CN114781450A - Train rolling bearing state identification method based on parameter optimization MOMEDA-MIA-CNN - Google Patents
Train rolling bearing state identification method based on parameter optimization MOMEDA-MIA-CNN Download PDFInfo
- Publication number
- CN114781450A CN114781450A CN202210433462.7A CN202210433462A CN114781450A CN 114781450 A CN114781450 A CN 114781450A CN 202210433462 A CN202210433462 A CN 202210433462A CN 114781450 A CN114781450 A CN 114781450A
- Authority
- CN
- China
- Prior art keywords
- rolling bearing
- train
- mia
- cnn
- momeda
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000005096 rolling process Methods 0.000 title claims abstract description 108
- 238000000034 method Methods 0.000 title claims abstract description 30
- 238000005457 optimization Methods 0.000 title claims abstract description 25
- 230000009467 reduction Effects 0.000 claims abstract description 11
- 238000007781 pre-processing Methods 0.000 claims abstract description 5
- 238000001914 filtration Methods 0.000 claims description 26
- 230000001133 acceleration Effects 0.000 claims description 13
- 230000006870 function Effects 0.000 claims description 13
- 238000012549 training Methods 0.000 claims description 12
- 238000002474 experimental method Methods 0.000 claims description 7
- 239000002245 particle Substances 0.000 claims description 6
- 230000003595 spectral effect Effects 0.000 claims description 6
- 238000012360 testing method Methods 0.000 claims description 6
- 238000004458 analytical method Methods 0.000 claims description 5
- 238000001228 spectrum Methods 0.000 claims description 4
- ORILYTVJVMAKLC-UHFFFAOYSA-N Adamantane Natural products C1C(C2)CC3CC1CC2C3 ORILYTVJVMAKLC-UHFFFAOYSA-N 0.000 claims description 3
- 230000004913 activation Effects 0.000 claims description 3
- 238000004364 calculation method Methods 0.000 claims description 3
- 238000013316 zoning Methods 0.000 claims description 2
- 230000005540 biological transmission Effects 0.000 abstract description 5
- 238000000605 extraction Methods 0.000 abstract description 4
- 238000013527 convolutional neural network Methods 0.000 description 17
- 230000008569 process Effects 0.000 description 5
- 230000000694 effects Effects 0.000 description 4
- 230000003044 adaptive effect Effects 0.000 description 2
- 230000004075 alteration Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000011176 pooling Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000008054 signal transmission Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
- 230000003313 weakening effect Effects 0.000 description 1
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
- G06F2218/12—Classification; Matching
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/02—Preprocessing
- G06F2218/04—Denoising
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Biophysics (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Computational Linguistics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Signal Processing (AREA)
- Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)
- Train Traffic Observation, Control, And Security (AREA)
Abstract
The invention relates to the technical field of running state recognition of train key components, and discloses a train rolling bearing state recognition method based on parameter optimization MOMEDA-MIA-CNN, which comprises the following steps: step (1), acquiring train rolling bearing data; step (2), optimizing MOMEDA parameters; step (3), preprocessing data; step (4), carrying out data modal compartmentalization by train rolling bearing; step (5), establishing an initial model of a rolling bearing MIA-CNN of the train; step (6), optimizing a train rolling bearing MIA-CNN model; and (7) identifying the running state of the rolling bearing of the train. The MIA-CNN model is used for carrying out self-adaptive feature extraction and state recognition on the interval signals subjected to noise reduction, end-to-end transmission of data is achieved, and human influence is reduced, so that the train rolling bearing operation state recognition result has good reliability.
Description
Technical Field
The invention relates to the technical field of running state recognition of train key components, in particular to a train rolling bearing state recognition method based on parameter optimization MOMEDA-MIA-CNN.
Background
The rapid development of rail transit brings infinite convenience to the lives of the public, and meanwhile, the safety of trains is concerned more and more. The rolling bearing is one of the most important parts in a transmission system as a key part of a bogie member and an axle, and the timely identification of the running state of the rolling bearing has important significance for ensuring the safe and stable running of a train and the life safety of train personnel.
Because the train rolling bearing bears huge load for a long time in the actual operation process of the train, and the complex and severe operation environment with high speed and high temperature in which the bearing is positioned makes the train bearing easily break down, and the rolling bearing is usually integrated in other mechanical equipment, the collected vibration signals contain a large amount of interference of impact noise and environmental noise besides the vibration signals of the bearing, and the interference brings great difficulty for accurately identifying the rolling bearing carrying state. A multipoint optimal adaptive minimum entropy deconvolution algorithm (MOMEDA) is a novel non-iterative blind deconvolution enhancing technology, the MOMEDA does not need to set iteration times in advance, the maximum multipoint kurtosis is used as a target function, an optimal deconvolution filter is found through a matrix operation method, a fault pulse signal is extracted from an actually measured rolling bearing vibration signal quickly and efficiently, and therefore noise reduction of the actually measured signal is achieved. Modal Interval Analysis (MIA), one of the effective tools to solve the uncertainty problem, converts the uncertainty into a modal interval form and quantifies the uncertainty by the width of the modal interval. The data are converted into a modal interval form, so that the information can be more abundant, and the reliability is higher. Convolutional Neural Networks (CNNs) are one of the typical deep learning algorithms, and generally include convolutional layers, pooled layers, fully-connected layers, and the like. The CNN convolutional layer can learn and extract key features in signals, the pooling layer screens fault features extracted by the convolutional layer, useless features are filtered, dimension reduction of the fault features is achieved, the fully-connected layer integrates the extracted features, and the softmax classifier identifies the operation state of equipment. The CNN (MIA-CNN) based on MIA has the same structure and working principle with the CNN, data characteristics and the like are converted into a mode of a mode interval from an original determined value with a mode interval theory in the process of establishing, optimizing and identifying a model, the mode interval comprises all possibilities of a real value of the data, and a final output interval also comprises all possible values of similarity with the corresponding model, so that some uncertain factors caused by noise interference, a transfer path and the like can be eliminated. The MOMEDA and the MIA-CNN are combined to identify the running state of the train rolling bearing, so that the reliability of the train rolling bearing identification result is greatly improved.
Disclosure of Invention
The invention aims to provide a train rolling bearing state identification method based on parameter optimization MOMEDA-MIA-CNN combination, aiming at the problem that the running state identification rate is not high due to the fact that a train rolling bearing vibration signal is influenced by a transmission path and background noise and fault characteristics are weak. Determining an optimal filtering period T and an optimal filter length L of a Multipoint Optimal Minimum Entropy Deconvolution (MOMEDA) by utilizing a particle swarm optimization algorithm, carrying out MOMEDA noise reduction on the collected train rolling bearing vibration signals by utilizing the obtained MOMEDA optimal parameters, converting the noise-reduced signals into a modal interval form, inputting a convolutional neural network model (MIA-CNN) based on modal interval analysis, and completing the extraction of the fault characteristics of the train rolling bearing and the identification of the running state.
In order to achieve the purpose, the invention provides the following technical scheme: a train rolling bearing state identification method based on parameter optimization MOMEDA-MIA-CNN comprises the following steps:
step (1), acquiring data of a train rolling bearing;
step (2), optimizing MOMEDA parameters;
optimizing a filtering period T and a filter length L of the MOMEDA algorithm by using the kurtosis spectral product as an optimization index and applying a particle swarm optimization algorithm, selecting the filtering period and the filter length when the kurtosis spectral product is maximum as optimal filtering parameters, and recording the optimal filtering parameters as Tu and Lu respectively;
step (3), preprocessing data;
carrying out MOMEDA noise reduction on the acquired train rolling bearing vibration signal data set { x (t) }accordingto the optimal filtering parameters Tu and Lu obtained in the step (2), reducing noise interference, and forming a train rolling bearing operation data set { y (t) } after noise reduction;
step (4), carrying out data modal compartmentalization by train rolling bearing;
converting the preprocessed train rolling bearing data set (y (t)) into a modal interval form y (t) according to an error theory and a modal interval theory,
to increase the reliability of the data set to be analyzed; whereiny(t) represents the lower bound of the rolling bearing operating data set,
representing the upper bound of the rolling bearing dataset.
