CN105045983B - A kind of bullet train axletree aging analysis method based on axle temperature data - Google Patents
A kind of bullet train axletree aging analysis method based on axle temperature data Download PDFInfo
- Publication number
- CN105045983B CN105045983B CN201510390862.4A CN201510390862A CN105045983B CN 105045983 B CN105045983 B CN 105045983B CN 201510390862 A CN201510390862 A CN 201510390862A CN 105045983 B CN105045983 B CN 105045983B
- Authority
- CN
- China
- Prior art keywords
- data
- axle temperature
- speed
- axle
- time
- 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.)
- Active
Links
- 230000032683 aging Effects 0.000 title claims abstract description 63
- 238000004458 analytical method Methods 0.000 title claims abstract description 35
- 230000009194 climbing Effects 0.000 claims abstract description 33
- 230000007704 transition Effects 0.000 claims description 42
- 238000000034 method Methods 0.000 claims description 30
- 238000005070 sampling Methods 0.000 claims description 23
- 230000008569 process Effects 0.000 claims description 19
- 238000004364 calculation method Methods 0.000 claims description 18
- 238000012545 processing Methods 0.000 claims description 16
- 101100442582 Neurospora crassa (strain ATCC 24698 / 74-OR23-1A / CBS 708.71 / DSM 1257 / FGSC 987) spe-1 gene Proteins 0.000 claims description 13
- 239000013589 supplement Substances 0.000 claims description 12
- 238000000605 extraction Methods 0.000 claims description 10
- 230000009466 transformation Effects 0.000 claims description 10
- 238000006243 chemical reaction Methods 0.000 claims description 6
- 238000000844 transformation Methods 0.000 claims description 5
- 230000008030 elimination Effects 0.000 claims description 4
- 238000003379 elimination reaction Methods 0.000 claims description 4
- 238000012216 screening Methods 0.000 claims description 4
- 230000000630 rising effect Effects 0.000 claims description 3
- 230000002277 temperature effect Effects 0.000 claims description 3
- 230000000717 retained effect Effects 0.000 claims 1
- 238000001514 detection method Methods 0.000 abstract description 3
- 238000010586 diagram Methods 0.000 description 8
- 238000005516 engineering process Methods 0.000 description 3
- 241000271559 Dromaiidae Species 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 238000002592 echocardiography Methods 0.000 description 2
- 238000009499 grossing Methods 0.000 description 2
- 230000000007 visual effect Effects 0.000 description 2
- 238000005299 abrasion Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000000205 computational method Methods 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000000354 decomposition reaction Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000003862 health status Effects 0.000 description 1
- 238000010438 heat treatment Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000005096 rolling process Methods 0.000 description 1
Landscapes
- Electric Propulsion And Braking For Vehicles (AREA)
- Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)
Abstract
The invention discloses a kind of bullet train axletree aging analysis method based on axle temperature data, specifically implement according to following steps：Step 1, train speed and axle temperature data are pre-processed；Step 2, pretreated axle temperature data progress smooth treatment is carried out to step 1；Step 3, calculate train axle temperature climbing speed；Step 4, the axle temperature climbing speed obtained using step 3 carry out the aging analysis of axletree, and the present invention solves the problems, such as that the detection of axle failures present in prior art is time-consuming and precision is low.
Description
Technical field
The invention belongs to fault diagnosis technology field, and in particular to a kind of bullet train axletree aging based on axle temperature data
Analysis method.
Background technology
With the fast development of high ferro, it is ensured that EMUs running order and improve the securities of EMUs, reliability by
Extensive concern.Axletree is the crucial load bearing component of rolling stock bogie, is the strength member for influenceing traffic safety, if axletree
Occur Aging Damage and extending, will because off-axis and caused by train derail, the consequence of bringing on a disaster property, therefore its safe handling
It is directly related to the safety in production of railway transportation.The main of axle failures detection at present includes two aspects, i.e., based on axle temperature threshold
It is worth and based on artificial maintenance.
First method is, by expertise, the corresponding axle of the setting train failure such as hot axle, warm axle in the process of walking
Warm threshold point.On this basis, the axletree temperature in train travelling process is monitored in real time using temperature sensor.When axle temperature reaches
During each threshold point, then the corresponding measure such as alarm, reduction of speed or parking is taken.Relative to the operating range of bullet train is remote, operation
The features such as velocity interval is big, it is the defects of the method maximum, fails to take into full account the factors such as environment temperature and the speed of service
Influence to axletree temperature, and the unsafe condition such as the health status of unpredictable axletree, the aging to axletree can not play it is pre-
Alert effect.
Second method is, after Train Stopping, assigns experienced service worker to be determined by the mode such as beaing, observing
The abrasion condition of axletree, and with the presence or absence of unsafe conditions such as slight cracks.The defects of the method maximum, is, due in axletree aging
It is difficult to detect in attribute display, so as to cause the aging conditions of axletree to be difficult to judge, and the uncertainty of desk checking.
The content of the invention
It is an object of the invention to provide a kind of bullet train axletree aging analysis method based on axle temperature data, solves existing
There is the problem of detection of axle failures present in technology is time-consuming and precision is low.
The technical solution adopted in the present invention is a kind of bullet train axletree aging analysis method based on axle temperature data,
Specifically implement according to following steps：
Step 1, train speed and axle temperature data are pre-processed；
Step 2, pretreated axle temperature data progress smooth treatment is carried out to step 1；
Step 3, calculate train axle temperature climbing speed；
Step 4, the axle temperature climbing speed obtained using step 3 carry out the aging analysis of axletree.
The features of the present invention also resides in,
Step 1 is specifically implemented according to following steps：
The supplement of step (1.1), shortage of data point：
Train speed spe (i) and axle temperature zw (i) is gathered, wherein i is corresponding sampling time point, and span is 1~n,
The train speed spe (i) and axle temperature zw (i) all values that collect are searched and handled respectively, if 2≤i≤n, i ∈ Z+, Z+
Positive integer is represented, if the i-th -1 point is non-null value point, and i-th point is null value point, i.e. zw (i-1) ≠ null, zw (i)=
Null, processing procedure are as follows：
If zw (i) is isolated missing point, i.e. zw (i-1) ≠ null, zw (i)=null, zw (i+1) ≠ null, then will
The value of i-th -1 point is assigned to i-th of value, that is, carries out zw (i)=zw (i-1) assignment operation；
If occurring continuous missing point in former data, its number is m (m>2), i.e. zw (i+1), zw (i+2) ..., zw (i+
M-1 average) is null, zw (i+m) ≠ null, then according to m≤n or m>N, a point situation is discussed, as follows：
(1) if m>N, then it represents that data are imperfect, and analysis terminates；
(2) if m≤n, linear interpolation processing is carried out, detailed process is as follows：
Step a:Order
Step b:Interpolation operation is carried out to all continuous missing points, i.e.,Zw (i+j)=zw
(i-1)+(j+1) × delta, is obtained
Zw (i)=zw (i-1)+delta,
Zw (i+1)=zw (i-1)+2 × delta,
Zw (i+m-1)=zw (i-1)+(m-1) × delta
Similarly, as above same operation is carried out to train speed data spe (i) according to above step, so as to by all numbers
Complete according to missing point supplement, axle temperature and speed data after supplement are designated as zw1 (i) and spe1 (i) respectively；
The elimination of step (1.2), isolated zero point：
Axle temperature data zw1 (i) that judgment step (1.1) obtains and speed data spe1 (i) whether be isolated zero point simultaneously
Handled, concrete operations are as follows：
If i-th of axle temperature data is not 0, i.e. zw1 (i) ≠ 0, then without any operation；
If i-th of axle temperature data is 0, i.e. zw1 (i)=0, then continue to judge the data zw1 (i-1) and zw1 (i+ that its is adjacent
1) whether also it is zero point, it is as follows：
(1) if the i-th -1 axle temperature data and i+1 axle temperature data one and only one be 0, i.e. zw1 (i-1)=0 or zw1 (i
+ 1)=0, then zw1 (i)=0 is considered as normal data, without any operation；
(2) if the i-th -1 axle temperature data are not 0 for 0 and i+1 axle temperature data yet, i.e. zw1 (i-1) ≠ 0 and zw1 (i+1)
≠ 0, then zw1 (i) is considered as isolated zero point, now
Similarly, identical is also carried out to speed data spe1 (i) according to above step to judge and handle, all are isolated
Zero point has been handled, and is eliminated the axle temperature data obtained after isolated point and is designated as zw2 (i), obtained speed data is designated as spe2 (i).
Step 2 is specifically implemented according to following steps：
Pretreated axle temperature data zw2 (i) is carried out to step 1 and carries out TIME layer scattering wavelet transformations, TIME layer scatterings
Wavelet transformation detailed process is as follows：
Step a:Aray variable data is set, and makes data=zw2 (i)；Intermediate variable time=0 is set；
Step b:Data in step a is updated according to equation below：And
Intermediate variable time is carried out plus 1 operates, i.e. time=time+1, wherein, H (k) represents low pass filter, and K represents H's (k)
Length, j represent array data lengthI.e.
Step c:If time<TIME, then return to step b, if time >=TIME, completes the smooth treatment of data,
And the axle temperature data after smooth treatment are designated as array DATA (i), even DATA (i)=data；
It is eliminated by the noise signal in above-mentioned conversion process axle temperature data, makes axle temperature curve more smooth, is temperature
The first transition of data, which obtains, provides basis.
In the step b of step 2, H (k) length K=length (data), represent identical with axle temperature data data length.
In the step c of step 2, TIME ∈ Z+。
Step 3 is specifically implemented according to following steps：
The extraction of step (3.1), axle temperature first transition：
To the axle temperature data after the step 2 processing, the extraction of first transition is carried out, it is specific as follows：By step 2 place
Axle temperature data after reason make the difference successively, and difference is designated as φ, i.e. φ=DATA (i)-DATA (i+1), if φ>When 0, illustrate the axle temperature
The axle temperature in section [i, i+1] is not in ascent stage, it may be possible to is declining or is not changing, is then continuing to make the difference extraction；If φ<0
When, the axle temperature section [i, i+1] is axle temperature first transition, stores the starting point i and terminal i+1 of the sampled point of first transition, and is deposited
In new array up；
Step (3.2), the screening of the speed of axle temperature first transition：
If threshold speed is V, if each first transition starting point i and terminal j in the step (3.1), during corresponding sampling
Between point be i × 2TIMEWith j × 2TIME, then corresponding velocity amplitude spe2 (i × 2TIME) and spe2 (j × 2TIME), if spe2 (i ×
2TIME)<V and spe2 (j × 2TIME)<V, then illustrate that the ascent stage [i, j] is shutdown phase, the train under ambient temperature effect
Axle temperature heats up, then does not retain the axle temperature first transition；If spe2 (i × 2TIME) >=V and spe2 (j × 2TIME) >=V, then explanation should
Axle temperature first transition when ascent stage [i, j] is train operation, retain and continue to judge；
The calculating of step (3.3), axle temperature climbing speed：
Climbing speed calculating is carried out to the axle temperature data after step (3.2) processing, calculating process is as follows：
Step (3.3.1), the axle temperature difference of each first transition of axle temperature being designated as to fz, corresponding each rise time is designated as t, and respectively
The average of the rate of climb is designated as v, and fz, t, v calculation formula difference are as follows：
Fz=DATA (j)-DATA (i),
T=(j-i) × 2TIME,
Wherein i and j is respectively the beginning and end of axle temperature first transition；
Step (3.3.2), axle temperature climbing speed is set as SSJSL, SSJSL calculation formula isThe ratio of axle temperature summation and speed average summation is designated as SSSSV, and its calculation formula is
In step (3.2), threshold speed V=2.
Step 4 is specifically implemented according to following steps：
Step (4.1), the high ferro axle temperature climbing speed SSJSL for calculating n days in the step 3i(i=1,2 ... n), calculates
Formula isWherein m is i-th day axle temperature data point number；
Step (4.2), by the high ferro axle temperature climbing speed SSJSL of n days obtained in step (4.1)i(i=1,2 ... n)
Rise reference speed rate SSJSL with the axletree axle temperature0Compare, axletree axle temperature rises reference speed rate SSJSL0Calculation formula isPolydispersity index number of days on the basis of k, k=100, the judgement of axletree aging
It is with foundation：
(1) as α >=a, the slight aging of train axle,
(2) as α >=b, train axle severe aging,
Wherein a<B, a, b ∈ R, a are slight aging coefficients, and a=1.1, b are severe aging coefficients, b=2.5.
The invention has the advantages that a kind of bullet train axletree aging analysis method based on axle temperature data, to high ferro
Train data is analyzed, and realizes the automatic calculating of axle temperature rate of change, and axletree aging analysis can be realized based on data, is used
Wavelet transform, after data are handled, allows axle temperature data trend to become apparent, and is easy to calculate the upper raising speed of data
Rate, it is good by data continuity after pretreatment, directly first transition can be judged by difference, be easy to automatically extract rising area
Between, climbing speed etc. is calculated, with reference to train speed, establishes temperature rate-of-rise computational methods, new method is more suitable for high ferro axle temperature
The calculating of climbing speed.
Brief description of the drawings
Fig. 1 is a kind of bullet train axletree aging analysis method flow diagram based on axle temperature data of the present invention；
Fig. 2 is that shortage of data point supplements in a kind of bullet train axletree aging analysis method based on axle temperature data of the present invention
Flow chart is eliminated with isolated zero point；
Fig. 3 is TIME layer scattering small echos in a kind of bullet train axletree aging analysis method based on axle temperature data of the present invention
The flow chart of conversion；
Fig. 4 is to judge first transition stream in a kind of bullet train axletree aging analysis method based on axle temperature data of the present invention
Cheng Tu；
Fig. 5 is on judgement meets the requirements in a kind of bullet train axletree aging analysis method based on axle temperature data of the present invention
Rise section and calculate total speed flow chart；
Fig. 6 is that axle temperature is calculated in a kind of bullet train axletree aging analysis method based on axle temperature data of the present invention with being averaged
Speed flow chart；
Fig. 7 is initial axle temperature signature tune in a kind of bullet train axletree aging analysis method based on axle temperature data of the present invention
Line schematic diagram；
Fig. 8 is the axle temperature letter after interpolation in a kind of bullet train axletree aging analysis method based on axle temperature data of the present invention
Number curve synoptic diagram；
Fig. 9 is TIME layer scattering small echos in a kind of bullet train axletree aging analysis method based on axle temperature data of the present invention
Axle temperature data and curves schematic diagram after conversion decomposition.
Embodiment
The present invention is described in detail with reference to the accompanying drawings and detailed description.
A kind of bullet train axletree aging analysis method based on axle temperature data of the present invention, flow chart is as shown in figure 1, specific
Implement according to following steps：
Step 1, train speed and axle temperature data are pre-processed, as shown in Figure 2：
The supplement of step (1.1), shortage of data point：
Train speed spe (i) and axle temperature zw (i) is gathered, wherein i is corresponding sampling time point, and span is 1~n,
The train speed spe (i) and axle temperature zw (i) all values that collect are searched and handled respectively, if 2≤i≤n, i ∈ Z+, Z+
Positive integer is represented, if the i-th -1 point is non-null value point, and i-th point is null value point, i.e. zw (i-1) ≠ null, zw (i)=
Null, processing procedure are as follows：
If zw (i) is isolated missing point, i.e. zw (i-1) ≠ null, zw (i)=null, zw (i+1) ≠ null, then will
The value of i-th -1 point is assigned to i-th of value, that is, carries out zw (i)=zw (i-1) assignment operation；
If occurring continuous missing point in former data, its number is m (m>2), i.e. zw (i+1), zw (i+2) ..., zw (i+
M-1 average) is null, zw (i+m) ≠ null, then according to m≤n or m>N, a point situation is discussed, as follows：
(1) if m>N, then it represents that data are imperfect, and analysis terminates；
(2) if m≤n, linear interpolation processing is carried out, detailed process is as follows：
Step a:Order
Step b:Interpolation operation is carried out to all continuous missing points, i.e.,Zw (i+j)=zw
(i-1)+(j+1) × delta, is obtained
Zw (i)=zw (i-1)+delta,
Zw (i+1)=zw (i-1)+2 × delta,
Zw (i+m-1)=zw (i-1)+(m-1) × delta
Similarly, as above same operation is carried out to train speed data spe (i) according to above step, so as to by all numbers
Complete according to missing point supplement, axle temperature and speed data after supplement are designated as zw1 (i) and spe1 (i) respectively；
The elimination of step (1.2), isolated zero point：
Axle temperature data zw1 (i) that judgment step (1.1) obtains and speed data spe1 (i) whether be isolated zero point simultaneously
Handled, concrete operations are as follows：
If i-th of axle temperature data is not 0, i.e. zw1 (i) ≠ 0, then without any operation；
If i-th of axle temperature data is 0, i.e. zw1 (i)=0, then continue to judge the data zw1 (i-1) and zw1 (i+ that its is adjacent
1) whether also it is zero point, it is as follows：
(1) if the i-th -1 axle temperature data and i+1 axle temperature data one and only one be 0, i.e. zw1 (i-1)=0 or zw1 (i
+ 1)=0, then zw1 (i)=0 is considered as normal data, without any operation；
(2) if the i-th -1 axle temperature data are not 0 for 0 and i+1 axle temperature data yet, i.e. zw1 (i-1) ≠ 0 and zw1 (i+1)
≠ 0, then zw1 (i) is considered as isolated zero point, now
Similarly, identical is also carried out to speed data spe1 (i) according to above step to judge and handle, all are isolated
Zero point has been handled, and is eliminated the axle temperature data obtained after isolated point and is designated as zw2 (i), obtained speed data is designated as spe2 (i)；
Step 2, pretreated axle temperature data progress smooth treatment is carried out to the step 1：
Pretreated axle temperature data zw2 (i) is carried out to step 1 and carries out TIME layer scattering wavelet transformations, as shown in figure 3,
TIME∈Z+, specific value is depending on actual conditions.TIME layer scattering wavelet transformation detailed processes are as follows：
Step a:Aray variable data is set, and makes data=zw2 (i)；Intermediate variable time=0 is set；
Step b:Data in step a is updated according to equation below：And
Intermediate variable time is carried out plus 1 operates, i.e. time=time+1, wherein, H (k) represents low pass filter, and K represents H's (k)
Length, H (k) length K=length (data), represent identical with axle temperature data data length, j represents array data lengthI.e.
Step c:If time<TIME, then return to step b, if time >=TIME, completes the smooth treatment of data,
And the axle temperature data after smooth treatment are designated as array DATA (i), even DATA (i)=data；
It is eliminated by the noise signal in above-mentioned conversion process axle temperature data, makes axle temperature curve more smooth, is temperature
The first transition of data, which obtains, provides basis；
Step 3, calculate train axle temperature climbing speed：
The extraction of step (3.1), axle temperature first transition：
To the axle temperature data after step 2 processing, the extraction of first transition is carried out, as shown in figure 4, specific as follows：Will
Axle temperature data after step 2 processing make the difference successively, and difference is designated as φ, i.e. φ=DATA (i)-DATA (i+1), if φ>When 0, say
The axle temperature in the bright axle temperature section [i, i+1] is not in ascent stage, it may be possible to is declining or is not changing, is then continuing to make the difference and carry
Take；If φ<When 0, the axle temperature section [i, i+1] is axle temperature first transition, stores the starting point i and terminal of the sampled point of first transition
I+1, and exist in new array up；
Step (3.2), the screening of the speed of axle temperature first transition, as shown in Figure 5：
If threshold speed is V, threshold speed V=2, in a practical situation, due to the difference of vehicle and bearing, institute be present
So that the value of threshold speed is not limited in V=2, can be with value threshold speed V=N+, N+For positive natural number, if step
(3.1) sampling time point corresponding to each the first transition starting point i and terminal j in is i × 2TIMEWith j × 2TIME, then corresponding speed
Value spe2 (i × 2TIME) and spe2 (j × 2TIME), if spe2 (i × 2TIME)<V and spe2 (j × 2TIME)<V, then illustrate the rising
Stage [i, j] is shutdown phase, the train axle temperature heating under ambient temperature effect, does not then retain the axle temperature first transition, if
spe2(i×2TIME) >=V and spe2 (j × 2TIME) >=V, then illustrate that the axle temperature when ascent stage [i, j] is train operation rises
Section, retain and continue to judge；
The calculating of step (3.3), axle temperature climbing speed, as shown in Figure 6：
Climbing speed calculating is carried out to the axle temperature data after step (3.2) processing, calculating process is as follows：
Step (3.3.1), the axle temperature difference of each first transition of axle temperature being designated as to fz, corresponding each rise time is designated as t, and respectively
The average of the rate of climb is designated as v, and fz, t, v calculation formula difference are as follows：
Fz=DATA (j)-DATA (i),
T=(j-i) × 2TIME,
Wherein i and j is respectively the beginning and end of axle temperature first transition；
Step (3.3.2), axle temperature climbing speed is set as SSJSL, SSJSL calculation formula isThe ratio of axle temperature summation and speed average summation is designated as SSSSV, and its calculation formula is
Step 4, the axle temperature climbing speed obtained using the step 3 carry out the aging analysis of axletree：
The high ferro axle temperature climbing speed SSJSL of n days in step (4.1), calculation procedure 3i(i=1,2 ... n), calculation formula
ForWherein m is i-th day axle temperature data point number；
Step (4.2), by the high ferro axle temperature climbing speed SSJSL of n days obtained in step (4.1)i(i=1,2 ... n)
Rise reference speed rate SSJSL with the axletree axle temperature0Compare, axletree axle temperature rises reference speed rate SSJSL0Calculation formula isPolydispersity index number of days on the basis of k, k=100, the judgement of axletree aging
It is with foundation：
(1) as α >=a, the slight aging of train axle；
(2) as α >=b, train axle severe aging,
Wherein a<B, a, b ∈ R, a are slight aging coefficients, and a=1.1, b are severe aging coefficients, b=2.5.
In the inventive method, slight aging coefficient a and severe aging coefficient b, threshold speed V, according to vehicle difference, axle
The different differences that can also exist numerically are held, but for the art, corresponding vehicle and bearing can correspond to a reality
Empirical value under the running situation of border, but no matter the slight aging coefficient and the value that severe aging coefficient and threshold speed are V have
Body is how many, and what is proposed in the application judges that bullet train axletree aging analysis method is still equally effective and feasible, so, though
Slight aging coefficient a=1.1, severe aging coefficient b=2.5, threshold speed V=2 in right the application, but can similarly cover
Any congeniality patent for analyzing bullet train axletree aging in this way.
A kind of bullet train axletree aging analysis method based on axle temperature data of the present invention, accurately to grasp train axle
Aging performance, foundation is provided for the fault pre-alarming of train axle, the service data based on train, by being pre-processed to data
And slickness is handled, the ascent stage Origin And Destination of each axletree temperature, and its corresponding time difference are then found out, is finally calculated
Axle temperature climbing speed, and the decision rule of axletree aging is formulated, whole method is according to clear so that observation is more blunt, helps
In bullet train operational management.
Embodiment
A kind of bullet train axletree aging analysis method based on axle temperature data of the present invention, it is specifically real according to following steps
Apply：
Step 1, train speed and axle temperature data are pre-processed：
The supplement of step (1.1), shortage of data point：
Gather train speed spe (i) and axle temperature zw (i), wherein i sampling time point, i=20, such as table 1 and table for corresponding to
Shown in 2, initial axle temperature signal curve schematic diagram is drawn,
Table 1：Axle temperature zw (i) sampled data
| Sampling time point | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 |
| Axle temperature | 28.67249 | 30.28081 | 30.28081 | 30.28081 | 31.88913 | 31.88913 | 0 | |
| Sampling time point | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 |
| Axle temperature | 33.49745 | 35.10577 | 35.10577 | 36.71409 | 36.71409 | 36.71409 | 38.32241 | 38.32241 |
| Sampling time point | 17 | 18 | 19 | 20 | ||||
| Axle temperature | 39.93074 | 39.93074 | 41.53906 | 41.53906 |
Table 2：Speed spe (i) sampled data
| Sampling time point | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 |
| Speed | 0.445477 | 2.545584 | 8.893636 | 23.88076 | 44.89775 | 76.57436 | 111.6239 | |
| Sampling time point | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 |
| Speed | 136.0297 | 161.0719 | 171.5724 | 183.7116 | 193.369 | 200.7352 | 213.3359 | 217.2815 |
| Sampling time point | 17 | 18 | 19 | 20 | ||||
| Speed | 0 | 268.8455 | 265.8067 | 259.0291 |
Respectively to the train speed spe (i) that collects and axle temperature zwi) all values are searched and are handled, and can be with by table 1
Find out, zw (5) is isolated missing point, i.e. zw (4) ≠ null, zw (5)=null, zw (6) ≠ null, then by the 4th point of value
The 5th value is assigned to, that is, zw (5)=zw (4) assignment operation is carried out, similarly, according to above step to train speed data spe (i)
As above same operation is carried out, complete so as to which all shortage of data points be supplemented, axle temperature and speed data after supplement are remembered respectively
For zw1 (i) and spe1 (i), as shown in Table 3 and Table 4,
Table 3：The train axle temperature zw1 (i) after shortage of data point is supplemented
| Sampling time point | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 |
| Axle temperature | 28.67249 | 30.28081 | 30.28081 | 30.28081 | 30.28081 | 31.88913 | 31.88913 | 0 |
| Sampling time point | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 |
| Axle temperature | 33.49745 | 35.10577 | 35.10577 | 36.71409 | 36.71409 | 36.71409 | 38.32241 | 38.32241 |
| Sampling time point | 17 | 18 | 19 | 20 | ||||
| Axle temperature | 39.93074 | 39.93074 | 41.53906 | 41.53906 |
Table 4：The train speed spe1 (i) after shortage of data point is supplemented
| Sampling time point | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 |
| Speed | 0.445477 | 2.545584 | 8.893636 | 23.88076 | 44.89775 | 44.89775 | 76.57436 | 111.6239 |
| Sampling time point | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 |
| Speed | 136.0297 | 161.0719 | 171.5724 | 183.7116 | 193.369 | 200.7352 | 213.3359 | 217.2815 |
| Sampling time point | 17 | 18 | 19 | 20 | ||||
| Speed | 0 | 268.8455 | 265.8067 | 259.0291 |
The elimination of step (1.2), isolated zero point：
Axle temperature data zw1 (i) that judgment step (1.1) obtains and speed data spe1 (i) whether be isolated zero point simultaneously
Handled, concrete operations are as follows：
8th axle temperature data are 0, i.e. zw1 (8)=0, then whether continue to judge data zw1 (7) that its is adjacent and zw1 (9)
Also it is zero point, it is as follows：
7th axle temperature data are not 0 for the 0 and the 9th axle temperature data yet, i.e. zw1 (7) ≠ 0 and zw1 (9) ≠ 0, then by zw1
(8) it is considered as isolated zero point, now
Similarly, identical is also carried out to speed data spe1 (i) according to above step to judge and handle, all are isolated
Zero point has been handled, and is eliminated the axle temperature data obtained after isolated point and is designated as zw2 (i), and obtained speed data is designated as spe2 (i),
As shown in table 5 and table 6：
Table 5：Eliminate the train axle temperature zw2 (i) after isolated zero point
Table 6：Eliminate the train speed spe2 (i) after isolated zero point
| Sampling time point | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 |
| Speed | 0.445477 | 2.545584 | 8.893636 | 23.88076 | 44.89775 | 44.89775 | 76.57436 | 111.6239 |
| Sampling time point | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 |
| Speed | 136.0297 | 161.0719 | 171.5724 | 183.7116 | 193.369 | 200.7352 | 213.3359 | 217.2815 |
| Sampling time point | 17 | 18 | 19 | 20 | ||||
| Speed | 243.0635 | 268.8455 | 265.8067 | 259.0291 |
Step 2, pretreated axle temperature data progress smooth treatment is carried out to step 1：
Pretreated axle temperature data zw2 (i) is carried out to step 1 and carries out TIME layer scattering wavelet transformations, in the present embodiment
In, TIME=1, TIME layer scattering wavelet transformation detailed processes are as follows：
Step a:Aray variable data is set, and makes data=zw2 (i)；Intermediate variable time=0 is set；
Step b:Data in step a is updated according to equation below：And
Intermediate variable time is carried out plus 1 operates, i.e. time=time+1, wherein, H (k) represents low pass filter, and K represents H's (k)
Length, H (k) length K=length (data), represent identical with axle temperature data data length, i.e. K=20, j represent array
Data lengthI.e.
Step c:If time<TIME, then return to step b, if time >=TIME, completes the smooth treatment of data,
And the axle temperature data after smooth treatment are designated as array DATA (i), even DATA (i)=data；
It is eliminated by the noise signal in above-mentioned conversion process axle temperature data, makes axle temperature curve more smooth, is temperature
The first transition of data, which obtains, provides basis, and the train axle temperature DATA (i) after smoothing processing is as shown in table 7：
Train axle temperature DATA (i) after the smoothing processing of table 7
Step 3, calculate train axle temperature climbing speed：
The extraction of step (3.1), axle temperature first transition：
To the axle temperature data after step 2 processing, the extraction of first transition is carried out, it is specific as follows：After step 2 is handled
Axle temperature data make the difference successively, difference is designated as φ, i.e. φ=DATA (i)-DATA (i+1), obtains result as shown in table 8:
The axle temperature difference table of table 8
| DATA(i)-DATA(i+1) | DATA(1)-DATA(2) | DATA(3)-DATA(2) | DATA(4)-DATA(3) |
| Axle temperature difference | -1.1372 | -1.1373 | -1.7059 |
| DATA(i)-DATA(i+1) | DATA(5)-DATA(4) | DATA(6)-DATA(5) | DATA(7)-DATA(6) |
| Axle temperature difference | -2.8431 | -2.2745 | -1.1373 |
| DATA(i)-DATA(i+1) | DATA(8)-DATA(7) | DATA(9)-DATA(8) | DATA(10)-DATA(9) |
| Axle temperature difference | -2.2745 | -2.2745 | -2.2745 |
By the axle temperature of table 8 make the difference it can be seen from difference be respectively less than 0, then belong to ascent stage, then by axle temperature rise correspondence
The beginning and end of sampling time ascent stage point is present in new array up, now up=[1,10]；
Step (3.2), the screening of the speed of axle temperature first transition：
If threshold speed is V, threshold speed V=2, to the array obtained in step (3.1), first transition is calculated
[2,20], corresponding speed spe2 (2)=2.545584 and spe2 (20)=259.0291 are found from table 6, it can be seen that spe2
(2) >=2 and spe2 (20) >=2, then the axle temperature first transition when ascent stage is train operation, now, array up=are illustrated
[1,10]；
The calculating of step (3.3), axle temperature climbing speed：
Climbing speed calculating is carried out to the axle temperature data after step (3.2) processing, calculating process is as follows：
Step (3.3.1), the axle temperature difference of each first transition of axle temperature being designated as to fz, corresponding each rise time is designated as t, and respectively
The average of the rate of climb is designated as v, and fz, t, v calculation formula difference are as follows：
fz1=DATA (j)-DATA (i)=58.7451-41.6863=17.0588,
t1=(j-i) × 2TIME=9 × 2=18,
Wherein i and j is respectively the beginning and end of axle temperature first transition；
Step (3.3.2), axle temperature climbing speed is set as SSJSL, SSJSL calculation formula isThe ratio of axle temperature summation and speed average summation is designated as SSSSV, its calculation formula
For
Step 4, the axle temperature climbing speed obtained using the step 3 carry out the aging analysis of axletree：
High ferro axle temperature climbing speed SSJSL and the axletree axle temperature are risen into reference speed rate SSJSL0Compare, axletree axle temperature
Rise reference speed rate SSJSL0For SSJSL0=0.8299, judgement and the foundation of axletree aging are：
Due to a=1.1, b=2.5, then a<α<B, it can be seen that train axle belongs to slight aging, according to actual conditions,
With train long-play, axle temperature climbing speed gradually increases.
Because data are less in above example, cause the curve visual effect drawn out and unobvious, in order to up to
To a kind of optimal visual effect, axle temperature gathered data amount is expanded as into individual sampled point more than 70,000, draws initial axle temperature signal curve
Schematic diagram, such as Fig. 7, after data enter row interpolation in Fig. 7, axle temperature signal curve schematic diagram is as shown in figure 8, to after pretreatment
Fig. 8 in axle temperature data carry out TIME layer scattering wavelet transformations, the axle temperature data after TIME layer scatterings wavelet transformation decomposes are bent
Line schematic diagram is smoothed as shown in figure 9, TIME is set into 6 herein.
Claims (7)
- A kind of 1. bullet train axletree aging analysis method based on axle temperature data, it is characterised in that specifically according to following steps Implement：Step 1, train speed and axle temperature data are pre-processed；Step 2, pretreated axle temperature data progress smooth treatment is carried out to the step 1；Step 3, train axle temperature climbing speed is calculated, specifically implemented according to following steps：The extraction of step (3.1), axle temperature first transition, detailed process are as follows：To the axle temperature data after the step 2 processing, the extraction of first transition is carried out, it is specific as follows：After step 2 is handled Axle temperature data make the difference successively, difference is designated as φ, φ=DATA (i)-DATA (i+1), if during φ ＞ 0, illustrating the axle temperature section The axle temperature of [i, i+1] is not in ascent stage, it may be possible to is declining or is not changing, is then continuing to make the difference extraction；If during φ ＜ 0, The axle temperature section [i, i+1] is axle temperature first transition, stores the starting point i and terminal i+1 of the sampled point of first transition, and is existed new In array up；Step (3.2), the screening of the speed of axle temperature first transition：If threshold speed is V, if sampling time point corresponding to each first transition starting point i and terminal j in the step (3.1) is i×2TIMEWith j × 2TIME, then corresponding velocity amplitude spe2 (i × 2TIME) and spe2 (j × 2TIME)；If spe2 (i × 2TIME) ＜ V And spe2 (j × 2TIME) ＜ V, then illustrate that the ascent stage [i, j] is probably shutdown phase, the train axle under ambient temperature effect Temperature rise temperature, then the axle temperature first transition is not retained；If spe2 (i × 2TIME) >=V and spe2 (j × 2TIME) >=V, then illustrate on this Axle temperature first transition when the stage of liter [i, j] is train operation, retains and continues to judge；The calculating of step (3.3), axle temperature climbing speed：Climbing speed calculating is carried out to the axle temperature data after step (3.2) processing, calculating process is as follows：Step (3.3.1), the axle temperature difference of each first transition of axle temperature is designated as to fz, corresponding each rise time is designated as t, and each rising The average of speed is designated as v, and fz, t, v calculation formula difference are as follows：Fz=DATA (j)-DATA (i),T=(j-i) × 2TIME,Wherein i and j is respectively the beginning and end of axle temperature first transition；Step (3.3.2), axle temperature climbing speed is set as SSJSL, SSJSL calculation formula isThe ratio of axle temperature summation and speed average summation is designated as SSSSV, and its calculation formula isStep 4, the axle temperature climbing speed obtained using the step 3 carry out the aging analysis of axletree.
- 2. a kind of bullet train axletree aging analysis method based on axle temperature data according to claim 1, its feature exist In the step 1 is specifically implemented according to following steps：The supplement of step (1.1), shortage of data point：Train speed spe (i) and axle temperature zw (i) is gathered, wherein i is corresponding sampling time point, and span is 1~n, respectively The train speed spe (i) and axle temperature zw (i) all values collected is searched and handled, if 2≤i≤n, i ∈ Z+, Z+Represent Positive integer, if the i-th -1 point is non-null value point, and i-th point is null value point, i.e. zw (i-1) ≠ null, zw (i)=null, place Reason process is as follows：If zw (i) is isolated missing point, i.e. zw (i-1) ≠ null, zw (i)=null, zw (i+1) ≠ null, then by i-th- 1 point of value is assigned to i-th of value, that is, carries out zw (i)=zw (i-1) assignment operation；If occurring continuous missing point in former data, its number is m, m ＞ 2, i.e. zw (i+1), zw (i+2) ..., zw (i+m-1) Average be null, zw (i+m) ≠ null, then according to m≤n or m ＞ n, a point situation is discussed, as follows：(1) if m ＞ n, then it represents that data are imperfect, and analysis terminates；(2) if m≤n, linear interpolation processing is carried out, detailed process is as follows：Step a:OrderStep b:Interpolation operation is carried out to all continuous missing points, i.e.,Zw (i+j)=zw (i-1) + (j+1) × delta, is obtainedSimilarly, as above same operation is carried out to train speed data spe (i) according to above step, so as to which all data be lacked Lose point supplement completely, axle temperature and speed data after supplement are designated as zw1 (i) and spe1 (i) respectively；The elimination of step (1.2), isolated zero point：Judge axle temperature data zw1 (i) that the step (1.1) obtains and speed data spe1 (i) whether be isolated zero point simultaneously Handled, concrete operations are as follows：If i-th of axle temperature data is not 0, i.e. zw1 (i) ≠ 0, then without any operation；If i-th axle temperature data are 0, i.e. zw1 (i)=0, then continue to judge that data zw1 (i-1) that its is adjacent and zw1 (i+1) are No is also zero point, as follows：(1) if the i-th -1 axle temperature data and i+1 axle temperature data one and only one be 0, i.e. zw1 (i-1)=0 or zw1 (i+1) =0, then zw1 (i)=0 is considered as normal data, without any operation；(2) if the i-th -1 axle temperature data are not 0 for 0 and i+1 axle temperature data yet, i.e. zw1 (i-1) ≠ 0 and zw1 (i+1) ≠ 0, Zw1 (i) is then considered as isolated zero point, nowSimilarly, identical is also carried out to speed data spe1 (i) according to above step to judge and handle, by all isolated zeros Point has been handled, and is eliminated the axle temperature data obtained after isolated point and is designated as zw2 (i), obtained speed data is designated as spe2 (i).
- 3. a kind of bullet train axletree aging analysis method based on axle temperature data according to claim 1, its feature exist In the step 2 is specifically implemented according to following steps：Pretreated axle temperature data zw2 (i) is carried out to the step 1 and carries out TIME layer scattering wavelet transformations, TIME layer scatterings Wavelet transformation detailed process is as follows：Step a:Aray variable data is set, and makes data=zw2 (i)；Intermediate variable time=0 is set；Step b:Data in the step a is updated according to equation below：And Intermediate variable time is carried out plus 1 operates, i.e. time=time+1, wherein, H (k) represents low pass filter, and j represents array Data lengthI.e.Step c:If time ＜ TIME, return to step b, if time >=TIME, the smooth treatment of data is completed, and will Axle temperature data after smooth treatment are designated as array DATA (i), even DATA (i)=data；It is eliminated by the noise signal in above-mentioned conversion process axle temperature data, makes axle temperature curve more smooth, is temperature data First transition obtain provide basis.
- 4. a kind of bullet train axletree aging analysis method based on axle temperature data according to claim 3, its feature exist In, in the step b of the step 2, H (k) length K=length (data), represent it is identical with axle temperature data data length.
- 5. a kind of bullet train axletree aging analysis method based on axle temperature data according to claim 3, its feature exist In, in the step c of the step 2, TIME ∈ Z+。
- 6. a kind of bullet train axletree aging analysis method based on axle temperature data according to claim 1, its feature exist In, in the step (3.2), threshold speed V=2.
- 7. a kind of bullet train axletree aging analysis method based on axle temperature data according to claim 1, its feature exist In the step 4 is specifically implemented according to following steps：Step (4.1), the high ferro axle temperature climbing speed SSJSL for calculating n days in the step 3i(i=1,2 ... n), calculation formula ForWherein m is i-th day axle temperature data point number；Step (4.2), by the high ferro axle temperature climbing speed SSJSL of n days obtained in the step (4.1)i(i=1,2 ... n) with The axletree axle temperature rises reference speed rate SSJSL0Compare, axletree axle temperature rises reference speed rate SSJSL0Calculation formula isPolydispersity index number of days on the basis of k, k=100, the judgement of axletree aging It is with foundation：(1) as α >=a, the slight aging of train axle；(2) as α >=b, train axle severe aging,Wherein a ＜ b, a, b ∈ R, a are slight aging coefficients, and b is severe aging coefficient, a=1.1, b=2.5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510390862.4A CN105045983B (en) | 2015-07-06 | 2015-07-06 | A kind of bullet train axletree aging analysis method based on axle temperature data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510390862.4A CN105045983B (en) | 2015-07-06 | 2015-07-06 | A kind of bullet train axletree aging analysis method based on axle temperature data |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105045983A CN105045983A (en) | 2015-11-11 |
| CN105045983B true CN105045983B (en) | 2018-02-02 |
Family
ID=54452526
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510390862.4A Active CN105045983B (en) | 2015-07-06 | 2015-07-06 | A kind of bullet train axletree aging analysis method based on axle temperature data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105045983B (en) |
Families Citing this family (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105976578B (en) * | 2016-07-21 | 2018-07-03 | 西安理工大学 | A kind of bullet train axle temperature dynamic alert threshold setting method based on monitoring data |
| CN106528940B (en) * | 2016-10-13 | 2019-04-02 | 中车长春轨道客车股份有限公司 | The method that evaluation and early warning are carried out to train axle performance based on mathematical model |
| CN106828106B (en) * | 2016-12-29 | 2019-07-09 | 北京交通大学 | A kind of fault early warning method towards EMU traction motor |
| CN107255560B (en) * | 2017-07-04 | 2019-11-01 | 西安理工大学 | A kind of failure modes of high-speed train braking pipe and diagnostic method based on pressure data |
| CN107527140A (en) * | 2017-07-28 | 2017-12-29 | 西安理工大学 | A kind of bullet train operating condition based on fuzzy membership determines method |
| CN107560848B (en) * | 2017-08-03 | 2019-11-22 | 北京交通大学 | Axis temperature variation model construction method and module, bearing health monitor method and system |
| CN109086832B (en) * | 2018-08-20 | 2021-11-16 | 西安理工大学 | Visual trend discrimination method for mass monitoring big data of train |
| CN111024416B (en) * | 2018-10-09 | 2021-08-10 | 株洲中车时代电气股份有限公司 | Fault diagnosis method and system for train traction system |
| CN110210174A (en) * | 2019-06-21 | 2019-09-06 | 西南交通大学 | A kind of EMU axle box bearing temperature predicting method based on multilayer LSTM |
| CN111537230A (en) * | 2020-04-27 | 2020-08-14 | 中车青岛四方机车车辆股份有限公司 | Train bearing temperature early warning method and device, electronic equipment and storage medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101181901A (en) * | 2007-12-03 | 2008-05-21 | 北京康拓红外技术有限公司 | Hotbox distinguishing method for train infrared shaft-temperature detecting system |
| CN102455220A (en) * | 2010-10-27 | 2012-05-16 | 北京南车时代信息技术有限公司 | Rail temperature monitoring method and device |
| KR101387018B1 (en) * | 2013-10-31 | 2014-04-18 | 한국철도공사 | Axle temperature measuring device in hbd electric rack and method thereof |
-
2015
- 2015-07-06 CN CN201510390862.4A patent/CN105045983B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101181901A (en) * | 2007-12-03 | 2008-05-21 | 北京康拓红外技术有限公司 | Hotbox distinguishing method for train infrared shaft-temperature detecting system |
| CN102455220A (en) * | 2010-10-27 | 2012-05-16 | 北京南车时代信息技术有限公司 | Rail temperature monitoring method and device |
| KR101387018B1 (en) * | 2013-10-31 | 2014-04-18 | 한국철도공사 | Axle temperature measuring device in hbd electric rack and method thereof |
Non-Patent Citations (2)
| Title |
|---|
| 基于DTW的车辆轴温监测方法;曹源 等;《交通运输工程学报》;20150630;第15卷(第3期);78-100 * |
| 高速铁路列车轴温探测系统的研究与设计;刘涛;《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》;20081215(第12期);C033-115 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105045983A (en) | 2015-11-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105045983B (en) | A kind of bullet train axletree aging analysis method based on axle temperature data | |
| WO2019080367A1 (en) | Method for evaluating health status of mechanical device | |
| CN103544389B (en) | Autocrane method for diagnosing faults based on fault tree and fuzzy neural network | |
| CN107797537A (en) | A kind of prognostic and health management method applied to automatic production line | |
| CN107145675A (en) | Diagnosing fault of power transformer device and method based on BP neural network algorithm | |
| CN104318347A (en) | Power transmission line icing state assessment method based on information fusion of multiple sensors | |
| CN110580387B (en) | DC protection system mixed Weibull reliability evaluation method based on entropy weight method | |
| CN107490479A (en) | Bearing residual life Forecasting Methodology and device | |
| CN107776606B (en) | Fault detection method for shaft temperature monitoring system | |
| CN110827270B (en) | Composite insulator defect diagnosis method and system | |
| CN111721834A (en) | Cable partial discharge online monitoring defect identification method | |
| CN105572492A (en) | City rail train auxiliary inverter fault diagnosis device | |
| CN110533115B (en) | Quantitative evaluation method for transmission characteristics of track circuit based on variational modal decomposition | |
| CN102692450B (en) | Method for identifying state of shaped crack of metal drawing part based on fuzzy comprehensive evaluation | |
| CN111860899A (en) | Motor operation and maintenance method, device, electronic equipment and system | |
| CN114781657A (en) | Power equipment maintenance system and method based on artificial intelligence | |
| CN112699483B (en) | Airplane structure DFR analysis method based on damage equivalent conversion | |
| CN103020346B (en) | Test method for physical design similarity of circuit | |
| CN112561257A (en) | Bridge structure safety evaluation method and device based on big data | |
| CN109359791B (en) | Mechanical system degradation trend prediction method and system | |
| CN115935162B (en) | Stray current online evaluation working system | |
| CN115650006B (en) | Elevator safety monitoring and early warning method and system based on big data | |
| CN104143158A (en) | Method for monitoring reliability of power system element under condition of multiple meteorological factors | |
| CN114509506B (en) | Online crack evaluation method based on guided wave time spectrum difference and convolutional neural network set | |
| CN107255560B (en) | A kind of failure modes of high-speed train braking pipe and diagnostic method based on pressure data |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20240204 Address after: 518000 1002, Building A, Zhiyun Industrial Park, No. 13, Huaxing Road, Henglang Community, Longhua District, Shenzhen, Guangdong Province Patentee after: Shenzhen Wanzhida Technology Co.,Ltd. Country or region after: China Address before: 710048 No. 5 Jinhua South Road, Shaanxi, Xi'an Patentee before: XI'AN University OF TECHNOLOGY Country or region before: China |