CN111898686A - Bearing fault identification method based on gated cyclic unit network - Google Patents
Bearing fault identification method based on gated cyclic unit network Download PDFInfo
- Publication number
- CN111898686A CN111898686A CN202010769848.6A CN202010769848A CN111898686A CN 111898686 A CN111898686 A CN 111898686A CN 202010769848 A CN202010769848 A CN 202010769848A CN 111898686 A CN111898686 A CN 111898686A
- Authority
- CN
- China
- Prior art keywords
- data
- model
- network
- weight
- unit network
- 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 30
- 125000004122 cyclic group Chemical group 0.000 title claims description 12
- 238000012360 testing method Methods 0.000 claims abstract description 22
- 238000012549 training Methods 0.000 claims abstract description 22
- 238000012795 verification Methods 0.000 claims abstract description 14
- 238000003745 diagnosis Methods 0.000 claims abstract description 10
- 230000001133 acceleration Effects 0.000 claims abstract description 6
- 230000000694 effects Effects 0.000 claims abstract description 5
- 210000002569 neuron Anatomy 0.000 claims description 8
- 238000013145 classification model Methods 0.000 claims description 6
- 238000007781 pre-processing Methods 0.000 claims description 6
- 238000000638 solvent extraction Methods 0.000 claims description 6
- 238000001514 detection method Methods 0.000 claims description 4
- 230000002159 abnormal effect Effects 0.000 claims description 3
- 238000010892 electric spark Methods 0.000 claims description 3
- 238000011156 evaluation Methods 0.000 claims description 3
- 238000002474 experimental method Methods 0.000 claims description 3
- 238000003754 machining Methods 0.000 claims description 3
- 239000011159 matrix material Substances 0.000 claims description 3
- 238000012545 processing Methods 0.000 claims description 3
- 238000010200 validation analysis Methods 0.000 claims description 3
- 238000013461 design Methods 0.000 abstract description 3
- 230000006870 function Effects 0.000 description 6
- 238000013528 artificial neural network Methods 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 3
- 230000005540 biological transmission Effects 0.000 description 2
- 210000004027 cell Anatomy 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000001413 cellular effect Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000008034 disappearance Effects 0.000 description 1
- 238000009776 industrial production Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 230000015654 memory Effects 0.000 description 1
- 230000000306 recurrent effect Effects 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 230000006403 short-term memory Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M13/00—Testing of machine parts
- G01M13/04—Bearings
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M13/00—Testing of machine parts
- G01M13/04—Bearings
- G01M13/045—Acoustic or vibration analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- 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
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/048—Activation functions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Molecular Biology (AREA)
- Computational Linguistics (AREA)
- Software Systems (AREA)
- Mathematical Physics (AREA)
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computing Systems (AREA)
- General Health & Medical Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Acoustics & Sound (AREA)
- Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)
Abstract
The invention designs a high-precision identification method aiming at the current situation that bearing faults are difficult to diagnose. According to the method, the time sequence data of the historical acceleration sensor is processed by innovatively applying a gating cycle unit-based network, so that the fault can be predicted and diagnosed. The method was set up by the Keras tool of PYTHON3.6.5, comprising the following steps: s1, dividing the time sequence data of the acceleration sensor into a training set (containing T group data), a verification set and a test set; s2, inputting the data of the training set into the gated circulation unit network for training until the network converges; s3, optimizing the parameters of the network on the verification set; s4, using the trained model for evaluating the model on the test set to obtain the effect of high accuracy; s5, model saving is applied to the actual situation. The method provides a new solution for the diagnosis of the rotary machine, and is further widely applied to the field of diagnosis of mechanical equipment.
Description
[ technical field ] A method for producing a semiconductor device
The invention relates to a bearing fault identification method based on a gated cycle unit network, and belongs to the field of fault diagnosis of rotary machinery.