Step (5), establishing an MIA-CNN initial model of a train rolling bearing;
dividing rolling bearing types according to the rolling bearing running states of the train, constructing an initial MIA-CNN model by combining a noise-reduced modal interval train rolling bearing data set, and coding the rolling bearing running state types of the train to form an ideal output target of the MIA-CNN model;
step (6), optimizing a rolling bearing MIA-CNN model of the train;
according to the CNN model training method, using part of each state of the train rolling axle bearing data set in the state interval in the step (4) as a training sample, and iteratively updating the weight and the offset of the MIA-CNN key point by adopting an Adam iterative updating algorithm; obtaining a loss function value and an accuracy rate of each iteration of training sample data through forward propagation calculation, updating model parameters through backward propagation, and finishing optimization of a MIA-CNN model of the train rolling bearing when the loss function value of the training sample data is low and the accuracy rate is high and an expected requirement is met;
step (7), identifying the running state of a train rolling bearing;
and (4) taking a part of the remaining signals of the train rolling bearing signals in the modal interval in the step (4) as test samples, inputting the test samples into an optimized train rolling bearing MIA-CNN model, calculating an output result of the MIA-CNN model, comparing the obtained result by using a modal interval size comparison rule, and obtaining the running state corresponding to the code corresponding to the maximum value of the modal interval, namely the train rolling bearing running state.
Preferably, in the step (1), a train running part experiment platform is built for the train rolling bearing, and the experiment platform comprises two one-way acceleration sensors, a data acquisition card, a three-phase asynchronous motor and a PC (personal computer), wherein the one-way acceleration sensors acquire running data of the rolling shaft of the train running part, the two one-way acceleration sensors are respectively installed in the horizontal direction and the vertical direction of an end cover of the experiment platform rolling bearing, vibration signals in the horizontal direction and the vertical direction acquired by the two one-way acceleration sensors are subjected to time domain analysis, the vibration signals with large amplitude are selected as an original signal data set for the train rolling bearing to operate, and the original signal data set is recorded as { x (t) }, wherein t is time.
Preferably, in step (2), the product of kurtosis and the peak factor of the envelope spectrum is taken as the kurtosis product.
Preferably, in step (4), the mathematical definition of the modal interval z is:
i.e. modal interval z is bounded by a pair of real numbersz，
Is defined in whichzThe lower bound is represented by the number of bits,
representing the upper bound, kR represents the set of real numbers for the modal closed space, where z is not subject to
The size of the upper and lower boundaries in the modal interval is not limited, and the theoretical basis of the mathematical operation of the modal interval is Kaucher algorithm.
Preferably, in the step (5), the initial parameter of MIA-CNN is initially set empirically, which includes: activation function, loss function, learning rate, batch size, optimizer, number of iterations, and Dropout layer.
The beneficial effects of the invention are as follows:
compared with the existing train rolling bearing state identification method, the invention provides the method that the MOMEDA algorithm is used for weakening the interference generated by a transmission path, background noise and the like, the filtering period and the filtering length of the MOMEDA algorithm are optimized through the particle swarm optimization algorithm, the vibration signal of the train rolling bearing is recovered from the noise signal to the greatest extent, meanwhile, the incompleteness of filtering is considered, the modal interval method is used for carrying out modal compartmentalization on the filtered signal, the uncertainty problem in the vibration signal of the train rolling bearing is further processed, and the MIA-CNN model is used for automatically completing the extraction of the fault characteristics of the train rolling bearing and the identification of the running state; because the method not only carries out filtering and denoising on the original signal, but also secondarily processes the uncertainty problem of the train rolling bearing data set through a modal interval method, the reliability of the state identification result is improved, and the method has the following advantages:
1. the two acceleration sensors are adopted to obtain the running data of the train rolling bearing, and the data with larger amplitude is selected as rolling bearing carrying data, so that the data source is more reliable;
2. the particle swarm optimization algorithm is adopted to optimize the parameters of the MOMEDA, so that the defect that the main influence parameters of the MOMEDA depend on manual experience selection is overcome, and the MOMEDA filtering effect is better;
3. the filtered train rolling bearing signals are converted into a modal interval form, and the problem of uncertainty in the input of a train rolling bearing state identification model can be effectively solved by secondarily processing the uncertainty problem existing in the signal acquisition process through a modal interval theory;
in conclusion, the MIA-CNN model is used for carrying out adaptive feature extraction and state recognition on the noise-reduced interval signals, end-to-end transmission of data is achieved, and human influence is reduced, so that the train rolling bearing operation state recognition result has good reliability.
Drawings
FIG. 1 is a flow chart of an embodiment of a train rolling bearing state identification method based on parameter optimization MOMEDA-MIA-CNN.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
In an embodiment, referring to fig. 1, the present invention provides a technical solution: a train rolling bearing state identification based on parameter optimization MOMEDA-MIA-CNN specifically comprises the following steps:
step (1), acquiring train rolling bearing data;
aiming at a train rolling bearing, a train running part experiment platform is built and comprises two one-way acceleration sensors PCB 352C33, a data acquisition card NI PXI-1042, a three-phase asynchronous motor, a PC (personal computer) and the like, wherein the one-way acceleration sensors PCB 352C33 acquire rolling bearing carrying data of the train running part, and the two one-way acceleration sensors PCB 352C33 are respectively installed in the horizontal direction and the vertical direction of an end cover of the experiment platform rolling bearing; performing time domain analysis on vibration signals in the horizontal direction and the vertical direction acquired by the two unidirectional acceleration sensors, selecting a vibration signal with large amplitude as an original signal data set for the operation of a train rolling bearing, and recording as { x (t) }, wherein t is time;
step (2), optimizing MOMEDA parameters;
the parameters influencing the filtering effect of the MOMEDA mainly comprise: a filter period T and a filter length L; when the set filtering period T is inconsistent with the actual filtering period, the obtained filtering result is very likely to be completely wrong, and the error is often hard to perceive; the length L of the filter is in direct proportion to the filtering effect of the MOMEDA algorithm, but when the length of the filter is too large, part of effective information of an original signal can be filtered;
therefore, the filtering period T and the filter length L are reasonably selected, and the decisive influence is exerted on the noise reduction effect of the MOMEDA;
the kurtosis is a time domain index capable of representing the impulse magnitude of a signal, the envelope spectrum peak factor is a frequency domain index capable of representing the periodicity of the signal and representing the impulse magnitude of the signal, the kurtosis product is a product of the kurtosis and the envelope spectrum peak factor, and when the kurtosis product is larger, the periodicity of the signal is more obvious and the impulse is larger; optimizing a filtering period T and a filter length L of an MOMEDA algorithm by using a particle swarm optimization algorithm by taking a kurtosis spectral product as an optimization index, taking T and L when the kurtosis spectral product is maximum as optimal filtering parameters, and respectively recording the optimal filtering parameters as Tu and Lu;
step (3), preprocessing data;
carrying out MOMEDA noise reduction on the acquired train rolling bearing vibration signal data set { x (t) }accordingto the optimal filtering parameters Tu and Lu obtained in the step (2), reducing noise interference, and forming a train rolling bearing operation data set { y (t) } after noise reduction;
step (4), carrying out data modal zoning on a rolling axle of the train;
considering the uncertainty problem in the vibration signal transmission path and the data preprocessing process of the train rolling bearing, converting a preprocessed train rolling bearing data set (y (t) into a modal interval form y (t)) according to a modal interval theory and an error theory,
to increase the reliability of the signal to be analyzed; whereiny(t) represents the lower bound of the rolling bearing operating data set,
representing a rolling bearing dataset upper bound;
the mathematical definition of the modal interval z is:
i.e. the modal interval z is defined by a pair of real numbersz，
Is defined in whichzThe lower bound is represented by the number of bits,
representing the upper bound, kR represents the set of real numbers for the modal closed interval, where z is independent of
The size of the upper and lower boundaries in the modal interval is not limited, and the theoretical basis of the mathematical operation of the modal interval is Kaucher algorithm.
