CN110441629A - Method for diagnosing faults and device based on point machine action current curve - Google Patents
Method for diagnosing faults and device based on point machine action current curve Download PDFInfo
- Publication number
- CN110441629A CN110441629A CN201910689553.5A CN201910689553A CN110441629A CN 110441629 A CN110441629 A CN 110441629A CN 201910689553 A CN201910689553 A CN 201910689553A CN 110441629 A CN110441629 A CN 110441629A
- Authority
- CN
- China
- Prior art keywords
- current curve
- current
- sample
- phase
- curve
- 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
- 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
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R19/00—Arrangements for measuring currents or voltages or for indicating presence or sign thereof
- G01R19/165—Indicating that current or voltage is either above or below a predetermined value or within or outside a predetermined range of values
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
Abstract
The embodiment of the invention provides a kind of method for diagnosing faults and device based on point machine action current curve, it is previously stored with the corresponding characteristic of current curve sample that fault type is marked, to the point machine of fault diagnosis to be carried out, obtain the target signature data of its action current curve, the corresponding characteristic of current curve sample that fault type according to target signature data and is respectively marked filters out target current curve sample similar with current curve from current curve sample, target faults type is determined according to the corresponding fault type of each target current curve sample, the target faults type is that there are the fault types of failure for point machine.The automatic identification to goat failure is realized by the comparison of the current curve sample with labeled fault type, the accuracy rate of fault identification is improved, reduces a possibility that failing to judge and judging by accident, improve fault identification efficiency.
Description
Technical field
The present invention relates to goat maintenance technology fields, are based on point machine action current curve more particularly, to one kind
Method for diagnosing faults and device.
Background technique
City rail signalling arrangement failure is to interfere one of the principal element of railway normal transport order, track signal equipment
If it is the influence that can greatly reduce to subway transport passenger's order that failure, which can be handled quickly and be restored,.Point machine
(for example, ZDJ9 type point machine) be track telecommunication and signaling branch safeguarded outdoor three-major-items (outdoor three-major-items include track switch turn
One of rut machine equipment, track circuit and semaphore), it is the infrastructure device for realizing interlocking of signals relationship.According to statistics, track switch conversion is set
Standby has been the highest equipment of telecommunication and signaling branch failure rate, and the situation that faulty rate rises, and switching device failure is
The undesirable influence such as track train operational efficiency is influenced, and has caused passenger train late.Goat is if there is exception, very
The major accidents such as it may cause off-line, derail, knock into the back, jeopardizing traffic safety.
Current track signal system maintenance mode uses always failure to repair and combines with periodic maintenance.Failure, which is repaired, refers to equipment
Troubleshooting maintenance is carried out after catastrophic failure.Periodic maintenance is the fault diagnosis maintenance side based on material lifetime analysis and evaluation
Formula, periodic maintenance refers to establish maintenance of equipment operation system based on the service life of the wear law of part or part, ties up
The planned maintenance activity that operation system is the interior progress of skylight point time range at the scene is repaired, again according to field investigation periodic maintenance
It is divided into signalling arrangement maintenance and maintenance.Maintenance is that equipment is tested and repaired, it covers the full content of maintenance, purpose
It is the degree to make equipment reach the technical parameter of service technique prescribed by standard, technical conditions tolerance.Maintenance is maintenance
With repairing, equipment is made to maintain good state.
The existing fault identification to goat and positioning mainly by comparing area under the curve ratio, average current value ratio and
The theoretical knowledge and field experience of maintenance personnel, the fault identification time is longer, low efficiency, and judge by accident, phenomenon of failing to judge occurs now and then.
And a large amount of manpower is also required to the maintenance mode of point machine and time cost has the possibility failed to judge or judged by accident,
In actual application, inventor has found that the existing identification to point machine failure relies primarily on artificial ginseng
With, a possibility that failing to judge and judging by accident big, the low efficiency of fault identification.
Summary of the invention
The embodiment of the present invention provides a kind of method for diagnosing faults and device based on point machine action current curve, uses
To solve a possibility that identification in the prior art to point machine failure relies primarily on artificial participation, fails to judge and judge by accident
Greatly, the problem of the low efficiency of fault identification.
Against the above technical problems, in a first aspect, the embodiment provides one kind based on point machine movement
The method for diagnosing faults of current curve, comprising:
The point machine action current curve for obtaining fault diagnosis to be carried out, extracts the characteristic of the current curve
According to as target signature data；
The corresponding characteristic of current curve sample of fault type is marked according to the target signature data and in advance
The current curve sample for being greater than default similarity with the similarity of the current curve is filtered out, as target current curve sample
This；
Target faults corresponding with the current curve are determined according to the corresponding fault type of each target current curve sample
Type, the point machine are the target faults type there are the fault type of failure.
Second aspect, the embodiment provides a kind of fault diagnosises based on point machine action current curve
Device, comprising:
Extraction module extracts the electric current for obtaining the point machine action current curve of fault diagnosis to be carried out
The characteristic of curve, as target signature data；
Screening module, for the current curve sample pair of fault type to be marked according to the target signature data and in advance
The characteristic answered filters out the current curve sample for being greater than default similarity with the similarity of the current curve, as target
Current curve sample；
Diagnostic module, for according to the corresponding fault type determination of each target current curve sample and the current curve pair
The target faults type answered, the point machine are the target faults type there are the fault type of failure.
The third aspect the embodiment provides a kind of electronic equipment, including memory, processor and is stored in
On reservoir and the computer program that can run on a processor, the processor realize any of the above item institute when executing described program
The step of method for diagnosing faults based on point machine action current curve stated.
Fourth aspect, the embodiment provides a kind of non-transient computer readable storage mediums, are stored thereon with
Computer program is realized when the computer program is executed by processor and acts electricity based on point machine described in any of the above item
The step of method for diagnosing faults of flow curve.
The embodiment provides a kind of method for diagnosing faults and dress based on point machine action current curve
It sets, the corresponding characteristic of current curve sample that fault type is marked is previously stored with, to the road of fault diagnosis to be carried out
Branch off goat, obtains the target signature data of its action current curve, fault type according to target signature data and is respectively marked
The corresponding characteristic of current curve sample that similar with current curve target current is filtered out from current curve sample is bent
Line sample determines target faults type according to the corresponding fault type of each target current curve sample, which is
For point machine, there are the fault types of failure.It is realized by the comparison of the current curve sample with labeled fault type
To the automatic identification of goat failure, the accuracy rate of fault identification is improved, a possibility that failing to judge and judging by accident is reduced, improved
Fault identification efficiency.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to do one simply to introduce, it should be apparent that, the accompanying drawings in the following description is this hair
Bright some embodiments for those of ordinary skill in the art without creative efforts, can be with root
Other attached drawings are obtained according to these attached drawings.
Fig. 1 is a kind of fault diagnosis side based on point machine action current curve provided by one embodiment of the present invention
The flow diagram of method；
Fig. 2 is that the electric current of a certain classification for the without failure point machine that another embodiment of the present invention provides is bent
Line schematic diagram；
Fig. 3 is that the electric current of the another category for the without failure point machine that another embodiment of the present invention provides is bent
Line schematic diagram；
Fig. 4 be another embodiment of the present invention provide there are the current curves of a certain classification of the point machine of failure
Schematic diagram；
Fig. 5 be another embodiment of the present invention provide there are the current curves of the another category of the point machine of failure
Schematic diagram；
Fig. 6 is another embodiment of the present invention offer to there are the electric current of a certain classification of the point machine of failure songs
Line carries out three-phase and restores schematic diagram；
Fig. 7 be another embodiment of the present invention provide there are the current curves of the another category of the point machine of failure
It carries out three-phase and restores schematic diagram；
Fig. 8 is the point machine action current fault diagnosis flow scheme schematic diagram that another embodiment of the present invention provides；
Fig. 9 is the method for diagnosing faults based on point machine action current curve that another embodiment of the present invention provides
The structural block diagram of device；
Figure 10 is the structural block diagram for the electronic equipment that another embodiment of the present invention provides.
Specific embodiment
In order to make the object, technical scheme and advantages of the embodiment of the invention clearer, below in conjunction with the embodiment of the present invention
In attached drawing, technical scheme in the embodiment of the invention is clearly and completely described, it is clear that described embodiment is
A part of the embodiment of the present invention, instead of all the embodiments.Based on the embodiments of the present invention, those of ordinary skill in the art
Every other embodiment obtained without making creative work, shall fall within the protection scope of the present invention.
Fig. 1 is a kind of process of method for diagnosing faults based on point machine action current curve provided in this embodiment
Schematic diagram, referring to Fig. 1, this method comprises:
101: obtaining the point machine action current curve of fault diagnosis to be carried out, extract the feature of the current curve
Data, as target signature data；
102: the corresponding feature of current curve sample of fault type being marked according to the target signature data and in advance
Data screening goes out to be greater than with the similarity of the current curve current curve sample of default similarity, as target current curve
Sample；
103: target corresponding with the current curve is determined according to the corresponding fault type of each target current curve sample
Fault type, the point machine are the target faults type there are the fault type of failure.
Method provided in this embodiment is executed by being equipped with the equipment for executing above-mentioned steps 101-103, which can be
Terminal or server, the present embodiment are not particularly limited this.The current curve of point machine refers to that goat carries out
By the movement of normotopia to antiposition, or the curve that the electric current that generate when the movement by antiposition to normotopia changes over time.By
It is the reaction whether goat is in normal operating conditions in the current curve of goat, therefore the present embodiment will turn according to track switch
Rut it is motor-driven as when the current curve that generates to point machine, there are the fault types of failure to identify.
Current curve is intended to carry out the point machine of fault diagnosis by normotopia to antiposition or by the mistake of antiposition to normotopia
The servo-actuated curve for making to realize variation of the electric current generated in journey.Target signature data be extracted from current curve can react electricity
The data acquisition system of flow curve feature.The current curve sample that fault type is marked by target signature data and in advance is corresponding
Characteristic can calculate the similarity between current curve and each current curve sample, obtain the biggish several electricity of similarity
Flow curve sample determines that track switch turns as target current curve sample, according to the corresponding fault type of each target current curve sample
The corresponding target faults type of rut machine.
Due to goat take be three-phase electricity power supply, since the current curve shape of three-phase is essentially identical, this reality
The method for applying example offer can carry out fault type recognition to goat by a certain phase current curve, can also be according to three-phase
Current curve carries out fault type recognition to goat.For example, target is special when current curve is only the current curve of a certain phase
Levying data is the feature only extracted from the phase current curve, when calculating and the similarity of current curve sample, the electricity of extraction
The characteristic of flow curve sample is also only to extract characteristic from a certain phase current curve of current curve sample.Work as electric current
When curve is only the current curve of three-phase, target signature data are the feature extracted from three current curve, in calculating and electricity
When the similarity of flow curve sample, the characteristic of the current curve sample of extraction is also the three-phase current from current curve sample
Characteristic is extracted in curve.
A kind of method for diagnosing faults based on point machine action current curve is present embodiments provided, is previously stored with
The corresponding characteristic of current curve sample of fault type is marked, to the point machine of fault diagnosis to be carried out, obtains
The current curve sample of fault type according to target signature data and is respectively marked in the target signature data of its action current curve
Corresponding characteristic filters out target current curve sample similar with current curve from current curve sample, according to each mesh
The corresponding fault type of mark current curve sample determines target faults type, which is that point machine exists
The fault type of failure.It is realized by the comparison of the current curve sample with labeled fault type to goat failure
Automatic identification improves the accuracy rate of fault identification, reduces a possibility that failing to judge and judging by accident, improves fault identification efficiency.
A kind of side specifically carrying out fault type recognition to goat (for example, ZDJ9 type point machine) provided herein
Method, this method are directed to the characteristics of original operation/maintenance data is without type label in advance and first carry out unsupervised learning analysis, are calculated using cluster
Method extracts the current curve of performance point machine inferior health and failure as training sample set.When carrying out failure mode analysis,
Feature extraction is carried out to the point machine action current data that subway line O&M generates, KNN is utilized to fault sample data set
Classifier carries out Analysis on Fault Diagnosis to the current data that O&M generates, and returns to fault type.Implementation method may include with
Under several steps:
The extraction of step (1) goat action current curvilinear characteristic vector；
Step (2), which is obtained, carries out the inferior health and fault type that clustering extracts to operation/maintenance data using clustering algorithm
Data sample, increase fault data sample set quantity；
Step (3) carries out Analysis on Fault Diagnosis to new data using KNN classifier, and returns to fault type.
Further, on the basis of the above embodiments, the point machine movement for obtaining fault diagnosis to be carried out
Current curve extracts the characteristic of the current curve, as target signature data, comprising:
The point machine action current curve for obtaining fault diagnosis to be carried out, the current curve of each phase is divided into
The current curve and fourth order of current curve, phase III generation that the current curve of first stage generation, second stage generate
The current curve of Duan Shengcheng；
To the current curve of each phase, first stage characteristic value, In are extracted in the current curve that generates in the first stage
Second stage characteristic value is extracted in the current curve that second stage generates, and extracts third in the current curve that the phase III generates
Phase characteristic value extracts fourth stage characteristic value in the current curve that fourth stage generates, and according to the electric current of each phase song
Line contains the three-phase characteristic value of relationship between three-phase current curve in the current curve extraction that second stage generates；
By the first stage characteristic value, the second stage characteristic value, the phase III characteristic value, the fourth order
The actuation time of section characteristic value, the three-phase characteristic value and point machine is as the target signature data；
It wherein, will be before first time point according to the time sequencing for generating current curve to the current curve of each phase
What the current curve and second stage that current maxima position was generated as the first stage in the current curve of generation generated
Current value in the current curve generated before the second time point is become the position of decline by the separation of current curve from rising
The separation for the current curve that the current curve generated as second stage and phase III generate, will be before third time point
The current curve and fourth stage that the position of current standard deviation from large to small was generated as the phase III in the current curve of generation
The separation of the current curve of generation；To the current curve of each phase, start the start time point for generating current curve and described
The ratio that duration between first time point accounts for current curve total duration is the first ratio, the start time point and described second
The ratio that duration between time point accounts for the total duration is the second ratio, the start time point and the third time point it
Between duration account for the total duration ratio be third ratio.
Further, first ratio is 10%, and second ratio is 80%, and the third ratio is 82%.
Method provided in this embodiment by three-phase current curve that point machine acts come to goat failure therefore
Barrier type is identified.The current curve of each phase is divided into 4 stages, extracts characteristic respectively for each stage
According to, then for the characteristic of relationship between the second stage extraction expression three-phase current of each phase current curve, by this of extraction
A little data and goat actuation time are as target signature data.Target signature data have reacted goat current curve feature,
The current curve and each current curve can be accurately calculated by target signature data and the characteristic of each current curve sample
The similarity of sample is laid a good foundation for the identification of fault type.
