Glass cutting device based on virtual simulation technology and error compensation method

Abstract

The invention discloses a glass cutting device based on a virtual simulation technology and an error compensation method, and relates to the technical field of glass cutting equipment. A raw material model is obtained through three-dimensional modeling, then a standard cutting path is planned according to the contour line of a sample model, a real-time cutting path is drawn according to real-time displacement, path calculation is performed according to the real-time cutting path and the standard cutting path to obtain a path deviation value, and meanwhile, feature extraction is performed on a real-time image to obtain a cutting trace. Performing image integration according to the cut time sequence characteristics to obtain a real-time sample trace graph, establishing a trace comparison model according to a convolutional neural network, importing the real-time sample trace graph into the trace comparison model, and outputting a trace deviation value; and the error degree in the cutting process is judged according to a preset path deviation judgment threshold value and a trace deviation judgment threshold value so as to generate a deviation correction signal, and a deviation correction scheme is determined according to the source of the deviation correction signal to conduct error compensation on the cutting process.

Description

Technical Field

[0001] The present invention relates to the technical field of glass cutting equipment, and in particular to a glass cutting device and an error compensation method based on virtual simulation technology. Background Art

[0002] Generally used glass cutting methods generally utilize the following processes: First, a scribing line is formed compared to completely cutting the glass, a fragile part is mechanically formed, and physical or thermal shock is applied to this part to cut it. In the existing processes for forming scribing lines on a glass substrate, there are mechanical processing methods and optical processing methods.

[0003] The mechanical processing method utilizes a cutting tool with a hardness stronger than that of the workpiece. For example, the following method is widely used: A diamond blade formed on the circumference of a disk with a predetermined diameter contacts the glass plate along the cutting path to form a scribing line on the surface of the glass substrate. However, the glass has a high hardness and is relatively fragile to the above mechanical shock, and there are disadvantages such as poor cutting states such as scratches in unwanted directions. Moreover, there is the complexity of dealing with the glass dust flying in the process, and there is also a cost burden based on the wear of the diamond wheel;

[0004] Existing cutting devices cut according to the pre-planned path. However, due to equipment failures or external interferences, the cutting path may deviate. At the same time, due to the deviation of the tool head angle during the cutting process, the cutting effect may also not meet the standard. However, it is difficult for the existing technology to detect deviations in a timely manner during the cutting process, resulting in a too high defective product rate;

[0005] In view of the above technical defects, a solution is now proposed. Summary of the Invention

[0006] The purpose of the present invention is to: judge the degree of error during the cutting process to generate a deviation correction signal, and determine a deviation correction scheme according to the source of the deviation correction signal to perform error compensation on the cutting process.

[0007] To achieve the above purpose, the present invention adopts the following technical solution: A glass cutting device based on virtual simulation technology, including a device body and a cutting tool head. A longitudinal axis driving mechanism is provided on the top surface of the device body. A transverse axis driving mechanism is fixedly provided on the top surface of the longitudinal axis driving mechanism. A tool head support is fixedly provided on the top surface of the transverse axis driving mechanism. A tool head driving mechanism is fixedly provided on the outer surface of the tool head support. A tool head fixing seat is fixedly provided on the outer surface of the output end of the tool head driving mechanism. The cutting tool head is connected to the inside of the tool head fixing seat through an adjustable mechanism. A purging mechanism is fixedly provided on the outer surface of the tool head support. A horizontal fixing plate is fixedly provided on the outer surface of the tool head fixing seat. A camera is fixedly provided on the bottom surface of the horizontal fixing plate.

[0008] Further, the adjustable mechanism includes a limit clamping shaft and a limit assembly. A limit clamping groove is formed on the bottom surface of the tool head fixing seat. The cutting tool head is movably connected to the inner wall of the limit clamping groove. A through groove is formed inside the cutting tool head. Both ends of the limit clamping shaft are movably connected to the inside of the tool head fixing seat and extend to the outer surface of the tool head fixing seat respectively. The limit assembly is fixedly arranged on one end surface of the limit clamping shaft. A passive gear is connected to the other end surface of the limit clamping shaft. A driving motor is fixedly arranged on the outer surface of the tool head fixing seat. A driving gear is connected to the outer surface of the output end of the driving motor. The driving gear meshes with the passive gear, and the radius of the driving gear is smaller than that of the passive gear.

[0009] Further, the limit assembly includes a first electromagnet and a magnet pressing ring. An activity long groove is formed on one end surface of the limit clamping shaft. A connecting shaft is fixedly arranged on the inner wall of the activity long groove. A reset spring is jointly connected between one end surface of the connecting shaft and the activity long groove. The first electromagnet is fixedly arranged on the other end surface of the connecting shaft. The magnet pressing ring is fixedly arranged on the outer surface of the limit clamping shaft. A second electromagnet is fixedly arranged inside the tool head fixing seat.

[0010] Further, the corresponding surfaces of the first electromagnet and the second electromagnet coincide with each other and the magnetic properties of the corresponding surfaces are opposite. The corresponding surfaces of the first electromagnet and the magnet ring coincide with each other and the magnetic properties of the corresponding surfaces are the same. The second electromagnet and the magnet ring do not coincide.

[0011] Further, the purging mechanism includes an air pump and a conveying pipeline. The two conveying pipelines are respectively connected to the outer surface of the output end of the air pump. Jet nozzles are fixedly arranged on the end surfaces of the two conveying pipelines. The two jet nozzles are arranged obliquely alternately.

[0012] An error compensation method for a glass cutting device based on virtual simulation technology includes the following steps:

[0013] Step 1: Obtain the size data of the glass raw material to be cut and the size data of the glass finished product. Perform three-dimensional modeling based on the size data of the glass raw material to be cut to obtain a raw material model. Divide a sample model on the raw material model according to the size data of the glass finished product. Plan a standard cutting path according to the contour line of the sample model.

[0014] Step 2: Obtain the real-time displacement of the cutting tool head through a displacement sensor arranged on the cutting tool head. Draw a real-time cutting path according to the real-time displacement. Calculate the path deviation value according to the real-time cutting path and the standard cutting path.

[0015] Step 3: Obtain the real-time image of glass cutting through the camera set outside the cutting tool head, extract the features of the real-time image to obtain the cutting marks, and perform image integration according to the timing characteristics of cutting to obtain the real-time sample mark diagram. Establish a mark comparison model based on the convolutional neural network, and import the real-time sample mark diagram into the mark comparison model to output the mark deviation value;

[0016] Step 4: Judge the error degree during the cutting process according to the preset path deviation judgment threshold and mark deviation judgment threshold to generate a deviation correction signal, and determine the deviation correction scheme according to the source of the deviation correction signal to perform error compensation on the cutting process.

[0017] Further, the specific process of obtaining the path deviation value is as follows:

[0018] S101: Establish a two-dimensional coordinate system according to the planar structure of the raw material model, divide the standard cutting path into several continuous cutting points Mk, mark the cutting points one by one on the two-dimensional coordinate system to obtain the position coordinates Mk(xm, ym) of several cutting points;

[0019] S102: Set the cutting tool head as a moving point on the two-dimensional coordinate system, obtain the real-time displacement of the cutting tool head according to the displacement sensor, and add the real-time displacement to the total displacement of the cutting tool head to calculate and obtain the position coordinates Mb(xb, yb) of the calibration point on the two-dimensional coordinate system;

[0020] S103: Calculate the path deviation value ΔS according to the following formula:

[0021] Further, the specific process of outputting the mark deviation value is as follows:

[0022] S201: Obtain the real-time image of glass cutting, perform impurity removal processing on the real-time image and then perform feature extraction, obtain multiple groups of real-time sample mark diagrams as training samples, and construct a mark comparison model between the sample mark diagram and the contour line of the sample model based on the convolutional neural network;

[0023] S2011: Determine the number of input layer nodes. The input layer is the contour line and contour feature vector of the sample model;

[0024] S2012: Determine the number of hidden layer nodes; determine the number of output layer nodes. The output layer is the mark deviation value;

[0025] S2013: Initialize the weights and biases; set the learning rate and the number of iterations;

[0026] S2014: Perform iterative training, and update the weights and biases through the backpropagation algorithm;

[0027] S2015: Verify the accuracy, precision and recall rate of the model;

[0028] S202. After inputting the real-time sample trace diagram into the deviation prediction model, a trace deviation value ΔG is obtained.

[0029] Further, the specific process of determining the error compensation for the cutting process by the rectification scheme is as follows:

[0030] S301. Obtain a preset path deviation judgment threshold and a trace deviation judgment threshold:

[0031] If the path deviation value ΔS is greater than or equal to the path deviation judgment threshold, a rectification signal is generated;

[0032] If the trace deviation value ΔG is greater than or equal to the trace deviation judgment threshold, a rectification signal is generated;

[0033] S302. If the source of the rectification signal is the path deviation, the rectification scheme is specifically to rectify the driving structure of the glass cutting device: obtain the real-time coordinates of the cutting tool head, that is, the calibration point position coordinates Mb(xb, yb), generate a corresponding calibration instruction according to the path deviation value ΔS, and control the driving structure of the glass cutting device to drive the cutting tool head to adjust through the calibration instruction, so that the real-time coordinates of the cutting tool head are consistent with the position coordinates Mk(xm, ym) of the cutting site;

[0034] If the source of the rectification signal is the image deviation, the rectification scheme is specifically to rectify the angle of the cutting tool head of the glass cutting device.

[0035] In summary, due to the adoption of the above technical solutions, the beneficial effects of the present invention are:

[0036] 1. For the glass cutting device based on virtual simulation technology, by controlling the first electromagnet and the second electromagnet to cut off the power supply, under the action of the return spring, the connecting shaft is ejected from the movable long groove, so that the limit clamping shaft is released from the limit. The driving motor drives the driving gear to rotate, and under the meshing action, the limit clamping shaft rotates synchronously. After adjusting the cutting tool head to an appropriate angle, control the first electromagnet and the second electromagnet to connect to the power supply. Under the action of the magnetic attraction force, the first electromagnet presses the magnet ring. At the same time, due to the repulsive force between the first electromagnet and the magnet ring, the pressing effect is enhanced, ensuring the positioning stability of the limit clamping shaft.

[0037] 2. For the glass cutting device based on virtual simulation technology, the air pump outputs gas into the conveying pipeline, and the gas sprays out from the air jet nozzle. Through the interactively set air jet nozzles, two counter-directional airflows distributed along the cutting path are formed to sweep the debris generated during the cutting process, which not only avoids interference with subsequent cutting, but also ensures the clarity of the real-time image of the glass cutting obtained by the camera.

[0038] 3. The error compensation method of the glass cutting device based on virtual simulation technology. A raw material model is obtained by performing three-dimensional modeling according to the size data of the glass raw material to be cut. Then, a standard cutting path is planned based on the contour line of the sample model. A real-time cutting path is drawn according to the real-time displacement. The path deviation value is obtained by performing path calculation based on the real-time cutting path and the standard cutting path. At the same time, feature extraction is performed on the real-time image to obtain cutting traces, and image integration is performed according to the timing characteristics of the cutting to obtain a real-time sample trace map. A trace comparison model is established based on a convolutional neural network, and the real-time sample trace map is imported into the trace comparison model to output the trace deviation value. The error degree during the cutting process is judged according to the preset path deviation judgment threshold and trace deviation judgment threshold to generate a deviation correction signal, and a deviation correction scheme is determined according to the source of the deviation correction signal to perform error compensation on the cutting process. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 Shows a schematic diagram of the overall external structure of the present invention;

[0040] Figure 2 Shows a schematic diagram of the external structure of the tool head fixing seat of the present invention;

[0041] Figure 3 Shows a schematic diagram of the internal structure of the adjustable mechanism of the present invention;

[0042] Figure 4 Shows a flowchart of the error correction method of the present invention;

[0043] Legend: 1. Device body; 2. Longitudinal axis drive mechanism; 3. Transverse axis drive mechanism; 4. Tool head support; 5. Tool head drive mechanism; 6. Tool head fixing seat; 7. Cutting tool head; 8. Horizontal fixing plate; 9. Camera; 10. Limit card slot; 11. Limit card shaft; 12. Passive gear; 13. Drive motor; 14. Drive gear; 15. Movable long slot; 16. Connecting shaft; 17. Return spring; 18. First electromagnet; 19. Magnet pressing ring; 20. Second electromagnet; 21. Air pump; 22. Delivery pipeline; 23. Jet nozzle. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0044] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0045] Embodiment 1:

[0046] As Figures 1-3As shown in the figure, a glass cutting device based on virtual simulation technology includes a device body 1 and a cutting tool head 7. On the top surface of the device body 1, a longitudinal axis driving mechanism 2 is provided. On the top surface of the longitudinal axis driving mechanism 2, a transverse axis driving mechanism 3 is fixedly installed. On the top surface of the transverse axis driving mechanism 3, a tool head support 4 is fixedly installed. On the outer surface of the tool head support 4, a tool head driving mechanism 5 is fixedly installed. On the outer surface of the output end of the tool head driving mechanism 5, a tool head fixing seat 6 is fixedly installed. The cutting tool head 7 is connected to the inside of the tool head fixing seat 6 through an adjustable mechanism. On the outer surface of the tool head support 4, a purging mechanism is fixedly installed. On the outer surface of the tool head fixing seat 6, a horizontal fixing plate 8 is fixedly installed. On the bottom surface of the horizontal fixing plate 8, a camera 9 is fixedly installed.

[0047] The adjustable mechanism includes a limit clamping shaft 11 and a limit component. A limit clamping groove 10 is opened on the bottom surface of the tool head fixing seat 6. The cutting tool head 7 is movably connected to the inner wall of the limit clamping groove 10. A through groove is opened inside the cutting tool head 7. Both ends of the limit clamping shaft 11 are movably connected to the inside of the tool head fixing seat 6 and respectively extend to the outer surface of the tool head fixing seat 6. The limit component is fixedly installed on one end surface of the limit clamping shaft 11. A passive gear 12 is connected to the other end surface of the limit clamping shaft 11. A driving motor 13 is fixedly installed on the outer surface of the tool head fixing seat 6. A driving gear 14 is connected to the outer surface of the output end of the driving motor 13. The driving gear 14 meshes with the passive gear 12, and the radius of the driving gear 14 is smaller than that of the passive gear 12.

[0048] The limit component includes a first electromagnet 18 and a magnet pressing ring 19. An activity long groove 15 is opened on one end surface of the limit clamping shaft 11. A connecting shaft 16 is fixedly installed on the inner wall of the activity long groove 15. A return spring 17 is jointly connected between one end surface of the connecting shaft 16 and the activity long groove 15. The first electromagnet 18 is fixedly installed on the other end surface of the connecting shaft 16. The magnet pressing ring 19 is fixedly installed on the outer surface of the limit clamping shaft 11. A second electromagnet 20 is fixedly installed inside the tool head fixing seat 6.

[0049] The corresponding surfaces of the first electromagnet 18 and the second electromagnet 20 overlap with each other and the magnetic properties of the corresponding surfaces are opposite. The corresponding surfaces of the first electromagnet 18 and the magnet ring overlap with each other and the magnetic properties of the corresponding surfaces are the same. The second electromagnet 20 does not overlap with the magnet ring.

[0050] The purging mechanism includes an air pump 21 and a conveying pipeline 22. Two conveying pipelines 22 are respectively connected to the outer surface of the output end of the air pump 21. Jet nozzles 23 are fixedly installed on the end surfaces of the two conveying pipelines 22. The two jet nozzles 23 are arranged obliquely alternately.

[0051] The working principle is as follows:

[0052] The longitudinal axis drive mechanism 2, the transverse axis drive mechanism 3, and the cutter head drive mechanism 5 are all prior arts, and their structures are publicly known and mature technologies, so no detailed description will be given. Through the mutual cooperation of the longitudinal axis drive mechanism 2 and the transverse axis drive mechanism 3, the cutting cutter head 7 is driven to move on the glass raw material along the cutting path, and the cutting process is completed by driving the cutting cutter head 7 through the cutter head drive mechanism 5;

[0053] During the cutting process, if the cutting cutter head 7 needs to be angle-adjusted, the control system is used to control the first electromagnet 18 and the second electromagnet 20 to cut off the power. Under the action of the reset spring 17, the connecting shaft 16 is ejected from the movable long groove 15, so that the limit clamping shaft 11 is released from the limit. The drive motor 13 is driven to drive the drive gear 14 to rotate. Under the meshing action, the limit clamping shaft 11 rotates synchronously. After the cutting cutter head 7 is adjusted to an appropriate angle, the first electromagnet 18 and the second electromagnet 20 are controlled to be powered on. Since the corresponding surfaces of the first electromagnet 18 and the second electromagnet 20 overlap with each other and the magnetic poles carried by the corresponding surfaces are opposite, under the action of the magnetic attraction force, the first electromagnet 18 moves along the outside of the limit clamping shaft 11, and then the first electromagnet 18 presses the magnet ring. At the same time, due to the repulsive force between the first electromagnet 18 and the magnet ring, the pressing effect is enhanced, ensuring the positioning stability of the limit clamping shaft 11;

[0054] During the cutting process, the air pump 21 outputs gas into the conveying pipeline 22, and the gas is ejected from the air jet nozzle 23. Two counter-directional airflows distributed along the cutting path are formed through the interactively arranged air jet nozzles 23 to clean the debris generated during the cutting process, which not only avoids interfering with the subsequent cutting, but also ensures the clarity of the real-time image of the glass cutting obtained by the camera 9.

[0055] Embodiment 2:

[0056] Such as Figure 4 shown, an error compensation method for a glass cutting device based on virtual simulation technology includes the following steps:

[0057] Step 1: Obtain the size data of the glass raw material to be cut, and obtain the size data of the glass finished product. According to the size data of the glass raw material to be cut, a raw material model is established by three-dimensional modeling, and a sample model is divided on the raw material model according to the size data of the glass finished product. A standard cutting path is planned according to the contour line of the sample model;

[0058] Step 2: Obtain the real-time displacement of the cutting cutter head 7 through the displacement sensor arranged on the cutting cutter head 7, and draw the real-time cutting path according to the real-time displacement. The path deviation value is obtained through path calculation based on the real-time cutting path and the standard cutting path;

[0059] The specific process of obtaining the path deviation value is as follows:

[0060] S101. Establish a two-dimensional coordinate system based on the planar structure of the raw material model, divide the standard cutting path into several continuous cutting points Mk, mark the cutting points one by one on the two-dimensional coordinate system, and obtain the position coordinates Mk(xm, ym) of several cutting points;

[0061] S102. Set the cutting tool head 7 as a moving point on the two-dimensional coordinate system, obtain the real-time displacement of the cutting tool head 7 according to the displacement sensor, and add the real-time displacement to the total displacement of the cutting tool head 7 to convert and obtain the position coordinates Mb(xb, yb) of the calibration point on the two-dimensional coordinate system;

[0062] S103. Calculate the path deviation value ΔS according to the following formula:

[0063] Step 3. Obtain the real-time image of glass cutting through the camera 9 set outside the cutting tool head 7, perform feature extraction on the real-time image to obtain cutting traces, and perform image integration according to the chronological features of cutting to obtain a real-time sample trace map. Establish a trace comparison model based on the convolutional neural network, and import the real-time sample trace map into the trace comparison model to output the trace deviation value;

[0064] The specific process of outputting the trace deviation value is as follows:

[0065] S201. Obtain the real-time image of glass cutting, perform impurity removal processing on the real-time image and then perform feature extraction, obtain multiple groups of real-time sample trace maps as training samples, and construct a trace comparison model between the sample trace map and the contour line of the sample model based on the convolutional neural network;

[0066] S2011. Determine the number of input layer nodes. The input layer is the contour line and contour feature vector of the sample model;

[0067] S2012. Determine the number of hidden layer nodes; determine the number of output layer nodes. The output layer is the trace deviation value;

[0068] S2013. Initialize the weights and biases; set the learning rate and the number of iterations;

[0069] S2014. Perform iterative training, and update the weights and biases through the backpropagation algorithm;

[0070] S2015. Verify the accuracy, precision, and recall rate of the model;

[0071] S202. After inputting the real-time sample trace map into the deviation prediction model, obtain the trace deviation value ΔG.

[0072] Step 4: Determine the error degree during the cutting process according to the preset path deviation judgment threshold and trace deviation judgment threshold to generate a deviation correction signal, and determine a deviation correction scheme according to the source of the deviation correction signal to perform error compensation on the cutting process.

[0073] The specific process of determining a deviation correction scheme to perform error compensation on the cutting process is as follows:

[0074] S301: Obtain the preset path deviation judgment threshold and trace deviation judgment threshold:

[0075] If the path deviation value ΔS is greater than or equal to the path deviation judgment threshold, generate a deviation correction signal;

[0076] If the trace deviation value ΔG is greater than or equal to the trace deviation judgment threshold, generate a deviation correction signal;

[0077] S302: If the source of the deviation correction signal is path deviation, the deviation correction scheme is specifically to correct the driving structure of the glass cutting device: obtain the real-time coordinates of the cutting tool head 7, that is, the calibration point position coordinates Mb(xb, yb), generate a corresponding calibration instruction according to the path deviation value ΔS, and control the driving structure of the glass cutting device to drive the cutting tool head 7 to adjust through the calibration instruction, so that the real-time coordinates of the cutting tool head 7 are consistent with the position coordinates Mk(xm, ym) of the cutting point;

[0078] If the source of the deviation correction signal is image deviation, the deviation correction scheme is specifically to correct the angle of the cutting tool head 7 of the glass cutting device.

[0079] The present invention performs three-dimensional modeling on the basis of the size data of the glass raw material to be cut to obtain a raw material model, and then plans a standard cutting path according to the contour line of the sample model, draws a real-time cutting path according to the real-time displacement, calculates the path deviation value according to the real-time cutting path and the standard cutting path, and at the same time extracts the features of the real-time image to obtain the cutting trace, and performs image integration according to the timing characteristics of the cutting to obtain a real-time sample trace diagram, establishes a trace comparison model according to the convolutional neural network, imports the real-time sample trace diagram into the trace comparison model to output the trace deviation value, determines the error degree during the cutting process according to the preset path deviation judgment threshold and trace deviation judgment threshold to generate a deviation correction signal, and determines a deviation correction scheme according to the source of the deviation correction signal to perform error compensation on the cutting process.

[0080] The setting of the interval and the size of the threshold is for the convenience of comparison. Regarding the size of the threshold, it depends on the amount of sample data and the base quantity set by those skilled in the art for each group of sample data; as long as it does not affect the proportional relationship between the parameters and the quantified values.

[0081] The above formulas are all dimensionless and take their numerical values for calculation. The formulas are obtained by collecting a large amount of data for software simulation to get a formula closest to the actual situation. The preset parameters in the formulas are set by those skilled in the art according to the actual situation;

[0082] In the two embodiments provided in the present application, it should be understood that the disclosed devices and systems can be implemented in other ways; for example, the device embodiments described above are only illustrative. For example, the division of the modules is only for logical function division. In actual implementation, there may be other division methods. For example, multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed; another point is that the displayed or discussed coupling or direct coupling or communication connection to each other can be through some interfaces. The indirect coupling or communication connection of the devices or modules can be in electrical, mechanical or other forms;

[0083] The above is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention, according to the technical solution and inventive concept of the present invention, makes equivalent substitutions or changes, and should be covered by the protection scope of the present invention.

Claims

1. A glass cutting device based on virtual simulation technology, characterized in that It includes a device body (1) and a cutting tool head (7). On the top surface of the device body (1), a longitudinal axis driving mechanism (2) is provided. On the top surface of the longitudinal axis driving mechanism (2), a transverse axis driving mechanism (3) is fixedly installed. On the top surface of the transverse axis driving mechanism (3), a tool head support (4) is fixedly installed. On the outer surface of the tool head support (4), a tool head driving mechanism (5) is fixedly installed. On the outer surface of the output end of the tool head driving mechanism (5), a tool head fixing seat (6) is fixedly installed. The cutting tool head (7) is connected to the inside of the tool head fixing seat (6) through an adjustable mechanism. On the outer surface of the tool head support (4), a purging mechanism is fixedly installed. On the outer surface of the tool head fixing seat (6), a horizontal fixing plate (8) is fixedly installed. On the bottom surface of the horizontal fixing plate (8), a camera (9) is fixedly installed.

2. The glass cutting device based on virtual simulation technology according to claim 1, characterized in that The adjustable mechanism includes a limit clamping shaft (11) and a limit component. On the bottom surface of the tool head fixing seat (6), a limit clamping groove (10) is opened. The cutting tool head (7) is movably connected to the inner wall of the limit clamping groove (10). A through groove is opened inside the cutting tool head (7). The two ends of the limit clamping shaft (11) are movably connected to the inside of the tool head fixing seat (6) and respectively extend to the outer surface of the tool head fixing seat (6). The limit component is fixedly installed on one end surface of the limit clamping shaft (11). On the other end surface of the limit clamping shaft (11), a passive gear (12) is connected. On the outer surface of the tool head fixing seat (6), a driving motor (13) is fixedly installed. On the outer surface of the output end of the driving motor (13), a driving gear (14) is connected. The driving gear (14) meshes with the passive gear (12), and the radius of the driving gear (14) is smaller than that of the passive gear (12).

3. The glass cutting device based on virtual simulation technology according to claim 2, characterized in that, The limit component includes a first electromagnet (18) and a magnet pressing ring (19). On one end surface of the limit clamping shaft (11), a movable long groove (15) is opened. On the inner wall of the movable long groove (15), a connecting shaft (16) is fixedly installed. Between one end surface of the connecting shaft (16) and the movable long groove (15), a return spring (17) is jointly connected. The first electromagnet (18) is fixedly installed on the other end surface of the connecting shaft (16). The magnet pressing ring (19) is fixedly installed on the outer surface of the limit clamping shaft (11). Inside the tool head fixing seat (6), a second electromagnet (20) is fixedly installed.

4. A glass cutting device based on virtual simulation technology according to claim 3, characterized in that, The corresponding surfaces of the first electromagnet (18) and the second electromagnet (20) coincide with each other and the magnetism of the corresponding surfaces is opposite. The corresponding surfaces of the first electromagnet (18) and the magnet ring coincide with each other and the magnetism of the corresponding surfaces is the same. The second electromagnet (20) does not coincide with the magnet ring.

5. A glass cutting device based on virtual simulation technology according to claim 1, characterized in that, The purging mechanism includes an air pump (21) and a conveying pipeline (22). The air pump (21) is fixedly installed on the outer surface of the tool head support (4). The two conveying pipelines (22) are respectively connected to the outer surface of the output end of the air pump (21). On the end surfaces of the two conveying pipelines (22), air jet nozzles (23) are fixedly installed. The two air jet nozzles (23) are arranged obliquely alternately.

6. An error compensation method for a glass cutting device based on virtual simulation technology, characterized in that, It includes the following steps: Step 1: Obtain the dimensional data of the glass raw material to be cut, and obtain the dimensional data of the glass finished product. Perform 3D modeling based on the dimensional data of the glass raw material to be cut to obtain a raw material model, divide a sample model on the raw material model according to the dimensional data of the glass finished product, and plan a standard cutting path according to the contour line of the sample model; Step 2: Obtain the real-time displacement of the cutting tool head (7) through the displacement sensor set on the cutting tool head (7), draw a real-time cutting path according to the real-time displacement, and perform path calculation based on the real-time cutting path and the standard cutting path to obtain a path deviation value; Step 3: Obtain a real-time image of glass cutting through the camera (9) set outside the cutting tool head (7), perform feature extraction on the real-time image to obtain cutting marks, perform image integration according to the timing characteristics of cutting to obtain a real-time sample mark diagram, establish a mark comparison model based on a convolutional neural network, and import the real-time sample mark diagram into the mark comparison model to output a mark deviation value; Step 4: Judge the error degree during the cutting process according to the preset path deviation judgment threshold and mark deviation judgment threshold to generate a deviation correction signal, and determine a deviation correction plan according to the source of the deviation correction signal to perform error compensation on the cutting process.

7. A method for error compensation of a glass cutting device based on virtual simulation technology according to claim 6, characterized in that, The specific process of obtaining the path deviation value is as follows: S101: Establish a two-dimensional coordinate system according to the planar structure of the raw material model, divide the standard cutting path into several continuous cutting points Mk, mark the cutting points one by one on the two-dimensional coordinate system to obtain the position coordinates Mk(xm, ym) of several cutting points; S102: Set the cutting tool head (7) as a moving point on the two-dimensional coordinate system, obtain the real-time displacement of the cutting tool head (7) according to the displacement sensor, and add the real-time displacement to the total displacement of the cutting tool head (7) to convert and obtain the position coordinates Mb(xb, yb) of the calibration point on the two-dimensional coordinate system; S103. Calculate the path deviation value ΔS according to the following formula:

8. A method for error compensation of a glass cutting device based on virtual simulation technology according to claim 6, characterized in that, The specific process of outputting the mark deviation value is as follows: S201: Obtain a real-time image of glass cutting, perform impurity removal processing on the real-time image and then perform feature extraction, obtain multiple groups of real-time sample mark diagrams as training samples, and construct a mark comparison model between the sample mark diagram and the contour line of the sample model based on a convolutional neural network; S2011: Determine the number of input layer nodes, and the input layer is the contour line and contour feature vector of the sample model; S2012: Determine the number of hidden layer nodes; determine the number of output layer nodes, and the output layer is the mark deviation value; S2013: Initialize the weights and biases; set the learning rate and the number of iterations; S2014: Perform iterative training, and update the weights and biases through the backpropagation algorithm; S2015: Verify the accuracy, precision and recall rate of the model; S202: After inputting the real-time sample mark diagram into the deviation prediction model, obtain the mark deviation value ΔG.

9. The error compensation method of a glass cutting device based on virtual simulation technology according to claim 6, characterized in that, The specific process of determining a deviation correction plan to perform error compensation on the cutting process is as follows: S301: Obtain the preset path deviation judgment threshold and mark deviation judgment threshold: If the path deviation value ΔS is greater than or equal to the path deviation judgment threshold, generate a deviation correction signal; If the mark deviation value ΔG is greater than or equal to the mark deviation judgment threshold, generate a deviation correction signal; S302. If the source of the deviation correction signal is the path deviation, the deviation correction solution is specifically to correct the driving structure of the glass cutting device; If the source of the deviation correction signal is the image deviation, the deviation correction solution is specifically to correct the angle of the cutting tool head (7) of the glass cutting device.

Images