Step (5), establishing an MIA-CNN initial model of a train rolling bearing;
constructing an initial MIA-CNN model according to the train running state expression type and input noise-reduced modal interval train rolling bearing signals, and coding the train running state expression type to form an ideal output target of the MIA-CNN model;
in the embodiment, the train rolling bearing operation states are preferably 5, namely normal, inner ring fault, outer ring fault, retainer fault and ball fault; sensitive characteristic data of each operation state is processed into a vector form of 1024 x 1 and input into the convolutional neural network model, 5 train rolling bearing operation states are coded at the same time to form an ideal output target of the convolutional neural network model,
wherein the normal code is ([1,1], [0,0], [0,0], [0,0], [0,0 ]);
the inner ring fault codes are ([0,0], [1,1], [0,0], [0,0], [0,0 ]);
the outer ring fault codes are ([0,0], [0,0], [1,1], [0,0], [0,0 ]);
the cage fault code is ([0,0], [0,0], [0,0], [1,1], [0,0 ]);
the ball fault codes are ([0,0], [0,0], [0,0], [0,0], [1,1 ]);
the initial parameters of the convolutional neural network are initially set according to experience, and the initial parameters comprise: an activation function, a loss function, a learning rate, a batch size, an optimizer, an iteration number, a Dropout layer and the like;
step (6), optimizing a rolling bearing MIA-CNN model of the train;
according to the CNN model training method, a part of each state of the train rolling axis carrying dataset in the state interval in the step (4) is used as a training sample, and the Adam iterative update algorithm is adopted to iteratively update the weight and the offset of the MIA-CNN key points; obtaining a loss function value and an accuracy rate of each iteration of training sample data through forward propagation calculation, updating model parameters through backward propagation, and finishing optimization of the MIA-CNN model of the train rolling bearing when the loss function value of the training sample data is low and the accuracy rate is high and an expected requirement is met;
step (7), identifying the running state of a rolling bearing of the train;
and (4) taking a part of the remaining signals of the train rolling bearing signals in the modal interval in the step (4) as test samples, inputting the test samples into an optimized train rolling bearing MIA-CNN model, calculating an output result of the MIA-CNN model, comparing the obtained result by using a modal interval size comparison rule, and obtaining the running state corresponding to the maximum value of the modal interval corresponding to the code, namely the train rolling bearing running state.
Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and alterations can be made in these embodiments without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.
Claims (5)
1. A train rolling bearing state identification method based on parameter optimization MOMEDA-MIA-CNN is characterized by comprising the following steps:
step (1), acquiring train rolling bearing data;
step (2), optimizing MOMEDA parameters;
optimizing a filtering period T and a filter length L of the MOMEDA algorithm by using the kurtosis spectral product as an optimization index and applying a particle swarm optimization algorithm, selecting the filtering period and the filter length when the kurtosis spectral product is maximum as optimal filtering parameters, and recording the optimal filtering parameters as Tu and Lu respectively;
step (3), preprocessing data;
carrying out MOMEDA noise reduction on the acquired train rolling bearing vibration signal data set { x (t) }accordingto the optimal filtering parameters Tu and Lu obtained in the step (2), reducing noise interference, and forming a train rolling bearing operation data set { y (t) } after noise reduction;
step (4), carrying out data modal zoning on a rolling axle of the train;
converting the preprocessed train rolling bearing data set (y (t)) into a modal interval form y (t) according to an error theory and a modal interval theory,
to increase the reliability of the data set to be analyzed; whereiny(t) represents the lower bound of the rolling bearing operating data set,
representing a rolling bearing dataset upper bound;
step (5), establishing an MIA-CNN initial model of a train rolling bearing;
dividing rolling bearing classes according to the rolling bearing operating state of the train, constructing an initial MIA-CNN model by combining a noise-reduced modal interval-based rolling bearing operating data set, and coding the rolling bearing operating state classes of the train to form an ideal output target of the MIA-CNN model;
step (6), optimizing a rolling bearing MIA-CNN model of the train;
according to the CNN model training method, using part of each state of the train rolling axle bearing data set in the state interval in the step (4) as a training sample, and iteratively updating the weight and the offset of the MIA-CNN key point by adopting an Adam iterative updating algorithm; obtaining a loss function value and an accuracy rate of each iteration of training sample data through forward propagation calculation, updating model parameters through backward propagation, and finishing optimization of a MIA-CNN model of the train rolling bearing when the loss function value of the training sample data is low and the accuracy rate is high and an expected requirement is met;
step (7), identifying the running state of a rolling bearing of the train;
and (5) taking a part of the rest signals of the train rolling bearing signals with the modal interval in the step (4) as test samples, inputting the test samples into an optimized train rolling bearing MIA-CNN model, calculating an output result of the MIA-CNN model, comparing the obtained result by using a modal interval size comparison rule, and determining the running state corresponding to the maximum value of the modal interval corresponding to the code as the train rolling bearing carrying running state.
2. The method for identifying the state of the train rolling bearing based on the parameter optimization MOMEDA-MIA-CNN as claimed in claim 1, wherein: in the step (1), a train running part experiment platform is built for a train rolling bearing, and the experiment platform comprises two one-way acceleration sensors, a data acquisition card, a three-phase asynchronous motor and a PC (personal computer), wherein the one-way acceleration sensors acquire running shaft carrying data of the train running part, the two one-way acceleration sensors are respectively installed in the horizontal direction and the vertical direction of an end cover of the train rolling bearing, vibration signals in the horizontal direction and the vertical direction acquired by the two one-way acceleration sensors are subjected to time domain analysis, the vibration signals with large amplitude are selected as an original signal data set for the train rolling bearing to operate, and the original signal data set is marked as { x (t) }, wherein t is time.