A kind of method for diagnosing faults based on point machine action current curve is present embodiments provided, according to electric current song
Current curve is divided into 4 stages by the feature of line, convenient for extracting characteristic quickly through the current curve in this 4 stages.
Further, on the basis of the various embodiments described above, the current curve to each phase is given birth in the first stage
At current curve in extract first stage characteristic value, second stage generate current curve in extract second stage feature
Value extracts phase III characteristic value in the current curve that the phase III generates, and mentions in the current curve that fourth stage generates
Fourth stage characteristic value is taken, and three-phase is contained in the current curve extraction that second stage generates according to the current curve of each phase
The three-phase characteristic value of relationship between current curve, comprising:
To the current curve of any first phase, obtained from the current curve that the first stage generates the first current maxima,
First current average and the first current standard deviation, as the first stage characteristic value；
To the current curve of the first phase, the second current maxima, first are obtained from the current curve that second stage generates
Current minimum, the second current average and the second current standard deviation, as the second stage characteristic value；
To the current curve of the first phase, third current maxima, second are obtained from the current curve that the phase III generates
Current minimum, third current average and third current standard deviation, as the phase III characteristic value；
To the current curve of the first phase, the 4th current maxima, third are obtained from the current curve that fourth stage generates
Current minimum, the 4th current average and the 4th current standard deviation, as the fourth stage characteristic value；
To the current curve of the second phase, the 5th current average is obtained from the current curve that second stage generates, and right
The current curve of third phase obtains the 6th current average from the current curve that second stage generates, and calculates second electricity
First difference of levelling mean value and the 5th current average, second current average and the 6th current average
The second difference, the third difference of the 5th current average and the 6th current average, by first difference, institute
The second difference and the third difference are stated as the three-phase characteristic value.
The acquisition methods for present embodiments providing each characteristic value in target signature data exist to the current curve of each phase
First stage of division extracts the first current maxima, the first current average and the first current standard deviation this 3 characteristic values,
The second current maxima, the first current minimum, the second current average and the second electric current mark are extracted in the second stage of division
Quasi- this 4 characteristic values of difference, extract third current maxima, the second current minimum, third electric current in the phase III of division
Average value and third current standard deviation this 4 characteristic values extract the 4th current maxima, third electricity in the fourth stage of division
Flow minimum value, the 4th current average and the 4th current standard deviation this 4 characteristic values.Along with 3 three-phase characteristic values and turn-out track
Machine actuation time, total 3* (3+4+4+4)+3+1=49 characteristic values.This 49 eigenvalue clusters are at matrix, and from each current curve
Sample is calculated using 49 characteristic values that same procedure is extracted, and obtains the phase of the current curve and each current curve sample
Like degree.Wherein, the current standard deviation in each stage is equal to the electric current of each sampled point and current average in the stage in the stage
Deviation square arithmetic average square root.
A kind of method for diagnosing faults based on point machine action current curve is present embodiments provided, individually according to each
Stage current curve realizes the extraction to characteristic, is extracted reaction three-phase current curve by relatively simple calculating
The data of feature.
To above-mentioned steps (1), by taking ZDJ9 type point machine as an example, when extracting characteristic, when each with current curve
Between status information extracts feature vector when track switch conversion in the stage.Wherein, current curve can be divided into according to conversion time sequence
Unlock-conversion-locking-is slow to put four time phases.Fig. 2 is a certain of without failure goat provided in this embodiment
The current curve schematic diagram of classification, Fig. 3 are the current curve of the another category of without failure goat provided in this embodiment
Schematic diagram.As shown in Figures 2 and 3, track switch unlocking phases when electric motor starting, have very big starting current, generate a peak value
Afterwards, track switch enters unlocked state quickly, and with the operation of equipment, after the completion of track switch unlock, resistance becomes smaller rapidly, current curve
It falls after rise rapidly, track switch enters conversion process.
The form of ZDJ9 goat action current curve can accurately show goat action process, therefore extract electric current
Need to be divided into before curvilinear characteristic 4 stages correspond to goat movement time series.The switching point in each stage is corresponding
The transfer point of the time phase of goat 4 movements.After action current curve is divided into 4 time sections according to the time, In
Characteristic value is extracted in each time section and sets up feature vector, thus can accurately capture each period relative to whole
The fainter fault eigenvalue of difference for a current data sample.By the consulting of site technology expert and from operation number
According to sampling analysis discovery: the sampling period for the electric current that is operating normally be 40ms and form it is roughly the same, act total time in 5.5s-
6.5s。
To above-mentioned steps (1), determine that the process in 4 stages is as follows:
A, preceding 10% (current curve generated before first time point) of obtaining current data, by comparing its determining electricity
Flow the division points that maximum value position is first stage and second stage.Preceding the 80% of obtaining current data is (i.e. in the second time
The current curve generated before point), it is second stage and phase III by comparing the position that determining current value declines suddenly
Transfer point.Preceding 82% (current curve generated before third time point) of obtaining current data, analyzes second inversion point
Data later, by comparing standard deviation, from large to small after position be determined as the transfer point of phase III and fourth stage.
B, to each phase current curve, the maximum value of current data, average and standard deviation conduct in the first stage are extracted
The characteristic value of first stage extracts in second stage the maximum value of current data, minimum value, average and standard deviation as second
The characteristic value in stage, the maximum value of current data, minimum value, average and standard deviation are as the phase III in the extraction phase III
Characteristic value, extract the spy as fourth stage of maximum value, minimum value, average and standard deviation of current data in fourth stage
Value indicative extracts the value for making difference two-by-two of second stage current average in three-phase current curve as characteristic value, and extracts movement
The duration of current data is as characteristic value；
C, by the characteristic value of the current curve four-stage of the three-phase of each action current, according to the current curve of three-phase
Two-stage calculate current average difference and actuation time totally 49 eigenvalue clusters at goat action current curve spy
It levies data (i.e. target signature data).
Further, described to be marked according to the target signature data and in advance on the basis of the various embodiments described above
The corresponding characteristic of current curve sample of fault type filters out similar greater than presetting to the similarity of the current curve
The current curve sample of degree, as target current curve sample, comprising:
The current curve for obtaining any phase of point machine movement in advance, as current curve sample, if to acquisition
Dry current curve sample carries out clustering, obtains the cluster being made of the current curve sample that quantity is less than preset quantity, makees
For fault sample cluster, the corresponding fault type of each current curve sample in the fault sample cluster is marked；
It is bent by the corresponding three-phase current of the current curve sample to each current curve sample that fault type is marked
Line obtains the corresponding characteristic of the current curve sample, corresponding according to the target signature data and each current curve sample
The characteristic Euclidean distance that calculates the current curve and be respectively marked between the current curve sample of fault type, screening
Euclidean distance is less than the current curve sample of default Euclidean distance out, as target current curve sample.
Further, on the basis of the various embodiments described above, the preparatory any phase for obtaining point machine movement
Current curve carries out clustering to several current curve samples of acquisition, obtains being less than by quantity as current curve sample
The cluster of the current curve sample composition of preset quantity, clusters as fault sample, comprising:
In advance obtain point machine movement any phase current curve, as current curve sample, from each electricity
Characteristic is extracted in flow curve sample, it is bent by several electric currents of the K-means algorithm to acquisition according to the characteristic of extraction
Line sample carries out clustering, the cluster being made of the current curve sample that quantity is less than preset quantity is obtained, as failure sample
This cluster.
