CN110561192A - Deep hole boring cutter state monitoring method based on stacking self-encoder - Google Patents
Deep hole boring cutter state monitoring method based on stacking self-encoder Download PDFInfo
- Publication number
- CN110561192A CN110561192A CN201910855352.8A CN201910855352A CN110561192A CN 110561192 A CN110561192 A CN 110561192A CN 201910855352 A CN201910855352 A CN 201910855352A CN 110561192 A CN110561192 A CN 110561192A
- Authority
- CN
- China
- Prior art keywords
- encoder
- self
- cutter
- data
- training
- 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.)
- Withdrawn
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
Landscapes
- Engineering & Computer Science (AREA)
- Mechanical Engineering (AREA)
- Automatic Control Of Machine Tools (AREA)
Abstract
the invention provides a deep hole boring cutter state monitoring method based on a stacked self-encoder, and belongs to the technical field of cutter state monitoring. The method comprises the following steps that firstly, two three-way acceleration sensors are respectively adsorbed outside two retainer bearing bushes of a deep hole boring bar through magnetic seats, a microphone is placed at a processing inlet of an inner hole of a workpiece, and vibration and acoustic signals in the boring process are collected; then, carrying out data preprocessing on the acquired data by adopting an amplitude limiting value filtering method; then, constructing a stacking self-encoder network, and training the stacking self-encoder by using the preprocessed data by adopting a greedy layer-by-layer method; and finally, inputting real-time vibration and sound signals in the boring machining process into a stacked self-encoder network after data preprocessing, and outputting the current state of the cutter by the network. The method can realize the real-time monitoring of the state of the deep hole boring cutter.
Description
Technical Field
The invention belongs to the technical field of cutter state monitoring, and particularly relates to a deep hole boring cutter state monitoring method based on a stacked self-encoder.
Background
In the machine manufacturing industry, cylindrical holes with a hole depth of more than 5 times the hole diameter are generally referred to as deep holes. The common processing methods for processing deep holes mainly include drilling, reaming, boring, reaming and the like. At present, a boring process is widely adopted for deep hole machining with a larger structure size, and deep holes obtained through boring are high in precision and low in cost, so that many manufacturing enterprises adopt the boring process. Compared with general turning and milling, the boring cutting area is positioned in the deep hole, and the abrasion and damage state of the cutter is difficult to observe by naked eyes. The machine tool operator can only determine the cutting state of the cutter in the deep hole by observing the flowing cuttings and the vibration of the boring bar through the touch perception of the hand by experience, and the real-time state of the cutter is difficult to accurately judge. When abnormal conditions such as cutter breakage, tipping and the like occur, the parts can be possibly scrapped due to the fact that judgment cannot be made in time and corresponding measures are taken.
In a patent 'a real-time monitoring method for the state of a complex structural member numerical control machining cutter based on deep learning' (CN201710739173.9), a two-stage deep learning model comprising a deep confidence network and a convolution neural network is constructed, and the deep learning network is trained based on a large number of numerical control machining monitoring signals, so that the real-time monitoring of the state of the numerical control machining cutter is realized. In the patent ' a digit control machine tool knife-breaking detection system and method based on deep learning ' (CN201910228970.X) ', an image data acquisition module is used for shooting a video of a cutter cutting workpiece in the cutting process, an image data preprocessing module is used for extracting an image in the video, the extracted image is positioned, cut and normalized, an edge calculation module integrated with a knife-breaking discriminator is used for receiving the processed image, and a pre-trained convolutional neural network forward reasoning is used for obtaining a knife-breaking discrimination result.
It can be known from the above patent that the tool state monitoring method based on image recognition or cutting force signal cannot be used for deep hole boring state monitoring. The invention provides a method for monitoring the state of a deep-hole boring cutter in real time based on a stacked self-encoder, aiming at the problem of monitoring the state of the deep-hole boring cutter.
Disclosure of Invention
the invention aims to provide a method for effectively monitoring the real-time state of a boring cutter, and solves the problem that the state of the boring cutter is difficult to monitor.
in order to solve the technical problems, the technical scheme of the invention is as follows: firstly, respectively adsorbing two three-way acceleration sensors outside two retainer bearing bushes of a deep hole boring bar through magnetic seats, placing a microphone at a processing inlet of an inner hole of a workpiece, and collecting vibration and acoustic signals in the boring process; then, carrying out data preprocessing on the acquired data by adopting an amplitude limiting value filtering method; then, constructing a stacking self-encoder network, and training a stacking self-encoder by using the preprocessed data by adopting a greedy layer-by-layer method; and finally, inputting real-time vibration and sound signals in the boring processing process into a stacked self-encoder network after data preprocessing, and outputting the current state of the cutter by the network, thereby realizing the real-time monitoring of the state of the deep-hole boring cutter.
a deep hole boring cutter state monitoring method based on a stacked self-encoder is characterized by comprising the following steps:
Firstly, collecting vibration and sound information in the process of deep hole boring
Adsorbing two three-way acceleration sensors on two retainer bearings of a deep hole boring bar through a magnetic seat, placing a microphone at one end of an inner hole of a workpiece, and collecting cutter bar vibration and cutting noise in the machining process;
Second, vibration and sound data preprocessing
segmenting data collected by the three-way acceleration sensor according to sampling frequency, setting each segment of data as x (n), and performing fast Fourier transform on each segment of data:
Wherein k is 0.., N-1;
Calculating the single-side amplitude spectrum of the single-side amplitude spectrum, and completely filtering data with the amplitude lower than 0.2; grouping the filtered data according to three states of normal, dull grinding and cutter breaking;
Third, construction and training of a stacked self-encoder network
taking sample data in the same number of three states as a training set, and taking the rest samples as a test set, wherein the training set and the test set both contain the sample data in the three states; let input data be x ═ x1,x2,…,xi]Then, the input-output relationship from the encoder is:
Wherein f (-) represents an activation function of a neuron,Representing the weights of the input layer and the hidden layer,the weights of the hidden layer and the output layer are represented,showing the bias of the neurons of the hidden layer,Representing the bias of each neuron of the output layer;
constructing a loss function:
in the formula (I), the compound is shown in the specification,Is the output data from the encoder; alpha is a sparse penalty term parameter which is used for controlling the proportion of the sparse penalty term in the loss function; γ is a sparsity parameter;Is the activation of the jth neuron of the hidden layer;
using a adam optimizer, minimizing a loss function, optimizing a weight wijand bias b:
In the formula, t is iteration times; eta is the learning rate;andRespectively a first-order momentum term and a second-order momentum term; lambda [ alpha ]1and λ2Respectively taking 0.9 and 0.999 for power value;andThe correction values of the first-order and second-order momentum terms respectively; (w)ij,b)tRepresenting model weights and biases at the t-th iteration; gt=ΔJ((wij,b)t) Representing a cost function of (w) over t iterationsij,b)ta gradient of (a); the epsilon is a constant, the denominator is avoided to be 0, and the value is very small;
performing unsupervised training on the self-encoder by adopting a greedy training method, inputting a training set into a first-layer self-encoder, training the self-encoder, and minimizing a Loss function Loss to obtain optimal weight and bias; taking the hidden layer output of the first layer of self-encoder as the input of the second layer of self-encoder, training the hidden layer output of the first layer of self-encoder, performing supervised training on the softmax classifier by using the output of the second layer of self-encoder after the training is finished, and stacking the trained two layers of self-encoders and the softmax classifier to obtain the trained stacked self-encoder; after training is finished, testing the stacked self-encoder by adopting a residual test set; when the test accuracy is higher than 90%, the model can be used for monitoring the state of the cutter;
fourthly, monitoring the state of the deep hole boring cutter in real time
in the actual processing process, inputting real-time data into a trained stacked self-encoder network model after data preprocessing, and outputting the real-time state of a cutter by the model; when the cutter state is normal, the model output is 1; when the cutter state is cutter breaking, the model output is 2; when the tool state is dull, the model output is 3.
the invention has the beneficial effects that: by the method, the state of the deep hole boring cutter is monitored in real time, dependence on experience of a machine tool operator is reduced, machining efficiency is improved, and rejection rate is reduced. Redundant data are removed through filtering of the original data, the training speed of the stacked self-encoder is increased, and the accuracy of prediction is improved; and adding a sparse penalty term into the loss function to prevent the overfitting of the model and improve the generalization capability of the model.
drawings
fig. 1 is a flow chart for monitoring the state of the deep hole boring cutter.
Fig. 2 is a schematic diagram of deep hole boring machine sensor arrangement.
Fig. 3 is a structural diagram of a first layer self-encoder.
fig. 4 is a structural view of a stacked self-encoder.
Fig. 5 is a broken blade monitoring waveform diagram.
FIG. 6 is a waveform of dull monitoring.
in the figure: 1-a workpiece; 2-machine tool gear box; 3-a microphone; 4-a lathe bed; 5-1# three-way acceleration sensor; 6-cutter bar; 7-2# three-way acceleration sensor.
Detailed Description
In order to make the technical scheme and the beneficial effects of the invention clearer, the invention is described in detail below with reference to the accompanying drawings by combining the specific implementation mode of deep hole boring cutter state monitoring. The present embodiment is based on the technical solution of the present invention, and a detailed implementation and a specific operation process are given, but the scope of the present invention is not limited to the following embodiments.
the embodiment of the present invention will be described in detail by taking an example of deep hole processing by a horizontal deep hole boring machine.
Firstly, collecting vibration and sound information in the process of deep hole boring
And fixing the microphone 3 at the workpiece processing inlet to be aligned with the inner hole of the workpiece. The #1 three-way acceleration sensor 5 and the #2 three-way acceleration sensor 7 are attached to the side surface of the bearing bush of the cutter bar holder by magnetic seats in the manner shown in fig. 2. And setting the sampling frequency to be 1000Hz, and collecting vibration and sound signals in the processing process.
second, vibration and sound data preprocessing
and performing fast Fourier transform on the acquired data, and calculating a single-side amplitude spectrum. And classifying the frequency domain data according to three states of normal, knife breaking and dull grinding, wherein each sample comprises 7000 data. Wherein, the number of samples under normal condition is 20000, and the number of samples under the broken knife and the wearing condition is 5000 and 3000 respectively.
Third, construction and training of a stacked self-encoder network
stacking two autoencoders and one softmax classifier builds a network of stacked autoencoders. The number of neurons in the input layer and the output layer of the first self-encoder is 7000, and the number of neurons in the hidden layer is 3000, as shown in fig. 3. The number of neurons in the input and output layers of the second self-encoder is 3000, and the number of neurons in the hidden layer is 1000. The number of input and output layer neurons of the third self-encoder is 1000, and the number of hidden layer neurons is 500. And randomly selecting 2000 samples in three states, training the self-encoder, and training the self-encoder layer by adopting a greedy training method. Stacking two self-encoders and one softmax sorter into a stack of self-encoders as shown in fig. 4, three states of the tool are output. And the network is tested by using the residual samples, the testing accuracy reaches 91.13%, and the model can be used for monitoring the state of the cutter.
fourthly, monitoring the state of the deep hole boring cutter in real time
And inputting the real-time vibration and sound data into a stacked self-encoder model after data preprocessing, and monitoring the state of the cutter in the boring machining process. As shown in fig. 5 and 6, the model can accurately judge the real-time state of the tool.
Claims (1)
1. a deep hole boring cutter state monitoring method based on a stacked self-encoder is characterized by comprising the following steps:
Firstly, collecting vibration and sound information in the process of deep hole boring
adsorbing two three-way acceleration sensors on two retainer bearings of a deep hole boring bar through a magnetic seat, placing a microphone at one end of an inner hole of a workpiece, and collecting cutter bar vibration and cutting noise in the machining process;
Second, vibration and sound data preprocessing
Segmenting data collected by the three-way acceleration sensor according to sampling frequency, setting each segment of data as x (n), and performing fast Fourier transform on each segment of data:
Wherein k is 0.., N-1;
Calculating the single-side amplitude spectrum of the single-side amplitude spectrum, and completely filtering data with the amplitude lower than 0.2; grouping the filtered data according to three states of normal, dull grinding and cutter breaking;
Third, construction and training of a stacked self-encoder network
Taking sample data in the same number of three states as a training set, and taking the rest samples as a test set, wherein the training set and the test set both contain the sample data in the three states; let input data be x ═ x1,x2,…,xi]Then from the input of the encoderThe output relation is as follows:
wherein f (-) represents an activation function of a neuron,Representing the weights of the input layer and the hidden layer,the weights of the hidden layer and the output layer are represented,showing the bias of the neurons of the hidden layer,Representing the bias of each neuron of the output layer;
constructing a loss function:
in the formula (I), the compound is shown in the specification,is the output data from the encoder; alpha is a sparse penalty term parameter which is used for controlling the proportion of the sparse penalty term in the loss function; γ is a sparsity parameter;is the activation of the jth neuron of the hidden layer;
using a adam optimizer, minimizing a loss function, optimizing a weight wijAnd bias b:
In the formula, t is iteration times; eta is the learning rate;andrespectively a first-order momentum term and a second-order momentum term; lambda [ alpha ]1And λ2Respectively taking 0.9 and 0.999 for power value;AndThe correction values of the first-order and second-order momentum terms respectively; (w)ij,b)tRepresenting model weights and biases at the t-th iteration; gt=ΔJ((wij,b)t) Representing a cost function of (w) over t iterationsij,b)tA gradient of (a); the epsilon is a constant, the denominator is avoided to be 0, and the value is very small;
Performing unsupervised training on the self-encoder by adopting a greedy training method, inputting a training set into a first-layer self-encoder, training the self-encoder, and minimizing a Loss function Loss to obtain optimal weight and bias; taking the hidden layer output of the first layer of self-encoder as the input of the second layer of self-encoder, training the hidden layer output of the first layer of self-encoder, performing supervised training on the softmax classifier by using the output of the second layer of self-encoder after the training is finished, and stacking the trained two layers of self-encoders and the softmax classifier to obtain the trained stacked self-encoder; after training is finished, testing the stacked self-encoder by adopting a residual test set; when the test accuracy is higher than 90%, the model can be used for monitoring the state of the cutter;
fourthly, monitoring the state of the deep hole boring cutter in real time
in the actual processing process, inputting real-time data into a trained stacked self-encoder network model after data preprocessing, and outputting the real-time state of a cutter by the model; when the cutter state is normal, the model output is 1; when the cutter state is cutter breaking, the model output is 2; when the tool state is dull, the model output is 3.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910855352.8A CN110561192A (en) | 2019-09-11 | 2019-09-11 | Deep hole boring cutter state monitoring method based on stacking self-encoder |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910855352.8A CN110561192A (en) | 2019-09-11 | 2019-09-11 | Deep hole boring cutter state monitoring method based on stacking self-encoder |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110561192A true CN110561192A (en) | 2019-12-13 |
Family
ID=68779170
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910855352.8A Withdrawn CN110561192A (en) | 2019-09-11 | 2019-09-11 | Deep hole boring cutter state monitoring method based on stacking self-encoder |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110561192A (en) |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111142001A (en) * | 2020-01-10 | 2020-05-12 | 三峡大学 | Transformer multi-source partial discharge mode identification method based on parallel characteristic domain |
CN111177911A (en) * | 2019-12-24 | 2020-05-19 | 大连理工大学 | Part surface roughness online prediction method based on SDAE-DBN algorithm |
CN111276187A (en) * | 2020-01-12 | 2020-06-12 | 湖南大学 | Gene expression profile feature learning method based on self-encoder |
CN111716150A (en) * | 2020-06-30 | 2020-09-29 | 大连理工大学 | Evolution learning method for intelligently monitoring cutter state |
CN111890127A (en) * | 2020-08-06 | 2020-11-06 | 南京航空航天大学 | Cutting state edge intelligent monitoring method based on online incremental wear evolution model |
CN113103068A (en) * | 2021-04-19 | 2021-07-13 | 大连理工大学 | Cutter state monitoring method based on deep migration learning |
WO2021138982A1 (en) * | 2020-01-10 | 2021-07-15 | 大连理工大学 | Elm-sdae algorithm-based cutter state monitoring method |
CN117961645A (en) * | 2024-03-29 | 2024-05-03 | 中国矿业大学 | Cutter abrasion soft measurement method for small sample noise data |
Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106408088A (en) * | 2016-11-22 | 2017-02-15 | 北京六合智汇技术有限责任公司 | Depth-learning-theory-based fault diagnosis method for rotating machinery |
CN107584334A (en) * | 2017-08-25 | 2018-01-16 | 南京航空航天大学 | A kind of complex structural member numerical control machining cutter status real time monitor method based on deep learning |
WO2018030033A1 (en) * | 2016-08-10 | 2018-02-15 | 三菱重工工作機械株式会社 | Abnormality-detecting device and method for tool of machine tool |
CN108747590A (en) * | 2018-06-28 | 2018-11-06 | 哈尔滨理工大学 | A kind of tool wear measurement method based on rumble spectrum and neural network |
CN108764588A (en) * | 2018-07-11 | 2018-11-06 | 天津工业大学 | A kind of temperature influence power prediction method based on deep learning |
CN108921233A (en) * | 2018-07-31 | 2018-11-30 | 武汉大学 | A kind of Raman spectrum data classification method based on autoencoder network |
US20190143469A1 (en) * | 2015-09-01 | 2019-05-16 | Jpw Industries Inc. | Power tool with digital variable reluctance motor control |
CN109822398A (en) * | 2019-03-25 | 2019-05-31 | 华中科技大学 | A kind of numerically-controlled machine tool breaking detection system and method based on deep learning |
-
2019
- 2019-09-11 CN CN201910855352.8A patent/CN110561192A/en not_active Withdrawn
Patent Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20190143469A1 (en) * | 2015-09-01 | 2019-05-16 | Jpw Industries Inc. | Power tool with digital variable reluctance motor control |
WO2018030033A1 (en) * | 2016-08-10 | 2018-02-15 | 三菱重工工作機械株式会社 | Abnormality-detecting device and method for tool of machine tool |
CN106408088A (en) * | 2016-11-22 | 2017-02-15 | 北京六合智汇技术有限责任公司 | Depth-learning-theory-based fault diagnosis method for rotating machinery |
CN107584334A (en) * | 2017-08-25 | 2018-01-16 | 南京航空航天大学 | A kind of complex structural member numerical control machining cutter status real time monitor method based on deep learning |
CN108747590A (en) * | 2018-06-28 | 2018-11-06 | 哈尔滨理工大学 | A kind of tool wear measurement method based on rumble spectrum and neural network |
CN108764588A (en) * | 2018-07-11 | 2018-11-06 | 天津工业大学 | A kind of temperature influence power prediction method based on deep learning |
CN108921233A (en) * | 2018-07-31 | 2018-11-30 | 武汉大学 | A kind of Raman spectrum data classification method based on autoencoder network |
CN109822398A (en) * | 2019-03-25 | 2019-05-31 | 华中科技大学 | A kind of numerically-controlled machine tool breaking detection system and method based on deep learning |
Non-Patent Citations (1)
Title |
---|
库祥臣等: "利用振动频谱预测刀具磨损量", 《机械设计与制造》 * |
Cited By (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111177911B (en) * | 2019-12-24 | 2022-09-20 | 大连理工大学 | Part surface roughness online prediction method based on SDAE-DBN algorithm |
CN111177911A (en) * | 2019-12-24 | 2020-05-19 | 大连理工大学 | Part surface roughness online prediction method based on SDAE-DBN algorithm |
CN111142001B (en) * | 2020-01-10 | 2022-04-22 | 三峡大学 | Transformer multi-source partial discharge mode identification method based on parallel characteristic domain |
WO2021138982A1 (en) * | 2020-01-10 | 2021-07-15 | 大连理工大学 | Elm-sdae algorithm-based cutter state monitoring method |
CN111142001A (en) * | 2020-01-10 | 2020-05-12 | 三峡大学 | Transformer multi-source partial discharge mode identification method based on parallel characteristic domain |
CN111276187B (en) * | 2020-01-12 | 2021-09-10 | 湖南大学 | Gene expression profile feature learning method based on self-encoder |
CN111276187A (en) * | 2020-01-12 | 2020-06-12 | 湖南大学 | Gene expression profile feature learning method based on self-encoder |
CN111716150A (en) * | 2020-06-30 | 2020-09-29 | 大连理工大学 | Evolution learning method for intelligently monitoring cutter state |
CN111890127A (en) * | 2020-08-06 | 2020-11-06 | 南京航空航天大学 | Cutting state edge intelligent monitoring method based on online incremental wear evolution model |
CN111890127B (en) * | 2020-08-06 | 2022-09-20 | 南京航空航天大学 | Cutting state edge intelligent monitoring method based on online incremental wear evolution model |
CN113103068A (en) * | 2021-04-19 | 2021-07-13 | 大连理工大学 | Cutter state monitoring method based on deep migration learning |
CN113103068B (en) * | 2021-04-19 | 2022-03-29 | 大连理工大学 | Cutter state monitoring method based on deep migration learning |
CN117961645A (en) * | 2024-03-29 | 2024-05-03 | 中国矿业大学 | Cutter abrasion soft measurement method for small sample noise data |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110561192A (en) | Deep hole boring cutter state monitoring method based on stacking self-encoder | |
CN110576336A (en) | Method for monitoring abrasion loss of deep hole machining tool based on SSAE-LSTM model | |
CN111136509B (en) | Cutter state monitoring method based on ELM-SDAE algorithm | |
CN111716150B (en) | Evolution learning method for intelligently monitoring cutter state | |
CN107584334B (en) | A kind of complex structural member numerical control machining cutter status real time monitor method based on deep learning | |
Segreto et al. | Multiple sensor monitoring in nickel alloy turning for tool wear assessment via sensor fusion | |
CN102091972B (en) | Numerical control machine tool wear monitoring method | |
CN106181579B (en) | A kind of Tool Wear Monitoring method based on multisensor current signal | |
JP4321581B2 (en) | Machine tool comprehensive monitoring device | |
CN104723171B (en) | Cutter wear monitoring method based on current and acoustic emission compound signals | |
CN110355608A (en) | Based on the tool abrasion prediction technique from attention mechanism and deep learning | |
CN105834834B (en) | Cutter wear state monitoring method based on drosophila optimized algorithm | |
CN103465107B (en) | A kind of Tool Wear Monitoring method | |
CN111037365A (en) | Cutter state monitoring data set enhancing method based on generative countermeasure network | |
CN110263474A (en) | A kind of cutter life real-time predicting method of numerically-controlled machine tool | |
CN106112697A (en) | A kind of milling parameter automatic alarm threshold setting method based on 3 σ criterions | |
CN111660141B (en) | Milling cutter wear state identification method based on spindle driving current and irrelevant to working conditions | |
CN110472800A (en) | A kind of machine tool method for predicting residual useful life based on LSTM+CNN | |
CN113378725A (en) | Cutter fault diagnosis method, equipment and storage medium based on multi-scale-channel attention network | |
Arapoğlu et al. | An ANN-based method to predict surface roughness in turning operations | |
CN112247673B (en) | Woodworker cutter wear state diagnosis method based on genetic BP neural network | |
CN109333159B (en) | Depth kernel extreme learning machine method and system for online monitoring of tool wear state | |
CN108393744B (en) | Multi-sensing monitoring method for cutter state | |
CN111814728A (en) | Method for recognizing wear state of cutting tool of numerical control machine tool and storage medium | |
CN104708497A (en) | Tool wear monitoring system based on current and sound emission composite signals |
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 | ||
WW01 | Invention patent application withdrawn after publication |
Application publication date: 20191213 |
|
WW01 | Invention patent application withdrawn after publication |