, line loosen etc. intermittent faults generation, to obtain the intermittent fault data of different elements
To carry out intermittent fault GHMM model trainings.
3) j layers are carried out to the output signal under normal condition, intermittent fault state and the permanent fault state in step 2)
WAVELET PACKET DECOMPOSITION extracts the energy feature of each inband signal as feature vector, and each state has K feature vector.
Intermittent fault be frequently experienced in failure early stage, have certain cumulative effect, be finally changed into it is irreversible forever
Its interval fault-signal of long failure is often mixed in normal signal, with normal state signal and permanent fault status signal phase
Than energy feature has differences in partial-band, according to this species diversity, chooses WAVELET PACKET DECOMPOSITION signal high and low frequency
Part is non-overlapping, is exhaustively decomposed, and extracts the energy feature of each frequency band to achieve the purpose that follow-up diagnosis intermittent fault.
Its wavelet packet decomposition algorithm is：
J is scale factor in formula, and n is modulation parameter or parameter of oscillation, dj,nIndicate the wavelet systems of jth n-th of sub-band of layer
Number, dj+1,2nAnd dj+1,2n+1Indicate that the wavelet coefficient of jth+1 layer of 2n and 2n+1 sub-band, m are translational movements, k becomes for function
Amount.
By taking three layers of WAVELET PACKET DECOMPOSITION as an example, WAVELET PACKET DECOMPOSITION structure chart is as shown in Figure 2.X (t) is original signal in figure,
d1,0(t) the first layer low frequency coefficient of WAVELET PACKET DECOMPOSITION, d are indicated1,1(t) expression first layer high frequency coefficient, other and so on.
The calculation formula of wavelet-packet energy is：
Wherein, k indicates that kth time Monte Carlo analyses, (j, i) indicate i-th of section of the jth layer of WAVELET PACKET DECOMPOSITION tree
Point,It is the reconstruction signal of corresponding node,For the energy of corresponding node.
The feature vector that each frequency band energy after signal decomposition is constituted is found out, then kth time Monte Carlo analyses correspond to
Feature vector is：
4) K feature vector under each state in step 3) is formed into observation sequence.Pass through the observation sequence of each state
Row are trained to obtain GHMM models (normal condition, intermittent fault state and permanent event corresponding to different elements different conditions
Barrier state), complete off-line modeling.The wavelet packet character that the signal for surveying circuit is carried out to step 3) simultaneously is extracted, construction feature
Vector is used with carrying out test identification.
Wherein, state snObservation sequence can be expressed as：
O(sn)={ T(1),T(2),…,T(K)} (4)
Wherein, simulation software training sample observation sequence is obtained to be denoted asThe test sample obtained by actual measurement circuit
Observation sequence is denoted as
If GHMM models have N number of state, Gaussian mixture number M, GHMM model that can indicate as follows：
λ=(π, A, C, μ, U) (5)
Wherein, initial state probabilities are distributed π, state-transition matrix A, mixed weight-value Matrix C, mean value vector μ, covariance square
Battle array U indicates as follows：
U=[Unm]N×M (10)
Wherein, qtIndicate the state under moment t, P is probability, cnmIt is state snThe weight of lower m-th of single Gauss,It is
State snThe D of m-th of single Gauss tie up mean value vector, UnmFor state snD × D of m-th of single Gauss tie up covariance.Then probability
Density function can be written as
Wherein,It is observation vectorIn state snGaussian density function.
GHMM is trained and diagnostic process is as shown in Figure 3.By training sampleIt is carried out using Baum-Welch algorithms initial
The training of model, the GHMM models after training are denoted asThe parameter re-evaluation method of GHMM model trainings is as follows：
In formulaIndicate that initial time is in state snProbability,Indicate that moment i is in state snIt is general
Rate,Indicate the state s of moment tiBecome state s in subsequent timejProbability,Expression moment t is shape
State snWhen, the output probability of m-th of single Gauss.By parameter revaluation, the parameter revaluation of complete paired systems initial model was trained
Journey terminates.
5) observation sequence of the measured signal of circuit is inputted in trained GHMM model libraries, by measured signal and respectively
The similarity of model comes the normal condition of identification circuit, intermittent fault state and permanent fault state, obtains diagnostic result, completes
The diagnosis of intermittent fault.
After its detailed process diagnosed by the measured data of circuit under test as shown in figure 3, carry out feature extraction, electricity to be measured
The observation sequence on roadIt is input in trained each GHMM models, by forward algorithms acquisition observation sequence and often
The similarity probability of a modelIt finds out maximumCorresponding model state is system shape
State obtains diagnostic result.
Claims (6)
1. a kind of analog circuit intermittent fault diagnostic method, which is characterized in that include the following steps：
1) it uses circuit simulating software to carry out sensitivity analysis to circuit under test, chooses the element of high sensitivity as intermittent fault
Tolerance with the happening part of permanent fault, each element of given circuit is 5%.
2) circuit under test is applied and is encouraged, in the off-line modeling stage, by circuit simulating software, for the normal shape of different elements
State, intermittent fault state and permanent fault state execute K Carlo points of Monte respectively according to the variation range of component parameters
Analysis finally respectively obtains K output signal of circuit under test under normal condition, intermittent fault state and permanent fault state.
3) j layers of small echo are carried out to the output signal under normal condition, intermittent fault state and the permanent fault state in step 2)
Packet decomposes, and extracts the energy feature of each inband signal as feature vector, each state has K feature vector.
4) K feature vector under each state in step 3) is formed into observation sequence.It is instructed by the observation sequence of each state
The corresponding GHMM models of different elements different conditions (normal condition, intermittent fault state and permanent fault state) are got, it is complete
At off-line modeling.The wavelet packet character that the signal for surveying circuit is carried out to step 3) simultaneously is extracted, construction feature vector, to carry out
Test identification uses.
5) observation sequence of the measured signal of circuit is inputted in trained GHMM model libraries, passes through measured signal and each model
Similarity come the normal condition of identification circuit, intermittent fault state and permanent fault state, obtain diagnostic result, complete interval
The diagnosis of failure.
2. a kind of according to claim 1, analog circuit intermittent fault diagnostic method, which is characterized in that the step 2)
The acquisition of data is emulated under its interval malfunction, it is contemplated that it is difficult to obtain large number of intermittently fault data in actual circuit, and
It is difficult to ensure the generation of intermittent fault in measurement process, therefore considers to obtain analog circuit intermittent fault offline by simulation software
Data.It is used as intermittent fault by the key components obtained in step 1) and node occurs, relay is added at key node
Equal switch elements simulate the interval events such as rosin joint, line loosening by the intermittent break-make in switch elements such as control relays
The generation of barrier, to obtain the intermittent fault data of different elements to carry out intermittent fault GHMM model trainings.
3. a kind of according to claim 1, analog circuit intermittent fault diagnostic method, which is characterized in that the step 3)
Wavelet-packet energy calculates as follows：
Wherein, k indicates kth time Monte Carlo analyses, and (j, i) indicates i-th of node of the jth layer of WAVELET PACKET DECOMPOSITION tree,It is the reconstruction signal of corresponding node,For the energy of corresponding node.Then kth time Monte Carlo analyze character pair to
Amount is：
4. a kind of according to claim 1, analog circuit intermittent fault diagnostic method, which is characterized in that the step 4)
State snObservation sequence can be expressed as：
O(sn)={ T(1),T(2),…,T(K)} (3)
Wherein, simulation software training sample observation sequence is obtained to be denoted asThe test sample observation obtained by actual measurement circuit
Sequence is denoted as
5. a kind of according to claim 1, analog circuit intermittent fault diagnostic method, which is characterized in that the step 4)
Training algorithm use Baum-Welch algorithms.GHMM models after it is trained are denoted as
6. a kind of according to claim 1, analog circuit intermittent fault diagnostic method, which is characterized in that the step 5)
Each GHMM models are calculated using forward algorithmsPosterior probability under measured signal observation sequence
It finds out maximumCorresponding model state is system mode, obtains diagnostic result, completes intermittent fault and examines
It is disconnected.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810343669.9A CN108563874A (en) | 2018-04-17 | 2018-04-17 | A kind of analog circuit intermittent fault diagnostic method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810343669.9A CN108563874A (en) | 2018-04-17 | 2018-04-17 | A kind of analog circuit intermittent fault diagnostic method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108563874A true CN108563874A (en) | 2018-09-21 |
Family
ID=63535609
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810343669.9A Pending CN108563874A (en) | 2018-04-17 | 2018-04-17 | A kind of analog circuit intermittent fault diagnostic method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108563874A (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109782157A (en) * | 2019-01-22 | 2019-05-21 | 北京旋极信息技术股份有限公司 | A kind of built-in test design method and system |
| CN110046663A (en) * | 2019-04-16 | 2019-07-23 | 重庆大学 | A kind of complex electromechanical systems fault critical state discrimination method |
| CN110058110A (en) * | 2019-04-16 | 2019-07-26 | 重庆大学 | A kind of active inverter intermittent fault diagnostic method |
| CN112098802A (en) * | 2020-07-31 | 2020-12-18 | 中国人民解放军国防科技大学 | Welding spot intermittent fault diagnosis method based on wavelet time-frequency entropy energy sequence |
| CN112104340A (en) * | 2020-09-08 | 2020-12-18 | 华北电力大学 | HMM model and Kalman filtering technology-based switching value input module BIT false alarm reduction method |
| CN112731098A (en) * | 2020-12-03 | 2021-04-30 | 西安电子科技大学 | Radio frequency low-noise discharge circuit fault diagnosis method, system, medium and application |
| CN113984103A (en) * | 2021-10-29 | 2022-01-28 | 自然资源部第二海洋研究所 | Automatic test system and test method for ocean observation |
| WO2023274121A1 (en) * | 2021-06-28 | 2023-01-05 | 中兴通讯股份有限公司 | Fault detection method and apparatus, and electronic device and computer-readable storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20020072872A1 (en) * | 2000-04-19 | 2002-06-13 | Abhijit Chatterjee | Method for diagnosing process parameter variations from measurements in analog circuits |
| CN102413005A (en) * | 2010-09-26 | 2012-04-11 | 北京旋极信息技术股份有限公司 | Fault injection method |
| CN103064008A (en) * | 2012-12-26 | 2013-04-24 | 浙江大学 | Nolinear analog circuit soft fault diagnostic method based on Hilbert-huang transform |
| CN104504296A (en) * | 2015-01-16 | 2015-04-08 | 湖南科技大学 | Gaussian mixture hidden Markov model and regression analysis remaining life prediction method |
| US20180039865A1 (en) * | 2015-08-07 | 2018-02-08 | Hefei University Of Technology | Analog circuit fault mode classification method |
-
2018
- 2018-04-17 CN CN201810343669.9A patent/CN108563874A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20020072872A1 (en) * | 2000-04-19 | 2002-06-13 | Abhijit Chatterjee | Method for diagnosing process parameter variations from measurements in analog circuits |
| CN102413005A (en) * | 2010-09-26 | 2012-04-11 | 北京旋极信息技术股份有限公司 | Fault injection method |
| CN103064008A (en) * | 2012-12-26 | 2013-04-24 | 浙江大学 | Nolinear analog circuit soft fault diagnostic method based on Hilbert-huang transform |
| CN104504296A (en) * | 2015-01-16 | 2015-04-08 | 湖南科技大学 | Gaussian mixture hidden Markov model and regression analysis remaining life prediction method |
| US20180039865A1 (en) * | 2015-08-07 | 2018-02-08 | Hefei University Of Technology | Analog circuit fault mode classification method |
Non-Patent Citations (2)
| Title |
|---|
| 孙业胜: "基于优选小波包的模拟电路故障深度特征提取方法", 《中国优秀博硕士学位论文全文数据库(硕士)信息科技辑》 * |
| 郭明威 等: "基于EMD_HMM的BIT间歇故障识别", 《振动、测试与诊断》 * |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109782157A (en) * | 2019-01-22 | 2019-05-21 | 北京旋极信息技术股份有限公司 | A kind of built-in test design method and system |
| CN110046663A (en) * | 2019-04-16 | 2019-07-23 | 重庆大学 | A kind of complex electromechanical systems fault critical state discrimination method |
| CN110058110A (en) * | 2019-04-16 | 2019-07-26 | 重庆大学 | A kind of active inverter intermittent fault diagnostic method |
| CN112098802A (en) * | 2020-07-31 | 2020-12-18 | 中国人民解放军国防科技大学 | Welding spot intermittent fault diagnosis method based on wavelet time-frequency entropy energy sequence |
| CN112098802B (en) * | 2020-07-31 | 2022-09-02 | 中国人民解放军国防科技大学 | Welding spot intermittent fault diagnosis method based on wavelet time-frequency entropy energy sequence |
| CN112104340A (en) * | 2020-09-08 | 2020-12-18 | 华北电力大学 | HMM model and Kalman filtering technology-based switching value input module BIT false alarm reduction method |
| CN112104340B (en) * | 2020-09-08 | 2024-04-16 | 华北电力大学 | HMM model and Kalman filtering technology-based switching value input module BIT false alarm reduction method |
| CN112731098A (en) * | 2020-12-03 | 2021-04-30 | 西安电子科技大学 | Radio frequency low-noise discharge circuit fault diagnosis method, system, medium and application |
| CN112731098B (en) * | 2020-12-03 | 2022-04-29 | 西安电子科技大学 | Radio frequency low-noise discharge circuit fault diagnosis method, system, medium and application |
| WO2023274121A1 (en) * | 2021-06-28 | 2023-01-05 | 中兴通讯股份有限公司 | Fault detection method and apparatus, and electronic device and computer-readable storage medium |
| CN113984103A (en) * | 2021-10-29 | 2022-01-28 | 自然资源部第二海洋研究所 | Automatic test system and test method for ocean observation |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108563874A (en) | A kind of analog circuit intermittent fault diagnostic method | |
| CN107894564B (en) | Analog circuit fault diagnosis method based on cross wavelet characteristics | |
| Beusch et al. | Emulating Earth system model temperatures with MESMER: from global mean temperature trajectories to grid-point-level realizations on land | |
| CN106197999B (en) | A kind of planetary gear method for diagnosing faults | |
| CN101246200B (en) | Analog PCB intelligent test system based on neural network | |
| CN108535635B (en) | EEMD and HMM based analog circuit intermittent fault diagnosis method | |
| CN101614786B (en) | Online intelligent fault diagnosis method of power electronic circuit based on FRFT and IFSVC | |
| CN101221213A (en) | Analogue circuit fault diagnosis neural network method based on particle swarm algorithm | |
| CN103064008B (en) | A kind of Nolinear analog circuit soft fault diagnostic method based on Hilbert-Huang transform | |
| CN103995237A (en) | Satellite power supply system online fault diagnosis method | |
| CN107909118A (en) | A kind of power distribution network operating mode recording sorting technique based on deep neural network | |
| CN103439647A (en) | Analog circuit fault diagnosis method for dynamic supply current monitoring | |
| CN101900789A (en) | Tolerance analog circuit fault diagnosing method based on wavelet transform and fractal dimension | |
| CH698898A2 (en) | System and method for improved condition monitoring of a plant system. | |
| CN108535648A (en) | Method of Motor Fault Diagnosis and system | |
| CN105574669B (en) | Power transmission and transformation equipment state abnormity detection method based on time-space joint data clustering analysis | |
| Xu et al. | A novel method for the diagnosis of the incipient faults in analog circuits based on LDA and HMM | |
| CN103018660A (en) | Multi-fault intelligent diagnosing method for artificial circuit utilizing quantum Hopfield neural network | |
| CN102262211A (en) | Analog circuit failure diagnosis method based on dynamic classification of echo state network | |
| CN113297922B (en) | High-voltage switch cabinet fault diagnosis method, device and storage medium | |
| CN112083328A (en) | Fault diagnosis method, system and device for high-voltage circuit breaker | |
| CN116610998A (en) | Switch cabinet fault diagnosis method and system based on multi-mode data fusion | |
| CN103064010A (en) | Parameter estimation method for artificial circuit fault component based on Hilbert-Huang transforming (HHT) | |
| Huang et al. | BIT-based intermittent fault diagnosis of analog circuits by improved deep forest classifier | |
| CN117289013A (en) | Data processing method and system for pulse current test |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180921 |