CN114091544A - TBM disc cutter abrasion identification system based on vibration signal and neural network - Google Patents
TBM disc cutter abrasion identification system based on vibration signal and neural network Download PDFInfo
- Publication number
- CN114091544A CN114091544A CN202111429678.8A CN202111429678A CN114091544A CN 114091544 A CN114091544 A CN 114091544A CN 202111429678 A CN202111429678 A CN 202111429678A CN 114091544 A CN114091544 A CN 114091544A
- Authority
- CN
- China
- Prior art keywords
- vibration
- neural network
- hob
- frequency domain
- wear
- 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.)
- Pending
Links
- 238000013528 artificial neural network Methods 0.000 title claims abstract description 36
- 238000005299 abrasion Methods 0.000 title claims abstract description 22
- 238000012545 processing Methods 0.000 claims abstract description 25
- 239000011435 rock Substances 0.000 claims abstract description 25
- 238000000034 method Methods 0.000 claims abstract description 16
- 230000001133 acceleration Effects 0.000 claims description 14
- 238000012360 testing method Methods 0.000 claims description 10
- 230000015654 memory Effects 0.000 claims description 8
- 238000005070 sampling Methods 0.000 claims description 7
- 230000008569 process Effects 0.000 claims description 6
- 125000004122 cyclic group Chemical group 0.000 claims description 3
- 230000035515 penetration Effects 0.000 claims description 3
- 238000000605 extraction Methods 0.000 abstract description 2
- 238000012549 training Methods 0.000 description 13
- 238000010586 diagram Methods 0.000 description 9
- 238000010276 construction Methods 0.000 description 5
- 230000006870 function Effects 0.000 description 4
- 238000012544 monitoring process Methods 0.000 description 4
- 238000012795 verification Methods 0.000 description 4
- 238000003745 diagnosis Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 238000003062 neural network model Methods 0.000 description 3
- 230000000737 periodic effect Effects 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 238000007689 inspection Methods 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 230000005641 tunneling Effects 0.000 description 2
- ORILYTVJVMAKLC-UHFFFAOYSA-N Adamantane Natural products C1C(C2)CC3CC1CC2C3 ORILYTVJVMAKLC-UHFFFAOYSA-N 0.000 description 1
- 229910001069 Ti alloy Inorganic materials 0.000 description 1
- 230000002411 adverse Effects 0.000 description 1
- 230000002238 attenuated effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000000919 ceramic Substances 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 239000000306 component Substances 0.000 description 1
- 239000008358 core component Substances 0.000 description 1
- 238000005520 cutting process Methods 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000013079 data visualisation Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000013210 evaluation model Methods 0.000 description 1
- 239000010438 granite Substances 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 230000007787 long-term memory Effects 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000000306 recurrent effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000007789 sealing Methods 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 230000006403 short-term memory Effects 0.000 description 1
- 238000010183 spectrum analysis Methods 0.000 description 1
- 238000003860 storage Methods 0.000 description 1
- 230000001360 synchronised effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
-
- E—FIXED CONSTRUCTIONS
- E21—EARTH OR ROCK DRILLING; MINING
- E21D—SHAFTS; TUNNELS; GALLERIES; LARGE UNDERGROUND CHAMBERS
- E21D9/00—Tunnels or galleries, with or without linings; Methods or apparatus for making thereof; Layout of tunnels or galleries
- E21D9/10—Making by using boring or cutting machines
- E21D9/11—Making by using boring or cutting machines with a rotary drilling-head cutting simultaneously the whole cross-section, i.e. full-face machines
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Computing Systems (AREA)
- Software Systems (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Mathematical Physics (AREA)
- Data Mining & Analysis (AREA)
- Health & Medical Sciences (AREA)
- Mining & Mineral Resources (AREA)
- Signal Processing (AREA)
- Environmental & Geological Engineering (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Life Sciences & Earth Sciences (AREA)
- Geochemistry & Mineralogy (AREA)
- Geology (AREA)
- Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)
- Measurement Of Mechanical Vibrations Or Ultrasonic Waves (AREA)
Abstract
The invention discloses a TBM disc cutter abrasion recognition system based on a vibration signal and a neural network, which comprises a vibration signal acquisition processing subsystem and a neural network recognition model, wherein the vibration signal acquisition processing subsystem comprises a sensor module, a data acquisition module and a data processing output module which are sequentially connected; the sensor module is arranged on the hob and used for acquiring a rock breaking vibration signal of the hob and converting the vibration signal into a piezoelectric signal; the data processing and outputting module calculates the time domain signals by adopting a fast Fourier method to obtain frequency domain signals and outputs frequency domain data; and the neural network identification model identifies the wear state of the single-blade disc cutter according to the frequency domain data. The recognition system has better stability and reliability, has good feature extraction and recognition capability, can comprehensively capture the features of the vibration frequency domain data, and achieves a good accuracy level.
Description
Technical Field
The invention belongs to the field of TBM hob abrasion state identification and diagnosis, and particularly relates to a TBM disc hob abrasion identification system based on a vibration signal and a neural network.
Background
The disc cutter is an important core component of the TBM, and plays a role in the most important rock breaking function, the state of the disc cutter directly influences whether the TBM can normally operate, and the disc cutter is a major risk source in tunneling construction. The conventional common method for diagnosing the abrasion of the TBM disc cutter is shutdown and open-bin inspection, and an operator must frequently shut down and enter a bin to manually overhaul a cutter head cutter so as to judge whether the cutter head needs to be repaired and replaced. High time cost and economic cost are brought by cutter replacement and maintenance due to shutdown inspection, and the health and safety of operating personnel cannot be effectively guaranteed due to the high-temperature and high-pressure environment in the bin. Therefore, the hob abrasion state identification system is established, and the hob abrasion state information is provided in time, so that the hob abrasion state identification system has important practical significance.
At present, most of the existing TBM hob abrasion identification and diagnosis methods are to indirectly calculate the integral abrasion condition of a hob from integral parameters of a hob head, such as torque, rotating speed, temperature rise and the like, or to establish a prediction model by means of geological parameters and tunneling parameters. However, these methods using indirect parameters are not suitable for evaluating the wear state of a single hob. Some studies have used sensors, such as eddy current sensors, ultrasonic sensors, laser sensors, etc., to evaluate the wear of a single knife by acquiring corresponding signals. However, due to the strong damp heat and dusty noise of the construction environment, the above electric, acoustic and optical signals have the problems of stability and reliability in application.
In the field of machine manufacturing, sensors using vibration as a measurement signal are robust and more suitable for industrial environments. The vibration signal is converted to extract characteristic information, so that the tool wear identification is convenient. With the development of AI technology, sufficient potential and advantages are gradually exhibited in the field of intelligent state monitoring, and one of typical representatives is a neural network algorithm. In 1987, the neural network technology is combined with the sensor technology for the first time, so that the reliability of cutter monitoring is improved. At present, a neural network algorithm is researched and applied more in the state monitoring of a common cutting tool, and in the aspect of TBM disc cutter abrasion, researchers establish an abrasion evaluation model by utilizing a one-dimensional convolution neural network and combining field parameters. Therefore, the neural network has potential and feasibility for identifying the abrasion state of the hob based on the vibration signal.
In the existing research, the vibration signal is rarely used for detecting the wear state of a single disc cutter, and part of the neural network algorithm is also primarily applied to the disc cutter state monitoring, but there is no example of combining the vibration signal and the neural network algorithm for identifying the wear state of the single disc cutter.
Disclosure of Invention
The invention aims to overcome the defects of the prior art and provide a TBM disc cutter abrasion recognition system based on vibration signals and a neural network, which takes vibration frequency domain signals as input data, can exert the advantages of firmness, durability and high reliability and stability of a vibration sensor, and enables the whole abrasion recognition system to have better stability and reliability.
The purpose of the invention is realized by the following technical scheme: the TBM disc cutter abrasion identification system based on the vibration signal and the neural network comprises a vibration signal acquisition processing subsystem and a neural network identification model, wherein the vibration signal acquisition processing subsystem comprises a sensor module, a data acquisition module and a data processing output module which are sequentially connected;
the sensor module is arranged on the hob and used for acquiring a rock breaking vibration signal of the hob and converting the vibration signal into a piezoelectric signal; the data processing and outputting module calculates the time domain signals by adopting a fast Fourier method to obtain frequency domain signals and outputs frequency domain data; and the neural network identification model identifies the wear state of the single-blade disc cutter according to the frequency domain data.
Further, the wear state of the TBM disc cutter comprises four states of a normal wear state, a wear failure state, an eccentric wear failure state and a tipping failure state.
Further, the hob rock breaking vibration signal is acquired by the vibration sensor at the positions of the hob ring, the cutter hub, the cutter shaft and the cutter holder respectively under the working condition that the TBM disc hob breaks rock on the hard rock sample according to set parameters. The vibration sensor adopts a three-axis acceleration vibration sensor.
Furthermore, the neural network recognition model adopts a long-short term memory network model (LSTM) in a cyclic neural network, the model recognizes the wear state according to the vibration frequency domain characteristics, and the recognition object is the overall wear state of the single hob.
Further, the construction process of the long-short term memory network model data set is as follows: respectively collecting vibration acceleration signals of hobs in four wear states of a normal wear state, a wear failure state, an eccentric wear failure state and a tipping failure state on a hard rock sample at the positions of a cutter ring, a cutter hub, a cutter shaft and a cutter seat of the TBM disc-shaped hob by using a vibration sensor, and respectively carrying out a rock breaking test on the hob with penetration degrees of 1mm, 2mm and 3mm and a feeding speed of 30mm/s, and converting the vibration acceleration signals into piezoelectric signals; sampling the piezoelectric signal by the data acquisition module at a sampling rate of 5000Hz to obtain a vibration time domain signal, and transmitting the vibration time domain signal to the data processing output module; the data processing and outputting module converts a time domain signal into a frequency domain signal with 1Hz resolution ratio by using a fast Fourier method, the frequency domain range is 0-2500 Hz, the time-frequency domain data is visually output to a PC (personal computer) end, signal segments with obvious time-frequency domain characteristics are selected, the selected signal segments are output as time sequence frequency domain data at 1s time intervals to form a data set, and the long-short term memory network model is trained by using the data set.
The invention has the beneficial effects that: the wear identification system provided by the invention can specifically identify the wear state of a single hob; the disc cutter abrasion identification model takes the vibration frequency domain signal as input data, can exert the advantages of firmness, durability and high reliability and stability of the vibration sensor, greatly avoid the adverse effect of the construction environment on the sensor and ensure that the whole abrasion identification system has better stability and reliability; the neural network model has good feature extraction and recognition capability, can comprehensively capture the features of vibration frequency domain data, and achieves a good accuracy level; the invention preliminarily verifies and explores the application feasibility and application effect of the wear identification system combining the vibration signal and the neural network model on the single disc cutter, and accumulates experience for subsequent perfection and deep application.
Drawings
FIG. 1 is a hob wear identification flow chart;
FIG. 2 is a schematic view of a vibration sensor;
FIG. 3 is a schematic diagram of a neural network model;
FIG. 4 is a schematic diagram of the position of the sensor to detect the vibrations of the hob;
FIG. 5 is a schematic axial diagram of a time domain signal X of a normal wear hob;
FIG. 6 is a schematic diagram of X-axis frequency domain signals of a normal wear hob and a wear failure hob;
FIG. 7 is a schematic diagram of X-axis time-frequency domain signals of an eccentric wear failure hob in an eccentric wear surface grinding condition.
Wherein: 1. the position of the cutter ring; 2. a cutter hub position; 3. the position of the cutter shaft; 4. a tool holder position.
Detailed Description
According to the invention, a triaxial vibration acceleration sensor is arranged on a hob, vibration signals generated when the hob breaks rock are directly collected, time domain analysis is carried out on the vibration signals, a periodic waveform section is taken for FFT calculation, frequency domain characteristics of the periodic waveform section are obtained, the frequency domain characteristics are output as time series data to train an LSTM (local state transform) model, and identification and diagnosis of the abrasion state of a single hob are realized. The technical scheme of the invention is further explained by combining the attached drawings.
As shown in fig. 1, the TBM disc cutter wear identification system based on vibration signals and neural network of the present invention includes a vibration signal acquisition processing subsystem and a neural network identification model, the vibration signal acquisition processing subsystem includes a sensor module, a data acquisition module and a data processing output module which are connected in sequence; the data processing output module and the neural network identification model are positioned at the PC end of the upper computer, and the sensor module and the data acquisition module are connected with the upper computer to transmit data and acquire electric energy;
the sensor module is arranged on the hob and used for acquiring a rock breaking vibration signal of the hob and converting the vibration signal into a piezoelectric signal; the data processing and outputting module calculates the time domain signals by adopting a Fast Fourier Transform (FFT) method to obtain frequency domain signals and outputs frequency domain data, and visually outputs the time domain signals and the frequency domain signals; and the neural network identification model identifies the wear state of the single-blade disc cutter according to the frequency domain data.
The wear state of the TBM disc cutter comprises a normal wear state, a wear failure state, an eccentric wear failure state and a tipping failure state. The hob rock breaking vibration signal is acquired by a vibration sensor at the positions of a hob ring, a hob hub, a hob shaft and a hob seat respectively under the working condition that a TBM disc hob breaks rock on a hard rock sample according to set parameters.
The sensor module is an IEPE triaxial acceleration vibration sensor, the sensor model is DYTRAN 3263A3, the mass of a single sensor is 5.6g, the sensitivity is 500mV/g, and the measurement range is +/-100 g; the sensor is schematically shown in fig. 2. The sensor is composed of a light titanium alloy shell, a shear type piezoelectric ceramic element, a low-noise JFET circuit and the like, has the characteristics of firmness, durability, strong anti-interference capability, good sealing performance and high reliability, and still has excellent performance under complex environmental conditions of high humidity, dustiness and the like.
Specifically, the model of the data acquisition module is a DEWEsoft SIRIUS multi-channel data acquisition module, each channel is synchronized with microsecond precision, and the sampling rate is set to be 5000 Hz; the sensor and the upper computer are connected by an additional data line to transmit signals, the upper computer provides electric energy through the USB port, and the sensor is powered.
Specifically, the data processing output module is a PC-side dewoft X312 software platform. The software integrates the functions of data storage, data analysis and data visualization, has the multithreading data calculation capability and a high-performance storage engine, has the FFT analyzer module with most of functions for spectrum analysis, and can perform frequency domain analysis on vibration signals.
The neural network identification model adopts a long-short term memory network (LSTM) model in a cyclic neural network, identifies the wear state according to the vibration frequency domain characteristics, and identifies the overall wear state of the single hob as an object; the LSTM has a gating mechanism (forgetting gate, input gate, output gate), can effectively solve the long-term dependence problem existing in the conventional recurrent neural network, and is suitable for processing time series events, and the LSTM model schematic diagram is shown in fig. 3 (a). The LSTM initial model is built on a pystorm platform by using a pytorech framework, the number of nodes of a double-layer, one-way and hidden layer is 128, a Dropout parameter is set to 0.3 to reduce the overfitting risk, the output of the LSTM model is imported to a Linear layer (Linear, input feature dimension 128, output feature dimension 3) for outputting a prediction result, and a schematic diagram of the Linear layer is shown in fig. 3 (b). And then, training and optimizing by using a training data set, wherein the batch size of input data is 128, the learning rate is 0.008 (the learning rate is attenuated, and the learning rate is reduced to 95 percent of the previous learning rate every two training periods), and the optimizer selects Adam and selects a cross entropy loss function to calculate a loss value. In the training process, once a period of training is completed, a verification data set is imported, and the verification accuracy is obtained. And if the verification accuracy rate is continuously reduced in 24 continuous periods, stopping the training process and storing the model parameters corresponding to the optimal verification accuracy rate. So as to ensure that the overfitting phenomenon does not occur in the training process.
The initial model is established on a Pycharm platform according to selected parameters by using a Pythrch frame, and then training optimization is carried out by using a training data set obtained by a rock breaking test. And importing the test data set after the training is finished to obtain the test accuracy so as to check the training effect of the model.
The construction process of the long-short term memory network model data set is as follows: using a triaxial acceleration vibration sensor to respectively collect vibration acceleration signals of a hob in four wear states of a normal wear state, a wear failure state, an eccentric wear failure state and a tipping failure state on a hard rock sample at the rock breaking test by respectively using penetration degrees of 1mm, 2mm and 3mm and a feed speed of 30mm/s at the positions of a cutter ring, a cutter hub, a cutter shaft and a cutter seat of the TBM disc hob (the positions of the sensors are shown in a figure 4(a) and a figure 4(b), wherein 1 is the position of the cutter ring, 2 is the position of the cutter hub, 3 is the position of the cutter shaft, and 4 is the position of the cutter seat), and converting the vibration acceleration signals into piezoelectric signals; sampling the piezoelectric signal by the data acquisition module at a sampling rate of 5000Hz to obtain a vibration time domain signal, and transmitting the vibration time domain signal to the data processing output module; the data processing and outputting module converts a time domain signal into a frequency domain signal with 1Hz resolution ratio by using a fast Fourier method, the frequency domain range is 0-2500 Hz, the time-frequency domain data is visually output to a PC (personal computer) end, signal segments with obvious time-frequency domain characteristics are selected, the selected signal segments are output as time sequence frequency domain data at 1s time intervals to form a data set, the data set is divided into training, verifying and testing data sets according to the proportion of 7:2:1, then a long-term and short-term memory network model is trained by using the training set data in the data set, the verifying set is used for verifying to prevent overfitting, and the testing data set is used for testing the accuracy of the trained model.
Referring to fig. 5, the vibration time domain signals are schematic, when a hob performs a rock breaking test on a granite sample, a characteristic waveform appears, when the rock sample is fractured in a powdery form, a stable periodic wave characteristic shown in the diagram (a) appears, and when the rock sample is fractured in a deep and long crack form, a single sawtooth wave characteristic shown in the diagram (b) appears; referring to fig. 6, it is shown that the acceleration components are distributed in a plurality of frequency bands in a concentrated manner, and the frequency band positions are different due to different wear states of the hob. Acceleration and frequency signals of the eccentric wear failure hob in the eccentric wear surface grinding condition are respectively shown in fig. 7(a) and 7(b), under the special working condition of the eccentric wear surface grinding, the vibration acceleration amplitude change in the time domain is far larger than that in other working conditions due to violent vibration, and a large amount of high-order harmonic waves are excited in the frequency domain.
It will be appreciated by those of ordinary skill in the art that the embodiments described herein are intended to assist the reader in understanding the principles of the invention and are to be construed as being without limitation to such specifically recited embodiments and examples. Those skilled in the art can make various other specific changes and combinations based on the teachings of the present invention without departing from the spirit of the invention, and these changes and combinations are within the scope of the invention.
Claims (6)
1. The TBM disc cutter abrasion identification system based on the vibration signal and the neural network is characterized by comprising a vibration signal acquisition and processing subsystem and a neural network identification model, wherein the vibration signal acquisition and processing subsystem comprises a sensor module, a data acquisition module and a data processing and outputting module which are sequentially connected;
the sensor module is arranged on the hob and used for acquiring a rock breaking vibration signal of the hob and converting the vibration signal into a piezoelectric signal; the data processing and outputting module calculates the time domain signals by adopting a fast Fourier method to obtain frequency domain signals and outputs frequency domain data; and the neural network identification model identifies the wear state of the single-blade disc cutter according to the frequency domain data.
2. The TBM disc cutter wear identification system based on the vibration signal and the neural network is characterized in that the wear states of the TBM disc cutter comprise four states, namely a normal wear state, a wear failure state, an eccentric wear failure state and a tipping failure state.
3. The TBM disc cutter abrasion recognition system based on the vibration signal and the neural network as claimed in claim 1, wherein the hob rock breaking vibration signal is acquired by the vibration sensor at the positions of a hob ring, a hob hub, a hob shaft and a hob seat respectively under the working condition that the TBM disc cutter breaks rock on a hard rock sample according to set parameters.
4. The TBM disc cutter wear identification system based on the vibration signal and the neural network is characterized in that the vibration sensor is a three-axis acceleration vibration sensor.
5. The TBM disc cutter wear identification system based on the vibration signal and the neural network is characterized in that the neural network identification model adopts a long-short term memory network model in a cyclic neural network, and identifies the wear state according to the vibration frequency domain characteristics, wherein the model identifies the overall wear state of the object as a single-handle disc cutter.
6. The TBM disc cutter wear identification system based on the vibration signal and the neural network is characterized in that the long-short term memory network model data set is constructed by the following process: respectively collecting vibration acceleration signals of hobs in four wear states of a normal wear state, a wear failure state, an eccentric wear failure state and a tipping failure state on a hard rock sample at the positions of a cutter ring, a cutter hub, a cutter shaft and a cutter seat of the TBM disc-shaped hob by using a vibration sensor, and respectively carrying out a rock breaking test on the hob with penetration degrees of 1mm, 2mm and 3mm and a feeding speed of 30mm/s, and converting the vibration acceleration signals into piezoelectric signals; sampling the piezoelectric signal by the data acquisition module at a sampling rate of 5000Hz to obtain a vibration time domain signal, and transmitting the vibration time domain signal to the data processing output module; the data processing and outputting module converts a time domain signal into a frequency domain signal with 1Hz resolution ratio by using a fast Fourier method, the frequency domain range is 0-2500 Hz, the time-frequency domain data is visually output to a PC (personal computer) end, signal segments with obvious time-frequency domain characteristics are selected, the selected signal segments are output as time sequence frequency domain data at 1s time intervals to form a data set, and the long-short term memory network model is trained by using the data set.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111429678.8A CN114091544A (en) | 2021-11-29 | 2021-11-29 | TBM disc cutter abrasion identification system based on vibration signal and neural network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111429678.8A CN114091544A (en) | 2021-11-29 | 2021-11-29 | TBM disc cutter abrasion identification system based on vibration signal and neural network |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114091544A true CN114091544A (en) | 2022-02-25 |
Family
ID=80305360
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111429678.8A Pending CN114091544A (en) | 2021-11-29 | 2021-11-29 | TBM disc cutter abrasion identification system based on vibration signal and neural network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114091544A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114595733A (en) * | 2022-05-10 | 2022-06-07 | 山东大学 | Bridge inhaul cable broken wire signal identification method and system based on long-term and short-term memory network |
Citations (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20170039799A (en) * | 2015-10-01 | 2017-04-12 | 현대건설주식회사 | Apparatus and method for evaluating wear and abrasion of slurry type TBM disc cutter |
| 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 |
| CN109048492A (en) * | 2018-07-30 | 2018-12-21 | 北京航空航天大学 | Cutting-tool wear state detection method, device and equipment based on convolutional neural networks |
| CN109307493A (en) * | 2018-09-12 | 2019-02-05 | 天津大学 | The real-time detection method of the normal section disk cutter wear condition of hard rock tunnel development machine |
| CN109590805A (en) * | 2018-12-17 | 2019-04-09 | 杭州国彪超声设备有限公司 | A kind of determination method and system of turning cutting tool working condition |
| CN110153802A (en) * | 2019-07-04 | 2019-08-23 | 西南交通大学 | A kind of cutting-tool wear state discrimination method based on convolutional neural networks and long Memory Neural Networks conjunctive model in short-term |
| CN111322082A (en) * | 2020-03-09 | 2020-06-23 | 西南交通大学 | TBM hob state monitoring and fault diagnosis method and system |
| CN111687689A (en) * | 2020-06-23 | 2020-09-22 | 重庆大学 | Cutter wear state prediction method and device based on LSTM and CNN |
| CN111761410A (en) * | 2020-07-20 | 2020-10-13 | 天津职业技术师范大学(中国职业培训指导教师进修中心) | Cutter wear state monitoring device and method based on vibration signals |
| CN112160761A (en) * | 2020-09-25 | 2021-01-01 | 上海交通大学 | Hard rock TBM cutter disc hob abrasion real-time evaluation method based on field parameters |
| CN112435363A (en) * | 2019-08-26 | 2021-03-02 | 贵州大学 | Cutter wear state real-time monitoring method |
| CN112692646A (en) * | 2020-12-31 | 2021-04-23 | 上海交通大学 | Intelligent assessment method and device for tool wear state |
| CN112766182A (en) * | 2021-01-22 | 2021-05-07 | 浙江工业大学 | Method for identifying wear state of broaching tool |
| CN113211189A (en) * | 2021-04-26 | 2021-08-06 | 上海大学 | Broaching tool wear and damage quantitative prediction system and method based on vibration signals |
| CN113496312A (en) * | 2021-07-05 | 2021-10-12 | 山东大学 | Cutter residual life prediction method and equipment based on multi-dimensional feature extraction fusion and long-term and short-term memory network and storage medium |
| CN113569903A (en) * | 2021-06-09 | 2021-10-29 | 西安电子科技大学 | Method, system, equipment, medium and terminal for predicting abrasion of numerical control machine tool cutter |
-
2021
- 2021-11-29 CN CN202111429678.8A patent/CN114091544A/en active Pending
Patent Citations (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20170039799A (en) * | 2015-10-01 | 2017-04-12 | 현대건설주식회사 | Apparatus and method for evaluating wear and abrasion of slurry type TBM disc cutter |
| 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 |
| CN109048492A (en) * | 2018-07-30 | 2018-12-21 | 北京航空航天大学 | Cutting-tool wear state detection method, device and equipment based on convolutional neural networks |
| CN109307493A (en) * | 2018-09-12 | 2019-02-05 | 天津大学 | The real-time detection method of the normal section disk cutter wear condition of hard rock tunnel development machine |
| CN109590805A (en) * | 2018-12-17 | 2019-04-09 | 杭州国彪超声设备有限公司 | A kind of determination method and system of turning cutting tool working condition |
| CN110153802A (en) * | 2019-07-04 | 2019-08-23 | 西南交通大学 | A kind of cutting-tool wear state discrimination method based on convolutional neural networks and long Memory Neural Networks conjunctive model in short-term |
| CN112435363A (en) * | 2019-08-26 | 2021-03-02 | 贵州大学 | Cutter wear state real-time monitoring method |
| CN111322082A (en) * | 2020-03-09 | 2020-06-23 | 西南交通大学 | TBM hob state monitoring and fault diagnosis method and system |
| CN111687689A (en) * | 2020-06-23 | 2020-09-22 | 重庆大学 | Cutter wear state prediction method and device based on LSTM and CNN |
| CN111761410A (en) * | 2020-07-20 | 2020-10-13 | 天津职业技术师范大学(中国职业培训指导教师进修中心) | Cutter wear state monitoring device and method based on vibration signals |
| CN112160761A (en) * | 2020-09-25 | 2021-01-01 | 上海交通大学 | Hard rock TBM cutter disc hob abrasion real-time evaluation method based on field parameters |
| CN112692646A (en) * | 2020-12-31 | 2021-04-23 | 上海交通大学 | Intelligent assessment method and device for tool wear state |
| CN112766182A (en) * | 2021-01-22 | 2021-05-07 | 浙江工业大学 | Method for identifying wear state of broaching tool |
| CN113211189A (en) * | 2021-04-26 | 2021-08-06 | 上海大学 | Broaching tool wear and damage quantitative prediction system and method based on vibration signals |
| CN113569903A (en) * | 2021-06-09 | 2021-10-29 | 西安电子科技大学 | Method, system, equipment, medium and terminal for predicting abrasion of numerical control machine tool cutter |
| CN113496312A (en) * | 2021-07-05 | 2021-10-12 | 山东大学 | Cutter residual life prediction method and equipment based on multi-dimensional feature extraction fusion and long-term and short-term memory network and storage medium |
Non-Patent Citations (5)
| Title |
|---|
| QIPENG CHEN 等: "Research on a Real-Time Monitoring Method for the Wear State of a Tool Based on a Convolutional Bidirectional LSTM Model" * |
| ZHIWEN HUANG 等: "Tool Wear Monitoring with Vibration Signals Based on Short-Time Fourier Transform and Deep Convolutional Neural Network in Milling" * |
| 刘思辰 等: "基于多传感器融合的刀具剩余寿命预测" * |
| 张晓波 等: "TBM 掘进刀具磨损实时监测技术及刀盘振动监测分析" * |
| 李增良: "盘形滚刀刀圈失效原因及成形方法研究" * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114595733A (en) * | 2022-05-10 | 2022-06-07 | 山东大学 | Bridge inhaul cable broken wire signal identification method and system based on long-term and short-term memory network |
| CN114595733B (en) * | 2022-05-10 | 2024-04-02 | 山东大学 | Bridge inhaul cable broken wire signal identification method and system based on long-short-term memory network |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109186744B (en) | Turbine blade crack evaluation system and turbine blade crack early warning method | |
| CN102226740A (en) | Bearing fault detection method based on manner of controlling stochastic resonance by external periodic signal | |
| CN110160765A (en) | A kind of shock characteristic recognition methods and system based on sound or vibration signal | |
| CN112098128A (en) | Power mechanical equipment fault and energy consumption analysis method based on noise and vibration | |
| CN102495320A (en) | Dynamic monitoring system and on-line monitoring method of transformer operation exception | |
| CN101794138A (en) | Dynamic characteristic test and analysis system for numerical control machine tool | |
| Sun et al. | Fault diagnosis of mechanical equipment in high energy consumption industries in China: A review | |
| CN113155469A (en) | Engine fault diagnosis alarm system and device | |
| CN110186549A (en) | Blade vibration recognition methods based on Tip timing sensor | |
| Saxena et al. | Assessment of gearbox fault detection using vibration signal analysis and acoustic emission technique | |
| CN107414600A (en) | The process monitoring method of internal thread low frequency exciting cold extrusion lathe based on multiple sensor signals | |
| CN114091544A (en) | TBM disc cutter abrasion identification system based on vibration signal and neural network | |
| CN202661269U (en) | Compound fault diagnosis test platform of diesel engine | |
| CN203965033U (en) | A kind of boom-type roadheader spindle vibration monitoring running state device | |
| Wang et al. | Condition monitoring on grease lubrication of rolling bearing using AE technology | |
| CN103878639A (en) | Static and dynamic tension detecting system for chains of chained tool magazines | |
| CN207600740U (en) | A kind of monitoring system for wind-driven generator | |
| CN103630604B (en) | The recognition methods of centrifugal compressor half-opened impeller crack fault | |
| CN201953977U (en) | On-line fault diagnosis system of decelerating clutch | |
| CN103091257A (en) | Apple storage period detection method based on spectrum analysis | |
| CN202948288U (en) | Sawing load detection device based on flutter characteristic of metal band saw blade | |
| CN104236871A (en) | Crack failure recognition method for semi-open type impeller of centrifugal compressor | |
| CN103454114B (en) | The diagnosis of mesolow gas pressure regulator, governor operation troubles and safety early warning device and method | |
| Yang et al. | Unknown Bearing Fault Recognition in Strong Noise Background | |
| Wändell | Multistage gearboxes: Vibration based quality control |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20220225 |