CN108343599B - A kind of water pump assembly intelligent failure diagnosis method based on more granularities cascade forest - Google Patents
A kind of water pump assembly intelligent failure diagnosis method based on more granularities cascade forest Download PDFInfo
- Publication number
- CN108343599B CN108343599B CN201810023728.4A CN201810023728A CN108343599B CN 108343599 B CN108343599 B CN 108343599B CN 201810023728 A CN201810023728 A CN 201810023728A CN 108343599 B CN108343599 B CN 108343599B
- Authority
- CN
- China
- Prior art keywords
- water pump
- pump assembly
- forest
- sample
- feature
- 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.)
- Expired - Fee Related
Links
- 235000019580 granularity Nutrition 0.000 title claims abstract description 39
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 title claims abstract description 37
- 238000003745 diagnosis Methods 0.000 title claims abstract description 32
- 238000000034 method Methods 0.000 title claims abstract description 31
- 238000006243 chemical reaction Methods 0.000 claims abstract description 10
- 238000007637 random forest analysis Methods 0.000 claims description 23
- 238000012549 training Methods 0.000 claims description 16
- 239000013598 vector Substances 0.000 claims description 15
- 230000001133 acceleration Effects 0.000 claims description 12
- 238000002790 cross-validation Methods 0.000 claims description 3
- 238000003066 decision tree Methods 0.000 claims description 3
- 238000000605 extraction Methods 0.000 claims description 3
- 241000287196 Asthenes Species 0.000 claims description 2
- 239000012141 concentrate Substances 0.000 claims description 2
- 238000005272 metallurgy Methods 0.000 abstract description 2
- 239000000126 substance Substances 0.000 abstract description 2
- 238000013024 troubleshooting Methods 0.000 abstract 1
- 238000013528 artificial neural network Methods 0.000 description 5
- 238000005096 rolling process Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 4
- 230000008569 process Effects 0.000 description 4
- 241000196324 Embryophyta Species 0.000 description 3
- 238000005086 pumping Methods 0.000 description 3
- 241000208340 Araliaceae Species 0.000 description 2
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 description 2
- 235000003140 Panax quinquefolius Nutrition 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000004422 calculation algorithm Methods 0.000 description 2
- 238000013135 deep learning Methods 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 238000002405 diagnostic procedure Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 235000008434 ginseng Nutrition 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 239000002245 particle Substances 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 230000003044 adaptive effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000006698 induction Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Classifications
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F04—POSITIVE - DISPLACEMENT MACHINES FOR LIQUIDS; PUMPS FOR LIQUIDS OR ELASTIC FLUIDS
- F04B—POSITIVE-DISPLACEMENT MACHINES FOR LIQUIDS; PUMPS
- F04B51/00—Testing machines, pumps, or pumping installations
Landscapes
- Engineering & Computer Science (AREA)
- Mechanical Engineering (AREA)
- General Engineering & Computer Science (AREA)
- Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)
- Monitoring And Testing Of Nuclear Reactors (AREA)
- Control Of Positive-Displacement Pumps (AREA)
Abstract
The invention discloses a kind of water pump assembly intelligent failure diagnosis methods based on more granularities cascade forest, are related to mechanical fault diagnosis and Artificial smart field.The described method includes: acquire water pump assembly to be measured it is in running order when working state signal, using working state signal as sample to be tested；The sample to be tested is input in more granularity cascade forest diagnostic models, Feature Conversion is carried out and feature diagnoses step by step, obtain the corresponding working condition label of the sample to be tested, the working condition for surveying water pump assembly is obtained according to the working condition label.Water pump assembly intelligent failure diagnosis method of the present invention based on more granularities cascade forest improves the accuracy and validity of water pump assembly intelligent trouble diagnosis, a kind of new effective way is provided to solve the troubleshooting issue of water pump assembly, can be widely applied in the complication system of each key areas such as electric power, machinery, metallurgy, chemical industry.
Description
Technical field
The present invention relates to mechanical fault diagnosis and Artificial smart field, art more particularly to one kind are based on more particle size fractions
Join the water pump assembly intelligent failure diagnosis method of forest.
Background technique
Water pump assembly is delivery equipment important in project of South-to-North water diversion as one of component most important in pumping plant,
Operating status has vital influence to the operation conditions of entire pumping plant.However water pump assembly is in heavy load, thump, height
Inevitably various failures under the bad working environments such as revolving speed, overall background noise, if fault type cannot be accurately identified
And take corresponding means, then it will affect the safe operation of pumping plant, cause national economy loss and personal safety.
Because fault characteristic information is present among vibration signal, therefore, it is water pump assembly state prison to analysis of vibration signal
Common method in survey.Conventional fault diagnosis method utilizes the feature that after signal processing technology manual extraction feature, will be extracted
As input, it is brought into the sorting algorithm of machine learning, which needs professional domain knowledge abundant,
And only for some specific problem, do not have universality.During the diagnosis of the physical fault of water pump assembly, diagnostic result
Depending on the quality of extracted feature, if extracted feature is of poor quality, the detection efficiency of traditional diagnosis method is used
Low and False Rate is high.
In recent years, as deep learning is in every field such as speech recognition, the development of Activity recognition etc. is based on deep learning
Theoretical intelligent Fault Diagnosis Technique has very big breakthrough, but existing intelligent Fault Diagnosis Technique is based on deep neural network
Diagnostic model adjusts ginseng sufficiently complex, and theory analysis is difficult, and in the case where fault sample data very little, classification accuracy
It is low.Therefore, it is necessary to a kind of diagnostic models being different from based on deep neural network, carry out feature learning and classification end to end.
Summary of the invention
The purpose of the present invention is to provide it is a kind of based on more granularities cascade forest water pump assembly intelligent failure diagnosis method,
To solve the problems, such as that the detection efficiency of conventional rolling bearing diagnostic method is low and False Rate is high.
To achieve the goals above, the water pump assembly intelligent trouble diagnosis side of the present invention based on more granularities cascade forest
Method, which comprises
S1, acquire water pump assembly to be measured it is in running order when working state signal, using working state signal as to
Test sample；
The sample to be tested is input in more granularities cascade forest diagnostic models by S2, carries out Feature Conversion and step by step
Feature diagnosis, obtains the corresponding working condition label of the sample to be tested, obtains the survey according to the working condition label
The working condition of water pump assembly；
It include establishing more granularity cascade forest diagnostic models before S1, specifically:
S01 obtains water pump assembly and is in the working state signal for working normally operating condition and various faults status condition, establishes
There is the sample data set of working condition label；The working condition label includes fault type label and normal operating conditions mark
Label；
S02, the sample signal that the sample data is concentrated realize Feature Conversion through excessive granularity scan phase, using
Cascade forest structure carries out training step by step for supervision, obtains more granularity cascade forest diagnostic models.
Preferably, S01, specifically:
S11, if water pump assembly, which is respectively at, works normally operating condition and various faults status condition；Pass through acceleration transducer
Acquisition water pump assembly is in the vibration acceleration signal under every kind of operating condition respectively；
S12 obtains working condition type corresponding to each vibration acceleration signal, and in statistical work Status Type
Fault type；
S13, adds corresponding working condition label on each vibration acceleration signal, and completion has working condition mark
The foundation of the sample data set of label.
Preferably, in S02, the sample signal that the sample data is concentrated realizes that feature turns through excessive granularity scan phase
It changes, specifically:
A1 generates the training example that sample data concentrates any one sample
Using the vibration letter of any one multidimensional characteristic in the different size of sliding characteristics window scanned samples data set of m kind
Number, generate the training sample of respective dimensions；If the vibration signal of multidimensional characteristic is [Y1,Y2] dimensional feature vibration signal；
It is with sizeTwo-dimentional sliding window for, sliding step is denoted asThen window scanning is appointed
Anticipate [a Y1,Y2] dimensional feature vibration signal, generating should [Y1,Y2] dimensional feature vibration signal training example number nwAre as follows:
A2 constructs two kinds of Random Forest models
Depth random forest includes standard random forest and completely random forest, and the decision tree inside standard random forest is logical
The mode that Gini criterion and random character extract is crossed to be constructed, completely random forest using all features and by Gini criterion into
Row building；
In more granularity scan phases, the output that the training example of generation is obtained by two kinds of random forests is swept as more granularities
The feature extraction in stage is retouched as a result, being denoted as more granularity scanning feature extracted vectors；
In the cascade sort stage, more granularity scanning feature extracted vectors substitute into the mould of each layer depth random forest as input
Type, the classification results of every layer model are merged with more granularity scanning feature extracted vectors, as next layer depth Random Forest model
Input continue to calculate, until convergence criterion meets condition, i.e. cross validation rate no longer increases, into A3；
A3, final result output
Model result in the last layer cascade forest is exported, the probability value meter of each failure modes is carried out using ballot method
It calculates, takes maximum probability as the classification results of final fault diagnosis.
