CN109333159A - The depth core extreme learning machine method and system of cutting-tool wear state on-line monitoring - Google Patents
The depth core extreme learning machine method and system of cutting-tool wear state on-line monitoring Download PDFInfo
- Publication number
- CN109333159A CN109333159A CN201811056778.9A CN201811056778A CN109333159A CN 109333159 A CN109333159 A CN 109333159A CN 201811056778 A CN201811056778 A CN 201811056778A CN 109333159 A CN109333159 A CN 109333159A
- Authority
- CN
- China
- Prior art keywords
- depth
- cutting
- signal
- tool wear
- 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.)
- Granted
Links
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B23—MACHINE TOOLS; METAL-WORKING NOT OTHERWISE PROVIDED FOR
- B23Q—DETAILS, COMPONENTS, OR ACCESSORIES FOR MACHINE TOOLS, e.g. ARRANGEMENTS FOR COPYING OR CONTROLLING; MACHINE TOOLS IN GENERAL CHARACTERISED BY THE CONSTRUCTION OF PARTICULAR DETAILS OR COMPONENTS; COMBINATIONS OR ASSOCIATIONS OF METAL-WORKING MACHINES, NOT DIRECTED TO A PARTICULAR RESULT
- B23Q17/00—Arrangements for observing, indicating or measuring on machine tools
- B23Q17/09—Arrangements for observing, indicating or measuring on machine tools for indicating or measuring cutting pressure or for determining cutting-tool condition, e.g. cutting ability, load on tool
- B23Q17/0952—Arrangements for observing, indicating or measuring on machine tools for indicating or measuring cutting pressure or for determining cutting-tool condition, e.g. cutting ability, load on tool during machining
- B23Q17/0957—Detection of tool breakage
Abstract
The invention discloses a kind of depth core extreme learning machine method and system of cutting-tool wear state on-line monitoring, acquire multi channel signals of the cutter under different state of wear by multiple sensors first;Signal is transported in DSP secondly by the port RS232 and carries out signal processing, calculates its multiple statistical nature parameter;Sample signal is mapped to depth nuclear space by the depth kernel function for then constructing data-driven, is formed training sample and is stored in RAM;When final online monitors, calculate the feature of collected multi channel signals, and be mapped to depth nuclear space and form test sample and be input in extreme learning machine simultaneously with the training sample being stored in RAM and classify, state of wear locating for current demand signal is exported, realizes cutting-tool wear state on-line monitoring.
Description
Technical field
The present invention relates to manufacturing processes to monitor field, in particular to a kind of depth core pole of cutting-tool wear state on-line monitoring
Limit learning machine method and system.
Background technique
With the continuous development of modern manufacturing industry, the degree of automation of manufacture system directly affects enterprise's productivity effect,
And the degree of automation of numerically-controlled machine tool is the component part of most manufacture systems.The cutter component most easily damaged as numerically-controlled machine tool
One of, the degree of wear and product quality and production efficiency are closely related.According to statistics, lathe is caused to be shut down due to tool failure
Account for nearly the 20% of lathe total down-time.Therefore, real-time monitoring is carried out to the state of wear of cutter, so as to the shape that notes abnormalities ahead of time
State is very valuable and realistic meaning.
It is difficult to realize monitor on-line due to measuring cutting-tool wear state by the direct method of measurement, therefore more and more scholars are special
Infuse the research in the indirect method of measurement.In product processing, single sensor acquires information related with tool wear
With limitation.It in order to enhance the reliability of sensor information under various regimes, reduces uncertain, it is necessary to more sensings
Device replaces single sensor.However, the kernel function that different physical field signal is suitble to is not when being classified with classifier
Together, mapping and unreasonable is carried out using single simple core.In recent years, the research about Multiple Kernel Learning method is more and more, including
It is layered kernel learning method (HKL), the multicore built-up pattern learning method based on Boosting, is based on the more of Semidefinite Programming (SDP)
Kernel learning method etc..Depth kernel function of the present invention by construction data-driven, the available optimal kernel function group of data source
It closes.
Summary of the invention
The present invention in view of the above shortcomings of the prior art, provides a kind of depth core pole of cutting-tool wear state on-line monitoring
Limit learning machine method and system.
In order to achieve the above object, the present invention provides a kind of depth core limit study of cutting-tool wear state on-line monitoring
Machine method, the following steps are included:
S1, model training;Specific step is as follows:
S11, the relevant signal of multiple channel cutting-tool wear states is acquired by sensor;
S12, pass through time and frequency domain analysis, its multiple statistical nature is calculated according to the collected signal of institute in step S11
Parameter;
S13, the depth kernel function for constructing data-driven;
S14, each data are mapped to by depth core k by the depth kernel function constructed in step S13MKLSpace forms instruction
Practice sample set Xtr;
S2, on-line monitoring;Specific step is as follows:
S21, the relevant signal of multiple channel cutting-tool wear states is acquired in real time by sensor;
S22, pass through time and frequency domain analysis, its multiple statistical nature is calculated according to the collected signal of institute in step S21
Parameter;
S23, each data are mapped to the data-driven depth core k constructed in model trainingMKLSpace forms test sample
Collect Xte;
S24, by training sample set Xtr, test sample collection XteAnd cutting-tool wear state is input in extreme learning machine, is realized
Cutting-tool wear state classification.
Further setting is that the step S11 is specifically included:
Acquire multi channel signals (X of the cutter in different state of wearj,Yj), J class physical field is shared, with jth class physics
For, time-domain signal mathematical form is
Xj=xij(n)
In above formula, Xj={ x1j,x2j,…,xmj}T∈Rm×nIt is m × n rank original signal sample matrix of acquisition, wherein n is
The sampling site number of signal, i=1,2 ..., m are signal acquisition number;
Yj={ y1j,y2j,…,ylj}T∈RlState of wear corresponding to signal is acquired for i-th, wherein yij∈{c1,
c2,…,cpRepresent cutter difference state of wear.
Further setting is that the step S13 is specifically included:
S131, the depth kernel function for constructing data-driven
η in above formulasFor weight shared by s-th of kernel function, α, β={ 1,2 ..., m };
S132, to weight ηsIt optimizes;
S133, optimized after depth kernel function
Further setting is that the step S132 is specifically included:
To weight ηsIt optimizes, obtains ηs*;
Condition distributionWherein γiIndicate cpKind tool wear
The conditional probability of state;
Optimize γ,
In above formula,Λ is regularization parameter, | | γ | |2Indicate γ's
2 norms;
Optimize ηs*,
In above formula, ds(γ *)=γ *TYKsYγ*,s∈{0,1,…,R}。
The present invention also provides a kind of depth core extreme learning machine systems of cutting-tool wear state on-line monitoring, include:
Model training unit acquires the associated analog signal of m cutter by multiclass sensor, defeated by RS232 interface
It send into DSP development board;
Statistical nature parameter extraction unit merges multiclass sensor feature, generates characteristic signal
Data-driven depth nuclear unit is constructed, the depth kernel function of data-driven is constructed, it is empty to map the data into depth core
Between, training sample is formed, and be stored in RAM;
It monitors unit on-line, acquires M tool wear coherent signal, utilize the depth core letter of the data-driven of RAM storage
Number maps the data into nuclear space and forms test sample;Training sample and test sample are directly inputted to core extreme learning machine
In, classification processing is carried out in dsp, obtains cutter state in which, and output display on a display screen;When cutting-tool wear state is aobvious
When being shown as severe abrasion, display screen is given a warning, and prompts the timely tool changing of operator.
Further setting is that the on-line monitoring unit is built by TL-6748F-EVM.
The beneficial effects of the present invention are:
One aspect of the present invention is classified using the extreme learning machine of base core, is replaced eigenmatrix operation with kernel function, is changed
It has been apt to the performance of classifier.On the other hand the data-driven depth kernel function suitable for isomeric data is constructed, use is avoided
Influence of the single kernel function to classification accuracy, improves the accuracy rate of classification.
Detailed description of the invention
Fig. 1 is a kind of process of the depth core extreme learning machine method of cutting-tool wear state on-line monitoring provided by the invention
Figure;
Fig. 2 is a kind of frame of the depth core extreme learning machine system of cutting-tool wear state on-line monitoring provided by the invention
Figure;
Fig. 3 is the simplified block diagram of digital information processing system in the present invention;
Fig. 4 is that cutting-tool wear state monitors flow chart on-line in the present invention;
Fig. 5 is testing classification result schematic diagram in the present invention;
Fig. 6 is trained and test result analysis schematic diagram in the present invention.
Specific embodiment
The invention will be described in further detail with reference to the accompanying drawing:
As shown in Figure 1, in the embodiment of the present invention, a kind of depth core limit of cutting-tool wear state on-line monitoring of proposition
Learning machine method, the following steps are included:
S1, model training;Specific step is as follows:
S11, the relevant signal of multiple channel cutting-tool wear states is acquired by sensor.It mainly includes:
Acquire multi-pass of the cutter at different state of wear (normal, mild wear, moderate abrasion, severe abrasion, breakage)
Road signal (Xj,Yj), share J class physical field, by taking jth class physical field as an example, time-domain signal mathematical form are as follows:
Xj=xij(n) (1)
In formula (1), Xj={ x1j,x2j,…,xmj}T∈Rm×nIt is m × n rank original signal sample matrix of acquisition, in which: n is
The sampling site number of signal, i=1,2 ..., m are signal acquisition number.
Yj={ y1j,y2j,…,ylj}T∈RlState of wear corresponding to signal is acquired for i-th, wherein yij∈{c1,
c2,…,cpRepresent cutter difference state of wear.
S12, pass through time and frequency domain analysis, its multiple statistical nature is calculated according to the collected signal of institute in step S11
Parameter.It mainly includes:
Wherein, i=1,2 ..., m represent the signal of i-th acquisition,For the time-domain signal x of i-th acquisitioni(n) equal
Value, siIt (k) is the corresponding frequency spectrum of time-domain signal of i-th acquisition, k=1,2 ..., K represent spectral line number, ui(k) it is adopted for i-th
The corresponding power spectrum of the time-domain signal of collection.
Generate characteristic signal group:
Wherein, l is the sum of single physical field signal complete characterization signal collection, and the feature of each physical field is merged,
The then feature of whole physical fields are as follows:
S13, the depth kernel function for constructing data-driven.It mainly includes:
a)
η in formula (3)sFor weight shared by s-th of kernel function, α, β={ 1,2 ..., m }.
B) to weight ηsIt optimizes, obtains ηs*
Condition distributionWherein γiIndicate cpKind tool wear
The conditional probability of state.
Optimize γ:
In formula (4),Λ is regularization parameter, | | γ | |2Indicate γ
2 norms.
Optimize ηs*:
D in formula (5)s(γ *)=γ *TYKsYγ*,s∈{0,1,…,R}。
C) the depth kernel function after being optimized
S14, each data are mapped to by depth core k by the depth kernel function constructed in step S13MKLSpace forms instruction
Practice sample set Xtr。
S2, on-line monitoring;Specific step is as follows:
S21, the relevant signal of multiple channel cutting-tool wear states is acquired in real time by sensor.It mainly includes:
Acquire different physical field signal of the cutter in processing, by taking jth class physical field as an example, time-domain signal mathematical form
Are as follows:
Xj'=xij′(n) (6)
In formula (6), Xj'={ x1j′,x2j′,…,xMj′}T∈RM×nIt is M × n rank original signal sample matrix of acquisition,
In: n is the sampling site number of signal, and i=1,2 ..., M are signal acquisition number.
Yj'={ y1j′,y2j′,…,yMj′}T∈RMState of wear corresponding to signal is acquired for i-th, wherein yij∈
{c1,c2,…,cp}。
S22, pass through time and frequency domain analysis, its multiple statistical nature is calculated according to the collected signal of institute in step S21
Parameter.It mainly includes:
Calculate separately different physical field time-frequency domain statistical nature parameter obtained in training pattern, and by different physical field
Feature merged, generate characteristic signal group:
X_T '={ x_t1′(i),x_t2′(i),…,x_tL′(i)} (7)
Wherein, i=1,2 ..., M, L=l × J are the sum of all features.
S23, each data are mapped to the data-driven depth core k constructed in model trainingMKLSpace forms test sample
Collect Xte。
S24, by training sample set Xtr, test sample collection XteAnd cutting-tool wear state is input in extreme learning machine, is realized
Cutting-tool wear state classification.It mainly includes:
A) algorithm inputs: { [Y, Xtr],[Y′,Xte],Λ}
B) depth core extreme learning machine is classified:
B-1 training set label matrix Y) is createdtr={ Ytr|Ytr(i,yi)=1, remaining Ytr=-1 }.Test label matrix Yte
={ Yte|Yte(i,y'i)=1, remaining Yte=-1 }.
B-2) core extreme learning machine training weight outputWherein E1For unit diagonal matrix.
B-3) training resultTest result
Wherein, For the sorted state of depth core extreme learning machine.
B-4) whenWhen classification it is correct.
C) algorithm exports: classification results and classification accuracy rate.
As shown in Fig. 2, in the embodiment of the present invention, a kind of depth core limit of cutting-tool wear state on-line monitoring provided
Learning machine system, comprising:
Model training unit 101 mainly includes:
The associated analog signal that m cutter is acquired by multiclass sensor, is delivered to DSP development board by RS232 interface
In, the simplified block diagram of digital information processing system is as shown in Figure 3.
Statistical nature parameter extraction unit 102 mainly includes:
Calculate the time domain index of time-domain signal x (n):
f1 i:
f2 i:
f3 i:
f4 i:
f5 i:
FFT transform is carried out to time-domain signal x (n) and obtains frequency spectrum s (k) and power spectrum u (k), calculates its frequency-domain index:
f6 i:
f7 i:
f8 i:
f9 i:
f10 i:
Multiclass sensor feature is merged, characteristic signal is generated
Data-driven depth nuclear unit 103 is constructed, mainly includes:
The depth kernel function for constructing data-driven maps the data into depth nuclear space, forms training sample, and be stored in
In RAM.
It monitors unit 104 on-line, mainly includes:
A) M tool wear coherent signal is acquired, using the depth kernel function of the data-driven of RAM storage, data are reflected
It is mapped to nuclear space and forms test sample.
B) training sample and test sample are directly inputted in core extreme learning machine, carry out classification processing in dsp, obtains
Cutter state in which, and output display on a display screen.
C) when cutting-tool wear state is shown as severe abrasion, display screen is given a warning, and prompts the timely tool changing of operator.
Further setting is that the on-line monitoring unit is built by TL-6748F-EVM development board, cutting-tool wear state
Flow chart is monitored on-line as shown in figure 4, TL-6748F-EVM development board has display screen and buzzer.
As shown in Figure 5 and Figure 6, using the tool wear data of PHM association data challenge in 2010 as acquisition signal, according to
The degree of wear of cutter is different, and being responded output Y to divide is that five classes (severe abrasion, break by normal, mild wear, moderate abrasion
Damage), i.e. Y={ y1,y2,…,ym, yi∈ { 1,2,3,4,5 }, time-domain signal mathematical form are as follows:
Xj=xij(n) (8)
In formula (8), X={ x1,x2,…,xm}T∈Rm×nIt is m × n rank original signal sample matrix of acquisition, in which: n is letter
Number sampling site number, i is times of collection, and j is physical field classification (including three-dimensional cutting force, three-way vibration, sound emission root mean square).
Signal is acquired 945 times altogether with the acquisition of the channel 50kHz/, and sampling site number is 10000, randomly selects 645 sample conducts
Training sample, remaining 200 acquisition signals are as test sample.
10 statistical nature parameters of sample signal are calculated, and the feature of 7 class different physical field signals is merged, are generated
Trained and test feature signal group:
X_Ttr={ x_ttr1(i'),x_ttr2(i'),…,x_ttr70(i')} (9)
X_Tte={ x_tte1(ii),x_tte2(ii),…,x_tte70(ii)} (10)
{ 1,2 ..., 645 } i'=in formula (9), (10), ii={ 1,2 ..., 200 }.
Condition distributionWherein γi′Indicate that cutting-tool wear state is
cpConditional probability.
The depth kernel function of data-driven is constructed, optimal conditions probability γ first:
In formula (11),KsFor Gaussian kernel, linear kernel, polynomial kernel
And their Polynomial combination core, Λ=1 are regularization parameter, | | γ | |2Indicate 2 norms of γ.
Then optimize core weight ηs*:
D in formula (12)s(γ *)=γ *TYKsYγ*,s∈{0,1,…,R}。
Obtain data-driven depth kernel functionSample data set is mapped to depth core
kMKLSpace is respectively formed training sample set XtrWith test sample Xte, and be input in core extreme learning machine, classify.Root
It is compared according to the classification results and cutter actual wear state of core extreme learning machine, cutting-tool wear state classification can be obtained just
True rate.Its testing classification result and actual result compare as shown in figure 5, repeating 20 times tests its training, test accuracy
As shown in Figure 6.
The above disclosure is only the preferred embodiments of the present invention, cannot limit the right model of the present invention with this certainly
It encloses, therefore equivalent changes made in accordance with the claims of the present invention, is still within the scope of the present invention.
Claims (6)
1. a kind of depth core extreme learning machine method of cutting-tool wear state on-line monitoring, which comprises the following steps:
S1, model training;Specific step is as follows:
S11, the relevant signal of multiple channel cutting-tool wear states is acquired by sensor;
S12, pass through time and frequency domain analysis, its multiple statistical nature parameter is calculated according to the collected signal of institute in step S11;
S13, the depth kernel function for constructing data-driven;
S14, each data are mapped to by depth core k by the depth kernel function constructed in step S13MKLSpace forms training sample
This collection Xtr;
S2, on-line monitoring;Specific step is as follows:
S21, the relevant signal of multiple channel cutting-tool wear states is acquired in real time by sensor;
S22, pass through time and frequency domain analysis, its multiple statistical nature parameter is calculated according to the collected signal of institute in step S21;
S23, each data are mapped to the data-driven depth core k constructed in model trainingMKLSpace forms test sample collection Xte;
S24, by training sample set Xtr, test sample collection XteAnd cutting-tool wear state is input in extreme learning machine, realizes cutter
State of wear classification.
2. the depth core extreme learning machine method of cutting-tool wear state on-line monitoring according to claim 1, feature exist
In the step S11 is specifically included:
Acquire multi channel signals (X of the cutter in different state of wearj,Yj), J class physical field is shared, is with jth class physical field
Example, time-domain signal mathematical form are
Xj=xij(n)
In above formula, Xj={ x1j,x2j,…,xmj}T∈Rm×nIt is m × n rank original signal sample matrix of acquisition, wherein n is signal
Sampling site number, i=1,2 ..., m are signal acquisition number;
Yj={ y1j,y2j,…,ylj}T∈RlState of wear corresponding to signal is acquired for i-th, wherein yij∈{c1,c2,…,
cpRepresent cutter difference state of wear.
3. the depth core extreme learning machine method of cutting-tool wear state on-line monitoring according to claim 1, feature exist
In the step S13 is specifically included:
S131, the depth kernel function for constructing data-driven
η in above formulasFor weight shared by s-th of kernel function, α, β={ 1,2 ..., m };
S132, to weight ηsIt optimizes;
S133, optimized after depth kernel function
4. the depth core extreme learning machine method of cutting-tool wear state on-line monitoring according to claim 3, feature exist
In the step S132 is specifically included:
To weight ηsIt optimizes, obtains ηs*;
Condition distributionWherein γiIndicate cpKind cutting-tool wear state
Conditional probability;
Optimize γ,
In above formula,Λ is regularization parameter, | | γ | |2Indicate 2 models of γ
Number;
Optimize ηs*,
In above formula,s∈{0,1,…,R}。
5. a kind of depth core extreme learning machine system of cutting-tool wear state on-line monitoring characterized by comprising
Model training unit is acquired the associated analog signal of m cutter by multiclass sensor, is delivered to by RS232 interface
In DSP development board;
Statistical nature parameter extraction unit merges multiclass sensor feature, generates characteristic signal
Data-driven depth nuclear unit is constructed, the depth kernel function of data-driven is constructed, maps the data into depth nuclear space, shape
At training sample, and it is stored in RAM;
It monitors unit on-line, acquires M tool wear coherent signal, it, will using the depth kernel function of the data-driven of RAM storage
Data are mapped to nuclear space and form test sample;Training sample and test sample are directly inputted in core extreme learning machine,
Classification processing is carried out in DSP, obtains cutter state in which, and output display on a display screen;When cutting-tool wear state is shown as
When severe is worn, display screen is given a warning, and prompts the timely tool changing of operator.
6. the depth core extreme learning machine system of cutting-tool wear state on-line monitoring according to claim 5, feature exist
In: the on-line monitoring unit is built by TL-6748F-EVM.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811056778.9A CN109333159B (en) | 2018-09-11 | 2018-09-11 | Depth kernel extreme learning machine method and system for online monitoring of tool wear state |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811056778.9A CN109333159B (en) | 2018-09-11 | 2018-09-11 | Depth kernel extreme learning machine method and system for online monitoring of tool wear state |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109333159A true CN109333159A (en) | 2019-02-15 |
CN109333159B CN109333159B (en) | 2021-04-13 |
Family
ID=65305301
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811056778.9A Active CN109333159B (en) | 2018-09-11 | 2018-09-11 | Depth kernel extreme learning machine method and system for online monitoring of tool wear state |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109333159B (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110057707A (en) * | 2019-05-31 | 2019-07-26 | 上海交通大学 | Carbon fibre reinforced composite/titanium alloy lamination drilling cutters biometrics method |
CN110091216A (en) * | 2019-05-13 | 2019-08-06 | 江苏师范大学 | The monitoring of milling noise and milling vibration and its correlation analysis system and method |
CN114800039A (en) * | 2022-04-08 | 2022-07-29 | 山东大学 | Characteristic strengthening method and system for on-line monitoring of state of milling cutter of thin-wall part |
CN117001423A (en) * | 2023-09-28 | 2023-11-07 | 智能制造龙城实验室 | Tool state online monitoring method based on evolutionary learning |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH01177950A (en) * | 1988-01-07 | 1989-07-14 | Nissan Motor Co Ltd | Method for detecting wear of cutting tool |
US4942387A (en) * | 1989-09-05 | 1990-07-17 | General Electric Company | Apparatus and method for tool break and wear detection |
CN102073300A (en) * | 2010-12-28 | 2011-05-25 | 华中科技大学 | Tool wear monitoring system capable of realizing self-learning in numerical control machining state |
CN105834835A (en) * | 2016-04-26 | 2016-08-10 | 天津大学 | Method for monitoring tool wear on line based on multiscale principal component analysis |
CN107345884A (en) * | 2017-06-03 | 2017-11-14 | 东北电力大学 | A kind of tunnel piercing rounding machine cutter on-Line Monitor Device and its method |
CN108107838A (en) * | 2017-12-27 | 2018-06-01 | 山东大学 | A kind of numerical control equipment tool wear monitoring method based on cloud knowledge base and machine learning |
CN108393744A (en) * | 2018-04-11 | 2018-08-14 | 温州大学 | A kind of more sensor monitoring methods of cutting tool state |
-
2018
- 2018-09-11 CN CN201811056778.9A patent/CN109333159B/en active Active
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH01177950A (en) * | 1988-01-07 | 1989-07-14 | Nissan Motor Co Ltd | Method for detecting wear of cutting tool |
US4942387A (en) * | 1989-09-05 | 1990-07-17 | General Electric Company | Apparatus and method for tool break and wear detection |
CN102073300A (en) * | 2010-12-28 | 2011-05-25 | 华中科技大学 | Tool wear monitoring system capable of realizing self-learning in numerical control machining state |
CN105834835A (en) * | 2016-04-26 | 2016-08-10 | 天津大学 | Method for monitoring tool wear on line based on multiscale principal component analysis |
CN107345884A (en) * | 2017-06-03 | 2017-11-14 | 东北电力大学 | A kind of tunnel piercing rounding machine cutter on-Line Monitor Device and its method |
CN108107838A (en) * | 2017-12-27 | 2018-06-01 | 山东大学 | A kind of numerical control equipment tool wear monitoring method based on cloud knowledge base and machine learning |
CN108393744A (en) * | 2018-04-11 | 2018-08-14 | 温州大学 | A kind of more sensor monitoring methods of cutting tool state |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110091216A (en) * | 2019-05-13 | 2019-08-06 | 江苏师范大学 | The monitoring of milling noise and milling vibration and its correlation analysis system and method |
CN110091216B (en) * | 2019-05-13 | 2021-06-01 | 江苏师范大学 | Milling noise and milling vibration monitoring and correlation analysis system and method |
CN110057707A (en) * | 2019-05-31 | 2019-07-26 | 上海交通大学 | Carbon fibre reinforced composite/titanium alloy lamination drilling cutters biometrics method |
CN110057707B (en) * | 2019-05-31 | 2021-06-25 | 上海交通大学 | Method for measuring service life of carbon fiber reinforced composite/titanium alloy laminated drilling tool |
CN114800039A (en) * | 2022-04-08 | 2022-07-29 | 山东大学 | Characteristic strengthening method and system for on-line monitoring of state of milling cutter of thin-wall part |
CN114800039B (en) * | 2022-04-08 | 2023-03-14 | 山东大学 | Characteristic strengthening method and system for on-line monitoring of state of milling cutter of thin-wall part |
CN117001423A (en) * | 2023-09-28 | 2023-11-07 | 智能制造龙城实验室 | Tool state online monitoring method based on evolutionary learning |
CN117001423B (en) * | 2023-09-28 | 2023-12-05 | 智能制造龙城实验室 | Tool state online monitoring method based on evolutionary learning |
Also Published As
Publication number | Publication date |
---|---|
CN109333159B (en) | 2021-04-13 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109158953B (en) | A kind of cutting-tool wear state on-line monitoring method and system | |
CN107584334B (en) | A kind of complex structural member numerical control machining cutter status real time monitor method based on deep learning | |
CN109333159A (en) | The depth core extreme learning machine method and system of cutting-tool wear state on-line monitoring | |
CN104407589B (en) | Workshop manufacturing process-oriented active sensing and anomaly analysis method of real-time production performance | |
CN106197999B (en) | A kind of planetary gear method for diagnosing faults | |
CN111687689A (en) | Cutter wear state prediction method and device based on LSTM and CNN | |
CN112692646B (en) | Intelligent assessment method and device for tool wear state | |
CN108399248A (en) | A kind of time series data prediction technique, device and equipment | |
CN110561191B (en) | Numerical control machine tool cutter abrasion data processing method based on PCA and self-encoder | |
CN114619292B (en) | Milling cutter wear monitoring method based on fusion of wavelet denoising and attention mechanism with GRU network | |
CN106842922A (en) | A kind of NC Machining Error optimization method | |
CN108581633A (en) | A method of based on the more sensor monitoring cutting tool states of genetic algorithm optimization | |
CN112541510A (en) | Intelligent fault diagnosis method based on multi-channel time series data | |
CN110533007A (en) | A kind of vehicle-mounted strain of bridge influences the identification of line feature intelligent and extracting method | |
CN108931621B (en) | Zinc ore grade soft measurement method based on process texture characteristics | |
CN115438726A (en) | Device life and fault type prediction method and system based on digital twin technology | |
CN111783616B (en) | Nondestructive testing method based on data-driven self-learning | |
CN109508753A (en) | A kind of on-line prediction method of Mineral Floating Process index | |
CN108647722B (en) | Zinc ore grade soft measurement method based on process size characteristics | |
CN110346130A (en) | A kind of boring flutter detection method based on empirical mode decomposition and time-frequency multiple features | |
de la Cruz et al. | Convolutional neural networks for the Hass avocado classification using LabVIEW in an agro-industrial plant | |
CN114714145A (en) | Method for enhancing, comparing, learning and monitoring tool wear state by using Gelam angular field | |
CN109389039A (en) | A kind of optimization core extreme learning machine on-line monitoring cutting-tool wear state method and system | |
CN116787225A (en) | Rotary cutter abrasion monitoring method based on sound vibration signal characteristic fusion graph | |
CN109886288A (en) | A kind of method for evaluating state and device for power transformer |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |