CN103264317B - A kind of appraisal procedure of Milling Process cutter operational reliability - Google Patents

A kind of appraisal procedure of Milling Process cutter operational reliability Download PDF

Info

Publication number
CN103264317B
CN103264317B CN201310181390.2A CN201310181390A CN103264317B CN 103264317 B CN103264317 B CN 103264317B CN 201310181390 A CN201310181390 A CN 201310181390A CN 103264317 B CN103264317 B CN 103264317B
Authority
CN
China
Prior art keywords
milling process
cutter
process cutter
running status
state
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
Application number
CN201310181390.2A
Other languages
Chinese (zh)
Other versions
CN103264317A (en
Inventor
刘繁茂
彭佑多
毛征宇
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Funing Jinyu Plastics Co.,Ltd.
Original Assignee
Hunan University of Science and Technology
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 Hunan University of Science and Technology filed Critical Hunan University of Science and Technology
Priority to CN201310181390.2A priority Critical patent/CN103264317B/en
Publication of CN103264317A publication Critical patent/CN103264317A/en
Application granted granted Critical
Publication of CN103264317B publication Critical patent/CN103264317B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Landscapes

  • Complex Calculations (AREA)

Abstract

The invention discloses a kind of appraisal procedure of Milling Process cutter operational reliability.The present invention first gathers machining tool main shaft and feed shaft three-phase current signal by Hall element, then extracts time domain characteristics of mean and tooth frequency energy feature by wavelet analysis.For these characteristic vectors, pass through mean algorithm, obtains observation vector output function: Gaussian mixtures density function.Then adopt Bao Mu-Wei Erqi (Baum-Welch) Algorithm for Training " hidden-semi-Markov " model continuously, obtain parameter estimation result, adopt Qie Puman-Andrei Kolmogorov differential equation, calculate Milling Process cutter operational reliability level.The present invention provides decision support for the preventative maintenance of Milling Process cutter.

Description

A kind of appraisal procedure of Milling Process cutter operational reliability
Technical field
The present invention relates to a kind of appraisal procedure of Milling Process cutter operational reliability.
Technical background
In modern manufacturing, cutting tool state is to guarantee crudy and boost productivity most important.But tool wear work in-process is inevitable again, it directly can affect machining accuracy and the surface roughness of workpiece, not only reduces crudy, also can affect the safety of system of processing and normal operation time serious.Therefore, good cutting tool state has become the necessary condition in modern mechanical manufacturing and automation processing, and it is also the key technology ensureing processing work quality.And artificial cognition cutting tool state has become the important bottleneck of restriction process industry development, therefore, in the urgent need to a kind of new method of carrying out intellectual monitoring and automatic evaluation for cutter running status of design and development, it does not need to wait for that Milling Process cutter breaks down through long-play, just can analyze exactly, assess Milling Process Tool Reliability change conditions, thus instruct user to take rational preventive measure in advance, prevent from affecting processing work quality and system of processing safe operation because of cutter problem.
Summary of the invention
For solving in prior art the above-mentioned technical problem existed, the invention provides a kind ofly can provide the Milling Process cutter Analysis of Running Reliability method of decision support for preventative maintenance.
The technical scheme solved the problems of the technologies described above comprise the following steps into:
1) according to design data and the use history of Milling Process cutter, its running status sum M is determined; Milling Process cutter running status set representations is S={s 1, s 2... s m... s m, s mfor Milling Process cutter is entirely ineffective.
2) for milling cutter processing operating mode, record cutting parameter, utilizes Hall element to gather machine tool chief axis and feed shaft three-phase current signal, and the average therefrom in extraction time domain and the energy value of tooth frequency signal are as characteristic quantity; Specifically comprise: 1. frequency analysis, carry out wavelet decomposition to spindle motor of machine tool current signal, the signal extracting tooth frequency content is reconstructed, and obtains tooth frequency curve, carries out the extraction of energy eigenvalue; 2. time-domain analysis: get feed shaft current signal three-phase current root-mean-square value and carry out wavelet analysis, extracts the characteristic quantity in signal time domain.
3) according to Milling Process cutter running status sum M, the gauss of distribution function number k corresponding to each Milling Process cutter running status is determined m; For a certain running status of Milling Process cutter, the characteristic signal of above-mentioned collection is averaged segmentation, and the parameter belonging to a section in characteristic signal is formed a large matrix, K mean cluster is carried out to the vector of each section, obtains the continuous mixed Gaussian probability density function corresponding to certain state of Milling Process cutter wherein: for t observation vector, k mfor state s mthe number of corresponding Gaussian component, for Gaussian Profile Mean Parameters, Σ mkfor Gaussian Profile covariance matrix, c mkstate s mthe weight of corresponding each Gaussian component; Repeat step 3), obtain the continuous mixed Gaussian probability density function of all running statuses of Milling Process cutter.
4) continuous hidden Semi-Markov Process is adopted to carry out reliability level estimation to cutter running status; First to the given initial value of continuous hidden Semi-Markov Process CHSMM, then adopt Baum-Welch algorithm to model training, complete parameter Estimation, carry out Milling Process cutter operational reliability level calculation;
Wherein, state duration model adopts Gamma distribution, namely v mfor form parameter, w mfor scale parameter; Observation model adopts Gaussian mixtures, namely for t observation vector, k mfor state s mthe number of corresponding Gaussian component, for Gaussian Profile Mean Parameters, Σ mkfor Gaussian Profile covariance matrix, c mkstate s mthe weight of corresponding each Gaussian component.
Above-mentioned steps 3) detailed process as follows:
1. model definition: being expressed as of continuous hidden-semi-Markov model wherein, M is cutter running status sum, and m is Milling Process cutter running status, initial state distribution π m=(π 1..., π m), state transition probability matrix a mnrepresent that cutter is from motion state s mjump to motion state s nprobability.
2. model training: adopt Bao Mu-Wei Erqi (Baum-Welch) algorithm to model training, namely solve the Parameter Estimation Problem of model, obtain model parameter estimate successively all running statuses of cutter are trained, obtain often kind of running status hidden-semi-Markov model.
3. operation reliability evaluation: after model training completes, according to the result of model parameter estimation, adopts Qie Puman-Andrei Kolmogorov differential equation, calculates Milling Process cutter operational reliability level.
Beneficial effect of the present invention has: the present invention is by gathering machining tool spindle motor current and feed shaft current signal, adopt wavelet analysis method to extract time domain and the frequency domain character information of above-mentioned signal, recycling continuous hidden Semi-Markov Process and Qie Puman-Andrei Kolmogorov differential equation calculate Milling Process cutter operational reliability level.This maintenance decision analysis being Milling Machining Center cutter provides important reference information.
Accompanying drawing explanation
Fig. 1 is flow chart of the present invention;
Fig. 2 is machine tool chief axis current signal time domain average time domain waveform in the present invention;
Fig. 3 is machine tool chief axis current signal frequency-domain waveform in the present invention;
Fig. 4 is the reliable linearity curve of Milling Process cutter of the present invention.
Detailed description of the invention
Below in conjunction with drawings and Examples, the present invention is described in more detail.First the inventive method gathers machining tool main shaft and feed shaft current signal by Hall element, then wavelet analysis method is adopted to carry out feature extraction, then parameter Estimation is carried out by continuous hidden Semi-Markov Process, and adopt Qie Puman-Andrei Kolmogorov differential equation to calculate Milling Process cutter operational reliability level, thus provide decision support for preventative maintenance.
As shown in Figure 1, the present invention includes following steps:
First step determination Milling Process cutter running status.
According to design data and the use history of Milling Process cutter, determine its running status sum M.Milling Process cutter running status set representations is S={s 1, s 2... s m... s m, s mfor Milling Process cutter is entirely ineffective.
Second step Milling Process cutter running status Characteristic Extraction.
Utilize Hall element to gather machining tool main shaft and feed shaft three-phase current signal, the group number of sampling interval and each collection signal can be determined according to the actual conditions of enterprise.Then the energy value of average in time domain and tooth frequency signal is therefrom extracted as characteristic quantity.Detailed process is: (1) frequency analysis, carries out wavelet decomposition to spindle motor of machine tool current signal, and the signal extracting tooth frequency content is reconstructed, and obtains tooth frequency curve, carries out the extraction of energy eigenvalue.(2) time-domain analysis: get feed shaft current signal three-phase current root-mean-square value and carry out wavelet analysis, extracts the characteristic quantity in signal time domain.
3rd step observation sample sequence probability density function extracts.
According to Milling Process cutter running status sum M, determine the gauss of distribution function number k corresponding to each Milling Process cutter running status m.For a certain running status of Milling Process cutter, the characteristic signal of above-mentioned collection is averaged segmentation, and the parameter belonging to a section in characteristic signal is formed a large matrix, K mean cluster is carried out to the vector of each section, calculates three key parameter c mk, and Σ mk, obtain the continuous mixed Gaussian probability density function corresponding to certain state of Milling Process cutter the like, obtain the continuous mixed Gaussian probability density function of all running statuses of Milling Process cutter.
4th step Milling Process estimating operation reliability of tool.
HSMM (HiddenSemi-MarkovModel, hidden-semi-Markov model) is the expansion of HMM (HiddenMarkovModel, hidden Markov model).Be the limitation of exponential distribution to improve state duration in hidden Markov model, on the basis of hidden Markov model, hidden-semi-Markov model allows according to the self-defined residence time distribution of practical problem.From the use history of Milling Process cutter, adopt Gamma distribution as state duration probability-distribution function.Meanwhile, observation sequence output function adopts Gaussian mixtures to carry out matching.
Step (a1): model training.First, initialize continuous hidden Semi-Markov Process parameter, wherein initial state distribution adopts and is uniformly distributed, and sets iterations and convergence error.Then adopt Baum-Welch algorithm to model training, obtain model parameter estimate successively all running statuses of cutter are trained, obtain the continuous hidden-semi-Markov model of often kind of running status.
Step (a2): Milling Process estimating operation reliability of tool.Make P j(t)=P (q t=s j) represent that being equipped in t is in s jshape probability of state.P'(t is had according to Qie Puman-Andrei Kolmogorov differential equation)=P (t) A, wherein P (t)=(p 0(t), P 1(t) ..., P k(t), P k+1(t)) be state vector, P'(t) be the first differential state vector of P (t), for state-transition matrix.Laplace transformation is carried out to the differential equation, can obtain: s · P 0 ( s ) P 1 ( s ) . . . P k ( s ) P k + 1 ( s ) - P 0 ( 0 ) P 1 ( 0 ) . . . P k ( 0 ) P k + 1 ( 0 ) = a ^ 00 0 ... 0 0 a ^ 01 a ^ 11 ... 0 0 . . . . . . . . . . . . a ^ 0 k a ^ 1 k ... a ^ k k 0 a ^ 0 ( k + 1 ) a ^ 1 ( k + 1 ) ... a ^ k ( k + 1 ) a ^ ( k + 1 ) ( k + 1 ) · P 0 ( s ) P 1 ( s ) . . . P k ( s ) P ( k + 1 ) ( s ) , Wherein Milling Process cutter is in normal condition when primary condition, can obtain P (0)=(p 0(0), P 1(0) ..., P k(0), P k+1(0))=(1,0,0,0).
There is P (s)=(p 0(s), P 1(s) ..., P k(s), P k+1(s)), Laplace Transform is carried out to P (s) and obtains being equipped in probability P (t)=(p that t is in different conditions 0(t), P 1(t) ..., P k(t), P k+1(t)), thus reliability R (the t)=1-P of Milling Process cutter t can be calculated k+1(t).
The present invention below in conjunction with object lesson makes further elaboration:
1: according to certain Milling Process Tool Design data and use history, determine that its running status adds up to 4, s 1for normal condition, s 2for minor degradation state, s 3for severe deterioration state, s 4for total failure mode.
2: this cutter is actual adds man-hour, cutting parameter is speed of mainshaft 130r/min, feed speed 140mm/min, cutting depth 1mm, and slotting cutter is processed, 6 teeth, sample frequency 1000Hz.Sampling interval is 48 hours, sampling total time T=960 hour.Wavelet analysis is adopted to carry out time domain and the extraction of frequency domain character amount.Fig. 2 is machine tool chief axis current signal time domain average time domain waveform; Fig. 3 is machine tool chief axis current signal frequency-domain waveform;
3: be equally divided into four sections to above-mentioned characteristic signal, corresponding 4 gauss of distribution function of each state, adopt K mean algorithm, obtain observation vector probability density function: mixed Gaussian density function.For saving space, here only list the some time and carve t, cutter is in the Gaussian mixtures function parameter corresponding to normal condition: weight parameter matrix is weight=[0.3450.2870.1290.239], Mean Matrix mean=[0.00990.00790.55680.3234], covariance matrix cov=[3.94483.86486.39457.4736].
4: Milling Process estimating operation reliability of tool, first to the given initial value of continuous hidden Semi-Markov Process CHSMM, if initial state distribution is for being uniformly distributed, cutter running status sum M=4.Then adopt Baum-Welch algorithm to model training, complete parameter Estimation, carry out Milling Process cutter operational reliability level calculation, Fig. 4 is milling cutter operational reliability horizontal line graph for this reason.
The present invention is not only confined to above-mentioned detailed description of the invention; persons skilled in the art are according to content disclosed by the invention; other multiple detailed description of the invention can be adopted to implement the present invention; therefore; every employing project organization of the present invention and thinking; do the design that some simply change or change, all fall into the scope of protection of the invention.

Claims (4)

1. an appraisal procedure for Milling Process cutter operational reliability, the steps include:
1) according to design data and the use history of Milling Process cutter, its running status sum M is determined; Milling Process cutter running status set representations is S={s 1, s 2... s m... s m, s mfor Milling Process cutter is entirely ineffective;
2) for milling cutter processing operating mode, record cutting parameter, utilizes sensor to gather machine tool chief axis and feed shaft three-phase current signal, and the average therefrom in extraction time domain and the energy value of tooth frequency signal are as characteristic quantity;
3) according to Milling Process cutter running status sum M, the gauss of distribution function number k corresponding to each Milling Process cutter running status is determined m, m is Milling Process cutter running status s msubscript, be used for representing s m; For a certain running status of Milling Process cutter, the characteristic signal of above-mentioned collection is averaged segmentation, and the parameter belonging to a section in characteristic signal is formed a large matrix, K mean cluster is carried out to the vector of each section, obtains the continuous mixed Gaussian probability density function corresponding to certain state of Milling Process cutter wherein: for t observation vector, t is the moment, k mfor state s mcorresponding gauss of distribution function number, for Gaussian Profile Mean Parameters, Σ mkfor Gaussian Profile covariance matrix, c mkstate s mthe weight of corresponding each Gaussian component; Repeat step 3), obtain the continuous mixed Gaussian probability density function of all running statuses of Milling Process cutter;
4) continuous hidden Semi-Markov Process is adopted to carry out reliability level estimation to cutter running status; First to the given initial value of continuous hidden Semi-Markov Process CHSMM, then adopt Baum-Welch algorithm to model training, complete parameter Estimation, carry out Milling Process cutter operational reliability level calculation;
Wherein, state duration model adopts Gamma distribution, namely v mfor form parameter, w mfor scale parameter; Observation model adopts Gaussian mixtures, namely
2. the appraisal procedure of a kind of Milling Process cutter operational reliability as claimed in claim 1, is characterized in that: step 2) comprising:
1) frequency analysis, carries out wavelet decomposition to spindle motor of machine tool current signal, and the signal extracting tooth frequency content is reconstructed, and obtains tooth frequency curve, carries out the extraction of energy eigenvalue;
2) time-domain analysis: get feed shaft current signal three-phase current root-mean-square value and carry out wavelet analysis, extracts the characteristic quantity in signal time domain.
3. the appraisal procedure of a kind of Milling Process cutter operational reliability as claimed in claim 1, is characterized in that: step 2) described in sensor be Hall element.
4. the appraisal procedure of a kind of Milling Process cutter operational reliability as described in claim arbitrary in claim 1-3, is characterized in that: step 4) detailed process as follows:
1) model definition: being expressed as of continuous hidden-semi-Markov model wherein, the quantity of hidden state and cutter running status is M, initial state distribution π m=(π 1..., π m); State transition probability matrix a mnrepresent that cutter is from motion state s mjump to motion state s nprobability, c mkstate s mthe weight of corresponding each Gaussian component, for Gaussian Profile Mean Parameters, Σ mkfor Gaussian Profile covariance matrix, v mfor form parameter, w mfor scale parameter;
2) model training: adopt Bao Mu-Wei Erqi (Baum-Welch) algorithm to model training, namely solve the Parameter Estimation Problem of model, obtain model parameter estimate successively all running statuses of cutter are trained, obtain often kind of running status hidden-semi-Markov model;
3) operation reliability evaluation: after model training completes, according to the result of model parameter estimation, adopts Qie Puman-Andrei Kolmogorov differential equation, calculates Milling Process cutter operational reliability level.
CN201310181390.2A 2013-05-16 2013-05-16 A kind of appraisal procedure of Milling Process cutter operational reliability Active CN103264317B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201310181390.2A CN103264317B (en) 2013-05-16 2013-05-16 A kind of appraisal procedure of Milling Process cutter operational reliability

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201310181390.2A CN103264317B (en) 2013-05-16 2013-05-16 A kind of appraisal procedure of Milling Process cutter operational reliability

Publications (2)

Publication Number Publication Date
CN103264317A CN103264317A (en) 2013-08-28
CN103264317B true CN103264317B (en) 2015-11-18

Family

ID=49008003

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201310181390.2A Active CN103264317B (en) 2013-05-16 2013-05-16 A kind of appraisal procedure of Milling Process cutter operational reliability

Country Status (1)

Country Link
CN (1) CN103264317B (en)

Families Citing this family (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104191301B (en) * 2014-08-02 2016-06-29 金华中科机电研究所 Anti-jamming control system and method thereof for common driller
CN105160147B (en) * 2015-07-07 2017-11-10 大连理工大学 A kind of cutter changing time decision-making technique based on state-space model and fuzzy theory
CN106002483B (en) * 2016-05-04 2018-02-02 北京信息科技大学 A kind of intelligent tool method for diagnosing faults
CN106181579B (en) * 2016-08-23 2018-04-17 西安交通大学 A kind of Tool Wear Monitoring method based on multisensor current signal
CN107877262B (en) * 2017-11-13 2019-06-11 华中科技大学 A kind of numerical control machine tool wear monitoring method based on deep learning
CN108122049A (en) * 2017-12-21 2018-06-05 郓城金河热电有限责任公司 Photovoltaic array operation and maintenance method based on the analysis of photovoltaic module failure rate
CN108620950B (en) * 2018-05-08 2021-02-02 华中科技大学无锡研究院 Method and system for monitoring machining state of turning tool
CN108956783B (en) * 2018-05-18 2020-04-21 南京大学 HDP-HSMM-based grinding sound grinding wheel passivation state detection method
CN109262369B (en) * 2018-09-13 2020-02-21 成都数之联科技有限公司 Cutter state detection system and method
CN109396956B (en) * 2018-11-06 2020-04-07 重庆大学 Intelligent monitoring method for hob state of numerical control gear hobbing machine
CN111085898A (en) * 2019-12-30 2020-05-01 南京航空航天大学 Working condition self-adaptive high-speed milling process cutter monitoring method and system
TWI749742B (en) * 2020-08-31 2021-12-11 國立虎尾科技大學 Machine tool spindle diagnosis method
CN112636642B (en) * 2020-12-17 2023-03-24 广东工业大学 Method and device for evaluating performance state of numerical control cutting tool bit made of flexible material

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE102008064391A1 (en) * 2007-12-27 2009-07-02 Okuma Corporation Machine diagnostic method and apparatus
CN101799674A (en) * 2010-03-28 2010-08-11 华中科技大学 Method for analyzing service state of numerical control equipment
CN101804580A (en) * 2010-03-29 2010-08-18 华中科技大学 Method for assessing reliability of technology for large numerical control machine
CN102176217A (en) * 2010-12-20 2011-09-07 西安瑞特快速制造工程研究有限公司 Method for estimating reliability of numerical control machine tool cutting tool based on logistic model
CN102254100A (en) * 2011-07-13 2011-11-23 西安工业大学 Proportional hazard rate model method for estimating operation reliability of tool
CN102689230A (en) * 2012-05-09 2012-09-26 天津大学 Tool wear condition monitoring method based on conditional random field model

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE102008064391A1 (en) * 2007-12-27 2009-07-02 Okuma Corporation Machine diagnostic method and apparatus
CN101799674A (en) * 2010-03-28 2010-08-11 华中科技大学 Method for analyzing service state of numerical control equipment
CN101804580A (en) * 2010-03-29 2010-08-18 华中科技大学 Method for assessing reliability of technology for large numerical control machine
CN102176217A (en) * 2010-12-20 2011-09-07 西安瑞特快速制造工程研究有限公司 Method for estimating reliability of numerical control machine tool cutting tool based on logistic model
CN102254100A (en) * 2011-07-13 2011-11-23 西安工业大学 Proportional hazard rate model method for estimating operation reliability of tool
CN102689230A (en) * 2012-05-09 2012-09-26 天津大学 Tool wear condition monitoring method based on conditional random field model

Also Published As

Publication number Publication date
CN103264317A (en) 2013-08-28

Similar Documents

Publication Publication Date Title
CN103264317B (en) A kind of appraisal procedure of Milling Process cutter operational reliability
CN103105820B (en) Machining cutter abrasion state identification method of numerical control machine tool
CN102284888B (en) Online monitoring method for turning stability of digital control machine tool
CN111582392B (en) Multi-working-condition health state online monitoring method for key components of wind turbine generator
CN102152172B (en) Cutter abrasion monitoring method based on co-integration modeling
CN109396953A (en) Lathe work condition intelligent identification system based on signal fused
CN101799674B (en) Method for analyzing service state of numerical control equipment
CN103646167A (en) Satellite abnormal condition detection system based on telemeasuring data
CN102495949A (en) Fault prediction method based on air data
CN113177537B (en) Fault diagnosis method and system for rotary mechanical equipment
CN102176217A (en) Method for estimating reliability of numerical control machine tool cutting tool based on logistic model
CN112883075B (en) Landslide universal type ground surface displacement monitoring data missing and outlier processing method
CN103345200A (en) Cutting flutter identification method based on generalized interval
CN112207631B (en) Method for generating tool detection model, method, system, device and medium for detecting tool detection model
CN114871850B (en) Tool wear state assessment method based on vibration signals and BP neural network
CN116893643A (en) Intelligent robot driving track safety control system based on data analysis
CN115741235A (en) Wear prediction and health management method based on five-axis machining center cutter
CN109580224A (en) Rolling bearing fault method of real-time
CN116690313B (en) Failure monitoring method for machining cutter of web plate of aircraft structural member
CN111993158B (en) Method for generating tool detection model, method, system, device and medium for detecting tool detection model
CN116204781A (en) Rotary machine fault migration diagnosis method and system
CN106881630B (en) High-speed milling flutter online recognition method based on adaptive-filtering Yu AR models
CN114638265B (en) Milling chatter judging method based on signal convolution neural network
CN114279704B (en) Method and system for monitoring planet-level running state of wind power gear box
Cong et al. Online milling chatter detection using deep residual network and transfer learning

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
TR01 Transfer of patent right

Effective date of registration: 20201224

Address after: 224400 No.10 Chunyuan Road, Xingou Town, Funing County, Yancheng City, Jiangsu Province

Patentee after: Funing Jinyu Plastics Co.,Ltd.

Address before: 411201 No. 2 stone wharf, Yuhu District, Hunan, Xiangtan

Patentee before: Hunan University of Science and Technology

TR01 Transfer of patent right