CN109649432A - Cloud platform rail integrity monitoring systems and method based on guided wave technology - Google Patents
Cloud platform rail integrity monitoring systems and method based on guided wave technology Download PDFInfo
- Publication number
- CN109649432A CN109649432A CN201910063557.2A CN201910063557A CN109649432A CN 109649432 A CN109649432 A CN 109649432A CN 201910063557 A CN201910063557 A CN 201910063557A CN 109649432 A CN109649432 A CN 109649432A
- Authority
- CN
- China
- Prior art keywords
- guided wave
- monitoring
- rail
- data
- matrix
- 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
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B61—RAILWAYS
- B61K—AUXILIARY EQUIPMENT SPECIALLY ADAPTED FOR RAILWAYS, NOT OTHERWISE PROVIDED FOR
- B61K9/00—Railway vehicle profile gauges; Detecting or indicating overheating of components; Apparatus on locomotives or cars to indicate bad track sections; General design of track recording vehicles
- B61K9/08—Measuring installations for surveying permanent way
Landscapes
- Engineering & Computer Science (AREA)
- Mechanical Engineering (AREA)
- Investigating Or Analyzing Materials By The Use Of Ultrasonic Waves (AREA)
Abstract
The invention discloses a kind of cloud platform rail integrity monitoring systems and method based on guided wave technology.Monitoring system includes front end monitoring modular, cloud monitoring server and browsing terminal, front end monitoring modular includes wave guide energy converter, temperature sensor and control cabinet, there is solar panel module, charge-discharge control circuit, energy-storage module, transmission interface circuit module, slave computer control circuit module, communication interface circuit module, guided wave transceiver module and signal conditioning module in control cabinet；The characteristic signal and coefficient matrix that strong correlation characterizes acoustic emission source at rail different structure will be extracted by noise reduction first, then carries out dimensionality reduction and noise reduction reconstruct obtains the guided wave signals of characterization rail failure, obtain damage position, realize rail Integrity Verification.The present invention can effectively realize the reliable monitoring of rail integrality, substantially increase the cross-region and real-time of monitoring, have important practical significance and engineering value.
Description
Technical field
The present invention relates to a kind of rail integrity monitoring systems and method more particularly to a kind of clouds based on guided wave technology
Platform rail integrity monitoring systems and method.
Background technique
In recent years, on the one hand with the fast development of national economy, the railway traffic concerning lifelines of the national economy is obtained
Unprecedented fast development；On the other hand with the fast development of railway traffic, rate of traffic flow, the speed of service and loading capacity
Etc. having been greatly improved, load, impact that the rail as track important component will certainly be subject to etc. similarly increases,
The probability for inevitably rail being caused to damage improves, and proposes tightened up requirement to reliable, the safe operation of track, is
This, there is an urgent need to propose it is effective, reliably for in-service railway track detection and military service status monitoring technology and method.
Supersonic guide-wave technology is widely used in each with the characteristics of its long range, a wide range of, total cross-section detection, single-ended transmitting-receiving
Among the non-destructive testing of each industry of row and on-line monitoring.Whether material or element type are all very suitable to using leading railway track
The monitoring and evaluation of wave technology progress working condition.However, it is contemplated that steel-rail structure feature is complex, fastener bonding state compared with
More, the point-to-point types nondestructiving detecting means such as traditional ultrasound detection, Magnetic Flux Leakage Inspecting, machine vision and Liquid penetrant testing are difficult to meet real
In the monitoring of border to timeliness, cloud is online, reliability, cross-region are strict on a large scale.Simultaneously, it is contemplated that railway line road transport
Row is busy, carries out the artificial detection under line to track using the skylight period and has been difficult to meet increasingly developed iron with monitoring technology
Road is actually detected and monitoring requirements.Here the related early stage research with above-mentioned drawback, Publication No. CN104535652A are provided
A kind of " electromagnetic acoustic technology steel rail defect detection side of " a kind of rail failure detection method ", Publication No. CN101398410A
Method and its device ", a kind of " rail examination of the image co-registration of laser-ultrasound and high-speed camera of Publication No. CN104237381A
Method " and " method and apparatus of live ultrasonic examination railway track " patent of Publication No. CN102084245A etc..Information skill
The tremendous development of today of the rapid development of art, especially internet, cloud computing, the 5th third-generation mobile communication technology etc. are based on cloud
End the online, trans-regional of platform, real time monitoring increasingly become the focus and forward position focus of all trades and professions application.Carry out base
In the popularization and application of the rail Integrity Verification method and system of cloud platform, also will be helpful to promote China railways track online
The intellectually and automatically degree of detection and monitoring technology.
Summary of the invention
The present invention for the problems in above-mentioned background technology and defect, proposes a kind of cloud platform based on guided wave technology
Rail integrity monitoring systems and method are able to achieve the real-time online cross-region monitoring to railway gauze track service state.
As shown in Fig. 2, the present invention is achieved through the following technical solutions:
One, a kind of cloud platform rail integrity monitoring systems based on guided wave technology:
The monitoring system includes front end monitoring modular, cloud monitoring server and browsing terminal, and the front end monitors mould
Block includes wave guide energy converter, temperature sensor and control cabinet, have in control cabinet solar panel module, charge-discharge control circuit,
Energy-storage module, transmission interface circuit module, slave computer control circuit module, communication interface circuit module, guided wave transceiver module and
Signal conditioning module；Wherein, wave guide energy converter, temperature sensor are mounted on rail to be measured, slave computer control circuit module
It is electrically connected, is led with temperature sensor, charge-discharge control circuit and guided wave transceiver module respectively by transmission interface circuit module
Wave transducer is electrically connected through signal conditioning module and guided wave transceiver module, solar panel module and charge-discharge control circuit electricity
Gas connection, charge-discharge control circuit and energy-storage module are electrically connected, and slave computer control circuit module passes through communication interface circuit mould
Block accesses cloud monitoring server.
The wave guide energy converter sends supersonic guide-wave to rail, and supersonic guide-wave encounters defect reflection after rail propagation and produces
Raw echo-signal, echo-signal guided wave energy converter receives the guided wave signals as monitoring rail, and is sent to slave computer control
Circuit module；The real time temperature of rail near the temperature sensor detection wave guide energy converter, is sent to slave computer control
Circuit module.
The described solar panel module acquisition solar energy is converted to electric energy, by charge-discharge control circuit to energy-storage module
Charging, then by energy-storage module be that wave guide energy converter, temperature sensor and entire control cabinet are powered.
The browsing terminal includes computer/mobile phone terminal.
Two, a kind of cloud platform rail Integrity Verification method based on guided wave technology, using above system, then method
It is specific as follows:
S1, monitoring method first part will extract the spy that strong correlation characterizes acoustic emission source at rail different structure by noise reduction
Reference number and coefficient matrix, steps are as follows:
S1.1: according to rail geometry to be measured and physical attribute, the mould of the used supersonic guide-wave of wave guide energy converter is preset
State and frequency preset the acquisition monitoring parameters of wave guide energy converter, are passed by interruption continuous acquisition wave guide energy converter and temperature
Sensor obtains the guided wave signals and temperature signal of monitoring rail；
The acquisition parameter of wave guide energy converter includes acquisition time duration A, data storage capacity B, times of collection C, collection period
Ts, monitoring temperature D, monitoring time interval E, distance threshold F, iteration total degree N, amplitude thresholds Z, sample frequency Fs and ultrasound are led
Wave velocity V.
It further include data storage capacity B in the acquisition parameter of the wave guide energy converter, data storage capacity B leads for what is stored
The data capacity of wave signal and temperature signal.
Acquisition time duration A is greater than collection period TS.
S1.2: number C, each collection period T are acquired altogether in acquisition time duration ASInterior guided wave signals X, leads
Wave signal X and collected temperature signal of same time, are transmitted to cloud monitoring server, according to C signal data dimension
Data format Y=(X1, X2..., Xc)TIt carries out storage and forms guided wave monitoring data, using times of collection as the dimension of signal data
Degree；
Time interval between adjacent acquisition time duration A is monitoring time interval E.
The guided wave signals substantial decomposition is to characterize the linear superposition of different sound-source signals: Y=MR+N (t), in order to ask
Weight matrix M harmony source signal data R is obtained, N (t) indicates noise signal data.The present invention does not need acquisition steel-rail structure here
Different complex environments are in when sufficiently complete descended the guided wave signals of redundancy to be handled, can be arbitrarily containing damage defect
It is directly extracted in guided wave signals and obtains sound-source signal data R.
S1.3: for acquiring two groups of guided wave monitoring data (infalls before and after acquisition at different acquisition time duration A
Reason), the signal data dimension of two groups of guided wave monitoring data is q and w, and two groups of guided wave monitoring data are superimposed jointly according to line direction
Then the data Z ' to be analyzed of the new more higher-dimension of composition carries out feature scaling and sparse to data Z ' to be analyzed according to following formula
The standardization of processing obtains the multidimensional data Z to be analyzed for the dimension normalization that mean value is 0, variance is 1:
Wherein, E (z) and σ is respectively the mean value and standard deviation of data Z ' to be analyzed, ZiIt indicates in multidimensional data Z to be analyzed
I-th group of signal, Zi' indicate i-th group of signal in data Z ' to be analyzed；
S1.4: carry out objective optimization: data Z to be analyzed to multidimensional carries out blind source separating, obtains the sound of characterization rail failure
Source signal data:
Weight coefficient of the sound-source signal in guided wave monitoring data is obtained using following formula iterative solution first:
S1.4.1: the weight coefficient W that one two norm of initialization are 10With iteration count n=1；
S1.4.2: solution is iterated according to following formula:
Wn=E { Z (WT n-1Z)3}-3Wn-1, n=n+1
Wherein, WnIndicate that the weight coefficient vector that nth iteration obtains, Z indicate multidimensional data to be analyzed, E { } indicates the phase
Hope function；δ1n~δLnRespectively indicate n-th of weight coefficient vector WnIn coefficient value, L indicate strong correlation weight coefficient sum；
S1.4.3: every time to weight coefficient vector W after iterative solutionnIt is normalized, then judges:
If | WT nWn-1| 1 is not converged on, then re-execute the steps S1.4.2；
If | WT nWn-1| converge on 1 and under the conditions of meeting the number of iterations ordinal number n less than iteration total degree N, then output is current
Weight coefficient vector W under the number of iterationsnAs strong correlation weight coefficient, and it is added in coefficient matrix W* and continues to walk
Q+w group strong correlation weight system is obtained until the number of iterations ordinal number n is equal to iteration total degree N then iteration ends in rapid S1.4.2
Number；
S1.4.4: final to obtain coefficient matrix W*, W*=(W1, W2..., Wq+w)T, characterized according to formula R=W* × Z
Sound-source signal data R, the R=(r of different sound sources in rail to be measured1,r2,...,rL)T, r1,r2,...,rLIndicate sound-source signal number
According to the sound source subsignal for corresponding to each guided wave signals of guided wave monitoring data in R；
S1.5: the generalized inverse matrix for solving coefficient matrix W* obtains weight matrix M, indicates are as follows:
In formula, β11~βqLThe q row coefficient value that the guided wave monitoring data that dimension is q correspond to sound-source signal is respectively indicated,
As one group of coefficient sets；α11~αwLIt indicates that dimension corresponds to the w row coefficient value of sound-source signal for the guided wave monitoring data of w, makees
For another group of coefficient sets；Q row coefficient value and w row coefficient value correspond respectively to the signal number of two groups of front and back guided wave monitoring data
According to dimension q and w；
Construction is below with reference to matrix K:
Wherein, the number of element -1 is q in each column and 1 number of element is w；
S1.6: carrying out similarity measurement, calculates separately a column p of weight matrix MiColumn k corresponding with R-matrix KiIt
Between similitude:
For each column vector p of weight matrix MiWith each column vector k of R-matrix KiUse and step S1.3 phase
Respective ranks vector p after being standardized to obtain standardization with mode* iAnd k* i, then obtained according to following formula
Obtain similarity distance:
θi=| 1-p* i×k* i|
In formula, θiA column vector p after indicating the standardization of weight matrix M* iStandardization corresponding with R-matrix K
A treated column vector k* iBetween similarity distance；
Then by all similarity distance result compositional similarity distance set ξ={ θ1,θ2,...,θL}；
S2, the guided wave signals for carrying out dimensionality reduction and noise reduction reconstruct acquisition characterization rail failure, obtain damage position:
S2.1: feature extraction dimension-reduction treatment is carried out to weight matrix M and obtains strong correlation sound-source signal:
The satisfaction relationships θ such as not is extracted from similarity distance set ξ according to distance threshold FiThe similarity distance θ of < FiIt is corresponding
Weight matrix M in each column vector and each sound source subsignal in sound-source signal data R, from the column that extract to
Amount forms following hurt matrix P(q+w)×H, following strong correlation sound source matrix is formed by the sound source subsignal extracted
Wherein, H indicates the quantity of strong correlation sound-source signal, as strong correlation sound source component；
The reconstruct sound-source signal of number needed for automatically extracting out may be implemented by threshold value F in the present invention, obtains damage letter
Number, similarity distance is shorter to show that required sound-source signal is more related to damage.