[ background of the invention ]
Rotating machines are one of the most common types of machinery used in industrial production and life, and therefore fault detection of rotating machinery is of great significance in practical application situations. As one of the important parts of automobile construction, bearings have a significant impact on the load-bearing capacity, the operating performance and the life of the automobile. But traditional methods tend to rely on traditional methods of manually extracting features, which would require a large amount of a priori knowledge. Because the manual feature extraction process is separated from the classification process, the influence of the feature extraction process on the classification process is difficult to judge, and the effect is to be further improved.
Common methods are largely divided into theoretical methods and statistical-based methods. Statistical-based classification methods are increasingly gaining attention because they do not require excessive prior knowledge and can efficiently and accurately obtain mapping relationships between inputs and outputs, such as artificial neural networks, clustering, regression, and the like. The neural network algorithm can realize automatic identification without extracting fault features in advance. In addition, long short term memory network (LSTM) based methods are also applied in the field of bearing fault identification.
In order to solve the gradient disappearance problem of the traditional recurrent neural network, the LSTM can continuously realize the transmission of timing information and the updating of the memory unit through three gates (an input gate, an output gate, and a forgetting gate). The GRU is an improved variant of the LSTM three-gate design, which can efficiently process long and short term timing data and obtain useful information. In contrast to LSTM, GRU has no cellular state and can directly use hidden states for information transmission. Therefore, the GRU has simpler structure, fewer parameters and faster training speed. In recent years, deep learning has shown great advantages in classification tasks, and with the help of this, the applicant proposes a bearing fault identification method based on a gated cycle unit network to solve the above problems, so as to improve the identification accuracy of bearing faults.
[ summary of the invention ]
Aiming at the defects in the prior art, the invention designs a bearing fault identification method based on a gated cyclic unit network. The technical system disclosed by the invention can well utilize the data of the acceleration sensor to realize the diagnosis of the bearing fault and is beneficial to the safe operation of the rotating machinery.
The technical scheme adopted for achieving the purpose of the invention is that a bearing fault identification method based on a gated cyclic unit network mainly comprises the following steps:
step 1: preprocessing is carried out after data acquisition to obtain time sequence data of the acceleration sensor, and a data set is divided into a training set, a verification set and a test set;
step 2: inputting data of a training set into a gated cyclic unit network for training until the network converges;
and step 3: inputting the data of the verification set into a network for verification, and adjusting the parameters of the network to finally obtain the optimal parameters;
and 4, step 4: and storing the final model, inputting the test set for recognition effect test, and using the final model in an actual rotary machine diagnosis link.
Further, the main steps of step 1 are as follows:
step 1.1: data acquisition: an SKF6205 bearing is adopted in an experiment, faults are introduced through electric spark machining, the fault diameters are 0.18mm, 0.36mm and 0.53mm respectively, and 4 kinds of vibration signal data of a normal state, a ball fault, an inner ring fault and an outer ring fault are collected;
step 1.2: data preprocessing: carrying out mean value missing value filling on the index historical time sequence data, and obtaining a final historical time sequence data set after processing and standardizing abnormal values;
step 1.3: data set partitioning: partitioning a historical time series data set into: the training set contains 1900 pieces of data, the validation set contains 500 pieces of data, and the test set contains 500 pieces of data.
Further, the main steps of step 2 are as follows:
step 2.1: and constructing a gating cycle unit network, wherein the gating cycle unit network comprises an input layer, a gating cycle unit layer, a full connection layer and an output layer. The parameters to be adjusted are the number of gating cycle units of each layer, a Dropout value, the selection of an optimizer and the number of neurons of a full connection layer;
step 2.2: initializing the hidden state of a gating cycle unit, and inputting data;
step 2.3: calculating reset gate r of current neurontUpdate gate ztWeight and alternative hidden state ofComprises the following steps:
zt=σ(Wxzxt+Whzht-1+bz)
rt=σ(Wxrxt+Whrht-1+br)
wherein, WhzRepresenting the weight between the last hidden state and the update gate, WxzRepresenting the weight between the current input and the update gate, WhrRepresenting the weight between the last hidden state and the reset gate, WxrRepresenting the weight between the current input and the reset gate, WxhRepresenting the weight between the current input and the current candidate hidden state, and W is a weight matrix; bzIs to update the bias term of the gate, brIs the bias term for the reset gate; "σ (-) denotes a sigmoid function, and" tanh (-) denotes a hyperbolic tangent function; "+" indicates point-by-point multiplication;
step 2.4: computing the hidden state h of the current neurontAnd passed on to the next neuron;
step 2.5: the training loop is cycled through the above steps until the model converges, using Multi-category entry as the loss function.
Further, the main steps of step 3 are as follows:
step 3.1: testing the trained model on the verification set;
step 3.2: adjusting parameters according to the model result;
step 3.3: and (5) completing parameter adjustment and saving the model.
Further, the main steps of step 4 are as follows:
step 4.1: using the test set to test the finally trained classification model;
step 4.2: comparing all the prediction results with the true values to obtain a final model evaluation index;
step 4.3: comparing the obtained result with other methods to obtain the superiority of the model;
step 4.4: the final classification model can be used for a fault detection link of an actual rotating machine.
The invention has the beneficial effects that: the technical scheme of the invention can well utilize the data of the acceleration sensor to realize the diagnosis of the bearing fault and is beneficial to the safe operation of the rotating machinery.
[ description of the drawings ]
FIG. 1 is a flow chart of the present invention;
FIG. 2 is a diagram of a gated loop cell network;
FIG. 3 is a GRU model training loss value;
FIG. 4 is GRU model accuracy.
[ detailed description ] embodiments
The present invention is further illustrated by the following examples. Referring to FIG. 1, which is a flow chart of the present invention, a gated loop cell network structure is shown in FIG. 2. The implementation mode mainly comprises the following steps:
step 1: preprocessing is carried out after data acquisition to obtain sensor time sequence data, and a data set is divided into a training set, a verification set and a test set;
step 2: inputting data of a training set into a gated cyclic unit network for training until the network converges;
and step 3: inputting the data of the verification set into a network for verification, and adjusting the parameters of the network to finally obtain the optimal parameters;
and 4, step 4: and storing the final model, inputting the test set for recognition effect test, and using the final model for an actual bearing fault diagnosis link.
The step 1 comprises the following steps:
step 1.1: data acquisition: the implementation method of the patent comprises the following steps of: the implementation method of the patent comprises the following steps of: the SKF6205 bearing is adopted in the experiment, faults are introduced through electric spark machining, and the fault diameters are 0.18mm, 0.36mm and 0.53mm respectively. 4 kinds of vibration signal data of a normal state, a ball fault, an inner ring fault and an outer ring fault are collected;
step 1.2: data preprocessing: and (4) carrying out mean value missing value filling on the index historical time sequence data, and processing and standardizing abnormal values to obtain a final historical time sequence data set.
Step 1.3: data set partitioning: partitioning a historical time series data set into: the training set contains 1900 pieces of data, the validation set contains 500 pieces of data, and the test set contains 500 pieces of data.
The step 2 comprises the following steps:
step 2.1: and constructing a gating cycle unit network, wherein the gating cycle unit network comprises an input layer, a gating cycle unit layer, a full connection layer and an output layer. The parameters to be adjusted are the number of gating cycle units of each layer, a Dropout value, the selection of an optimizer and the number of neurons of a full connection layer;
step 2.2: and initializing the hidden state of the gating cycle unit and inputting data. The dimension of each sample is 1 × 1600, the data dimension input to the GRU is 160 × 10, i.e. 160 time steps, the input dimension for each time step is 10. Training internal parameters of the model by using back propagation;
step 2.3: calculating the reset gate, the weight of the update gate and the alternative hidden state of the current neuron; inputting x of current neurontAnd hidden state of previous neuronht-1When inputting, update the gate ztAnd a reset gate rtIs calculated as:
zt=σ(Wxzxt+Whzht-1+bz)
rt=σ(Wxrxt+Whrht-1+br)
wherein, WhzRepresenting the weight between the last hidden state and the update gate, WxzRepresenting the weight between the current input and the update gate, WhrRepresenting the weight between the last hidden state and the reset gate, WxrRepresenting the weight between the current input and the reset gate; bzIs to update the bias term of the gate, brIs the bias term for the reset gate; "σ (-) denotes a sigmoid function.
Step 2.4: calculating the hidden state of the current neuron and transmitting the hidden state to the next neuron;
wherein, WxhRepresenting the weight between the current input and the current candidate hidden state; w is a weight matrix; "tanh (. circle.)" represents a hyperbolic tangent function; "+" indicates point-by-point multiplication.
Step 2.5: the training loop is cycled through the above steps until the model converges, using Multi-category entry as the loss function.
The step 3 comprises the following steps:
step 3.1: verifying the trained model on a verification set;
step 3.2: adjusting parameters according to the model result;
step 3.3: and (5) completing parameter adjustment and saving the model.
The step 4 comprises the following steps:
step 4.1: using the test set to test the finally trained classification model;
step 4.2: comparing all the prediction results with the true values to obtain a final model evaluation index;
step 4.3: comparing the obtained result with other methods to obtain the superiority of the model;
step 4.4: and adopting a final classification model for a fault detection link of the actual rotating machinery.
The comparison result of the final model is shown in table 1, the training loss value of the GRU model is shown in fig. 3, and the accuracy of the GRU model is shown in fig. 4, so that the algorithm in the patent has good practical prospect and generalization capability.
TABLE 1 Fault diagnosis result Table
It should be noted that the above embodiments are only used for explaining the bearing fault diagnosis of the present invention, and are not intended to limit the present invention. It should be understood by those skilled in the art that modifications and equivalents may be made to the technical solution of the present invention without departing from the spirit and scope of the technical solution of the present invention, and all such modifications and equivalents are intended to be included within the scope of the claims of the present invention.
Claims (5)
1. A bearing fault identification method based on a gated cyclic unit network is characterized by mainly comprising the following steps:
step 1: preprocessing is carried out after data acquisition to obtain time sequence data of the acceleration sensor, and a data set is divided into a training set, a verification set and a test set;
step 2: inputting data of a training set into a gated cyclic unit network for training until the network converges;
and step 3: inputting the data of the verification set into a network for verification, and adjusting the parameters of the network to finally obtain the optimal parameters;
and 4, step 4: and storing the final model, inputting the test set for recognition effect test, and using the final model in an actual rotary machine diagnosis link.
2. The gated cyclic unit network-based bearing fault identification method according to claim 1, wherein the main steps of the step 1 are as follows:
step 1.1: data acquisition: an SKF6205 bearing is adopted in an experiment, faults are introduced through electric spark machining, the fault diameters are 0.18mm, 0.36mm and 0.53mm respectively, and 4 kinds of vibration signal data of a normal state, a ball fault, an inner ring fault and an outer ring fault are collected;
step 1.2: data preprocessing: carrying out mean value missing value filling on the index historical time sequence data, and obtaining a final historical time sequence data set after processing and standardizing abnormal values;
step 1.3: data set partitioning: partitioning a historical time series data set into: the training set contains 1900 pieces of data, the validation set contains 500 pieces of data, and the test set contains 500 pieces of data.
3. The gated cyclic unit network-based bearing fault identification method according to claim 1, wherein the main steps of the step 2 are as follows:
step 2.1: and constructing a gating cycle unit network, wherein the gating cycle unit network comprises an input layer, a gating cycle unit layer, a full connection layer and an output layer. The parameters to be adjusted are the number of gating cycle units of each layer, a Dropout value, the selection of an optimizer and the number of neurons of a full connection layer;
step 2.2: initializing the hidden state of a gating cycle unit, and inputting data;
step 2.3: calculating the weight of the current neuronDoor rtUpdate gate ztWeight and alternative hidden state ofComprises the following steps:
zt=σ(Wxzxt+Whzht-1+bz)
rt=σ(Wxrxt+Whrht-1+br)
wherein, WhzRepresenting the weight between the last hidden state and the update gate, WxzRepresenting the weight between the current input and the update gate, WhrRepresenting the weight between the last hidden state and the reset gate, WxrRepresenting the weight between the current input and the reset gate, WxhRepresenting the weight between the current input and the current candidate hidden state, and W is a weight matrix; bzIs to update the bias term of the gate, brIs the bias term for the reset gate; "σ (-) denotes a sigmoid function, and" tanh (-) denotes a hyperbolic tangent function; "+" indicates point-by-point multiplication;
step 2.4: computing the hidden state h of the current neurontAnd passed on to the next neuron;
step 2.5: the training loop is cycled through the above steps until the model converges, using Multi-category entry as the loss function.
4. The gated cyclic unit network-based bearing fault identification method according to claim 1, wherein the main steps of the step 3 are as follows:
step 3.1: testing the trained model on the verification set;
step 3.2: adjusting parameters according to the model result;
step 3.3: and (5) completing parameter adjustment and saving the model.
5. The gated cyclic unit network-based bearing fault identification method according to claim 1, wherein the main steps of the step 4 are as follows:
step 4.1: using the test set to test the finally trained classification model;
step 4.2: comparing all the prediction results with the true values to obtain a final model evaluation index;
step 4.3: comparing the obtained result with other methods to obtain the superiority of the model;
step 4.4: the final classification model can be used for a fault detection link of an actual rotating machine.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010769848.6A CN111898686A (en) | 2020-08-04 | 2020-08-04 | Bearing fault identification method based on gated cyclic unit network |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010769848.6A CN111898686A (en) | 2020-08-04 | 2020-08-04 | Bearing fault identification method based on gated cyclic unit network |
Publications (1)
Publication Number | Publication Date |
---|---|
CN111898686A true CN111898686A (en) | 2020-11-06 |
Family
ID=73183624
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010769848.6A Pending CN111898686A (en) | 2020-08-04 | 2020-08-04 | Bearing fault identification method based on gated cyclic unit network |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111898686A (en) |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112328588A (en) * | 2020-11-27 | 2021-02-05 | 哈尔滨工程大学 | Industrial fault diagnosis unbalanced time sequence data expansion method |
CN112763967A (en) * | 2020-12-11 | 2021-05-07 | 国网辽宁省电力有限公司鞍山供电公司 | BiGRU-based intelligent electric meter metering module fault prediction and diagnosis method |
CN112836577A (en) * | 2020-12-30 | 2021-05-25 | 中南大学 | Intelligent traffic unmanned vehicle fault gene diagnosis method and system |
CN113486291A (en) * | 2021-06-18 | 2021-10-08 | 电子科技大学 | Petroleum drilling machine micro-grid fault prediction method based on deep learning |
CN114545899A (en) * | 2022-02-10 | 2022-05-27 | 上海交通大学 | Priori knowledge-based multi-sensor fault signal reconstruction method for gas turbine system |
CN114819102A (en) * | 2022-05-19 | 2022-07-29 | 东南大学溧阳研究院 | GRU-based air conditioning equipment fault diagnosis method |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108960077A (en) * | 2018-06-12 | 2018-12-07 | 南京航空航天大学 | A kind of intelligent failure diagnosis method based on Recognition with Recurrent Neural Network |
CN111026095A (en) * | 2019-12-30 | 2020-04-17 | 太原科技大学 | Fault diagnosis method with noise label based on recurrent neural network |
CN111459144A (en) * | 2020-05-01 | 2020-07-28 | 西北工业大学 | Airplane flight control system fault prediction method based on deep cycle neural network |
-
2020
- 2020-08-04 CN CN202010769848.6A patent/CN111898686A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108960077A (en) * | 2018-06-12 | 2018-12-07 | 南京航空航天大学 | A kind of intelligent failure diagnosis method based on Recognition with Recurrent Neural Network |
CN111026095A (en) * | 2019-12-30 | 2020-04-17 | 太原科技大学 | Fault diagnosis method with noise label based on recurrent neural network |
CN111459144A (en) * | 2020-05-01 | 2020-07-28 | 西北工业大学 | Airplane flight control system fault prediction method based on deep cycle neural network |
Non-Patent Citations (1)
Title |
---|
汪洋 等: "一种新型DSCNN-GRU结构的减速机轴承故障诊断方法", 《机械科学与技术》, vol. 39, no. 2, pages 258 - 266 * |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112328588A (en) * | 2020-11-27 | 2021-02-05 | 哈尔滨工程大学 | Industrial fault diagnosis unbalanced time sequence data expansion method |
CN112763967A (en) * | 2020-12-11 | 2021-05-07 | 国网辽宁省电力有限公司鞍山供电公司 | BiGRU-based intelligent electric meter metering module fault prediction and diagnosis method |
CN112836577A (en) * | 2020-12-30 | 2021-05-25 | 中南大学 | Intelligent traffic unmanned vehicle fault gene diagnosis method and system |
CN112836577B (en) * | 2020-12-30 | 2024-02-20 | 中南大学 | Intelligent traffic unmanned vehicle fault gene diagnosis method and system |
CN113486291A (en) * | 2021-06-18 | 2021-10-08 | 电子科技大学 | Petroleum drilling machine micro-grid fault prediction method based on deep learning |
CN113486291B (en) * | 2021-06-18 | 2023-04-07 | 电子科技大学 | Petroleum drilling machine micro-grid fault prediction method based on deep learning |
CN114545899A (en) * | 2022-02-10 | 2022-05-27 | 上海交通大学 | Priori knowledge-based multi-sensor fault signal reconstruction method for gas turbine system |
CN114819102A (en) * | 2022-05-19 | 2022-07-29 | 东南大学溧阳研究院 | GRU-based air conditioning equipment fault diagnosis method |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111898686A (en) | Bearing fault identification method based on gated cyclic unit network | |
US11454567B2 (en) | Fault diagnosis method of reciprocating machinery based on keyphasor-free complete-cycle signal | |
CN111046945B (en) | Fault type and damage degree diagnosis method based on combined convolutional neural network | |
CN111460728B (en) | Method and device for predicting residual life of industrial equipment, storage medium and equipment | |
CN112883994B (en) | Rotating machinery variable working condition fault diagnosis method with balanced distribution adaptation | |
CN111562108A (en) | Rolling bearing intelligent fault diagnosis method based on CNN and FCMC | |
CN110567720A (en) | method for diagnosing depth confrontation of fault of fan bearing under unbalanced small sample scene | |
CN113188794B (en) | Gearbox fault diagnosis method and device based on improved PSO-BP neural network | |
CN109946080B (en) | Mechanical equipment health state identification method based on embedded circulation network | |
CN113203566B (en) | Motor bearing fault diagnosis method based on one-dimensional data enhancement and CNN | |
CN111459144A (en) | Airplane flight control system fault prediction method based on deep cycle neural network | |
Miao et al. | A novel real-time fault diagnosis method for planetary gearbox using transferable hidden layer | |
CN112257767B (en) | Product key part state classification method aiming at class imbalance data | |
Zhao et al. | Multiscale graph-guided convolutional network with node attention for intelligent health state diagnosis of a 3-PRR planar parallel manipulator | |
CN111680788A (en) | Equipment fault diagnosis method based on deep learning | |
CN111190349A (en) | Method, system and medium for monitoring state and diagnosing fault of ship engine room equipment | |
CN113139278A (en) | Deep learning-based bearing residual life prediction method and system | |
CN112729834B (en) | Bearing fault diagnosis method, device and system | |
CN114626625B (en) | Bearing life prediction method based on multi-module U-BiLSTM network | |
CN116821697A (en) | Mechanical equipment fault diagnosis method based on small sample learning | |
Wenqiang et al. | Remaining useful life prediction for mechanical equipment based on temporal convolutional network | |
CN115795397A (en) | Gearbox temperature prediction method based on 1DCNN-LSTM and BiLSTM parallel network | |
Zhang et al. | A flexible monitoring framework via dynamic-multilayer graph convolution network | |
CN115099264A (en) | Ship part fault diagnosis method and device, computer and computer storage medium | |
CN114841063A (en) | Aero-engine residual life prediction method based on deep learning |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination |