CN111967364A - Composite fault diagnosis method, device, electronic equipment and storage medium - Google Patents

Composite fault diagnosis method, device, electronic equipment and storage medium Download PDF

Info

Publication number
CN111967364A
CN111967364A CN202010800302.2A CN202010800302A CN111967364A CN 111967364 A CN111967364 A CN 111967364A CN 202010800302 A CN202010800302 A CN 202010800302A CN 111967364 A CN111967364 A CN 111967364A
Authority
CN
China
Prior art keywords
representing
mll
neural network
sample
network model
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
CN202010800302.2A
Other languages
Chinese (zh)
Other versions
CN111967364B (en
Inventor
张亦萱
杨瑞
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.)
Xian Jiaotong Liverpool University
Original Assignee
Xian Jiaotong Liverpool University
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 Xian Jiaotong Liverpool University filed Critical Xian Jiaotong Liverpool University
Priority to CN202010800302.2A priority Critical patent/CN111967364B/en
Publication of CN111967364A publication Critical patent/CN111967364A/en
Application granted granted Critical
Publication of CN111967364B publication Critical patent/CN111967364B/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
    • G06F2218/08—Feature extraction
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
    • G01M13/00—Testing of machine parts
    • G01M13/02—Gearings; Transmission mechanisms
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
    • G01M13/00—Testing of machine parts
    • G01M13/02—Gearings; Transmission mechanisms
    • G01M13/021—Gearings
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
    • G01M13/00—Testing of machine parts
    • G01M13/02—Gearings; Transmission mechanisms
    • G01M13/028—Acoustic or vibration analysis
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
    • G01M13/00—Testing of machine parts
    • G01M13/04—Bearings
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
    • G01M13/00—Testing of machine parts
    • G01M13/04—Bearings
    • G01M13/045—Acoustic or vibration analysis
    • 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
    • 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
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G06N3/084—Backpropagation, e.g. using gradient descent

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Artificial Intelligence (AREA)
  • General Engineering & Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Evolutionary Computation (AREA)
  • Biophysics (AREA)
  • Molecular Biology (AREA)
  • Health & Medical Sciences (AREA)
  • Computational Linguistics (AREA)
  • Acoustics & Sound (AREA)
  • Software Systems (AREA)
  • General Health & Medical Sciences (AREA)
  • Biomedical Technology (AREA)
  • Computing Systems (AREA)
  • Mathematical Physics (AREA)
  • Signal Processing (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Evolutionary Biology (AREA)
  • Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)
  • Testing And Monitoring For Control Systems (AREA)

Abstract

The embodiment of the application provides a composite fault diagnosis method and device, electronic equipment and a storage medium, wherein a characteristic signal data set is obtained by acquiring a plurality of paths of working state signals of a rotary machine to be detected and preprocessing the plurality of paths of working state signals, the characteristic signal data set is analyzed according to a BP-MLL neural network model, the composite fault type of the rotary machine to be detected is determined, the composite fault diagnosis of the rotary machine is realized, and the diagnosis accuracy of the composite fault is improved.

Description

Composite fault diagnosis method, device, electronic equipment and storage medium
Technical Field
The embodiment of the application relates to a mechanical fault diagnosis technology, in particular to a composite fault diagnosis method, a composite fault diagnosis device, electronic equipment and a storage medium.
Background
A rotary machine plays an important role in a machine as a power transmission device in various machines, and fault location of the rotary machine is a major point of management and maintenance work of the machine. The composite fault is a concept opposite to a single fault, and due to the complex structure and working environment of a rotating machine, the composite fault is positioned by the related algorithm of the current multi-combination neural network.
In the prior art, vibration acceleration signals of a rotary machine under single-fault and compound-fault working conditions are collected, a certain sample extraction parameter is set, a plurality of samples are extracted in a cut-off mode, for each sample, a single label is given to a single fault, a plurality of labels are given to compound faults, then a sample set of the given labels is randomly divided into a training set and a testing set according to a certain proportion, a deep one-dimensional convolutional neural network is built by using Keras, an output layer activation function is set as a Sigmoid activation function, a cost function is set as a boundary loss function, under the condition that no sample is preprocessed, vibration data of the training set is directly input into the one-dimensional convolutional neural network for training, an optimal model is selected through Grid Search and is applied to the testing set, and a fault state classification result is obtained.
Although the prior art can diagnose the compound fault, the prior art only distinguishes the compound fault by a plurality of labels and does not consider the correlation among a plurality of faults, so that the prior art has the problem of low accuracy when the prior art model is adopted to diagnose the compound fault of the rotating machine.
Disclosure of Invention
The application provides a composite fault diagnosis method, a composite fault diagnosis device, electronic equipment and a storage medium, and aims to solve the problem that in the prior art, the diagnosis accuracy is not high.
In a first aspect, an embodiment of the present application provides a composite fault diagnosis method, including:
acquiring a multi-path working state signal of a rotary machine to be detected;
preprocessing the multi-channel working state signals to obtain a characteristic signal data set;
and analyzing the characteristic signal data set according to a back propagation multi-label learning BP-MLL neural network model, and determining the composite fault type of the rotary machine to be detected.
Optionally, the BP-MLL neural network model is trained according to a BP-MLL improvement algorithm, and the BP-MLL improvement algorithm is configured to determine a distance function according to values of the sample-dependent label set and values of the sample-independent label set:
Figure BDA0002627136840000021
wherein,
Figure BDA0002627136840000022
Figure BDA0002627136840000023
i denotes a sample number, k denotes a sample number of a correlation tag, l denotes a sample number of an uncorrelated tag,
Figure BDA0002627136840000024
output values representing relevant label position output layer neurons,
Figure BDA0002627136840000025
output values, Y, representing neurons of the output layer at unrelated label positionsiA set of labels associated with the sample is represented,
Figure BDA0002627136840000026
representing a set of labels, | Y, independent of the exemplariL represents the value of the labelset associated with the exemplar,
Figure BDA0002627136840000027
value representing a set of labels independent of the sample, fdistanceRepresenting a distance function.
Optionally, the BP-MLL refinement algorithm is configured to determine the difference between the error function:
Figure BDA0002627136840000028
adjusting weight values of a neural network, wherein E represents an error, I represents a logarithm of a sample data set, β represents a regularization coefficient, VtdRepresenting weight values between the t-th hidden layer neuron and the d-th input layer neuron, WstRepresenting the weight value between the s-th output layer neuron and the t-th hidden layer neuron, N representing the number of input layer neurons, M representing the number of hidden layer neurons, and J representing the number of output layer neurons.
Optionally, before analyzing the feature signal data set according to the back propagation multi-label learning BP-MLL neural network model and determining the type of the compound fault of the rotating machine to be detected, the method further includes:
acquiring a training sample set and a testing sample set, wherein data in the training sample set and the testing sample set are obtained by preprocessing a pre-acquired working state signal of a rotary machine;
training the training sample set by using a BP-MLL improved algorithm to obtain an initial neural network model;
and testing the initial neural network model according to the test sample set to obtain the BP-MLL neural network model.
Optionally, the preprocessing the multiple working state signals to obtain a feature signal data set includes:
extracting the characteristics of the time domain and the frequency domain of the multi-channel working state signal to obtain a time-frequency domain characteristic parameter set of the multi-channel working state signal;
and constructing the characteristic signal data set by a condensation degree evaluation algorithm according to the time-frequency domain characteristic parameter set.
In a second aspect, an embodiment of the present application provides a composite fault diagnosis apparatus, including:
the acquisition module is used for acquiring a plurality of paths of working state signals of the rotary machine to be detected;
the processing module is used for preprocessing the multi-channel working state signals to obtain a characteristic signal data set; and analyzing the characteristic signal data set according to a back propagation multi-label learning BP-MLL neural network model to determine the composite fault type of the rotary machine to be detected, wherein the BP-MLL neural network model is obtained by training according to a BP-MLL improved algorithm.
Optionally, the BP-MLL neural network model is trained according to a BP-MLL improvement algorithm, and the BP-MLL improvement algorithm is configured to determine a distance function according to values of the sample-dependent label set and values of the sample-independent label set:
Figure BDA0002627136840000041
wherein,
Figure BDA0002627136840000042
Figure BDA0002627136840000043
i denotes a sample number, k denotes a sample number of a correlation tag, l denotes a sample number of an uncorrelated tag,
Figure BDA0002627136840000044
output values representing relevant label position output layer neurons,
Figure BDA0002627136840000045
output values, Y, representing neurons of the output layer at unrelated label positionsiA set of labels associated with the sample is represented,
Figure BDA0002627136840000046
representing a set of labels, | Y, independent of the exemplariL represents the value of the labelset associated with the exemplar,
Figure BDA0002627136840000047
value representing a set of labels independent of the sample, fdistanceRepresenting a distance function;
the BP-MLL refinement algorithm is used to pass an error function:
Figure BDA0002627136840000048
adjusting weight values of a neural network, wherein E represents an error, I represents a logarithm of a sample data set, β represents a regularization coefficient, VtdRepresenting weight values between the t-th hidden layer neuron and the d-th input layer neuron, WstRepresenting the weight value between the s-th output layer neuron and the t-th hidden layer neuron, N representing the number of input layer neurons, M representing the number of hidden layer neurons, and J representing the number of output layer neurons.
Optionally, the obtaining module is further configured to obtain a training sample set and a testing sample set, where data in the training sample set and the testing sample set are obtained by preprocessing a working state signal of a rotary machine acquired in advance;
the processing module is further configured to train the training sample set by using a BP-MLL improved algorithm to obtain an initial neural network model; and testing the initial neural network model according to the test sample set to obtain the BP-MLL neural network model.
In a third aspect, an embodiment of the present application provides an electronic device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and when the processor executes the computer program, the composite fault diagnosis method according to the first aspect is implemented.
In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, on which a computer program is stored, which when executed by a processor implements the composite fault diagnosis method according to the first aspect.
The composite fault diagnosis method, the composite fault diagnosis device, the electronic equipment and the storage medium provided by the embodiment of the application can obtain the multi-path working state signals of the rotary machine to be detected, preprocessing the multi-channel working state signals to obtain a characteristic signal data set, analyzing the characteristic signal data set according to a BP-MLL neural network model to determine the composite fault type of the rotary machine to be detected, realizing the composite fault diagnosis of the rotary machine, the BP-MLL neural network model is obtained by training according to the BP-MLL improved algorithm, so that the correlation among the labels (the characteristic of the BP-MLL algorithm) is reserved, and the BP-MLL neural network model is suitable for diagnosing compound faults under multiple conditions (including the conditions of full marks and empty marks) through an improved error formula, and the diagnosis accuracy of the compound faults is greatly improved.
Drawings
Fig. 1 is a schematic flowchart of a composite fault diagnosis method according to an embodiment of the present application;
FIG. 2 is a schematic diagram of a three-layer neural network structure provided in an embodiment of the present application;
FIG. 3 is a logic flow diagram of performing a composite fault diagnosis according to an embodiment of the present disclosure;
fig. 4 is a schematic flowchart of a composite fault diagnosis method according to a second embodiment of the present application;
FIG. 5 is a schematic diagram of a training logic of a BP-MLL neural network model provided in the second embodiment of the present application;
FIG. 6 is a schematic diagram illustrating variation of error values with rounds according to a third embodiment of the present disclosure;
FIG. 7 is a schematic diagram of a classification result of a BP-MLL neural network model provided in the third embodiment of the present application;
fig. 8 is a schematic structural diagram of a composite fault diagnosis device according to a fourth embodiment of the present application;
fig. 9 is a schematic structural diagram of an electronic device according to a fifth embodiment of the present application.
Detailed Description
The present application will be described in further detail with reference to the following drawings and examples. It is to be understood that the specific embodiments described herein are merely illustrative of the application and are not limiting of the application. It should be further noted that, for the convenience of description, only some of the structures related to the present application are shown in the drawings, not all of the structures.
First, terms related to embodiments of the present application are explained:
a rotating machine: the rotating machine is a machine which can complete specific functions by means of rotation, such as a steam turbine, a gas turbine, a compressor, a fan, a generator and the like, and generally comprises a rotating shaft and various disc-shaped parts fixed on the rotating shaft, such as a bearing, a gear box, an impeller and the like.
Compound failure: it refers to two or more faults occurring simultaneously, such as a ball bearing fault and missing teeth, an inner bearing fault and tooth breakage, etc. in a rotating machine.
In the prior art, when a composite fault is diagnosed through multiple labels, only simple superposition of multiple faults is considered, and incidence relation among the multiple faults is not considered, so that the problem of low diagnosis accuracy exists in the prior art. Based on the technical problems in the prior art, the embodiment of the application provides a composite fault diagnosis method, a BP-MLL neural network model is constructed based on a BP-MLL improved algorithm, the correlation between labels is kept by performing exponential operation on the difference value of neurons marked by an output layer and neurons not marked by the output layer, the characteristic of multi-label learning is captured, the diagnosis accuracy of the composite fault is improved, meanwhile, the distance between the actual output value of the relevant label neurons and the actual output value of neurons which are not related labels is increased as much as possible by improving the error function in the original BP-MLL algorithm, and the diagnosis accuracy of the composite fault is further improved.
Example one
Fig. 1 is a schematic flowchart of a composite fault diagnosis method according to an embodiment of the present disclosure, where the method may be executed by an electronic device with a composite fault diagnosis function, such as a rotary machine management/control platform, a terminal device, and the like. As shown in fig. 1, the method of this embodiment specifically includes the following steps:
and S110, acquiring a plurality of paths of working state signals of the rotary machine to be detected.
In this step, for example, the electronic device for performing the composite fault diagnosis may be connected to a multi-channel sensor, where the multi-channel sensor is configured to measure a working state signal of the to-be-detected rotary machine, such as a force signal, a moment signal, a vibration signal, and the like.
The rotary machine to be detected is a rotary machine which needs to be subjected to compound fault diagnosis.
The working state signal is a parameter related to the mechanical performance of the rotary machine, for example, the multiple working state signals may include a force signal, a moment signal, and a vibration signal at the same time, and the force signal, the moment signal, and the vibration signal may be multiple paths, for example, the multiple working state signals may include 2 paths of force signals, 3 paths of moment signals, and 2 paths of vibration signals, which may be specifically determined according to the structure and actual analysis requirements of the rotary machine.
In the embodiment, the diagnosis of the compound fault is performed by acquiring the multi-path working state signal of the rotary machine to be detected, so that the diagnosis accuracy of the compound fault is improved.
And S120, preprocessing the multi-channel working state signals to obtain a characteristic signal data set.
In this step, an input feature signal data set satisfying the BP-MLL neural network model is obtained by preprocessing the multi-path working state information acquired in S110.
Optionally, for example, preprocessing the multiple working status signals to obtain a feature signal data set, may be implemented by:
A. and sequentially extracting the characteristics of the time domain and the frequency domain of each path of working state signal to obtain a time-frequency domain characteristic parameter set of the multi-path working state signal.
The feature extraction is a key factor affecting the accuracy and efficiency of the model, and in this step, for example, the feature extraction may be performed by combining fast fourier transform, wavelet transform and a feature parameter calculation method when performing the feature extraction on the time domain and the frequency domain of each path of working state signal.
B. And constructing a characteristic signal data set by a condensation degree evaluation algorithm according to the time-frequency domain characteristic parameter set.
In the step, firstly, the sensitivity factor of each time-frequency domain characteristic parameter is calculated according to the cohesion degree evaluation algorithm, and then on the basis of sorting the sensitivity factors, the first d parameters with larger sensitivity factors are selected to construct a sensitivity vector to obtain a characteristic signal data set.
Illustratively, the sensitivity factor is represented by η, and the characteristic signal data set is constructed by sorting the sensitivity factors by ν -sort (η)
Figure BDA0002627136840000081
S130, analyzing the characteristic signal data set according to the BP-MLL neural network model, and determining the composite fault type of the rotary machine to be detected.
In the step, the BP-MLL neural network model is a neural network model obtained by training according to a BP-MLL improved algorithm in advance, in the step, the characteristic signal data set obtained by preprocessing in the step S120 is input into the BP-MLL neural network model, and the characteristic signal data set is analyzed through the BP-MLL neural network model to realize characteristic classification, so that the composite fault type of the rotary machine to be detected is determined.
The BP-MLL improved algorithm is obtained by improving an error formula of the BP-MLL algorithm, specifically, a regular term is added into an original error formula of the BP-MLL algorithm, and a distance function of the original error formula is adjusted.
Exemplarily, fig. 2 is a schematic diagram of a three-layer neural network structure provided in an embodiment of the present application, and a principle of implementing feature classification for a BP-MLL improved algorithm is described by taking the neural network structure shown in fig. 2 as an example, it can be understood that the technical solution of the present application is also applicable when an implicit layer is two or more layers.
The three-layer neural network structure shown in FIG. 2 comprises an input layer consisting of d neurons, a hidden layer consisting of t neurons, and an output layer consisting of s neurons, wherein the neurons in each layer are represented by hollow circles and solid circles, and a1，a2，……，adCorresponding input layer neurons, { a1，a2，……，adDenotes the input values of input layer neurons, b1，……，btCorresponding hidden layer neurons, { b1，……，btDenotes the output value of the hidden layer neuron, c1，……，csCorresponding output layer neurons, { c1，……，csDenotes the output value of the output layer neuron, a0、b0Corresponding bias neuron, a0Value of a bias term representing the hidden layer, b0Representing the value of the bias term of the output layer. Except for the bias neuron, all the hidden layer neurons and the input layer neurons are connected pairwise, and the weight value is VtdTables, e.g. V12Representing the weight value between the hidden layer neuron 1 and the input layer neuron 2, connecting all the output layer neurons and the hidden layer neurons in pairs, wherein the weight value is WstOf the tabular form, e.g. W12Representing the weight values between output layer neuron 1 and hidden layer neuron 2.
Assume that a dataset has I-to-multi-label samples { (X)1,Y1),…,(Xi,Yi) Where I is 1, …, I denotes a sample number, I is a positive integer of 1 or more, and each XiIs a d-dimensional input feature vector, corresponding to an output vector YiIs a set of labels, distinguished from XiRelated tags and unrelated tags, with YiIs represented by the formula XiRelated set of tags, using
Figure BDA0002627136840000101
Representing a set of labels independent of the exemplar, accordingly, | YiL represents the value of the labelset associated with the exemplar,
Figure BDA0002627136840000102
a value representing a set of labels independent of the sample.
Each layer is completely connected with the next layer and has a weight Vtd,Wst](d ═ 1,2, …, N; (t ═ 1,2, …, M; (s ═ 1,2, …, J)), where N, M, J denotes the number of input layer neurons, hidden layer neurons, and output layer neurons, respectively, N, M, J are all positive integers equal to or greater than 1, and d, t, and s are the numbers of input layer neurons, hidden layer neurons, and output layer neurons, respectively.
(1) The BP-MLL refinement algorithm defines the distance function as follows:
Figure BDA0002627136840000103
wherein,
Figure BDA0002627136840000104
Figure BDA0002627136840000105
Figure BDA0002627136840000106
output values representing relevant label position output layer neurons,
Figure BDA0002627136840000107
output value, | Y, representing an irrelevant label position output layer neuroniL represents the value of the labelset associated with the exemplar,
Figure BDA0002627136840000108
value representing a set of labels independent of the sample, fdistanceRepresenting a distance function.
Figure BDA0002627136840000109
Denotes when YiWhen | ═ 0 (meaning all labels are not related to the sample, i.e., the case of empty labels)Let us order
Figure BDA00026271368400001010
Taken as 1, when | YiIf | ≠ 0 (indicating not the case of an empty tag), order
Figure BDA00026271368400001011
Equal to the output value of the output layer
Figure BDA00026271368400001012
Figure BDA00026271368400001013
Is shown as
Figure BDA00026271368400001014
(indicating that all tags are associated with the sample, i.e., the case of full tags), let
Figure BDA00026271368400001015
Get 1 when
Figure BDA00026271368400001016
(indicating not the case of a full tag), let
Figure BDA00026271368400001017
Equal to the output value of the output layer
Figure BDA00026271368400001018
It follows that if a sample does not have any label mark (empty label), i.e. | YiI.e. only irrelevant tags are present, the distance function is then:
Figure BDA0002627136840000111
if all tags are marked in a sample, i.e.
Figure BDA0002627136840000112
I.e. all tagers are relevant (full tag), when the distance function is:
Figure BDA0002627136840000113
ensuring correlation between labels by distance function, making output value of related label position neuron larger than that of unrelated label position neuron, and obtaining correlation by using YiI and YiI pair
Figure BDA0002627136840000114
And
Figure BDA0002627136840000115
the values are classified and discussed, so that the BP-MLL neural network model is also suitable for the conditions of empty labels and full labels, and the diagnosis accuracy of the BP-MLL neural network model on the composite faults is improved.
(2) The BP-MLL refinement algorithm defines the error function as follows:
Figure BDA0002627136840000116
wherein E represents an error;
beta represents a regularization coefficient and,
Figure BDA0002627136840000117
the error value is a regular term, and the regular term is used for balancing errors, so that the situations of infinite error value or infinitesimal error value and the like are avoided;
max(1,|Yi|) represents taking 1 and | YiThe larger of the values of l is,
Figure BDA0002627136840000118
is expressed by taking 1 and
Figure BDA0002627136840000119
greater of (d), by max (1, | Y)iI) and
Figure BDA00026271368400001110
the multiplied value is used as the denominator of the first item, so that the denominator of the first item in the error formula is not 0, the error formula can be suitable for common conditions and conditions of full labels and empty labels, and the applicability of the algorithm is improved.
In the BP-MLL modified algorithm, an error value between an output value of an output layer and a desired value is calculated by the above error formula 4, and a weight [ V ] is adjusted according to the error valuetd,Wst]Until the error value meets the preset condition.
(3) Updating of weights
Order to
Figure BDA0002627136840000121
(wherein c issRepresents the output value of the s-th output layer neuron, ScsAs a weighted sum of the output values of s output layer neurons, EiError values representing i samples), h is the activation function, derived as:
Figure BDA0002627136840000122
where h' is the derivative of the activation function, clOutput values representing neurons at unrelated label positions, ckRepresenting the output value of the relevant tag location neuron.
In this embodiment, the weight value W for the neural network is calculated according to the following formulastUpdating:
Wst＝Wst+ΔWstequation 6
Figure BDA0002627136840000123
Wherein α is a learning rate, btThe output value for the t-th hidden layer neuron.
In addition, V istdUpdate method of (1) and (W)stIn the updating mannerSimilarly, no further description is provided herein.
(3) The output results are classified according to the following formula:
Figure BDA0002627136840000124
where, the threshold value of the output value is indicated, res indicates the flag result. The above formula indicates that when the output value of the output layer is greater than the threshold value, it is marked as 1, and when the output value of the output layer is less than or equal to the threshold value, it is marked as-1.
In this embodiment, the BP-MLL neural network model can be obtained by collecting the working state signal of the rotary machine and training the working state signal according to the BP-MLL improved algorithm, and when the feature signal data set of the rotary machine to be detected is input to the BP-MLL neural network model, diagnosis of the composite fault type of the rotary machine to be detected can be realized. Fig. 3 is a logic flow diagram illustrating a composite fault diagnosis according to an embodiment of the present application.
In the embodiment, the multi-channel working state signals of the rotary machine to be detected are obtained, the multi-channel working state signals are preprocessed to obtain the characteristic signal data set, the characteristic signal data set is analyzed according to the BP-MLL neural network model, the composite fault type of the rotary machine to be detected is determined, and the composite fault diagnosis of the rotary machine is realized.
Example two
In the first embodiment, before analyzing the feature signal data set according to the BP-MLL neural network model and determining the type of the complex fault of the rotary machine to be detected, a process for constructing the BP-MLL neural network model is further included, and a specific embodiment will be described below with respect to the process for constructing the BP-MLL neural network model.
Fig. 4 is a schematic flow chart of a composite fault diagnosis method provided in the second embodiment of the present application, and as shown in fig. 4, in this embodiment, constructing a BP-MLL neural network model includes:
s210, acquiring a training sample set and a testing sample set.
In this step, the data in the training sample set and the test sample set is obtained by preprocessing the working state signal of the rotary machine acquired in advance, and the implementation flow of the preprocessing is similar to the preprocessing of the multi-path working state signal of the rotary machine to be detected in the first embodiment, and details are not repeated here.
It is understood that the rotary machine generating the working status signals of the training sample set and the test sample set may be the same as or different from the rotary machine to be tested, and the working status information of the rotary machine used to obtain the training sample set and the test sample set is collected under the condition that the fault type is known.
S220, training the training sample set by using a BP-MLL improved algorithm to obtain an initial neural network model.
In this step, illustratively, the training sample set is trained by using a BP-MLL improved algorithm, and the initial neural network model is obtained by the following steps:
A. determining the structure of a neural network, the number of neurons in each layer and an activation function, initializing weight values and bias item values connected among the neurons in each layer, and setting an update step length, a turn and an error threshold of the weight;
B. executing a forward propagation process, calculating an output value of an output layer and a loss value between the output value and an actual value according to a formula 4, namely an error, according to an input sample, a weight value connected among all layers of neurons and a value of a bias term, obtaining an initial neural network model if the error is smaller than a preset threshold value, and stopping training; if the error is larger than or equal to the preset threshold value, executing the step C;
C. and executing a back propagation process, and updating the weight values of the connections between the neurons in each layer according to a weight updating formula 6 and a weight updating formula 7.
D. Step B, C is repeated until a set maximum turn is reached or the error between the output value of the output layer and the actual value is less than the error threshold.
And S230, testing the initial neural network model according to the test sample set to obtain the BP-MLL neural network model.
In the step, a test sample set is input into an initial neural network model, the initial neural network model is tested, whether the output of the initial neural network model is effectively classified is judged, if the multiple test results are effectively classified, the initial neural network model is used as a BP-MLL neural network model, and if the multiple test results cannot meet the requirement of effective classification, the initial neural network model is adjusted according to the test results to obtain the BP-MLL neural network model.
Specifically, according to the weight value of the last round in S220, a forward propagation process is performed on the test sample set, the confidence of the predicted label of the test sample set is obtained, and the predicted multi-label-set label is obtained. Fig. 5 is a schematic diagram of a training logic of the BP-MLL neural network model according to the second embodiment of the present application.
In the embodiment, a training sample set and a test sample set are obtained, data in the training sample set and the test sample set are obtained by preprocessing a pre-collected working state signal of a rotary machine, the training sample set is trained by using a BP-MLL improved algorithm to obtain an initial neural network model, the initial neural network model is tested according to the test sample set to obtain the BP-MLL neural network model, and the reliability and the accuracy of the BP-MLL neural network model are effectively ensured through preprocessing, algorithm improvement and measurement, so that the accuracy and the efficiency of diagnosing a composite fault of the rotary machine are improved.
EXAMPLE III
The construction and classification effects of the BP-MLL neural network model will be described below, taking the complex fault diagnosis of a rotating machine having a bearing and a gearbox as an example.
TABLE 1
Type (B) Number of
Normal (without failure) 48
Failure and missing teeth of ball bearing 48
Outer bearing failure and missing teeth 48
Outer bearing failure and tooth breakage 48
Inner bearing failure and tooth breakage 48
The training sample set was 240 groups of samples (including 48 non-failures) of 5 failure types, as shown in table 1. The test sample set is again another 240 sets of samples for 5 fault types.
And training an initial neural network model on the training sample set by adopting a BP-MLL improved algorithm, then checking the accuracy of fault detection on the testing sample set, and adjusting the initial neural network model according to the effective classification condition to obtain the BP-MLL neural network model.
By using a personal computer (8G RAM and
Figure BDA0002627136840000161
i5-7200U CPU @2.70GHz), the BP-MLL neural network model can successfully classify the 5 types shown in the table 1Multiple failures, 100% accuracy and 55.3102 seconds total training time. Table 2 shows the configuration of the neural network structure in the BP-MLL neural network model. In Table 2, the learning rate α is an exponentially decaying learning rate, and the learning rate initial value α0Related to round correlation (epochs), and is formulated as 0.95 ═ alphaepochs*α0。
TABLE 2
Item Configuration of
Number of neurons in input layer 32
Number of neurons in hidden layer 44
Number of neurons in output layer 5
Learning rate (alpha) 0.95epochs*α0
Rounds (epochs) 100
Fig. 6 is a schematic diagram of variation of the error value with the number of rounds according to the third embodiment of the present application, and it can be seen from fig. 6 that the error value becomes smaller with the number of rounds, and the error value tends to be stable when the number of rounds reaches 20.
Fig. 7 is a schematic diagram of a classification result of a BP-MLL neural network model provided in the third embodiment of the present application, and fig. 7 is a diagram obtained by reducing an output result of the BP-MLL neural network model from 5 dimensions to 3 dimensions by using a t-distributed random neighborhood embedding (t-SNE) algorithm. As can be seen from FIG. 7, successful classification of a composite fault can be achieved by using the BP-MLL neural network model.
Example four
Fig. 8 is a schematic structural diagram of a composite fault diagnosis device according to the fourth embodiment of the present application, and as shown in fig. 8, the composite fault diagnosis device 10 in this embodiment includes:
an acquisition module 11 and a processing module 12.
The acquisition module 11 is used for acquiring a plurality of paths of working state signals of the rotary machine to be detected;
the processing module 12 is configured to pre-process the multiple paths of working state signals to obtain a feature signal data set; and analyzing the characteristic signal data set according to a back propagation multi-label learning BP-MLL neural network model to determine the composite fault type of the rotary machine to be detected, wherein the BP-MLL neural network model is obtained by training according to a BP-MLL improved algorithm.
Optionally, the BP-MLL neural network model is trained according to a BP-MLL improvement algorithm, and the BP-MLL improvement algorithm is configured to determine a distance function according to values of the sample-dependent label set and values of the sample-independent label set:
Figure BDA0002627136840000171
wherein,
Figure BDA0002627136840000172
Figure BDA0002627136840000173
i denotes a sample number, k denotes a sample number of a correlation tag, l denotes a sample number of an uncorrelated tag,
Figure BDA0002627136840000174
output values representing relevant label position output layer neurons,
Figure BDA0002627136840000175
output values, Y, representing neurons of the output layer at unrelated label positionsiA set of labels associated with the sample is represented,
Figure BDA0002627136840000176
representing a set of labels, | Y, independent of the exemplariL represents the value of the labelset associated with the exemplar,
Figure BDA0002627136840000177
value representing a set of labels independent of the sample, fdistanceRepresenting a distance function;
the BP-MLL refinement algorithm is used to pass an error function:
Figure BDA0002627136840000178
adjusting weight values of a neural network, wherein E represents an error, I represents a logarithm of a sample data set, β represents a regularization coefficient, VtdRepresenting weight values between the t-th hidden layer neuron and the d-th input layer neuron, WstRepresenting the weight value between the s-th output layer neuron and the t-th hidden layer neuron, N representing the number of input layer neurons, M representing the number of hidden layer neurons, and J representing the number of output layer neurons.
Optionally, the obtaining module 11 is further configured to obtain a training sample set and a testing sample set, where data in the training sample set and the testing sample set is obtained by preprocessing a working state signal of a rotary machine acquired in advance;
the processing module 12 is further configured to train the training sample set by using a BP-MLL improved algorithm to obtain an initial neural network model; and testing the initial neural network model according to the test sample set to obtain the BP-MLL neural network model.
Optionally, the processing module 12 is specifically configured to:
extracting the characteristics of the time domain and the frequency domain of the multi-channel working state signal to obtain a time-frequency domain characteristic parameter set of the multi-channel working state signal;
and constructing the characteristic signal data set by a condensation degree evaluation algorithm according to the time-frequency domain characteristic parameter set.
The composite fault diagnosis device provided by the embodiment of the application can execute the composite fault diagnosis method provided by any method embodiment of the application, and has corresponding functional modules and beneficial effects of the execution method. The implementation principle and technical effect of this embodiment are similar to those of the above method embodiments, and are not described in detail here.
EXAMPLE five
Fig. 9 is a schematic structural diagram of an electronic device according to a fifth embodiment of the present application, and as shown in fig. 9, the electronic device 20 includes a memory 21, a processor 22, and a computer program stored in the memory and executable on the processor; the number of the processors 22 of the electronic device 20 may be one or more, and one processor 22 is taken as an example in fig. 9; the processor 22 and the memory 21 in the electronic device 20 may be connected by a bus or other means, and fig. 9 illustrates the connection by the bus as an example.
The memory 21 is a computer-readable storage medium, and can be used for storing software programs, computer-executable programs, and modules, such as program instructions/modules corresponding to the acquisition module 11 and the processing module 12 in the embodiment of the present application. The processor 22 executes various functional applications of the device/terminal/server and data processing by executing software programs, instructions and modules stored in the memory 21, that is, implements the above-described composite fault diagnosis method.
The memory 21 may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required for at least one function; the storage data area may store data created according to the use of the terminal, and the like. Further, the memory 21 may include high speed random access memory, and may also include non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other non-volatile solid state storage device. In some examples, the memory 21 may further include memory located remotely from the processor 22, which may be connected to the device/terminal/server via a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.
EXAMPLE six
Embodiment D of the present application also provides a computer-readable storage medium having stored thereon a computer program for executing a composite fault diagnosis method when executed by a computer processor, the method including:
acquiring a multi-path working state signal of a rotary machine to be detected;
preprocessing the multi-channel working state signals to obtain a characteristic signal data set;
and analyzing the characteristic signal data set according to a back propagation multi-label learning BP-MLL neural network model to determine the composite fault type of the rotary machine to be detected, wherein the BP-MLL neural network model is obtained by training according to a BP-MLL improved algorithm.
Of course, the computer program of the computer-readable storage medium provided in this embodiment of the present application is not limited to the method operations described above, and may also perform related operations in the composite fault diagnosis method provided in any embodiment of the present application.
From the above description of the embodiments, it is obvious for those skilled in the art that the present application can be implemented by software and necessary general hardware, and certainly can be implemented by hardware, but the former is a better embodiment in many cases. Based on such understanding, the technical solutions of the present application may be embodied in the form of a software product, which may be stored in a computer-readable storage medium, such as a floppy disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a FLASH Memory (FLASH), a hard disk or an optical disk of a computer, and includes several instructions for enabling a computer device (which may be a personal computer, a server, or a network device) to execute the methods described in the embodiments of the present application.
It should be noted that, in the embodiment of the above search apparatus, each included unit and module are merely divided according to functional logic, but are not limited to the above division as long as the corresponding functions can be implemented; in addition, specific names of the functional units are only used for distinguishing one functional unit from another, and are not used for limiting the protection scope of the application.
It is to be noted that the foregoing is only illustrative of the preferred embodiments of the present application and the technical principles employed. It will be understood by those skilled in the art that the present application is not limited to the particular embodiments described herein, but is capable of various obvious changes, rearrangements and substitutions as will now become apparent to those skilled in the art without departing from the scope of the application. Therefore, although the present application has been described in more detail with reference to the above embodiments, the present application is not limited to the above embodiments, and may include other equivalent embodiments without departing from the spirit of the present application, and the scope of the present application is determined by the scope of the appended claims.

Claims (10)

1. A composite fault diagnostic method, comprising:
acquiring a multi-path working state signal of a rotary machine to be detected;
preprocessing the multi-channel working state signals to obtain a characteristic signal data set;
and analyzing the characteristic signal data set according to a back propagation multi-label learning BP-MLL neural network model, and determining the composite fault type of the rotary machine to be detected.
2. The method of claim 1, wherein the BP-MLL neural network model is trained according to a BP-MLL refinement algorithm for determining a distance function based on values of the sample-dependent label set and values of the sample-independent label set:
Figure FDA0002627136830000011
wherein,
Figure FDA0002627136830000012
i denotes a sample number, k denotes a sample number of a correlation tag, l denotes a sample number of an uncorrelated tag,
Figure FDA0002627136830000013
output values representing relevant label position output layer neurons,
Figure FDA0002627136830000014
output values, Y, representing neurons of the output layer at unrelated label positionsiA set of labels associated with the sample is represented,
Figure FDA0002627136830000015
representing a set of labels, | Y, independent of the exemplariL represents the value of the labelset associated with the exemplar,
Figure FDA0002627136830000016
value representing a set of labels independent of the sample, fdistanceRepresenting a distance function.
3. The method of claim 2, wherein the BP-MLL refinement algorithm is configured to perform the following steps by an error function:
Figure FDA0002627136830000017
adjusting weight values of a neural network, wherein E represents an error, I represents a logarithm of a sample data set, β represents a regularization coefficient, VtdRepresenting between the t-th hidden layer neuron and the d-th input layer neuronWeight value, WstRepresenting the weight value between the s-th output layer neuron and the t-th hidden layer neuron, N representing the number of input layer neurons, M representing the number of hidden layer neurons, and J representing the number of output layer neurons.
4. The method according to any one of claims 1 to 3, wherein before analyzing the characteristic signal data set according to a back-propagation multi-label learning BP-MLL neural network model and determining the type of the composite fault of the rotary machine to be detected, the method further comprises:
acquiring a training sample set and a testing sample set, wherein data in the training sample set and the testing sample set are obtained by preprocessing a pre-acquired working state signal of a rotary machine;
training the training sample set by using a BP-MLL improved algorithm to obtain an initial neural network model;
and testing the initial neural network model according to the test sample set to obtain the BP-MLL neural network model.
5. The method according to any one of claims 1-3, wherein said preprocessing said plurality of operating condition signals to obtain a signature data set comprises:
extracting the characteristics of the time domain and the frequency domain of the multi-channel working state signal to obtain a time-frequency domain characteristic parameter set of the multi-channel working state signal;
and constructing the characteristic signal data set by a condensation degree evaluation algorithm according to the time-frequency domain characteristic parameter set.
6. A composite failure diagnosis apparatus characterized by comprising:
the acquisition module is used for acquiring a plurality of paths of working state signals of the rotary machine to be detected;
the processing module is used for preprocessing the multi-channel working state signals to obtain a characteristic signal data set; and analyzing the characteristic signal data set according to a back propagation multi-label learning BP-MLL neural network model, and determining the composite fault type of the rotary machine to be detected.
7. The apparatus of claim 6, wherein the BP-MLL neural network model is trained according to a BP-MLL refinement algorithm configured to determine a distance function based on values of the sample-dependent label set and values of the sample-independent label set:
Figure FDA0002627136830000031
wherein,
Figure FDA0002627136830000032
i denotes a sample number, k denotes a sample number of a correlation tag, l denotes a sample number of an uncorrelated tag,
Figure FDA0002627136830000033
output values representing relevant label position output layer neurons,
Figure FDA0002627136830000034
output values, Y, representing neurons of the output layer at unrelated label positionsiA set of labels associated with the sample is represented,
Figure FDA0002627136830000035
representing a set of labels, | Y, independent of the exemplariL represents the value of the labelset associated with the exemplar,
Figure FDA0002627136830000036
value representing a set of labels independent of the sample, fdistanceRepresenting a distance function;
the BP-MLL refinement algorithm is used to pass an error function:
Figure FDA0002627136830000037
adjusting weight values of a neural network, wherein E represents an error, I represents a logarithm of a sample data set, β represents a regularization coefficient, VtdRepresenting weight values between the t-th hidden layer neuron and the d-th input layer neuron, WstRepresenting the weight value between the s-th output layer neuron and the t-th hidden layer neuron, N representing the number of input layer neurons, M representing the number of hidden layer neurons, and J representing the number of output layer neurons.
8. The apparatus according to claim 6 or 7,
the acquisition module is further used for acquiring a training sample set and a test sample set, wherein data in the training sample set and the test sample set are obtained by preprocessing a pre-acquired working state signal of the rotary machine;
the processing module is further configured to train the training sample set by using a BP-MLL improved algorithm to obtain an initial neural network model; and testing the initial neural network model according to the test sample set to obtain the BP-MLL neural network model.
9. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the composite fault diagnosis method according to any one of claims 1 to 5 when executing the program.
10. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out a composite fault diagnosis method according to any one of claims 1 to 5.
CN202010800302.2A 2020-08-11 2020-08-11 Composite fault diagnosis method, device, electronic equipment and storage medium Active CN111967364B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010800302.2A CN111967364B (en) 2020-08-11 2020-08-11 Composite fault diagnosis method, device, electronic equipment and storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010800302.2A CN111967364B (en) 2020-08-11 2020-08-11 Composite fault diagnosis method, device, electronic equipment and storage medium

Publications (2)

Publication Number Publication Date
CN111967364A true CN111967364A (en) 2020-11-20
CN111967364B CN111967364B (en) 2023-11-21

Family

ID=73365453

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010800302.2A Active CN111967364B (en) 2020-08-11 2020-08-11 Composite fault diagnosis method, device, electronic equipment and storage medium

Country Status (1)

Country Link
CN (1) CN111967364B (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113361557A (en) * 2020-12-21 2021-09-07 南京仁智网络科技有限公司 Training method of neural network for underground coal mine fire extinguishing control based on vibration data

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2001001344A2 (en) * 1999-06-26 2001-01-04 Axeon Limited Neural network for performing real-time channel equalisation
US20160086185A1 (en) * 2014-10-15 2016-03-24 Brighterion, Inc. Method of alerting all financial channels about risk in real-time
CN108664924A (en) * 2018-05-10 2018-10-16 东南大学 A kind of multi-tag object identification method based on convolutional neural networks
CN109635677A (en) * 2018-11-23 2019-04-16 华南理工大学 Combined failure diagnostic method and device based on multi-tag classification convolutional neural networks
CN109765054A (en) * 2019-01-22 2019-05-17 上海海事大学 A kind of Fault Diagnosis of Roller Bearings
CN110907176A (en) * 2019-09-30 2020-03-24 合肥工业大学 Wasserstein distance-based fault diagnosis method for deep countermeasure migration network

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2001001344A2 (en) * 1999-06-26 2001-01-04 Axeon Limited Neural network for performing real-time channel equalisation
US20160086185A1 (en) * 2014-10-15 2016-03-24 Brighterion, Inc. Method of alerting all financial channels about risk in real-time
CN108664924A (en) * 2018-05-10 2018-10-16 东南大学 A kind of multi-tag object identification method based on convolutional neural networks
CN109635677A (en) * 2018-11-23 2019-04-16 华南理工大学 Combined failure diagnostic method and device based on multi-tag classification convolutional neural networks
CN109765054A (en) * 2019-01-22 2019-05-17 上海海事大学 A kind of Fault Diagnosis of Roller Bearings
CN110907176A (en) * 2019-09-30 2020-03-24 合肥工业大学 Wasserstein distance-based fault diagnosis method for deep countermeasure migration network

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
YIXUAN ZHANG 等: "An Improved Simultaneous Fault Diagnosis Method based on Cohesion Evaluation and BP-MLL for Rotating Machinery", 《2020 IEEE 29TH INTERNATIONAL SYMPOSIUM ON INDUSTRIAL ELECTRONICS (ISIE)》 *
单东 等: "基于径向基神经网络和正则化极限学习机的多标签学习模型", 《模式识别与人工智能》, pages 833 - 840 *

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113361557A (en) * 2020-12-21 2021-09-07 南京仁智网络科技有限公司 Training method of neural network for underground coal mine fire extinguishing control based on vibration data

Also Published As

Publication number Publication date
CN111967364B (en) 2023-11-21

Similar Documents

Publication Publication Date Title
CN109655259A (en) Combined failure diagnostic method and device based on depth decoupling convolutional neural networks
CN110110768B (en) Rolling bearing fault diagnosis method based on parallel feature learning and multiple classifiers
CN116450399B (en) Fault diagnosis and root cause positioning method for micro service system
CN113111820B (en) Rotary part fault diagnosis method and device based on improved CNN and relation module
CN113485302B (en) Vehicle operation process fault diagnosis method and system based on multivariate time sequence data
CN112729834B (en) Bearing fault diagnosis method, device and system
CN109858352A (en) A kind of method for diagnosing faults based on compressed sensing and the multiple dimensioned network of improvement
CN110823576A (en) Mechanical anomaly detection method based on generation of countermeasure network
CN108919059A (en) A kind of electric network failure diagnosis method, apparatus, equipment and readable storage medium storing program for executing
CN112465124B (en) Twin depth space-time neural network model acquisition/fault diagnosis method and device
CN114091504B (en) Rotary machine small sample fault diagnosis method based on generation countermeasure network
CN110689130B (en) Bearing fault diagnosis method
CN111898644B (en) Intelligent identification method for health state of aerospace liquid engine under fault-free sample
CN108153987A (en) A kind of hydraulic pump Multiple faults diagnosis approach based on the learning machine that transfinites
CN116168720A (en) Motor sound abnormality fault diagnosis method, system and storable medium
CN111967364B (en) Composite fault diagnosis method, device, electronic equipment and storage medium
CN116541771A (en) Unbalanced sample bearing fault diagnosis method based on multi-scale feature fusion
CN116222753A (en) Rotor system fault sensitivity feature extraction method and system
Hao et al. New fusion features convolutional neural network with high generalization ability on rolling bearing fault diagnosis
CN114186593A (en) Bearing fault identification method, device, equipment and storage medium under noise condition
CN116451105A (en) Wind turbine generator system gearbox fault diagnosis method based on ICEEMDAN-MPE-RF and SVM
CN114528906A (en) Fault diagnosis method, device, equipment and medium for rotary machine
Gelman et al. Novel anomaly detection technique based on the nearest neighbour and sequential methods
CN118035873B (en) Fault diagnosis method for parallel convolution improved triple network
Lyu et al. A Novel Fault Diagnosis Method Based on Feature Fusion and Model Agnostic Meta-Learning

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