The beneficial effects of the present invention are:
Compared with existing fault diagnosis technology, the water pump assembly intelligence event of the present invention based on more granularities cascade forest
Hinder accuracy and validity that diagnostic method improves water pump assembly intelligent trouble diagnosis, is asked to solve the fault diagnosis of water pump assembly
Topic provides a kind of new effective way, can be widely applied to the complication system of each key areas such as electric power, machinery, metallurgy, chemical industry
In the middle.Specifically:
1, the water pump assembly intelligent failure diagnosis method of the present invention based on more granularities cascade forest utilizes more particle size fractions
Connection forest algorithm directly adaptively learns fault signature from original vibration acceleration signal, excavates and is hidden in vibration acceleration
The characteristic information in signal is spent, and realizes failure modes and identification automatically.This method is got rid of in conventional fault diagnosis method to letter
Number processing knowledge and diagnostician's experience dependence, save a large amount of labour cost and time.
2, the water pump assembly intelligent failure diagnosis method of the present invention based on more granularities cascade forest is used based on random
The diagnostic model of forest is different from traditional depth model based on artificial neural network, avoids deep neural network and be applied to
Complicated theory analysis, cumbersome tune in fault diagnosis field join process, and to the defect of the low classification accuracy of small sample,
And the method for the invention energy adaptive determining trains the number of plies, insensitive to Parameters variation, can obtain parameter constant
Very high accuracy rate of diagnosis.
Detailed description of the invention
Fig. 1 is the block diagram that the water pump assembly intelligent failure diagnosis method of forest is cascaded based on more granularities；
Fig. 2 is the detailed process schematic diagram that the water pump assembly intelligent failure diagnosis method of forest is cascaded based on more granularities；
Fig. 3 is the flow diagram for realizing Feature Conversion；
Fig. 4 is that each Feature Conversion signal carries out the flow diagram trained step by step for having supervision by cascade forest structure.
Specific embodiment
In order to make the objectives, technical solutions, and advantages of the present invention clearer, below in conjunction with attached drawing, to the present invention into
Row is further described.It should be appreciated that the specific embodiments described herein are only used to explain the present invention, it is not used to
Limit the present invention.
Bearing is one of critical component of water pump assembly, with reference to the accompanying drawing, the bearing event provided with Xi Chu university, the U.S.
Barrier data instance illustrates a specific embodiment of the invention.
The basic of water pump assembly data center, Xi Chu university testing stand is laid out by one 2 horsepowers of three phase induction motor,
One acceleration transducer and a power meter composition.The data of collection come from a motor-driven mechanical system, sampling frequency
Rate is every channel 48kHz.Under three kinds of different operating conditions, Single Point of Faliure data set is obtained by experimental system, three kinds of operating conditions are respectively as follows:
(1) band outer ring failure；(2) band inner ring failure；(3) band rolling element failure.The flaw size (diameter, depth) of three kinds of failures point
It Wei not be 0.007,0.014 or 0.021 inch.
1, the sample data under 9 kinds of different faults operating conditions of collection, i.e. outer ring failure, inner ring failure, rolling element failure, every kind
It respectively corresponds 0.007,0.014 or 0.021 inch of failure depth, adds fault type label, 0~8, establish training sample
Collection.
2, more granularity cascade forest models are established, which is broadly divided into two parts, and training sample is scanned through excessive granularity
Stage realizes Feature Conversion, carries out training step by step for supervision using cascade forest structure.Final prediction result passes through poly-
Collect the class vector of four respective dimensions of afterbody, and is maximized as final classification result.
More granularity cascade forest models are specific, and specific step is as follows:
More granularity scan phases:
Using 3 kinds of different size of sliding windows, respectively 250,500,1000 dimensions go to scan original 4096 dimensional feature
Vibration signal generates the training example of respective dimensions.
A training one random forest and completely random tree forest is removed with these training examples, each random forest and complete
It include respectively 500 decision trees in full random tree forest, for each example, each tree can generate the estimated value of class distribution,
These estimated values form a class vector.Wherein, random forest passes through the random selection best Geordie coefficient value of 64 feature selectings
A feature divided, the number of the tree in each forest is a hyper parameter.Completely random tree forest be then by with
Machine selects a feature in the enterprising line splitting of each node of tree, until each leaf node only includes same class example.In order to subtract
The risk of few over-fitting, the class vector that each forest generates are generated by 3 folding cross validations.
The class vector that the same window generates respectively after two kinds of random forests is connected, as the feature after conversion
Vector.
Cascade the forest stage:
With step B1-B4, by the feature vector sample after above-mentioned transformation be brought into two random forests and two completely with
Machine tree forest generates the class vector of respective dimensions.The class vector of generation and the converting characteristic vector of upper level link together, and make
The input of forest is cascaded for next stage.
The process of above-mentioned feature learning step by step is repeated, until verifying collection convergence.
By using above-mentioned technical proposal disclosed by the invention, following beneficial effect has been obtained: of the present invention to be based on
The water pump assembly intelligent failure diagnosis method of more granularity cascade forests, using more granularities cascade forest method directly from original vibration
In acceleration signal carry out layer and layer feature learning and failure modes, avoid manual designs and extract feature limitation and
Complicated tune ginseng process based on artificial neural network, so that the intelligent diagnostics of rolling bearing fault are realized, to improve rolling bearing
The accuracy and validity of fault diagnosis.
The above is only a preferred embodiment of the present invention, it is noted that for the ordinary skill people of the art
For member, various improvements and modifications may be made without departing from the principle of the present invention, these improvements and modifications are also answered
Depending on protection scope of the present invention.
Claims (1)
1. a kind of water pump assembly intelligent failure diagnosis method based on more granularities cascade forest, which is characterized in that the method packet
It includes:
S1, acquire water pump assembly to be measured it is in running order when working state signal, using working state signal as to be tested
Sample；
The sample to be tested is input in more granularity cascade forest diagnostic models, carries out Feature Conversion and step by step feature by S2
Diagnosis, obtains the corresponding working condition label of the sample to be tested, obtains the survey water pump according to the working condition label
The working condition of unit；
It include establishing more granularity cascade forest diagnostic models before S1, specifically:
S01 obtains water pump assembly and is in the working state signal for working normally operating condition and various faults status condition, and foundation has work
Make the sample data set of state tag；The working condition label includes fault type label and normal operating conditions label；
S02, the sample signal that the sample data is concentrated realizes Feature Conversion through excessive granularity scan phase, using cascade
Forest structure carries out training step by step for supervision, obtains more granularity cascade forest diagnostic models；
S01, specifically:
S11, if water pump assembly, which is respectively at, works normally operating condition and various faults status condition；Distinguished by acceleration transducer
Acquire the vibration acceleration signal that water pump assembly is under every kind of operating condition；
S12 obtains working condition type corresponding to each vibration acceleration signal, and the failure in statistical work Status Type
Type；
S13, adds corresponding working condition label on each vibration acceleration signal, and completion has working condition label
The foundation of sample data set；
In S02, the sample signal that the sample data is concentrated realizes Feature Conversion through excessive granularity scan phase, specifically:
A1 generates the training example that sample data concentrates any one sample
Using the vibration signal of any one multidimensional characteristic in the different size of sliding characteristics window scanned samples data set of m kind,
Generate the training sample of respective dimensions；If the vibration signal of multidimensional characteristic is [Y1, Y2] dimensional feature vibration signal；
It is with sizeTwo-dimentional sliding window for, sliding step is denoted asThen window scanning is any one
A [Y1, Y2] dimensional feature vibration signal, generating should [Y1, Y2] dimensional feature vibration signal training example number nwAre as follows:
A2 constructs two kinds of Random Forest models
Depth random forest includes standard random forest and completely random forest, and the decision tree inside standard random forest passes through
The mode that Gini criterion and random character extract is constructed, and completely random forest is carried out using all features and by Gini criterion
Building；
Rank is scanned as more granularities by the output that two kinds of random forests obtain in the training example of more granularity scan phases, generation
The feature extraction of section is as a result, be denoted as more granularity scanning feature extracted vectors；
In the cascade sort stage, more granularity scanning feature extracted vectors substitute into the model of each layer depth random forest as input,
The classification results of every layer model are merged with more granularity scanning feature extracted vectors, as next layer depth Random Forest model
Input continues to calculate, and until convergence criterion meets condition, i.e. cross validation rate no longer increases, into A3；
A3, final result output
Model result in the last layer cascade forest is exported, is calculated, is taken using the probability value that ballot method carries out each failure modes
Classification results of the maximum probability as final fault diagnosis.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810023728.4A CN108343599B (en) | 2018-01-10 | 2018-01-10 | A kind of water pump assembly intelligent failure diagnosis method based on more granularities cascade forest |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810023728.4A CN108343599B (en) | 2018-01-10 | 2018-01-10 | A kind of water pump assembly intelligent failure diagnosis method based on more granularities cascade forest |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108343599A CN108343599A (en) | 2018-07-31 |
| CN108343599B true CN108343599B (en) | 2019-04-23 |
Family
ID=62961404
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810023728.4A Expired - Fee Related CN108343599B (en) | 2018-01-10 | 2018-01-10 | A kind of water pump assembly intelligent failure diagnosis method based on more granularities cascade forest |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108343599B (en) |
Families Citing this family (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109102032A (en) * | 2018-09-03 | 2018-12-28 | 中国水利水电科学研究院 | A kind of pumping plant unit diagnostic method based on depth forest and oneself coding |
| CN109297689B (en) * | 2018-09-11 | 2020-11-06 | 中国水利水电科学研究院 | Large-scale hydraulic machinery intelligent diagnosis method introducing weight factors |
| CN110020637B (en) * | 2019-04-16 | 2023-04-07 | 重庆大学 | Analog circuit intermittent fault diagnosis method based on multi-granularity cascade forest |
| CN110298485A (en) * | 2019-05-29 | 2019-10-01 | 国电联合动力技术有限公司 | Based on the pitch-controlled system failure prediction method for improving depth random forests algorithm |
| CN110375987B (en) * | 2019-06-24 | 2021-10-22 | 昆明理工大学 | Mechanical bearing fault detection method based on deep forest |
| CN111412948A (en) * | 2020-03-31 | 2020-07-14 | 南京云思顿环保科技有限公司 | Kitchen waste equipment fault diagnosis method based on deep forest |
| CN111582362A (en) * | 2020-05-06 | 2020-08-25 | 河海大学常州校区 | Farmland water pump fault diagnosis method and device based on random forest algorithm |
| CN111476383B (en) * | 2020-05-08 | 2023-06-02 | 中国水利水电科学研究院 | Dynamic decision method for state maintenance of pump station unit |
| CN111722046B (en) * | 2020-07-01 | 2021-05-18 | 昆明理工大学 | Transformer fault diagnosis method based on deep forest model |
| CN112116058B (en) * | 2020-09-16 | 2022-05-31 | 昆明理工大学 | Transformer fault diagnosis method for optimizing multi-granularity cascade forest model based on particle swarm algorithm |
| CN113485287B (en) * | 2021-06-23 | 2024-03-26 | 佛山科学技术学院 | Multi-granularity measurement dynamic system fault diagnosis method and system |
| CN115062553B (en) * | 2022-08-17 | 2022-11-25 | 浪潮通用软件有限公司 | Water pump working condition degradation detection method, equipment and medium based on multi-model fusion |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2009072098A1 (en) * | 2007-12-04 | 2009-06-11 | University College Dublin, National University Of Ireland | Method and system for image analysis |
| CN105971901A (en) * | 2016-05-03 | 2016-09-28 | 北京航空航天大学 | Centrifugal pump fault diagnosis method based on complete ensemble empirical mode decomposition and random forest |
| CN106897109A (en) * | 2017-02-13 | 2017-06-27 | 云南大学 | Based on the virtual machine performance Forecasting Methodology that random forest is returned |
| CN107506796A (en) * | 2017-08-25 | 2017-12-22 | 电子科技大学 | A kind of alzheimer disease sorting technique based on depth forest |
-
2018
- 2018-01-10 CN CN201810023728.4A patent/CN108343599B/en not_active Expired - Fee Related
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2009072098A1 (en) * | 2007-12-04 | 2009-06-11 | University College Dublin, National University Of Ireland | Method and system for image analysis |
| CN105971901A (en) * | 2016-05-03 | 2016-09-28 | 北京航空航天大学 | Centrifugal pump fault diagnosis method based on complete ensemble empirical mode decomposition and random forest |
| CN106897109A (en) * | 2017-02-13 | 2017-06-27 | 云南大学 | Based on the virtual machine performance Forecasting Methodology that random forest is returned |
| CN107506796A (en) * | 2017-08-25 | 2017-12-22 | 电子科技大学 | A kind of alzheimer disease sorting technique based on depth forest |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108343599A (en) | 2018-07-31 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108343599B (en) | A kind of water pump assembly intelligent failure diagnosis method based on more granularities cascade forest | |
| CN108241873B (en) | A kind of intelligent failure diagnosis method towards pumping plant main equipment | |
| CN109102032A (en) | A kind of pumping plant unit diagnostic method based on depth forest and oneself coding | |
| CN109555566B (en) | Steam turbine rotor fault diagnosis method based on LSTM | |
| CN107273920A (en) | A kind of non-intrusion type household electrical appliance recognition methods based on random forest | |
| CN107784325A (en) | Spiral fault diagnosis model based on the fusion of data-driven increment | |
| CN110044623A (en) | The rolling bearing fault intelligent identification Method of empirical mode decomposition residual signal feature | |
| CN108827605A (en) | A kind of mechanical breakdown characteristic automatic extraction method based on improvement sparseness filtering | |
| CN106092578A (en) | A kind of machine tool mainshaft bearing confined state online test method based on wavelet packet and support vector machine | |
| CN109297689A (en) | A kind of large-scale hydraulic intelligent diagnosing method introducing weight factor | |
| CN105424366A (en) | Bearing fault diagnosis method based on EEMD adaptive denoising | |
| CN105973595A (en) | Diagnosis method of rolling bearing fault | |
| CN105678343A (en) | Adaptive-weighted-group-sparse-representation-based diagnosis method for noise abnormity of hydroelectric generating set | |
| CN108869145A (en) | Pumping plant unit diagnostic method based on compound characteristics index and depth limit learning machine | |
| CN110020637A (en) | A kind of analog circuit intermittent fault diagnostic method based on more granularities cascade forest | |
| CN113869339A (en) | Deep learning classification model for fault diagnosis and fault diagnosis method | |
| CN116956215A (en) | Fault diagnosis method and system for transmission system | |
| Whitehill et al. | Whosecough: In-the-wild cougher verification using multitask learning | |
| Chen et al. | An adversarial learning framework for zero-shot fault recognition of mechanical systems | |
| CN113627358A (en) | Multi-feature fusion turnout intelligent fault diagnosis method, system and equipment | |
| CN113946920A (en) | Rolling bearing fault diagnosis method with unbalanced data and data set deviation | |
| CN102778632A (en) | Double normalization recognition method for directly forecasting and recognizing transformer winding fault type | |
| CN116484258A (en) | Elevator traction machine bearing fault diagnosis method | |
| CN116541771A (en) | Unbalanced sample bearing fault diagnosis method based on multi-scale feature fusion | |
| CN110490218A (en) | A kind of rolling bearing fault self-learning method based on two-stage DBN |
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 | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20190423 |