3. The method for identifying the state of the train rolling bearing based on the parameter optimization MOMEDA-MIA-CNN as claimed in claim 1, wherein: in the step (2), the product of the kurtosis and the peak factor of the envelope spectrum is used as the kurtosis product.
4. The method for identifying the state of the train rolling bearing based on the parameter optimization MOMEDA-MIA-CNN as claimed in claim 1, wherein the method comprises the following steps: in the step (4), the mathematical definition of the modal interval z is as follows:
i.e. modal interval z is bounded by a pair of real numbersz，
Is defined in whichzThe lower bound is represented by the number of bits,
representing the upper bound, kR represents the set of real numbers for the modal closed interval, where z is independent of
The size of the upper and lower boundaries in the modal interval is not limited, and the theoretical basis of the mathematical operation of the modal interval is Kaucher algorithm.
5. The method for identifying the state of the train rolling bearing based on the parameter optimization MOMEDA-MIA-CNN as claimed in claim 1, wherein the method comprises the following steps: in the step (5), the initial parameter of the MIA-CNN is initially set according to experience, and the method comprises the following steps: activation function, loss function, learning rate, batch size, optimizer, number of iterations, and Dropout layer.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210433462.7A CN114781450B (en) | 2022-04-24 | 2022-04-24 | Train rolling bearing state identification method based on parameter optimization MOMEDA-MIA-CNN |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210433462.7A CN114781450B (en) | 2022-04-24 | 2022-04-24 | Train rolling bearing state identification method based on parameter optimization MOMEDA-MIA-CNN |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114781450A true CN114781450A (en) | 2022-07-22 |
| CN114781450B CN114781450B (en) | 2024-03-29 |
Family
ID=82433397
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210433462.7A Active CN114781450B (en) | 2022-04-24 | 2022-04-24 | Train rolling bearing state identification method based on parameter optimization MOMEDA-MIA-CNN |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114781450B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115859091A (en) * | 2022-11-01 | 2023-03-28 | 哈尔滨工业大学 | Bearing fault feature extraction method, electronic device and storage medium |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104502126A (en) * | 2014-12-28 | 2015-04-08 | 华东交通大学 | Modal intervals-based high-speed train bogie fault diagnosis method |
| CN108168886A (en) * | 2017-12-22 | 2018-06-15 | 合肥工业大学 | Rolling Bearing Fault Character extracts and method for monitoring operation states |
| CN109635334A (en) * | 2018-11-12 | 2019-04-16 | 武汉科技大学 | Fault Diagnosis of Roller Bearings, system and medium based on particle group optimizing |
| US20200200648A1 (en) * | 2018-02-12 | 2020-06-25 | Dalian University Of Technology | Method for Fault Diagnosis of an Aero-engine Rolling Bearing Based on Random Forest of Power Spectrum Entropy |
| WO2021128510A1 (en) * | 2019-12-27 | 2021-07-01 | 江苏科技大学 | Bearing defect identification method based on sdae and improved gwo-svm |
| WO2021135630A1 (en) * | 2019-12-31 | 2021-07-08 | 福州大学 | Rolling bearing fault diagnosis method based on grcmse and manifold learning |
| WO2021143128A1 (en) * | 2020-01-16 | 2021-07-22 | 佛山科学技术学院 | Deep learning-based bearing state identification method and system |
| CN113743010A (en) * | 2021-08-31 | 2021-12-03 | 三峡大学 | Rolling bearing running state evaluation method based on EEMD energy entropy |
| CN113919220A (en) * | 2021-10-08 | 2022-01-11 | 西安科技大学 | Intelligent fault diagnosis method for rolling bearing driven by vibration image |
-
2022
- 2022-04-24 CN CN202210433462.7A patent/CN114781450B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104502126A (en) * | 2014-12-28 | 2015-04-08 | 华东交通大学 | Modal intervals-based high-speed train bogie fault diagnosis method |
| CN108168886A (en) * | 2017-12-22 | 2018-06-15 | 合肥工业大学 | Rolling Bearing Fault Character extracts and method for monitoring operation states |
| US20200200648A1 (en) * | 2018-02-12 | 2020-06-25 | Dalian University Of Technology | Method for Fault Diagnosis of an Aero-engine Rolling Bearing Based on Random Forest of Power Spectrum Entropy |
| CN109635334A (en) * | 2018-11-12 | 2019-04-16 | 武汉科技大学 | Fault Diagnosis of Roller Bearings, system and medium based on particle group optimizing |
| WO2021128510A1 (en) * | 2019-12-27 | 2021-07-01 | 江苏科技大学 | Bearing defect identification method based on sdae and improved gwo-svm |
| WO2021135630A1 (en) * | 2019-12-31 | 2021-07-08 | 福州大学 | Rolling bearing fault diagnosis method based on grcmse and manifold learning |
| WO2021143128A1 (en) * | 2020-01-16 | 2021-07-22 | 佛山科学技术学院 | Deep learning-based bearing state identification method and system |
| CN113743010A (en) * | 2021-08-31 | 2021-12-03 | 三峡大学 | Rolling bearing running state evaluation method based on EEMD energy entropy |
| CN113919220A (en) * | 2021-10-08 | 2022-01-11 | 西安科技大学 | Intelligent fault diagnosis method for rolling bearing driven by vibration image |
Non-Patent Citations (2)
| Title |
|---|
| 张龙;蔡秉桓;吴佳敏;熊国良;: "滚动轴承故障特征自适应增强的相关峭度联合降噪方法", 机械设计与研究, no. 02, 20 April 2020 (2020-04-20) * |
| 李长青;林建辉;胡永旭;: "优化参数VMD和MED在列车齿轮箱滚动轴承故障诊断中的应用", 机车电传动, no. 03, 10 May 2020 (2020-05-10) * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115859091A (en) * | 2022-11-01 | 2023-03-28 | 哈尔滨工业大学 | Bearing fault feature extraction method, electronic device and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114781450B (en) | 2024-03-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109102005B (en) | Small sample deep learning method based on shallow model knowledge migration | |
| CN111504676B (en) | Equipment fault diagnosis method, device and system based on multi-source monitoring data fusion | |
| CN107657250B (en) | Bearing fault detection and positioning method and detection and positioning model implementation system and method | |
| CN112784965A (en) | Large-scale multi-element time series data abnormity detection method oriented to cloud environment | |
| CN111914873A (en) | Two-stage cloud server unsupervised anomaly prediction method | |
| CN106124212A (en) | Based on sparse coding device and the Fault Diagnosis of Roller Bearings of support vector machine | |
| CN104502126B (en) | A kind of high-speed train bogie method for diagnosing faults based on Modal Intervals | |
| CN110766192A (en) | Drilling well leakage prediction system and method based on deep learning | |
| CN111507046B (en) | Method and system for predicting remaining service life of electric gate valve | |
| CN114978613B (en) | Network intrusion detection method based on data enhancement and self-supervision feature enhancement | |
| CN117421684B (en) | Abnormal data monitoring and analyzing method based on data mining and neural network | |
| CN113779882A (en) | Method, device, equipment and storage medium for predicting residual service life of equipment | |
| WO2023231374A1 (en) | Semi-supervised fault detection and analysis method and apparatus for mechanical device, terminal, and medium | |
| CN116415200A (en) | Abnormal vehicle track abnormality detection method and system based on deep learning | |
| CN115169430A (en) | Cloud network end resource multidimensional time sequence anomaly detection method based on multi-scale decoding | |
| CN114781450A (en) | Train rolling bearing state identification method based on parameter optimization MOMEDA-MIA-CNN | |
| CN112560948A (en) | Eye fundus map classification method and imaging method under data deviation | |
| CN115290326A (en) | Rolling bearing fault intelligent diagnosis method | |
| CN116702090A (en) | Multi-mode data fusion and uncertain estimation water level prediction method and system | |
| CN116992757A (en) | Wellhead pressure prediction method and device based on deep learning and rolling optimization | |
| CN115883424B (en) | Method and system for predicting flow data between high-speed backbone networks | |
| CN116467652A (en) | Bearing fault prediction method and system based on convolution noise reduction self-encoder | |
| CN111858343A (en) | Countermeasure sample generation method based on attack capability | |
| CN111639609A (en) | Intelligent identification system for metal fracture types based on machine vision and deep learning | |
| CN116205135A (en) | SO based on data decomposition and neural network 2 Emission prediction method and system |
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 |