CN110068760A - A kind of Induction Motor Fault Diagnosis based on deep learning - Google Patents
A kind of Induction Motor Fault Diagnosis based on deep learning Download PDFInfo
- Publication number
- CN110068760A CN110068760A CN201910325479.9A CN201910325479A CN110068760A CN 110068760 A CN110068760 A CN 110068760A CN 201910325479 A CN201910325479 A CN 201910325479A CN 110068760 A CN110068760 A CN 110068760A
- Authority
- CN
- China
- Prior art keywords
- fault diagnosis
- deep learning
- induction motor
- diagnosis based
- data
- 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
- 238000003745 diagnosis Methods 0.000 title claims abstract description 33
- 238000013135 deep learning Methods 0.000 title claims abstract description 20
- 230000006698 induction Effects 0.000 title claims abstract description 14
- 238000000034 method Methods 0.000 claims abstract description 20
- 238000004088 simulation Methods 0.000 claims abstract description 11
- 238000010276 construction Methods 0.000 claims abstract description 7
- 238000013528 artificial neural network Methods 0.000 claims abstract description 6
- 239000000284 extract Substances 0.000 claims abstract description 5
- 230000001133 acceleration Effects 0.000 claims description 8
- 239000000203 mixture Substances 0.000 claims description 4
- 238000013461 design Methods 0.000 abstract description 3
- 238000012549 training Methods 0.000 description 10
- 230000004913 activation Effects 0.000 description 5
- 230000008569 process Effects 0.000 description 5
- 238000004804 winding Methods 0.000 description 5
- 230000006870 function Effects 0.000 description 3
- 238000004519 manufacturing process Methods 0.000 description 3
- 230000006378 damage Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000000644 propagated effect Effects 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 230000004888 barrier function Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000009776 industrial production Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 230000036651 mood Effects 0.000 description 1
- 238000013021 overheating Methods 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000000452 restraining effect Effects 0.000 description 1
- 230000035939 shock Effects 0.000 description 1
- 238000001228 spectrum Methods 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M15/00—Testing of engines
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/34—Testing dynamo-electric machines
- G01R31/343—Testing dynamo-electric machines in operation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2413—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on distances to training or reference patterns
- G06F18/24147—Distances to closest patterns, e.g. nearest neighbour classification
Abstract
The invention discloses a kind of Induction Motor Fault Diagnosis based on deep learning comprising the steps of: A, the simulation experiment platform for building asynchronous motor operating;B, the acquisition of current signal and vibration signal is carried out to the electrical fault state simulated;C, it extracts and collects the internal characteristics of data and carry out corresponding label, complete the building of data set;D, the fault diagnosis model based on stack encoder is constructed;Successively train the network and classifier;E, the deep neural network built is trained using the data set of construction, and the method for diagnosing faults proposed is verified in conjunction with simulation experiment platform.Present invention introduces deep learning theory, the system that accurate, sensitive, effective diagnosis asynchronous motor complex fault is capable of in design solves the problems, such as present on Asynchronous Motor Fault Diagnosis, adapts to the requirement of the electric system continued to develop.
Description
Technical field
The present invention relates to a kind of computer, specifically a kind of Induction Motor Fault Diagnosis based on deep learning.
Background technique
With the raising of the progress of modern science and technology and the development of production system and device fabrication level, motor conduct
In the world using most common, quantity is most power supply unit and dynamic power machine, all spectra has almost been captured.It is clear that
It is very heavy that the normal work of motor runs meaning to safe and efficient, quick, the high-quality and low consumption guaranteed during manufacturing
Greatly.Wherein asynchronous motor is most widely used, a kind of motor for having the call in various motor.In the total load of power grid
In, asynchronous motor dosage accounts for 60% or more.It is in current industrial production activities and daily life most important motive power and
Driving device.
