CN108267684A - A kind of Converter Fault Diagnosis method - Google Patents
A kind of Converter Fault Diagnosis method Download PDFInfo
- Publication number
- CN108267684A CN108267684A CN201810035745.XA CN201810035745A CN108267684A CN 108267684 A CN108267684 A CN 108267684A CN 201810035745 A CN201810035745 A CN 201810035745A CN 108267684 A CN108267684 A CN 108267684A
- Authority
- CN
- China
- Prior art keywords
- frequency converter
- fault
- image
- simulation
- circuit
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- 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
- G01R31/28—Testing of electronic circuits, e.g. by signal tracer
- G01R31/282—Testing of electronic circuits specially adapted for particular applications not provided for elsewhere
- G01R31/2829—Testing of circuits in sensor or actuator systems
Abstract
The invention discloses a kind of Converter Fault Diagnosis methods, it includes the following steps：Step 1：Circuit simulation is carried out to frequency converter；Step 2：The waveform image that will be obtained in above-mentioned fault simulation circuit extracts the individual features per piece image by feature extracting method；Step 3：Bayes classifier is constructed, determines the distributed model of described image feature；Step 4：By summarizing to the data in above-mentioned steps 1 to 3, fault database data are established.The present invention carries out fault simulation using the method for Computer Simulation, can improve the efficiency and quality of work.It can not need to carry out equipment investigation one by one when frequency converter really breaks down, but the rule emulated according to previous failures is timely repaired as a result, position and reason that failure occurs rapidly are judged.
Description
Technical field
The present invention relates to Converter Fault Diagnosis method fields, specially a kind of to be based on circuit simulation and Bayes classifier
Converter Fault Diagnosis method.
Background technology
Frequency converter has many advantages, such as control performance is superior, efficient, small, cheap, energy-saving effect is notable,
Current optimal speed regualtion of AC motor control device is acknowledged as, has been widely used in numerous areas.But frequency converter
Residing scene often bad environments, high temperature exothermic, grease are dirty, the electromagnetic interference of dust and alternation etc. both influences frequency converter
Frequency converter failure increasing with application of frequency converter, frequency converter also can easily be caused how to be diagnosed after breaking down and such as
What determines that abort situation just becomes to be highly desirable.Although there are many type of frequency converter, internal structure is also different, big
Most frequency converters are all made of rectification circuit, direct-flow intermediate circuit, inverter, four part of control circuit, their difference is only
Only it is control circuit and the realization difference of detection circuit and the difference of control algolithm.
When component aging, misoperation, environmental disturbances or artificial destruction, frequency converter also will appear various events
Hinder, can cause a serious accident under serious situation, bring the economy that can not be made up or social loss.It is main in frequency conversion speed-adjusting system
There are three types of fault modes:(1) electric fault, such as stator failure (disconnected phase, short circuit, electric leakage), rotor fault (rotor broken bar, ring
End cracking etc.), fault of converter (output short-circuit, IGBT open circuit etc.), dc-link capacitance aging or damage etc..(2) machinery event
Barrier, such as bearing fault, air gap eccentric centre.(3) sensor fault, such as current sensor or velocity sensor open circuit, sensor
Susceptibility reduction etc..Since main circuit is the position that frequency converter most easily breaks down, operating status is directly related to entire change
The safety and reliability of frequency device, and the structure of this part is relatively easy, therefore at present to the research of Converter Fault Diagnosis
It is concentrated mainly on main circuit part.
This just needs us when producing frequency converter using effective method for diagnosing faults, fast when frequency converter failure occurs
It is diagnosed to be its failure cause fastly, and protect and handle accordingly, equipment is made not shut down suddenly.Operating personnel are just in this way
It can have ample time to carry out maintenance of equipment, quickly to resume production, avoid accident, loss is preferably minimized.
Therefore, the fault diagnosis technology of frequency converter is studied, is had for the reliability and maintainability that improve frequency conversion speed-adjusting system great
Realistic meaning, be a urgent problem.
Invention content
The purpose of the present invention is to provide a kind of Converter Fault Diagnosis sides based on circuit simulation and Bayes classifier
Method is solved when frequency converter breaks down, it is impossible to rapidly carry out abort situation judgement, can not on-call maintenance the problem of.
To achieve the above object, technical solution provided by the invention：It includes the following steps：
Step 1：Circuit simulation is carried out to frequency converter, with reference to the principle that different frequency converter failures generate, failure is passed through not
Same mode is realized in artificial circuit to be come out, according to its generation as a result, obtaining original analysis data；
Step 2：The waveform image that will be obtained in above-mentioned fault simulation circuit extracts each width by feature extracting method
The individual features of image, and these features are combined as a feature vector, to represent corresponding image；
Step 3：Bayes classifier is constructed, the distributed model of described image feature is determined, so as to obtain the elder generation of image class
Test probability and class conditional probability；Corresponding posterior probability is calculated using Bayesian formula；Decision function is recycled to be expired
The image retrieval data required enough；
Step 4：By summarizing to the data in above-mentioned steps 1 to 3, fault database data are established, in the fault database
The phenomenon that being corresponded to containing different faults, the frequency converter by comparing phenomenon of the failure and fault database determine failure after breaking down
Type.