Further, on the basis of the various embodiments described above, each electricity marked in the fault sample cluster
The corresponding fault type of flow curve sample, comprising:
It is poly- that current curve when having sent out the fault type of failure according to goat and having sent out failure analyzes the fault sample
The corresponding fault type of each current curve sample in class marks each current curve sample in the fault sample cluster
This corresponding fault type.
It should be noted that when carrying out clustering by K-means algorithm, it can be to from each current curve sample
The characteristic extracted in monophase current curve carries out clustering, can also be from the three-phase current song of each current curve sample
The characteristic extracted in line carries out clustering, and the present embodiment is not particularly limited this.
It includes: that the monophase current curve is divided into aforementioned four stage generation that characteristic is extracted from monophase current curve
Current curve, it is average that current maxima, current minimum, electric current are then extracted from the current curve that each stage generates respectively
Value and current standard deviation, the actuation time with point machine is together as the characteristic for carrying out clustering.
It is upper that extraction characteristic, which includes: by each phase current curve-equipartition of the current curve, from three-phase current curve
The current curve for stating four-stage generation extracts current maxima, current average from the first stage to each phase current curve
And current standard deviation, current maxima, current minimum, current average and current standard are extracted respectively from other three phases
It is average to calculate any two-phase second stage electric current then further according to the current average of the second stage of each phase current curve for difference
The difference of value turns the current maxima of extraction, current minimum, current average, current standard deviation, three differences and track switch
The actuation time of rut machine is together as the characteristic for carrying out clustering.
Further, current curve when having sent out the fault type of failure according to point machine and having sent out failure analyzes institute
State the corresponding fault type of each current curve sample in fault sample cluster, comprising: to any current curve sample, if through
The feature of the analysis expert current curve sample and a certain current curve when having sent out failure have same fault type, then basis
The fault type marks the current curve sample.That is, to current curve sample in fault sample cluster in the present embodiment
Label can be labelled with combination technology analysis expert fault type, increase fault type data by repeating this work
Amount.
There are three states for point machine, are normal condition, Subhealthy Status and fault condition respectively.The present embodiment pair
How obtaining abnormal condition, (i.e. both of these case is referred to as goat herein and there is event by Subhealthy Status and fault condition
The case where barrier) current curve be introduced, since the form of the three-phase action current curve of goat is almost the same, thus
When extracting the current curve of abnormal condition, only analyzed with the current curve of a certain phase.
When carrying out clustering and obtaining fault sample cluster, by the characteristic of each current curve sample of extraction into
Row cluster operation, the method for the characteristic of each current curve sample of extraction include:
By the electric current song that each current curve sample is divided into the current curve of first stage generation, second stage generates
The current curve that the current curve and fourth stage that line, phase III generate generate, from the current curve that the first stage generates
Current maxima, current average and current standard deviation are obtained, it is maximum that electric current is obtained from the current curve that second stage generates
It is maximum to obtain electric current from the current curve that the phase III generates for value, current minimum, current average and current standard deviation
It is maximum to obtain electric current from the current curve that fourth stage generates for value, current minimum, current average and current standard deviation
Value, current minimum, current average and current standard deviation, then goat actuation time is obtained, the spy that will be obtained from each stage
Value indicative and goat actuation time are as the characteristic extracted from current curve sample.
Wherein, the method to current curve sample progress divided stages and the above-mentioned side that divided stages are carried out to current curve
Method is identical, and details are not described herein.
Each current curve sample standard deviation to corresponding one group of characteristic, to the corresponding characteristic of each current curve sample into
Row cluster calculation (for example, the Euclidean distance between each group characteristic is calculated, the closer current curve sample aggregation of Euclidean distance
It is clustered for one).Since in most cases goat is in the case where operating normally, by a fairly large number of current curve
The cluster that sample is formed regard as be the generation of trouble-free goat current curve, by the current curve sample shape of negligible amounts
At cluster regard as there are the goat of failure generate current curve.Fig. 4 turns to be provided in this embodiment there are failure
The current curve schematic diagram of a certain classification of rut machine, Fig. 5 are that provided in this embodiment there are the another categories of the goat of failure
Current curve schematic diagram.As shown in Figure 4 and Figure 5, the current curve quantity that trouble-free goat is formed is more, and failure turns
The current curve that rut machine generates is less, clusters by the cluster of a small number of current curve sample aggregations as fault sample.To event
Hinder the mark that each current curve sample in sample clustering carries out fault type, to be used for subsequent analysis.
The present embodiment carries out clustering, reason using K-means algorithm are as follows: for common three kinds of clustering algorithms (spectrum
Cluster, DBSCAN and K-means) it is compared using silhouette coefficient as measurement standard, find the profile system of three kinds of clustering algorithms
Numerical value is substantially at same standard.But spectral clustering time complexity is higher, the algorithm training process time is longer, and
DBSCAN needs to select there are two parameter, and adjustment parameter process is complex, therefore the present embodiment determines and passes through K-means algorithm
It is clustered and the current curve of the goat of failure is screened.
Specifically, for above-mentioned steps (2), with wherein point machine 2 months 120,000 operation/maintenance datas into
For row clustering, by K-means algorithm obtain fault sample cluster after, by fault sample cluster in current curve pair
The fault type answered is divided into 37 classifications.
A certain current curve and each current curve Sample Similarity are calculated by then passing through the current curve of three-phase, into
Before row similarity calculation, three-phase reduction need to be carried out to the current curve sample that fault type is marked.Fig. 6 mentions for the present embodiment
Supply to there are the current curve of a certain classification of the goat of failure carry out three-phase restore schematic diagram, wherein draw above
A kind of other current curve carries out the three-phase current curve after three-phase reduction, and figure below is that the current curve of a certain classification carries out
The current curve of a certain phase before three-phase restores.Fig. 7 is that provided in this embodiment there are the electricity of the another category of the goat of failure
Flow curve carries out three-phase and restores schematic diagram, wherein draw above is three carried out after three-phase reduction to the current curve of the category
Phase current curve, figure below are the current curve that a certain phase before three-phase restores is carried out to the current curve of the category.Such as Fig. 6 and
Shown in Fig. 7, goat actuation time power at 2.7 seconds, rotation is shown as disconnected display without obviously increasing in monitoring system.Fig. 7
In, goat AC phase duration is 0.8 second, and B phase broken phase current is 0A, and power is without obviously increasing when rotation, in monitoring system
It is shown as disconnected display.
A kind of method for diagnosing faults based on point machine action current curve is present embodiments provided, it is true by clustering
The fixed current curve sample for being used to carry out fault diagnosis, by between the characteristic and target signature data of current curve sample
The calculating of Euclidean distance, it is determining with the more similar a plurality of current curve sample of target signature data, as current curve sample,
Analysis for consequent malfunction type is laid a good foundation.
Further, on the basis of the various embodiments described above,
It is described that target corresponding with the current curve is determined according to the corresponding fault type of each target current curve sample
Fault type, the point machine are the target faults type there are the fault type of failure, comprising:
Fault type corresponding to each target current curve sample counts the corresponding target current curve of each fault type
The quantity of sample accounts for the 4th ratio of target current curve total sample number amount, and the corresponding fault type of maximum 4th ratio is made
For the target faults type, the point machine is the target faults type there are the fault type of failure.
The step can be realized by k nearest neighbor algorithm KNN.Wherein, KNN is that a simple and classical machine learning is calculated
Method classifies to sample by measurement " data to be sorted " and " classification known sample data " distance.Algorithm flow are as follows: calculate
Distance of the sample point of unknown classification to known class point；According to distance-taxis calculated；Count each classification in K point
Number, and by the highest classification for making unknown sample data of the frequency of occurrences in K sample point.Most important super ginseng in KNN algorithm
Number is exactly K value.The invention patent K value is determined by common trellis search method.Because parameter only one therefore adjust joined
Journey is simple and quick.Goat action current data, which can have both been inputted, after true defining K value carries out failure modes diagnosis
Specifically, to above-mentioned steps (3), for example, fault type is in the corresponding fault type of each current curve sample
A's accounts for 50%, and fault type accounts for 20% for B's, and fault type accounts for 10% for C's, and fault type accounts for 20% for D's, then target
Fault type is A.
Fig. 8 is goat action current fault diagnosis flow scheme schematic diagram provided in this embodiment, referring to Fig. 8, first passes through and turns
Rut machine operation/maintenance data and artificial fault data obtain the current curve sample for carrying out fault diagnosis.When input new data (electricity
Flow curve) when, by KNN classifier, fault type belonging to new data can be analyzed.
Method for diagnosing faults provided in this embodiment based on point machine action current curve determines effectively and efficient
Clustering algorithm (K-means) to initial data carry out clustering, by clustering extract fault type data, repeat this
Fault type data sample is collected in work, and the k nearest neighbor algorithm being introduced into machine learning acts electricity to goat from the angle of mathematics
Flow curve is classified, its normal or fault category belonged to is divided, and realizes the fault diagnosis of goat action current curve
Analysis.That it changes the maintenance modes of " failure is repaired " of failure outside current goat room, reduce the human cost of maintenance failure.
Live operation maintenance personnel is simplified to the accident analysis process of goat, failure point can be positioned accurately, track switch is promoted and turns
The O&M efficiency of rut machine, reduces the wasting of resources of breakdown maintenance, increases economic efficiency.
Fig. 9 is the structural frames of the trouble-shooter provided in this embodiment based on point machine action current curve
Figure, referring to Fig. 9, which includes extraction module 901, screening module 902 and diagnostic module 903, wherein
Extraction module 901 extracts the electricity for obtaining the point machine action current curve of fault diagnosis to be carried out
The characteristic of flow curve, as target signature data；
Screening module 902, for the current curve sample of fault type to be marked according to the target signature data and in advance
This corresponding characteristic filters out the current curve sample for being greater than default similarity with the similarity of the current curve, as
Target current curve sample；
Diagnostic module 903, for determining bent with the electric current according to the corresponding fault type of each target current curve sample
The corresponding target faults type of line, the point machine are the target faults type there are the fault type of failure.
Trouble-shooter provided in this embodiment based on point machine action current curve is suitable for above-mentioned implementation
The method for diagnosing faults based on point machine action current curve in example, details are not described herein.
A kind of trouble-shooter based on point machine action current curve is present embodiments provided, is previously stored with
The corresponding characteristic of current curve sample of fault type is marked, to the point machine of fault diagnosis to be carried out, obtains
The current curve sample of fault type according to target signature data and is respectively marked in the target signature data of its action current curve
Corresponding characteristic filters out target current curve sample similar with current curve from current curve sample, according to each mesh
The corresponding fault type of mark current curve sample determines target faults type, which is that point machine exists
The fault type of failure.It is realized by the comparison of the current curve sample with labeled fault type to goat failure
Automatic identification improves the accuracy rate of fault identification, reduces a possibility that failing to judge and judging by accident, improves fault identification efficiency.
Figure 10 is the structural block diagram of electronic equipment provided in this embodiment.
Referring to Fig.1 0, the electronic equipment includes: processor (processor) 1010, communication interface
(Communications Interface) 1020, memory (memory) 1030 and communication bus 1040, wherein processor
1010, communication interface 1020, memory 1030 completes mutual communication by communication bus 1040.Processor 1010 can be adjusted
With the logical order in memory 1030, to execute following method: the point machine movement for obtaining fault diagnosis to be carried out is electric
Flow curve extracts the characteristic of the current curve, as target signature data；According to the target signature data and in advance
The corresponding characteristic of current curve sample that fault type is marked is filtered out to be greater than in advance with the similarity of the current curve
If the current curve sample of similarity, as target current curve sample；According to the corresponding failure of each target current curve sample
Type determines target faults type corresponding with the current curve, and the point machine is institute there are the fault type of failure
State target faults type.
In addition, the logical order in above-mentioned memory 1030 can be realized by way of SFU software functional unit and conduct
Independent product when selling or using, can store in a computer readable storage medium.Based on this understanding, originally
Substantially the part of the part that contributes to existing technology or the technical solution can be in other words for the technical solution of invention
The form of software product embodies, which is stored in a storage medium, including some instructions to
So that a computer equipment (can be personal computer, server or the network equipment etc.) executes each implementation of the present invention
The all or part of the steps of example the method.And storage medium above-mentioned include: USB flash disk, mobile hard disk, read-only memory (ROM,
Read-Only Memory), random access memory (RAM, Random Access Memory), magnetic or disk etc. it is various
It can store the medium of program code.
The present embodiment provides a kind of non-transient computer readable storage mediums, are stored thereon with computer program, the calculating
Machine program is executed by processor following method: obtaining the point machine action current curve of fault diagnosis to be carried out, extracts institute
The characteristic for stating current curve, as target signature data；Failure classes are marked with preparatory according to the target signature data
The corresponding characteristic of current curve sample of type filters out the electricity for being greater than default similarity with the similarity of the current curve
Flow curve sample, as target current curve sample；According to the corresponding fault type determination of each target current curve sample and institute
The corresponding target faults type of current curve is stated, the point machine is the target faults class there are the fault type of failure
Type.
The present embodiment discloses a kind of computer program product, and the computer program product includes being stored in non-transient calculating
Computer program on machine readable storage medium storing program for executing, the computer program include program instruction, when described program instruction is calculated
When machine executes, computer is able to carry out method provided by above-mentioned each method embodiment, it may for example comprise: obtain failure to be carried out
The point machine action current curve of diagnosis, extracts the characteristic of the current curve, as target signature data；According to
The target signature data and the corresponding characteristic of current curve sample that fault type is marked in advance filter out with it is described
The similarity of current curve is greater than the current curve sample of default similarity, as target current curve sample；According to each target
The corresponding fault type of current curve sample determines target faults type corresponding with the current curve, the point machine
It is the target faults type there are the fault type of failure.
The embodiments such as electronic equipment described above are only schematical, wherein it is described as illustrated by the separation member
Unit may or may not be physically separated, and component shown as a unit may or may not be object
Manage unit, it can it is in one place, or may be distributed over multiple network units.It can select according to the actual needs
Some or all of the modules therein is selected to achieve the purpose of the solution of this embodiment.Those of ordinary skill in the art are not paying wound
In the case where the labour for the property made, it can understand and implement.
Through the above description of the embodiments, those skilled in the art can be understood that each embodiment can
It realizes by means of software and necessary general hardware platform, naturally it is also possible to pass through hardware.Based on this understanding, on
Stating technical solution, substantially the part that contributes to existing technology can be embodied in the form of software products in other words, should
Computer software product may be stored in a computer readable storage medium, such as ROM/RAM, magnetic disk, CD, including several fingers
It enables and using so that a computer equipment (can be personal computer, server or the network equipment etc.) executes each implementation
Method described in certain parts of example or embodiment.
Finally, it should be noted that the above various embodiments is only to illustrate the technical solution of the embodiment of the present invention, rather than it is right
It is limited；Although the embodiment of the present invention is described in detail referring to foregoing embodiments, the ordinary skill of this field
Personnel are it is understood that it is still possible to modify the technical solutions described in the foregoing embodiments, or to part
Or all technical features are equivalently replaced；And these are modified or replaceed, it does not separate the essence of the corresponding technical solution
The range of each embodiment technical solution of the embodiment of the present invention.
Claims (10)
1. a kind of method for diagnosing faults based on point machine action current curve characterized by comprising
The point machine action current curve for obtaining fault diagnosis to be carried out, extracts the characteristic of the current curve, makees
For target signature data；
The corresponding characteristic screening of current curve sample of fault type is marked according to the target signature data and in advance
It is greater than the current curve sample of default similarity with the similarity of the current curve out, as target current curve sample；
Target faults type corresponding with the current curve is determined according to the corresponding fault type of each target current curve sample,
The point machine is the target faults type there are the fault type of failure.
2. the method for diagnosing faults according to claim 1 based on point machine action current curve, which is characterized in that
The point machine action current curve for obtaining fault diagnosis to be carried out, extracts the characteristic of the current curve, makees
For target signature data, comprising:
The point machine action current curve for obtaining fault diagnosis to be carried out, is divided into first for the current curve of each phase
Current curve, the current curve that second stage generates, the current curve of phase III generation and the fourth stage that stage generates are raw
At current curve；
To the current curve of each phase, first stage characteristic value is extracted in the current curve that generates in the first stage, second
Second stage characteristic value is extracted in the current curve that stage generates, and extracts the phase III in the current curve that the phase III generates
Characteristic value extracts fourth stage characteristic value in the current curve that fourth stage generates, and is existed according to the current curve of each phase
The current curve extraction that second stage generates contains the three-phase characteristic value of relationship between three-phase current curve；
The first stage characteristic value, the second stage characteristic value, the phase III characteristic value, the fourth stage is special
The actuation time of value indicative, the three-phase characteristic value and point machine is as the target signature data；
Wherein, the current curve of each phase will be generated according to the time sequencing for generating current curve before first time point
Current curve in the electric current that generates of the current curve that generates as the first stage of current maxima position and second stage
The separation of curve, using current value in the current curve generated before the second time point from rise become decline position as
The separation for the current curve that the current curve and phase III that second stage generates generate, will generate before third time point
Current curve in the current curve that generates as the phase III of the position of current standard deviation from large to small and fourth stage generate
Current curve separation；To the current curve of each phase, start the start time point and described first for generating current curve
The ratio that duration between time point accounts for current curve total duration is the first ratio, the start time point and second time
The ratio that duration between point accounts for the total duration is the second ratio, between the start time point and the third time point
The ratio that duration accounts for the total duration is third ratio.
3. the method for diagnosing faults according to claim 2 based on point machine action current curve, which is characterized in that
The current curve to each phase extracts first stage characteristic value in the current curve generated in the first stage, second
Second stage characteristic value is extracted in the current curve that stage generates, and extracts the phase III in the current curve that the phase III generates
Characteristic value extracts fourth stage characteristic value in the current curve that fourth stage generates, and is existed according to the current curve of each phase
The current curve extraction that second stage generates contains the three-phase characteristic value of relationship between three-phase current curve, comprising:
To the current curve of any first phase, the first current maxima, first are obtained from the current curve that the first stage generates
Current average and the first current standard deviation, as the first stage characteristic value；
To the current curve of the first phase, the second current maxima, the first electric current are obtained from the current curve that second stage generates
Minimum value, the second current average and the second current standard deviation, as the second stage characteristic value；
To the current curve of the first phase, third current maxima, the second electric current are obtained from the current curve that the phase III generates
Minimum value, third current average and third current standard deviation, as the phase III characteristic value；
To the current curve of the first phase, the 4th current maxima, third electric current are obtained from the current curve that fourth stage generates
Minimum value, the 4th current average and the 4th current standard deviation, as the fourth stage characteristic value；
To the current curve of the second phase, the 5th current average is obtained from the current curve that second stage generates, and to third
The current curve of phase obtains the 6th current average from the current curve that second stage generates, and it is flat to calculate second electric current
First difference of mean value and the 5th current average, the of second current average and the 6th current average
Two differences, the third difference of the 5th current average and the 6th current average, by first difference, described
Two differences and the third difference are as the three-phase characteristic value.
4. the method for diagnosing faults according to claim 1 based on point machine action current curve, which is characterized in that
The corresponding characteristic screening of current curve sample that fault type is marked according to the target signature data and in advance
It is greater than the current curve sample of default similarity with the similarity of the current curve out, as target current curve sample, packet
It includes:
The current curve for obtaining any phase of point machine movement in advance, as current curve sample, to several electricity of acquisition
Flow curve sample carries out clustering, obtains the cluster being made of the current curve sample that quantity is less than preset quantity, as event
Hinder sample clustering, marks the corresponding fault type of each current curve sample in the fault sample cluster；
To each current curve sample that fault type is marked, obtained by the corresponding three-phase current curve of the current curve sample
To the corresponding characteristic of the current curve sample, according to the target signature data and the corresponding spy of each current curve sample
The Euclidean distance that sign data calculate the current curve and are respectively marked between the current curve sample of fault type, filters out Europe
Formula distance is less than the current curve sample of default Euclidean distance, as target current curve sample.
5. the method for diagnosing faults according to claim 1 based on point machine action current curve, which is characterized in that
It is described that target faults type corresponding with the current curve is determined according to the corresponding fault type of each target current curve sample,
The point machine is the target faults type there are the fault type of failure, comprising:
Fault type corresponding to each target current curve sample counts the corresponding target current curve sample of each fault type
Quantity account for the 4th ratio of target current curve total sample number amount, using the corresponding fault type of maximum 4th ratio as institute
Target faults type is stated, the point machine is the target faults type there are the fault type of failure.
6. the method for diagnosing faults according to claim 4 based on point machine action current curve, which is characterized in that
The current curve of the preparatory any phase for obtaining point machine movement, as current curve sample, to several electricity of acquisition
Flow curve sample carries out clustering, obtains the cluster being made of the current curve sample that quantity is less than preset quantity, as event
Hinder sample clustering, comprising:
The current curve for obtaining any phase of point machine movement in advance, as current curve sample, it is bent from each electric current
Characteristic is extracted in line sample, according to the characteristic of extraction by K-means algorithm to several current curve samples of acquisition
This progress clustering obtains the cluster being made of the current curve sample that quantity is less than preset quantity, poly- as fault sample
Class.
7. the method for diagnosing faults according to claim 4 based on point machine action current curve, which is characterized in that
The corresponding fault type of each current curve sample marked in the fault sample cluster, comprising:
It is poly- that current curve when having sent out the fault type of failure according to point machine and having sent out failure analyzes the fault sample
The corresponding fault type of each current curve sample in class marks each current curve sample in the fault sample cluster
This corresponding fault type.
8. a kind of trouble-shooter based on point machine action current curve characterized by comprising
Extraction module extracts the current curve for obtaining the point machine action current curve of fault diagnosis to be carried out
Characteristic, as target signature data；
Screening module, the current curve sample for fault type to be marked according to the target signature data and in advance are corresponding
Characteristic filters out the current curve sample for being greater than default similarity with the similarity of the current curve, as target current
Curve sample；
Diagnostic module, for corresponding with the current curve according to the corresponding fault type determination of each target current curve sample
Target faults type, the point machine are the target faults type there are the fault type of failure.
9. a kind of electronic equipment including memory, processor and stores the calculating that can be run on a memory and on a processor
Machine program, which is characterized in that the processor realizes as described in any one of claim 1 to 7 be based on when executing described program
The step of method for diagnosing faults of point machine action current curve.
10. a kind of non-transient computer readable storage medium, is stored thereon with computer program, which is characterized in that the computer
It is realized when program is executed by processor as described in any one of claim 1 to 7 based on point machine action current curve
The step of method for diagnosing faults.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910689553.5A CN110441629B (en) | 2019-07-29 | 2019-07-29 | Fault diagnosis method and device based on turnout switch machine operating current curve |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910689553.5A CN110441629B (en) | 2019-07-29 | 2019-07-29 | Fault diagnosis method and device based on turnout switch machine operating current curve |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110441629A true CN110441629A (en) | 2019-11-12 |
| CN110441629B CN110441629B (en) | 2022-02-15 |
Family
ID=68431955
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910689553.5A Active CN110441629B (en) | 2019-07-29 | 2019-07-29 | Fault diagnosis method and device based on turnout switch machine operating current curve |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110441629B (en) |
Cited By (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110927497A (en) * | 2019-12-09 | 2020-03-27 | 交控科技股份有限公司 | Point switch fault detection method and device |
| CN110988557A (en) * | 2019-12-20 | 2020-04-10 | 中国人民解放军陆军军医大学第一附属医院 | Equipment fault detection device based on real-time current detection |
| CN110988560A (en) * | 2019-12-20 | 2020-04-10 | 中国人民解放军陆军军医大学第一附属医院 | Medical equipment fault detection system and method based on real-time current |
| CN111881950A (en) * | 2020-07-10 | 2020-11-03 | 交控科技股份有限公司 | Method and device for representing characteristics of current time sequence of turnout switch machine |
| CN111907562A (en) * | 2020-06-24 | 2020-11-10 | 中铁第一勘察设计院集团有限公司 | Real-time monitoring and control system for railway communication state |
| CN112036505A (en) * | 2020-09-17 | 2020-12-04 | 广西交控智维科技发展有限公司 | Method and device for determining equipment state of turnout switch machine and electronic equipment |
| CN112214634A (en) * | 2020-09-24 | 2021-01-12 | 交控科技股份有限公司 | Processing method and system for switch conversion sound |
| CN112319554A (en) * | 2020-10-30 | 2021-02-05 | 交控科技股份有限公司 | Multipoint traction turnout synchronous fault monitoring method and device and electronic equipment |
| CN113534776A (en) * | 2021-07-16 | 2021-10-22 | 珠海丽珠试剂股份有限公司 | Data processing method and device and pipeline system |
| CN113581253A (en) * | 2021-07-26 | 2021-11-02 | 中国铁路兰州局集团有限公司 | Method and device for determining state of electric empty switch machine |
| CN113859304A (en) * | 2021-09-23 | 2021-12-31 | 通号城市轨道交通技术有限公司 | Turnout monitoring and controlling method and device |
| CN114325188A (en) * | 2021-12-28 | 2022-04-12 | 国网河北省电力有限公司经济技术研究院 | Fault detection method and device and server |
| CN114655285A (en) * | 2022-02-25 | 2022-06-24 | 北京全路通信信号研究设计院集团有限公司 | Point switch health degree evaluation method and device, terminal equipment and storage medium |
| CN114692969A (en) * | 2022-03-29 | 2022-07-01 | 西门子交通技术(北京)有限公司 | Point switch failure prediction method, device, electronic equipment and storage medium |
| CN116482457A (en) * | 2023-03-24 | 2023-07-25 | 四川众合智控科技有限公司 | Intelligent turnout diagnosis method based on relay interlocking |
| CN116628446A (en) * | 2023-05-24 | 2023-08-22 | 成都工业职业技术学院 | Intelligent analysis method and system for turnout health standard value based on clustering algorithm |
| CN112214634B (en) * | 2020-09-24 | 2024-04-23 | 交控科技股份有限公司 | Method and system for processing switch conversion sound |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH09145772A (en) * | 1995-11-20 | 1997-06-06 | Furukawa Electric Co Ltd:The | Method and device for locating transmission/distribution line electric wire failure section |
| CN104280668A (en) * | 2014-11-05 | 2015-01-14 | 广东电网有限责任公司佛山供电局 | Failure type identifying method and system of power distribution network |
| CN104360262A (en) * | 2014-10-29 | 2015-02-18 | 国家电网公司 | Method for opening-closing coil current comparison of circuit breaker operating mechanisms on basis of feature points |
| CN105184084A (en) * | 2015-09-14 | 2015-12-23 | 深圳供电局有限公司 | Fault type predicting method and system for automatic electric power measurement terminals |
| CN106896323A (en) * | 2017-04-17 | 2017-06-27 | 天津商业大学 | Switched reluctance machines asymmetrical half-bridge type power inverter main switch fault detection method |
| CN107203746A (en) * | 2017-05-12 | 2017-09-26 | 同济大学 | A kind of switch breakdown recognition methods |
| CN107590506A (en) * | 2017-08-17 | 2018-01-16 | 北京航空航天大学 | A kind of complex device method for diagnosing faults of feature based processing |
| CN108416362A (en) * | 2018-01-29 | 2018-08-17 | 同济大学 | A kind of track switch abnormity early warning and method for diagnosing faults |
| CN108761226A (en) * | 2018-04-02 | 2018-11-06 | 浙江众合科技股份有限公司 | A kind of circuit and method for realizing three-phase alternating current five-wire system goat analog simulation |
-
2019
- 2019-07-29 CN CN201910689553.5A patent/CN110441629B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH09145772A (en) * | 1995-11-20 | 1997-06-06 | Furukawa Electric Co Ltd:The | Method and device for locating transmission/distribution line electric wire failure section |
| CN104360262A (en) * | 2014-10-29 | 2015-02-18 | 国家电网公司 | Method for opening-closing coil current comparison of circuit breaker operating mechanisms on basis of feature points |
| CN104280668A (en) * | 2014-11-05 | 2015-01-14 | 广东电网有限责任公司佛山供电局 | Failure type identifying method and system of power distribution network |
| CN105184084A (en) * | 2015-09-14 | 2015-12-23 | 深圳供电局有限公司 | Fault type predicting method and system for automatic electric power measurement terminals |
| CN106896323A (en) * | 2017-04-17 | 2017-06-27 | 天津商业大学 | Switched reluctance machines asymmetrical half-bridge type power inverter main switch fault detection method |
| CN107203746A (en) * | 2017-05-12 | 2017-09-26 | 同济大学 | A kind of switch breakdown recognition methods |
| CN107590506A (en) * | 2017-08-17 | 2018-01-16 | 北京航空航天大学 | A kind of complex device method for diagnosing faults of feature based processing |
| CN108416362A (en) * | 2018-01-29 | 2018-08-17 | 同济大学 | A kind of track switch abnormity early warning and method for diagnosing faults |
| CN108761226A (en) * | 2018-04-02 | 2018-11-06 | 浙江众合科技股份有限公司 | A kind of circuit and method for realizing three-phase alternating current five-wire system goat analog simulation |
Non-Patent Citations (8)
| Title |
|---|
| OU DX ETC.: "HYBRID FAULT DIAGNOSIS OF RAILWAY SWITCHES BASED ON THE SEGMENTATION OF MONITORING CURVES", 《EKSPLOATACJA I NIEZAWODNOSC-MAINTENANCE AND RELIABILITY》 * |
| 周旋: "电动道岔动作电流曲线特性及分析", 《鄂钢科技》 * |
| 杨云国等: "基于动作电流波形树表达的道岔转辙机状态检测方法", 《上海铁道大学学报》 * |
| 许庆阳: "道岔故障诊断及健康状态预测", 《中国优秀博硕士学位论文全文数据库(硕士)工程科技Ⅱ辑》 * |
| 赵振翔: "模块化的多机牵引提速道岔转辙机智能测试仪的研究与设计", 《中国优秀博硕士学位论文全文数据库(硕士) 工程科技Ⅱ辑》 * |
| 钟志旺等: "基于SVDD的道岔故障检测和健康评估方法", 《西南交通大学学报》 * |
| 陈亭: "基于道岔动作电流的故障诊断方法研究", 《中国优秀博硕士学位论文全文数据库(硕士) 工程科技Ⅱ辑》 * |
| 黄世泽等: "基于弗雷歇距离的道岔故障诊断方法", 《同济大学学报（自然科学版）》 * |
Cited By (21)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110927497A (en) * | 2019-12-09 | 2020-03-27 | 交控科技股份有限公司 | Point switch fault detection method and device |
| CN110988557A (en) * | 2019-12-20 | 2020-04-10 | 中国人民解放军陆军军医大学第一附属医院 | Equipment fault detection device based on real-time current detection |
| CN110988560A (en) * | 2019-12-20 | 2020-04-10 | 中国人民解放军陆军军医大学第一附属医院 | Medical equipment fault detection system and method based on real-time current |
| CN110988557B (en) * | 2019-12-20 | 2022-06-10 | 中国人民解放军陆军军医大学第一附属医院 | Equipment fault detection device based on real-time current detection |
| CN111907562A (en) * | 2020-06-24 | 2020-11-10 | 中铁第一勘察设计院集团有限公司 | Real-time monitoring and control system for railway communication state |
| CN111881950A (en) * | 2020-07-10 | 2020-11-03 | 交控科技股份有限公司 | Method and device for representing characteristics of current time sequence of turnout switch machine |
| CN112036505A (en) * | 2020-09-17 | 2020-12-04 | 广西交控智维科技发展有限公司 | Method and device for determining equipment state of turnout switch machine and electronic equipment |
| CN112214634A (en) * | 2020-09-24 | 2021-01-12 | 交控科技股份有限公司 | Processing method and system for switch conversion sound |
| CN112214634B (en) * | 2020-09-24 | 2024-04-23 | 交控科技股份有限公司 | Method and system for processing switch conversion sound |
| CN112319554A (en) * | 2020-10-30 | 2021-02-05 | 交控科技股份有限公司 | Multipoint traction turnout synchronous fault monitoring method and device and electronic equipment |
| CN113534776A (en) * | 2021-07-16 | 2021-10-22 | 珠海丽珠试剂股份有限公司 | Data processing method and device and pipeline system |
| CN113581253A (en) * | 2021-07-26 | 2021-11-02 | 中国铁路兰州局集团有限公司 | Method and device for determining state of electric empty switch machine |
| CN113581253B (en) * | 2021-07-26 | 2023-10-03 | 中国铁路兰州局集团有限公司 | State determination method and device for electric air switch machine |
| CN113859304A (en) * | 2021-09-23 | 2021-12-31 | 通号城市轨道交通技术有限公司 | Turnout monitoring and controlling method and device |
| CN113859304B (en) * | 2021-09-23 | 2024-01-02 | 通号城市轨道交通技术有限公司 | Switch monitoring and controlling method and device |
| CN114325188A (en) * | 2021-12-28 | 2022-04-12 | 国网河北省电力有限公司经济技术研究院 | Fault detection method and device and server |
| CN114655285A (en) * | 2022-02-25 | 2022-06-24 | 北京全路通信信号研究设计院集团有限公司 | Point switch health degree evaluation method and device, terminal equipment and storage medium |
| CN114655285B (en) * | 2022-02-25 | 2024-01-19 | 北京全路通信信号研究设计院集团有限公司 | Switch machine health evaluation method and device, terminal equipment and storage medium |
| CN114692969A (en) * | 2022-03-29 | 2022-07-01 | 西门子交通技术(北京)有限公司 | Point switch failure prediction method, device, electronic equipment and storage medium |
| CN116482457A (en) * | 2023-03-24 | 2023-07-25 | 四川众合智控科技有限公司 | Intelligent turnout diagnosis method based on relay interlocking |
| CN116628446A (en) * | 2023-05-24 | 2023-08-22 | 成都工业职业技术学院 | Intelligent analysis method and system for turnout health standard value based on clustering algorithm |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110441629B (en) | 2022-02-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110441629A (en) | Method for diagnosing faults and device based on point machine action current curve | |
| Kezunovic et al. | Detect and classify faults using neural nets | |
| CN108416362B (en) | Turnout abnormity early warning and fault diagnosis method | |
| CN105260595B (en) | The feature extracting method and switch breakdown diagnostic method of track switch action current curve | |
| CN107203746A (en) | A kind of switch breakdown recognition methods | |
| CN105184084B (en) | A kind of automatic power-measuring terminal fault type prediction method and system | |
| CN105045256A (en) | Rail traffic real-time fault diagnosis method and system based on data comparative analysis | |
| CN105787511A (en) | Track switch fault diagnosis method and system based on support vector machine | |
| CN109501834A (en) | A kind of point machine failure prediction method and device | |
| CN109633369B (en) | Power grid fault diagnosis method based on multi-dimensional data similarity matching | |
| CN108256738B (en) | Turnout action reference curve selection method and application thereof | |
| CN107909118A (en) | A kind of power distribution network operating mode recording sorting technique based on deep neural network | |
| CN110726898B (en) | Power distribution network fault type identification method | |
| CN112036505A (en) | Method and device for determining equipment state of turnout switch machine and electronic equipment | |
| CN107328560A (en) | A kind of load ratio bridging switch diagnostic method and device | |
| CN108319789A (en) | In conjunction with transformer methods of risk assessment, device, equipment and the medium of abnormal failure | |
| CN110580492A (en) | Track circuit fault precursor discovery method based on small fluctuation detection | |
| CN108919044A (en) | A kind of unit style distribution network failure active identification method based on mutual verification scheme | |
| CN111157850A (en) | Mean value clustering-based power grid line fault identification method | |
| CN111915061A (en) | Switch action current curve prediction method and fault discrimination method thereof | |
| CN113641486B (en) | Intelligent turnout fault diagnosis method based on edge computing network architecture | |
| CN111999591A (en) | Method for identifying abnormal state of primary equipment of power distribution network | |
| CN108238066B (en) | Track switch operation curve template choosing method and its application | |
| CN107064745B (en) | Stagewise method for diagnosing faults based on transient current information and Wavelet Entropy | |
| CN110598750B (en) | Working condition identification method based on switch machine action curve similarity characteristics |
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 | ||
| CB03 | Change of inventor or designer information | ||
| CB03 | Change of inventor or designer information |
Inventor after: Gao Chunhai Inventor after: Zhang Xuan Inventor after: Liu Feng Inventor after: Yang Xinchao Inventor before: Gao Chunhai |
|
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20230721 Address after: No.2 and No.3 building, Beijing headquarters international, No.6 Haiying Road, science and Technology Park, Fengtai District, Beijing 100070 Patentee after: TRAFFIC CONTROL TECHNOLOGY Co.,Ltd. Patentee after: Beijing Yunjie Technology Co.,Ltd. Address before: No.2 and No.3 building, Beijing headquarters international, No.6 Haiying Road, science and Technology Park, Fengtai District, Beijing 100070 Patentee before: TRAFFIC CONTROL TECHNOLOGY Co.,Ltd. |