Currently, asynchronous motor typical fault is rotor broken bar, stator winding inter-turn short circuit and bearing fault.Statistical result
Show that these three types of failures account for about the 80% of asynchronous motor failure sum: the probability of happening of rotor bar breaking fault up to 10% or so,
The probability of happening of stator winding inter-turn short circuit and its dependent failure is about 30%, and bearing fault accounts for about 38%.Operating experience table
Bright, rotor bar breaking fault will lead to motor output drop, if cannot find that failure further expansion can be made in time;Stator winding circle
Between short trouble often lead to phase fault or ground short circuit failure, harm is serious;Bearing fault can lead to motor and cannot open
In dynamic, operation sound exception, operational shock, operating slowly, motor overheating or even burn.
For a long time, for the various catastrophe failures of motor, the relay that people mainly take various mature and reliables is protected
Shield measure, as overcurrent protection, mistake/low-voltage protection, differential protection, passive moods, ground protection, vibration are transfinited protection and mistake
Thermal protection etc..Although relay protection function is all very perfect, supervised it must be recognized that relay protection system is only worked as
It meets or exceeds relay setting depending on parameter just to work, the relay protection of equipment can only take action after the accident;
It is the ambulance under cliffs and precipices, rather than the fence of pedestrian is protected on cliffs and precipices top.
The failure of motor is out of service, can not only damage motor itself, but also will affect the normal of whole system
Work, or even meeting crisis personal safety, cause huge economic loss.Asynchronous Motor Fault Diagnosis system can be accurate and reliable
Reflection motor failure and abnormal operating condition, fast trip or alarm are taken to different degrees of failure according to the actual situation
Equal measures, prevent accident further expansion, to avoid economic loss caused by serious accident and unnecessary shutdown, guarantee life
Produce safe operation;For the failure having occurred and that, fault diagnosis can be effectively carried out rapidly, judges fault type and failure
Degree, to investigate thoroughly that cause of accident and maintenance provide foundation afterwards.Therefore, Asynchronous Motor Fault Diagnosis technology is studied, in motor
Failure is found in time and is repaired there is great theory significance and practical significance when generating failure.Currently, being directed to asynchronous electricity
The common fault type of machine, traditional method for diagnosing faults have been able to effective solution, but because electric machine structure is complicated, vibration
The problems such as asynchronous machine fault diagnosis caused by the factors such as the unstable and mechanical big data of signal is difficult is still without very perfect solution
Certainly scheme.Therefore, it is theoretical to introduce deep learning, design being capable of accurate, sensitive, the effective complicated event of diagnosis asynchronous motor
The system of barrier solves the problems, such as present on Asynchronous Motor Fault Diagnosis, adapts to the requirement of the electric system continued to develop, research
Reliable and effective Asynchronous Motor Fault Diagnosis system has become the hot subject that domestic and foreign scholars study.
Summary of the invention
The purpose of the present invention is to provide a kind of Induction Motor Fault Diagnosis based on deep learning, to solve
State the problem of proposing in background technique.
In order to achieve the object, the invention provides the following technical scheme:
A kind of Induction Motor Fault Diagnosis based on deep learning comprising the steps of:
A, the simulation experiment platform of asynchronous motor operating is built;
B, the acquisition of current signal and vibration signal is carried out to the electrical fault state simulated;
C, it extracts and collects the internal characteristics of data and carry out corresponding label, complete the building of data set;
D, the fault diagnosis model based on stack encoder is constructed;Successively train the network and classifier;
E, the deep neural network built is trained using the data set of construction, and combines simulation experiment platform to institute
The method for diagnosing faults of proposition is verified.
As further scheme of the invention: the simulation experiment platform in the step A is controlled by computer, load
Device, asynchronous machine, tachometer, current sensor, acceleration transducer and NI data collecting card composition.
As further scheme of the invention: the current signal is adopted by current sensor and acceleration transducer
Collection.
As further scheme of the invention: the vibration signal passes through acceleration transducer and acquires.
As further scheme of the invention: the data of the step B acquisition also need to be transferred to computer.
As further scheme of the invention: the step D is specifically: construction is by multilayer autocoder stack first
The stack encoder deep layer network structure being formed by stacking constitutes fault diagnosis models of classifying then in conjunction with softmax classifier more,
The data of vibration signal and current signal are combined as input, the network and classifier are successively trained, and have pair of supervision
Whole network is finely adjusted.
As further scheme of the invention: the verifying index of the step E includes accuracy and speed.
Compared with prior art, the beneficial effects of the present invention are: design can be accurate present invention introduces deep learning theory
, the system of sensitive, effective diagnosis asynchronous motor complex fault, solve to ask present on Asynchronous Motor Fault Diagnosis
Topic adapts to the requirement of the electric system continued to develop.
Detailed description of the invention
Fig. 1 is autocoder structure chart.
Fig. 2 is stack encoder fault diagnostic model figure.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts every other
Embodiment shall fall within the protection scope of the present invention.
Embodiment 1: a kind of Induction Motor Fault Diagnosis based on deep learning referring to FIG. 1-2, comprising following
Step:
A, the simulation experiment platform of asynchronous motor operating is built;Simulation experiment platform is mainly controlled by computer, load
The composition such as device, asynchronous machine, tachometer, current sensor, acceleration transducer, NI data collecting card.What plan can diagnose
Failure includes rotor unbalance, stator winding faults, stator winding disconnected circle, bearing fault, rotor bow, rotor broken bar.To protect
The diversity for demonstrate,proving experimental data will simulate a variety of different work shapes by the revolving speed and loading condition of change asynchronous motor
State.
B, the acquisition of current signal and vibration signal is carried out to the electrical fault state simulated;The input sample to be acquired
It should as far as possible include all features of fault-signal, since vibration signal includes complicated bearing information, current signal includes rich
Rich rotor characteristic, therefore cooperate NI data collecting card to the electrical fault simulated by current sensor and acceleration transducer
State carries out the acquisition of current signal and vibration signal, then by collected data transmission into computer.
C, it extracts and collects the internal characteristics of data and carry out corresponding label, complete the building of data set;
D, the fault diagnosis model based on stack encoder is constructed;Successively train the network and classifier;First construction by
The stack encoder deep layer network structure that multilayer autocoder stack is formed by stacking is constituted then in conjunction with softmax classifier
More classification fault diagnosis models.The data of vibration signal and current signal are combined as input, successively train the network and
Classifier, and have being finely adjusted to whole network for supervision;
E, the deep neural network built is trained using the data set of construction, and combines simulation experiment platform to institute
The method for diagnosing faults of proposition is verified.First configure training environment, using ubuntu system, tensorflow frame,
Then python interpreter is trained the deep neural network built using the data set of construction, and combine simulated experiment
Platform verifies the method for diagnosing faults proposed, assesses this method by accuracy and speed two indices.
2, autocoder is studied
Stack is one of current mainstream deep learning network from coding, and autocoder is that the basic composition of the network is single
Member.As shown in Figure 1, autocoder network is a kind of 3 layers of unsupervised network.
If the training sample data collection of autocoder network is { x(1),x(2),...,x(n), which is tieed up by n m
Vector x ∈ [0,1]m×1It constitutes.Coding is that sample x is propagated to hidden layer from input layer, by Sigmoid activation primitive (see formula
1) it is mapped to the process of k dimensional vector, as shown in Equation 2.
In formula: x is input sample;F () is activation primitive;θ={ w, b } is network parameter, and w is weight, and b is biasing.
Decoding is that feature coding is propagated to output layer from hidden layer, by activation primitive be mapped to m dimensional vector y ∈ [0,
1]m×1.The process of reconstructed sample x is as follows:
In formula: y is the reconstruct to sample x;F () is activation primitive, θ '={ w', b'} are reconstructed network parameter, and w' is
Weight is reconstructed, b' is reconstruct biasing.
The training objective of autocoder network is by finding one group of optimal parameter θ*, so that output data and input
Error between data is as small as possible, i.e. realization loss function L (θ) minimizes, and loss function expression formula is as follows:
3, stack encoder is studied
Stack encoder is cascaded by multiple autocoders, using the hidden layer of upper level autocoder under
The input layer of level-one successively extracts abstract characteristics to form multi-level network structure.
Assuming that using w(k,1), w(k,2), b(k,1), b(k,2)The weight and deviation for indicating k-th of autocoder, then for one
A n-layer stack encoder, cataloged procedure are exactly according to suitable from first layer autocoder to the last layer autocoder
Sequence executes the coding step of each layer of autocoder:
a(l)=f (z(l)) (5)
z(l+1)=w(l, 1)a(l)+b(l, 1)
Similar, the decoding process of stack encoder executes each layer of autocoder according to sequence from back to front.
a(n+l)=f (z(n+l)) (6)
z(n+l+1)=w(n-l, 2)a(n+l)+b(n-l, 2)
4, training and verifying
A kind of method of relatively good acquisition stack encoder neural network parameter is carried out using successively greedy coaching method
Training.The first layer that network is trained first with being originally inputted, obtains its parameter;Then network first tier, which will be originally inputted, turns
Chemical conversion is the vector that is made of hidden unit activation value, then assigns the vector as the input of the second layer, continues trained to obtain the
Two layers of parameter;Finally, successively being trained to the strategy that subsequent each layer equally uses.
For above-mentioned training method, when training each layer parameter, other each layer parameters can be fixed and remained unchanged.Institute
With if expected preferably as a result, can be adjusted simultaneously by back-propagation algorithm after the completion of above-mentioned pre-training process
For all layers of parameter to improve as a result, this process is generally known as finely tuning, fine tuning can significantly promote deep learning network
Performance, be the common method of deep learning network reference services.In fact, using layer-by-layer greedy training method by parameter training
To when soon restraining, it is necessary to use fine tuning., whereas if using fine tuning directly on the initial weight of randomization, then can obtain
To bad as a result, because parameter can converge to local optimum.
According to the above feature of stack encoder, a kind of event based on stack encoder in conjunction with softmax classifier is proposed
Hinder the fault diagnosis that diagnostic model is used for asynchronous machine.It is exactly by the output quantity a of stack encoder(n)As softmax classifier
Input, to achieve the effect that classification, as shown in Figure 2.
It is obvious to a person skilled in the art that invention is not limited to the details of the above exemplary embodiments, Er Qie
In the case where without departing substantially from spirit or essential attributes of the invention, the present invention can be realized in other specific forms.Therefore, no matter
From the point of view of which point, the present embodiments are to be considered as illustrative and not restrictive, and the scope of the present invention is by appended power
Benefit requires rather than above description limits, it is intended that all by what is fallen within the meaning and scope of the equivalent elements of the claims
Variation is included within the present invention.Any reference signs in the claims should not be construed as limiting the involved claims.
In addition, it should be understood that although this specification is described in terms of embodiments, but not each embodiment is only wrapped
Containing an independent technical solution, this description of the specification is merely for the sake of clarity, and those skilled in the art should
It considers the specification as a whole, the technical solutions in the various embodiments may also be suitably combined, forms those skilled in the art
The other embodiments being understood that.
Claims (7)
1. a kind of Induction Motor Fault Diagnosis based on deep learning, which is characterized in that comprise the steps of:
Build the simulation experiment platform of asynchronous motor operating;
The acquisition of current signal and vibration signal is carried out to the electrical fault state simulated;
It extracts the internal characteristics for collecting data and carries out corresponding label, complete the building of data set;
Construct the fault diagnosis model based on stack encoder;Successively train the network and classifier;
Using the data set of construction the deep neural network built is trained, and combines simulation experiment platform to being proposed
Method for diagnosing faults is verified.
2. a kind of Induction Motor Fault Diagnosis based on deep learning according to claim 1, which is characterized in that
Simulation experiment platform in the step A is by computer, load controller, asynchronous machine, tachometer, current sensor, acceleration
Spend sensor and NI data collecting card composition.
3. a kind of Induction Motor Fault Diagnosis based on deep learning according to claim 1, which is characterized in that
The current signal is acquired by current sensor and acceleration transducer.
4. a kind of Induction Motor Fault Diagnosis based on deep learning according to claim 1, which is characterized in that
The vibration signal is acquired by acceleration transducer.
5. a kind of Induction Motor Fault Diagnosis based on deep learning according to claim 1, which is characterized in that
The data of the step B acquisition also need to be transferred to computer.
6. a kind of Induction Motor Fault Diagnosis based on deep learning according to claim 1, which is characterized in that
The step D is specifically: the stack encoder deep layer network structure being formed by stacking by multilayer autocoder stack is constructed first,
It constitutes fault diagnosis models of classifying then in conjunction with softmax classifier more, the data of vibration signal and current signal is combined
As input, the network and classifier are successively trained, and has being finely adjusted to whole network for supervision.
7. -6 any a kind of Induction Motor Fault Diagnosis based on deep learning according to claim 1, special
Sign is that the verifying index of the step E includes accuracy and speed.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910325479.9A CN110068760A (en) | 2019-04-23 | 2019-04-23 | A kind of Induction Motor Fault Diagnosis based on deep learning |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910325479.9A CN110068760A (en) | 2019-04-23 | 2019-04-23 | A kind of Induction Motor Fault Diagnosis based on deep learning |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110068760A true CN110068760A (en) | 2019-07-30 |
Family
ID=67368464
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910325479.9A Pending CN110068760A (en) | 2019-04-23 | 2019-04-23 | A kind of Induction Motor Fault Diagnosis based on deep learning |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110068760A (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110596586A (en) * | 2019-08-14 | 2019-12-20 | 河海大学常州校区 | Motor current waveform analysis method based on Alexnet algorithm |
CN111766067A (en) * | 2020-07-10 | 2020-10-13 | 中国人民解放军空军工程大学 | Aircraft outfield aircraft engine fault prediction method based on deep learning |
CN112052796A (en) * | 2020-09-07 | 2020-12-08 | 电子科技大学 | Permanent magnet synchronous motor fault diagnosis method based on deep learning |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107957551A (en) * | 2017-12-12 | 2018-04-24 | 南京信息工程大学 | Stacking noise reduction own coding Method of Motor Fault Diagnosis based on vibration and current signal |
CN108062572A (en) * | 2017-12-28 | 2018-05-22 | 华中科技大学 | A kind of Fault Diagnosis Method of Hydro-generating Unit and system based on DdAE deep learning models |
US20180144465A1 (en) * | 2016-11-23 | 2018-05-24 | General Electric Company | Deep learning medical systems and methods for medical procedures |
JP2018091640A (en) * | 2016-11-30 | 2018-06-14 | 株式会社デンソー | Apparatus inspection device and apparatus inspection method |
CN108710771A (en) * | 2018-05-31 | 2018-10-26 | 西安电子科技大学 | Mechanized equipment service reliability appraisal procedure based on the integrated extraction of depth characteristic |
CN109060347A (en) * | 2018-10-25 | 2018-12-21 | 哈尔滨理工大学 | Based on the planetary gear fault recognition method for stacking de-noising autocoder and gating cycle unit neural network |
CN109145886A (en) * | 2018-10-12 | 2019-01-04 | 西安交通大学 | A kind of asynchronous machine method for diagnosing faults of Multi-source Information Fusion |
-
2019
- 2019-04-23 CN CN201910325479.9A patent/CN110068760A/en active Pending
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20180144465A1 (en) * | 2016-11-23 | 2018-05-24 | General Electric Company | Deep learning medical systems and methods for medical procedures |
JP2018091640A (en) * | 2016-11-30 | 2018-06-14 | 株式会社デンソー | Apparatus inspection device and apparatus inspection method |
CN107957551A (en) * | 2017-12-12 | 2018-04-24 | 南京信息工程大学 | Stacking noise reduction own coding Method of Motor Fault Diagnosis based on vibration and current signal |
CN108062572A (en) * | 2017-12-28 | 2018-05-22 | 华中科技大学 | A kind of Fault Diagnosis Method of Hydro-generating Unit and system based on DdAE deep learning models |
CN108710771A (en) * | 2018-05-31 | 2018-10-26 | 西安电子科技大学 | Mechanized equipment service reliability appraisal procedure based on the integrated extraction of depth characteristic |
CN109145886A (en) * | 2018-10-12 | 2019-01-04 | 西安交通大学 | A kind of asynchronous machine method for diagnosing faults of Multi-source Information Fusion |
CN109060347A (en) * | 2018-10-25 | 2018-12-21 | 哈尔滨理工大学 | Based on the planetary gear fault recognition method for stacking de-noising autocoder and gating cycle unit neural network |
Non-Patent Citations (2)
Title |
---|
朱煜奇等, 制造业自动化,基于栈式降噪自编码的故障诊断, vol. 39, no. 3, 31 March 2017 (2017-03-31), pages 152 - 155 * |
王丽华等, 西安交通大学学报,采用深度学习的异步电机故障诊断方法, vol. 51, no. 10, 31 October 2017 (2017-10-31), pages 128 - 133 * |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110596586A (en) * | 2019-08-14 | 2019-12-20 | 河海大学常州校区 | Motor current waveform analysis method based on Alexnet algorithm |
CN110596586B (en) * | 2019-08-14 | 2021-12-10 | 河海大学常州校区 | Motor current waveform analysis method based on Alexnet algorithm |
CN111766067A (en) * | 2020-07-10 | 2020-10-13 | 中国人民解放军空军工程大学 | Aircraft outfield aircraft engine fault prediction method based on deep learning |
CN112052796A (en) * | 2020-09-07 | 2020-12-08 | 电子科技大学 | Permanent magnet synchronous motor fault diagnosis method based on deep learning |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
EP3780307B1 (en) | Online dynamic decision-making method and system for unit restoration | |
CN110068760A (en) | A kind of Induction Motor Fault Diagnosis based on deep learning | |
CN106443297B (en) | The decision tree SVM method for diagnosing faults of photovoltaic diode Clamp three-level inverter | |
CN105403811B (en) | Wind power plant power grid fault diagnosis method and device | |
CN104459388B (en) | Permanent magnetic direct-drive wind power generation system integrated fault diagnosis method | |
CN108680358A (en) | A kind of Wind turbines failure prediction method based on bearing temperature model | |
CN102221655B (en) | Random-forest-model-based power transformer fault diagnosis method | |
CN102799722B (en) | A kind of wind power plant low voltage ride-through capability emulation verification method | |
CN104200396B (en) | A kind of wind turbine component fault early warning method | |
CN108732528A (en) | A kind of digitalized electrical energy meter method for diagnosing faults based on depth confidence network | |
CN106786530A (en) | A kind of method for adapting to the fast quick-recovery of power network under extreme weather conditions | |
CN107563069A (en) | A kind of wind power generating set intelligent fault diagnosis method | |
CN107422266A (en) | A kind of method for diagnosing faults and device of high capacity cell energy-storage system | |
Haque et al. | Deep learning based micro-grid fault detection and classification in future smart vehicle | |
CN102609764A (en) | CPN neural network-based fault diagnosis method for stream-turbine generator set | |
CN109633370A (en) | A kind of electric network failure diagnosis method based on fault message coding and fusion method | |
CN109061391A (en) | A kind of electric network failure diagnosis method and system based on computer vision tidal current chart | |
CN105117591A (en) | Power device running status analysis method | |
CN106230378A (en) | A kind of diagnostic method of photovoltaic plant group string fault | |
CN103852255B (en) | Based on neutral net wind power generating set typical case's drive failures intelligent diagnosing method | |
CN106383947A (en) | Method for quickly obtaining dynamic equivalence parameters of power collection network of wind power plant | |
CN106292631A (en) | A kind of PWM rectifier fault diagnosis system based on neutral net | |
Hare et al. | A review of faults and fault diagnosis in micro-grids electrical energy infrastructure | |
CN113468728B (en) | Variable pitch system fault prediction method based on neural network | |
CN105954616B (en) | Photovoltaic module method for diagnosing faults based on external characteristics electric parameter |
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 |