CN105882687A - Method for analyzing categories of faults of point machines - Google Patents
Method for analyzing categories of faults of point machines Download PDFInfo
- Publication number
- CN105882687A CN105882687A CN201610316708.7A CN201610316708A CN105882687A CN 105882687 A CN105882687 A CN 105882687A CN 201610316708 A CN201610316708 A CN 201610316708A CN 105882687 A CN105882687 A CN 105882687A
- Authority
- CN
- China
- Prior art keywords
- point machine
- fault category
- signal
- audio
- described step
- 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
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B61—RAILWAYS
- B61L—GUIDING RAILWAY TRAFFIC; ENSURING THE SAFETY OF RAILWAY TRAFFIC
- B61L5/00—Local operating mechanisms for points or track-mounted scotch-blocks; Visible or audible signals; Local operating mechanisms for visible or audible signals
- B61L5/06—Electric devices for operating points or scotch-blocks, e.g. using electromotive driving means
Landscapes
- Engineering & Computer Science (AREA)
- Mechanical Engineering (AREA)
- Measurement Of Mechanical Vibrations Or Ultrasonic Waves (AREA)
Abstract
The invention discloses a method for analyzing categories of faults of point machines. The method has the advantages that the method does not need to intervene in electric control systems of the point machines, results are reliable and convictive, properties of the faults and locations of the faults of the point machines can be determined by the method, and accordingly effective predication, analysis and judgment of faults of equipment can be conducted to help maintenance staff eliminate the faults in advance and avoid possible hazards and loss, and theoretical data can be provided for analyzing frequently-occurred faults of the point machines.
Description
Technical field
The present invention relates to a kind of analysis method, the fault category being specifically related to point machine analyzes method；Belong to track
Technical field of transportation.
Background technology
Along with city underground and the fast development of high-speed railway, rail transit network complicates all the more, in track lane change or fall
The position of head generally requires the one or more point machines of use and carries track switch, thus, the use of point machine
Very high frequency, easy fatigue damage.In prior art, the operation maintenance to point machine lacks effective early warning and analysis side
Method, it is impossible to realize preventative maintenance and protection, can only rush to repair after fault occurs laggard behaviour, cause traffic network to there will be not
The predictable short time paralyses；And, the scheme afterwards rushed to repair cannot provide any reference to the generation of fault and prevention, unfavorable
Technological improvement in goat.Along with the encryption of track traffic running interval, to the maintenance of goat and quality requirement increasingly
Height, traditional " mending the fold after the sheep is lost " mode is needed change badly, the most also to be provided more preferable thinking to the trouble-saving of point machine.
Summary of the invention
For solving the deficiencies in the prior art, it is an object of the invention to provide the fault category analysis of a kind of point machine
Method, to providing theoretical foundation for the trouble-saving of goat and analysis.
In order to realize above-mentioned target, the present invention adopts the following technical scheme that:
The fault category of the point machine of the present invention analyzes method, comprises the steps:
S1, utilize audio collection module gather point machine action time audio signal, send into signal amplification circuit enter
Row signal amplifies, and then stores to memory module after analog-to-digital conversion；
S2, signal processing module carry out band-pass filter, signal framing, FFT one by one to the signal in memory module
Conversion, normalization and cepstrum promote, and finally calculate condition code, are supported vector machine training according to condition code and generate model literary composition
Part also carries out classification storage；
S3, audio collection module Real-time Collection point machine work time audio signal and simultaneously caching the n second, use and
The method that step S2 is identical calculates the condition code of real-time audio signal, and transfers model file and compare, and is sentenced by signal
Disconnected module judges whether point machine is in normal operating conditions；
S4, etc. next section audio data to be collected repeat step S3；If point machine is in malfunction, then same
Time transfer the n second audio frequency cached before, and the fault category of this audio frequency judged and combine, exporting failure code.
Preferably, in abovementioned steps S3, audio buffer time n≤10s, so facilitate follow-up carry out fault category analyze time
Transfer original audio in time.
Specifically, in abovementioned steps S1, audio collection module includes a sound pick-up and mounting bracket, audio collection process
In, the outer wall of goat is close to by sound pick-up, it is not necessary to accesses the electric control system of point machine, substantially increases testing result
Reliability and accuracy.
More preferably, in abovementioned steps S2, model file is divided into following two classes: point machine normally works and track switch turns
Rut machine is in malfunction.When being in malfunction, fault category includes but not limited to following 8 classes: point machine does not pushes away
Inlet bit, point machine push excessively, pine crossed by the first bolt, the first bolt tension, the second bolt cross pine, the second bolt mistake
Tightly, interorbital card have foreign matter and point machine work time do not send sound, every class knocking noise and position are the poorest
Not, the condition code therefore calculated also is different, gives the feasibility in theory that fault category is analyzed.
It is further preferred that in abovementioned steps S3, carried out in real time by the decoding call back function of SDK bag in DVR
The collection of audio signal.
Further, in abovementioned steps S2, the detailed process of signal framing is: N number of sample point first assembles a sight
Surveying unit, referred to as sound frame, have one section of overlapping area between two adjacent tone frames, this overlapping area contains M sample point, M's
Value is the 1/2 or 1/3 of N.
Yet further, in abovementioned steps S2, Energy distribution audio signal being transformed on frequency domain by FFT is come
Analyzing, use and take advantage of Hamming window to add the continuity of forte frame left and right end, the signal of sound frame is S (n), n=0,1 ... N-1, N
For sound frame size, being S ' (n)=S (n) * W (n) after taking advantage of Hamming window, the formula of Hamming window W (n) is as follows: W (n, a)=(1-a)-a*
Cos (2pi*n/ (N-1)), wherein, a represents the constant parameter of setting, 0 ＜ a ＜ 1.
Further, in abovementioned steps S2, normalized detailed process is: enter in the range of data are mapped to (-1,1)
Row processes, and dimension expression formula will be had to be changed into dimensionless expression formula:Its
In, D ' (i) represents the result after normalized, and D (i) represents current value, and D represents the data set needing normalized, U
Representing the normalized data upper limit, L represents normalized data lower limit.
Additionally, in abovementioned steps S2, the detailed process that cepstrum promotes is: energy frequency spectrum energy to be multiplied by one group 20 the logical filters of band
Ripple device, tries to achieve the logarithmic energy of each bandpass filter output
Wherein, M represents the quantity of bandpass filter, HmK () is the frequency response of bandpass filter,
In formula
In formula, x (n) is the audio signal of input, this
Place N represents counting of Fourier transform；Then, 20 logarithmic energy are obtained MFCC coefficient through discrete cosine transform, obtains the cepstrum on L rank
Parameter, L value is 12, and discrete cosine transform formula is as follows:
Show that the feature of each sound frame has 13 dimensions, comprise 1 logarithmic energy and 12 parameters of cepstrums, then ask parameters of cepstrum relative to
The slope of time, formula is as follows:Wherein, CtRepresent t position
Cepstrum coefficient；Finally, add that residual quantity computing i.e. produces the characteristic vector of 26 dimensions, obtain condition code.
The invention have benefit that: the fault category of the present invention analyzes method without getting involved the electrical control of goat
System, result is relatively reliable and convincing, be can determine nature of trouble and the position of point machine by the method, thus
The fault of equipment is effectively predicted, analyzes and judged, contributes to maintainer and eliminate fault in advance, it is to avoid possible harm
And loss, gross data can also be provided for the accident analysis occurred frequently of goat simultaneously.
Accompanying drawing explanation
Fig. 1 is the flow chart of the fault category analysis method of the point machine of the present invention.
Detailed description of the invention
Below in conjunction with the drawings and specific embodiments, the present invention made concrete introduction.
Seeing Fig. 1, the fault category of the point machine of the present invention is analyzed method and is comprised the steps:
S1, utilize audio collection module gather point machine action time audio signal, send into signal amplification circuit enter
Row signal amplifies, and then stores to memory module after analog-to-digital conversion.
Specifically, audio collection module includes a sound pick-up and mounting bracket, and during audio collection, sound pick-up is close to turn
The outer wall of rut machine, it is not necessary to access the electric control system of point machine, substantially increases the confidence level of testing result with accurate
Property.
S2, signal processing module carry out band-pass filter one by one to the signal in memory module, to carry out frequency spectrum
Smoothing, and harmonic carcellation, highlight the formant of original voice, can reduce subsequent arithmetic amount simultaneously；Then through signal framing,
FFT, normalization and cepstrum promote, and finally calculate condition code, are supported vector machine training according to condition code and generate mould
Type file also carries out classification storage；Here model file can be divided into following two classes: point machine normally works and track switch turns
Rut machine is in malfunction.
Wherein, when point machine is in malfunction, classify according to fault category, fault category include but not
It is defined in following 8 classes: point machine does not pushes away inlet bit, point machine pushes excessively, the first bolt crosses pine, the first bolt mistake
Tightly, pine crossed by the second bolt, the second bolt tension, interorbital card have foreign matter and do not sends sound during point machine work, real
During the detection of border, if any other fault types, it is directly added into a new fault category and failure code.
S3, audio collection module are by the decoding call back function Real-time Collection point machine of SDK bag in DVR
Audio signal during work simultaneously caching n second (n≤10s), uses the method identical with step S2 to calculate real-time audio and believes
Number condition code, and transfer model file and compare, judge whether point machine is just in by signal judge module
Often duty；
S4, etc. next section audio data to be collected repeat step S3；If point machine is in malfunction, then same
Time transfer the n second audio frequency cached before, and the fault category of this audio frequency judged and combine, exporting failure code.
In the present invention, the calculating process of condition code is innovated, so can be obtained the feature disturbed less, precision is high
Code, thus improve the accuracy of judgement, specifically include following aspects:
(1), in step s 2, the detailed process of signal framing is: first by N (N is usually 256 or 512) individual sampling point set
Synthesize an observation unit, referred to as sound frame (Frame), there is between two adjacent tone frames one section of overlapping area, this overlapping area bag
Having contained M sample point, the value of M is the 1/2 or 1/3 of N, is thus avoided that the change of adjacent two sound frames is excessive, improves sound
Frame continuity, thus improve the accuracy that condition code is extracted further.
(2), due to audio signal change in time domain it is difficult to find out its characteristic, thus, in step s 2, pass through FFT
The Energy distribution that audio signal is transformed on frequency domain by conversion is analyzed, and uses and takes advantage of Hamming window to add the continuous of forte frame left and right end
Property, the signal of sound frame is S (n), n=0,1 ... N-1, N are sound frame size, are S ' (n)=S (n) * W (n) after taking advantage of Hamming window,
The formula of Hamming window W (n) is as follows: (n, a)=(1-a)-a*cos (2pi*n/ (N-1)), wherein, a represents the constant ginseng of setting to W
Number, 0 ＜ a ＜ 1.
(3), normalization: process in the range of data are mapped to (-1,1), dimension expression formula will be had to be changed into immeasurable
Guiding principle expression formula:Wherein, D ' (i) represents the result after normalized, D
I () represents current value, D represents the data set needing normalized, and U represents the normalized data upper limit 1, and L represents normalizing
The data lower limit-1 changed.
(4), cepstrum promotes (merging residual quantity parameters of cepstrum): energy frequency spectrum energy is multiplied by one group of 20 bandpass filtering
Device, tries to achieve the logarithmic energy of each bandpass filter output
Wherein, M represents the quantity of bandpass filter,For the frequency response of bandpass filter,
In formula
In formula, x (n) is the audio signal of input, herein
N represents counting of Fourier transform；Then, 20 logarithmic energy are obtained MFCC coefficient through discrete cosine transform, obtains the cepstrum on L rank
Parameter, L value is 12, and discrete cosine transform formula is as follows:
Show that the feature of each sound frame has 13 dimensions, comprise 1 logarithmic energy and 12 parameters of cepstrums, then ask parameters of cepstrum relative to
The slope of time, formula is as follows:Wherein, CtRepresent t position
Cepstrum coefficient；Finally, add that residual quantity computing i.e. produces the characteristic vector of 26 dimensions, obtain condition code.
To sum up, the fault category analysis method of the present invention is without getting involved the electric control system of goat, and result more may be used
By with convincing, being can determine nature of trouble and the position of point machine by the method, thus the fault of equipment entered
Row is effectively predicted, analyzes and is judged, contributes to maintainer and eliminates fault in advance, it is to avoid possible harm and loss, also simultaneously
Gross data can be provided for the accident analysis occurred frequently of goat.
The general principle of the present invention, principal character and advantage have more than been shown and described.The technical staff of the industry should
Understanding, above-described embodiment limits the present invention the most in any form, and the mode of all employing equivalents or equivalent transformation is obtained
Technical scheme, all falls within protection scope of the present invention.
Claims (10)
1. the fault category of point machine analyzes method, it is characterised in that comprise the steps:
S1, utilize audio collection module gather point machine action time audio signal, send into signal amplification circuit carry out letter
Number amplify, then store to memory module after analog-to-digital conversion；
S2, signal processing module the signal in memory module is carried out one by one band-pass filter, signal framing, FFT,
Normalization and cepstrum promote, and finally calculate condition code, are supported vector machine training according to condition code and generate model file also
Carry out classification storage；
S3, audio collection module Real-time Collection point machine work time audio signal and simultaneously caching the n second, employing and step
Method identical for S2 calculates the condition code of real-time audio signal, and transfers model file and compare, and judges mould by signal
Block judges whether point machine is in normal operating conditions；
S4, etc. next section audio data to be collected repeat step S3；If point machine is in malfunction, adjust the most simultaneously
Take the n second audio frequency cached before, and the fault category of this audio frequency is judged and combines, export failure code.
The fault category of point machine the most according to claim 1 analyzes method, it is characterised in that described step S3
In, audio buffer time n≤10s.
The fault category of point machine the most according to claim 1 analyzes method, it is characterised in that described step S1
In, audio collection module includes that a sound pick-up and mounting bracket, described sound pick-up are close to the outer wall of goat.
The fault category of point machine the most according to claim 1 analyzes method, it is characterised in that described step S2
In, model file is divided into following two classes: point machine normally works and point machine is in malfunction.
The fault category of point machine the most according to claim 1 analyzes method, it is characterised in that described step S4
In, fault category includes following 8 classes: point machine does not pushes away inlet bit, point machine pushes excessively, pine crossed by the first bolt,
Pine crossed by first bolt tension, the second bolt, the second bolt tension, interorbital card have foreign matter and do not sends out during point machine work
Go out sound.
The fault category of point machine the most according to claim 1 analyzes method, it is characterised in that described step S3
In, carry out the collection of real-time audio signal by the decoding call back function of SDK bag in DVR.
The fault category of point machine the most according to claim 1 analyzes method, it is characterised in that described step S2
In, the detailed process of signal framing is: N number of sample point first assembles an observation unit, referred to as sound frame, two adjacent tone frames it
Between there is one section of overlapping area, this overlapping area contains M sample point, and the value of M is the 1/2 or 1/3 of N.
The fault category of point machine the most according to claim 1 analyzes method, it is characterised in that described step S2
In, Energy distribution audio signal being transformed on frequency domain by FFT is analyzed, and uses and takes advantage of Hamming window left to add forte frame
The continuity of right-hand member, the signal of sound frame is S (n), n=0,1 ... N-1, N are sound frame size, take advantage of after Hamming window for S ' (n)=
S (n) * W (n), the formula of Hamming window W (n) is as follows: (n, a)=(1-a)-a*cos (2pi*n/ (N-1)), wherein, a represents and sets W
Fixed constant parameter, 0 ＜ a ＜ 1.
The fault category of point machine the most according to claim 1 analyzes method, it is characterised in that described step S2
In, normalized detailed process is: process in the range of data are mapped to (-1,1), and dimension expression formula will be had to be changed into nothing
Dimension expression formula:Wherein, D ' (i) represents the result after normalized,
D (i) represents current value, and D represents the data set needing normalized, and U represents the normalized data upper limit, and L represents normalizing
The data lower limit changed.
The fault category of point machine the most according to claim 1 analyzes method, it is characterised in that described step S2
In, the detailed process that cepstrum promotes is: energy frequency spectrum energy is multiplied by one group of 20 bandpass filter, tries to achieve each bandpass filtering
The logarithmic energy of device outputWherein, M represents that band is logical
The quantity of wave filter, HmK () is the frequency response of bandpass filter,
In formula, x (n) is the audio signal of input, and N represents counting of Fourier transform herein；Then, by 20 logarithmic energy through discrete
Cosine transform obtains MFCC coefficient, obtains the cepstrum parameter on L rank, and L value is 12, and discrete cosine transform formula is as follows:Draw the feature of each sound frame
Having 13 dimensions, comprise 1 logarithmic energy and 12 parameters of cepstrums, then ask parameters of cepstrum relative to the slope of time, formula is such as
Under:Wherein, CtRepresent the cepstrum coefficient of t position；Finally, add
Upper residual quantity computing i.e. produces the characteristic vector of 26 dimensions, obtains condition code.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610316708.7A CN105882687A (en) | 2016-05-13 | 2016-05-13 | Method for analyzing categories of faults of point machines |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610316708.7A CN105882687A (en) | 2016-05-13 | 2016-05-13 | Method for analyzing categories of faults of point machines |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105882687A true CN105882687A (en) | 2016-08-24 |
Family
ID=56716269
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610316708.7A Pending CN105882687A (en) | 2016-05-13 | 2016-05-13 | Method for analyzing categories of faults of point machines |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105882687A (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107215357A (en) * | 2017-05-25 | 2017-09-29 | 同济大学 | A kind of switch breakdown Forecasting Methodology |
| CN108269249A (en) * | 2017-12-11 | 2018-07-10 | 深圳市智能机器人研究院 | A kind of bolt detecting system and its implementation |
| CN108583629A (en) * | 2018-05-04 | 2018-09-28 | 兰州容大信息科技有限公司 | A kind of railcar business fault handling method |
| CN109034046A (en) * | 2018-07-20 | 2018-12-18 | 国网重庆市电力公司电力科学研究院 | Foreign matter automatic identifying method in a kind of electric energy meter based on Acoustic detection |
| CN109242131A (en) * | 2017-07-10 | 2019-01-18 | 比亚迪股份有限公司 | Track switch information processing method and device |
| CN111684213A (en) * | 2018-10-22 | 2020-09-18 | 深圳配天智能技术研究院有限公司 | Robot fault diagnosis method, system and storage device |
| CN112067961A (en) * | 2020-10-13 | 2020-12-11 | 哈尔滨工业大学（深圳） | Arc fault detection method, system and storage medium |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2954578B1 (en) * | 1998-08-07 | 1999-09-27 | 東日本旅客鉄道株式会社 | Train approach warning receiver |
| CN101893667A (en) * | 2010-05-26 | 2010-11-24 | 广州市地下铁道总公司 | Fault detection system of AC turnout switch machine and method thereof |
| CN102156262A (en) * | 2011-03-24 | 2011-08-17 | 西南交通大学 | On-line monitoring system and method of railway switch machine motor |
| CN102930870A (en) * | 2012-09-27 | 2013-02-13 | 福州大学 | Bird voice recognition method using anti-noise power normalization cepstrum coefficients (APNCC) |
| CN103810374A (en) * | 2013-12-09 | 2014-05-21 | 中国矿业大学 | Machine fault prediction method based on MFCC feature extraction |
| JP5970327B2 (en) * | 2012-10-12 | 2016-08-17 | 公益財団法人鉄道総合技術研究所 | Abnormality detection apparatus, information processing apparatus, and abnormality detection system |
-
2016
- 2016-05-13 CN CN201610316708.7A patent/CN105882687A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2954578B1 (en) * | 1998-08-07 | 1999-09-27 | 東日本旅客鉄道株式会社 | Train approach warning receiver |
| CN101893667A (en) * | 2010-05-26 | 2010-11-24 | 广州市地下铁道总公司 | Fault detection system of AC turnout switch machine and method thereof |
| CN102156262A (en) * | 2011-03-24 | 2011-08-17 | 西南交通大学 | On-line monitoring system and method of railway switch machine motor |
| CN102930870A (en) * | 2012-09-27 | 2013-02-13 | 福州大学 | Bird voice recognition method using anti-noise power normalization cepstrum coefficients (APNCC) |
| JP5970327B2 (en) * | 2012-10-12 | 2016-08-17 | 公益財団法人鉄道総合技術研究所 | Abnormality detection apparatus, information processing apparatus, and abnormality detection system |