As a kind of preferred embodiment of the present invention, in the step 1, by Matlab softwares to the frequency converter
Carry out circuit simulation.
Compared with prior art, beneficial effects of the present invention are as follows：
1. Bayesian network model is combined with converter circuit emulation, fault diagnosis is carried out：
Fault simulation is carried out using the method for Computer Simulation, the efficiency and quality of work can be improved.When frequency converter is true
It is positive not need to carry out equipment investigation one by one when breaking down, but emulated according to previous failures
Rule as a result, to failure occur position and reason rapidly judged, timely repair, so as to ensure that system is normal
Work, does not delay production, while also save maintenance cost.
2. emulating image is classified using Bayes classifier
The classification of multiple and different features is carried out to emulation atlas, can not only improve whole classification performance, but also different
Grader performance in its respective proper subspace can be optimal, therefore also be able to embody preferable part differentiation energy
Power.
3. instructing state opportunity maintenance using Bayesian network diagnostic result, maintenance cost is substantially reduced：
State machine meeting maintenance policy, the advantages of being a kind of dynamic maintenance policy, combine State Maintenance and opportunity maintenance.
Since the implementation of strategies needs to be carried out according to the real-time state monitoring value of each component of frequency converter group, can avoid well
" cross and repair " and " owing to repair " problem.In addition, the strategy can be realized multipart while be repaired, can save well repair into
This, and fault power time can be saved.
Description of the drawings
Fig. 1 is the flow chart of the Converter Fault Diagnosis method of the present invention.
Specific embodiment
The specific embodiment of the disclosure is illustrated with reference to Fig. 1：
A kind of Converter Fault Diagnosis method of the present invention, includes the following steps：
Step 1：Circuit simulation is carried out to frequency converter, with reference to the principle that different frequency converter failures generate, failure is passed through not
Same mode is realized in artificial circuit to be come out, according to its generation as a result, obtaining original analysis data；
Step 2：The waveform image that will be obtained in above-mentioned fault simulation circuit extracts each width by feature extracting method
The individual features of image, and these features are combined as a feature vector, to represent corresponding image；
Step 3：Bayes classifier is constructed, the distributed model of described image feature is determined, so as to obtain the elder generation of image class
Test probability and class conditional probability；Corresponding posterior probability is calculated using Bayesian formula；Decision function is recycled to be expired
The image retrieval data required enough；
Step 4：By the summary to above-mentioned fault experience data, fault database data are established, are contained not in the fault database
The phenomenon that being corresponded to failure, comparison phenomenon of the failure determines fault type with fault database after the frequency converter breaks down.
In step 1, circuit simulation is carried out to the frequency converter by Matlab softwares.In Matlab softwares, malfunctioning module
Failure can not directly be generated.In physical fault, the external failure of single IGBT only has switching tube breakdown and switch on the whole
Two kinds of pipe open circuit.Switching tube breakdown shows as uncontrolled short circuit phenomenon, and switching tube open circuit be usually generated heat by switching tube it is excessive
Caused, remaining external performance phenomenon between two kinds of extreme failures is similar.Therefore, for the two of single switching tube
Kind failure, directly emulates its short trouble and open circuit to connect conducting wire by-pass switch pipe in a model and directly cut off switching tube
Failure.
For power electronic devices IGBT widely used at present, since in its simulation model, some necessity are joined
Number device production producers do not provide, so with emulating to study its switching process be not a nothing the matter, it is contained in
It is implemented in PWM frequency converter control systems just more difficult.Developing the AC system using single-phase precision voltage frequency converter
During control, since research object is frequency converter and its control system rather than the switching process of IGBT, effect conditionally
IGBT does not consider the reverse recovery characteristic of fast recovery diode, will restore two soon by four IGBT and four as perfect switch
The two-arm of pole pipe composition is treated as a subsystem, has received preferable effect.It is by voltage frequency converter topological structure
PWM single-phase bridges after its state equation is obtained, are imitated control system with the Simulink tool boxes in Matlab
Very, so that it is determined that the control block diagram of system, wave filter and regulator parameter, have finally carried out material object to emulating determining system
Experiment.Experimental waveform coincide substantially with simulation waveform, demonstrates the credibility of emulation.
