CN112308038A - Mechanical equipment fault signal identification method based on classroom type generation confrontation network model - Google Patents

Mechanical equipment fault signal identification method based on classroom type generation confrontation network model Download PDF

Info

Publication number
CN112308038A
CN112308038A CN202011340437.1A CN202011340437A CN112308038A CN 112308038 A CN112308038 A CN 112308038A CN 202011340437 A CN202011340437 A CN 202011340437A CN 112308038 A CN112308038 A CN 112308038A
Authority
CN
China
Prior art keywords
training
discriminator
mechanical equipment
generator
value
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
Application number
CN202011340437.1A
Other languages
Chinese (zh)
Other versions
CN112308038B (en
Inventor
林琳
吕彦诚
郭丰
钟诗胜
刘杰
郭昊
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Harbin Institute of Technology
Original Assignee
Harbin Institute of Technology
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Harbin Institute of Technology filed Critical Harbin Institute of Technology
Priority to CN202011340437.1A priority Critical patent/CN112308038B/en
Publication of CN112308038A publication Critical patent/CN112308038A/en
Application granted granted Critical
Publication of CN112308038B publication Critical patent/CN112308038B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
    • G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
    • G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/044—Recurrent networks, e.g. Hopfield networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/045—Combinations of networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
    • G06F2218/12—Classification; Matching

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Artificial Intelligence (AREA)
  • General Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Evolutionary Computation (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Computational Linguistics (AREA)
  • Evolutionary Biology (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Health & Medical Sciences (AREA)
  • Biomedical Technology (AREA)
  • Biophysics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • General Health & Medical Sciences (AREA)
  • Mathematical Physics (AREA)
  • Computing Systems (AREA)
  • Molecular Biology (AREA)
  • Software Systems (AREA)
  • Signal Processing (AREA)
  • Measurement Of Mechanical Vibrations Or Ultrasonic Waves (AREA)

Abstract

A mechanical equipment fault signal identification method based on a classroom type generation confrontation network model relates to the field of machine fault signal identification. The invention aims to solve the problem that the existing mechanical equipment fault signal identification method is low in accuracy. The invention discloses a classroom-based generation countermeasure network model recognition mechanical equipment fault signal method comprising a generator and a plurality of discriminators, which comprises the following steps: acquiring a normal vibration signal of mechanical equipment and a fault vibration signal of the mechanical equipment; dividing the acquired mechanical equipment signals into a test set and a training set; setting classroom type generation confrontation network structure parameters; obtaining a batch of samples; calculating an increase value of the generating capacity; calculating the influence weight of each generator on the loss function value of the discriminator; calculating a loss function of the discriminator; calculating a loss function of the generator; testing the accuracy of the discriminator; and inputting the vibration signal of the mechanical equipment into the classification model with the highest accuracy to obtain an identification result.

Description

Mechanical equipment fault signal identification method based on classroom type generation confrontation network model
Technical Field
The invention belongs to the field of fault signal identification, and particularly relates to a mechanical equipment fault signal identification method based on a classroom type generation confrontation network model.
Background
In modern industry, the occurrence of mechanical equipment failure can bring great potential safety hazard to equipment, so in order to ensure the safety of the equipment, it is necessary to effectively analyze failure information, understand and grasp the state of the equipment in the using process, determine whether the whole or part of the equipment is normal or abnormal, find the failure early, and judge and eliminate the occurrence of the failure.
In the existing fault diagnosis, the fault detection and positioning are most commonly carried out according to the vibration signal characteristics of equipment fault signals. And judging whether the fault occurs or not according to the abnormal vibration signal characteristics of the equipment when the fault occurs. Most of traditional fault signal identification methods based on machine learning are based on balance data, but in real life, fault signal data of mechanical equipment are difficult to collect, and the quantity of fault signal data is small, so that the problem that the accuracy of the existing fault signal identification method of the mechanical equipment is low is caused.
Disclosure of Invention
The invention aims to solve the problem that the existing mechanical equipment fault signal identification method is low in accuracy, and provides a mechanical equipment fault signal identification method for generating a confrontation network model based on a classroom mode.
A mechanical equipment fault signal identification method based on a classroom type generation confrontation network model comprises the following specific processes:
acquiring a normal vibration signal of mechanical equipment and a fault vibration signal of the mechanical equipment;
step two, dividing the vibration signals of the mechanical equipment obtained in the step one into a test set and a training set;
dividing normal vibration signals of the mechanical equipment into a training set and a testing set in a random sampling mode;
all vibration signals of the mechanical equipment faults are used as a test set;
step three, setting classroom type generation confrontation network structure parameters;
sampling from the training set according to prior probability distribution to obtain a batch of samples;
fifthly, training the generators by using the obtained samples and calculating the improvement value of the generating capacity of each generator in the previous training;
the value of the increase in the generation capability is a value at which the loss function value discriminated by the discriminator for each generation data changes before and after the previous training.
Step six, calculating the influence weight lambda of each generator on the loss function value of the discriminator by using a weight function according to the lifting value of the generating capacity calculated in the step fivei,t；
Step seven, calculating a loss function of the discriminator by using the obtained sample, and training the discriminator by using the loss function value;
step eight, calculating a loss function value of each generator by using the obtained samples and training each generator by using the loss function value;
classifying the test set data by using a discriminator, and testing the accuracy of the discriminator;
step ten, acquiring a vibration signal of the mechanical equipment to be detected and inputting the vibration signal into a discriminator model with the highest classification accuracy of the data of the test set.
The invention has the beneficial effects that:
the invention improves the existing generation countermeasure network model, constructs the generation countermeasure network model into a model with a plurality of generators and a discriminator, and provides a weight distribution function to adaptively adjust the influence weight of each generator on a discriminator loss function, so that the generators cooperate together, the fit degree of the discriminator and a training sample space is improved, a discriminator with excellent performance is obtained by training, and the discriminator is applied to a mechanical equipment fault signal recognition task, thereby improving the accuracy of mechanical equipment fault signal recognition.
Drawings
FIG. 1 is a diagram of a classroom-based generation of a confrontation network model architecture;
fig. 2 is a diagram of a task structure for identifying a fault signal of mechanical equipment by an arbiter.
Detailed Description
The first embodiment is as follows: the specific process of the mechanical equipment fault signal identification method for generating the confrontation network model based on the classroom type in the embodiment is as follows:
acquiring a normal vibration signal of mechanical equipment and a fault vibration signal of the mechanical equipment;
step two, dividing the vibration signals of the mechanical equipment obtained in the step one into a test set and a training set;
dividing the normal vibration signal of the mechanical equipment into a training set and a test set in a random sampling mode;
all the mechanical equipment fault vibration signals are used as a test set;
step three, setting classroom type generation confrontation network structure parameters:
step three, establishing a classroom type generation confrontation network model:
the classroom-type generation confrontation network model comprises a discriminator and a plurality of generators:
a generator for constructing a model:
Figure BDA0002798446820000021
where G is the generator, N is the number of generators, X is the generated data, Z is the noise variance, G is the noise varianceiIs the ith generator;
wherein, the generators share input data and a discrimination network, and the generator mixed structure provides a learning signal for the discriminator;
step two, setting the classroom type generation confrontation network structure parameters, including: the number of generators, the classification model structure of the countermeasure network based on classroom type generation, the training times, the starting training times, the parameter adjustment and the batch size.
Step four, randomly sampling from the training set according to prior probability distribution to obtain a batch of samples:
Figure BDA0002798446820000031
Figure BDA0002798446820000032
where x is the mechanical device vibration signal in the training set, z is the noise sample, i is the sample number, pxIs the probability distribution of x, pzIs the probability distribution of z, and m is the batch.
Step five, training the generators by using the acquired samples, and calculating the improvement value of the generating capacity of each generator in the previous training from the third training (setting Q in the previous training)i,t0):
Figure BDA0002798446820000033
wherein D ist-1Is the discriminator after the t-1 training, Dt-2Is the discriminator after the t-2 training, Gi,t-1Is the ith generator after the t-1 training, Gi,t-2Is the ith generator after the t-2 training, zt-1Is the noise sample sampled in the t-1 st training; z is a radical oft-2Is the noise sample sampled in the t-2 training,
Figure BDA0002798446820000034
is when z ist-1～pzThe expected value of the time of day,
Figure BDA0002798446820000035
is when z ist-2～pzThe expected value of the time.
Step six, according to the improvement value of the generating capacity calculated in the step fiveCalculating the weight lambda of each generator influencing the loss function value of the discriminator by using the weight functioni,t：
Figure BDA0002798446820000036
Wherein Q isi,tThe generation capacity improvement value of the ith generator before the t training is shown, alpha is more than or equal to 0 and is a super parameter for adjustment, and N is the number of generators in the model.
Step seven, calculating a loss function of the discriminator by using the obtained samples and training the discriminator by using the loss function value, wherein the loss function of the discriminator is as follows:
Figure BDA0002798446820000037
wherein,
Figure BDA0002798446820000041
is the loss function of the discriminator, Gi(z) is a generated sample of the i-th generator, Dt-1(x) Is the predicted value of the discriminator after the t-1 training on the source of the input sample, ztIs the noise sample sampled in the t-th training;
when T < Tstart-upEach generator has the same weight of influence on the value of the discriminator loss function, i.e.
Figure BDA0002798446820000042
That is, if the super parameter α is set to 0, then:
Figure BDA0002798446820000043
wherein, Tstart-upIs the number of training times in the starting phase of the model training, DlossIs T < Tstart-upThe penalty function of the time arbiter.
Step eight, calculating a loss function value of each generator by using the obtained samples and training each generator by using the loss function value, wherein the loss function of each generator is as follows:
Figure BDA0002798446820000044
wherein z istIs the noise sample sampled in the t-th training, Gi,t-1Is the ith generator after the t-1 training, DtIs the discriminator after the t-th training,
Figure BDA0002798446820000045
is a loss function of the generator.
Step nine, classifying the test set data by using a discriminator, and testing the accuracy of the discriminator:
inputting a test sample into a discriminator, wherein if the output value is 1, the sample is a normal vibration signal of mechanical equipment, and if the output value is 0, the sample is a fault vibration signal of the mechanical equipment;
and comparing the output result of the discriminator with the label of the real sample, judging accurately if the output result is the same as the label of the real sample, judging wrongly if the output result is different from the label of the real sample, and then calculating the average accuracy of all samples.
Step ten, acquiring a vibration signal of the mechanical equipment to be detected and inputting the vibration signal into a discriminator model with the highest classification accuracy of the data of the test set.
Example (b):
according to the technical scheme of the specific implementation mode, a classroom-based generation confrontation network model is obtained, and a mechanical fault signal identification task is completed:
aiming at the classroom type generation countermeasure network structure provided by the invention, a data set provided by a bearing data center of the University of Kaiser Western University (CWRU) is used for experimental analysis so as to verify the capability of a discriminator obtained by training the classroom type generation countermeasure network to identify a fault signal of mechanical equipment. The method comprises the steps that a CWRU data set supports a motor rotating shaft through a detected bearing, a driving end bearing is SKF6205, a fan end bearing is SKF6203, the bearings are subjected to electric spark machining single-point damage, acceleration sensors are respectively arranged above a fan end and a driving end bearing seat of the motor and used for collecting vibration acceleration signals of the bearings, the machined bearings are installed in a test motor, vibration acceleration signal data are collected under the motor load working conditions of 0 horsepower, 1 horsepower, 2 and horsepower, the sampling frequency is 12KHz, in the method, 5 common neural network structures, namely, a feedforward neural network (FF), a fed-forward neural network (0.3556mm) and a deconvolution network (DN, Deconvolvulnal network), a long-term memory network (LSTM), long short-term memory), Radial Basis Function (RBF) and Residual Network (RN) as generator models, and Convolutional Neural Network (CNN) is used as the discriminator network.
Performing experiments on CWRU data, wherein the classification accuracy of each finally obtained discriminator on the class data (non-fault bearing vibration data signal data) participating in training and the class data (fault bearing vibration data signal data) not participating in training is shown in Table 1, DC, FF, LSTM, RBF and RN represent generator models used in the experiments, the lower color represents the use condition of the generator in the models, for example, the generator model used in the model with the number 17 comprises DC, FF and RBF, the three accuracy rates are the classification accuracy rates of the discriminator on normal samples and fault samples with three different fault diameters, rank 1 represents the classification accuracy rate of the model in all models, and rank 2 represents the ranking condition of the model in the generator models with the same number; the last two rows show the classification results of a single-classification Support Vector Machine (OCSVM) as a contrast classifier and a Support Vector Data Description (SVDD) on a dataset. Table 2 counts the training time and prediction time for the test samples of the most complicated class-wise generation countermeasure network (number of generators is 5, and number is 31 in table 1 and table 2), ocsvvm and SVDD.
TABLE 1
Figure BDA0002798446820000051
Figure BDA0002798446820000061
TABLE 2
Figure BDA0002798446820000062
The experimental result shows that the classifier obtained by the classroom type generation confrontation network training can effectively distinguish the vibration data signal of the fault-free bearing from the vibration data signal of the fault bearing, and a better identification effect is obtained. As can be seen from the effect of classifying the normal samples and the fault samples with the fault diameter of 0.014in the experiment, the generation performance of the class-wise generation countermeasure network with RN, DC and LSTM combinations named 1, 2 and 4 (number 21) is better than that with RN, RBF and LSTM combinations named 1, 3 and 4 (number 25), and similarly, the DC, LSTM and FF combinations named 2, 4 and 5 (number 16) are better than those with RBF, LSTM and FF combinations named 3, 4 and 5 (number 22), because the generation space of the better performing generator is closer to the no-fault bearing vibration data signal sample space when fitting the no-fault bearing vibration data signal sample space, the resulting discriminator is more "fitted" to the training sample space, and the signal recognition performance is better.
The combination of generators with smaller performance difference can obtain unexpected effect, the corresponding arbiter of the class-type generation countermeasure network has extremely strong signal identification capability, for example, in the classification result of the arbiter on the normal sample and the fault sample with the fault diameter of 0.014in the experiment, the generation performance of the DC and the RBF are ranked as 2 (accuracy 96.070%) and 3 (accuracy 95.985%), respectively, however, the arbiter performance of the class-type generation countermeasure network (number 1) combined by the two is ranked as 1 (accuracy 98.985%) in all the dual generator models, the identification accuracy is improved to a certain extent and is better than the combination of the RN (ranking as 1, accuracy 96.855%) and the RBF (ranking as 8, accuracy 94.605%), because the generators with performance are close to the vibration data signal sample space of the fault-free bearing at similar speed in the training process, all generators cooperate with each other, the generation performance is improved, and the corresponding discriminators also have better classification capability.
For the generator combination with larger performance difference, because the time for achieving the optimal performance of the discriminator or the space difference of the generator to fit the vibration data signal sample of the fault-free bearing is larger, the situation that the generator is synchronously close to the vibration data signal sample space of the fault-free bearing in the training process cannot be achieved, and the classification capability of the obtained discriminator is not ideal, even inferior to that of the discriminator in a single generator model. In the classification result of the discriminator on the normal sample and the fault sample with the fault diameter of 0.021in the experiment, the generation performance (accuracy rate 83.750%) of FF is greatly different from the generation performance of DC (accuracy rate 98.500%) and RN (accuracy rate 99.005%), the performance of the discriminator of the dual-generator classroom generation countermeasure network of the combination of the generators and FF is not as good as that of the discriminator in the single-generator model, and therefore, the selection of the correct generator combination has a large influence on the improvement of the classification capability of the discriminator.

Claims (8)

1. The mechanical equipment fault signal identification method based on the classroom type generation confrontation network model is characterized by comprising the following steps of:
acquiring a normal vibration signal of mechanical equipment and a fault vibration signal of the mechanical equipment;
step two, dividing the vibration signals of the mechanical equipment obtained in the step one into a test set and a training set;
dividing the normal vibration signals of the mechanical equipment into a training set and a test set in a random sampling mode;
all the mechanical equipment fault vibration signals are used as a test set;
step three, setting classroom type generation confrontation network structure parameters;
step four, sampling from the training set line according to prior probability distribution to obtain a batch of samples;
fifthly, training the generators by using the obtained samples and calculating the improvement value of the generating capacity of each generator in the previous training;
the raising value of the generating capacity is a value of the change of the loss function value judged by the judger on each generating data before and after the previous training;
step six, calculating the influence weight lambda of each generator on the loss function value of the discriminator by using a weight function according to the lifting value of the generating capacity calculated in the step fivei,t；
Step seven, calculating a loss function of the discriminator by using the obtained samples and training the discriminator by using the loss function value;
step eight, calculating a loss function value of each generator by using the obtained samples and training each generator by using the loss function value;
classifying the test set data by using a discriminator, and testing the accuracy of the discriminator to obtain a discriminator model with the highest accuracy;
step ten, acquiring a vibration signal of the mechanical equipment to be detected and inputting the vibration signal into a discriminator model with the highest classification accuracy of the data of the test set.
2. The method for identifying fault signals of mechanical equipment based on the classroom-type generated countermeasure network model as claimed in claim 1, wherein the classroom-type generated countermeasure network structure parameters are set in the third step, and the specific process is as follows:
step three, establishing a classroom-based generation confrontation network model:
the classroom-based generation confrontation network model comprises a discriminator and a plurality of generators:
a generator for constructing a model:
Figure FDA0002798446810000011
where G is the generator, N is the number of generators, X is the generated data, Z is the noise variance, G is the noise varianceiIs the ith studentForming a device;
wherein, the generators share input data and a discrimination network, and the generator mixed structure provides a learning signal for the discriminator;
step two, setting the classroom type generation confrontation network structure parameters, including: the number of generators, a class-based generation of a classification model mechanism of the countermeasure network, training times, starting training times, parameter adjustment and batch size.
3. The method for identifying fault signals of mechanical equipment based on classroom-based generation of countermeasure network models as claimed in claim 2, wherein the step four is randomly sampling from the training set according to prior distribution to obtain a batch of samples:
Figure FDA0002798446810000021
Figure FDA0002798446810000022
where x is the true mechanical equipment vibration signal in the training set, z is the noise sample, i is the sample number, pxIs the probability distribution of x, pzIs the probability distribution of z, and m is the batch.
4. The method for identifying fault signals of mechanical equipment based on classroom-type generation countermeasure network model according to claim 3, wherein the fifth step is to calculate the improvement value of the generating capacity of each generator in the previous training by:
Figure FDA0002798446810000023
wherein D ist-1Is the discriminator after the t-1 training, Dt-2Is the discriminator after the t-2 training, Gi,t-1Is the ith generator after the t-1 training, Gi,t-2Is the ith generator after the t-2 training, zt-1Is the noise sample sampled in the t-1 st training; z is a radical oft-2Is the noise sample sampled in the t-2 training,
Figure FDA0002798446810000024
is when z ist-1～pzThe expected value of the time of day,
Figure FDA0002798446810000025
is when z ist-2～pzThe expected value of the time.
5. The method as claimed in claim 4, wherein the sixth step calculates the weight λ of the effect of each generator on the function value of the loss function of the discriminator by using a weight function according to the boost value of the generation capability calculated in the fifth stepi,tThe specific process is as follows:
Figure FDA0002798446810000026
wherein Q isi,tThe generation capacity improvement value of the ith generator before the t training is shown, alpha is more than or equal to 0 and is a super parameter for adjustment, and N is the number of generators in the model.
6. The class-based method for identifying faults of mechanical equipment for generating countermeasure network models according to claim 5, wherein the loss function of the discriminator in the seventh step is:
Figure FDA0002798446810000031
wherein,
Figure FDA0002798446810000032
is the loss function of the arbiter, D (x) is the predicted value of the arbiter for the sample source, Gi(z) is a generated sample of the i-th generator, Dt-1(x) Is the predicted value of the discriminator after the t-1 training on the source of the input sample, ztIs the noise sample sampled in the t-th training;
when T < Tstart-upEach generator has the same weight of influence on the value of the discriminator loss function, i.e.
Figure FDA0002798446810000033
That is, if the super parameter α is set to 0, then:
Figure FDA0002798446810000034
wherein, Tstart-upIs the number of training times in the starting phase of the model training, DlossIs T < Tstart-upThe penalty function of the time arbiter.
7. The class-based generation of mechanical equipment failure signal identification of countermeasure network models of claim 6, wherein the loss function of each generator in step eight is:
Figure FDA0002798446810000035
wherein z istIs the noise sample sampled in the t-th training, Gi,t-1Is the ith generator after the t-1 training, DtIs the discriminator after the t-th training,
Figure FDA0002798446810000036
is a loss function of the generator.
8. The class-based mechanical equipment fault signal identification method for generating the countermeasure network model according to claim 7, wherein in the ninth step, a discriminator is used for classifying the test set data, and the accuracy of the discriminator is tested by the following specific processes:
inputting a test sample into a discriminator, wherein if the output value is 1, the sample is a normal signal of the mechanical equipment, and if the output value is 0, the sample is a fault signal of the mechanical equipment;
and comparing the output result of the discriminator with the label of the real sample, judging accurately if the output result is the same as the label of the real sample, judging wrongly if the output result is different from the label of the real sample, and then calculating the average accuracy of all samples.
CN202011340437.1A 2020-11-25 2020-11-25 Mechanical equipment fault signal identification method based on classroom type generation confrontation network model Active CN112308038B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202011340437.1A CN112308038B (en) 2020-11-25 2020-11-25 Mechanical equipment fault signal identification method based on classroom type generation confrontation network model

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202011340437.1A CN112308038B (en) 2020-11-25 2020-11-25 Mechanical equipment fault signal identification method based on classroom type generation confrontation network model

Publications (2)

Publication Number Publication Date
CN112308038A true CN112308038A (en) 2021-02-02
CN112308038B CN112308038B (en) 2022-09-27

Family

ID=74335607

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202011340437.1A Active CN112308038B (en) 2020-11-25 2020-11-25 Mechanical equipment fault signal identification method based on classroom type generation confrontation network model

Country Status (1)

Country Link
CN (1) CN112308038B (en)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113221946A (en) * 2021-04-02 2021-08-06 中国人民解放军92578部队 Method for diagnosing fault types of mechanical equipment
CN113222964A (en) * 2021-05-27 2021-08-06 推想医疗科技股份有限公司 Method and device for generating coronary artery central line extraction model
CN113899393A (en) * 2021-11-29 2022-01-07 武汉飞恩微电子有限公司 MEMS sensor detection method, device, equipment and medium based on neural network
CN114124676A (en) * 2021-11-19 2022-03-01 南京邮电大学 Fault root cause positioning method and system for network intelligent operation and maintenance system

Citations (18)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2004072858A1 (en) * 2003-01-17 2004-08-26 Ho Jinyama Condition identifying metod and condition identifying system
US20070234161A1 (en) * 2006-01-10 2007-10-04 Blanton Ronald D Using neighborhood functions to extract logical models of physical failures using layout based diagnosis
JP2010086160A (en) * 2008-09-30 2010-04-15 Nec Corp Fault analysis system, fault analysis method, and program for fault analysis
JP2018013978A (en) * 2016-07-21 2018-01-25 Ｎｅｃプラットフォームズ株式会社 Fault content specifying device, fault content specifying method, and fault content specifying program
CN109033930A (en) * 2018-05-07 2018-12-18 北京化工大学 Mechanical equipment fault diagnosis method based on fault mechanism and statistical model online learning
US20190155709A1 (en) * 2017-11-21 2019-05-23 Siemens Healthcare Gmbh Automatic failure detection in magnetic resonance apparatuses
US20190166645A1 (en) * 2017-11-29 2019-05-30 Qualcomm Incorporated Determining beam candidates for transmitting beam failure recovery signal
CN110567720A (en) * 2019-08-07 2019-12-13 东北电力大学 method for diagnosing depth confrontation of fault of fan bearing under unbalanced small sample scene
CN110617966A (en) * 2019-09-23 2019-12-27 江南大学 Bearing fault diagnosis method based on semi-supervised generation countermeasure network
CN110647923A (en) * 2019-09-04 2020-01-03 西安交通大学 Variable working condition mechanical fault intelligent diagnosis method based on self-learning under small sample
CN110823576A (en) * 2019-11-18 2020-02-21 苏州大学 Mechanical anomaly detection method based on generation of countermeasure network
CN110823574A (en) * 2019-09-30 2020-02-21 安徽富煌科技股份有限公司 Fault diagnosis method based on semi-supervised learning deep countermeasure network
US20200183032A1 (en) * 2018-12-11 2020-06-11 Exxonmobil Upstream Research Company Training machine learning systems for seismic interpretation
CN111337243A (en) * 2020-02-27 2020-06-26 上海电力大学 ACGAN-based wind turbine generator planet wheel gearbox fault diagnosis method
CN111458142A (en) * 2020-04-02 2020-07-28 苏州智传新自动化科技有限公司 Sliding bearing fault diagnosis method based on generation of countermeasure network and convolutional neural network
CN111800811A (en) * 2020-09-08 2020-10-20 中国人民解放军国防科技大学 Unsupervised detection method, unsupervised detection device, unsupervised detection equipment and storage medium for frequency spectrum abnormality
CN111914705A (en) * 2020-07-20 2020-11-10 华中科技大学 Signal generation method and device for improving health state evaluation accuracy of reactor
CN112543941A (en) * 2018-06-13 2021-03-23 科斯默人工智能-Ai有限公司 System and method for training generative confrontation networks and using trained generative confrontation networks

Patent Citations (18)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2004072858A1 (en) * 2003-01-17 2004-08-26 Ho Jinyama Condition identifying metod and condition identifying system
US20070234161A1 (en) * 2006-01-10 2007-10-04 Blanton Ronald D Using neighborhood functions to extract logical models of physical failures using layout based diagnosis
JP2010086160A (en) * 2008-09-30 2010-04-15 Nec Corp Fault analysis system, fault analysis method, and program for fault analysis
JP2018013978A (en) * 2016-07-21 2018-01-25 Ｎｅｃプラットフォームズ株式会社 Fault content specifying device, fault content specifying method, and fault content specifying program
US20190155709A1 (en) * 2017-11-21 2019-05-23 Siemens Healthcare Gmbh Automatic failure detection in magnetic resonance apparatuses
US20190166645A1 (en) * 2017-11-29 2019-05-30 Qualcomm Incorporated Determining beam candidates for transmitting beam failure recovery signal
CN109033930A (en) * 2018-05-07 2018-12-18 北京化工大学 Mechanical equipment fault diagnosis method based on fault mechanism and statistical model online learning
CN112543941A (en) * 2018-06-13 2021-03-23 科斯默人工智能-Ai有限公司 System and method for training generative confrontation networks and using trained generative confrontation networks
US20200183032A1 (en) * 2018-12-11 2020-06-11 Exxonmobil Upstream Research Company Training machine learning systems for seismic interpretation
CN110567720A (en) * 2019-08-07 2019-12-13 东北电力大学 method for diagnosing depth confrontation of fault of fan bearing under unbalanced small sample scene
CN110647923A (en) * 2019-09-04 2020-01-03 西安交通大学 Variable working condition mechanical fault intelligent diagnosis method based on self-learning under small sample
CN110617966A (en) * 2019-09-23 2019-12-27 江南大学 Bearing fault diagnosis method based on semi-supervised generation countermeasure network
CN110823574A (en) * 2019-09-30 2020-02-21 安徽富煌科技股份有限公司 Fault diagnosis method based on semi-supervised learning deep countermeasure network
CN110823576A (en) * 2019-11-18 2020-02-21 苏州大学 Mechanical anomaly detection method based on generation of countermeasure network
CN111337243A (en) * 2020-02-27 2020-06-26 上海电力大学 ACGAN-based wind turbine generator planet wheel gearbox fault diagnosis method
CN111458142A (en) * 2020-04-02 2020-07-28 苏州智传新自动化科技有限公司 Sliding bearing fault diagnosis method based on generation of countermeasure network and convolutional neural network
CN111914705A (en) * 2020-07-20 2020-11-10 华中科技大学 Signal generation method and device for improving health state evaluation accuracy of reactor
CN111800811A (en) * 2020-09-08 2020-10-20 中国人民解放军国防科技大学 Unsupervised detection method, unsupervised detection device, unsupervised detection equipment and storage medium for frequency spectrum abnormality

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
BIN LIAO: "《Fusion of Infrared-Visible Images in UE-IoT for Fault Point Detection Based on GAN》", 《IEEE ACCESS》 *
何威: "《基于视频特征的模拟电路故障诊断方法研究》", 《中国优秀博士学位论文全文数据库》 *
邵思羽: "《基于深度学习的旋转机械故障诊断方法研究》", 《中国优秀博士学位论文全文数据库》 *

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113221946A (en) * 2021-04-02 2021-08-06 中国人民解放军92578部队 Method for diagnosing fault types of mechanical equipment
CN113221946B (en) * 2021-04-02 2022-03-25 中国人民解放军92578部队 Method for diagnosing fault types of mechanical equipment
CN113222964A (en) * 2021-05-27 2021-08-06 推想医疗科技股份有限公司 Method and device for generating coronary artery central line extraction model
CN114124676A (en) * 2021-11-19 2022-03-01 南京邮电大学 Fault root cause positioning method and system for network intelligent operation and maintenance system
CN114124676B (en) * 2021-11-19 2024-04-02 南京邮电大学 Fault root positioning method and system for network intelligent operation and maintenance system
CN113899393A (en) * 2021-11-29 2022-01-07 武汉飞恩微电子有限公司 MEMS sensor detection method, device, equipment and medium based on neural network
CN113899393B (en) * 2021-11-29 2024-03-19 武汉飞恩微电子有限公司 Detection method, device, equipment and medium of MEMS sensor based on neural network

Also Published As

Publication number Publication date
CN112308038B (en) 2022-09-27

Similar Documents

Publication Publication Date Title
CN112308038B (en) Mechanical equipment fault signal identification method based on classroom type generation confrontation network model
CN107228766B (en) Based on the Fault Diagnosis of Roller Bearings for improving multiple dimensioned fuzzy entropy
CN111914883B (en) Spindle bearing state evaluation method and device based on deep fusion network
CN111046945B (en) Fault type and damage degree diagnosis method based on combined convolutional neural network
CN112257530B (en) Rolling bearing fault diagnosis method based on blind signal separation and support vector machine
CN110110768B (en) Rolling bearing fault diagnosis method based on parallel feature learning and multiple classifiers
CN112308147B (en) Rotary machinery fault diagnosis method based on multi-source domain anchor adapter integrated migration
CN111562108A (en) Rolling bearing intelligent fault diagnosis method based on CNN and FCMC
CN110823576B (en) Mechanical anomaly detection method based on generation of countermeasure network
CN109186964B (en) rotary machine fault diagnosis method based on angle resampling and ROC-SVM
WO2024065777A1 (en) Method, apparatus, electronic device, and storage medium for diagnosing industrial fault
CN110595778B (en) Wind turbine generator bearing fault diagnosis method based on MMF and IGRA
CN110044620A (en) A kind of Fault Diagnosis of Roller Bearings based on analysis of vibration signal
CN111382943A (en) Fault diagnosis and evaluation method based on weighted grey correlation analysis
CN115856623A (en) Motor fault diagnosis method based on uniformity and kurtosis calculation
CN115753101A (en) Bearing fault diagnosis method based on weight adaptive feature fusion
CN115034137A (en) RVM and degradation model-based two-stage hybrid prediction method for residual life of bearing
CN109726770A (en) A kind of analog circuit fault testing and diagnosing method
CN114088400A (en) Rolling bearing fault diagnosis method based on envelope permutation entropy
CN114895222B (en) Diagnostic method for identifying various faults and multiple faults of transformer
CN114383846B (en) Bearing composite fault diagnosis method based on fault label information vector
CN113588266B (en) Rolling bearing composite fault diagnosis method with embedded fault semantic space
CN116166980A (en) Fault diagnosis method and device for power equipment
CN110874088B (en) Monitoring method of ship key equipment system based on multi-dimensional vector model
Bonart et al. Enhancing end-of-line defect classifications and evaluating early testability for inline test stands using NVH measurements

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
GR01 Patent grant
GR01 Patent grant