S2.2: hurt matrix P(q+w)×HTwo groups of coefficient sets in the relatively small coefficient sets of matrix system numerical value are given up, protect
The relatively large coefficient sets of matrix coefficient are stayed, thus by hurt matrix P(q+w)×HIt extracts and obtains construction coefficient matrix Pw×H；
Below by taking the coefficient sets of w coefficient value are corresponding as an example, i.e., its coefficient value is relatively large:
S2.3: it is reconfigured according to construction coefficient matrix and strong correlation sound source matrix super containing track damage characteristic information
Guided Waves damage signal Ydefect；
To construction coefficient matrix Pw×HEach column coefficient value is carried out according to the following equation to take mean value that interference is gone to handle, is obtained new
One-dimensional row vector P1×H NEW, it thus is avoided that the abnormal interference for being mixed into noise and calculated result:
The supersonic guide-wave damage signal Y containing track damage characteristic information is acquired further according to following formuladefect:
S2.4: according to damage signal YdefectProcessing obtains envelope information,
If envelope range value thinks that rail to be measured has damage greater than the part of amplitude thresholds Z in envelope information；
If part of the envelope range value no more than amplitude thresholds Z thinks that damage is not present in rail to be measured in envelope information；
In rail to be measured there are under degree of impairment, supersonic guide-wave damage signal YdefectIt is right to there is damage institute in rail to be measured
The acquisition moment t that should locate obtains the positioning of rail failure using following formula processing:
S=t × V/2
T=1/Fs × i, i=0,1 ..., num, num=Ts × Fs
Wherein, Ts indicates that the collection period of wave guide energy converter, Fs indicate the sample frequency of wave guide energy converter, and V indicates guided wave
The supersonic guide-wave speed that energy converter issues, i indicate that the ordinal number of i-th of acquisition, num indicate that the sum all acquired, s indicate rail
Existing distance of the damage in position to wave guide energy converter；
Rail Integrity Verification of the invention is using rail failure and its position location as the characterization of integrality.
The interruption continuous acquisition is continuously to carry out number C, data at each acquisition interval duration A according to monitoring requirements
Length is the guided wave signals of Ts × Fs and the acquisition of temperature.
When the data of guided wave signals and temperature signal that the slave computer control circuit module stores reach data storage
When measuring B, monitoring gained supersonic guide-wave big data will be excavated, be failed to report using following setting cross processing analysis method reduction
Rate finds out damage:
A: the guided wave signals at identical monitoring temperature D but different time carry out processing analysis according to step, obtain rail
The positioning of damage；
B: at fixed monitoring time interval E, two groups of guided waves that the time difference is monitoring time interval E between monitor number
Processing analysis is carried out according to according to step, obtains the positioning of rail failure；
C: randomly selecting in the data of storage two groups of guided wave monitoring data according to step and carry out processing analysis, obtains rail damage
The positioning of wound.
The beneficial effects of the present invention are:
The present invention have it is online in real time, cross-region, low cost, high robust the characteristics of, not only greatly improve tradition
Detection and deficiency and problem present in monitoring means and technology, at the same improve the assessment monitoring of railway gauze service state certainly
Dynamicization and intelligent level；Detection monitoring efficiency, the range of railroad track will be greatly improved in the intrinsic advantages of supersonic guide-wave technology
And system reliability；The professional platform independence and far of rail on-line monitoring will be greatly improved in exploitation based on cloud monitoring algorithm and application
Journey convenience.
The monitoring processing of fracture, disconnected line class defect not only may be implemented in the present invention, while may be implemented in slow transformation
Corrosion class defect information monitoring trend evaluation, in particular by cross processing method substantially increase to erosion type lesions
Differentiation and tracking, improve monitoring accuracy, avoid in conventional monitoring methods to corrosion growth course in monitoring accuracy it is inadequate
The problem of with being difficult to find.
Monitoring system intelligent degree of the invention is high, by cross processing analysis method, may be implemented to grow corrosion
The tracking and monitoring of information, real-time perception are conducive to carry out real-time long term monitoring to rail Life cycle.
On the one hand the present invention obtains the mutually indepedent sound-source signal of de-redundancy by iteration by algorithm optimization, realize drop
It makes an uproar, the signal-to-noise ratio of the signal greatly improved, to improve the precision of monitoring algorithm；On the other hand, opposite to pass monitoring for method, needle
Hurt is reconstructed, the present invention has done triple dimension-reduction treatment, greatly reduces computation complexity, reduces monitoring time, improves prison
The efficiency and reliability of method of determining and calculating, this finds that the requirement of real-time of hurt is of great significance to rail in military service.First is that by
The sound-source signal algorithm iteration that original guided wave signals carry out extracts, the heavy dimensionality reduction of the one of realization；Second is that believing obtained a large amount of sound sources
Number, further according to the relationship of similarity distance proposed by the present invention, double noise reduction has been carried out, has further obtained strong correlation hurt
Signal, reduce the redundancy of conventional monitoring methods；Third is that in hurt signal reconstruction step, according to two groups in hurt matrix
The relative size relationship of coefficient value, further lesser coefficient value is contributed in removal, and extraction obtains the construction coefficient matrix of more low-dimensional,
Realize triple noise reductions.In an embodiment of the present invention, quadruple dimensionality reduction it has been further noted that, i.e., to the construction coefficient finally obtained
Matrix carries out the extraction of a certain row vector, simplifies the operation that mean value goes interference.To reduce computation complexity, greatly improve
The efficiency and reliability of monitoring algorithm, this monitors gained big data analysis to rail guided wave, is of great significance.
The present invention has been considered and combined the Mining Problems of rail guided wave big data obtained by long term monitoring, passes through cross processing point
Analysis reduces hurt rate of false alarm, improves monitoring accuracy.Meanwhile algorithmically innovating, in the selection of two groups of monitoring data,
Hurt comparison can only be carried out with the guided wave database under pattern library, that is, rail complete health working condition compared to conventional method, this
The processing analysis of different monitoring moment guided wave data may be implemented in invention, and ingenious construction R-matrix is realized to sound obtained by iteration
The measurement and extraction of source signal handle removing by the averaging method of setting to the coefficient value of construction coefficient matrix
Interference, and the wrong report of the rail defects and failures as caused by accidentalia is reduced, improve precision.
The present invention does not need to establish that rail is on active service initial stage crosses redundant reference library simultaneously, with with monitoring, from working as monitoring
The preceding moment carries out various dimensions monitoring analysis, may be implemented to the assessment differentiation of multiple defects and damage reason location, using benchmark
When the mode of library, due to when in detection signal there are when multiple damages, being difficult to carry out reconstruction signal damage differentiation in practice, and it is real
In the monitoring of border, the latent lesion of rail is not only more, but also form is variant, and traditional benchmark library method will be difficult to realize.
It can be seen that the present invention can effectively realize the reliable monitoring of rail integrality, the cross-region of monitoring is substantially increased
Property and real-time, have important practical significance and engineering value.
Detailed description of the invention
Fig. 1 is the system block diagram of present system.
Fig. 2 is the flow chart of the method for the present invention.
Fig. 3 is the correlation waveform diagram of the embodiment of the present invention.
Fig. 4 is the damage signal waveform diagram of the embodiment of the present invention.
Fig. 5 is the damage signal and envelope diagram of the embodiment of the present invention.
Fig. 6 is the envelope diagram of the embodiment of the present invention.
Fig. 7 is the damage reason location figure of the embodiment of the present invention.
Fig. 8 is the damage signal of the embodiment of the present invention.
Fig. 9 is the damage reason location figure of the embodiment of the present invention.
Figure 10 is the damage signal figure of the embodiment of the present invention.
Figure 11 is the practical rail monitoring signals figure of the embodiment of the present invention.
Specific embodiment
Present invention will be further explained below with reference to the attached drawings and examples.
As shown in Figure 1, monitoring system includes front end monitoring modular, cloud monitoring server and browsing terminal, the front end
Monitoring modular includes wave guide energy converter, temperature sensor and control cabinet, there is solar panel module, charge and discharge control in control cabinet
Circuit, energy-storage module, transmission interface circuit module, slave computer control circuit module, communication interface circuit module, guided wave receive and dispatch mould
Block and signal conditioning module；Wherein, wave guide energy converter, temperature sensor are mounted on rail to be measured, slave computer control circuit
Module is electrically connected with temperature sensor, charge-discharge control circuit and guided wave transceiver module respectively by transmission interface circuit module
It connects, wave guide energy converter is electrically connected through signal conditioning module and guided wave transceiver module, solar panel module and charge and discharge control
Circuit electrical connection, charge-discharge control circuit and energy-storage module are electrically connected, and slave computer control circuit module passes through communication interface
Circuit module accesses cloud monitoring server.
Solar panel module acquisition solar energy is converted to electric energy, charges by charge-discharge control circuit to energy-storage module,
It is again that wave guide energy converter, temperature sensor and entire control cabinet are powered by energy-storage module.Wave guide energy converter sends ultrasound to rail
Guided wave, supersonic guide-wave encounter defect reflection after rail propagation and generate echo-signal, and echo-signal guided wave energy converter, which receives, to be made
For monitor rail guided wave signals, and successively through signal conditioning module conditioning, guided wave transceiver module receive after be sent to slave computer
Received data are transferred to cloud monitoring server and stored by control circuit module, slave computer control circuit module；Temperature
Sensor detects the real time temperature of the rail near wave guide energy converter, is sent to slave computer control circuit module.Browsing terminal packet
Computer/mobile phone terminal, computer/mobile phone terminal connected reference cloud monitoring server are included, and receives alarm and message information.
As shown in Fig. 2, as follows according to the embodiment that complete method of the present invention is implemented:
Embodiment 1:
Such as Fig. 3 to Fig. 7, cloud platform is monitored processing analysis according to monitoring front end monitoring gained guided wave data.
Here further to a large amount of guided wave monitoring data, i.e. guided wave big data is excavated, and is carried out by cross processing
Analysis, is divided into two groups of web of the rail monitoring data Y of E between taking hereiAnd YjIt is monitored analysis, wherein i=22, j=22 monitoring
Data are standardized by step S1.3, after step S1.4.1 to the step S1.4.3 processing in claim 5,
The dimension of obtained coefficient matrix W* is 44 × 23, to the guided wave signals of 44 dimensions, by step S1.4.4 in claim 5
After processing, 23 feature sound source informations are obtained, after can programming in MATLAB, to every group of coefficient number of this feature signal
According to being visualized, totally 23 groups.
θ in order to obtaini=| 1-p* i×k* i|, it needs exist for standardizing to data, to avoid due to data itself
The inaccuracy of result caused by size is lack of standardization, illustrates so that covariance calculates as an example here, enables Y* i=k × Yi,Y* j=k × Yj, show
So there is cov (Y* i, Y* j)=k2cov (Yi, Yj), it can be seen that data itself are not unified enough, and the data of same characteristic is caused to have
Different results.In order to avoid pi,kiInfluence of numerical values recited to similitude itself, needs exist for pi,kiAccording to step
Standardization formula in S1.3 is handled, the data p that obtains that treated* iAnd k* i.It here need to be according to step in claim 5
Standardization formula in S1.3 is handled, and new data p is obtained* iAnd k* i。
Further according to step S1.6, similarity distance set ξ={ θ is obtained1,θ2,...,θL, obtain walking for similitude
Power curve is shown, as shown in Figure 3 from low to high according to correlation distance.Black dotted lines are step S2.1 setting apart from threshold in Fig. 3
Parameter θ is depicted in value F, Fig. 3 simultaneouslyi, can intuitively find out and meet setting relationship θiThe part coefficient of < F, it is smaller to show phase
Answer coefficient and the sign damage of sound source subscale in weight matrix M better.The damage signal obtained then according to step S2.1 to 2.3
Ydefect, as shown in Figure 4.
Further step 2.4 obtains clearly defective locations waveform diagram, as shown in Figure 5 and Figure 6, according to the amplitude threshold of setting
Value Z determines that rail is had damage, and provides damage position, shown in Fig. 7, completes positioning assessment.
Further it is actually detected in, for step S2.3, in order to improve monitoring efficiency, it is also possible that implement, Ke Yiyou
Construct coefficient matrix Pw×HA certain row coefficient value carry out hurt signal construction, by taking the coefficient sets of w coefficient value correspondence as an example, table
Show as follows:
Here can simplify mean value goes interference operation to construct hurt by extracting any row vector, here to extract structure
Coefficient matrix the i-th behavior example is made, new one-dimensional row vector P is obtained1×H NEW=[αi1 αii … αiH], then according toIt can be in the hope of the supersonic guide-wave damage signal Y containing track damage characteristic informationdefect, further walk
Rapid S2.4 is again by YdefectProcessing obtains envelope information, obtains hurt location information by the way that threshold value is arranged, completes monitoring.It can drop
The computation complexity of low operation improves calculating speed.
Embodiment 2:
Figure 11 show actual steel rail web monitoring signals, and noise is relatively low, it is difficult to differentiate damage.
The present invention poly-injury situation can also be monitored assessment and positioning, illustrate here respectively to containing by 4 damage,
The rail actual conditions for being grown to 5 damages carry out the detection and analysis based on the present invention and traditional benchmark library method and compare, that is, damage
Hurt the process increased to be monitored by cross processing analysis.In actual monitoring, guided wave acquisition mode is set as rail one end
Transducer excitation transmitting, other end energy converter receive rail guided wave signals, obtain acquisition data, reach cloud by front end monitoring modular
Hold monitoring server.
The present invention specific implementation use cross processing method, can select four damage guided wave signals with five damage when
Guided wave signals carry out processing analysis, and actually such case can be equivalent to 0 damage and 1 degree of impairment, pass through this method
After processing, feature sound-source signal is completed according to step S1.3 to step S1.6 and coefficient extracts, step S2.1 to step S2.3
Fig. 8 reconstruction signal is obtained, step S2.4 obtains Fig. 9 signal, completes hurt judgement positioning, available i.e. the 5th newly increased
The signal of damage, therefore available preferable processing result.
Analogized with secondary, when there is more damages to generate, such as nine, ten etc., this method is analyzed by cross processing,
Different analysis data are selected, can preferably be differentiated.And when using traditional analysis method based on pattern library, due to benchmark
The guided wave signals in library are the guided wave monitoring signals of rail when being completely free of to have damage, therefore in analysis containing there are five damage
When situation, 0 damage and five degree of impairment are equivalent to, need to find out five kinds of damage informations simultaneously, it is contemplated that in actual propagation,
The multiple roundtrip of guided wave signals (different acoustic emission sources), superposition are offset and conversion, after leading to five feature reconstructions, Wu Facong
What is obtained obtains position of each damage, Y as shown in Figure 10 containing there are five in the guided wave signals of featuredefect, all signals are mixed
It stacks, mutually superimposed, counteracting, it is difficult to judge damage position.
And so on, traditional monitoring method based on pattern library or sample database can be completed when for less damage
The assessment of damage, and in practical application, with the continuation of monitoring, the increase of data volume, the growth of damage with increase, it will be difficult to it is complete
It is analyzed at the reliable monitoring of rail failure.
The step S1.5 to step S1.6 of the present invention simultaneously, definition pass through similarity distance θi=| 1- | p* i×k* i| | to spend
Source of students signal and rail defects and failures similarity based method are measured, is made by the mode of taking absolute value to two groups of guided wave numbers in step S1.3
According to the not requirement of sequence and signal source, conventional method requires this two groups of data to have sequence and source to require, because of tradition side
Method is all based on pattern library, and perhaps sample database i.e. first group data need to be the data in pattern library or sample database, and second group
The data of hurt monitoring are come from, i.e. first group of data of conventional method default are undamaged, and second group of data is only needs
The hurt signal of analysis, this is restricted the application range of conventional method, and the present invention is not necessarily to consider, the present invention is for two
The sequence of group data is without particular/special requirement.
It can be summarized as, monitoring data cross processing, according to presetting, do lateral comparison at given monitoring temperature D,
Processing analysis is carried out according to monitoring step to different monitoring period of time rail guided wave signals, obtains rail monitoring state, to realize not
In same default monitoring period of time, under the conditions of identical temperature, the detection and positioning of service state；It can also supervised according to presetting
It surveys under time interval E, two groups of monitoring data Y of E is divided between pairiAnd YjIt is monitored analysis, obtains rail analyzing service state knot
Fruit；Two groups of data Y in storing data can also be randomly selected according to presettingiAnd YjAccording to the processing method of the monitoring
Cross processing is carried out, obtains the service state of monitoring rail.
Above-mentioned specific embodiment is used to illustrate the present invention, rather than limits the invention, of the invention
In spirit and scope of protection of the claims, to any modifications and changes that the present invention makes, protection model of the invention is both fallen within
It encloses.
Claims (7)
1. a kind of cloud platform rail integrity monitoring systems based on guided wave technology, it is characterised in that: the monitoring system packet
Front end monitoring modular, cloud monitoring server and browsing terminal are included, the front end monitoring modular includes wave guide energy converter, temperature biography
Sensor and control cabinet have solar panel module, charge-discharge control circuit, energy-storage module, transmission interface circuit mould in control cabinet
Block, slave computer control circuit module, communication interface circuit module, guided wave transceiver module and signal conditioning module；Wherein, guided wave changes
Energy device, temperature sensor are mounted on rail to be measured, and slave computer control circuit module is distinguished by transmission interface circuit module
With temperature sensor, charge-discharge control circuit and guided wave transceiver module be electrically connected, wave guide energy converter through signal conditioning module with
The electrical connection of guided wave transceiver module, solar panel module and charge-discharge control circuit are electrically connected, charge-discharge control circuit with
Energy-storage module electrical connection, slave computer control circuit module access cloud monitoring server by communication interface circuit module.
2. a kind of cloud platform rail integrity monitoring systems based on guided wave technology according to claim 1, feature
Be: the wave guide energy converter sends supersonic guide-wave to rail, and supersonic guide-wave encounters defect reflection after rail propagation and generates
Echo-signal, echo-signal guided wave energy converter receive the guided wave signals as monitoring rail, and are sent to slave computer control electricity
Road module；The real time temperature of rail near the temperature sensor detection wave guide energy converter, is sent to slave computer control electricity
Road module.
3. a kind of cloud platform rail integrity monitoring systems based on guided wave technology according to claim 1, feature
Be: the solar panel module acquisition solar energy is converted to electric energy, fills by charge-discharge control circuit to energy-storage module
Electricity, then by energy-storage module be that wave guide energy converter, temperature sensor and entire control cabinet are powered.
4. a kind of cloud platform rail integrity monitoring systems based on guided wave technology according to claim 1, feature
Be: the browsing terminal includes computer/mobile phone terminal.
5. being applied to a kind of cloud platform rail Integrity Verification based on guided wave technology of any system of claim 1-4
Method, it is characterised in that:
S1, steps are as follows:
S1.1: according to rail geometry to be measured and physical attribute, preset the used supersonic guide-wave of wave guide energy converter mode and
Frequency presets the acquisition monitoring parameters of wave guide energy converter, passes through interruption continuous acquisition wave guide energy converter and temperature sensor
Obtain the guided wave signals and temperature signal of monitoring rail；
S1.2: number C, each collection period T are acquired altogether in acquisition time duration ASInterior guided wave signals X, guided wave signals
X and collected temperature signal of same time, are transmitted to cloud monitoring server, according to the data lattice of C signal data dimension
Formula Y=(X1, X2..., Xc)TIt carries out storage and forms guided wave monitoring data；
S1.3: for acquiring two groups of guided wave monitoring data before and after acquisition, two groups of guided waves monitorings at different acquisition time duration A
The signal data dimension of data is q and w, and two groups of guided wave monitoring data are collectively constituted new more higher-dimension according to line direction superposition
Then data Z ' to be analyzed carries out the standardization of feature scaling and sparse processing according to following formula to data Z ' to be analyzed
Obtain the multidimensional data Z to be analyzed for the dimension normalization that mean value is 0, variance is 1:
Wherein, E (z) and σ is respectively the mean value and standard deviation of data Z ' to be analyzed, ZiIndicate i-th in multidimensional data Z to be analyzed
Group signal, Zi' indicate i-th group of signal in data Z ' to be analyzed；
S1.4: carry out objective optimization: data Z to be analyzed to multidimensional carries out blind source separating, obtains the sound source letter of characterization rail failure
Number:
Weight coefficient of the sound-source signal in guided wave monitoring data is obtained using following formula iterative solution first:
S1.4.1: the weight coefficient W that one two norm of initialization are 10With iteration count n=1；
S1.4.2: solution is iterated according to following formula:
Wn=E { Z (WT n-1Z)3}-3Wn-1, n=n+1
Wherein, WnIndicate that the weight coefficient vector that nth iteration obtains, Z indicate multidimensional data to be analyzed, E { } indicates expectation letter
Number；δ1n~δLnRespectively indicate n-th of weight coefficient vector WnIn coefficient value, L indicate strong correlation weight coefficient sum；
S1.4.3: every time to weight coefficient vector W after iterative solutionnIt is normalized, then judges:
If | WT nWn-1| 1 is not converged on, then re-execute the steps S1.4.2；
If | WT nWn-1| it converges on 1 and under the conditions of meeting the number of iterations ordinal number n less than iteration total degree N, then exports current iteration
Weight coefficient vector W under numbernAs strong correlation weight coefficient, and it is added in coefficient matrix W* and continues step
Q+w group strong correlation weight coefficient is obtained until the number of iterations ordinal number n is equal to iteration total degree N then iteration ends in S1.4.2；
S1.4.4: final to obtain coefficient matrix W*, W*=(W1, W2..., Wq+w)T, obtain characterizing according to formula R=W* × Z to be measured
Sound-source signal data R, the R=(r of different sound sources in rail1,r2,...,rL)T, r1,r2,...,rLIndicate sound-source signal data R
The sound source subsignal of the middle corresponding each guided wave signals of guided wave monitoring data；
S1.5: the generalized inverse matrix for solving coefficient matrix W* obtains weight matrix M, indicates are as follows:
In formula, β11~βqLThe q row coefficient value that the guided wave monitoring data that dimension is q correspond to sound-source signal is respectively indicated, as
One group of coefficient sets；α11~αwLIndicate that dimension corresponds to the w row coefficient value of sound-source signal for the guided wave monitoring data of w, as another
One group of coefficient sets；Q row coefficient value and w row coefficient value correspond respectively to the signal data dimension of two groups of front and back guided wave monitoring data
Spend q and w；
Construction is below with reference to matrix K:
Wherein, the number of element -1 is q in each column and 1 number of element is w；
S1.6: carrying out similarity measurement, calculates separately a column p of weight matrix MiColumn k corresponding with R-matrix KiBetween
Similitude:
For each column vector p of weight matrix MiWith each column vector k of R-matrix KiUse be standardized
Respective ranks vector p after to standardization* iAnd k* i, then similarity distance is obtained according to following formula:
θi=| 1-p* i×k* i|
In formula, θiA column vector p after indicating the standardization of weight matrix M* iStandardization corresponding with R-matrix K
A column vector k afterwards* iBetween similarity distance；
Then by all similarity distance result compositional similarity distance set ξ={ θ1,θ2,...,θL}；
S2, the guided wave signals for carrying out dimensionality reduction and noise reduction reconstruct acquisition characterization rail failure, obtain damage position:
S2.1: feature extraction dimension-reduction treatment is carried out to weight matrix M and obtains strong correlation sound-source signal:
The satisfaction relationships θ such as not is extracted from similarity distance set ξ according to distance threshold FiThe similarity distance θ of < FiCorresponding power
Column vector in weight matrix M and the sound source subsignal in sound-source signal data R, by the Column vector groups that extract at following wound
Damage matrix P(q+w)×H, following strong correlation sound source matrix is formed by the sound source subsignal extracted
Wherein, H indicates the quantity of strong correlation sound-source signal, as strong correlation sound source component；
S2.2: hurt matrix P(q+w)×HTwo groups of coefficient sets in the relatively small coefficient sets of matrix system numerical value are given up, retain square
The battle array relatively large coefficient sets of coefficient, thus by hurt matrix P(q+w)×HIt extracts and obtains construction coefficient matrix Pw×H；
Below by taking the coefficient sets of w coefficient value are corresponding as an example, i.e., its coefficient value is relatively large:
S2.3: it reconfigures the ultrasound containing track damage characteristic information according to construction coefficient matrix and strong correlation sound source matrix and leads
Wave damage signal Ydefect；
To construction coefficient matrix Pw×HEach column coefficient value is carried out according to the following equation to take mean value that interference is gone to handle, obtains new one
Tie up row vector P1×H NEW:
The supersonic guide-wave damage signal Y containing track damage characteristic information is acquired further according to following formuladefect:
S2.4: according to damage signal YdefectProcessing obtains envelope information,
If envelope range value thinks that rail to be measured has damage greater than the part of amplitude thresholds Z in envelope information；
If part of the envelope range value no more than amplitude thresholds Z thinks that damage is not present in rail to be measured in envelope information；
In rail to be measured there are under degree of impairment, supersonic guide-wave damage signal YdefectThere is the corresponding place of damage in rail to be measured
Acquisition moment t using following formula processing obtain rail failure positioning:
S=t × V/2
T=1/Fs × i, i=0,1 ..., num, num=Ts × Fs
Wherein, Ts indicates that the collection period of wave guide energy converter, Fs indicate the sample frequency of wave guide energy converter, and V indicates guided wave energy exchange
The supersonic guide-wave speed that device issues, i indicate that the ordinal number of i-th of acquisition, num indicate that the sum all acquired, s indicate that rail exists
Damage in position to wave guide energy converter distance.
6. a kind of cloud platform rail Integrity Verification method and system based on guided wave technology according to claim 5,
It is characterized by: the interruption continuous acquisition be continuously carried out at each acquisition interval duration A according to monitoring requirements number C,
Data length is the guided wave signals of Ts × Fs and the acquisition of temperature.
7. a kind of cloud platform rail Integrity Verification method and system based on guided wave technology according to claim 5,
It is characterized by: when the guided wave signals of slave computer control circuit module storage and the data of temperature signal reach data and deposit
When reserves B, rate of failing to report is reduced using following setting cross processing analysis method and finds out damage:
A: the guided wave signals at identical monitoring temperature D but different time carry out processing analysis according to step, obtain rail failure
Positioning；
B: at fixed monitoring time interval E, the time difference is that two groups of guided wave monitoring data of monitoring time interval E are pressed between
Processing analysis is carried out according to step, obtains the positioning of rail failure；
C: two groups of guided wave monitoring data are randomly selected in the data of storage according to step and carry out processing analysis, obtain rail failure
Positioning.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910063557.2A CN109649432B (en) | 2019-01-23 | 2019-01-23 | System and method for monitoring integrity of steel rail of cloud platform based on guided wave technology |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910063557.2A CN109649432B (en) | 2019-01-23 | 2019-01-23 | System and method for monitoring integrity of steel rail of cloud platform based on guided wave technology |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109649432A true CN109649432A (en) | 2019-04-19 |
| CN109649432B CN109649432B (en) | 2020-06-23 |
Family
ID=66120787
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910063557.2A Active CN109649432B (en) | 2019-01-23 | 2019-01-23 | System and method for monitoring integrity of steel rail of cloud platform based on guided wave technology |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109649432B (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110428072A (en) * | 2019-08-16 | 2019-11-08 | 苏州富欣智能交通控制有限公司 | A kind of streetcar track health monitoring systems |
| CN110568084A (en) * | 2019-09-19 | 2019-12-13 | 哈尔滨工业大学 | Method for extracting low signal-to-noise ratio guided wave signal reaching time suitable for guided wave transducer array |
| CN110940732A (en) * | 2019-12-30 | 2020-03-31 | 华东交通大学 | Steel rail structure corrosion online monitoring and evaluation method |
| CN111077220A (en) * | 2019-12-28 | 2020-04-28 | 陕西泰诺特检测技术有限公司 | Low-frequency ultrasonic guided wave detection device and method |
| CN111896625A (en) * | 2020-08-17 | 2020-11-06 | 中南大学 | Real-time monitoring method and monitoring system for rail damage |
| CN113715868A (en) * | 2021-06-17 | 2021-11-30 | 上海应用技术大学 | Remote track detection system based on time-space coupling |
| CN114330455A (en) * | 2022-01-05 | 2022-04-12 | 哈尔滨工业大学 | Steel rail acoustic emission signal rapid high-precision reconstruction method based on compressed sensing |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101398410A (en) * | 2008-11-07 | 2009-04-01 | 哈尔滨工业大学 | Steel rail defect detection method by electromagnetical ultrasonic technology and device thereof |
| CN102879472A (en) * | 2012-09-22 | 2013-01-16 | 华南理工大学 | Adaptive steel rail ultrasonic flaw detection method and device based on frequency spectrum recognition |
| CN104020221A (en) * | 2014-05-30 | 2014-09-03 | 杨媛 | Real-time broken-rail detecting and positioning system based on ultrasonic guided waves |
| CN104614105A (en) * | 2013-12-05 | 2015-05-13 | 北京交通大学 | Device and method for monitoring rail stress based on ultrasonic guided-wave |
| CN104634872A (en) * | 2015-01-10 | 2015-05-20 | 哈尔滨工业大学（威海） | Online high-speed railway steel rail damage monitoring method |
| CN105292176A (en) * | 2015-11-25 | 2016-02-03 | 株洲时代电子技术有限公司 | Testing task electrical system of rail-defect detector car |
| CN105564459A (en) * | 2015-12-14 | 2016-05-11 | 中国铁路总公司 | Seamless steel rail stress detection device and method based on ultrasonic guided waves and strain gauge |
| CN105651859A (en) * | 2016-01-12 | 2016-06-08 | 杭州浙达精益机电技术股份有限公司 | Ultrasonic guided wave device and method for monitoring corrosion of pipeline |
| CN106228107A (en) * | 2016-06-30 | 2016-12-14 | 杭州浙达精益机电技术股份有限公司 | A kind of supersonic guide-wave rail break monitoring algorithm based on independent component analysis |
| CN106248801A (en) * | 2016-09-06 | 2016-12-21 | 哈尔滨工业大学 | A kind of Rail crack detection method based on many acoustie emission events probability |
| CN106828535A (en) * | 2016-12-14 | 2017-06-13 | 中国铁道科学研究院电子计算技术研究所 | A kind of track rail break monitoring system and rail break method for detecting position |
| CN207579878U (en) * | 2017-11-06 | 2018-07-06 | 成都治通弘瑞信息科技有限公司 | A kind of information system of intellectual monitoring rail damage |
-
2019
- 2019-01-23 CN CN201910063557.2A patent/CN109649432B/en active Active
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101398410A (en) * | 2008-11-07 | 2009-04-01 | 哈尔滨工业大学 | Steel rail defect detection method by electromagnetical ultrasonic technology and device thereof |
| CN102879472A (en) * | 2012-09-22 | 2013-01-16 | 华南理工大学 | Adaptive steel rail ultrasonic flaw detection method and device based on frequency spectrum recognition |
| CN104614105A (en) * | 2013-12-05 | 2015-05-13 | 北京交通大学 | Device and method for monitoring rail stress based on ultrasonic guided-wave |
| CN104020221A (en) * | 2014-05-30 | 2014-09-03 | 杨媛 | Real-time broken-rail detecting and positioning system based on ultrasonic guided waves |
| CN104634872A (en) * | 2015-01-10 | 2015-05-20 | 哈尔滨工业大学（威海） | Online high-speed railway steel rail damage monitoring method |
| CN105292176A (en) * | 2015-11-25 | 2016-02-03 | 株洲时代电子技术有限公司 | Testing task electrical system of rail-defect detector car |
| CN105564459A (en) * | 2015-12-14 | 2016-05-11 | 中国铁路总公司 | Seamless steel rail stress detection device and method based on ultrasonic guided waves and strain gauge |
| CN105651859A (en) * | 2016-01-12 | 2016-06-08 | 杭州浙达精益机电技术股份有限公司 | Ultrasonic guided wave device and method for monitoring corrosion of pipeline |
| CN106228107A (en) * | 2016-06-30 | 2016-12-14 | 杭州浙达精益机电技术股份有限公司 | A kind of supersonic guide-wave rail break monitoring algorithm based on independent component analysis |
| CN106248801A (en) * | 2016-09-06 | 2016-12-21 | 哈尔滨工业大学 | A kind of Rail crack detection method based on many acoustie emission events probability |
| CN106828535A (en) * | 2016-12-14 | 2017-06-13 | 中国铁道科学研究院电子计算技术研究所 | A kind of track rail break monitoring system and rail break method for detecting position |
| CN207579878U (en) * | 2017-11-06 | 2018-07-06 | 成都治通弘瑞信息科技有限公司 | A kind of information system of intellectual monitoring rail damage |
Non-Patent Citations (5)
| Title |
|---|
| 冯志红: "大型锻件超声检测方法及信号处理算法研究", 《中国博士学位论文全文数据库（工程科技二辑）》 * |
| 刘国华: "声发射信号处理关键技术研究", 《中国博士学位论文全文数据库（工程科技二辑）》 * |
| 吴建春: "《光伏发电系统建设实用技术》", 30 April 2015 * |
| 杨臻垚: "基于超声导波的钢轨轨底结构健康监测技术研究", 《中国优秀硕士学位论文全文数据库（工程科技二辑）》 * |
| 胡剑虹: "基于超声导波的钢轨轨底无损检测关键技术研究", 《中国优秀硕士学位论文全文数据库（工程科技二辑）》 * |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110428072A (en) * | 2019-08-16 | 2019-11-08 | 苏州富欣智能交通控制有限公司 | A kind of streetcar track health monitoring systems |
| CN110568084A (en) * | 2019-09-19 | 2019-12-13 | 哈尔滨工业大学 | Method for extracting low signal-to-noise ratio guided wave signal reaching time suitable for guided wave transducer array |
| CN110568084B (en) * | 2019-09-19 | 2020-07-24 | 哈尔滨工业大学 | Method for extracting low signal-to-noise ratio guided wave signal reaching time suitable for guided wave transducer array |
| CN111077220A (en) * | 2019-12-28 | 2020-04-28 | 陕西泰诺特检测技术有限公司 | Low-frequency ultrasonic guided wave detection device and method |
| CN110940732A (en) * | 2019-12-30 | 2020-03-31 | 华东交通大学 | Steel rail structure corrosion online monitoring and evaluation method |
| CN110940732B (en) * | 2019-12-30 | 2022-03-01 | 华东交通大学 | Steel rail structure corrosion online monitoring and evaluation method |
| CN111896625A (en) * | 2020-08-17 | 2020-11-06 | 中南大学 | Real-time monitoring method and monitoring system for rail damage |
| CN111896625B (en) * | 2020-08-17 | 2023-07-14 | 中南大学 | Rail damage real-time monitoring method and monitoring system thereof |
| CN113715868A (en) * | 2021-06-17 | 2021-11-30 | 上海应用技术大学 | Remote track detection system based on time-space coupling |
| CN114330455A (en) * | 2022-01-05 | 2022-04-12 | 哈尔滨工业大学 | Steel rail acoustic emission signal rapid high-precision reconstruction method based on compressed sensing |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109649432B (en) | 2020-06-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109649432A (en) | Cloud platform rail integrity monitoring systems and method based on guided wave technology | |
| CN102288883B (en) | Oscillation wave partial discharge identifying and positioning method for asynchronous double-end power cable | |
| CN100427961C (en) | Virtual instrument technique based gas insulation combined electric appliances online detecting method | |
| CN206114822U (en) | Many information detection means of power transformer winding deformation state | |
| CN109324241B (en) | Corrosion diagnosis early warning method and system for transformer substation grounding grid | |
| CN111679166A (en) | Switch cabinet partial discharge fault multi-source information fusion detection early warning system and method based on wireless transmission technology | |
| CN108008287A (en) | Track circuit failure monitoring platform and its monitoring method based on frequency-shift signaling | |
| CN103487749A (en) | On-line monitoring and diagnosing system and method for mechanical state of high-voltage circuit breaker | |
| CN105997072A (en) | Electrode contact state detecting method for electrical impedance detection based on measurement precision | |
| CN109615273B (en) | Method and system for evaluating state of electric vehicle charging facility | |
| CN110261746B (en) | Cable defect detection method based on periodic attenuation characteristics of oscillating wave voltage | |
| CN109212378A (en) | Middle and low voltage network Fault Locating Method | |
| CN103439631A (en) | Method and system for detecting corrosion state of grounding grid | |
| WO2023284127A1 (en) | Vibration-signal-based online monitoring system for gil defects, and method | |
| CN116610998A (en) | Switch cabinet fault diagnosis method and system based on multi-mode data fusion | |
| CN104655967B (en) | Distribution transformer basket vibration signal characteristic quantity extracting method | |
| CN117289067A (en) | Transformer running state on-line monitoring system | |
| CN115935264A (en) | Power grid transient voltage stability assessment method based on time-space information synchronous learning | |
| CN111999614A (en) | Local discharge multi-source fusion detection early warning method for high-voltage switch cabinet | |
| CN109443190A (en) | Transient traveling wave-based transformer winding deformation online monitoring method and device | |
| CN105203937A (en) | Internal discharge mode recognition method and fault diagnosis system for transformer | |
| CN103267652B (en) | Intelligent online diagnosis method for early failures of equipment | |
| CN216848010U (en) | Cable partial discharge online monitoring device for edge calculation | |
| CN201628754U (en) | Wideband partial discharge detection device for power equipment | |
| CN105891685A (en) | All-weather ultrasonic state monitoring device for substation equipment |
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 |