In step 2 and three, for the waveform image that fault simulation obtains, we will be extracted every by feature extracting method
The individual features of piece image, such as shape, and feature is combined as a feature vector, to represent corresponding image.In structure
When making Bayes classifier, need to determine a kind of distributed model of characteristics of image first, so as to obtain the prior probability of image class
And class conditional probability.Bayesian formula is recycled to calculate corresponding posterior probability.Finally met using decision function
It is required that image searching result collection.
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 each embodiment can also be properly combined, forms those skilled in the art
The other embodiment being appreciated that.
Claims (2)
- A kind of 1. Converter Fault Diagnosis method, it is characterised in that：It includes the following steps：Step 1：Circuit simulation is carried out to frequency converter, with reference to the principle that different frequency converter failures generate, failure is passed through different Mode is realized in artificial circuit and is come out, and according to the waveform image of its generation, obtains original analysis data；Step 2：The waveform image that will be obtained in above-mentioned fault simulation circuit is extracted by feature extracting method per piece image Individual features, and these features are combined as a feature vector, obtain corresponding image data；Step 3：Construct Bayes classifier, determine the distributed model of characteristics of image, so as to obtain the prior probability of image and Class conditional probability；Corresponding posterior probability is calculated using Bayesian formula；The figure that decision function is recycled to be met the requirements As retrieval data；Step 4：By summarizing to the data in above-mentioned steps 1 to 3, fault database data are established, are contained in the fault database The phenomenon that different faults correspond to, the frequency converter by comparing phenomenon of the failure and fault database determine failure classes after breaking down Type.
- It is 2. according to claim 1, it is characterised in that：In step 1, the frequency converter is carried out by Matlab softwares Circuit simulation.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810035745.XA CN108267684B (en) | 2018-01-15 | 2018-01-15 | Frequency converter fault diagnosis method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810035745.XA CN108267684B (en) | 2018-01-15 | 2018-01-15 | Frequency converter fault diagnosis method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108267684A true CN108267684A (en) | 2018-07-10 |
| CN108267684B CN108267684B (en) | 2020-03-10 |
Family
ID=62775711
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810035745.XA Expired - Fee Related CN108267684B (en) | 2018-01-15 | 2018-01-15 | Frequency converter fault diagnosis method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108267684B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109521299A (en) * | 2018-11-27 | 2019-03-26 | 河南理工大学 | A kind of method of inverter intelligent fault reasoning |
| CN115703116A (en) * | 2021-08-10 | 2023-02-17 | 唐山新海纳自动化有限公司 | Deep maintenance process for prolonging service life of frequency converter |
| CN117638799A (en) * | 2023-12-14 | 2024-03-01 | 温岭市恒亚机电有限公司 | Remote variable frequency controller with leakage protection function and method |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20020144168A1 (en) * | 2001-03-30 | 2002-10-03 | Bleakley Thomas E. | Circuit for producing a variable frequency clock signal having a high frequency low jitter pulse component |
| CN101833324A (en) * | 2010-05-14 | 2010-09-15 | 武汉理工大学 | Intelligent fault diagnosis system in tread extrusion process and diagnosis method thereof |
| CN102253338A (en) * | 2011-05-30 | 2011-11-23 | 新疆大学 | Intelligent failure diagnosis method for frequency converter of wind power unit |
| CN102570392A (en) * | 2012-01-17 | 2012-07-11 | 上海电力学院 | Method for identifying exciting inrush current of transformer based on improved probability neural network |
| CN102722107A (en) * | 2011-03-29 | 2012-10-10 | 上海雷诺尔科技股份有限公司 | Intelligent integrated development and test system for high-voltage frequency converter and establishment method thereof |
| US20130257334A1 (en) * | 2012-03-28 | 2013-10-03 | Joy Mm Delaware, Inc. | Ground fault detection methods on variable frequency drive systems |
| US20140222212A1 (en) * | 2013-01-24 | 2014-08-07 | Trane International Inc. | Dynamic simulated motor |
| CN104698397A (en) * | 2015-03-16 | 2015-06-10 | 浙江万里学院 | Fault diagnosis method of multi-level inverter |
| CN107202956A (en) * | 2017-05-27 | 2017-09-26 | 武汉科技大学 | A kind of fault diagnosis for frequency conversion drives method based on m-Acetyl chlorophosphonazo feature |
-
2018
- 2018-01-15 CN CN201810035745.XA patent/CN108267684B/en not_active Expired - Fee Related
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20020144168A1 (en) * | 2001-03-30 | 2002-10-03 | Bleakley Thomas E. | Circuit for producing a variable frequency clock signal having a high frequency low jitter pulse component |
| CN101833324A (en) * | 2010-05-14 | 2010-09-15 | 武汉理工大学 | Intelligent fault diagnosis system in tread extrusion process and diagnosis method thereof |
| CN102722107A (en) * | 2011-03-29 | 2012-10-10 | 上海雷诺尔科技股份有限公司 | Intelligent integrated development and test system for high-voltage frequency converter and establishment method thereof |
| CN102253338A (en) * | 2011-05-30 | 2011-11-23 | 新疆大学 | Intelligent failure diagnosis method for frequency converter of wind power unit |
| CN102570392A (en) * | 2012-01-17 | 2012-07-11 | 上海电力学院 | Method for identifying exciting inrush current of transformer based on improved probability neural network |
| US20130257334A1 (en) * | 2012-03-28 | 2013-10-03 | Joy Mm Delaware, Inc. | Ground fault detection methods on variable frequency drive systems |
| US20140222212A1 (en) * | 2013-01-24 | 2014-08-07 | Trane International Inc. | Dynamic simulated motor |
| CN104698397A (en) * | 2015-03-16 | 2015-06-10 | 浙江万里学院 | Fault diagnosis method of multi-level inverter |
| CN107202956A (en) * | 2017-05-27 | 2017-09-26 | 武汉科技大学 | A kind of fault diagnosis for frequency conversion drives method based on m-Acetyl chlorophosphonazo feature |
Non-Patent Citations (1)
| Title |
|---|
| DANJIANG CHEN 等: "Fault Diagnosis of Three-level Inverter Based on Wavelet Analysis and Bayesian Classifier", 《2013 25TH CHINESE CONTROL AND DECISION CONFERENCE》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109521299A (en) * | 2018-11-27 | 2019-03-26 | 河南理工大学 | A kind of method of inverter intelligent fault reasoning |
| CN109521299B (en) * | 2018-11-27 | 2020-11-03 | 河南理工大学 | Intelligent fault reasoning method for inverter |
| CN115703116A (en) * | 2021-08-10 | 2023-02-17 | 唐山新海纳自动化有限公司 | Deep maintenance process for prolonging service life of frequency converter |
| CN117638799A (en) * | 2023-12-14 | 2024-03-01 | 温岭市恒亚机电有限公司 | Remote variable frequency controller with leakage protection function and method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108267684B (en) | 2020-03-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106443297B (en) | The decision tree SVM method for diagnosing faults of photovoltaic diode Clamp three-level inverter | |
| CN105548792B (en) | Matrix converter switch open method for diagnosing faults based on PREDICTIVE CONTROL | |
| Chen et al. | Logic-based methods for intelligent fault diagnosis and recovery in power electronics | |
| CN103235219B (en) | A kind of sub-module fault diagnostic method of modularization multi-level converter | |
| Duan et al. | Open-switch fault diagnosis and system reconfiguration of doubly fed wind power converter used in a microgrid | |
| CN103901301B (en) | The method for diagnosing faults of converter | |
| CN103728514B (en) | A kind of low voltage frequency converter method for diagnosing faults | |
| CN108267684A (en) | A kind of Converter Fault Diagnosis method | |
| Haque et al. | Deep learning based micro-grid fault detection and classification in future smart vehicle | |
| CN102567783A (en) | Expert fault analytical and diagnostic method of parallel mixed type power quality regulator | |
| CN106353573A (en) | Overcurrent fault monitoring protection device and method for flexible direct-current transmission inverter station | |
| Rajeswaran et al. | Hybrid artificial intelligence based fault diagnosis of svpwm voltage source inverters for induction motor | |
| Xiao et al. | Review of fault diagnosis and fault-tolerant control methods of the modular multilevel converter under submodule failure | |
| Xiong et al. | A monitoring and diagnostics method based on FPGA-digital twin for power electronic transformer | |
| Zhang | Current behavior-based open-switch fault on-line diagnosis of inverters in PMSM drive systems | |
| CN108963980A (en) | A kind of multi-modal failure separation method based on Fault Isolation library | |
| Guerrero et al. | A novel noninvasive failure-detection system for high-power converters based on SCRs | |
| Rayudu | A knowledge-based architecture for distributed fault analysis in power networks | |
| CN106019117A (en) | Open-circuit fault diagnosis method of cascaded STATCOM system IGBT | |
| Li et al. | Fault diagnosis and fault-tolerant control of photovoltaic micro-inverter | |
| CN211627761U (en) | High-speed permanent magnet synchronous motor loading device | |
| Mwangi et al. | Open Switch Fault Detection in Electric Vehicle Drives Using Support Vector Machine | |
| Ginart et al. | Self-healing from a PHM perspective | |
| Kezunovic et al. | Automated analysis of protective relay data | |
| Li et al. | A simple ANN-Based diagnosis method for open-switch faults in power converters |
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 |
Granted publication date: 20200310 Termination date: 20210115 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |