CN114799495A - Control method and related device for laser cutting - Google Patents
Control method and related device for laser cutting Download PDFInfo
- Publication number
- CN114799495A CN114799495A CN202111627695.2A CN202111627695A CN114799495A CN 114799495 A CN114799495 A CN 114799495A CN 202111627695 A CN202111627695 A CN 202111627695A CN 114799495 A CN114799495 A CN 114799495A
- Authority
- CN
- China
- Prior art keywords
- parameters
- cutting
- glass
- acoustic emission
- 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.)
- Granted
Links
Images
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B23—MACHINE TOOLS; METAL-WORKING NOT OTHERWISE PROVIDED FOR
- B23K—SOLDERING OR UNSOLDERING; WELDING; CLADDING OR PLATING BY SOLDERING OR WELDING; CUTTING BY APPLYING HEAT LOCALLY, e.g. FLAME CUTTING; WORKING BY LASER BEAM
- B23K26/00—Working by laser beam, e.g. welding, cutting or boring
- B23K26/36—Removing material
- B23K26/38—Removing material by boring or cutting
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B23—MACHINE TOOLS; METAL-WORKING NOT OTHERWISE PROVIDED FOR
- B23K—SOLDERING OR UNSOLDERING; WELDING; CLADDING OR PLATING BY SOLDERING OR WELDING; CUTTING BY APPLYING HEAT LOCALLY, e.g. FLAME CUTTING; WORKING BY LASER BEAM
- B23K26/00—Working by laser beam, e.g. welding, cutting or boring
- B23K26/02—Positioning or observing the workpiece, e.g. with respect to the point of impact; Aligning, aiming or focusing the laser beam
- B23K26/06—Shaping the laser beam, e.g. by masks or multi-focusing
- B23K26/062—Shaping the laser beam, e.g. by masks or multi-focusing by direct control of the laser beam
- B23K26/0622—Shaping the laser beam, e.g. by masks or multi-focusing by direct control of the laser beam by shaping pulses
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B23—MACHINE TOOLS; METAL-WORKING NOT OTHERWISE PROVIDED FOR
- B23K—SOLDERING OR UNSOLDERING; WELDING; CLADDING OR PLATING BY SOLDERING OR WELDING; CUTTING BY APPLYING HEAT LOCALLY, e.g. FLAME CUTTING; WORKING BY LASER BEAM
- B23K2103/00—Materials to be soldered, welded or cut
- B23K2103/50—Inorganic material, e.g. metals, not provided for in B23K2103/02 – B23K2103/26
- B23K2103/54—Glass
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P40/00—Technologies relating to the processing of minerals
- Y02P40/50—Glass production, e.g. reusing waste heat during processing or shaping
- Y02P40/57—Improving the yield, e-g- reduction of reject rates
Abstract
The application provides a control method and a related device for glass laser cutting, which comprise the following steps: collecting cutting parameters of sample glass with different component proportions, acoustic emission signals of the sample glass in a cutting process and surface characteristic parameters of the sample glass after cutting is finished; training a convolutional neural network model by taking the cutting parameters, the acoustic emission signals and the surface characteristic parameters as a data training set to obtain a target model in which the cutting parameters, the acoustic emission signals and the surface characteristic parameters are associated with one another; and finishing the cutting of the glass to be processed based on the target model. The prediction of the processing quality of the product under the condition of known process parameters and material characteristics is realized, when the characteristics of the processed material are unknown, the material characteristics can be analyzed according to the characteristics of the collected acoustic emission signals, the optimal laser cutting process parameters are optimized according to the processing requirements, and the processing quality is improved.
Description
Technical Field
The application relates to the technical field of laser processing of hard and brittle materials, in particular to a control method and a related device for glass laser cutting.
Background
The high-transparency glass laser cutting adopts a layer-by-layer processing method from bottom to top, the local temperature of the glass absorbing laser energy rapidly rises near the laser focus range, generates great temperature difference with the peripheral unirradiated area, generates internal stress around the focus, enables the glass to be broken to realize material removal, and completes the three-dimensional cutting of the high-transparency glass along with the planar movement and height change of a laser spot. Because the glass is a typical heterogeneous hard and brittle material, the defects of edge breakage, glass burst, local adhesion and the like can be generated on different glasses by the same laser process parameters, and the real-time monitoring of the whole laser processing process of the high-transparency glass is an effective way for improving the processing quality.
The traditional processing quality monitoring is mainly laser beam quality monitoring and optical microscope observation, the laser beam quality monitoring is to monitor laser power, repetition frequency, pulse width and other beam quality parameters in real time by utilizing a photosensitive resistor and a thermistor in the processing process, and the influence of laser parameter change on product quality is reduced by utilizing the monitoring of the laser beam quality; the optical microscope observation method is used for observing the product quality by adopting an optical microscope after the product is processed, and adaptively adjusting parameters of the processing process. The two methods can improve the processing quality of the product to a certain extent, but the real-time performance and the reliability of monitoring when the product has defects due to the characteristics of the high-transparency glass are difficult to guarantee.
Disclosure of Invention
In view of this, an object of the present application is to provide a control method and a related device for glass laser cutting, which achieve high-performance processing of high-transmittance glass, real-time monitoring of product state, adaptive adjustment of process parameters, and improve product processing quality and processing efficiency. In the process of processing the high-transparency glass by using the laser, the impact of laser pulse, the transfer and release of the internal stress of the material, the material crack and the like affect the generation of elastic waves, different processing technologies and real-time states of products generate different elastic waves, and the acoustic emission sensor monitors the processing quality of the products by measuring the elastic waves in the products in real time. Compared with the traditional measuring method, the acoustic emission sensor monitoring method can effectively reflect the processing state in real time, and remarkably improves the quality and efficiency of the laser processing of the high-transparency glass.
The embodiment of the application provides a control method for glass laser cutting, which comprises the following steps:
collecting cutting parameters of sample glass with different component proportions, acoustic emission signals of the sample glass in a cutting process and surface characteristic parameters of the sample glass after cutting is finished;
training a convolutional neural network model by taking the cutting parameters, the acoustic emission signals and the surface characteristic parameters as a data training set to obtain a target model in which the cutting parameters, the acoustic emission signals and the surface characteristic parameters are associated with one another;
and finishing the cutting of the glass to be processed based on the target model.
Optionally, the step of collecting the cutting parameters of the sample glass with different component ratios, the acoustic emission signal of the sample glass in the cutting process, and the surface characteristic parameters of the sample glass after the cutting is completed includes:
collecting cutting parameters of sample glass with different component proportions under the condition that the laser spot diameters are the same, wherein the cutting parameters comprise: laser process parameters, material characteristic parameters and processing shapes;
collecting acoustic emission signals of the sample glass in a cutting process;
extracting the characteristics of the acoustic emission signals to obtain characteristic data;
and detecting the cut sample glass to obtain the surface characteristic parameters of the sample glass.
Optionally, before the step of training a convolutional neural network model by using the cutting parameter, the acoustic emission signal, and the surface feature parameter as a data training set to obtain a target model in which the cutting parameter, the acoustic emission signal, and the surface feature parameter are associated with each other, the method further includes:
and matching and comparing the cutting parameters, the characteristic data and the surface characteristic characterization parameters based on different target processing quality requirements of sample glass with different component proportions, and establishing the corresponding relation among the target processing quality, the cutting parameters, the characteristic data and the surface characteristic characterization parameters.
Optionally, the step of performing feature extraction on the acoustic emission signal to obtain feature data includes:
respectively carrying out peak frequency operation, short-time Fourier transform operation and root-mean-square value operation on the acoustic emission signal, and respectively acquiring the peak frequency, a time-frequency distribution graph and an RMS (root mean square) value of the acoustic emission signal;
and taking the peak frequency, the time frequency distribution graph and the RMS value as the characteristic data.
Optionally, the step of training a convolutional neural network model by using the cutting parameter, the acoustic emission signal, and the surface feature parameter as a data training set to obtain a target model in which the cutting parameter, the acoustic emission signal, and the surface feature parameter are associated with each other includes:
taking the peak frequency, the time-frequency distribution graph, the RMS value, the laser process parameter, the material characteristic parameter, the machining shape and the surface feature parameter as input parameters;
taking the cutting quality and the cutting efficiency as output parameters;
and training the convolutional neural network model based on the input parameters and the output parameters to obtain the target model.
Optionally, the step of training the convolutional neural network model based on the input parameters and the output parameters to obtain the target model includes:
and training the convolutional neural network model based on the input parameters and the output parameters combined with a gradient propagation algorithm to obtain the target model.
Optionally, the step of training the convolutional neural network model based on the input parameters and the output parameters as well as combining a gradient propagation algorithm to obtain the target model includes:
running forward propagation algorithm to obtain input of each neuron
And output
And calculating an error according to the Loss and the true value on an output layer:
the back propagation algorithm, starting from the output layer, calculates the error of each layer from the front to the back:
calculate the gradient of the weight:
training parameters by a gradient descent optimizer:
wherein, eta is the learning rate,
represents the input of the ith neuron of the l layer in the hidden layer,
sigma represents an activation function for the output of the j-th nerve unit of the (l-1) -th layer after activation;
and finishing the training of the convolutional neural network model through training parameters to obtain the target model.
Optionally, the step of completing the cutting of the glass to be processed based on the target model includes:
when the material of the glass to be processed is unknown, acquiring an acoustic emission signal of the glass to be processed;
substituting the acoustic emission signal of the glass to be processed into the target model to determine the material characteristics of the glass to be processed;
and determining the cutting parameters of the glass to be processed based on the material characteristics of the glass to be processed.
The embodiment of the application also provides a control system for glass laser cutting, which comprises a glass transmission processing mechanical device, a high-performance nanosecond green laser processing device and a control device for glass laser cutting for operating the method.
Optionally, the control device for glass laser cutting includes:
the data acquisition module is used for acquiring cutting parameters of sample glass with different component proportions, acoustic emission signals of the sample glass in the cutting process and surface characteristic parameters of the sample glass after cutting is finished;
the training module is used for training a convolutional neural network model by taking the cutting parameters, the acoustic emission signals and the surface characteristic parameters as a data training set to obtain a target model with the cutting parameters, the acoustic emission signals and the surface characteristic parameters correlated with each other;
a control module for completing the cutting of the glass to be processed based on the target model
An embodiment of the present application further provides an electronic device, including: the control method comprises a processor, a memory and a bus, wherein the memory stores machine readable instructions executable by the processor, the processor and the memory are communicated through the bus when the electronic device runs, and the machine readable instructions are executed by the processor to execute the steps of the control method for glass laser cutting.
Embodiments of the present application also provide a computer-readable storage medium, on which a computer program is stored, and the computer program is executed by a processor to execute the steps of the control method for glass laser cutting as described above.
The control method and device for glass laser cutting provided by the embodiment of the application realize high-performance processing of high-transparency glass, real-time monitoring of product states and adaptive adjustment of process parameters, and improve the processing quality and processing efficiency of products. In the process of processing the high-transparency glass by using the laser, the impact of laser pulse, the transfer and release of the internal stress of the material, the material crack and the like affect the generation of elastic waves, different processing technologies and real-time states of products generate different elastic waves, and the acoustic emission sensor monitors the processing quality of the products by measuring the elastic waves in the products in real time. Compared with the traditional measuring method, the acoustic emission sensor monitoring method can effectively reflect the processing state in real time, and remarkably improves the quality and efficiency of the laser processing of the high-transparency glass.
In order to make the aforementioned objects, features and advantages of the present application more comprehensible, preferred embodiments accompanied with figures are described in detail below.
Drawings
To more clearly illustrate the technical solutions of the embodiments of the present application, the drawings needed in the embodiments will be briefly described below, it should be understood that the following drawings only illustrate some embodiments of the present application and therefore should not be considered as limiting the scope, and those skilled in the art can also obtain other related drawings based on the drawings without inventive efforts.
FIG. 1 is a flow chart illustrating a control method for laser cutting of glass according to an embodiment of the present disclosure;
FIG. 2 is a flow chart illustrating model training in a control method for glass laser cutting according to an embodiment of the present disclosure;
FIG. 3 is a schematic structural diagram of a control device for laser cutting of glass according to an embodiment of the present disclosure;
FIG. 4 is a schematic diagram illustrating the operation of another control system for laser cutting of glass according to an embodiment of the present disclosure;
fig. 5 shows a schematic structural diagram of an electronic device provided in an embodiment of the present application.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all the embodiments. The components of the embodiments of the present application, generally described and illustrated in the figures herein, can be arranged and designed in a wide variety of different configurations. Thus, the following detailed description of the embodiments of the present application, presented in the accompanying drawings, is not intended to limit the scope of the claimed application, but is merely representative of selected embodiments of the application. Every other embodiment that can be obtained by a person skilled in the art without making creative efforts based on the embodiments of the present application falls within the protection scope of the present application.
First, an application scenario to which the present application is applicable will be described. The method and the device can be applied to laser processing of hard and brittle material scenes.
Research shows that most of the traditional processing quality monitoring is laser beam quality monitoring and optical microscope observation, the laser beam quality monitoring is to utilize a photosensitive resistor and a thermistor to monitor laser power, repetition frequency, pulse width and other beam quality parameters in real time in the processing process, and the monitoring on the laser beam quality is utilized to reduce the influence of the laser parameter change on the product quality; the optical microscope observation method is used for observing the product quality by adopting an optical microscope after the product is processed, and adaptively adjusting parameters of the processing process. The two methods can improve the processing quality of the product to a certain extent, but the real-time performance and the reliability of monitoring when the product has defects due to the characteristics of the high-transparency glass are difficult to guarantee.
Based on this, the embodiment of the application provides a control method for glass laser cutting, so as to realize high-performance processing of high-transparency glass, real-time monitoring of product state, adaptive adjustment of process parameters, and improve product processing quality and processing efficiency.
The control method for glass laser cutting provided by the embodiment of the application is shown in fig. 1 and comprises the following steps:
s101, collecting cutting parameters of sample glass with different component proportions, acoustic emission signals of the sample glass in a cutting process and surface characteristic parameters of the sample glass after cutting is finished;
s102, training a convolutional neural network model by taking the cutting parameters, the acoustic emission signals and the surface characteristic parameters as a data training set to obtain a target model with the cutting parameters, the acoustic emission signals and the surface characteristic parameters correlated with each other;
s103, finishing the cutting of the glass to be processed based on the target model.
Illustratively, high-transparency glass (such as soda-lime glass, high-alumina glass and the like) with different component ratios is selected and cut into rectangular blocks with the size of 100 x 200mm, and the experimental material is placed below the galvanometer and the acoustic emission sensor is fixed. Traversing laser process parameters (repetition frequency, pulse width, power, scanning path distance, elevation height of each layer of focus, scanning speed), processing shapes (circle, rectangle, triangle and polygon), carrying out laser cutting high-transparency glass experiment, controlling each layer of a galvanometer to carry out scanning cutting and lifting laser focus layer by layer to form a complete cut hole, acquiring acoustic emission signals generated by the glass due to the influences of laser impact, transfer release of material internal stress, material crack and the like in the experimental process, wherein the acquired signals are X (t), and carrying out peak frequency, short-time Fourier transform (STFT) and root mean square value (RMS) operation:
and solving the peak frequency, the time-frequency distribution graph and the RMS of each group of experiment acoustic emission signals, carrying out measurement and characterization on an optical microscope, an SEM and residual stress of the processed experiment material, and measuring the edge breakage amount, the adhesion degree, the breakage area, the surface roughness, the surface microstructure, the residual stress and other data of the experiment sample.
Model training is carried out on all the collected data, and the measured data are integrated into a data training set { (x) (1) ,y (1) ,…)…(x (m) ,y (m) …), first initialize the matrix theta between each layer, where
In order to avoid the gradient descent failure caused by the initialization of the weight or parameter to 0, the symmetry is broken by initializing Θ through a random Gaussian distribution, wherein the weight is connected from the jth nerve unit of the (l-1) th layer to the ith nerve unit of the l-th layer.
In a possible embodiment, the step of acquiring the cutting parameters of the sample glass with different composition ratios, the acoustic emission signal of the sample glass during cutting and the surface characteristic parameters of the sample glass after cutting is completed comprises the following steps:
collecting cutting parameters of sample glass with different component proportions under the condition that the laser spot diameters are the same, wherein the cutting parameters comprise: laser process parameters, material characteristic parameters and processing shapes;
collecting acoustic emission signals of the sample glass in a cutting process;
extracting the characteristics of the acoustic emission signals to obtain characteristic data;
and detecting the cut sample glass to obtain the surface characteristic parameters of the sample glass.
In a possible embodiment, before the step of training the convolutional neural network model using the cutting parameter, the acoustic emission signal, and the surface feature parameter as a data training set to obtain a target model in which the cutting parameter, the acoustic emission signal, and the surface feature parameter are associated with each other, the method further includes:
and matching and comparing the cutting parameters, the characteristic data and the surface characteristic characterization parameters based on different target processing quality requirements of sample glass with different component proportions, and establishing the corresponding relation among the target processing quality, the cutting parameters, the characteristic data and the surface characteristic characterization parameters.
In a possible embodiment, the step of extracting the feature of the acoustic emission signal to obtain the feature data includes:
respectively carrying out peak frequency operation, short-time Fourier transform operation and root-mean-square value operation on the acoustic emission signal, and respectively acquiring the peak frequency, a time-frequency distribution graph and an RMS (root mean square) value of the acoustic emission signal;
and taking the peak frequency, the time frequency distribution graph and the RMS value as the characteristic data.
In one possible embodiment, the step of training a convolutional neural network model using the cutting parameters, the acoustic emission signals, and the surface feature parameters as a data training set to obtain an object model in which the cutting parameters, the acoustic emission signals, and the surface feature parameters are associated with each other includes:
taking the peak frequency, the time-frequency distribution graph, the RMS value, the laser process parameter, the material characteristic parameter, the machining shape and the surface feature parameter as input parameters;
taking the cutting quality and the cutting efficiency as output parameters;
and training the convolutional neural network model based on the input parameters and the output parameters to obtain the target model.
In one possible embodiment, the step of training the convolutional neural network model based on the input parameters and the output parameters to obtain the target model includes:
and training the convolutional neural network model based on the input parameters and the output parameters combined with a gradient propagation algorithm to obtain the target model.
In a possible embodiment, the step of training the convolutional neural network model based on the input parameters and the output parameters as well as combining a gradient propagation algorithm to obtain the target model includes:
running forward propagation algorithm to obtain input of each neuron
And output
And calculating an error according to the Loss and the true value on an output layer:
the back propagation algorithm, starting from the output layer, calculates the error of each layer from the front to the back:
calculate the gradient of the weight:
training parameters by a gradient descent optimizer:
wherein, eta is the learning rate,
represents the input of the ith neuron of the l layer in the hidden layer,
sigma represents an activation function for the output of the j-th nerve unit of the (l-1) -th layer after activation;
and finishing the training of the convolutional neural network model through training parameters to obtain the target model.
In a possible embodiment, the step of completing the cutting of the glass to be processed based on the target model comprises:
when the material of the glass to be processed is unknown, acquiring an acoustic emission signal of the glass to be processed;
substituting the acoustic emission signal of the glass to be processed into the target model to determine the material characteristics of the glass to be processed;
and determining the cutting parameters of the glass to be processed based on the material characteristics of the glass to be processed.
Illustratively, the acquired acoustic emission signal peak frequency, the time-frequency distribution graph, the RMS value, the laser process parameter, the material characteristic parameter, and the processed product measurement characterization parameter are used as input, the cutting quality and the cutting efficiency are used as output, and a gradient propagation algorithm is combined to train a deep learning model, as shown in fig. 3.
The trained model can obtain unknown processing technology state and product quality when the processing parameters are input, and the prediction of the product processing quality under the condition of known technology parameters and material characteristics is realized; by utilizing the mathematical model, when the characteristics of the processed material are unknown, the characteristics of the material can be analyzed according to the characteristics of the collected acoustic emission signals, the optimal laser cutting process parameters are optimized according to the processing requirements, and the processing quality is improved.
After the high-transparency glass reaches the position to be processed, the supporting cylinder jacks up, the negative pressure pipeline vacuumizes to adsorb the sucker on the glass, and then the servo motor rotates to convey the glass to the processing station under the support of the supporting wheel through the linear module. In the conveying process, the control system controls the linear motor to move the laser and the galvanometer to be right above the processing hole position, and the glass is processed after being moved in place.
After the glass moves to the processing position, the clamp drives the acoustic emission sensor to descend, and the sensor is fixed on the glass. The control system controls the laser to emit light according to the original set parameters, controls the galvanometer to reflect the laser to the lower surface of the glass according to the preset processing pattern, and lifts the focus according to layers along with the processing.
Before the machining starts, an acoustic emission sensor is started to measure acoustic emission signals, the measured signals are processed in real time, peak frequency, short-time Fourier transform (STFT) and root mean square value (RMS) operation is carried out, various calculated numerical values are substituted into a previously established convolution neural network calculation model, and the current high-transmittance glass laser machining quality is predicted. And when the system judges that the processing quality does not meet the requirement, the laser processing parameters are adjusted on line, and the process self-adaptive optimization is carried out.
And after the processing is finished, the linear module sends away the glass, and the whole process of laser cutting of the high-transparency glass is finished.
And high-quality and high-efficiency processing of high-transparency glass laser is realized.
The embodiment of the application also provides a control system for glass laser cutting, which comprises a glass transmission processing mechanical device, a high-performance nanosecond green laser processing device and a control device for glass laser cutting for operating the method.
As shown in fig. 4, the high-precision high-transparency glass stable conveying mechanical device comprises: the casting production reference platform is as the supporting component of device, on work platform installation and supporting platform, eight aluminium alloy and 48 group supporting wheel structure installations and work platform, four two liang of unanimous sharp module evenly distributed installations and work platform of length, install the support frame on every sharp module to with support cylinder, negative sucker fastening together, supporting conveyor installs the servo motor, positive pressure gas circuit pipeline, negative pressure gas circuit pipeline and the supporting beam of drive sharp module. The conveying device conveys the glass to a laser processing working area to perform laser cutting and acoustic emission signal acquisition work after the glass is firmly adsorbed by adopting different washing trays aiming at the adaptability of the high-transparency glass with different models.
The high-performance nanosecond green laser cutting device comprises: a marble platform for supporting installs on work platform, and the marble platform is inside to be inlayed the permanent magnet and is used for drive control linear electric motor, and linear electric motor installation and marble platform are last, and the permanent magnet in inside with the nestification mutually supports and realizes reciprocating motion, and the laser instrument passes through the installation version fastening on linear electric motor, and two-sided speculum passes through the jackscrew and installs on the mounting panel, shakes mirror installation and second speculum the place ahead. High-frequency pulse laser emitted by a laser enters the vibrating mirror through the two-sided reflector, the vibrating mirror focuses light beams on the lower surface of the processed glass through the internal reflector and the dynamic focusing shaft, and the light spots move according to a preset track along with the processing to realize the cutting of the high-transmittance glass. The matched industrial personal computer and the control board card are used for controlling the light-emitting technological parameters of the laser, and high-quality and high-efficiency processing of products is realized by adjusting the technological parameters.
The control device for glass laser cutting comprises: the acoustic emission sensor is arranged on the clamp, when the high-transparency glass is positioned at the processing station, the clamp is put down, the acoustic emission sensor is tightly attached to the surface of the glass, the preamplifier is connected with the sensor through a data line, the multi-channel data acquisition card is arranged on the industrial personal computer and connected with the preamplifier, and the acoustic emission sensor acquires acoustic emission signals and stores the data in the industrial personal computer through the preamplifier and the signal acquisition card in the process of processing the high-transparency glass by laser. By covering all processing parameters with the same laser spot diameter: the method comprises the following steps of carrying out laser cutting high-transparency glass experiment on laser process parameters (repetition frequency, pulse width, power, scanning path distance, elevation height of each layer of focus, scanning speed), material characteristic parameters (density, component proportion, microstructure, light transmittance and thickness) and processing shapes (circle, rectangle, triangle and polygon), collecting acoustic emission signals in the experiment process, carrying out feature extraction on the collected acoustic emission signals, carrying out edge breakage amount, adhesion degree, breakage area, surface roughness, surface microstructure and residual stress detection experiments on processed products, and obtaining surface feature characterization parameters of the processed products. And matching and comparing the process parameters, the acoustic emission signal characteristic parameters and the surface characteristic characterization parameters to establish the corresponding relation among the product processing quality, the process parameters, the material characteristics and the acoustic emission signal characteristics. And (3) carrying out model construction on the parameters by utilizing deep learning, and establishing a mathematical model with highly-associated processing technological parameters, acoustic emission signal characteristics and processing characterization quality. By utilizing the mathematical model, the processing technology state and the product quality of the position can be obtained when the processing parameters are input, and the prediction of the product processing quality under the condition of known technology parameters and material characteristics is realized; by utilizing the mathematical model, when the characteristics of the processed material are unknown, the characteristics of the material can be analyzed according to the characteristics of the collected acoustic emission signals, the optimal laser cutting process parameters are optimized according to the processing requirements, and the processing quality is improved.
Further, the linear module and the linear motor convey the glass to different positions according to different processing hole site requirements of the high-transparency glass, the laser moves to the corresponding area, and the full-width high-precision processing of the high-transparency glass is achieved.
Furthermore, the control system controls the galvanometer to complete different scanning paths according to the processing requirements of different product patterns, and the multi-pattern self-adaptive processing is realized.
In a possible embodiment, the control device for glass laser cutting, as shown in fig. 3, comprises:
the data acquisition module 201 is used for acquiring cutting parameters of sample glass with different component proportions, acoustic emission signals of the sample glass in a cutting process and surface characteristic parameters of the sample glass after cutting is finished;
a training module 202, configured to train a convolutional neural network model using the cutting parameter, the acoustic emission signal, and the surface feature parameter as a data training set, so as to obtain a target model in which the cutting parameter, the acoustic emission signal, and the surface feature parameter are associated with each other;
and the control module 203 is used for finishing the cutting of the glass to be processed based on the target model.
Specifically, the laser cutting control device includes: the system comprises an acoustic emission sensor, a clamp, a multi-channel data acquisition card, a preamplifier and an industrial personal computer;
the acoustic emission sensor is mounted on the clamp, when glass to be processed is located at a processing position, the clamp is put down, the acoustic emission sensor is attached to the surface of the glass to be processed, the preamplifier is connected with the acoustic emission sensor through a data line, the multi-channel data acquisition card is mounted on the industrial personal computer and connected with the preamplifier, the acoustic emission sensor acquires acoustic emission signals in the process of laser processing of the glass to be processed, and the acoustic emission sensor stores the acquired moral acoustic emission signals in the industrial personal computer through the preamplifier and the signal acquisition card.
Referring to fig. 5, fig. 5 is a schematic structural diagram of an electronic device according to an embodiment of the present disclosure. As shown in fig. 5, the electronic device 500 includes a processor 510, a memory 520, and a bus 530.
The memory 520 stores machine-readable instructions executable by the processor 510, when the electronic device 500 runs, the processor 510 communicates with the memory 520 through the bus 530, and when the machine-readable instructions are executed by the processor 510, the steps of the laser cutting control method in the embodiment of the method shown in fig. 1 may be executed.
An embodiment of the present application further provides a computer-readable storage medium, where a computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, the steps of the laser cutting control method in the method embodiment shown in fig. 1 may be executed.
It is clear to those skilled in the art that, for convenience and brevity of description, the specific working processes of the above-described systems, apparatuses and units may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.
In the several embodiments provided in the present application, it should be understood that the disclosed system, apparatus and method may be implemented in other ways. The above-described embodiments of the apparatus are merely illustrative, and for example, the division of the units is only one logical division, and there may be other divisions when actually implemented, and for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection of devices or units through some communication interfaces, and may be in an electrical, mechanical or other form.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
In addition, functional units in the embodiments of the present application may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit.
The functions, if implemented in the form of software functional units and sold or used as a stand-alone product, may be stored in a non-volatile computer-readable storage medium executable by a processor. Based on such understanding, the technical solution of the present application or portions thereof that substantially contribute to the prior art may be embodied in the form of a software product stored in a storage medium and including instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present application. And the aforementioned storage medium includes: various media capable of storing program codes, such as a usb disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk, or an optical disk.
Finally, it should be noted that: the above-mentioned embodiments are only specific embodiments of the present application, and are used for illustrating the technical solutions of the present application, but not limiting the same, and the scope of the present application is not limited thereto, and although the present application is described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: any person skilled in the art can modify or easily conceive the technical solutions described in the foregoing embodiments or equivalent substitutes for some technical features within the technical scope disclosed in the present application; such modifications, changes or substitutions do not depart from the spirit and scope of the exemplary embodiments of the present application, and are intended to be covered by the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.
Claims (10)
1. A control method for glass laser cutting is characterized by comprising the following steps:
collecting cutting parameters of sample glass with different component proportions, acoustic emission signals of the sample glass in a cutting process and surface characteristic parameters of the sample glass after cutting is finished;
training a convolutional neural network model by taking the cutting parameters, the acoustic emission signals and the surface characteristic parameters as a data training set to obtain a target model in which the cutting parameters, the acoustic emission signals and the surface characteristic parameters are associated with one another;
and finishing the cutting of the glass to be processed based on the target model.
2. The method for controlling the glass laser cutting according to claim 1, wherein the step of collecting the cutting parameters of the sample glass with different component proportions, the acoustic emission signal of the sample glass during the cutting process and the surface characteristic parameters of the sample glass after the cutting is completed comprises the following steps:
collecting cutting parameters of sample glass with different component proportions under the condition that the laser spot diameters are the same, wherein the cutting parameters comprise: laser process parameters, material characteristic parameters and processing shapes;
collecting acoustic emission signals of the sample glass in a cutting process;
extracting the characteristics of the acoustic emission signals to obtain characteristic data;
and detecting the cut sample glass to obtain the surface characteristic parameters of the sample glass.
3. The method for controlling glass laser cutting according to claim 2, wherein before the step of training the convolutional neural network model with the cutting parameters, the acoustic emission signals and the surface feature parameters as a data training set to obtain a target model with the cutting parameters, the acoustic emission signals and the surface feature parameters being associated with each other, the method further comprises:
and matching and comparing the cutting parameters, the characteristic data and the surface characteristic characterization parameters based on different target processing quality requirements of sample glass with different component proportions, and establishing the corresponding relation among the target processing quality, the cutting parameters, the characteristic data and the surface characteristic characterization parameters.
4. The method for controlling glass laser cutting according to claim 2, wherein the step of extracting the characteristics of the acoustic emission signal to obtain characteristic data comprises:
respectively carrying out peak frequency operation, short-time Fourier transform operation and root-mean-square value operation on the acoustic emission signal, and respectively acquiring the peak frequency, a time-frequency distribution graph and an RMS (root mean square) value of the acoustic emission signal;
and taking the peak frequency, the time frequency distribution graph and the RMS value as the characteristic data.
5. The method for controlling glass laser cutting according to claim 4, wherein the step of training a convolutional neural network model with the cutting parameters, the acoustic emission signals and the surface feature parameters as a data training set to obtain a target model with the cutting parameters, the acoustic emission signals and the surface feature parameters correlated with each other comprises:
taking the peak frequency, the time-frequency distribution graph, the RMS value, the laser process parameter, the material characteristic parameter, the machining shape and the surface feature parameter as input parameters;
taking the cutting quality and the cutting efficiency as output parameters;
and training the convolutional neural network model based on the input parameters and the output parameters to obtain the target model.
6. The method of claim 5, wherein the step of training the convolutional neural network model based on the input parameters and the output parameters to obtain the target model comprises:
and training the convolutional neural network model based on the input parameters and the output parameters combined with a gradient propagation algorithm to obtain the target model.
7. The method for controlling glass laser cutting according to claim 6, wherein the step of training the convolutional neural network model based on the input parameters and the output parameters in combination with a gradient propagation algorithm to obtain the target model comprises:
running forward propagation algorithm to obtain input of each neuron
And output
And calculating an error according to the Loss and the true value on an output layer:
the back propagation algorithm, starting from the output layer, calculates the error of each layer from the front to the back:
calculate the gradient of the weight:
training parameters by a gradient descent optimizer:
wherein, eta is the learning rate,
represents the input of the ith neuron of the l layer in the hidden layer,
sigma represents an activation function for the output of the j-th nerve unit of the (l-1) -th layer after activation;
and finishing the training of the convolutional neural network model through training parameters to obtain the target model.
8. The method for controlling glass laser cutting according to claim 1, wherein the step of completing the cutting of the glass to be processed based on the target model comprises:
when the material of the glass to be processed is unknown, acquiring an acoustic emission signal of the glass to be processed;
substituting the acoustic emission signal of the glass to be processed into the target model to determine the material characteristics of the glass to be processed;
and determining the cutting parameters of the glass to be processed based on the material characteristics of the glass to be processed.
9. A control system for laser cutting of glass comprising glass transport processing mechanical means, high performance nanosecond green laser processing means and control means for laser cutting of glass for operating the method of claims 1-8.
10. Control of glass laser cutting according to claim 9, characterized by comprising:
the data acquisition module is used for acquiring cutting parameters of sample glass with different component ratios, acoustic emission signals of the sample glass in a cutting process and surface characteristic parameters of the sample glass after cutting is finished;
the training module is used for training a convolutional neural network model by taking the cutting parameters, the acoustic emission signals and the surface characteristic parameters as a data training set to obtain a target model with the cutting parameters, the acoustic emission signals and the surface characteristic parameters correlated with each other;
and the control module is used for finishing the cutting of the glass to be processed based on the target model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111627695.2A CN114799495B (en) | 2021-12-28 | 2021-12-28 | Laser cutting control method and related device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111627695.2A CN114799495B (en) | 2021-12-28 | 2021-12-28 | Laser cutting control method and related device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114799495A true CN114799495A (en) | 2022-07-29 |
| CN114799495B CN114799495B (en) | 2023-06-13 |
Family
ID=82527826
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111627695.2A Active CN114799495B (en) | 2021-12-28 | 2021-12-28 | Laser cutting control method and related device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114799495B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115859572A (en) * | 2022-11-08 | 2023-03-28 | 深圳市深科达智能装备股份有限公司 | Material cutting scheme creation method, readable storage medium, and computer device |
| CN116944700A (en) * | 2023-08-31 | 2023-10-27 | 深圳市军志丰精密科技有限公司 | Detection and control method and system for laser cutting |
Citations (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20040251290A1 (en) * | 2001-07-25 | 2004-12-16 | Kondratenko Vladimir Stepanovich | Cutting method for brittle non-metallic materials (two variants) |
| CN101160607A (en) * | 2005-04-19 | 2008-04-09 | 康宁股份有限公司 | Acoustic emission system and method for on-line measurement of glass break energy |
| WO2008142745A1 (en) * | 2007-05-10 | 2008-11-27 | Toray Engineering Co., Ltd. | Laser scribing device and laser scribing method |
| CN103776905A (en) * | 2014-02-11 | 2014-05-07 | 哈尔滨工业大学 | Laser-induced hard and brittle material thermal-cracking cutting apparatus and crack detection method |
| CN103852525A (en) * | 2012-11-29 | 2014-06-11 | 沈阳工业大学 | Acoustic emission signal identification method based on AR-HMM |
| CN104044069A (en) * | 2014-07-03 | 2014-09-17 | 上海理工大学 | Optical workpiece subsurface damage depth prediction method based on acoustic emission signal |
| CN107199397A (en) * | 2016-03-17 | 2017-09-26 | 发那科株式会社 | Machine learning device, laser-processing system and machine learning method |
| WO2017216063A1 (en) * | 2016-06-15 | 2017-12-21 | Empa Eidgenössische Materialprüfungs- Und Forschungsanstalt | Quality control of laser welding process |
| 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 |
| RU2688656C1 (en) * | 2018-07-06 | 2019-05-21 | Общество с ограниченной ответственностью "Научно-исследовательское предприятие Лазерные технологии" | Method of cutting brittle non-metallic materials |
| CN109848141A (en) * | 2019-01-14 | 2019-06-07 | 中国电力科学研究院有限公司 | Insulator RTV coating laser sweep-out method and device |
| CN110799830A (en) * | 2017-04-27 | 2020-02-14 | 多元生物学公司 | Orthogonal multi-biometric sensing and imaging system |
| CN111251474A (en) * | 2020-02-21 | 2020-06-09 | 山东理工大学 | Ceramic laser turning composite plastic processing method based on acoustic emission signal characteristic identification and automatic matching of processing parameters |
| JP2020191597A (en) * | 2019-05-23 | 2020-11-26 | 太陽誘電株式会社 | Elastic wave device and manufacturing method thereof |
| CN213052896U (en) * | 2020-08-14 | 2021-04-27 | 湖南化工职业技术学院（湖南工业高级技工学校） | Milling system |
| US20220129757A1 (en) * | 2020-10-26 | 2022-04-28 | Jiangxi University Of Science And Technology | Prediction method, device and system for rock mass instability stages |
| US20220411314A1 (en) * | 2019-09-30 | 2022-12-29 | Hegla Gmbh & Co. Kg | Cutting method and cutting device for scoring components made of glass or ceramic, and method for splitting components made of glass or ceramic |
-
2021
- 2021-12-28 CN CN202111627695.2A patent/CN114799495B/en active Active
Patent Citations (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20040251290A1 (en) * | 2001-07-25 | 2004-12-16 | Kondratenko Vladimir Stepanovich | Cutting method for brittle non-metallic materials (two variants) |
| CN101160607A (en) * | 2005-04-19 | 2008-04-09 | 康宁股份有限公司 | Acoustic emission system and method for on-line measurement of glass break energy |
| WO2008142745A1 (en) * | 2007-05-10 | 2008-11-27 | Toray Engineering Co., Ltd. | Laser scribing device and laser scribing method |
| CN103852525A (en) * | 2012-11-29 | 2014-06-11 | 沈阳工业大学 | Acoustic emission signal identification method based on AR-HMM |
| CN103776905A (en) * | 2014-02-11 | 2014-05-07 | 哈尔滨工业大学 | Laser-induced hard and brittle material thermal-cracking cutting apparatus and crack detection method |
| CN104044069A (en) * | 2014-07-03 | 2014-09-17 | 上海理工大学 | Optical workpiece subsurface damage depth prediction method based on acoustic emission signal |
| CN107199397A (en) * | 2016-03-17 | 2017-09-26 | 发那科株式会社 | Machine learning device, laser-processing system and machine learning method |
| WO2017216063A1 (en) * | 2016-06-15 | 2017-12-21 | Empa Eidgenössische Materialprüfungs- Und Forschungsanstalt | Quality control of laser welding process |
| CN110799830A (en) * | 2017-04-27 | 2020-02-14 | 多元生物学公司 | Orthogonal multi-biometric sensing and imaging system |
| RU2688656C1 (en) * | 2018-07-06 | 2019-05-21 | Общество с ограниченной ответственностью "Научно-исследовательское предприятие Лазерные технологии" | Method of cutting brittle non-metallic materials |
| 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 |
| CN109848141A (en) * | 2019-01-14 | 2019-06-07 | 中国电力科学研究院有限公司 | Insulator RTV coating laser sweep-out method and device |
| JP2020191597A (en) * | 2019-05-23 | 2020-11-26 | 太陽誘電株式会社 | Elastic wave device and manufacturing method thereof |
| US20220411314A1 (en) * | 2019-09-30 | 2022-12-29 | Hegla Gmbh & Co. Kg | Cutting method and cutting device for scoring components made of glass or ceramic, and method for splitting components made of glass or ceramic |
| CN111251474A (en) * | 2020-02-21 | 2020-06-09 | 山东理工大学 | Ceramic laser turning composite plastic processing method based on acoustic emission signal characteristic identification and automatic matching of processing parameters |
| CN213052896U (en) * | 2020-08-14 | 2021-04-27 | 湖南化工职业技术学院（湖南工业高级技工学校） | Milling system |
| US20220129757A1 (en) * | 2020-10-26 | 2022-04-28 | Jiangxi University Of Science And Technology | Prediction method, device and system for rock mass instability stages |
Non-Patent Citations (1)
| Title |
|---|
| 王春明: "国外激光切割实时监测技术的最新进展", 激光技术, vol. 25, no. 6, pages 412 - 416 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115859572A (en) * | 2022-11-08 | 2023-03-28 | 深圳市深科达智能装备股份有限公司 | Material cutting scheme creation method, readable storage medium, and computer device |
| CN115859572B (en) * | 2022-11-08 | 2023-09-26 | 深圳市深科达智能装备股份有限公司 | Material cutting scheme creation method, readable storage medium, and computer device |
| CN116944700A (en) * | 2023-08-31 | 2023-10-27 | 深圳市军志丰精密科技有限公司 | Detection and control method and system for laser cutting |
| CN116944700B (en) * | 2023-08-31 | 2024-02-09 | 深圳市军志丰精密科技有限公司 | Detection and control method and system for laser cutting |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114799495B (en) | 2023-06-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114799495A (en) | Control method and related device for laser cutting | |
| CN106018288B (en) | The method of the online non-destructive testing increasing material manufacturing part of laser-ultrasound | |
| WO2018000586A1 (en) | Laser processing method and device for removing thin film or coating | |
| JP7407108B2 (en) | Apparatus and method for performing shock excitation techniques | |
| TW200822382A (en) | Method for structuring solar modules and structuring device | |
| JP2015518535A5 (en) | ||
| CN105382259A (en) | Selective laser fusion composite manufacturing machine tool and workpiece composite manufacturing method | |
| US20130081246A1 (en) | Systems and methods for in-process non-contact optical surface roughness measurement | |
| CN106324101A (en) | Laser shock strengthening on-line quality monitoring system and method based on sound pressure characteristic | |
| JP2002246304A5 (en) | ||
| CN104697444A (en) | Detecting system | |
| CN107543830B (en) | Detection device and detection method for surface defects of spherical optical element | |
| CN102419157A (en) | Micro-depth-dimension automatic image measuring system | |
| CN102494712B (en) | Method and device for detecting loading process and attenuation law of laser shock wave | |
| TW202146140A (en) | Laser processing apparatus facilitating directed inspection of laser-processed workpieces and methods of operating the same | |
| CN116346870B (en) | Industrial Internet of things system based on real-time monitoring of processing quality and control method | |
| WO2020199207A1 (en) | Surface defect optical inspection method and related device | |
| WO2018054449A1 (en) | Method for characterising a material with layered structure and a measuring system | |
| CN108871245B (en) | Measuring device | |
| CN113960032B (en) | Online laser cleaning effect detection method and three-dimensional detection device | |
| CN112666254B (en) | Active safety detection method for service state of glass curtain wall based on intelligent vision and big data | |
| JP2007333545A (en) | Sound absorption characteristic measuring method and sound absorption characteristic measuring apparatus | |
| EP2893339B1 (en) | Floating head contour following holder for ultrasonic testing | |
| IT201900018086A1 (en) | SHEET METAL WORKING MACHINE AND RELATIVE PROCESSING METHOD | |
| JP2013139362A (en) | Device for assisting with setting of manufacturing condition for silica glass crucible, and data obtainable by means of the device |
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 |