CN101149387A - Method for high precision non-contact measurement for rotation speed - Google Patents
Method for high precision non-contact measurement for rotation speed Download PDFInfo
- Publication number
- CN101149387A CN101149387A CNA2007100562884A CN200710056288A CN101149387A CN 101149387 A CN101149387 A CN 101149387A CN A2007100562884 A CNA2007100562884 A CN A2007100562884A CN 200710056288 A CN200710056288 A CN 200710056288A CN 101149387 A CN101149387 A CN 101149387A
- Authority
- CN
- China
- Prior art keywords
- signal
- laser
- high precision
- small echo
- displacement sensor
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Abstract
A kind of high precision non-contact rotate speed measurement method which relates to a new method in the test field, especially relates to a new high precision non-contact rotate speed measurement method to overcome the error of the rotate speed measurement in the exist measurement method. The method first uses the laser displacement sensor collect the signal, the processes the collected signal in the computer by measurement software, the said laser displacement sensor use high precision laser CCD displacement sensor, uses the laser displacement sensor scan the surface of the tested whirlabout to collect the distance ri,i=1,2,3......,that a series of laser CCD reference zero point to the surface of the object, i is the serial number of each scanning point, doing date process of these ri signal, distills the main frequency of the change signal which is also named the rotation frequency according to the periodicity change of the signal to get the rotation speed.
Description
Technical field
The present invention relates to a kind of new method of field tests, particularly relates to a kind of new method of novel high precision noncontact tachometric survey.
Background technology
Tachometric survey is very important in processing and manufacturing is measured, and rotatation speed test method has a lot, tests the speed as (1) Permanent Magnet DC Tachogeneator, and it is highly sensitive, and linear error is little, is its intrinsic defective but brush configuration is arranged; (2) photoelectric velocity measurement is a kind of more general speed-measuring method, and precision was low when the normal optical method for electrically was surveyed low speed, with photoelectric code disk rate accuracy height, but costs an arm and a leg; (3) indirect velocimetry generates tach signal by rotor-position signal, has saved special electromagnetism or photoelectric velocity measurement parts, and the utilization integrated circuit is realized testing the speed indirectly, and conveniently, simply, velocimetry mainly is used in miniature motor indirectly; (4) brushless DC tachogenerator tests the speed, as Hall brushless DC velocity measuring and the incorporate square wave brushless DC velocity measuring of motor etc., and modern design, simple and compact for structure, volume is little, better performances; (5) have rotor position detector class motor, (6) induction motor speed-measuring method, adopt photoelectric encoder (Pulse Light Generator, PLG) detect motor speed, (7) the contactless infrared electro tachometric survey of on-line operation, (8) based on the measurement of rotating speed method of speed probe, (9) micrometering technology and ccd image sensor integration imaging technique are used for the tachometric survey of microelectromechanical systems (MEMS), (10) realize tachometric survey with digital torque speed sensors.
Whether contact with the equipment of testing the speed according to tested rotary body, be divided into contact type measurement and non-contact measurement.Tradition tests the speed and generally adopts contact type measurement, and it is big to occur tested rotary body energy loss in measurement easily, the error that the mechanical wear and the usually generation of skidding are difficult to eliminate.Shortcomings such as measuring equipment mobility and flexibility difference.
Summary of the invention
The objective of the invention is to overcome the aforesaid error that in tachometric survey, exists at present,, provide a kind of high precision non-contact measurement for rotation speed method to satisfy the needs of processing manufacturing industry.
The technical scheme that realizes above-mentioned purpose of the present invention is:
At first use the laser displacement sensor acquired signal, then the signal of gathering is carried out data processing with Survey Software on computers, said laser displacement sensor adopts high-precision laser CCD displacement transducer, scan tested rotating object surface with laser displacement sensor, collect a series of laser and CCD reference zero to body surface apart from r
i, i=1,2,3......, i are the sequence number of each analyzing spot, to these r
iSignal advances data processing, according to the cyclical variation of signal, therefrom extracts the dominant frequency of variable signal, i.e. the rotational frequency of testee, and then try to achieve rotating speed.
Said laser and CCD displacement transducer adopts the laser triangulation principle, precision≤0.6 μ m, and valid analysing range is ± 40mm that the signals collecting minimum interval is 20 μ s.
Said laser and CCD displacement transducer sample frequency adopts 0.5 μ s and 1.0 μ s, scanning cross-section sampling number weekly: 500 point ~ 12000 point, measurand diameter range: 5mm～2000mm measures rotating speed: 5 rev/mins～1500 rev/mins.
Said data processing method is carried out the separation and the rejecting of error for adopting small echo multifrequency analytical algorithm or small echo-principal component analysis algorithm.
Said computer measurement software comprises the control module and the real time data processing calculating rotating speed module of carrying out data acquisition based on the USB communication port, wherein comprises small echo multifrequency analytical algorithm or small echo-principal component analysis algorithm in the data processing module.
Concrete measuring method provided by the invention is to utilize high-precision laser CCD displacement transducer scanning rotating object surface (as shown in Figure 1), collect a series of laser and CCD reference zero to body surface apart from r
i(i=1,2,3......), i is the sequence number of each analyzing spot.Get first detected value r in the scanning cross-section
0As initial value, this cross section difference of other point measurement values with respect to first is turning radius difference Δ R
i:
ΔR
i＝r
i-r
0(i＝1，2........n) (1)
Δ R
iBe the variable quantity of the turning radius, its size has reflected the deviation from circular from of testee, is cyclical variation.The precision of laser and CCD displacement transducer is 0.6 μ m, and valid analysing range is ± 40mm that the signals collecting minimum interval is 20 μ s, can detect testee surface subtle change or small off-centre.Object surface shape just can be detected when sampled point quantity is abundant.
After the raw data that collects is carried out the data pre-service, carry out separation and rejecting that small echo multifrequency analytical algorithm or small echo-principal component analysis (PCA) algorithm carries out error again, signal after the processing that obtains has periodically variable characteristics, therefrom extracting the dominant frequency of variable signal, is exactly the rotational frequency of testee.Can record rotating speed thus.
The analysis of small echo multifrequency is the way of thinking that adopts signal analysis, and utilization wavelet multiresolution analysis theory selects signal decomposition that suitable wavelet basis function will contain harmonic wave to become the signal of different frequency, constructs a wave filter according to actual needs.If f is the signal that records, with multiresolution analysis low frequency part is carried out some grades of decomposition, many details effect is good more more to decompose level in theory, gets final product but reach requirement in actual applications.Select the suitable wavelet decomposition exponent number data-signal f that scanning obtains to laser and CCD to decompose according to the feature of the measurand of reality and the objective condition of test among the present invention, and rejecting high-frequency interferencing signal, be reconstructed the signal after the acquisition error separating then.This signal can carry out the calculating of rotating speed.
The thought of wavelet packet is the detail signal that obtains to be separated on each rank also further decompose, and promptly the high-frequency signal to details also further decomposes, and considers this high-frequency information in restructuring procedure.Though most of noise and undesired signal are at HFS, all high-frequency signals can not be rejected without exception, might lose the Useful Information of original signal like this.So the thought of wavelet packet and arithmetic accuracy all more go a step further on the basis of simple small echo.
But each rank is decomposed and is obtained to such an extent that HFS and low frequency part all may contain undesired signal in the algorithm of wavelet packet, how selecting suitable standard to extract that wherein useful information is reconstructed is a complicated problems, generally all is to determine according to real data and test.And the computational algorithm complexity of wavelet packet is bigger, and speed improves difficulty.The present invention attempts addressing this problem in conjunction with the thought of principal component analysis.
The main thought of small echo--main composition is, all uses the thought of principal component analysis each rank low frequency signal that obtains from the analysis of small echo multifrequency and the high-frequency signal, all carries out the extraction of main composition.Regulate the quantity of information that this signal is contributed by the number of regulating main composition in reconstruct during reconstruct, pairing approximation signal and own interested signal can be given more main composition number, give less main composition number for the signal that needs suppress, even given 0 main composition, promptly this signal is all rejected.
Among the present invention acquired signal f is carried out small echo--during main composition error separating, the detail signal of every layer of wavelet decomposition is carried out in the leaching process of main composition factorial analysis and adopt kaiser method of inspection (Kaiser rule thumb) to select to carry out, promptly accumulate contribution rate decides main composition greater than 0.85 method choice.All select the main composition of certain number to extract calculating to all detail signals, a last approximate signal and reconstruction signal in the analysis, test findings shows better than using common small echo directly to reject the result of high frequency and reconstruct.Signal message amount after rejecting error that this method obtains is at last disturbed can reach 99.9736% of the signal message amount that common wavelet method obtains of using.Promptly obtain good computing velocity in the stick signal quantity of information to greatest extent, and reducing the complexity of calculating.
According to the surface quality situation of actual measured workpiece and the objective environment of measurement, carried out the analysis of use small echo multifrequency respectively and carried out error separating and use small echo--the test of main composition error separating among the present invention.The result shows the big measured workpiece of surface roughness of poor quality, uses the not obviously difference on computational accuracy of these two kinds of methods, and the exponent number of the wavelet decomposition of use also is identical.And to the little workpiece of the good roughness of high-precision surface quality, result calculated has bigger deviation when using the analysis of small echo multifrequency to carry out the method for error separating, and must use more wavelet decomposition exponent number.And the employing small echo--the result of main composition error separating algorithm not only as a result high the but also wavelet decomposition exponent number that use of precision significantly be less than the former.So small echo--main composition error separating algorithm has wider usable range, and has good precision and computing velocity.
Description of drawings
Fig. 1 is the tachometric survey schematic diagram.Laser and CCD scanning surface of the work carries out tachometric survey.0 is the rotation center of measured workpiece, and the A point is the reference zero of laser and CCD displacement transducer, P
i(Δ
i, φ
i) (i=1,2,3......n is the sequence number of the point that scans on each cross section) for being scanned some r on the surface of the work
iBe the laser and CCD reading of this point, φ
iIt is this angle that turns over respect to scanning cross-section starting point workpiece.
Fig. 2 is rotating speed and wavelet decomposition progression relation;
Fig. 3 is rotating speed and small echo master composition decomposed class relation;
Fig. 4 is rotating speed and wavelet decomposition progression relation;
Fig. 5 separates rotating speed and wavelet decomposition progression relation for not using multiple error;
Fig. 6 is rotating speed and small echo master composition decomposed class relation;
Fig. 7 is raw data wavelet decomposition progression and rotation speed relation;
Fig. 8 is principal component analysis algorithm decomposed class and rotation speed relation.
Embodiment
As shown in Figure 1, utilize high-precision laser CCD displacement transducer scanning rotating object surface, can access a series of laser and CCD reference zero to the body surface of scanning apart from r
i(i=1,2,3......), by r
iSequence constitutes signal f={r
i, i=1,2,3......n}.Get first detected value r in the scanning cross-section
0As initial value, this cross section difference of other point measurement values with respect to first is turning radius difference Δ R
i:
ΔR
i＝r
i-r
0(i＝1，2........n)
Δ R
iBe the variable quantity of the turning radius, its size has reflected the circularity of testee, is cyclical variation.Object surface shape just can be detected when sampled point quantity is abundant.Use common small echo to carry out error separating or small echo after the original signal f pre-service that collects--after the main composition error separating, signal all is cyclical variation, therefrom extracts the dominant frequency of variable signal, is exactly the rotational frequency of testee.
Embodiment 1
Application in the tachometric survey of high precision stepper motor.Measuring object is selected Korea S AUTONICS A16K-569 type high precision stepper motor for use, and minimum step angle is 0.009 degree.Keyway is arranged in the rotating shaft,, directly just can realize error separating in this case with small echo multifrequency error separating algorithm so measuring-signal has clearly periodically.
With rotating speed is 90 rev/mins data instance, and it decomposes the back with common small echo multifrequency analytical algorithm and rejects high-frequency signal, extracts rotating speed with reconstruct data then, the relation of rotating speed and wavelet decomposition exponent number as shown in Figure 2:
The rotating speed that extracts from single order wavelet decomposition reconstruct data just converges to 90 rev/mins as can be seen from Figure 2, and same The data is carried out error separating based on the algorithm of small echo-principal component analysis, the result who extracts rotating speed with reconstruct data as shown in Figure 3:
Both results do not have difference as can be seen.Reason mainly is in the tested stepper motor rotating shaft keyway to be arranged, periodically clearly.The result calculated of small echo-principal component analysis algorithm and common small echo multifrequency analytical algorithm is identical in this case, can both obtain satisfied measuring accuracy.
Application in the tachometric survey of large-scale high-precision filmatic bearing.To the very low workpiece of the such surface roughness value of filmatic bearing, and the lathe rotating speed is very accurate, and spindle rotation accuracy is very high, must use the multiple error separation algorithm.What measure in the experiment is the outside of bearing, does not have oil groove, thus used first order difference error separating algorithm, 53H error separating algorithm, 25 point data smoothing algorithms, and then carry out error separating with db3 small echo multifrequency analytical algorithm.As can be seen from Figure 4 carry out parasitic error and separate later data through three grades of error separating algorithms, carry out ten rank wavelet decomposition after reconstruct data extract stabilization of speed at 5.0093 rev/mins, it is 5 rev/mins that grinding machine is set rotating speed, relative error is 0.17%.
As can be seen from Figure 5, if do not use the multiple error separation algorithm, wavelet decomposition progression is very high in the time of then can having does not restrain yet, convergent series instability sometimes, before convergence and fluctuation big.Majority is to converge to 0 always, causes determining decomposed class, and result of calculation is very unstable.
As can be seen from Figure 6 for the measurand of same precision, carry out parasitic error through three grades of error separating algorithms and separate later data, carry out meticulous error separating with small echo-principal component analysis algorithm, extract rotating speed with the reconstruct of single order decomposition data and just obtained result accurately, rotating speed converges on 5.0093 rev/mins, and its speed of convergence is more effective than small echo multifrequency analytical algorithm.
Application in the engine lathe tachometric survey.CA6140 engine lathe rotating speed is measured, set 10 rev/mins of rotating speeds, the small echo low frequency decomposes 7 rank and restrains later on as can be seen in Fig. 7, is stabilized to 10.0334 rev/mins of rotating speeds.If to data filtering gross error, then the db3 wavelet filtering just converges to 10.0334 since first rank, if test findings explanation measured workpiece roughness is not very high, data to be carried out being easy to measure rotating speed after gross error is handled, it is just the same to measure 10 groups of data computation results in the test.
Result such as Fig. 8 for same The data small echo-principal component analysis algorithm computation, as can be seen when measurand precision when not being very high, to not passing through any pretreated data, directly adopt small echo-principal component analysis algorithm to carry out error separating, only just obtained accurate result, illustrated that its speed of convergence and precision all are higher than small echo multifrequency analytical algorithm with the single order decomposition.
Claims (5)
1. high precision noncontact rotating speed measurement method, it is characterized in that at first using the laser displacement sensor acquired signal, then the signal of gathering is carried out data processing with Survey Software on computers, said laser displacement sensor adopts high-precision laser CCD displacement transducer, scan tested rotating object surface with laser displacement sensor, collect a series of laser and CCD reference zero to body surface apart from r
i, i=1,2,3......, i are the sequence number of each analyzing spot, to these r
iSignal advances data processing, according to the cyclical variation of signal, therefrom extracts the dominant frequency of variable signal, i.e. the rotational frequency of testee, and then try to achieve rotating speed.
2. the method for high precision non-contact measurement for rotation speed according to claim 1, it is characterized in that said laser and CCD displacement transducer adopts the laser triangulation principle, precision≤0.6 μ m, valid analysing range are ± 40mm that the signals collecting minimum interval is 20 μ s.
3. the method for high precision non-contact measurement for rotation speed according to claim 1, it is characterized in that said laser and CCD displacement transducer sample frequency adopts 0.5 μ s and 1.0 μ s, scanning cross-section sampling number weekly: 500 point ~ 12000 point, measurand diameter range: 5mm～2000mm measures rotating speed: 5 rev/mins～1500 rev/mins.
4. the method for high precision non-contact measurement for rotation speed according to claim 1 is characterized in that said data processing method is to adopt small echo multifrequency analytical algorithm or small echo-principal component analysis algorithm to carry out the separation and the rejecting of error.
5. the method for high precision non-contact measurement for rotation speed according to claim 1, it is characterized in that said computer measurement software comprises the control module and the real time data processing calculating rotating speed module of carrying out data acquisition based on the USB communication port, wherein comprises small echo multifrequency analytical algorithm or small echo-principal component analysis algorithm in the data processing module.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNA2007100562884A CN101149387A (en) | 2007-11-08 | 2007-11-08 | Method for high precision non-contact measurement for rotation speed |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNA2007100562884A CN101149387A (en) | 2007-11-08 | 2007-11-08 | Method for high precision non-contact measurement for rotation speed |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN101149387A true CN101149387A (en) | 2008-03-26 |
Family
ID=39250014
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNA2007100562884A Pending CN101149387A (en) | 2007-11-08 | 2007-11-08 | Method for high precision non-contact measurement for rotation speed |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101149387A (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101876666A (en) * | 2010-05-21 | 2010-11-03 | 中钢集团天澄环保科技股份有限公司 | Non-contact type acceleration measurement method and device of filter-bag |
| CN102785169A (en) * | 2012-08-03 | 2012-11-21 | 东风康明斯发动机有限公司 | Double-speed-measuring safety-protecting method and device of numerically-controlled cylindrical grinder |
| CN105301276A (en) * | 2015-11-19 | 2016-02-03 | 南宁学院 | Rotation-speed measurement system of low rotation-speed electric motor |
| CN103712561B (en) * | 2014-01-20 | 2017-10-13 | 上海博泽电机有限公司 | Cooling unit method of testing and device with bounce and measurement of rotating speed function |
| CN107505475A (en) * | 2017-10-13 | 2017-12-22 | 河海大学 | A kind of measurement apparatus and method of rotating shaft transient speed and throw based on laser |
| CN109085374A (en) * | 2018-07-27 | 2018-12-25 | 江苏科技大学 | The multiple spot speed measuring device and its speed-measuring method for slow-speed of revolution system based on kinect |
| CN109782011A (en) * | 2019-02-18 | 2019-05-21 | 江苏大学 | A kind of rotation-speed measuring device and method of sea water desalination high-pressure pump and turbine all-in-one machine |
| CN110702936A (en) * | 2019-10-14 | 2020-01-17 | 深圳东方锅炉控制有限公司 | Device and method for measuring rotating speed of air preheater based on laser image recognition |
| CN111473762A (en) * | 2019-01-24 | 2020-07-31 | 长春工业大学 | Method for acquiring multipoint data with uniform excircle profile of high-speed rotor |
-
2007
- 2007-11-08 CN CNA2007100562884A patent/CN101149387A/en active Pending
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101876666A (en) * | 2010-05-21 | 2010-11-03 | 中钢集团天澄环保科技股份有限公司 | Non-contact type acceleration measurement method and device of filter-bag |
| CN102785169A (en) * | 2012-08-03 | 2012-11-21 | 东风康明斯发动机有限公司 | Double-speed-measuring safety-protecting method and device of numerically-controlled cylindrical grinder |
| CN103712561B (en) * | 2014-01-20 | 2017-10-13 | 上海博泽电机有限公司 | Cooling unit method of testing and device with bounce and measurement of rotating speed function |
| CN105301276A (en) * | 2015-11-19 | 2016-02-03 | 南宁学院 | Rotation-speed measurement system of low rotation-speed electric motor |
| CN107505475A (en) * | 2017-10-13 | 2017-12-22 | 河海大学 | A kind of measurement apparatus and method of rotating shaft transient speed and throw based on laser |
| CN109085374A (en) * | 2018-07-27 | 2018-12-25 | 江苏科技大学 | The multiple spot speed measuring device and its speed-measuring method for slow-speed of revolution system based on kinect |
| CN109085374B (en) * | 2018-07-27 | 2020-06-16 | 江苏科技大学 | Kinect-based multi-point speed measuring device for low-rotating-speed system and speed measuring method thereof |
| CN111473762A (en) * | 2019-01-24 | 2020-07-31 | 长春工业大学 | Method for acquiring multipoint data with uniform excircle profile of high-speed rotor |
| CN111473762B (en) * | 2019-01-24 | 2022-05-17 | 长春工业大学 | High-speed rotor excircle contour uniform multi-point data acquisition method |
| CN109782011A (en) * | 2019-02-18 | 2019-05-21 | 江苏大学 | A kind of rotation-speed measuring device and method of sea water desalination high-pressure pump and turbine all-in-one machine |
| CN110702936A (en) * | 2019-10-14 | 2020-01-17 | 深圳东方锅炉控制有限公司 | Device and method for measuring rotating speed of air preheater based on laser image recognition |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101149387A (en) | Method for high precision non-contact measurement for rotation speed | |
| Marsh | Precision spindle metrology | |
| CN101762253B (en) | System and method for multi-sensor-based on-line multi-dimension measurement of special-shaped shaft-type workpieces | |
| CN104482849B (en) | Testing system and testing method for dynamic rotation precision of main shaft | |
| CN105234745B (en) | Machine tool spindle thermal error, deviation from circular from are separated and processing method with turn error | |
| CN109947047B (en) | Electric spindle unbalance fault diagnosis method | |
| CN106168464A (en) | A kind of main shaft dynamic rotation method for testing precision based on machine vision | |
| CN110514164B (en) | Based on synchronous motion formula circularity error separator | |
| EP1549960A1 (en) | Speed sensing method and apparatus | |
| Li et al. | Online chatter detection in milling process based on VMD and multiscale entropy | |
| CN108629864A (en) | A kind of electro spindle radial accuracy characterizing method and its system based on vibration | |
| Noguchi et al. | Evaluation method to determine radial accuracy of high-precision rotating spindle units | |
| CN111259765A (en) | Order analysis method based on numerical control machine tool spindle current signal | |
| CN110118582A (en) | A kind of rotating machinery fault diagnosis method and system | |
| CN108593955A (en) | For moment rotatation speed test method in the case of cyclic irregularity | |
| CN1156695C (en) | System based on array-type flexible electric eddy sensor for monitoring gap between spherical layers | |
| Orton et al. | Automatic self-calibration of an incremental motion encoder | |
| CN107869949A (en) | Axial displacement detection method, device and shaft position sensor | |
| JP2011185675A (en) | Tolerance detection method and tolerance detection device for shape measuring apparatus | |
| Cheng et al. | An effective coaxiality measurement for twist drill based on line structured light sensor | |
| CN201532393U (en) | Laser rotating speed measuring instrument | |
| CN113758695B (en) | System and method for realizing rotation axis order analysis by adopting visual composite position-sensing stripes | |
| CN103743330B (en) | Comb-tooth-type cylindrical capacitive sensor based on mutual capacitance measuring principle | |
| US20050016279A1 (en) | Measurement of motions of rotating shafts using non-vibrating contact potential difference sensor | |
| Zeng et al. | A method combining order tracking and fuzzy c-means for diesel engine fault detection and isolation |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication |