CN102837711A - Infrared waveform based intelligent identification method for railway bearing - Google Patents
Infrared waveform based intelligent identification method for railway bearing Download PDFInfo
- Publication number
- CN102837711A CN102837711A CN2011101669334A CN201110166933A CN102837711A CN 102837711 A CN102837711 A CN 102837711A CN 2011101669334 A CN2011101669334 A CN 2011101669334A CN 201110166933 A CN201110166933 A CN 201110166933A CN 102837711 A CN102837711 A CN 102837711A
- Authority
- CN
- China
- Prior art keywords
- infrared
- normalization method
- squiggle
- temperature
- waveform
- 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
Images
Abstract
The invention relates to a method of identifying bearing failures, in particular to an infrared waveform based intelligent identification method for a railway bearing. The method is based on a dynamic standard waveform of a specific detecting spot, a vertical scan in the original method is substituted with a once horizontal scan for processing the infrared waveform, and therefore, the nature of distinguishing a normal infrared waveform from an abnormal infrared waveform is reflected. With the adoption of the technical scheme, the intelligent identification method based on waveform characteristics is standard, concise and good in effect, and the workload can be effectively reduced on the premise that the effect of the method is better or not worse than that of artificial identification.
Description
Technical field
The present invention relates to a kind of detection recognition methods of judging bearing fault, more particularly, be meant the method for the infrared waveform Intelligent Recognition of a kind of railway bearing.
Background technology
Bearing is the critical component of vehicle, and bearing fault is one of major failure source in the train operation.Because bearing life has certain discreteness; In order to prevent to fire the generation of cutting an accident; China railways has been set up infrared ray vehicle axle temperature intelligent detecting system; Operating rolling stock is carried out online bearing temperature monitoring, and system adopts dispersion to detect, networking is concentrated, the operating mode of the artificial warning of road bureau, and large tracts of land has effectively reduced the generation of firing an accident of cutting after promoting.
Infrared ray vehicle axle temperature intelligent detecting system approximately whenever sets up 1 unattended acquisition station to carry out axle temperature detection at a distance from about 30 kilometers along the railway; Site environment is complicated, abominable; Various factors such as vibration, impact, unexpected thermal source, sleet mist etc. all possibly influence the axle temperature detection quality; Cause unusual waveforms, other factors such as system design or defective mounting, power supply voltage instability etc. also can produce and detect unusually.Common waveform have unusually all kinds of burrs, sunlight disturb, fullly press, not full, be shifted, narrow down, broaden etc.
For solving interference problem, each manufacturer of railway axle temperature intelligent detecting system has adopted many measures at acquisition station, has obtained certain effect; But in the infrared networking concentrated alarm of road bureau, still have the abnormal data of larger amt; Strengthened infrared watch keeper's work capacity, in case the strong thermal warning that swashs is arranged, infrared watch keeper must be in very short time; Check the infrared ray waveform, make right judgement.Along with infrared watch keeper is responsible for the increase of equipment, abnormal data has disturbed normal work, and the intelligent identification of infrared waveform is extremely important.
Along with development of technology; China has carried out research of hot box integrated alarm and utilization; This moment, infrared axle temperature information was still important judgment rules, and its correctness is significant to the enforcement of integrated alarm; Be necessary to realize its intelligent identification Method, and then effectively realize the comprehensive auto alarm of hot box.
Original infrared axle temperature method for waveform identification mainly contains following several kinds:
A, difference and many (being generally less than 4) step method of finite difference; Calculate the difference of the adjacent multi-point of axle temperature curve; Judge whether that through the size of multistep difference and distribute (peak value number) waveform is unusual; This method generally is used to handle burr type unusual waveforms, and waveform rises the multistep difference in order to calculate, falling edge provides mark simultaneously;
B, area-method, the calculated characteristics position (as all, front portion, middle part, rear portion) area that comprises of temperature curve and relative size thereof concerns that to judge whether waveform unusual, this method generally is used to handle the sunlight interference waveform;
C, Furthest Neighbor calculate the distance of squiggle and reference waveform, judge whether bearing type reaches normally;
D, correlation method calculate the coefficient of correlation of squiggle and reference waveform, judge whether bearing type reaches normally;
E, other, like neural network method etc.
But disclosed method mainly has in the above-mentioned prior art: 1. be provided with to different unusual characteristics; 2. original algorithm is mainly according to the order of axle temperature data, the method more complicated of reflection data characteristics, as the difference that is primarily aimed at all kinds of burrs just is divided into a variety of; 3. calibrated curve is from all waveforms, to obtain, and specific aim is not strong, passes judgment on threshold and confirms difficulty; 4. original algorithm does not have effectively to embed the characteristic of normal waveform.
Summary of the invention
Based on the weak point of disclosed method in the above-mentioned prior art, goal of the invention of the present invention has been to provide a kind of intelligent identification Method based on wave character, and the method standard is succinct, and is respond well.
Another goal of the invention of the present invention in addition is the waveform intelligent identification Method that adopts the present invention to propose, can effect be superior to or prerequisite no less than artificial congnition under, effectively reduce work capacity.
In order to realize above-mentioned goal of the invention, the present invention adopts following technical scheme:
The method of the infrared waveform Intelligent Recognition of a kind of railway bearing comprises following step:
(1) preset reference normalization method temperature Y
0
The current infrared normalization method temperature curve and the said benchmark normalization method temperature Y of axle temperature transfinite when (2) certain bearing being passed through
0Relatively, obtain all and be not less than said benchmark normalization method temperature Y
0The position of point for measuring temperature;
(3) if all that obtain are not less than said benchmark normalization method temperature Y
0The point for measuring temperature position continuous, and do not comprise the beginning or end of infrared normalization method temperature curve, then carry out the operation of (4), otherwise directly be judged to unusual waveforms;
(4) said current infrared normalization method temperature curve is carried out fixed mode interpolation, obtain the standardization squiggle and the characteristic parameter thereof of this current infrared normalization method temperature curve;
(5) current standardization squiggle and characteristic parameter and measuring point dynamic standard squiggle and measuring point characteristic parameter are compared judge.
The invention described above adopt improve one's methods in generate said measuring point dynamic standard squiggle and measuring point characteristic parameter the standardize method basically identical of squiggle of step and above-mentioned formation, specifically comprise:
(1) preset reference normalization method temperature Y
0
(2) with certain the infrared normalization method temperature curve and the said benchmark normalization method temperature Y of certain measuring point
0Relatively, obtain all and be not less than said benchmark normalization method temperature Y
0The position of point for measuring temperature;
(3) if all that obtain are not less than said benchmark normalization method temperature Y
0The point for measuring temperature position continuous, and do not comprise the beginning or end of infrared normalization method temperature curve, then carry out the operation of (4);
(4) said certain infrared normalization method temperature curve is carried out fixed mode interpolation, obtain the standardization squiggle and the characteristic parameter thereof of this current infrared normalization method temperature curve;
(5) the said standardization squiggle to this measuring point specified requirements carries out the pointwise statistics, obtains measuring point dynamic standard squiggle;
(6) the said standardization squiggle characteristic parameter of this measuring point specified requirements is added up obtained the measuring point characteristic parameter.
Among the present invention, if sign point for measuring temperature position is that the corresponding normalization method temperature of abscissa X, point for measuring temperature is ordinate Y, above-described fixed mode interpolation method can specifically be expressed as:
(1) obtains the said relatively benchmark normalization method of infrared normalization method temperature curve temperature Y
0Just pass through point (X
0+, Y
0) and the negative point (X that passes through
0-, Y
0);
(2) to infrared normalization method temperature curve at a positive and negative abscissa [X that passes through
0+, X
0-] between carry out the equidistant interpolation of n point, generate the standardization squiggle, wherein the span of n is between 18-24.
The wave character parameter of extracting has standardization squiggle center (X
0++ X
0-)/2 and standardization squiggle width (X
0--X
0+)/2.
Carrying out current standardization squiggle and characteristic parameter and measuring point dynamic standard squiggle and measuring point characteristic parameter when relatively passing judgment on; Center of curve and these two characteristic parameters of curve width be to assess on the one hand, current standardization waveform and the maxim difference of the corresponding interpolation point of measuring point dynamic standard waveform and the coefficient of correlation between current standardization waveform and measuring point dynamic standard waveform also will be assessed on the other hand.
Need to prove; Described measuring point dynamic standard squiggle and characteristic parameter upgrade when reaching preset time threshold values or quantity threshold values and carry out with the same category of device reference waveform of infrared central store and the characteristic parameter comparison is confirmed, described current standardization squiggle and characteristic parameter all compare judge with the measuring point dynamic standard squiggle and the characteristic parameter of corresponding measuring point latest update affirmation.The comformability and the real-time of measuring point dynamic standard waveform can be guaranteed so on the one hand basically, the maintenance and the state assurance of equipment can also be monitored, be beneficial on the other hand the state of the art of measuring point.
Among the present invention, can be directed against the measuring point standard specifications waveform that special vehicle and bearing type generate more refinement, to realize the identification of more refinement.
In the process that realizes the method that the present invention is above-mentioned, preferably, described benchmark normalization method temperature Y
0Value is between 50-70.More preferably, described benchmark normalization method temperature Y
0Value is 60.
Through adopting above-mentioned technical scheme, the invention provides a kind of intelligent identification Method based on wave character, method is succinct, and is respond well.Waveform intelligent identification Method of the present invention in addition, can effect be superior to or prerequisite no less than artificial congnition under, effectively reduce work capacity.
Description of drawings
What show among Fig. 1 is the existing method flow diagram of the measurement railway bearing of available technology adopting;
What show among Fig. 2 is the diagram of circuit of the method for the measurement railway bearing that adopts among the present invention;
What show among Fig. 3 is the normal basically infrared normalization method temperature curve sample of certain measuring point in the embodiment of the invention;
That show among Fig. 4 is the infrared normalization method temperature curve sample relative datum normalization method temperature Y in the embodiment of the invention
0A positive and negative situation of passing through;
What show among Fig. 5 is to carry out the process that equidistant interpolation generates the standardization waveform between passing through a little positive and negative in the embodiment of the invention;
What show among Fig. 6 is the measuring point standard waveform parameter statistical graph in the embodiment of the invention;
What show among Fig. 7 is the pointwise statistics generation of the measuring point dynamic standard waveform in the embodiment of the invention;
That show among Fig. 8-Figure 12 is the sample figure that is judged as obviously unusual or unusual waveforms in the embodiment of the invention; Wherein long and short dash line (---*---) is 32 current infrared normalization method temperature curves; Basin type smooth curve is corresponding measuring point dynamic standard squiggle, irises wipe line (---о---) if any the 3rd and then is the standardization squiggle of current infrared normalization method temperature curve.
The specific embodiment
The invention reside in provides a kind of intelligent identification Method based on wave character, and the method standard is succinct, and is respond well.Waveform intelligent identification Method of the present invention, can effect be superior to or prerequisite no less than artificial congnition under, effectively reduce work capacity.
Carry out detailed explanation below in conjunction with the Figure of description specific embodiments of the invention.
What Fig. 1 showed is the infrared axle temperature method for waveform identification flow process (drawing from " train axle temperature detection system data processing algorithm and realization ") of available technology adopting, and algorithm flow comprises: spike differentiation, surge detection, tub curve are differentiated, waveform becomes fat differentiation, waveform moves differentiation and characteristic distance calculating etc.The algorithm emphasis is the pretreatment to various unusual waveforms, and the structure clear process of main program is the process of an order execution subroutine from top to bottom.
Can see with determination system of train wheel axles data processing algorithm block diagram from above analysis: disclosed method mainly has the prior art: 1. be provided with to different unusual characteristics; 2. original algorithm is mainly according to the order of axle temperature data, the method more complicated of reflection data characteristics, as the difference that is primarily aimed at all kinds of burrs just is divided into a variety of; 3. calibrated curve is added up from the waveform of the infrared measuring point of typical case and is obtained and apply to all measuring points, and specific aim is not strong, the threshold wide ranges; 4. original algorithm does not have effectively to embed the characteristic of normal waveform.
Fig. 2 is the flow process of the infrared waveform intelligent identification Method of railway bearing of the present invention, contrast prior art diagram of circuit (Fig. 1) can be seen: existing infrared method for waveform identification is based on the unusual waveforms characteristic, so the method that needs is many; Adopted unified calibrated curve in the existing method, and actual waveform and measuring point relation are bigger, make that the setting of threshold value is difficult in the prior art, can only provide a very wide scope; The present invention proposes the intelligent identification Method based on the normal wave character of measuring point, standard is succinct, and is respond well.
New mode is based on following normal waveform essential characteristic: the point for measuring temperature greater than certain suitable temperature value (or normalization method temperature value) is continuous and is positioned at the middle part; For same measuring point, the number of these points for measuring temperature changes in a small range; For same measuring point, the position of these points for measuring temperature changes in a small range; For same measuring point, the curve that these points for measuring temperature are formed is similar and close in value.
New mode is different from has 2 points the most at all in original method: the 1st, and normal waveform has solved threshold problem has been set to concrete measuring point; The 2nd, adopted once laterally " scanning " to substitute in original method repeatedly vertical " scanning " to infrared waveform, reflected that normal infrared waveform distinguishes the essence of unusual waveforms, the method standard, succinctly, comformability is strong.
New mode mainly contains two big links: the dynamic standard waveform based on measuring point generates; Pass judgment on according to the infrared waveform of the axle temperature transfinite of normal waveform model.
And concrete grammar of the present invention comprises following step:
(1) preset reference normalization method temperature Y
0
The current infrared normalization method temperature curve and the said benchmark normalization method temperature Y of axle temperature transfinite when (2) certain bearing being passed through
0Relatively, obtain all and be not less than said benchmark normalization method temperature Y
0The position of point for measuring temperature;
(3) if all that obtain are not less than said benchmark normalization method temperature Y
0The point for measuring temperature position continuous, and do not comprise the beginning or end of infrared normalization method temperature curve, then carry out the operation of (4), otherwise directly be judged to unusual waveforms;
(4) said current infrared normalization method temperature curve is carried out fixed mode interpolation, obtain the standardization squiggle and the characteristic parameter thereof of this current infrared normalization method temperature curve;
(5) current standardization squiggle and characteristic parameter and measuring point dynamic standard squiggle and measuring point characteristic parameter are compared judge.
Because the formation step of measuring point dynamic standard squiggle is with current basic identical by the formation step than squiggle, therefore this only casehistory form the step of measuring point dynamic standard squiggle.Fig. 3-Fig. 7 has illustrated each step of generation measuring point dynamic standard squiggle, below in conjunction with the explanation of the squiggle among Fig. 3-Fig. 7 concrete grammar that the present invention adopted.
1.1 certain measuring point axle temperature data is carried out preliminary differentiation, obtains the set of the basic normal waveform of this measuring point;
1.1.1 set benchmark normalization method temperature Y
0, benchmark normalization method temperature can be set between the 50-70;
1.1.2 32 axle temperature normalization method temperature curves that will record and benchmark normalization method temperature Y
0Compare, obtain all and be not less than benchmark normalization method temperature Y
0The position of point for measuring temperature;
1.1.3 if all are not less than benchmark normalization method temperature Y
0The position of point for measuring temperature is (spacing is 1) and do not comprise starting point and terminal point continuously, thinks that then these 32 axle temperature normalization method temperature curves are basic normal waveform.
1.2 statistics is extracted this measuring point ordinary wave shape parameter, ripple is wide, center, reference waveform etc.;
1.2.1 every 32 axle temperature curves in 1.1 are handled, are obtained the positive and negative point (X that passes through of curve relative datum normalization method temperature
0+, Y
0) and (X
0-, Y
0);
1.2.2 obtain width and center that this curve is not less than benchmark normalization method temperature, the center equals (X
0++ X
0-)/2, width equals (X
0--X
0+)/2;
1.2.3 to infrared normalization method temperature curve at a positive and negative abscissa [X that passes through
0+, X
0-] between carry out equidistantly interpolation of n point (like 20 points, contain positive and negative passing through a little), obtain the standardization waveform of this curve;
1.2.4 1.2.2 under this measuring point specified requirements is obtained the curve width and the center is added up, obtains corresponding dynamic benchmark;
1.2.5 the standardization waveform of the regular length that 1.2.3 under this measuring point specified requirements is obtained carries out the pointwise statistics, obtains the dynamic standard standardization waveform of certain measuring point.
1.3 can be directed against the measuring point standard specifications waveform that special vehicle and bearing type generate more refinement, this moment only need be with the data in 1.1 by vehicle and bearing type grouping;
1.4 measuring point standard specifications waveform that extracts and characteristic parameter are compared with infrared center same category of device standard specifications waveform and characteristic parameter, confirm correctness.
1.5 measuring point standard specifications waveform can carry out regularly as required, quantitative or Real-time and Dynamic generation.
Generation is that reference waveform and measuring point equipment have certain correlativity based on the reason of the dynamic standard standardization waveform of measuring point; If adopt unified reference waveform will make differing greatly between normal waveform and reference waveform; Be unfavorable for the setting of threshold value, be unfavorable for judge based on normal waveform model; A potential utilization is, through the contrast of certain measuring point dynamic standard waveform and other same category of device reference waveforms of infrared center, can judge the measuring point equipment state, instruct adjustment, the maintenance of measuring point equipment.
Further according to the infrared waveform intelligent identification Method of railway bearing of normal waveform model,, center wide and carry out the infrared waveform judge of axle temperature transfinite with relevant, the corresponding interpolation point maximum amplitude difference of measuring point dynamic standard waveform according to waveform continuity, ripple.
2.1 to the infrared waveform of axle temperature transfinite, the method by 1.1 confirms whether it is basic normal waveform;
2.2 if the step that the then similar 1.2 measuring point reference waveforms of basic normal waveform are produced extracts that ripple is wide, center and generate the standardization waveform of current waveform;
2.3 calculate current standard waveform and the coefficient of correlation of measuring point dynamic standard standardization waveform, the maxim of corresponding interpolation point difference in magnitude;
2.4 result according to 2.1,2.2,2.3---whether normal basically, ripple is wide, the maxim of center, coefficient of correlation and corresponding interpolation point difference in magnitude is passed judgment on the waveform correctness.
In addition, for technique effect of the present invention more specifically is described, and the technical scheme of clear and definite this method, collected Qun Dynasty's railway on November 6,2007 30 days to 2010 July in little, strong, swash thermal warning and process information compares assessment.
Total sample number 16228, wherein 223 strong, swash the thermal warning samples, in practice,, have 121 warnings to carry out blocking to stop and feed back through manual intervention (waveform artificial judgment, traffic control etc.).Below from strong swashing the thermal warning sample, blocking that the waveform intelligent appraisement result who stops sample constitutes, two aspects of evaluation result feedback statistics are assessed the waveform intelligent appraisement effect that this paper proposes.
Consider practice, critical parameter (center, ripple are wide, coefficient of correlation, pointwise difference maxim etc.) has adopted interval setting, has promptly increased and has passed judgment on uncertain type, and such waveform needs manual work to confirm again.According to each parameter distribution that statistics obtains, can confirm the interval of passing judgment on, as parameter 223 are swashed thermal warning samples, 121 and block and stop sample and carried out intelligent distinguishing, result such as table 1 by force:
By force sharp heat of table 1 and infrared manual work are blocked and are stopped sample Intelligent Recognition result contrast
Annotate: obviously unusually the finger waveform do not satisfy condition-discontinuous or comprise the waveform beginning or end greater than the point of benchmark normalization method temperature
Can see:
1, the unusual waveforms ratio in the infrared strong sharp thermal warning of Qun Dynasty's line is than higher, and in 223 samples, 50 obviously unusual, accounts for 22.42%; Unusual 29, account for 13.00%; Toatl proportion is about 35%.
2, present manual work confirms that screening is significant, swashs thermal warning to strong, and after artificial the judge, the obviously unusual ratio of waveform reduces to 8.26% from 22.42%, descends 63.14% relatively; The unusual waveforms ratio reduces to 8.26% from 13.00%, descends 36.45% relatively; Meanwhile, normal waveform ratio rises to 71.07% from 51.12%, rises 39.03% relatively; Unascertainable ratio drops to 12.40% from 12.56%, descends 1.27% relatively.
If the evaluation result of the infrared waveform intelligent identification Method of a kind of railway bearing that this patent proposes is reliable, will effectively reduce staff's work capacity so, the staff only needs unascertainable portion waveshape to be confirmed the ratio of minimizing is not less than 85%.
Whether the result who judges intelligent appraisement is reliable, need block the feedback information analysis of stopping to corresponding.Block at present and stop feedback and have 6 kinds: 1-blocks and stops the back and let pass, and 2-lets pass at the back of uncoupling, and the 3-bearing has fault after moving back and unloading, and 4-changes wheel and do not move back and unload, 5-bearing trouble free, 6-other 1 other 2; Because other situation information of 6-is indeterminate; The preceding 5 total situation of can mainly assessing (are annotated: change wheel and do not move back to unload and do not mean that bearing has fault; According to Ministry of Railways's ad hoc survey of implementing by the four directions, judge that at the scene bearing has the less than 20% of obvious hurt in the problematic axle).
Table 1 Intelligent Recognition result stops the feedback relationship table with blocking
Table 1 pairs 121 blocks and stops sample and carried out waveform recognition, does not consider other situation of 6-, and 13 result is to block to park row in the obviously unusual and unusual individual sample of 18 (20-2); 4 bearings that normal sample has comprised all affirmations move back and have unloaded the fault feedback; Its centre bearer has fault and changes wheel and do not move back and unload 52; Ratio is (4+48)/(86-5-2)=52/79=65.82%, is superior to artificial congnition result (4+3+48+2+9)/(121-5-2-1-1-1-1)=66/110=60.00%.
It is unusual or be recorded as unusually and change wheel and do not move back the waveform sample that unloads for obviously that Fig. 8-Figure 12 has listed all 5 Intelligent Recognition, can see that these waveforms obviously are problematic.
So, the waveform intelligent identification Method that adopts this method to propose, can effect be superior to or prerequisite no less than artificial congnition under, effectively reduce work capacity.
Protection scope of the present invention is not limited to disclosed specific embodiment in the above-mentioned specific embodiment, as long as but the combination of satisfying technical characterictic in the claim of the present invention just fallen within protection scope of the present invention.
Claims (9)
1. the method for the infrared waveform Intelligent Recognition of railway bearing is characterized in that, described method comprises following step:
(1) preset reference normalization method temperature Y
0
The current infrared normalization method temperature curve and the said benchmark normalization method temperature Y of axle temperature transfinite when (2) bearing being passed through
0Relatively, obtain all and be not less than said benchmark normalization method temperature Y
0The position of point for measuring temperature;
(3) if all that obtain are not less than said benchmark normalization method temperature Y
0The point for measuring temperature position continuous, and do not comprise the beginning or end of infrared normalization method temperature curve, then carry out the operation of (4), otherwise directly be judged to unusual waveforms;
(4) said current infrared normalization method temperature curve is carried out fixed mode interpolation, obtain the standardization squiggle and the characteristic parameter thereof of this current infrared normalization method temperature curve;
(5) the standardization squiggle of current normalization method temperature curve and characteristic parameter and measuring point dynamic standard squiggle and measuring point characteristic parameter are compared judge.
2. the method for the infrared waveform Intelligent Recognition of a kind of railway bearing according to claim 1 is characterized in that, the step that forms said measuring point dynamic standard squiggle comprises:
(1) preset reference normalization method temperature Y
0
(2) with the infrared normalization method temperature curve and the said benchmark normalization method temperature Y of measuring point
0Relatively, obtain all and be not less than said benchmark normalization method temperature Y
0The position of point for measuring temperature;
(3) if all that obtain are not less than said benchmark normalization method temperature Y
0The point for measuring temperature position continuous, and do not comprise the beginning or end of infrared normalization method temperature curve, then carry out the operation of (4);
(4) said certain infrared normalization method temperature curve is carried out fixed mode interpolation, obtain the standardization squiggle and the characteristic parameter thereof of this current infrared normalization method temperature curve;
(5) the said standardization squiggle to this measuring point specified requirements carries out the pointwise statistics, obtains measuring point dynamic standard squiggle;
(6) the said standardization squiggle characteristic parameter of this measuring point specified requirements is added up obtained the measuring point characteristic parameter.
3. the method for the infrared waveform Intelligent Recognition of a kind of railway bearing according to claim 1 and 2 is characterized in that, if sign point for measuring temperature position is that the corresponding normalization method temperature of abscissa X, point for measuring temperature is ordinate Y, the step of described fixed mode interpolation comprises:
(1) obtains the said relatively benchmark normalization method of infrared normalization method temperature curve temperature Y
0Just pass through point (X
0+, Y
0) and the negative point (X that passes through
0-, Y
0);
(2) to infrared normalization method temperature curve at a positive and negative abscissa [X that passes through
0+, X
0-] between carry out the equidistant interpolation of n point, generate the standardization squiggle.
4. the method for the infrared waveform Intelligent Recognition of a kind of railway bearing according to claim 3 is characterized in that, described to infrared normalization method temperature curve at a positive and negative abscissa [X that passes through
0+, X
0-] between carry out the equidistant interpolation of n point, the n value is between 18-24.
5. the method for the infrared waveform Intelligent Recognition of a kind of railway bearing according to claim 3 is characterized in that the characteristic parameter of described extraction comprises:
(1) standardization squiggle center (X
0++ X
0-)/2;
(2) standardization squiggle width (X
0--X
0+)/2.
6. the method for the infrared waveform Intelligent Recognition of a kind of railway bearing according to claim 1; It is characterized in that described current standardization squiggle compares to pass judgment on measuring point dynamic standard squiggle and comprises the maximum amplitude difference of corresponding interpolation point and the comparison of the coefficient of correlation between curve.
7. the method for the infrared waveform Intelligent Recognition of a kind of railway bearing according to claim 1 and 2; It is characterized in that; Described measuring point dynamic standard squiggle upgrades when reaching preset time threshold values or quantity threshold values and carries out with the same category of device reference waveform of infrared central store and the characteristic parameter comparison is confirmed, described current standardization squiggle and characteristic parameter all compare judge with the measuring point dynamic standard squiggle and the characteristic parameter of corresponding measuring point latest update affirmation.
8. the method for the infrared waveform Intelligent Recognition of a kind of railway bearing according to claim 1 and 2 is characterized in that, can generate the more accurate measuring point standard specifications waveform of parameter in order to the identification that realizes more refinement to different automobile types and bearing type.
9. the method for the infrared waveform Intelligent Recognition of a kind of railway bearing according to claim 8 is characterized in that, described benchmark normalization method temperature Y
0Value is between 50-70.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110166933.4A CN102837711B (en) | 2011-06-21 | 2011-06-21 | Infrared waveform based intelligent identification method for railway bearing |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110166933.4A CN102837711B (en) | 2011-06-21 | 2011-06-21 | Infrared waveform based intelligent identification method for railway bearing |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102837711A true CN102837711A (en) | 2012-12-26 |
| CN102837711B CN102837711B (en) | 2015-04-08 |
Family
ID=47365508
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201110166933.4A Active CN102837711B (en) | 2011-06-21 | 2011-06-21 | Infrared waveform based intelligent identification method for railway bearing |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102837711B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104679982A (en) * | 2014-12-29 | 2015-06-03 | 北京康拓红外技术股份有限公司 | THDS system detecting result effectiveness identification method based on feature value detection |
| CN105466927A (en) * | 2014-07-02 | 2016-04-06 | 深圳迈瑞生物医疗电子股份有限公司 | Identification method, correction method and alarm method for turbidimetry abnormal reaction curve |
| CN109812334A (en) * | 2019-03-27 | 2019-05-28 | 潍柴动力股份有限公司 | A kind of main bearing of engine fault monitoring method and device |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US2856540A (en) * | 1954-12-30 | 1958-10-14 | Gen Electric | Infrared detection apparatus |
| US4313583A (en) * | 1980-03-31 | 1982-02-02 | Servo Corporation Of America | Railroad car wheel bearing heat signal processing circuit |
| US4659043A (en) * | 1981-10-05 | 1987-04-21 | Servo Corporation Of America | Railroad hot box detector |
| 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 |
| WO2008091444A1 (en) * | 2007-01-26 | 2008-07-31 | General Electric Company | Method and apparatus for monitoring bearings |
| CN101716945A (en) * | 2010-01-14 | 2010-06-02 | 广州科易光电技术有限公司 | Railway locomotive axle infrared thermal image monitoring method and system |
-
2011
- 2011-06-21 CN CN201110166933.4A patent/CN102837711B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US2856540A (en) * | 1954-12-30 | 1958-10-14 | Gen Electric | Infrared detection apparatus |
| US4313583A (en) * | 1980-03-31 | 1982-02-02 | Servo Corporation Of America | Railroad car wheel bearing heat signal processing circuit |
| US4659043A (en) * | 1981-10-05 | 1987-04-21 | Servo Corporation Of America | Railroad hot box detector |
| WO2008091444A1 (en) * | 2007-01-26 | 2008-07-31 | General Electric Company | Method and apparatus for monitoring bearings |
| 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 |
| CN101716945A (en) * | 2010-01-14 | 2010-06-02 | 广州科易光电技术有限公司 | Railway locomotive axle infrared thermal image monitoring method and system |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105466927A (en) * | 2014-07-02 | 2016-04-06 | 深圳迈瑞生物医疗电子股份有限公司 | Identification method, correction method and alarm method for turbidimetry abnormal reaction curve |
| CN105466927B (en) * | 2014-07-02 | 2020-04-10 | 深圳迈瑞生物医疗电子股份有限公司 | Method for identifying, correcting and alarming abnormal reaction curve of turbidimetry |
| CN104679982A (en) * | 2014-12-29 | 2015-06-03 | 北京康拓红外技术股份有限公司 | THDS system detecting result effectiveness identification method based on feature value detection |
| CN104679982B (en) * | 2014-12-29 | 2017-08-08 | 北京康拓红外技术股份有限公司 | A kind of recognition methods of the THDS system result of detection validity of feature based value detection |
| CN109812334A (en) * | 2019-03-27 | 2019-05-28 | 潍柴动力股份有限公司 | A kind of main bearing of engine fault monitoring method and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102837711B (en) | 2015-04-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105652154A (en) | Safety monitoring analysis system for contact net running state | |
| CN102628814B (en) | Automatic detection method of steel rail light band abnormity based on digital image processing | |
| CN103034850B (en) | Trouble of moving freight car detection system (TFDS) block key loss fault automatic identification method | |
| CN108197817A (en) | A kind of method of the non-intrusion type load transient state monitoring based on big data | |
| CN106056821A (en) | Power-transmission-line foreign-matter-invasion intelligent-video on-line monitoring evaluation system | |
| CN102633173A (en) | System and method for monitoring operation state of elevator car | |
| CN108731953B (en) | Online detection method for polygonal fault of train wheel set | |
| CN102837711A (en) | Infrared waveform based intelligent identification method for railway bearing | |
| CN104820168A (en) | Lightning stroke fault determination method based on waveform difference degree and lightning stroke fault sample database | |
| CN105911424A (en) | Identification method based on fault indicator false alarm signal | |
| CN105354827A (en) | Method and system for intelligently identifying clamp nut shedding in catenary image | |
| CN104677997A (en) | Transformer oil chromatography online monitoring differential early warning method | |
| CN104805742B (en) | Railway switch state dynamic testing method and device | |
| CN114386451A (en) | Contact net dropper fault diagnosis and alarm method based on sensor information perception | |
| CN116227752B (en) | Park facility management system based on Internet of things | |
| CN105241679A (en) | Method for detecting hidden fault of motor train unit | |
| CN113177646A (en) | Power distribution equipment online monitoring method and system based on self-adaptive edge proxy | |
| CN114878796A (en) | Assessment monitoring platform based on road maintenance | |
| CN109298700B (en) | Method and system for judging abnormal change of operation parameters of thermal power generating unit in real time | |
| CN106325258B (en) | Relay protection device state evaluation method based on online monitoring information | |
| CN103217590A (en) | Method automatically acquiring atmospheric electric field thunder and lightning early-warning characteristic parameter threshold values | |
| CN115358278B (en) | State quantity distributed monitoring method and system for electricity collector | |
| CN112285111A (en) | Pantograph front carbon sliding plate defect detection method, device, system and medium | |
| CN109190924B (en) | Video number plate data quality analysis method | |
| CN103487275B (en) | Based on coal mine equipment state recognition and the method for early warning of two-dimensional projections |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant |