CN102825504B - State detection method for main shaft of numerically-controlled machine tool - Google Patents

State detection method for main shaft of numerically-controlled machine tool Download PDF

Info

Publication number
CN102825504B
CN102825504B CN201210346785.9A CN201210346785A CN102825504B CN 102825504 B CN102825504 B CN 102825504B CN 201210346785 A CN201210346785 A CN 201210346785A CN 102825504 B CN102825504 B CN 102825504B
Authority
CN
China
Prior art keywords
time
signal
domain analysis
frequency
domain
Prior art date
Application number
CN201210346785.9A
Other languages
Chinese (zh)
Other versions
CN102825504A (en
Inventor
王雪
李益才
陈洪
Original Assignee
重庆科技学院
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 重庆科技学院 filed Critical 重庆科技学院
Priority to CN201210346785.9A priority Critical patent/CN102825504B/en
Publication of CN102825504A publication Critical patent/CN102825504A/en
Application granted granted Critical
Publication of CN102825504B publication Critical patent/CN102825504B/en

Links

Abstract

The invention discloses a state detection method for a main shaft of a numerically-controlled machine tool, belonging to the field of machine tool failure diagnosis. The method comprises the following steps of: acquiring a signal transmitted by a sensor system through a host; and analyzing the signal acquired according to the step 1, determining a failure characteristic parameter, processing the obtained failure characteristic parameter through a specialist system, an artificial nerve network system and a support vector respectively to obtain three failure diagnosis results, and fusing the three failure diagnosis results through a genetic algorithm. By fusing the diagnosis results of the specialist system, the artificial nerve network system and the support vector, an artificial intelligence effect can be achieved during field failure diagnosis, so that the failure diagnosis accuracy of the numerically-controlled machine tool is increased; moreover, a displacement signal is detected by adopting a CCD (Charge Coupled Device) laser displacement sensor, so that the obtained run-out accuracy of the main shaft can be up to 0.5 micron, and the failure diagnosis accuracy of the numerically-controlled machine tool is further ensured.

Description

State detection

Technical field

The invention belongs to machine failure diagnostic field, particularly relate to a kind of State detection.

Background technology

Digit Control Machine Tool is a kind of automated machine tool that program control system is housed, and this program control system can logically process has control coding or other symbolic instruction regulated procedures, and by its decoding, thus machine tool action is processed machine components.Main shaft of numerical control machine tool refers to the axle that lathe drives workpiece or cutter to rotate, usually spindle unit is formed by main shaft, bearing and driving member such as gear or belt wheel etc., except the lathe that the main motion such as planer, broaching machine is rectilinear motion, most of lathe has spindle unit, and the kinematic accuracy of spindle unit and the rigidity of structure are the key factors determining crudy and stock-removing efficiency.Weigh index mainly rotating accuracy, rigidity and the velocity adaptive of spindle unit performance, what occur on the direction affecting machining accuracy when rotating accuracy refers to that main shaft rotates radial and axially beats, and depends mainly on manufacture and the assembly quality of main shaft and bearing; Dynamic and static rigidity depends mainly on the bending stiffness of main shaft, the rigidity of bearing and damping; Velocity adaptive refers to maximum speed and the range of speeds of permission, depends mainly on structure and the lubrication of bearing, and radiating condition.

The quality of main shaft runnability is the important key factor determining Digit Control Machine Tool crudy, if when installation and debugging or maintenance, drive disk assembly in main spindle box is (as tooth form synchronous belt, gear and shaft coupling etc.) parameter adjustment is improper, or the part distortion owing to causing in use, wearing and tearing etc., the decline of main shaft performance is caused in capital, main shaft is caused to beat overproof in machining accuracy direction, crudy is declined, machining accuracy exceeds standard, even completely can not processing parts, therefore, ensure that main shaft runnability is the prerequisite that Digit Control Machine Tool carries out normal process at kilter, and the running status of main shaft is monitored accurately, can accurately diagnose when going wrong, its running precision is predicted to be the basic condition ensureing that main shaft runnability is good.

The outer detection and diagnosis technology to Digit Control Machine Tool of Present Domestic, mainly concentrate on the aspects such as the driver part of Digit Control Machine Tool, PLC control logic and electric fault, to the diagnosis of main shaft and the investigation and application of prediction aspect less, because the jerk value of main shaft own is very little on the one hand, general between 0.5 ~ 1 micron, need the displacement transducer that precision is very high; Need on the other hand on hardware and software, ensure that the antijamming capability tested is strong.

The running status of main shaft can be reflected by the vibration of the jerk value of main shaft and Vibration Condition, main spindle box and temperature conditions, for the actual conditions of above-mentioned main tapping, mainly through ANN (Artificial Neural Networks in prior art, artificial neural network), ES (Expert System, expert system) or SVM (support vector machine, SVMs) Analysis on Fault Diagnosis is carried out to Digit Control Machine Tool.

The feature of ANN is that Modling model is relatively easy, but needs to carry out parameter differentiation, and the layering of its neutral net there is no guidance method at present, is not easy to obtain optimized level, and whether restrains and need to carry out computational analysis.

The birth of SVM is that the solution solving failure modes problem opens new approach, and Statistical Learning Theory is on the basis of structural risk minimization, specially a set of new theoretical system set up for the Machine Learning Problems of small sample.At present, the many sorting techniques of SVM have successfully been applied in many practical problems, as text identification, recognition of face, Three-dimension object recognition, speech recognition, Images Classification, gene sequencing etc.

Expert system based on sign computation and reasoning is widely used in process failure diagnosis field, expert system can high efficiency, carry out work accurately, thoughtfully, rapidly and tirelessly, but expert system can not obtain new knowledge from the example of diagnosis, there is the bottleneck of knowledge acquisition, and on knowledge-based reasoning, traditional expert system adopts serial mode, and inference method is single, control strategy is dumb, easily occurs the problems such as matching conflict, multiple shot array, infinite recurrence.

SVMs has good extensive Generalization Ability than neutral net, for the mapping of each higher dimensional space in this space, how to determine that this mapping i.e. kernel function also do not have suitable method now, so for general problem, SVM has just transferred the difficulty of the complexity of higher dimensional space the difficulty asking kernel function to, even if after definite kernel function, when Solve problems is classified, solve the quadratic programming of kernel function, this just needs a large amount of memory spaces, this is also a problem of SVM, and artificial neural network system just can head it off, therefore, how dexterously by expert system, artificial neural network system and SVMs join together to carry out intelligent diagnostics, it is the key problem of this project research.

Therefore those skilled in the art are devoted to develop a kind of State detection, can Fusion Expert System, artificial neural network system and SVMs diagnostic result, thus improve the accuracy of diagnosing faults of numerical control machine.

Summary of the invention

Because the above-mentioned defect of prior art, technical problem to be solved by this invention be to provide a kind of can Fusion Expert System, artificial neural network system and SVMs diagnostic result, thus improve the accuracy of diagnosing faults of numerical control machine.

For achieving the above object, the invention provides a kind of State detection, comprise the following steps:

The signal that step one, host computer pick-up transducers system send;

Step 2, analyze the signal collected according to step one, determine Fault characteristic parameters;

Step 3, Fault characteristic parameters step 2 obtained are processed by expert system, draw Fisrt fault diagnostic result; Fault characteristic parameters step 2 obtained is processed by artificial neural network system, draws the second fault diagnosis result; Fault characteristic parameters step 2 obtained is processed by SVMs, draws the 3rd fault diagnosis result;

Step 4, described Fisrt fault diagnostic result, the second fault diagnosis result and the 3rd fault diagnosis result to be merged by genetic algorithm.

In described step 4, described Fisrt fault diagnostic result, the second fault diagnosis result and the 3rd fault diagnosis result are undertaken merging by genetic algorithm and carry out according to the following steps:

A1, described Fisrt fault diagnostic result, the second fault diagnosis result and the 3rd fault diagnosis result to be encoded, each fault diagnosis result is converted to the genotypic population be made up of chromosome;

A2, calculate all chromosomal fitness values;

A2, to be formed based on each chromosomal fitness value and select wheel disc;

A3, from wheel disc the pairing individuality of each genotype individuals of Stochastic choice carry out single-point intersection and variation process, form new individuality;

A4, when set the fitness value of described genotypic population in algebraically do not change or the fitness value of described genotypic population is positioned at the precision threshold of setting time, genetic algorithm terminates.

Preferably, described sensing system comprises temperature sensor, and described temperature sensor sends temperature value signal to described host computer; Described host computer also comprises the step of the temperature value signal collected being carried out to time-domain analysis after gathering the signal that described sensing system sends;

Time-domain analysis is carried out to the temperature value signal collected and comprises peak-to-peak value, value, kurtosis and the waveform of analyzing this signal, by drawing the Fault characteristic parameters of this temperature value signal after carrying out time-domain analysis to the temperature value signal collected.

Further, described sensing system also comprises acceleration transducer, and described acceleration transducer sends acceleration signal to described host computer; Described host computer also comprises the step of the acceleration signal collected being carried out to time-domain analysis and frequency-domain analysis after gathering the signal that described sensing system sends;

Time-domain analysis is carried out to the acceleration signal collected and comprises peak-to-peak value, value, kurtosis and the waveform of analyzing this signal, after time-domain analysis completes, time domain data is transformed to frequency domain, frequency-domain analysis is carried out to the acceleration signal collected and comprises single channel and multichannel stable state spectrum analysis and time frequency analysis; The Fault characteristic parameters of this acceleration signal is drawn after time-domain analysis and frequency-domain analysis are carried out to the acceleration signal collected.

Further, described sensing system also comprises displacement transducer, and preferably, institute's displacement sensors is CCD laser displacement sensor; Institute's displacement sensors sends displacement signal to described host computer; Described host computer also comprises the jerk value calculating main shaft and the step of displacement signal being carried out to time domain, frequency-domain analysis after gathering the signal that described sensing system sends:

Described calculating main axis jerking amount is carried out according to the following steps:

The time domain data of institute's displacement signal is transformed to frequency domain, then carries out High frequency filter at frequency domain, and calculate the main axis jerk value of a week by the time domain point in the monocycle;

Time-domain analysis is carried out to the displacement signal collected and comprises peak-to-peak value, value, kurtosis and the waveform of analyzing this signal, after time-domain analysis completes, time domain data is transformed to frequency domain, frequency-domain analysis is carried out to the displacement signal collected and comprises single channel and multichannel stable state spectrum analysis and time frequency analysis; The Fault characteristic parameters of this displacement signal is drawn after time-domain analysis and frequency-domain analysis are carried out to the displacement signal collected.But repeatable accuracy due to CCD laser displacement sensor reaches 0.01 micron brings great interfering signal, and the signal therefore recorded seriously drifts about, and the main axis jerking accuracy of measurement obtained by above method can reach 0.05 micron.

The invention has the beneficial effects as follows: the present invention is by the diagnostic result of Fusion Expert System, artificial neural network system and SVMs, can at the scene fault diagnosis time play the effect of artificial intelligence, thus improve the accuracy of diagnosing faults of numerical control machine, the present invention adopts CCD laser displacement sensor detecting position shifting signal in addition, the main axis jerking accuracy of measurement obtained can reach 0.05 micron, further ensure that the accuracy of diagnosing faults of numerical control machine.

Accompanying drawing explanation

Fig. 1 is the structural representation of the device for main shaft of numerical control machine tool state-detection.

Fig. 2 is the schematic flow sheet of the embodiment of the present invention one.

Fig. 3 is the schematic flow sheet of the embodiment of the present invention two.

Detailed description of the invention

Below in conjunction with drawings and Examples, the invention will be further described:

Embodiment one: as shown in Figure 1, a kind of device for main shaft of numerical control machine tool state-detection, comprises host computer 1 and sensing system 2; Described sensing system 2 comprises temperature sensor 201, acceleration transducer 202 and displacement transducer 203, the output of described temperature sensor 201 connects the first input end of described host computer 1 by RS232 data wire, the output of described acceleration transducer 202 connects the second input of described host computer 1 by capture card 204, the output of institute's displacement sensors 203 connects the 3rd input of described host computer 1 by displacement controller 205.

As shown in Figure 2, a kind of State detection, comprises the following steps:

S1, described host computer 1 gather the signal that described sensing system 2 sends.

S2, time-domain analysis is carried out to the temperature value signal collected; Time-domain analysis is carried out to the temperature value signal collected and comprises peak-to-peak value, value, kurtosis and the waveform of analyzing this signal, by drawing the Fault characteristic parameters index of this temperature value signal after carrying out time-domain analysis to the temperature value signal collected.

S3, time-domain analysis and frequency-domain analysis are carried out to the acceleration signal collected; Time-domain analysis is carried out to the acceleration signal collected and comprises peak-to-peak value, value, kurtosis and the waveform of analyzing this signal, after time-domain analysis completes, time domain data is transformed to frequency domain, frequency-domain analysis is carried out to the acceleration signal collected and comprises single channel and multichannel stable state spectrum analysis and time frequency analysis; The Fault characteristic parameters index of this acceleration signal is drawn after time-domain analysis and frequency-domain analysis are carried out to the acceleration signal collected.

S4, calculate main shaft jerk value and time domain, frequency-domain analysis are carried out to displacement signal; Described calculating main axis jerking amount is carried out according to the following steps:

The time domain data of institute's displacement signal is transformed to frequency domain, then carries out High frequency filter at frequency domain, and calculate the main axis jerk value of a week by the time domain point in the monocycle; Repeatedly calculate the main axis jerk value of a week in the present embodiment and obtain the main axis jerk value mean value of one week; Time-domain analysis is carried out to the displacement signal collected and comprises peak-to-peak value, value, kurtosis and the waveform of analyzing this signal, after time-domain analysis completes, time domain data is transformed to frequency domain, frequency-domain analysis is carried out to the displacement signal collected and comprises single channel and multichannel stable state spectrum analysis and time frequency analysis; The Fault characteristic parameters index of this displacement signal is drawn after time-domain analysis and frequency-domain analysis are carried out to the displacement signal collected.

S5, obtain Fault characteristic parameters by the Fault characteristic parameters index of the Fault characteristic parameters index of jerk value, acceleration signal, the Fault characteristic parameters index of temperature value signal and displacement signal, i.e. the temperature value of the jerk value of main shaft of numerical control machine tool, the vibratory output of main shaft of numerical control machine tool, the vibratory output of main spindle box of numerical control body and main spindle box of numerical control body.

S6, the Fault characteristic parameters obtained by step S5 are processed by expert system, draw Fisrt fault diagnostic result; The Fault characteristic parameters obtained by step S5 is processed by artificial neural network system, draws the second fault diagnosis result; The Fault characteristic parameters obtained by step S5 is processed by SVMs, draws the 3rd fault diagnosis result.

S7, described Fisrt fault diagnostic result, the second fault diagnosis result and the 3rd fault diagnosis result to be merged by genetic algorithm.

In described step S7, described Fisrt fault diagnostic result, the second fault diagnosis result and the 3rd fault diagnosis result are undertaken merging by genetic algorithm and carry out according to the following steps:

A1, described Fisrt fault diagnostic result, the second fault diagnosis result and the 3rd fault diagnosis result to be encoded, each fault diagnosis result is converted to the genotypic population be made up of chromosome;

A2, calculate all chromosomal fitness values;

A2, to be formed based on each chromosomal fitness value and select wheel disc;

A3, from wheel disc the pairing individuality of each genotype individuals of Stochastic choice carry out single-point intersection and variation process, form new individuality;

A4, when set the fitness value of described genotypic population in algebraically do not change or the fitness value of described genotypic population is positioned at the precision threshold of setting time, genetic algorithm terminates.

Be provided with two acceleration transducers 202 in the present embodiment, acceleration transducer 202 adopts ICP acceleration transducer, after carrying out signal anti-aliasing filtering and AD conversion, becomes data signal and enters host computer 1 by capture card 204.Temperature sensor 201 adopts infrared temperature sensor, the CCD laser displacement sensor that institute's displacement sensors 203 adopts Japanese Keyemce company to produce, its repeatable accuracy is 0.05 micron, displacement transducer 203 transmits data to host computer 1 by displacement controller 205, in the present embodiment, host computer 1 adopts computer to realize, certainly, described host computer 1 also can adopt the realization such as notebook computer, panel computer.

The artificial neural network system's model adopted in the present embodiment is BP model, certain described artificial neural network system's model can also be RBF model and Elman model, for ensureing the convergence of genetic algorithm, in the present embodiment, genetic algorithm adopts the method for roulette, individuality the highest for fitness is directly copied in follow-on population, both can improve pace of learning, the optimal solution that can retain again genotypic population exempts from degeneration.

It should be noted that the order of step S2, S3 and S4 can be exchanged arbitrarily, after performing step S1, perform S3, S4 and S2 successively, then perform step S5, S6 and S7 successively.

Embodiment two, as shown in Figure 3, the present embodiment is substantially identical with the flow process of embodiment one, and difference is:

S1, described host computer 1 gather the signal that described sensing system 2 sends.

S2, time-domain analysis is carried out to the temperature value signal collected; Time-domain analysis and frequency-domain analysis are carried out to the acceleration signal collected; Calculate the jerk value of main shaft and time domain, frequency-domain analysis are carried out to displacement signal; Time-domain analysis is carried out to the temperature value signal collected and comprises peak-to-peak value, value, kurtosis and the waveform of analyzing this signal, by drawing the Fault characteristic parameters index of this temperature value signal after carrying out time-domain analysis to the temperature value signal collected.

Time-domain analysis is carried out to the acceleration signal collected and comprises peak-to-peak value, value, kurtosis and the waveform of analyzing this signal, after time-domain analysis completes, time domain data is transformed to frequency domain, frequency-domain analysis is carried out to the acceleration signal collected and comprises single channel and multichannel stable state spectrum analysis and time frequency analysis; The Fault characteristic parameters index of this acceleration signal is drawn after time-domain analysis and frequency-domain analysis are carried out to the acceleration signal collected.

Described calculating main axis jerking amount is carried out according to the following steps: the time domain data of institute's displacement signal is transformed to frequency domain, then carries out High frequency filter at frequency domain, and calculates the main axis jerk value of a week by the time domain point in the monocycle; Repeatedly calculate the main axis jerk value of a week in the present embodiment and obtain the main axis jerk value mean value of one week.

Time-domain analysis is carried out to the displacement signal collected and comprises peak-to-peak value, value, kurtosis and the waveform of analyzing this signal, after time-domain analysis completes, time domain data is transformed to frequency domain, frequency-domain analysis is carried out to the displacement signal collected and comprises single channel and multichannel stable state spectrum analysis and time frequency analysis; The Fault characteristic parameters index of this displacement signal is drawn after time-domain analysis and frequency-domain analysis are carried out to the displacement signal collected.

Fault characteristic parameters is obtained, i.e. the temperature value of the jerk value of main shaft of numerical control machine tool, the vibratory output of main shaft of numerical control machine tool, the vibratory output of main spindle box of numerical control body and main spindle box of numerical control body by the Fault characteristic parameters index of the Fault characteristic parameters index of main axis jerking amount, acceleration signal, the Fault characteristic parameters index of temperature value signal and displacement signal.

S3, the Fault characteristic parameters obtained by S2 are processed by expert system, draw Fisrt fault diagnostic result; The Fault characteristic parameters obtained by step S2 is processed by artificial neural network system, draws the second fault diagnosis result; The Fault characteristic parameters obtained by step S2 is processed by SVMs, draws the 3rd fault diagnosis result.

S4, described Fisrt fault diagnostic result, the second fault diagnosis result and the 3rd fault diagnosis result to be merged by genetic algorithm.

More than describe preferred embodiment of the present invention in detail.Should be appreciated that those of ordinary skill in the art just design according to the present invention can make many modifications and variations without the need to creative work.Therefore, all technical staff in the art, all should by the determined protection domain of claims under this invention's idea on the basis of existing technology by the available technical scheme of logical analysis, reasoning, or a limited experiment.

Claims (6)

1. a State detection, comprises host computer (1) and sensing system (2); The output of described sensing system (2) connects the input of described host computer (1); It is characterized in that comprising the following steps:
Step one, described host computer (1) gather the signal that described sensing system (2) sends;
Step 2, analyze the signal collected according to step one, determine Fault characteristic parameters;
Step 3, Fault characteristic parameters step 2 obtained are processed by expert system, draw Fisrt fault diagnostic result; Fault characteristic parameters step 2 obtained is processed by artificial neural network system, draws the second fault diagnosis result; Fault characteristic parameters step 2 obtained is processed by SVMs, draws the 3rd fault diagnosis result;
Step 4, described Fisrt fault diagnostic result, the second fault diagnosis result and the 3rd fault diagnosis result to be merged by genetic algorithm.
2. State detection as claimed in claim 1, is characterized in that: described Fisrt fault diagnostic result, the second fault diagnosis result and the 3rd fault diagnosis result are undertaken merging by genetic algorithm in described step 4 and carry out according to the following steps:
A1, described Fisrt fault diagnostic result, the second fault diagnosis result and the 3rd fault diagnosis result to be encoded, each fault diagnosis result is converted to the genotypic population be made up of chromosome;
A2, calculate all chromosomal fitness values;
A2, to be formed based on each chromosomal fitness value and select wheel disc;
A3, from wheel disc the pairing individuality of each genotype individuals of Stochastic choice carry out single-point intersection and variation process, form new individuality;
A4, when set the fitness value of described genotypic population in algebraically do not change or the fitness value of described genotypic population is positioned at the precision threshold of setting time, genetic algorithm terminates.
3. State detection as claimed in claim 1, is characterized in that: described sensing system (2) comprises temperature sensor (201), and described temperature sensor (201) sends temperature value signal to described host computer (1); Described host computer (1) also comprises the step of the temperature value signal collected being carried out to time-domain analysis after gathering the signal that described sensing system (2) sends;
Time-domain analysis is carried out to the temperature value signal collected and comprises peak-to-peak value, value, kurtosis and the waveform of analyzing this signal, by drawing the Fault characteristic parameters of this temperature value signal after carrying out time-domain analysis to the temperature value signal collected.
4. the State detection as described in claim 1 or 3, is characterized in that: described sensing system (2) also comprises acceleration transducer (202), and described acceleration transducer (202) sends acceleration signal to described host computer (1); Described host computer (1) also comprises the step of the acceleration signal collected being carried out to time-domain analysis and frequency-domain analysis after gathering the signal that described sensing system (2) sends;
Time-domain analysis is carried out to the acceleration signal collected and comprises peak-to-peak value, value, kurtosis and the waveform of analyzing this signal, after time-domain analysis completes, time domain data is transformed to frequency domain, frequency-domain analysis is carried out to the acceleration signal collected and comprises single channel and multichannel stable state spectrum analysis and time frequency analysis; The Fault characteristic parameters of this acceleration signal is drawn after time-domain analysis and frequency-domain analysis are carried out to the acceleration signal collected.
5. State detection as claimed in claim 4, is characterized in that: described sensing system (2) also comprises displacement transducer (203), and institute's displacement sensors (203) sends displacement signal to described host computer (1); Described host computer (1) also comprises the jerk value calculating main shaft and the step of displacement signal being carried out to time domain, frequency-domain analysis after gathering the signal that described sensing system (2) sends:
Described calculating main axis jerking amount is carried out according to the following steps:
The time domain data of institute's displacement signal is transformed to frequency domain, then carries out High frequency filter at frequency domain, and calculate the main axis jerk value of a week by the time domain point in the monocycle;
Time-domain analysis is carried out to the displacement signal collected and comprises peak-to-peak value, value, kurtosis and the waveform of analyzing this signal, after time-domain analysis completes, time domain data is transformed to frequency domain, frequency-domain analysis is carried out to the displacement signal collected and comprises single channel and multichannel stable state spectrum analysis and time frequency analysis; The Fault characteristic parameters of this displacement signal is drawn after time-domain analysis and frequency-domain analysis are carried out to the displacement signal collected.
6. State detection as claimed in claim 5, is characterized in that: institute's displacement sensors (203) is CCD laser displacement sensor.
CN201210346785.9A 2012-09-18 2012-09-18 State detection method for main shaft of numerically-controlled machine tool CN102825504B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201210346785.9A CN102825504B (en) 2012-09-18 2012-09-18 State detection method for main shaft of numerically-controlled machine tool

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201210346785.9A CN102825504B (en) 2012-09-18 2012-09-18 State detection method for main shaft of numerically-controlled machine tool

Publications (2)

Publication Number Publication Date
CN102825504A CN102825504A (en) 2012-12-19
CN102825504B true CN102825504B (en) 2014-12-24

Family

ID=47328924

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201210346785.9A CN102825504B (en) 2012-09-18 2012-09-18 State detection method for main shaft of numerically-controlled machine tool

Country Status (1)

Country Link
CN (1) CN102825504B (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP3425460A1 (en) * 2017-07-04 2019-01-09 Siemens Aktiengesellschaft Device and method for determining the condition of a spindle of a machine tool

Families Citing this family (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103624633A (en) * 2013-12-09 2014-03-12 大连理工大学 Micro-milling vibration precision measurement system taking laser micro-displacement sensor as measuring element
CN105269404B (en) * 2014-11-20 2018-11-23 电子科技大学 Numerically-controlled machine tool point of a knife dynamic characteristic accuracy detecting device and its method
CN104742153A (en) * 2015-04-01 2015-07-01 中国计量学院 Fault predication device of six-axis multi-joint industrial robot
CN104942655A (en) * 2015-07-20 2015-09-30 徐文欢 Main shaft rotary error accuracy testing device
CN105204436B (en) * 2015-08-26 2016-10-26 山东省计算中心(国家超级计算济南中心) Diagnosing faults of numerical control machine method based on grading forewarning system
CN107199506B (en) * 2017-06-05 2019-05-21 苏州微著设备诊断技术有限公司 A kind of grinding trembling detection method based on stack self-encoding encoder and support vector machines
CN108760813A (en) * 2018-06-05 2018-11-06 哈尔滨工程大学 A kind of turbine blade of gas turbine health monitoring systems and method based on temperature signal
CN109396953B (en) * 2018-12-05 2020-03-31 上海交通大学 Machine tool working state intelligent identification system based on signal fusion
CN110347114A (en) * 2019-07-23 2019-10-18 清华大学 A kind of Cnc ReliabilityintelligeNetwork Network data acquisition and analysis system

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6948381B1 (en) * 2002-04-09 2005-09-27 Rockwell Automation Technologies, Inc. System and method for sensing torque on a rotating shaft
CN101722438A (en) * 2008-10-28 2010-06-09 大隈株式会社 Vibration suppressing method and vibration suppressing device for machine tool
CN102452018A (en) * 2010-10-27 2012-05-16 株式会社捷太格特 Spindle condition detection device for machine tool
CN102490086A (en) * 2011-10-28 2012-06-13 浙江大学 System for monitoring working state of boring rod in real time

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2754760B2 (en) * 1989-07-13 1998-05-20 オムロン株式会社 Tool damage detection device

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6948381B1 (en) * 2002-04-09 2005-09-27 Rockwell Automation Technologies, Inc. System and method for sensing torque on a rotating shaft
CN101722438A (en) * 2008-10-28 2010-06-09 大隈株式会社 Vibration suppressing method and vibration suppressing device for machine tool
CN102452018A (en) * 2010-10-27 2012-05-16 株式会社捷太格特 Spindle condition detection device for machine tool
CN102490086A (en) * 2011-10-28 2012-06-13 浙江大学 System for monitoring working state of boring rod in real time

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
JP平3-49850A 1991.03.04 *
张建民等.《面向机电系统状态监测与故障诊断的现代技术》.《北京理工大学学报》.2004,第24卷(第9期), *
许志军.《基于粒子群算法优化支持向量机的数控机床状态预测》.《现代制造工程》.2011,(第7期), *

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP3425460A1 (en) * 2017-07-04 2019-01-09 Siemens Aktiengesellschaft Device and method for determining the condition of a spindle of a machine tool
WO2019007820A1 (en) * 2017-07-04 2019-01-10 Siemens Aktiengesellschaft Device and method for determining the status of a spindle of a machine tool

Also Published As

Publication number Publication date
CN102825504A (en) 2012-12-19

Similar Documents

Publication Publication Date Title
CN104057290B (en) A kind of robotic asssembly method and system of view-based access control model and force-feedback control
Mikołajczyk et al. Predicting tool life in turning operations using neural networks and image processing
CN105171525B (en) The diagnostic method and system of lathe
CN103056759B (en) Robot grinding system based on feedback of sensor
Stavropoulos et al. Monitoring and control of manufacturing processes: A review
CN103465107B (en) A kind of Tool Wear Monitoring method
Chen et al. Multi-fault diagnosis study on roller bearing based on multi-kernel support vector machine with chaotic particle swarm optimization
Lamraoui et al. Indicators for monitoring chatter in milling based on instantaneous angular speeds
CN107003663B (en) The monitoring of device with movable part
CN102449457B (en) For the system and method for gearbox health monitoring
Martins et al. Tool condition monitoring of single-point dresser using acoustic emission and neural networks models
JP6140331B1 (en) Machine learning device and machine learning method for learning failure prediction of main shaft or motor driving main shaft, and failure prediction device and failure prediction system provided with machine learning device
Prickett et al. An overview of approaches to end milling tool monitoring
ES2776651T3 (en) Procedure for monitoring rotating machines
JP6196093B2 (en) Vibration analysis method for bearing device, vibration analysis device for bearing device, and state monitoring device for rolling bearing
DE4447288B4 (en) Machine fault diagnosis system and method
Aliustaoglu et al. Tool wear condition monitoring using a sensor fusion model based on fuzzy inference system
US10525563B2 (en) Abnormality-detecting device and method for tool of machine tool
Rehorn et al. State-of-the-art methods and results in tool condition monitoring: a review
Benardos et al. Predicting surface roughness in machining: a review
Khorasani et al. Chatter prediction in turning process of conical workpieces by using case-based resoning (CBR) method and taguchi design of experiment
Chen et al. CPS modeling of CNC machine tool work processes using an instruction-domain based approach
Bediaga et al. An automatic spindle speed selection strategy to obtain stability in high-speed milling
Caixu et al. A review of chatter vibration research in milling
US7571022B2 (en) System and method for monitoring machine health

Legal Events

Date Code Title Description
PB01 Publication
C06 Publication
SE01 Entry into force of request for substantive examination
C10 Entry into substantive examination
GR01 Patent grant
C14 Grant of patent or utility model
CF01 Termination of patent right due to non-payment of annual fee

Granted publication date: 20141224

Termination date: 20150918

EXPY Termination of patent right or utility model