CN103042436B - Spindle turning error source tracing method based on shaft center orbit manifold learning - Google Patents
Spindle turning error source tracing method based on shaft center orbit manifold learning Download PDFInfo
- Publication number
- CN103042436B CN103042436B CN201310021815.3A CN201310021815A CN103042436B CN 103042436 B CN103042436 B CN 103042436B CN 201310021815 A CN201310021815 A CN 201310021815A CN 103042436 B CN103042436 B CN 103042436B
- Authority
- CN
- China
- Prior art keywords
- spindle
- main shaft
- orbit
- shaft center
- vibration signal
- 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
Abstract
The invention relates to a spindle turning error source tracing method based on shaft center orbit manifold learning. The method includes the following step that (1) two electrical vortex sensors are arranged on the periphery of a spindle at intervals and used for collecting spindle vibration signals; (2) the detected spindle vibration signals are processed to judge an operation state of the spindle; (3) the spindle vibration signals intersect at one point on the same plane, and a shaft center orbit is obtained after continuous sampling; (4) error separation is conducted on a spindle center orbit to obtain spindle actual rotation precision A; (5) a mapping function atlas data base Q:{f(i)=Qij|A} is obtained according to the spindle actual rotation precision A and a manifold sensitive characteristic Qij; and (6) if the spindle actual rotation precision A>=etaE, eta=0.8-1, the mapping function atlas data base Q is called, source tracing of spindle rotation errors is conducted, and corresponding faults are maintained; and if the spindle actual rotation precision A>=etaE, eta=0.6-0.8, source tracing analysis monitoring is conducted on the spindle rotation errors, wherein E is spindle rotation precision of a machine tool leaving a factory.
Description
Technical field
The present invention relates to a kind of electromechanical equipment spindle rotation error source tracing method, particularly about a kind of spindle rotation error source tracing method based on orbit of shaft center manifold learning.
Background technology
Along with the development of the emerge science technology such as atomic energy, space technology, microelectronics, information technology and bioengineering, more and more higher to the requirement of machining accuracy, from millimeter to micron, sub-micron, develop into nanometer level now, and stride forward towards atomic lattice size (Ya Na meter) level gradually, be referred to as Ultra-precision Turning.Precision machine tool realizes precision machined primary basic condition.Current high speed Ultra-precision CNC Machine becomes the manufacturing crucial production equipment of modernization, improves the reliability of high speed Ultra-precision CNC Machine precision in processing running, stability and maintainability, more and more important to enterprise competitiveness.Survey of product life prediction and safe military service basic theory are the important research contents in " 12 " Country science and technology plan advanced manufacturing technology field, and diagnosing faults of numerical control machine and early warning technology are one of core technologies ensureing lathe reliability service, improve lathe military service performance.High-grade Ultra-precision CNC Machine due to complex structure, transmission link more, causing trouble can not accurately be located, and overhauling blindly can cause installation accuracy error, lathe military service hydraulic performance decline and reliability to reduce.Therefore Real-Time Monitoring, diagnosis and early warning are carried out to the duty of numerical control equipment extremely important.
The quality of ultra-precision machine tool, depend on the quality of its critical component, spindle unit is the core ensureing ultra-precision machine tool machining accuracy, also be one of position of the most easily losing efficacy, the quality of its dynamic property all has a great impact the cutting vibration resistance of lathe, machining accuracy and surface roughness, is the key factor of condition number controlled machine machining accuracy and service efficiency.Experiment shows: the deviation from circular from of precision turning about has 30% ~ 70% to be because the turn error of main shaft causes, and the precision of processing is higher, and shared ratio is larger.Orbit of shaft center during axis system revolution contains a large amount of information relevant with rotary part duty with axis system state of the art, is the information source of machine tool accuracy degeneration research and state analysis.Therefore, for the rotating accuracy deterioration produced due to a variety of causes in high speed Ultra-precision Turning main shaft of numerical control machine tool process, that afunction has a strong impact on a series of technical barriers such as the machining accuracy of product parts and quality is urgently to be resolved hurrily, effectively to improve lathe service reliability.
According to statistics, the production loss caused due to Precision of NC Machine Tool deterioration and fault every year reaches hundreds billion of RMB, and the problem such as reliability, fault diagnosis and fault prediction, Performance Evaluation of lathe obtains the extensive concern of Chinese scholars.Rotating accuracy at present for machine-tool spindle system detects the rotating accuracy detection of laying particular stress on static state; Main shaft status monitoring and prediction, more concentrate on typical fault pattern-recognition.The two is independently disconnected from each other, can not provide assessment and the judgement of real-time accuracy deterioration for produced on-site is actual.The rotating accuracy of axis system has a strong impact on machining accuracy, the leading indicator of reflection lathe dynamic property and the important means in analytical error source.
Summary of the invention
For the problems referred to above, the object of this invention is to provide a kind of spindle rotation error source tracing method based on orbit of shaft center manifold learning, it can effectively improve lathe service reliability, and ensures machining accuracy and production efficiency.
For achieving the above object, the present invention takes following technical scheme: a kind of spindle rotation error source tracing method based on orbit of shaft center manifold learning, it comprises the following steps: 1) arrange two current vortex sensors in main shaft periphery to interval, two current vortex sensors are 90 ° along spindle axis and are crisscross arranged, and gather spindle vibration signal by two current vortex sensors; 2) the spindle vibration signal detected is processed, and then the running status of main shaft is judged, it comprises the following steps: the vibration signal that two record in 90 ° of current vortex sensors be crisscross arranged is labeled as X and Y by (1) respectively, adopts Mean-Variance standardized method to be normalized pretreatment to X, Y vibration signal; (2) EEMD noise reduction process is carried out to X, Y vibration signal after normalization; (3) the some orbit of shaft center jointly formed by X, Y vibration signal are extracted, using the discrete point on each orbit of shaft center as a dimension, structure high-dimensional feature space; (4) ISOMAP, LLE or LTSA manifold learning arithmetic is adopted to extract orbit of shaft center two-dimensional manifold as sensitive features; (5) basis can obtain the stream shape sensitive features Q of different faults state after being processed by step (4)
ij, according to stream shape sensitive features Q
ijcarry out main shaft malfunction f (i) to judge; 3) in step 1), two current vortex sensors collect phase difference is that the spindle vibration signal of 90 degree meets at a bit in same plane coordinate system, obtains orbit of shaft center after continuous sampling; 4) to step 3) in obtain spindle axis track carry out error separate, to obtain the actual rotating accuracy A of main shaft; 5) according to actual rotating accuracy A and the stream shape sensitive features Q of main shaft
ijobtain mapping function spectrum data storehouse Q:{f (i)=Q
ij| A}; Wherein, i represents the kind of main shaft state: normal condition, spindle eccentricity, bearing heating wear-out failure state; J=1,2,3, represent the manifold learning arithmetic that ISOMAP, LLE and LTSA tri-kinds is different; 6) if actual rotating accuracy A>=η E of main shaft, η=0.8 ~ 1, then call mapping function spectrum data storehouse Q, carry out tracing to the source of spindle rotation error, and keep in repair corresponding failure; If actual rotating accuracy A>=η E of main shaft, η=0.6 ~ 0.8, then carry out Source Tracing monitoring to spindle rotation error; Wherein E is the spindle rotation accuracy of this lathe when dispatching from the factory.
In described step 4), described spindle axis trajectory error is separated and adopts line-of-sight course roundness fault separating method to carry out being separated of circularity shape and turn error.
Described spindle axis trajectory error is separated and adopts frequency domain line-of-sight course.
The present invention is owing to taking above technical scheme, it has the following advantages: the present invention is bridge owing to adopting with orbit of shaft center, set up associating and mapping of machine tool chief axis running status and spindle rotation accuracy, spindle rotation accuracy deterioration source tracing method based on orbit of shaft center manifold learning is provided, the rotating accuracy degradation trend of main shaft is predicted.Can be foundation raising machining accuracy corresponding real-time compensation method foundation to be provided, to provide for produced on-site is actual real-time accuracy deterioration to judge, realize the deterioration of main shaft precision can recall, can effectively improve lathe service reliability, to promote Digit Control Machine Tool processing precision reliability and production efficiency is significant and using value.The present invention can extensively apply in various Digit Control Machine Tool.
Accompanying drawing explanation
Fig. 1 is overall flow schematic diagram of the present invention;
Fig. 2 is two current vortex sensor structural representations of the present invention;
Fig. 3 is main shaft running state recognition schematic flow sheet of the present invention;
Fig. 4 is that the present invention adopts orbit of shaft center schematic diagram before and after EEMD denoising; Wherein, Fig. 4 (a) is the orbit of shaft center that main shaft normal condition carries out before and after EEMD noise reduction process; Fig. 4 (b) is the orbit of shaft center that main shaft condition of misalignment carries out before and after EEMD noise reduction process;
Fig. 5 two-dimensional manifold feature schematic diagram that to be main shaft of the present invention adopt three kinds of manifold learning arithmetic to obtain when normal, condition of misalignment respectively.
Detailed description of the invention
Below in conjunction with drawings and Examples, the present invention is described in detail.
As shown in Figure 1, the invention provides a kind of spindle rotation error source tracing method based on orbit of shaft center manifold learning, it comprises the following steps:
1) arrange two current vortex sensors, 2, two current vortex sensors 2 to be 90 ° along main shaft 1 axle center to be crisscross arranged (as shown in Figure 2) at the outer circumferential interval of main shaft 1, gather main shaft 1 vibration signal by two current vortex sensors 2.
2) process the spindle vibration signal detected, and then judge (as shown in Figure 3) the running status of main shaft 1, it comprises the following steps:
(1) vibration signal that two record in 90 ° of current vortex sensors be crisscross arranged 2 is labeled as X and Y respectively, adopts Mean-Variance standardized method to be normalized pretreatment to X, Y vibration signal;
(2) EEMD noise reduction process is carried out to X, Y vibration signal after normalization;
(3) the some orbit of shaft center (as shown in Figure 4) jointly formed by X, Y vibration signal are extracted, using the discrete point on each orbit of shaft center as a dimension, structure high-dimensional feature space; Normal and misalign two states for main shaft 1, carry out the orbit of shaft center (as shown in Figure 4) before and after EEMD noise reduction process, this shows that the orbit of shaft center after EEMD noise reduction is more level and smooth.
(4) adopt ISOMAP(Isometric Maps algorithm again), LLE(Local Liner Prediction) or LTSA(local tangent space alignment algorithm) manifold learning arithmetic extracts orbit of shaft center two-dimensional manifold as sensitive features; Wherein often kind of stream shape habit algorithm correspondence represents that the curve map of main shaft 1 running status is different respectively, normal and misalign two states and carry out stream shape sensitive features and extract, the low dimensional manifold figure difference (as shown in Figure 5) of three kinds of manifold learning arithmetic extraction different conditions for main shaft 1.
(5) basis can obtain the stream shape sensitive features Q of different faults state after being processed by step (4)
ij, according to stream shape sensitive features Q
ijcarry out main shaft malfunction f (i) to judge.
3) in step 1), two current vortex sensors 2 collect phase difference is that the spindle vibration signal of 90 degree meets at a bit in same plane coordinate system, can obtain orbit of shaft center (as shown in Figure 4) after continuous sampling.
4) to step 3) in obtain spindle axis track carry out error separate, to obtain the actual rotating accuracy of main shaft.Existing line-of-sight course roundness fault separating method (EST) is adopted to carry out being separated of circularity shape and turn error; The present invention adopts frequency domain line-of-sight course to realize the separation of error, obtains the actual rotating accuracy A of main shaft 1.
Calculating due to spindle rotation error is a kind of method solving its kinematic error size under the condition recording spindle rotation error movement locus quantitatively, represents the size of spindle rotation error with the characteristic value-rotating accuracy of spindle rotation error.The definition mode of spindle rotation accuracy: using the roundness error of circular image as spindle rotation accuracy, after adopting frequency domain line-of-sight course roundness fault separating method, using the roundness error of circular image as spindle rotation accuracy, calculate the actual rotating accuracy A of institute's test main shaft 1.
5) according to actual rotating accuracy A and the stream shape sensitive features Q of main shaft 1
ijobtain mapping function spectrum data storehouse Q:{f (i)=Q
ij| A}; Wherein, i represents the kind of main shaft state, as malfunctions such as normal condition, spindle eccentricity, bearing heating wearing and tearing; J=1,2,3, represent the manifold learning arithmetic that ISOMAP, LLE and LTSA tri-kinds is different; Each main shaft state, all corresponding three kinds of stream shape sensitive features.
6) if actual rotating accuracy A >=η E of main shaft 1, η=0.8 ~ 1, then call mapping function spectrum data storehouse Q, carry out tracing to the source of spindle rotation error, and keep in repair corresponding failure; If actual rotating accuracy A >=η E of main shaft 1, η=0.6 ~ 0.8, then carry out Source Tracing monitoring to spindle rotation error; Wherein E is the spindle rotation accuracy of this lathe when dispatching from the factory.
In sum, the present invention, according to the orbit of shaft center of spindle rotation accuracy, calls the mapping function spectrum data storehouse Q of stream shape sensitive features and malfunction, can be easy to the malfunction determining main shaft 1, achieve tracing to the source of spindle rotation error according to sensitive features.
The various embodiments described above are only for illustration of the present invention; each step all can change to some extent; on the basis of technical solution of the present invention, all improvement of carrying out separate step according to the principle of the invention and equivalents, all should not get rid of outside protection scope of the present invention.
Claims (3)
1., based on a spindle rotation error source tracing method for orbit of shaft center manifold learning, it comprises the following steps:
1) arrange two current vortex sensors in main shaft periphery to interval, two current vortex sensors are 90 ° along spindle axis and are crisscross arranged, and gather spindle vibration signal by two current vortex sensors;
2) process the spindle vibration signal detected, and then judge the running status of main shaft, it comprises the following steps:
(1) vibration signal that two record in 90 ° of current vortex sensors be crisscross arranged is labeled as X and Y respectively, adopts Mean-Variance standardized method to be normalized pretreatment to X, Y vibration signal;
(2) EEMD noise reduction process is carried out to X, Y vibration signal after normalization;
(3) the some orbit of shaft center jointly formed by X, Y vibration signal are extracted, using the discrete point on each orbit of shaft center as a dimension, structure high-dimensional feature space;
(4) ISOMAP, LLE or LTSA manifold learning arithmetic is adopted to extract orbit of shaft center two-dimensional manifold as sensitive features;
(5) basis can obtain the stream shape sensitive features Q of different faults state after being processed by step (4)
ij, according to stream shape sensitive features Q
ijcarry out main shaft malfunction f (i) to judge;
3) through step 1) in two current vortex sensors to collect phase difference be that the spindle vibration signal of 90 degree meets at a bit in same plane coordinate system, obtain orbit of shaft center after continuous sampling;
4) to step 3) in obtain spindle axis track carry out error separate, to obtain the actual rotating accuracy A of main shaft;
5) according to actual rotating accuracy A and the stream shape sensitive features Q of main shaft
ijobtain mapping function spectrum data storehouse Q:{f (i)=Q
ija}; Wherein, i represents the kind of main shaft state: normal condition, spindle eccentricity, bearing heating wear-out failure state; J=1,2,3, represent the manifold learning arithmetic that ISOMAP, LLE and LTSA tri-kinds is different;
6) if actual rotating accuracy A >=η E of main shaft, η=0.8 ~ 1, then call mapping function spectrum data storehouse Q, carry out tracing to the source of spindle rotation error, and keep in repair corresponding failure; If actual rotating accuracy A >=η E of main shaft, η=0.6 ~ 0.8, then carry out Source Tracing monitoring to spindle rotation error; Wherein E is the spindle rotation accuracy of this main shaft place lathe when dispatching from the factory.
2. a kind of spindle rotation error source tracing method based on orbit of shaft center manifold learning as claimed in claim 1, it is characterized in that: described step 4) in, described spindle axis trajectory error is separated and adopts line-of-sight course roundness fault separating method to carry out being separated of circularity shape and turn error.
3. a kind of spindle rotation error source tracing method based on orbit of shaft center manifold learning as claimed in claim 2, is characterized in that: described spindle axis trajectory error is separated and adopts frequency domain line-of-sight course.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310021815.3A CN103042436B (en) | 2013-01-21 | 2013-01-21 | Spindle turning error source tracing method based on shaft center orbit manifold learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310021815.3A CN103042436B (en) | 2013-01-21 | 2013-01-21 | Spindle turning error source tracing method based on shaft center orbit manifold learning |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103042436A CN103042436A (en) | 2013-04-17 |
| CN103042436B true CN103042436B (en) | 2014-12-24 |
Family
ID=48055420
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310021815.3A Active CN103042436B (en) | 2013-01-21 | 2013-01-21 | Spindle turning error source tracing method based on shaft center orbit manifold learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103042436B (en) |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103345196B (en) * | 2013-06-14 | 2015-12-02 | 西安交通大学 | A kind of interlock of the numerically-controlled machine feed shaft based on analysis of spectrum performance estimating method |
| CN104400560B (en) * | 2014-11-07 | 2016-11-23 | 西安交通大学 | A kind of numerical control machine tool cutting operating mode lower main axis orbit of shaft center On-line Measuring Method |
| CN104483118B (en) * | 2014-12-08 | 2017-04-19 | 西安交通大学 | Rotor dynamic and static rub impact fault diagnosis method based on instantaneous frequency shaft centerline orbit |
| CN107036817B (en) * | 2017-04-05 | 2019-03-08 | 哈尔滨理工大学 | SVR rolling bearing performance decline prediction technique based on krill group's algorithm |
| CN108629864B (en) * | 2018-04-27 | 2020-08-21 | 北京科技大学 | Vibration-based electric spindle radial precision characterization method and system |
| CN112846938B (en) * | 2021-01-05 | 2022-09-16 | 北京信息科技大学 | Main shaft rotation precision degradation traceability system under cutting working condition |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101799368A (en) * | 2010-01-27 | 2010-08-11 | 北京信息科技大学 | Electromechanical device nonlinear failure prediction method |
| CN102682074A (en) * | 2012-03-09 | 2012-09-19 | 浙江大学 | Product implicit attribute recognition method based on manifold learning |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8064697B2 (en) * | 2007-10-12 | 2011-11-22 | Microsoft Corporation | Laplacian principal components analysis (LPCA) |
| US9311567B2 (en) * | 2010-05-10 | 2016-04-12 | Kuang-chih Lee | Manifold learning and matting |
| TWI419059B (en) * | 2010-06-14 | 2013-12-11 | Ind Tech Res Inst | Method and system for example-based face hallucination |
-
2013
- 2013-01-21 CN CN201310021815.3A patent/CN103042436B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101799368A (en) * | 2010-01-27 | 2010-08-11 | 北京信息科技大学 | Electromechanical device nonlinear failure prediction method |
| CN102682074A (en) * | 2012-03-09 | 2012-09-19 | 浙江大学 | Product implicit attribute recognition method based on manifold learning |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103042436A (en) | 2013-04-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103042436B (en) | Spindle turning error source tracing method based on shaft center orbit manifold learning | |
| CN103522013B (en) | The unmanned flexible mixed production method of aluminium alloy wheel hub of vehicle and production system | |
| CN103753353B (en) | A kind of non-contact laser measuring method of Fast Measurement milling cutter bias | |
| Shakya et al. | Bearing diagnosis based on Mahalanobis–Taguchi–Gram–Schmidt method | |
| CN105571531A (en) | Dynamic detecting device and adjusting method for misalignment of rotating machine | |
| CN102368156A (en) | On-line estimating method for workpiece processing quality of numerical control machine based on position information | |
| CN102637017A (en) | Real-time monitoring device and method for spindle performance and calibrating experimental device and method thereof | |
| CN110806723A (en) | Novel method for state evaluation of numerical control machine tool | |
| CN105205221A (en) | Precision reliability analysis method for heavy numerical control machine tool | |
| Godreau et al. | Continuous improvement of HSM process by data mining | |
| CN113126563A (en) | Numerical control machine tool data management system and method | |
| CN108257365A (en) | A kind of industrial alarm designs method based on global nonspecific evidence dynamic fusion | |
| Zhang et al. | Complexity analysis of distributed measuring and sensing network in multistage machining processes | |
| CN103341788A (en) | Ultra-precise static-pressure main shaft dynamic characteristic online testing method capable of eliminating measuring basis installation error | |
| Nishida et al. | Sequence planning of on-machine measurement and re-machining | |
| CN101628396B (en) | Method for measuring thermal error of grinding wheel spindle of grinder | |
| CN202676060U (en) | Tool for position measuring of hole and face of connecting rod bolt | |
| Liu et al. | Fault diagnosis of diesel engine information fusion based on adaptive dynamic weighted hybrid distance-taguchi method (ADWHD-T) | |
| Johansson et al. | Context driven remaining useful life estimation | |
| GB2573273A (en) | System And Method For Monitoring Characteristics Of A Rotary Table | |
| CN108303251A (en) | Rigidity modeling and Indirect Detecting Method under a kind of electro spindle rotary state | |
| CN200999636Y (en) | Vehicle lubricant oil abrasive grain monitor | |
| CN208772643U (en) | Collector ring detecting tool | |
| CN108145531A (en) | A kind of device and method for detecting machine tool motion characteristic and precision deterioration law | |
| Tan | Fast and Accurate Position Determination of Multimodal Couplings of Steel Rolling Machinery and Equipment Combined with Multimedia Optimized Particle Filters |
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 |