CN107315892A - A kind of Method for Bearing Fault Diagnosis based on extreme learning machine - Google Patents
A kind of Method for Bearing Fault Diagnosis based on extreme learning machine Download PDFInfo
- Publication number
- CN107315892A CN107315892A CN201710679461.XA CN201710679461A CN107315892A CN 107315892 A CN107315892 A CN 107315892A CN 201710679461 A CN201710679461 A CN 201710679461A CN 107315892 A CN107315892 A CN 107315892A
- Authority
- CN
- China
- Prior art keywords
- msub
- mrow
- msup
- sigma
- connection weight
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 38
- 238000003745 diagnosis Methods 0.000 title claims abstract description 30
- 238000000354 decomposition reaction Methods 0.000 claims abstract description 32
- 238000012549 training Methods 0.000 claims abstract description 30
- 238000003062 neural network model Methods 0.000 claims abstract description 21
- 238000012360 testing method Methods 0.000 claims abstract description 14
- 230000001133 acceleration Effects 0.000 claims abstract description 10
- 210000002569 neuron Anatomy 0.000 claims description 14
- 230000006870 function Effects 0.000 claims description 13
- 239000011159 matrix material Substances 0.000 claims description 12
- 238000013528 artificial neural network Methods 0.000 claims description 11
- 238000005096 rolling process Methods 0.000 claims description 11
- 238000001228 spectrum Methods 0.000 claims description 7
- 230000004913 activation Effects 0.000 claims description 5
- 238000003491 array Methods 0.000 claims description 3
- 238000013135 deep learning Methods 0.000 claims description 3
- TWAWPCGFTAOEND-UHFFFAOYSA-N muk-1 Chemical compound N=1C=2C(C(=O)OC)=CC=CC=2OC=1C(C=1N=2)=CC=CC=1OC=2C1=CC=CC=C1OC TWAWPCGFTAOEND-UHFFFAOYSA-N 0.000 claims description 3
- 238000010586 diagram Methods 0.000 description 11
- 238000001514 detection method Methods 0.000 description 3
- 238000000926 separation method Methods 0.000 description 3
- 238000002474 experimental method Methods 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000010835 comparative analysis Methods 0.000 description 1
- 125000004122 cyclic group Chemical group 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 239000006185 dispersion Substances 0.000 description 1
- 238000005461 lubrication Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000010998 test method Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/10—Geometric CAD
- G06F30/17—Mechanical parametric or variational design
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Geometry (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Health & Medical Sciences (AREA)
- Biophysics (AREA)
- Pure & Applied Mathematics (AREA)
- Mathematical Optimization (AREA)
- Mathematical Analysis (AREA)
- Computational Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Computer Hardware Design (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Measurement Of Mechanical Vibrations Or Ultrasonic Waves (AREA)
Abstract
The invention provides a kind of Method for Bearing Fault Diagnosis based on extreme learning machine, belong to technology for mechanical fault diagnosis field, including IMF modal components are obtained to vibration acceleration signal progress variation mode decomposition；The singular value of IMF modal components is obtained by singular value decomposition svd algorithm；The singular value of IMF modal components is divided into training sample and test sample two parts；Using the singular value of training sample as the input value of extreme learning machine ELM neural network models, input connection weight, bias and the optimal output connection weight of ELM neural network models are determined；Using the singular value of test sample as the input value for the ELM neural network models that input connection weight, bias and optimal output connection weight is determined, output result is bearing failure diagnosis result.The present invention can accurately realize that signal is efficiently separated, and the convergence of component signal mode is fast, robustness is high, and Fault Identification speed is fast, and accuracy rate is high, without setting up model, professional requirement is reduced, suitable for commercial Application.
Description
Technical Field
The invention relates to the technical field of mechanical fault diagnosis, in particular to a bearing fault diagnosis method based on an extreme learning machine.
Background
The rolling bearing has a remarkable feature that the degree of life dispersion is very large. It is not scientific to perform regular maintenance of bearings only rigidly according to the design life. When the bearing is used, working conditions are monitored and faults are judged at any time. Therefore, the working precision of the equipment can be prevented from being reduced, the probability of accidents is reduced, the working potential of the bearing can be exerted to the maximum extent, and the expenditure is saved.
A slightly damaged bearing can be inferred from its true cause of failure from the use, in particular from the signs of wear conditions, wear tracks, etc. of the bearing working surfaces. The severely damaged bearing is the bearing which is completely scrapped due to an accident, the final damaged condition often covers the initial damaged trace, and only the phenomena of final seizure and burning of the bearing and the broken bearing parts are exposed. These causes make it easy to confuse the most important sources of bearing damage, which can only be inferred from the working conditions of the bearing, the lubrication conditions, the overall structure of the bearing and the form of the damage and verified by other scientific analytical methods.
Therefore, the running condition of the important bearing in the mechanical equipment can be accurately and timely known, and the method has very important significance for guaranteeing the normal operation of the mechanical equipment. The method comprises the steps of monitoring vibration of the bearing through a sensor, obtaining a large amount of information of bearing faults, analyzing fault characteristics of the bearing based on the mechanism of the bearing faults, and thus scientifically judging the bearing faults.
Disclosure of Invention
The invention aims to provide a bearing fault diagnosis method based on a limit learning machine, which can accurately realize effective signal separation, has high fault recognition speed and high accuracy, does not need to establish a model, and has quick and convenient diagnosis, so as to solve the problems of untimely diagnosis and low accuracy of establishing a diagnosis model in the background technology.
In order to achieve the purpose, the invention adopts the following technical scheme:
a bearing fault diagnosis method based on an extreme learning machine comprises the steps of obtaining vibration acceleration signals under four working conditions through a vibration acceleration sensor, wherein the four working conditions comprise normal operation, inner ring fault operation, rolling fault operation and outer ring fault operation, and the method comprises the following steps:
step S110: decomposing the vibration acceleration signal through a variational modal decomposition VMD algorithm to obtain K IMF modal components;
step S120: singular values of the IMF modal components are obtained through a Singular Value Decomposition (SVD) algorithm;
step S130: dividing the singular value of the IMF modal component into a training sample and a testing sample;
step S140: taking the singular value of the training sample as the input value of the extreme learning machine ELM neural network model, performing deep learning training, and determining the input connection weight, the offset value and the optimal output connection weight of the ELM neural network model;
step S150: and taking the singular value of the test sample as the input value of the ELM neural network model for determining the input connection weight, the bias value and the optimal output connection weight, and performing learning training, wherein the output result of the network training is the bearing fault diagnosis result.
Further, the obtaining K IMF modal components by the VMD algorithm in step S110 includes:
step S111: for each IMF modal component function μk(t) performing Hilbert transform to obtain an analytic signal of IMF modal component, wherein the expression isWherein σtDenotes a unit pulse function, j ═ 1, 2.... k);
step S112: estimating the center frequency of the analytic signal of each IMF modal componentMixing, modulating the frequency spectrum of each IMF modal component to a corresponding base frequency band,
step S113: calculating the square L of the gradient of the analytic signal of each IMF modal component determining the fundamental frequency band2Norm, and obtaining corresponding IMF modal component expression as
Wherein,denotes a partial derivative, and μ K ═ { μ 1, μ 2.. μ K } denotes K IMF modal components obtained by decomposition, ωkRepresenting the center frequency of the IMF modal components, and f represents the sum of all IMF modal components;
step S114: introducing a secondary penalty factor alpha and a Lagrang multiplication operator lambda to obtain an expanded Lagrange algorithm, wherein the expression is as follows,
step S115: and (4) solving the saddle point of the expanded Lagrange expression by using an alternative direction multiplier Algorithm (ADMM) to obtain K IMF modal components.
Further, the finding of the saddle point of the extended Lagrange expression in step S115 includes,
the method comprises the following steps: initializing muk 1，ωk 1，λ1；
Step two: and (3) executing a loop: n is n + 1;
step three: updating muk:
Updating omegak:
Step four: updating lambda:
step five: repeating the first step to the fourth step until the iteration stop condition is met
And finishing the iteration to obtain the saddle point of the expanded Lagrange expression.
Further, the obtaining the singular value of the IMF modal component by the SVD algorithm in step S120 includes:
constructing a signal data m multiplied by n order matrix H by K IMF modal components
Wherein, U ∈ Rm×mAnd V ∈ Rn×nAre all orthogonal matrices and are all provided with a matrix,Ar＝diag(σ1,σ2,…,σr)，σi(i ═ 1,2, …, r) denotes the singular value of H, and σ denotes1≥…≥σrR is equal to or more than 0 and represents the rank of H, mui、νiAre respectively square arrays HHTAnd HTH ith feature vector.
Further, the determining the input connection weight, the offset value and the optimal output connection weight of the ELM neural network model in step S140 includes:
calculating an output matrix F of an ELM neural network hidden layer, wherein an output expression of the ELM neural network model can be abbreviated as F beta as Y,
wherein, beta represents the output connection weight of hidden layer neurons, L represents the number of ELM neural network hidden layer neurons, N represents the number of training samples, and Y is an expected output value;
determining β a least squares solution, the output formula being:wherein, F+Moore-Penrose generalized inverse, least squares solution representing hidden layer output matrix FI.e. the optimal output connection weight βi；
The output expression of the ELM neural network model is
Wherein x isi(i ═ 1,2, …, N) represents the input vector of singular value components of the training samples, yi(i ═ 1,2, …, N) represents the training sample network output vector, αiIs the input connection weight connecting the i-th hidden layer neuron, biIs the bias value for the ith hidden layer neuron, and G represents the activation function.
The invention has the beneficial effects that: in the VMD decomposition process, the optimal solution of the constraint variation problem is solved through cyclic iteration to determine the frequency center and the bandwidth of the natural modal component obtained through decomposition, so that the effective separation of each frequency component of the signal is realized, and the decomposed signal has the advantages of fast convergence and high robustness; the SVD method is adopted to further extract fault characteristics of the signals obtained by VMD decomposition, so that the purposes of extracting signal essential characteristics and reducing dimension are achieved, the speed and the accuracy of fault identification are improved, the fault detection and identification can be realized without establishing a model, the professional requirements are reduced, and the engineering applicability is increased.
Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the description of the embodiments are briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
Fig. 1 is a flowchart of a fault diagnosis method for a rolling bearing according to an embodiment of the present invention.
Fig. 2 is a diagram of an ELM neural network structure model according to an embodiment of the present invention.
Fig. 3 is a diagram of a vibration signal spectrum obtained by VMD decomposition in a normal operation state according to an embodiment of the present invention.
Fig. 4 is a diagram of a vibration signal spectrum obtained by VMD decomposition in an inner ring fault operation state according to an embodiment of the present invention.
Fig. 5 is a diagram illustrating a vibration signal spectrum obtained by VMD decomposition in a rolling fault operation state according to an embodiment of the present invention.
Fig. 6 is a diagram illustrating a vibration signal spectrum obtained by VMD decomposition in an outer ring fault operation state according to an embodiment of the present invention.
Fig. 7 is a comparison graph of the fault diagnosis result according to the embodiment of the present invention.
Detailed Description
Reference will now be made in detail to embodiments of the present invention, examples of which are illustrated in the accompanying drawings, wherein like reference numerals refer to the same or similar elements or elements having the same or similar function throughout. The embodiments described below with reference to the accompanying drawings are illustrative only for the purpose of explaining the present invention, and are not to be construed as limiting the present invention.
As used herein, the singular forms "a", "an", "the" and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will be further understood that the terms "comprises" and/or "comprising," when used in this specification, specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof. It will be understood that when an element is referred to as being "connected" or "coupled" to another element, it can be directly connected or coupled to the other element or intervening elements may also be present. Further, "connected" or "coupled" as used herein may include wirelessly connected or coupled. As used herein, the term "and/or" includes any and all combinations of one or more of the associated listed items.
It will be understood by those skilled in the art that, unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. It will be further understood that terms, such as those defined in commonly used dictionaries, should be interpreted as having a meaning that is consistent with their meaning in the context of the prior art and will not be interpreted in an idealized or overly formal sense unless expressly so defined herein.
For the convenience of understanding the embodiments of the present invention, the following description will be further explained by taking several specific embodiments as examples in conjunction with the drawings, and the embodiments are not to be construed as limiting the embodiments of the present invention. Fig. 1 is a flowchart of a fault diagnosis method for a rolling bearing according to an embodiment of the present invention; FIG. 2 is a diagram of an ELM neural network structure model according to an embodiment of the present invention; FIG. 3 is a diagram of the vibration signal decomposed by the VMD under normal operation according to the embodiment of the present invention; FIG. 4 is a diagram of a vibration signal decomposed by a VMD under an inner ring faulty operation state according to an embodiment of the present invention; FIG. 5 is a diagram of a vibration signal decomposed by a VMD in a rolling fault operating state according to an embodiment of the present invention; FIG. 6 is a diagram of a vibration signal decomposed by a VMD under an outer ring fault operation state according to an embodiment of the present invention; fig. 7 is a comparison graph of the fault diagnosis result according to the embodiment of the present invention.
Those of ordinary skill in the art will understand that: the figures are merely schematic representations of one embodiment, and the blocks or flow diagrams in the figures are not necessarily required to practice the present invention.
As shown in fig. 1, a bearing fault diagnosis method based on an extreme learning machine according to an embodiment of the present invention includes acquiring a vibration acceleration signal under four operating conditions by a vibration acceleration sensor, where the four operating conditions include normal operation, inner ring fault operation, rolling fault operation, and outer ring fault operation, and the method includes the following steps:
step S110: decomposing the vibration acceleration signal through a variational modal decomposition VMD algorithm to obtain K IMF modal components;
step S120: singular values of the IMF modal components are obtained through a Singular Value Decomposition (SVD) algorithm;
step S130: dividing the singular value of the IMF modal component into a training sample and a testing sample;
step S140: taking the singular value of the training sample as the input value of the extreme learning machine ELM neural network model, performing deep learning training, and determining the input connection weight, the offset value and the optimal output connection weight of the ELM neural network model;
step S150: and taking the singular value of the test sample as the input value of the ELM neural network model for determining the input connection weight, the bias value and the optimal output connection weight, and performing learning training, wherein the output result of the network training is the bearing fault diagnosis result.
In an embodiment of the present invention, the obtaining K IMF modal components by the VMD algorithm includes:
for each IMF modal component function μk(t) performing Hilbert transform to obtain an analytic signal of IMF modal component, wherein the expression isWherein σtDenotes a unit pulse function, j ═ 1, 2.... k);
resolution of each IMF modal componentSignal estimated central frequencyMixing, modulating the frequency spectrum of each IMF modal component to a corresponding base frequency band,
calculating the square L of the gradient of the analytic signal of each IMF modal component determining the fundamental frequency band2Norm, and obtaining corresponding IMF modal component expression as
Wherein,denotes a partial derivative, and μ K ═ { μ 1, μ 2.. μ K } denotes K IMF modal components obtained by decomposition, ωkRepresenting the center frequency of the IMF modal components, and f represents the sum of all IMF modal components;
introducing a secondary penalty factor alpha and a Lagrang multiplication operator lambda to obtain an expanded Lagrange algorithm, wherein the expression is as follows,
and (4) solving the saddle point of the expanded Lagrange expression by using an alternative direction multiplier Algorithm (ADMM) to obtain K IMF modal components.
In one embodiment of the invention, the saddle points of the extended Lagrange expression include,
the method comprises the following steps: initializing muk 1，ωk 1，λ1；
Step two: and (3) executing a loop: n is n + 1;
step three: updating muk:
Updating omegak:
Step four: updating lambda:
step five: repeating the first step to the fourth step until the iteration stop condition is met
And finishing the iteration to obtain the saddle point of the expanded Lagrange expression.
In an embodiment of the present invention, the obtaining the singular values of the IMF modal components by the SVD algorithm includes:
constructing a signal data m multiplied by n order matrix H by K IMF modal components
Wherein, U ∈ Rm×mAnd V ∈ Rn×nAre all orthogonal matrices and are all provided with a matrix,Ar＝diag(σ1,σ2,…,σr)，σi(i ═ 1,2, …, r) denotes the singular value of H, and σ denotes1≥…≥σrR is equal to or more than 0 and represents the rank of H, mui、νiAre respectively square arrays HHTAnd HTH ith feature vector.
In an embodiment of the present invention, the determining the input connection weight, the bias value, and the optimal output connection weight of the ELM neural network model includes:
calculating an output matrix F of an ELM neural network hidden layer, wherein an output expression of the ELM neural network model can be abbreviated as F beta as Y,
wherein, beta represents the output connection weight of hidden layer neuron, L represents the number of ELM neural network hidden layer neuron, N represents the number of training sample, M represents the number of output segment, and Y is the expected output value;
determining β a least squares solution, the output formula being:wherein, F+Moore-Penrose generalized inverse, least squares solution representing hidden layer output matrix FI.e. the optimal output connection weight βi；
The output expression of the ELM neural network model is
Wherein x isi(i ═ 1,2, …, N) represents the input vector of singular value components of the training samples, yi(i ═ 1,2, …, N) represents the training sample network output vector, αiIs the input connection weight connecting the i-th hidden layer neuron, biIs the bias value for the ith hidden layer neuron, and G represents the activation function.
As shown in fig. 2 to 7, the specific test procedures and results of the present invention are as follows:
the rolling bearing fault signal that this experiment adopted washington universities' universities of universities bearing data center to provide verifies. The bearing fault diagnosis method based on VMD, SVD and ELM of the invention is detected and verified by respectively using sample signals under the four states of normal, inner ring fault, outer ring fault and rolling element fault, and comprises the following specific steps:
step one, VMD decomposition is carried out on the bearing vibration signal.
The number of signal samples in the four states is shown in table 1.
TABLE 1 number of samples in four states
| Is normal | Failure of rolling body | Inner ring failure | Outer ring failure | |
| Number of samples | 24 | 12 | 12 | 12 |
The VMD decomposition is performed by predetermining the number K of the IMF modal components obtained by the decomposition and determining K by observing whether the center frequency is over-decomposed. The center frequencies of the IMF modal components obtained corresponding to different K values in four different states are shown in table 2.
TABLE 2 center frequency of each IMF modal component
As can be seen from table 2, in the four different states, when K is 5, the modal components start to appear with similar center frequencies, a phenomenon called over-decomposition, and thus the value K assumes 4.
Substituting K-4 into VMD procedure, the signal decomposition results in four different states are shown in fig. 3, fig. 4, fig. 5, and fig. 6.
Step two, further extracting fault characteristics by singular value decomposition:
singular values obtained by singular value decomposition in four different states are shown in Table 3
Table 3 singular values obtained by SVD decomposition
And step three, carrying out neural network training through an ELM algorithm to realize bearing fault identification and diagnosis.
Singular values obtained through singular value decomposition in four different states are divided into two parts to be respectively used as a training sample and a testing sample, and the two parts are shown in table 4.
TABLE 4 training sample and test sample numbers
The ELM algorithm needs to specify parameters of an activation function and the number of hidden neurons, in the experiment, the activation function G selects a Sigmoidal function, and the number L of the hidden neurons selects 10. The test results are shown in FIG. 7. As can be seen from fig. 7, the actual fault type curve completely coincides with the identified fault type curve, which indicates that the fault diagnosis accuracy of the method is 100%.
In order to illustrate the effectiveness of the method, an SVM model and a BP neural network are introduced for comparative analysis. The test results are shown in table 5.
TABLE 5 test comparison results
| Network type | Number of training samples | Number of samples tested | Run time | Rate of identification accuracy |
| ELM | 40 | 20 | 0.0312 | 100％ |
| SVM | 40 | 20 | 0.658 | 90％ |
| BPNN | 40 | 20 | 0.9414 | 80％ |
According to the test results, the fault diagnosis method based on the ELM is obviously superior to SVM and BP neural networks. The ELM has higher superiority in terms of running time and recognition accuracy.
The method can realize the fault detection and classification of the bearing, has high fault identification success rate and short time consumption, and has obvious practical application value.
In summary, the VMD decomposition algorithm is used to decompose the signal, and in the method, the optimal solution of the constraint variation problem is solved through loop iteration in the decomposition process to determine the frequency center and the bandwidth of the natural modal component obtained through decomposition, so as to realize effective separation of each frequency component of the signal, and compared with the EMD and the LMD, the VMD decomposition signal has the characteristics of fast convergence and high robustness; the SVD method is adopted to further extract fault characteristics of the signals obtained by VMD decomposition, so that the purposes of extracting signal essential characteristics and reducing dimensions are achieved; compared with other fault diagnosis methods, the extreme learning machine algorithm has remarkable superiority in fault identification speed and identification accuracy; the method of the invention can realize the detection and the identification of the fault by using the sample without establishing a model, thereby reducing the professional requirements and increasing the engineering applicability.
From the above description of the embodiments, it is clear to those skilled in the art that the present invention can be implemented by software plus necessary general hardware platform. Based on such understanding, the technical solutions of the present invention may be embodied in the form of a software product, which may be stored in a storage medium, such as ROM/RAM, magnetic disk, optical disk, etc., and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the method according to the embodiments or some parts of the embodiments.
The embodiments in the present specification are described in a progressive manner, and the same and similar parts among the embodiments are referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, for apparatus or system embodiments, since they are substantially similar to method embodiments, they are described in relative terms, as long as they are described in partial descriptions of method embodiments. The above-described embodiments of the apparatus and system are merely illustrative, and the units described as separate parts may or may not be physically separate, and the parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment. One of ordinary skill in the art can understand and implement it without inventive effort.
The above description is only for the preferred embodiment of the present invention, but the scope of the present invention is not limited thereto, and any changes or substitutions that can be easily conceived by those skilled in the art within the technical scope of the present invention are included in the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (5)
1. A bearing fault diagnosis method based on an extreme learning machine comprises the steps of obtaining vibration acceleration signals under four working conditions through a vibration acceleration sensor, wherein the four working conditions comprise normal operation, inner ring fault operation, rolling fault operation and outer ring fault operation, and the method is characterized by comprising the following steps:
step S110: decomposing the vibration acceleration signal through a variational modal decomposition VMD algorithm to obtain K IMF modal components;
step S120: singular values of the IMF modal components are obtained through a Singular Value Decomposition (SVD) algorithm;
step S130: dividing the singular value of the IMF modal component into a training sample and a testing sample;
step S140: taking the singular value of the training sample as the input value of the extreme learning machine ELM neural network model, performing deep learning training, and determining the input connection weight, the offset value and the optimal output connection weight of the ELM neural network model;
step S150: and taking the singular value of the test sample as the input value of the ELM neural network model for determining the input connection weight, the bias value and the optimal output connection weight, and performing learning training, wherein the output result of the network training is the bearing fault diagnosis result.
2. The extreme learning machine-based bearing fault diagnosis method according to claim 1, wherein the obtaining K IMF modal components by the VMD algorithm in step S110 comprises:
step S111: for each IMF modal component function μk(t) performing Hilbert transform to obtain an analytic signal of IMF modal component, wherein the expression isWherein σtDenotes a unit pulse function, j ═ 1, 2.... k);
step S112: estimating the center frequency of the analytic signal of each IMF modal componentMixing, modulating the frequency spectrum of each IMF modal component to a corresponding base frequency band,
<mrow> <mo>&lsqb;</mo> <mrow> <mo>(</mo> <msub> <mi>&sigma;</mi> <mi>t</mi> </msub> <mo>+</mo> <mfrac> <mi>j</mi> <mrow> <mi>&pi;</mi> <mi>t</mi> </mrow> </mfrac> <mo>)</mo> </mrow> <mo>&times;</mo> <msub> <mi>&mu;</mi> <mi>j</mi> </msub> <mrow> <mo>(</mo> <mi>t</mi> <mo>)</mo> </mrow> <mo>&rsqb;</mo> <msup> <mi>e</mi> <mrow> <mo>-</mo> <msub> <mi>jt&omega;</mi> <mi>k</mi> </msub> </mrow> </msup> <mo>;</mo> </mrow>
step S113: calculating the square L of the gradient of the analytic signal of each IMF modal component determining the fundamental frequency band2Norm, and obtaining corresponding IMF modal component expression as
<mrow> <mfenced open = "{" close = ""> <mtable> <mtr> <mtd> <mrow> <munder> <mrow> <mi>m</mi> <mi>i</mi> <mi>n</mi> </mrow> <mrow> <msub> <mi>&mu;</mi> <mi>k</mi> </msub> <mo>,</mo> <msub> <mi>&omega;</mi> <mi>k</mi> </msub> </mrow> </munder> <mo>{</mo> <msub> <mo>&Sigma;</mo> <mi>k</mi> </msub> <mo>|</mo> <mo>|</mo> <msub> <mo>&part;</mo> <mi>t</mi> </msub> <mo>&lsqb;</mo> <mrow> <mo>(</mo> <msub> <mi>&sigma;</mi> <mi>t</mi> </msub> <mo>+</mo> <mfrac> <mi>j</mi> <mrow> <mi>&pi;</mi> <mi>t</mi> </mrow> </mfrac> <mo>)</mo> </mrow> <msub> <mi>&mu;</mi> <mi>k</mi> </msub> <mrow> <mo>(</mo> <mi>t</mi> <mo>)</mo> </mrow> <mo>&rsqb;</mo> <msup> <mi>e</mi> <mrow> <mo>-</mo> <msub> <mi>jt&omega;</mi> <mi>k</mi> </msub> </mrow> </msup> <mo>|</mo> <msubsup> <mo>|</mo> <mn>2</mn> <mn>2</mn> </msubsup> <mo>}</mo> </mrow> </mtd> </mtr> <mtr> <mtd> <mrow> <mi>s</mi> <mo>.</mo> <mi>t</mi> <munder> <mo>&Sigma;</mo> <mi>k</mi> </munder> <msub> <mi>&mu;</mi> <mi>k</mi> </msub> <mo>=</mo> <mi>f</mi> </mrow> </mtd> </mtr> </mtable> </mfenced> <mo>,</mo> </mrow>
Wherein,denotes a partial derivative, and μ K ═ { μ 1, μ 2.. μ K } denotes K IMF modal components obtained by decomposition, ωkRepresenting the center frequency of the IMF modal components, and f represents the sum of all IMF modal components;
step S114: introducing a secondary penalty factor alpha and a Lagrang multiplication operator lambda to obtain an expanded Lagrange algorithm, wherein the expression is as follows,
<mrow> <mi>L</mi> <mrow> <mo>(</mo> <msub> <mi>&mu;</mi> <mi>k</mi> </msub> <mo>,</mo> <msub> <mi>&omega;</mi> <mi>k</mi> </msub> <mo>,</mo> <mi>&lambda;</mi> <mo>)</mo> </mrow> <mo>=</mo> <mi>&alpha;</mi> <munder> <mo>&Sigma;</mo> <mi>k</mi> </munder> <mo>|</mo> <mo>|</mo> <msub> <mo>&part;</mo> <mi>t</mi> </msub> <mo>&lsqb;</mo> <mrow> <mo>(</mo> <msub> <mi>&sigma;</mi> <mi>t</mi> </msub> <mo>+</mo> <mfrac> <mi>j</mi> <mrow> <mi>&pi;</mi> <mi>t</mi> </mrow> </mfrac> <mo>)</mo> </mrow> <msub> <mi>&mu;</mi> <mi>k</mi> </msub> <mrow> <mo>(</mo> <mi>t</mi> <mo>)</mo> </mrow> <mo>&rsqb;</mo> <msup> <mi>e</mi> <mrow> <mo>-</mo> <msub> <mi>jt&omega;</mi> <mi>k</mi> </msub> </mrow> </msup> <mo>|</mo> <msubsup> <mo>|</mo> <mn>2</mn> <mn>2</mn> </msubsup> <mo>+</mo> <mo>|</mo> <mo>|</mo> <mi>f</mi> <mo>-</mo> <mo>&Sigma;</mo> <msub> <mi>&mu;</mi> <mi>k</mi> </msub> <mo>|</mo> <msubsup> <mo>|</mo> <mn>2</mn> <mn>2</mn> </msubsup> <mo>+</mo> <mo><</mo> <mi>&lambda;</mi> <mo>,</mo> <mi>f</mi> <mo>-</mo> <msub> <mi>&Sigma;&mu;</mi> <mi>k</mi> </msub> <mo>></mo> </mrow>
step S115: and (4) solving the saddle point of the expanded Lagrange expression by using an alternative direction multiplier Algorithm (ADMM) to obtain K IMF modal components.
3. The extreme learning machine-based bearing fault diagnosis method according to claim 2, wherein the finding of the saddle point of the extended Lagrange expression in the step S115 comprises,
the method comprises the following steps: initializing muk 1，ωk 1，λ1；
Step two: and (3) executing a loop: n is n + 1;
step three: updating muk:
Updating omegak:
Step four: updating lambda:
step five: repeating the first step to the fourth step until the iteration stop condition is met
<mrow> <munder> <mo>&Sigma;</mo> <mi>k</mi> </munder> <mo>|</mo> <mo>|</mo> <msup> <msub> <mi>&mu;</mi> <mi>k</mi> </msub> <mrow> <mi>n</mi> <mo>+</mo> <mn>1</mn> </mrow> </msup> <mo>-</mo> <msup> <msub> <mi>&mu;</mi> <mi>k</mi> </msub> <mi>n</mi> </msup> <mo>|</mo> <msubsup> <mo>|</mo> <mn>2</mn> <mn>2</mn> </msubsup> <mo>/</mo> <mo>|</mo> <mo>|</mo> <msup> <msub> <mi>&mu;</mi> <mi>k</mi> </msub> <mi>n</mi> </msup> <mo>|</mo> <msubsup> <mo>|</mo> <mn>2</mn> <mn>2</mn> </msubsup> <mo><</mo> <mi>&epsiv;</mi> <mo>,</mo> </mrow>
And finishing the iteration to obtain the saddle point of the expanded Lagrange expression.
4. The extreme learning machine-based bearing fault diagnosis method according to claim 1, wherein the step S120 of obtaining the singular values of the IMF modal components through the SVD algorithm comprises:
constructing a signal data m multiplied by n order matrix H by K IMF modal components
<mrow> <mi>H</mi> <mo>=</mo> <msup> <mi>UAV</mi> <mi>T</mi> </msup> <mo>=</mo> <munder> <mo>&Sigma;</mo> <mrow> <mi>i</mi> <mo>=</mo> <mn>1</mn> </mrow> </munder> <msub> <mi>&sigma;</mi> <mi>i</mi> </msub> <msub> <mi>&mu;</mi> <mi>i</mi> </msub> <msup> <msub> <mi>v</mi> <mi>i</mi> </msub> <mi>T</mi> </msup> <mo>=</mo> <munder> <mo>&Sigma;</mo> <mrow> <mi>i</mi> <mo>=</mo> <mn>1</mn> </mrow> </munder> <msub> <mi>&sigma;</mi> <mi>i</mi> </msub> <msub> <mi>B</mi> <mi>i</mi> </msub> <mo>,</mo> </mrow>
Wherein, U ∈ Rm×mAnd V ∈ Rn×nAre all orthogonal matrices and are all provided with a matrix,Ar＝diag(σ1,σ2,…,σr)，σi(i ═ 1,2, …, r) denotes the singular value of H, and σ denotes1≥…≥σrR is equal to or more than 0 and represents the rank of H, mui、viAre respectively square arrays HHTAnd HTH ith feature vector.
5. The extreme learning machine-based bearing fault diagnosis method according to claim 4, wherein the determining the input connection weight, the offset value and the optimal output connection weight of the ELM neural network model in step S140 comprises:
calculating an output matrix F of an ELM neural network hidden layer, wherein an output expression of the ELM neural network model can be abbreviated as F beta as Y,
wherein, beta represents the output connection weight of hidden layer neurons, L represents the number of ELM neural network hidden layer neurons, N represents the number of training samples, and Y is an expected output value;
determining β a least squares solution, the output formula being:wherein, F+Moore-Penrose generalized inverse, least squares solution representing hidden layer output matrix FI.e. the optimal output connection weight βi；
The output expression of the ELM neural network model is
<mrow> <msub> <mi>y</mi> <mi>i</mi> </msub> <mo>=</mo> <munderover> <mo>&Sigma;</mo> <mrow> <mi>i</mi> <mo>=</mo> <mn>1</mn> </mrow> <mi>L</mi> </munderover> <msub> <mi>&beta;</mi> <mi>i</mi> </msub> <mi>G</mi> <mrow> <mo>(</mo> <msub> <mi>&alpha;</mi> <mi>i</mi> </msub> <mo>,</mo> <msub> <mi>b</mi> <mi>i</mi> </msub> <mo>,</mo> <msub> <mi>x</mi> <mi>i</mi> </msub> <mo>)</mo> </mrow> <mo>,</mo> </mrow>
Wherein x isi(i ═ 1,2, …, N) represents the input vector of singular value components of the training samples, yi(i ═ 1,2, …, N) represents the training sample network output vector, αiIs the input connection weight to connect the ith hidden layer neuron,biis the bias value for the ith hidden layer neuron, and G represents the activation function.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710679461.XA CN107315892A (en) | 2017-08-10 | 2017-08-10 | A kind of Method for Bearing Fault Diagnosis based on extreme learning machine |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710679461.XA CN107315892A (en) | 2017-08-10 | 2017-08-10 | A kind of Method for Bearing Fault Diagnosis based on extreme learning machine |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107315892A true CN107315892A (en) | 2017-11-03 |
Family
ID=60175181
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710679461.XA Pending CN107315892A (en) | 2017-08-10 | 2017-08-10 | A kind of Method for Bearing Fault Diagnosis based on extreme learning machine |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107315892A (en) |
Cited By (26)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107832525A (en) * | 2017-11-07 | 2018-03-23 | 昆明理工大学 | A kind of method and its application of information entropy optimization VMD extractions bearing fault characteristics frequency |
| CN108171263A (en) * | 2017-12-26 | 2018-06-15 | 合肥工业大学 | Based on the Fault Diagnosis of Roller Bearings for improving variation mode decomposition and extreme learning machine |
| CN108197648A (en) * | 2017-12-28 | 2018-06-22 | 华中科技大学 | A kind of Fault Diagnosis Method of Hydro-generating Unit and system based on LSTM deep learning models |
| CN108647807A (en) * | 2018-04-10 | 2018-10-12 | 武汉理工大学 | The prediction technique of river discharge |
| CN108869145A (en) * | 2018-04-26 | 2018-11-23 | 中国水利水电科学研究院 | Pumping plant unit diagnostic method based on compound characteristics index and depth limit learning machine |
| CN109100146A (en) * | 2018-08-29 | 2018-12-28 | 北京信息科技大学 | For detecting the method and device of motor bearings failure |
| CN109145999A (en) * | 2018-09-06 | 2019-01-04 | 福州大学 | A kind of panel switches vibration signal fault signature extracts and discrimination method |
| CN109187025A (en) * | 2018-09-19 | 2019-01-11 | 哈尔滨理工大学 | A kind of rolling bearing remaining life prediction technique of integrated KELM |
| CN109443752A (en) * | 2018-10-17 | 2019-03-08 | 北京信息科技大学 | A kind of Gearbox vibration signal noise reduction and method for diagnosing faults based on VMD |
| CN109540522A (en) * | 2018-11-16 | 2019-03-29 | 北京航空航天大学 | Bearing health quantifies modeling method, device and server |
| CN109977920A (en) * | 2019-04-11 | 2019-07-05 | 福州大学 | Fault Diagnosis of Hydro-generator Set method based on time-frequency spectrum and convolutional neural networks |
| CN110161125A (en) * | 2019-06-17 | 2019-08-23 | 哈尔滨工业大学 | The Aeroengine Smart monitoring method combined based on acceleration with sound emission cognition technology |
| CN110286286A (en) * | 2019-05-30 | 2019-09-27 | 昆明理工大学 | A kind of VSC-HVDC converter station fault identification device and method based on VMD-ELM |
| CN110780343A (en) * | 2020-01-02 | 2020-02-11 | 四川大学 | Automatic microseismic signal identification method based on waveform frequency band characteristics |
| CN110907177A (en) * | 2019-12-10 | 2020-03-24 | 北京交通大学 | Bearing fault diagnosis method based on layered extreme learning machine |
| CN111190429A (en) * | 2020-01-13 | 2020-05-22 | 南京航空航天大学 | Unmanned aerial vehicle active fault-tolerant control method based on reinforcement learning |
| CN111795826A (en) * | 2020-06-29 | 2020-10-20 | 南京航空航天大学 | Fault diagnosis method for abnormal oil injection of small two-stroke piston engine |
| CN111879514A (en) * | 2020-07-31 | 2020-11-03 | 南京机电职业技术学院 | Brushless direct current motor bearing fault diagnosis method based on ELM model |
| CN112051480A (en) * | 2020-08-06 | 2020-12-08 | 国网安徽省电力有限公司电力科学研究院 | Neural network power distribution network fault diagnosis method and system based on variational modal decomposition |
| CN112100767A (en) * | 2020-09-02 | 2020-12-18 | 西北工业大学 | Aero-engine service life prediction method based on singular value decomposition and GRU |
| CN112798280A (en) * | 2021-02-05 | 2021-05-14 | 山东大学 | Rolling bearing fault diagnosis method and system |
| CN113340625A (en) * | 2021-04-21 | 2021-09-03 | 北京交通大学 | Bogie fault diagnosis method |
| CN113435255A (en) * | 2021-06-01 | 2021-09-24 | 西北工业大学 | Engine fault diagnosis method based on feature fusion |
| CN113740671A (en) * | 2020-10-22 | 2021-12-03 | 国网河北省电力有限公司雄安新区供电公司 | Fault arc identification method based on VMD and ELM |
| CN114675547A (en) * | 2022-05-30 | 2022-06-28 | 华中科技大学 | MIMO active vibration reduction control method and system with deep learning automatic diagnosis mechanism |
| CN116703624A (en) * | 2023-05-22 | 2023-09-05 | 淮阴工学院 | Block chain-based enterprise operation diagnosis system and method |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103048137A (en) * | 2012-12-20 | 2013-04-17 | 北京航空航天大学 | Fault diagnosis method of rolling bearing under variable working conditions |
| CN104408302A (en) * | 2014-11-19 | 2015-03-11 | 北京航空航天大学 | Bearing variable-condition fault diagnosis method based on LMD-SVD (Local Mean Decomposition-Singular Value Decomposition) and extreme learning machine |
| CN104794368A (en) * | 2015-05-15 | 2015-07-22 | 哈尔滨理工大学 | Rolling bearing fault classifying method based on FOA-MKSVM (fruit fly optimization algorithm-multiple kernel support vector machine) |
| CN105740619A (en) * | 2016-01-28 | 2016-07-06 | 华南理工大学 | On-line fault diagnosis method of weighted extreme learning machine sewage treatment on the basis of kernel function |
| CN105758644A (en) * | 2016-05-16 | 2016-07-13 | 上海电力学院 | Rolling bearing fault diagnosis method based on variation mode decomposition and permutation entropy |
-
2017
- 2017-08-10 CN CN201710679461.XA patent/CN107315892A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103048137A (en) * | 2012-12-20 | 2013-04-17 | 北京航空航天大学 | Fault diagnosis method of rolling bearing under variable working conditions |
| CN104408302A (en) * | 2014-11-19 | 2015-03-11 | 北京航空航天大学 | Bearing variable-condition fault diagnosis method based on LMD-SVD (Local Mean Decomposition-Singular Value Decomposition) and extreme learning machine |
| CN104794368A (en) * | 2015-05-15 | 2015-07-22 | 哈尔滨理工大学 | Rolling bearing fault classifying method based on FOA-MKSVM (fruit fly optimization algorithm-multiple kernel support vector machine) |
| CN105740619A (en) * | 2016-01-28 | 2016-07-06 | 华南理工大学 | On-line fault diagnosis method of weighted extreme learning machine sewage treatment on the basis of kernel function |
| CN105758644A (en) * | 2016-05-16 | 2016-07-13 | 上海电力学院 | Rolling bearing fault diagnosis method based on variation mode decomposition and permutation entropy |
Non-Patent Citations (2)
| Title |
|---|
| XIN YU ET AL.: ""A novel roller bearing fault diagnosis method based on the wavelet extreme learning machine"", 《2017 PROGNOSTICS AND SYSTEM HEALTH MANAGEMENT CONFERENCE(PHM-HARBIN)》 * |
| 秦波 等: ""基于VMD与分层极限学习机的滚动轴承故障诊断方法研究"", 《组合机床与自动化加工技术》 * |
Cited By (35)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107832525B (en) * | 2017-11-07 | 2021-01-05 | 昆明理工大学 | Method for extracting bearing fault characteristic frequency by optimizing VMD (variable minimum vector machine) through information entropy and application of method |
| CN107832525A (en) * | 2017-11-07 | 2018-03-23 | 昆明理工大学 | A kind of method and its application of information entropy optimization VMD extractions bearing fault characteristics frequency |
| CN108171263A (en) * | 2017-12-26 | 2018-06-15 | 合肥工业大学 | Based on the Fault Diagnosis of Roller Bearings for improving variation mode decomposition and extreme learning machine |
| CN108197648A (en) * | 2017-12-28 | 2018-06-22 | 华中科技大学 | A kind of Fault Diagnosis Method of Hydro-generating Unit and system based on LSTM deep learning models |
| CN108197648B (en) * | 2017-12-28 | 2020-06-05 | 华中科技大学 | Hydroelectric generating set fault diagnosis method and system based on LSTM deep learning model |
| CN108647807A (en) * | 2018-04-10 | 2018-10-12 | 武汉理工大学 | The prediction technique of river discharge |
| CN108647807B (en) * | 2018-04-10 | 2021-08-13 | 武汉理工大学 | River flow prediction method |
| CN108869145A (en) * | 2018-04-26 | 2018-11-23 | 中国水利水电科学研究院 | Pumping plant unit diagnostic method based on compound characteristics index and depth limit learning machine |
| CN109100146A (en) * | 2018-08-29 | 2018-12-28 | 北京信息科技大学 | For detecting the method and device of motor bearings failure |
| CN109145999A (en) * | 2018-09-06 | 2019-01-04 | 福州大学 | A kind of panel switches vibration signal fault signature extracts and discrimination method |
| CN109187025B (en) * | 2018-09-19 | 2020-04-07 | 哈尔滨理工大学 | Method for predicting residual service life of rolling bearing integrated with KELM |
| CN109187025A (en) * | 2018-09-19 | 2019-01-11 | 哈尔滨理工大学 | A kind of rolling bearing remaining life prediction technique of integrated KELM |
| CN109443752A (en) * | 2018-10-17 | 2019-03-08 | 北京信息科技大学 | A kind of Gearbox vibration signal noise reduction and method for diagnosing faults based on VMD |
| CN109540522A (en) * | 2018-11-16 | 2019-03-29 | 北京航空航天大学 | Bearing health quantifies modeling method, device and server |
| CN109977920A (en) * | 2019-04-11 | 2019-07-05 | 福州大学 | Fault Diagnosis of Hydro-generator Set method based on time-frequency spectrum and convolutional neural networks |
| CN109977920B (en) * | 2019-04-11 | 2022-06-07 | 福州大学 | Water turbine set fault diagnosis method based on time-frequency spectrogram and convolutional neural network |
| CN110286286A (en) * | 2019-05-30 | 2019-09-27 | 昆明理工大学 | A kind of VSC-HVDC converter station fault identification device and method based on VMD-ELM |
| CN110286286B (en) * | 2019-05-30 | 2021-04-30 | 昆明理工大学 | VSC-HVDC converter station fault identification device and method based on VMD-ELM |
| CN110161125A (en) * | 2019-06-17 | 2019-08-23 | 哈尔滨工业大学 | The Aeroengine Smart monitoring method combined based on acceleration with sound emission cognition technology |
| CN110907177A (en) * | 2019-12-10 | 2020-03-24 | 北京交通大学 | Bearing fault diagnosis method based on layered extreme learning machine |
| CN110780343A (en) * | 2020-01-02 | 2020-02-11 | 四川大学 | Automatic microseismic signal identification method based on waveform frequency band characteristics |
| CN111190429A (en) * | 2020-01-13 | 2020-05-22 | 南京航空航天大学 | Unmanned aerial vehicle active fault-tolerant control method based on reinforcement learning |
| CN111190429B (en) * | 2020-01-13 | 2022-03-18 | 南京航空航天大学 | Unmanned aerial vehicle active fault-tolerant control method based on reinforcement learning |
| CN111795826B (en) * | 2020-06-29 | 2021-12-21 | 南京航空航天大学 | Fault diagnosis method for abnormal oil injection of small two-stroke piston engine |
| CN111795826A (en) * | 2020-06-29 | 2020-10-20 | 南京航空航天大学 | Fault diagnosis method for abnormal oil injection of small two-stroke piston engine |
| CN111879514A (en) * | 2020-07-31 | 2020-11-03 | 南京机电职业技术学院 | Brushless direct current motor bearing fault diagnosis method based on ELM model |
| CN112051480A (en) * | 2020-08-06 | 2020-12-08 | 国网安徽省电力有限公司电力科学研究院 | Neural network power distribution network fault diagnosis method and system based on variational modal decomposition |
| CN112100767A (en) * | 2020-09-02 | 2020-12-18 | 西北工业大学 | Aero-engine service life prediction method based on singular value decomposition and GRU |
| CN113740671A (en) * | 2020-10-22 | 2021-12-03 | 国网河北省电力有限公司雄安新区供电公司 | Fault arc identification method based on VMD and ELM |
| CN112798280B (en) * | 2021-02-05 | 2022-01-04 | 山东大学 | Rolling bearing fault diagnosis method and system |
| CN112798280A (en) * | 2021-02-05 | 2021-05-14 | 山东大学 | Rolling bearing fault diagnosis method and system |
| CN113340625A (en) * | 2021-04-21 | 2021-09-03 | 北京交通大学 | Bogie fault diagnosis method |
| CN113435255A (en) * | 2021-06-01 | 2021-09-24 | 西北工业大学 | Engine fault diagnosis method based on feature fusion |
| CN114675547A (en) * | 2022-05-30 | 2022-06-28 | 华中科技大学 | MIMO active vibration reduction control method and system with deep learning automatic diagnosis mechanism |
| CN116703624A (en) * | 2023-05-22 | 2023-09-05 | 淮阴工学院 | Block chain-based enterprise operation diagnosis system and method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107315892A (en) | A kind of Method for Bearing Fault Diagnosis based on extreme learning machine | |
| Schlegel et al. | Towards a rigorous evaluation of XAI methods on time series | |
| Liu et al. | An audio-based fault diagnosis method for quadrotors using convolutional neural network and transfer learning | |
| Su et al. | A novel method based on deep transfer unsupervised learning network for bearing fault diagnosis under variable working condition of unequal quantity | |
| CN107066759B (en) | Steam turbine rotor vibration fault diagnosis method and device | |
| US11709979B1 (en) | Bridge damage identification method considering uncertainty | |
| CN111651937A (en) | Method for diagnosing similar self-adaptive bearing fault under variable working conditions | |
| CN113111820B (en) | Rotary part fault diagnosis method and device based on improved CNN and relation module | |
| Guo et al. | Deep convolution feature learning for health indicator construction of bearings | |
| CN107766668B (en) | Complex simulation model verification method based on neural network | |
| CN105606914A (en) | IWO-ELM-based Aviation power converter fault diagnosis method | |
| Huang et al. | Machine fault detection for intelligent self-driving networks | |
| CN103412557A (en) | Industrial fault detection and diagnostic method suitable for nonlinear process on-line monitoring | |
| Lu et al. | Fault detection for hydraulic pump based on chaotic parallel RBF network | |
| Cheng et al. | A Koopman operator approach for machinery health monitoring and prediction with noisy and low-dimensional industrial time series | |
| CN113792758A (en) | Rolling bearing fault diagnosis method based on self-supervision learning and clustering | |
| CN104634603A (en) | Early fault diagnosis method for complex equipment | |
| CN112149884A (en) | Academic early warning monitoring method for large-scale students | |
| Ding et al. | A zero-shot soft sensor modeling approach using adversarial learning for robustness against sensor fault | |
| CN115935277A (en) | Supercharged boiler fault diagnosis method based on small sample learning and training method and testing method of fault diagnosis model | |
| CN114742122A (en) | Equipment fault diagnosis method and device, electronic equipment and storage medium | |
| CN114358085A (en) | Line fault diagnosis method and device based on heterogeneous model fusion | |
| Youn et al. | A generic Bayesian framework for real-time prognostics and health management (PHM) | |
| Jin et al. | Transfer learning with quantile regression | |
| CN113627539A (en) | Rolling bearing composite fault diagnosis method and system, storage medium and computing equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20171103 |
|
| RJ01 | Rejection of invention patent application after publication |