| CN103810374A (en) * | 2013-12-09 | 2014-05-21 | 中国矿业大学 | Machine fault prediction method based on MFCC feature extraction |
Non-Patent Citations (2)
| Title |
|---|
| 北京信息科技大学研究生院: "《研究生教育教学改革理论与实践》", 31 January 2016 * |
| 北方工业大学教务处: "《创想启动未来 北方工业大学2011年"北京市大学生科学研究与创业行动计划"研究报告论文集》", 30 November 2012 * |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107215357A (en) * | 2017-05-25 | 2017-09-29 | 同济大学 | A kind of switch breakdown Forecasting Methodology |
| CN107215357B (en) * | 2017-05-25 | 2018-10-26 | 同济大学 | A kind of switch breakdown prediction technique |
| CN109242131A (en) * | 2017-07-10 | 2019-01-18 | 比亚迪股份有限公司 | Track switch information processing method and device |
| CN108269249A (en) * | 2017-12-11 | 2018-07-10 | 深圳市智能机器人研究院 | A kind of bolt detecting system and its implementation |
| CN108583629A (en) * | 2018-05-04 | 2018-09-28 | 兰州容大信息科技有限公司 | A kind of railcar business fault handling method |
| CN109034046A (en) * | 2018-07-20 | 2018-12-18 | 国网重庆市电力公司电力科学研究院 | Foreign matter automatic identifying method in a kind of electric energy meter based on Acoustic detection |
| CN111684213A (en) * | 2018-10-22 | 2020-09-18 | 深圳配天智能技术研究院有限公司 | Robot fault diagnosis method, system and storage device |
| CN112067961A (en) * | 2020-10-13 | 2020-12-11 | 哈尔滨工业大学（深圳） | Arc fault detection method, system and storage medium |
| CN112067961B (en) * | 2020-10-13 | 2023-08-15 | 哈尔滨工业大学（深圳） | Arc fault detection method, system and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105882687A (en) | Method for analyzing categories of faults of point machines | |
| CN111354366B (en) | Abnormal sound detection method and abnormal sound detection device | |
| CN101907437B (en) | Wavelet difference algorithm-based cable fault localization method | |
| CN111626153A (en) | Integrated learning-based partial discharge fault state identification method | |
| CN108802535B (en) | Screening method, main interference source identification method and device, server and storage medium | |
| CN106017954A (en) | Turnout point machine fault early warning system and method based on audio analysis | |
| CN108169639A (en) | Method based on the parallel long identification switch cabinet failure of Memory Neural Networks in short-term | |
| CN107423692A (en) | A kind of rail corrugation fault detection method based on wavelet-packet energy entropy | |
| CN113297922B (en) | High-voltage switch cabinet fault diagnosis method, device and storage medium | |
| CN107798283A (en) | A kind of neural network failure multi classifier based on the acyclic figure of decision-directed | |
| CN107844067A (en) | A kind of gate of hydropower station on-line condition monitoring control method and monitoring system | |
| CN116778964A (en) | Power transformation equipment fault monitoring system and method based on voiceprint recognition | |
| CN106771598A (en) | A kind of Adaptive spectra kurtosis signal processing method | |
| CN103267652B (en) | Intelligent online diagnosis method for early failures of equipment | |
| CN108844612A (en) | A kind of identification method of transformer internal faults based on mathematical statistics probabilistic model | |
| CN112233695A (en) | Oiling machine abnormal sound analysis and fault early warning system based on artificial intelligence and big data | |
| CN116705039A (en) | AI-based power equipment voiceprint monitoring system and method | |
| CN1908987A (en) | Method for diagnosing accidents of waveshape recognition based substation locomotive traction circuit | |
| CN115376526A (en) | Power equipment fault detection method and system based on voiceprint recognition | |
| CN117668751B (en) | High-low voltage power system fault diagnosis method and device | |
| CN115078912A (en) | Method and system for detecting abnormity of roof high-voltage cable connector in real time and train | |
| WO2024119983A1 (en) | Anomality detection method and apparatus for cable terminal of rail vehicle | |
| CN116593829B (en) | Transmission line hidden danger on-line monitoring system based on data analysis | |
| CN117292713A (en) | Transformer voiceprint monitoring method and system based on end Bian Yun cooperation | |
| CN116524273A (en) | Method, device, equipment and storage medium for detecting draft tube of power station |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20160824 |
|
| RJ01 | Rejection of invention patent application after publication |