Glass laser engraving equipment

Abstract

The invention relates to the technical field of glass product engraving, in particular to glass cup laser engraving equipment. The laser engraving equipment comprises a workbench, a positioning base is arranged on the workbench, and a positioning groove is formed in the positioning base; a support is installed on the workbench, a driving motor is installed on one side of the support, and a first bearing seat is installed on the other side of the support. The output end of the driving motor is connected with a driving shaft in a matched mode, the other end of the driving shaft is connected with the first bearing seat in a matched mode, and a first driving belt wheel and a second driving belt wheel are installed on the driving shaft. By adjusting the motion parameters of the five-axis linkage mechanical arm and arranging the positioning and clamping system, effective carving operation can be conducted on glass cups of various specifications, and the universality and production efficiency of the equipment are improved.

Description

Technical Field

[0001] The invention relates to the technical field of glass product engraving, in particular to a glass cup laser engraving device. Background Art

[0002] With the continuous development of laser technology, laser engraving technology has gradually been applied to the decoration of glass cups. Laser engraving uses a high-energy-density laser beam to irradiate the glass surface, causing the material to partially melt, vaporize or chemically change, thereby forming a preset pattern or text. Compared with traditional processes, laser engraving has significant advantages such as non-contact processing, high precision, high speed, flexible and changeable patterns, and environmental protection. It can achieve fine engraving of complex patterns without causing mechanical stress damage to the glass cup, thereby improving the product yield.

[0003] However, the existing laser engraving equipment for glass cups still has some shortcomings in practical applications. First, for glass cups of different shapes and sizes, existing equipment often needs to replace the fixture or perform complex parameter adjustments, which is cumbersome to operate and affects production efficiency. Secondly, the clamping method used by some equipment may not be stable enough, which may easily cause the glass cup to shift or vibrate during the engraving process, affecting the engraving accuracy. In addition, for glass cups with complex curved surfaces, existing engraving equipment is difficult to achieve all-round and high-precision engraving, which limits its scope of application. Finally, equipment that lacks real-time monitoring and feedback control systems cannot dynamically adjust parameters according to the actual situation during the engraving process, and it is difficult to ensure the stability and consistency of the engraving quality. Summary of the invention

[0004] The invention overcomes the shortcomings of the prior art and provides a glass cup laser engraving device.

[0005] To achieve the above-mentioned purpose, the technical solution adopted by the present invention is:

[0006] The present invention discloses a glass cup laser engraving device, the laser engraving device comprises a workbench, a positioning base is arranged on the workbench, and a positioning groove is arranged on the positioning base; a bracket is installed on the workbench, a driving motor is installed on one side of the bracket, and a first bearing seat is installed on the other side; a driving shaft is matched and connected with the output end of the driving motor, and the other end of the driving shaft is matched and connected with the first bearing seat, and a first driving pulley and a second driving pulley are installed on the driving shaft;

[0007] The workbench is also symmetrically provided with a first bidirectional threaded slide rod and a second bidirectional threaded slide rod, and a first slider and a second slider are slidably connected between the first bidirectional threaded slide rod and the second bidirectional threaded slide rod; a first clamping block is fixedly installed on the first slider, and a second clamping block is fixedly installed on the second slider;

[0008] A first driven pulley is disposed at one end of the first bidirectional threaded slide rod, and a second driven pulley is disposed at one end of the second bidirectional threaded slide rod; the first driving pulley and the first driven pulley are connected via a first belt transmission, and the second driving pulley and the second driven pulley are connected via a second belt transmission;

[0009] The workbench is also provided with a five-axis linkage mechanical arm, on which a laser engraving head is installed.

[0010] Furthermore, the cross-sectional shape of the positioning groove is the same as the cross-sectional shape of the bottom of the glass to be engraved.

[0011] Furthermore, the first bidirectional threaded slide rod is rotatably mounted on the workbench through a group of second bearing seats; the second bidirectional threaded slide rod is rotatably mounted on the workbench through a group of third bearing seats.

[0012] Furthermore, the first clamping block and the second clamping block are both provided with pressure sensors, and the pressure sensors are communicatively connected with the driving motor.

[0013] The present invention also discloses a control method for a glass laser engraving device, which is applied to any one of the glass laser engraving devices described above. The control method comprises the following steps:

[0014] Acquire etching feature data of the laser engraving device under various preset engraving parameters, and construct a dictionary tree search map according to the etching feature data of the laser engraving device under various preset engraving parameters;

[0015] Acquire real-time engraving parameters of a laser engraving device at a preset time node, and acquire predicted etching feature data of the glass at the current preset time node based on the real-time engraving parameters and in combination with a dictionary tree search graph;

[0016] The laser engraving equipment is adjusted and optimized based on the predicted etching feature data of the glass at the current preset time node.

[0017] Furthermore, the etching feature data of the laser engraving device under various preset engraving parameters are obtained, and a dictionary tree search map is constructed according to the etching feature data of the laser engraving device under various preset engraving parameters, specifically:

[0018] Testing the laser engraving device to obtain etching feature data of the laser engraving device under various preset engraving parameters;

[0019] Constructing a dictionary tree, converting various preset engraving parameters into branch nodes of the dictionary tree, and converting various etching feature data into leaf nodes of the dictionary tree;

[0020] According to the etching feature data of the laser engraving device under various preset engraving parameters obtained by the test, a directed description relationship between various preset engraving parameters and various etching feature data is established;

[0021] According to the directed description relationship between various preset engraving parameters and various etching feature data, each branch node and each leaf node are feature-combined to form a dictionary tree search map of various preset engraving parameters and various etching feature data;

[0022] Among them, the preset engraving parameters include the laser power, moving speed and moving direction of the laser engraving head; the etching feature data includes the etching depth, etching width and etching shape corresponding to the corresponding position node of the glass cup after the laser engraving head acts on the corresponding preset engraving parameters.

[0023] Furthermore, the real-time engraving parameters of the laser engraving device are obtained at a preset time node, and the predicted etching feature data of the glass at the current preset time node is obtained according to the real-time engraving parameters and combined with the dictionary tree search map, specifically:

[0024] Acquire real-time engraving parameters of a laser engraving device at a preset time node, and import the real-time engraving parameters into the dictionary tree search graph;

[0025] The real-time engraving parameters are converted into the root node of the dictionary tree, a cosine similarity algorithm is introduced, and the cosine similarity between the root node and each branch node in the dictionary tree search graph is calculated based on the cosine similarity algorithm;

[0026] Constructing a size sorting table, importing the cosine similarity between the root node and each branch node in the dictionary tree search graph into the size sorting table for size sorting, and sorting to obtain the highest cosine similarity;

[0027] The branch nodes corresponding to the highest cosine similarity are calibrated, the leaf nodes belonging to the calibrated branch nodes are extracted, and the feature information in the extracted leaf nodes is interpreted to obtain the etching feature data corresponding to the extracted leaf nodes;

[0028] The etching feature data corresponding to the extracted leaf node is output as the predicted etching feature data of the glass at the preset time node.

[0029] Furthermore, the laser engraving equipment is adjusted and optimized according to the predicted etching feature data of the glass at the current preset time node, specifically:

[0030] Obtaining a preset engraving process scheme for the glass, and obtaining preset engraving feature data range information of the glass at a current preset time node according to the preset engraving process scheme;

[0031] Determine the maximum preset engraving feature data and the minimum preset engraving feature data of the glass at the current preset time node according to the preset engraving feature data range information;

[0032] The maximum preset engraving feature data of the glass at the current preset time node is used as the upper boundary condition, and the minimum preset engraving feature data of the glass at the current preset time node is used as the lower boundary condition;

[0033] Setting a definition domain according to the upper boundary condition and the lower boundary condition, and determining whether the predicted etching feature data of the glass cup at the current preset time node completely falls within the definition domain;

[0034] If the predicted etching feature data of the glass cup at the current preset time node completely falls within the definition domain, the real-time engraving parameters of the laser engraving device are not adjusted, and the real-time engraving parameters of the laser engraving device are continuously monitored at the next preset time node;

[0035] If the predicted etching feature data of the glass cup at the current preset time node partially or completely falls outside the definition domain, the particle swarm optimization algorithm is introduced, and the real-time engraving parameters of the laser engraving device are iteratively optimized based on the particle swarm optimization algorithm, and the real-time engraving parameters of the laser engraving device at the next preset time node are readjusted so that the predicted etching feature data at the next time node falls within the definition domain again.

[0036] Furthermore, if the predicted etching feature data of the glass at the current preset time node partially or completely falls outside the definition domain, the particle swarm optimization algorithm is introduced, and the real-time engraving parameters of the laser engraving device are iteratively optimized based on the particle swarm optimization algorithm, and the real-time engraving parameters of the laser engraving device at the next preset time node are readjusted so that the predicted etching feature data at the next time node falls within the definition domain again, specifically:

[0037] Determine the particle population size in the particle swarm optimization algorithm, where each particle corresponds to a set of real-time engraving parameters of the laser engraving device, including the laser power, moving speed, and moving direction of the laser engraving head;

[0038] Initialize the position and velocity vectors of the particles, where the position vector represents a specific combination of engraving parameters, and the velocity vector represents the adjustment direction and amplitude of the parameters;

[0039] A fitness function is constructed based on the adaptability of the predicted etching feature data and the definition domain, and the fitness value of each particle is calculated. The fitness value reflects the effectiveness of the engraving parameter combination represented by the particle in making the predicted etching feature data fall into the definition domain.

[0040] For each particle, compare its current fitness value with the individual historical optimal fitness value. If it is better, update the individual optimal position; at the same time, compare in the entire particle group and update the global optimal position;

[0041] According to the particle's current position, speed, individual optimal position and global optimal position, continue to update the particle's speed and position vector to generate a new carving parameter combination;

[0042] The above steps are continuously iterated until the predicted etching feature data completely falls into the definition domain, and the engraving parameters corresponding to the global optimal position are set as the real-time engraving parameters of the next preset time node, thereby ensuring that the predicted etching feature data of the next time node falls into the definition domain again.

[0043] The present invention solves the technical defects existing in the background technology, and has the following beneficial effects: First, by adjusting the motion parameters of the five-axis linkage robot arm and the setting of the positioning and clamping system, it is possible to effectively engrave glass cups of various specifications, thereby improving the versatility and production efficiency of the equipment. Secondly, the control method can construct a dictionary tree search map by pre-acquiring etching feature data under different engraving parameters, and then predicting etching feature data in combination with real-time engraving parameters, thereby realizing timely adjustment and optimization of the laser engraving equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, drawings of other embodiments can be obtained based on these drawings without paying creative work.

[0045] Figure 1 is a first three-dimensional structural schematic diagram of the device;

[0046] Figure 2 for Figure 1 The enlarged schematic diagram of the AA in the middle;

[0047] Figure 3 is a second three-dimensional structural schematic diagram of the device;

[0048] Figure 4 It is a side view structural schematic diagram of the device;

[0049] The reference numerals in the accompanying drawings are explained as follows: 101, workbench; 102, positioning base; 103, positioning groove; 104, bracket; 105, driving motor; 106, first bearing seat; 107, driving shaft; 108, first driving pulley; 109, second driving pulley; 201, first bidirectional threaded slide rod; 202, second bidirectional threaded slide rod; 203, first slider; 204, second slider; 205, first clamping block; 206, second clamping block; 207, first driven pulley; 208, second driven pulley; 209, first belt; 301, second belt; 302, five-axis linkage robot arm; 303, laser engraving head; 304, second bearing seat; 305, third bearing seat. DETAILED DESCRIPTION

[0050] In order to facilitate the understanding of the present invention, the present invention will be described more fully below with reference to the relevant drawings. The preferred embodiments of the present invention are given in the drawings. However, the present invention can be implemented in many different forms and is not limited to the embodiments described herein. On the contrary, the purpose of providing these embodiments is to make the disclosure of the present invention more thoroughly understood.

[0051] The present invention discloses a glass cup laser engraving device, such as Figure 1 As shown, the laser engraving device includes a workbench 101, on which a positioning base 102 is provided, and on which a positioning groove 103 is provided; a bracket 104 is installed on the workbench 101, on one side of the bracket 104 a driving motor 105 is installed, and on the other side a first bearing seat 106 is installed; the output end of the driving motor 105 is matched and connected with a driving shaft 107, and the other end of the driving shaft 107 is matched and connected with the first bearing seat 106, and on the driving shaft 107 a first driving pulley 108 and a second driving pulley 109 are installed;

[0052] like Figure 2 , Figure 3 , Figure 4 As shown, the workbench 101 is also symmetrically provided with a first bidirectional threaded slide bar 201 and a second bidirectional threaded slide bar 202, and a first slider 203 and a second slider 204 are slidably connected between the first bidirectional threaded slide bar 201 and the second bidirectional threaded slide bar 202; a first clamping block 205 is fixedly installed on the first slider 203, and a second clamping block 206 is fixedly installed on the second slider 204;

[0053] A first driven pulley 207 is disposed at one end of the first bidirectional threaded slide bar 201, and a second driven pulley 208 is disposed at one end of the second bidirectional threaded slide bar 202; the first driving pulley 108 and the first driven pulley 207 are connected to each other through a first belt 209, and the second driving pulley 109 and the second driven pulley 208 are connected to each other through a second belt 301;

[0054] A five-axis linkage mechanical arm 302 is also provided on the workbench 101 , and a laser engraving head 303 is installed on the five-axis linkage mechanical arm 302 .

[0055] The cross-sectional shape of the positioning groove 103 is the same as the cross-sectional shape of the bottom of the glass to be engraved.

[0056] The first bidirectional threaded slide bar 201 is rotatably mounted on the workbench 101 via a set of second bearing seats 304 ; the second bidirectional threaded slide bar 202 is rotatably mounted on the workbench 101 via a set of third bearing seats 305 .

[0057] The first clamping block 205 and the second clamping block 206 are both provided with pressure sensors, and the pressure sensors are in communication connection with the driving motor 105 .

[0058] The positioning base 102 is detachable, and the user can install positioning bases of different shapes and specifications according to specific requirements, such as positioning bases with positioning grooves 103 of different diameters; such as positioning bases with positioning grooves 103 of various shapes. The shapes of the positioning grooves include but are not limited to square, circular, triangular, and elliptical.

[0059] It should be noted that the glass to be engraved is placed in the positioning groove 103 of the positioning base 102. Since the cross-sectional shape of the positioning groove 103 is the same as the cross-sectional shape of the bottom of the glass, it can ensure that the glass is initially positioned accurately on the horizontal plane and prevent it from unnecessary translation in subsequent operations. The drive motor 105 is started, and the drive motor 105 drives the drive shaft 107 to rotate. The first active pulley 108 and the second active pulley 109 on the drive shaft 107 rotate synchronously with the drive shaft 107. The first active pulley 108 drives the first driven pulley 207 at one end of the first bidirectional threaded slide 201 to rotate through the first belt 209, and the second active pulley 109 drives the second driven pulley 208 at one end of the second bidirectional threaded slide 202 to rotate through the second belt 301. Since the first bidirectional threaded slide 201 and the second bidirectional threaded slide 202 are bidirectional thread structures, as they rotate, the first slider 203 and the second slider 204 will move toward each other on the slide. The first clamping block 205 on the first slider 203 and the second clamping block 206 on the second slider 204 also move toward each other to clamp the glass in the positioning groove 103. During the clamping process, the pressure sensors on the first clamping block 205 and the second clamping block 206 monitor the clamping pressure in real time. When the pressure reaches the set value, the pressure sensor transmits a signal to the drive motor 105, and the drive motor 105 stops rotating, thereby ensuring that the clamping force can not only firmly fix the glass, but also does not damage the glass due to excessive pressure.

[0060] The laser engraving head 303 is moved to the initial engraving position above the glass by the five-axis linkage robot 302. The five-axis linkage can realize the precise movement of the laser engraving head 303 in space to meet the complex curved surface engraving requirements of the glass. The laser engraving head 303 emits a laser beam to engrave the glass according to the set engraving content such as patterns, text or textures. During the engraving process, the five-axis linkage robot 302 adjusts the position, angle and other parameters of the laser engraving head 303 in real time according to the requirements of the engraving path to ensure the accuracy and quality of the engraving.

[0061] After the engraving is completed, the driving motor 105 rotates in the opposite direction. The first active pulley 108 and the second active pulley 109 rotate in the opposite direction, driving the first bidirectional threaded slide bar 201 and the second bidirectional threaded slide bar 202 to rotate in the opposite direction through the belt, so that the first slider 203 and the second slider 204 move in opposite directions, and the first clamping block 205 and the second clamping block 206 release the clamping of the glass. Then, the engraved glass can be taken out of the positioning groove 103 by an automatic unloading robot or manually, thereby completing an engraving process.

[0062] In general, the design of the positioning groove 103 matching the shape of the bottom of the glass ensures the precise initial positioning of the glass on the horizontal plane, providing a stable foundation for subsequent clamping and engraving operations. The bidirectional threaded slide rod structure cooperates with the clamping system of the pressure sensor to achieve stable clamping of the glass. The bidirectional threaded slide rod enables the first clamping block 205 and the second clamping block 206 to move toward each other, ensuring that the glass is evenly clamped from both sides. The presence of the pressure sensor can accurately control the clamping force to avoid damage to the glass due to excessive clamping force, thereby improving the applicability of the equipment to glasses of different materials and thicknesses. The use of the five-axis linkage robot 302 enables the laser engraving head 303 to perform complex movements in space and accurately engrave the glass according to the design requirements. Whether it is the curved surface part or the flat surface part of the glass, high-precision engraving can be achieved to meet the engraving requirements of complex patterns and fine textures. By adjusting the motion parameters of the five-axis linkage robot 302 and the setting of the positioning and clamping system, effective engraving operations can be performed on glasses of various specifications, thereby improving the versatility and production efficiency of the equipment.

[0063] The present invention also discloses a control method for a glass laser engraving device, which is applied to any one of the glass laser engraving devices described above. The control method comprises the following steps:

[0064] Acquire etching feature data of the laser engraving device under various preset engraving parameters, and construct a dictionary tree search map according to the etching feature data of the laser engraving device under various preset engraving parameters;

[0065] Acquire real-time engraving parameters of a laser engraving device at a preset time node, and acquire predicted etching feature data of the glass at the current preset time node based on the real-time engraving parameters and in combination with a dictionary tree search graph;

[0066] The laser engraving equipment is adjusted and optimized based on the predicted etching feature data of the glass at the current preset time node.

[0067] It should be noted that the etching feature data of the laser engraving device under various preset engraving parameters are obtained. Then a dictionary tree search map is constructed based on these data. The dictionary tree search map is a data structure that can effectively organize and store the relationship between these etching feature data and engraving parameters. In this map, different engraving parameter combinations and their corresponding etching feature data are stored in a tree structure to facilitate subsequent query and analysis. The real-time engraving parameters of the laser engraving device are obtained at the preset time node. This real-time engraving parameter reflects the current working state of the device, such as the current laser power, engraving speed and other actual operating parameters. Then, based on these real-time engraving parameters and combined with the previously constructed dictionary tree search map, the predicted etching feature data of the glass at the current preset time node can be obtained. This utilizes the relationship between the previously stored engraving parameters and the etching feature data, and searches for the corresponding etching feature data in the dictionary tree search map through the known real-time engraving parameters, so as to predict the current engraving effect. Finally, the laser engraving device is adjusted and optimized according to the predicted etching feature data. If the predicted etching feature data shows that there may be unsatisfactory engraving results, such as etching depth that is too deep or too shallow, unclear etching patterns, etc., the engraving parameters of the equipment can be adjusted in time, such as adjusting the laser power, changing the engraving speed, etc., to avoid a wide range of poor engraving effects. This control method can construct a dictionary tree search map by pre-acquiring etching feature data under different engraving parameters, and then predict the etching feature data in combination with real-time engraving parameters, thereby achieving timely adjustment and optimization of the laser engraving equipment. It can effectively improve the quality of laser engraving of glass cups and avoid poor etching effects caused by improper engraving parameter settings, such as uneven etching depth, unclear patterns, etc. At the same time, this control method based on data structure and prediction can improve the degree of automation of the equipment, reduce manual intervention, improve production efficiency, and can adapt to different engraving requirements and characteristics of glass cups, enhancing the versatility and flexibility of the equipment.

[0068] Furthermore, the etching feature data of the laser engraving device under various preset engraving parameters are obtained, and a dictionary tree search map is constructed according to the etching feature data of the laser engraving device under various preset engraving parameters, specifically:

[0069] Testing the laser engraving device to obtain etching feature data of the laser engraving device under various preset engraving parameters;

[0070] Constructing a dictionary tree, converting various preset engraving parameters into branch nodes of the dictionary tree, and converting various etching feature data into leaf nodes of the dictionary tree;

[0071] According to the etching feature data of the laser engraving device under various preset engraving parameters obtained by the test, a directed description relationship between various preset engraving parameters and various etching feature data is established;

[0072] According to the directed description relationship between various preset engraving parameters and various etching feature data, each branch node and each leaf node are feature-combined to form a dictionary tree search map of various preset engraving parameters and various etching feature data;

[0073] Among them, the preset engraving parameters include the laser power, moving speed and moving direction of the laser engraving head; the etching feature data includes the etching depth, etching width and etching shape corresponding to the corresponding position node of the glass cup after the laser engraving head acts on the corresponding preset engraving parameters.

[0074] It should be noted that the laser engraving equipment must first be tested in order to obtain etching feature data under different preset engraving parameters. The preset engraving parameters here clearly include the laser power, moving speed and moving direction of the laser engraving head. These parameters are key factors affecting the engraving effect. Different powers, speeds and directions will cause differences in the effects of the laser on the surface of the glass cup. During the test, by changing these preset engraving parameters, the etching conditions on the surface of the glass cup under each parameter combination are recorded, that is, the etching feature data, including etching depth, etching width and etching shape. The etching depth reflects the degree of laser removal of glass materials, the etching width reflects the lateral scale of the engraving traces, and the etching shape is the pattern outline finally formed on the glass surface. When constructing the dictionary tree, various preset engraving parameters are converted into branch nodes of the dictionary tree, and various etching feature data are converted into leaf nodes of the dictionary tree. This structure is similar to a tree structure, and the branch nodes are path branches leading to the leaf nodes, just like in a decision tree, different decision conditions (here are preset engraving parameters) will lead to different results (here are etching feature data). According to the data obtained from the test, a directed description relationship between various preset engraving parameters and various etching feature data is constructed. This means that it is clear how different engraving parameters affect the final etching features, which is a description of cause and effect. For example, a specific laser power, moving speed and direction (preset engraving parameters) will lead to a specific etching depth, width and shape (etching feature data). Finally, according to this directed description relationship, each branch node and each leaf node are combined with features to form a dictionary tree search map of various preset engraving parameters and various etching feature data. This map fully presents the mapping relationship between the preset engraving parameters and the etching feature data, just like a lookup table, which can quickly locate the possible etching feature data through given engraving parameters.

[0075] The dictionary tree search map constructed through the above series of operations can clearly present the relationship between the preset engraving parameters of the laser engraving equipment and the etching feature data. This allows the possible etching effect to be quickly predicted based on the current engraving parameters during the actual engraving process, providing a basis for optimizing the engraving process. This relationship map helps to improve the accuracy and controllability of laser engraving, and can better adjust the engraving parameters according to needs to obtain the ideal etching depth, width and shape. It also provides effective data support for the automatic control and process optimization of the equipment.

[0076] Furthermore, the real-time engraving parameters of the laser engraving device are obtained at a preset time node, and the predicted etching feature data of the glass at the current preset time node is obtained according to the real-time engraving parameters and combined with the dictionary tree search map, specifically:

[0077] Acquire real-time engraving parameters of a laser engraving device at a preset time node, and import the real-time engraving parameters into the dictionary tree search graph;

[0078] The real-time engraving parameters are converted into the root node of the dictionary tree, a cosine similarity algorithm is introduced, and the cosine similarity between the root node and each branch node in the dictionary tree search graph is calculated based on the cosine similarity algorithm;

[0079] Constructing a size sorting table, importing the cosine similarity between the root node and each branch node in the dictionary tree search graph into the size sorting table for size sorting, and sorting to obtain the highest cosine similarity;

[0080] The branch nodes corresponding to the highest cosine similarity are calibrated, the leaf nodes belonging to the calibrated branch nodes are extracted, and the feature information in the extracted leaf nodes is interpreted to obtain the etching feature data corresponding to the extracted leaf nodes;

[0081] The etching feature data corresponding to the extracted leaf node is output as the predicted etching feature data of the glass at the preset time node.

[0082] It should be noted that the real-time engraving parameters of the laser engraving equipment are obtained at the preset time node. These real-time engraving parameters reflect the current working state of the equipment, such as the actual operating parameter values ​​of laser power, engraving speed, and moving direction. Then these real-time engraving parameters are imported into the previously constructed dictionary tree search map. This dictionary tree search map contains the relationship between various preset engraving parameters and etching feature data, which is an important basis for predicting etching feature data. After converting the real-time engraving parameters into the root node of the dictionary tree, the cosine similarity algorithm is introduced. The cosine similarity algorithm is used to measure the similarity between two vectors. Here, the root node (representing the real-time engraving parameter) and each branch node in the dictionary tree search map (representing different preset engraving parameters) are regarded as vectors, and the cosine similarity between them is calculated to determine the similarity between the real-time engraving parameters and the preset engraving parameters. Construct a size sorting table, and import the calculated cosine similarity values ​​between the root node and each branch node into it for size sorting. After sorting, the highest cosine similarity value is obtained. The branch node corresponding to this highest cosine similarity value represents the preset engraving parameter combination that is most similar to the current real-time engraving parameter. The branch node corresponding to the highest cosine similarity is calibrated, and then the leaf node to which the calibrated branch node belongs is extracted. These leaf nodes contain etching feature data related to the branch node (preset engraving parameters). The feature information in the extracted leaf node is interpreted, that is, the etching feature data information stored in the leaf node is converted into an understandable and usable form. Finally, the corresponding etching feature data in the interpreted leaf node is output as the predicted etching feature data of the glass at the preset time node. This predicted etching feature data can reflect the etching effects such as etching depth, etching width, and etching shape that may appear on the glass under the current real-time engraving parameters. According to the real-time engraving parameters of the laser engraving equipment, the dictionary tree search graph and the cosine similarity algorithm can be used to accurately predict the etching feature data of the glass at the current moment. This helps to understand the possible etching effects in advance during the engraving process, so as to adjust the engraving parameters in time to achieve the ideal engraving effect, improve the accuracy and controllability of laser engraving, reduce the engraving failure or poor quality caused by improper parameter settings, and provide strong technical support for the efficient and precise operation of laser engraving equipment.

[0083] Furthermore, the laser engraving equipment is adjusted and optimized according to the predicted etching feature data of the glass at the current preset time node, specifically:

[0084] Obtaining a preset engraving process scheme for the glass, and obtaining preset engraving feature data range information of the glass at a current preset time node according to the preset engraving process scheme;

[0085] Determine the maximum preset engraving feature data and the minimum preset engraving feature data of the glass at the current preset time node according to the preset engraving feature data range information;

[0086] The maximum preset engraving feature data of the glass at the current preset time node is used as the upper boundary condition, and the minimum preset engraving feature data of the glass at the current preset time node is used as the lower boundary condition;

[0087] Setting a definition domain according to the upper boundary condition and the lower boundary condition, and determining whether the predicted etching feature data of the glass cup at the current preset time node completely falls within the definition domain;

[0088] If the predicted etching feature data of the glass cup at the current preset time node completely falls within the definition domain, the real-time engraving parameters of the laser engraving device are not adjusted, and the real-time engraving parameters of the laser engraving device are continuously monitored at the next preset time node;

[0089] If the predicted etching feature data of the glass cup at the current preset time node partially or completely falls outside the definition domain, the particle swarm optimization algorithm is introduced, and the real-time engraving parameters of the laser engraving device are iteratively optimized based on the particle swarm optimization algorithm, and the real-time engraving parameters of the laser engraving device at the next preset time node are readjusted so that the predicted etching feature data at the next time node falls within the definition domain again.

[0090] It should be noted that the preset engraving process scheme of the glass is first obtained. This scheme stipulates the ideal engraving effect of the glass at each preset time node, that is, the preset engraving feature data range information. For example, there may be an ideal interval range for the etching depth, and the etching width and etching shape also have corresponding required ranges. According to this range information, the maximum preset engraving feature data and the minimum preset engraving feature data at the current preset time node can be determined. These two data represent the upper and lower limits allowed by the process scheme, respectively. The maximum preset engraving feature data is used as the upper boundary condition, and the minimum preset engraving feature data is used as the lower boundary condition, so as to set a definition domain. This definition domain is the value range of the predicted etching feature data that meets the preset engraving process scheme at the current preset time node. Then, it is determined whether the predicted etching feature data of the glass at the current preset time node completely falls within this definition domain. If the predicted etching feature data completely falls within the definition domain, it means that the real-time engraving parameter setting of the current laser engraving device is reasonable and can meet the preset engraving process scheme requirements. Therefore, there is no need to regulate the real-time engraving parameters of the laser engraving device. It is only necessary to continue to monitor the real-time engraving parameters at the next preset time node to ensure that the engraving process continues to meet the requirements. If the predicted etching feature data partially or completely falls outside the definition domain, it means that the current engraving parameter settings may cause the engraving result to not conform to the preset process plan. At this time, the particle swarm optimization algorithm is introduced, which is an optimization algorithm based on swarm intelligence. The real-time engraving parameters of the laser engraving equipment are iteratively optimized based on the particle swarm optimization algorithm. During the iteration process, the predicted etching feature data is recalculated by continuously adjusting the real-time engraving parameters of the laser engraving equipment (such as laser power, engraving speed, moving direction, etc.) until the predicted etching feature data at the next preset time node falls back into the definition domain, so that the engraving process returns to the track that conforms to the preset process plan.

[0091] Through the above operations, this adjustment optimization processing method can effectively monitor and adjust the real-time engraving parameters of the laser engraving equipment according to the preset engraving process plan. While ensuring that the engraving process meets the process requirements, it reduces unnecessary parameter adjustments and improves the stability and efficiency of equipment operation. When the predicted etching feature data does not meet the requirements, the particle swarm optimization algorithm can intelligently adjust the engraving parameters to make the engraving results as close as possible to the requirements of the preset process plan, thereby improving the quality and yield of glass laser engraving and ensuring the accuracy and controllability of the entire engraving process.

[0092] Furthermore, if the predicted etching feature data of the glass at the current preset time node partially or completely falls outside the definition domain, the particle swarm optimization algorithm is introduced, and the real-time engraving parameters of the laser engraving device are iteratively optimized based on the particle swarm optimization algorithm, and the real-time engraving parameters of the laser engraving device at the next preset time node are readjusted so that the predicted etching feature data at the next time node falls within the definition domain again, specifically:

[0093] Determine the particle population size in the particle swarm optimization algorithm, where each particle corresponds to a set of real-time engraving parameters of the laser engraving device, including the laser power, moving speed, and moving direction of the laser engraving head;

[0094] Initialize the position and velocity vectors of the particles, where the position vector represents a specific combination of engraving parameters, and the velocity vector represents the adjustment direction and amplitude of the parameters;

[0095] A fitness function is constructed based on the adaptability of the predicted etching feature data and the definition domain, and the fitness value of each particle is calculated. The fitness value reflects the effectiveness of the engraving parameter combination represented by the particle in making the predicted etching feature data fall into the definition domain.

[0096] For each particle, compare its current fitness value with the individual historical optimal fitness value. If it is better, update the individual optimal position; at the same time, compare in the entire particle group and update the global optimal position;

[0097] According to the particle's current position, speed, individual optimal position and global optimal position, continue to update the particle's speed and position vector to generate a new carving parameter combination;

[0098] The above steps are continuously iterated until the predicted etching feature data completely falls into the definition domain, and the engraving parameters corresponding to the global optimal position are set as the real-time engraving parameters of the next preset time node, thereby ensuring that the predicted etching feature data of the next time node falls into the definition domain again.

[0099] It should be noted that the particle population size in the particle swarm optimization algorithm is first determined. Each particle here corresponds to a set of real-time engraving parameters of the laser engraving device, which include the laser power, moving speed and moving direction of the laser engraving head, which are factors that have an important impact on the engraving effect. This means that different engraving parameter combination spaces are explored through multiple particles to find the optimal combination that can make the predicted etching feature data fall into the definition domain. The position and velocity vectors of the particles are initialized. The position vector represents a specific engraving parameter combination, which determines the initial position of each particle in the parameter space, that is, a set of initial laser power, moving speed and moving direction values. The velocity vector represents the adjustment direction and amplitude of the parameters, which determines how the particles move in the parameter space in each iteration, that is, how to change the value of the engraving parameters. The fitness function is constructed based on the adaptability of the predicted etching feature data to the definition domain. This fitness function is the key to evaluating the quality of particles. It reflects the effectiveness of the engraving parameter combination represented by the particle in making the predicted etching feature data fall into the definition domain. By calculating the fitness value of each particle, the degree to which the engraving parameter combination distance corresponding to each particle meets the requirements (making the predicted etching feature data fall into the definition domain) can be measured.

[0100] For each particle, its current fitness value is compared with the individual historical optimal fitness value. If the current fitness value is better, the individual optimal position is updated, which means that a better engraving parameter combination for the particle has been found. At the same time, the global optimal position is updated by comparing in the entire particle swarm. This global optimal position represents the most effective engraving parameter combination currently found in the entire particle swarm to make the predicted etching feature data fall into the definition domain. According to the current position, speed, individual optimal position and global optimal position of the particle, the particle's speed and position vector are continued to be updated. This step is the core operation of the particle swarm optimization algorithm. Through the comprehensive use of this information, particles can move in the direction where it is more likely to find the optimal solution (making the predicted etching feature data fall into the definition domain), thereby generating a new engraving parameter combination.

[0101] Continue to iterate the above steps, constantly adjust the position and speed of the particles, and explore different combinations of engraving parameters until the predicted etching feature data completely falls into the definition domain. Once this condition is met, the engraving parameters corresponding to the global optimal position are set as the real-time engraving parameters of the next preset time node, thereby ensuring that the predicted etching feature data at the next time node falls into the definition domain again.

[0102] Through this iterative optimization process based on the particle swarm optimization algorithm, the real-time engraving parameters of the laser engraving equipment can be intelligently adjusted. When the predicted etching feature data partially or completely falls outside the definition domain, multiple particles are explored and optimized in the engraving parameter combination space to find the parameter combination that is most conducive to making the predicted etching feature data fall back into the definition domain. The adaptability of the laser engraving equipment to the engraving process requirements is improved, and the situation where the engraving effect does not meet the expectations due to improper parameter settings is effectively avoided, thereby improving the quality and accuracy of glass cup laser engraving and ensuring that the engraving process can be carried out according to the preset process plan.

[0103] In addition, in practical applications, the control method may further include the following steps:

[0104] Obtaining a working calendar of the laser engraving device, obtaining a fault condition event of the laser engraving device according to the working calendar, obtaining fault characteristic data of the laser engraving device when the fault condition event occurs, and obtaining normal characteristic data of the laser engraving device during a preset time period before the fault condition event occurs;

[0105] A fuzzy evaluation algorithm is introduced, and based on the fuzzy evaluation algorithm, a fuzzy evaluation is performed on the fault feature data of the laser engraving device when the fault condition event occurs and the normal feature data obtained during a preset time period before the fault condition event occurs, so as to obtain the fuzziness of the fault transfer feature when various fault condition events occur in the laser engraving device;

[0106] Constructing a fault transfer feature fuzziness matrix based on the fault transfer feature fuzziness of various fault conditions of the laser engraving device; constructing a Bayesian network model, and embedding the fault transfer feature fuzziness matrix into the Bayesian network model for encoding learning;

[0107] Acquire the real-time working parameters of the laser engraving device at a preset time node, import the real-time working parameters into the Bayesian network model for evaluation, and acquire the fault transfer feature fuzziness of the laser engraving device at the current preset time node;

[0108] If the fuzziness of the fault transfer feature of the laser engraving device at the current preset time node is greater than the preset threshold, the laser engraving device is controlled to shut down and a fault warning message is generated;

[0109] If the fault transfer feature fuzziness of the laser engraving device at the current preset time node is not greater than the preset threshold, further obtaining the fault transfer feature fuzziness of the laser engraving device at several future time nodes based on the Bayesian network model;

[0110] If the laser engraving device has a fault transfer feature fuzziness greater than a preset threshold at at least one future time node, the laser engraving device is controlled to shut down and a fault warning message is generated.

[0111] It should be noted that the working calendar of the laser engraving equipment is first obtained to find out the fault condition events. For each fault condition event, the fault feature data when the fault occurs and the normal feature data of the preset time period before the fault occur are collected. These data are the basis for the subsequent analysis of equipment failures. The fault feature data include abnormal operating parameters of the equipment (such as laser power fluctuations, abnormal movement of the robotic arm, etc.), and the normal feature data reflects the state of the equipment when it is operating normally. The fuzzy evaluation algorithm is introduced to process the fault feature data and normal feature data. The fuzzy evaluation algorithm can handle the uncertainty and ambiguity in the data. Through this algorithm, the fault transfer feature fuzziness of various fault condition events of the laser engraving equipment is obtained. This fuzziness quantifies the fuzzy characteristics in the process of transition from the normal state to the fault state, reflecting the possibility and trend of the fault. The fault transfer feature fuzziness matrix is ​​constructed based on the calculated fault transfer feature fuzziness of various fault condition events. This matrix systematically organizes the fuzziness information under different fault conditions. Then a Bayesian network model is constructed, and the fault transfer feature fuzziness matrix is ​​embedded in it for encoding learning. The Bayesian network model is a probabilistic graph model that can represent the probabilistic relationship between variables. Through this embedded learning, the model can learn the complex relationship between fault feature data, normal feature data and fault transfer feature fuzziness. The real-time working parameters of the laser engraving device at the preset time node are obtained and imported into the Bayesian network model for evaluation, so as to obtain the fault transfer feature fuzziness of the device at the current preset time node. Then compare this fuzziness with the preset threshold. If it is greater than the preset threshold, it means that the device is currently in a high fault risk state, so the device is controlled to shut down and generate fault warning information to avoid possible damage or poor engraving results. If the fault transfer feature fuzziness of the current preset time node is not greater than the preset threshold, the fault transfer feature fuzziness of the device at several future time nodes is further obtained based on the Bayesian network model. If there is at least one future time node whose fault transfer feature fuzziness is greater than the preset threshold, the device is also controlled to shut down and generate fault warning information. This step realizes the prediction of future equipment failures, takes measures in advance to prevent failures, reduces equipment downtime and maintenance costs, and improves equipment reliability and production efficiency. Through the above series of steps, this control method can comprehensively analyze the fault conditions of laser engraving equipment. From collecting fault and normal status data to building a fault prediction system using fuzzy evaluation algorithms and Bayesian network models, we can not only accurately assess the current fault risk of the equipment, but also predict the possibility of future faults. This helps to timely discover potential equipment failures, take shutdown and early warning measures in advance, and avoid the impact of equipment failures on production, such as engraving interruptions and product quality degradation, thereby improving the stability, reliability and production efficiency of the equipment and reducing maintenance costs.

[0112] In addition, in practical applications, the control method may further include the following steps:

[0113] When the laser engraving head has finished engraving a certain position node of the glass and the laser engraving head has moved to the next position node for engraving, the CCD camera is used to obtain an actual engraving working condition image of the certain position node that has been engraved;

[0114] Performing matrix conversion processing on the actual engraving working condition image to obtain the actual engraving working condition gray level co-occurrence matrix of the certain position node; calculating the structural similarity index between the actual engraving working condition gray level co-occurrence matrix and the preset engraving working condition gray level co-occurrence matrix;

[0115] If the structural similarity index is not greater than the preset index value, the element values ​​of the same matrix position in the gray-level co-occurrence matrix of the actual engraving condition are compared with the element values ​​of the gray-level co-occurrence matrix of the preset engraving condition, and the matrix positions with different element values ​​are marked to obtain the defect position area of ​​the certain position node;

[0116] Determine whether the defective location region of the certain location node has extended to the non-engraving region; if the defective location region of the certain location node has extended to the non-engraving region, segment the region image corresponding to the defective location region in the actual engraving working condition image based on the watershed algorithm;

[0117] Perform feature recognition on the regional image corresponding to the defect position area to obtain the defect type of the defect position area. If the defect type of the defect position area is a concave defect (such as a crack or a etched hole), the laser engraving device is controlled to stop and the current engraved semi-finished product is scrapped; if the defect type of the defect position area is a convex defect (such as welding slag or a bulge), the laser engraving device is controlled to continue working and the current engraved semi-finished product is marked as a product to be repaired;

[0118] If the defective position area of ​​the certain position node does not extend to the non-engraving area, the laser engraving device is controlled to continue working, and the current engraved semi-finished product is marked as a product to be repaired.

[0119] It should be noted that when the laser engraving head completes the engraving of a position node on the glass and transfers to the next position node, the CCD camera is used to obtain the actual engraving condition image of the position node that has been engraved. The CCD camera can capture the actual surface condition of the position node after engraving, and provide original image data for subsequent quality inspection. The obtained actual engraving condition image is subjected to matrix transformation processing to obtain the actual engraving condition grayscale co-occurrence matrix. The grayscale co-occurrence matrix can reflect the spatial correlation characteristics of the pixel grayscale in the image. Then calculate the structural similarity index between the actual engraving condition grayscale co-occurrence matrix and the preset engraving condition grayscale co-occurrence matrix. The preset engraving condition grayscale co-occurrence matrix is ​​a pre-set matrix representing the ideal engraving effect. By comparing the structural similarity indexes of the two, the degree of difference between the actual engraving effect and the ideal effect can be preliminarily judged. If the structural similarity index is not greater than the preset index value, it indicates that there may be problems with the actual engraving effect. At this time, the element values ​​of the same matrix position in the grayscale co-occurrence matrix of the actual engraving condition are compared with the element values ​​of the preset engraving condition grayscale co-occurrence matrix, and the matrix positions with different element values ​​are marked. These marked positions constitute the defect position area of ​​the position node. This step accurately finds the area where the actual engraving is different from the ideal state by comparing the matrix elements. Then determine whether the defect position area has extended to the non-engraving area. This judgment is to distinguish the severity and scope of the defect. If the defect has extended to the non-engraving area, it will have a more serious impact on the quality and structural integrity of the entire engraving.

[0120] If the defect location area has extended to the non-engraving area, the regional image corresponding to the defect location area is segmented in the actual engraving working condition image based on the watershed algorithm. The watershed algorithm is an image segmentation algorithm that can divide the image according to different feature areas. Then the segmented regional image is feature identified to determine the defect type. If it is a concave defect (such as cracks, corrosion holes), this type of defect will seriously affect the strength and appearance of the glass, so the laser engraving equipment is controlled to stop and the current engraved semi-finished product is scrapped; if it is a convex defect (such as welding slag, bulge), although it also affects the appearance, it has less impact on the product structure than concave defects, so the laser engraving equipment is controlled to continue working and the current engraved semi-finished product is marked as a product to be repaired.

[0121] If the defective location area of ​​the certain location node does not extend to the non-engraving area, it means that the impact range of the defect is relatively small. At this time, the laser engraving device is controlled to continue working, and the current engraved semi-finished product is marked as a product to be repaired.

[0122] Through the above series of operations, the quality of the engraved position nodes can be inspected in real time during the laser engraving process. By comparing the actual engraving conditions with the preset conditions, the possible defective areas can be accurately found, and the treatment method for the engraved semi-finished products can be determined based on whether the defects extend to the non-engraving area and the type of defects. This helps to timely discover and deal with quality problems in the engraving process, reduce the scrap rate, and improve production efficiency. At the same time, it can reasonably deal with defects of different types and severity, ensuring the quality of laser engraving products and the stability of the production process.

[0123] The above is based on the ideal embodiment of the present invention, and its description is relatively specific and detailed, but it cannot be understood as limiting the scope of the invention patent. It should be pointed out that for ordinary technicians in this field, several modifications and improvements can be made without departing from the concept of the present invention, which all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be based on the attached claims.

Claims

1. A glass cup laser engraving device, characterized in that: The laser engraving device comprises a workbench, a positioning base is arranged on the workbench, and a positioning groove is arranged on the positioning base; a bracket is installed on the workbench, a driving motor is installed on one side of the bracket, and a first bearing seat is installed on the other side; a driving shaft is matched and connected with the output end of the driving motor, and the other end of the driving shaft is matched and connected with the first bearing seat, and a first driving pulley and a second driving pulley are installed on the driving shaft; The workbench is also symmetrically provided with a first bidirectional threaded slide rod and a second bidirectional threaded slide rod, and a first slider and a second slider are slidably connected between the first bidirectional threaded slide rod and the second bidirectional threaded slide rod; a first clamping block is fixedly installed on the first slider, and a second clamping block is fixedly installed on the second slider; A first driven pulley is disposed at one end of the first bidirectional threaded slide rod, and a second driven pulley is disposed at one end of the second bidirectional threaded slide rod; the first driving pulley and the first driven pulley are connected via a first belt transmission, and the second driving pulley and the second driven pulley are connected via a second belt transmission; The workbench is also provided with a five-axis linkage mechanical arm, on which a laser engraving head is installed.

2. The glass cup laser engraving device according to claim 1, characterized in that: The cross-sectional shape of the positioning groove is the same as the cross-sectional shape of the bottom of the glass to be engraved.

3. The glass cup laser engraving device according to claim 1, characterized in that: The first bidirectional threaded slide rod is rotatably mounted on the workbench through a group of second bearing seats; the second bidirectional threaded slide rod is rotatably mounted on the workbench through a group of third bearing seats.

4. The glass cup laser engraving device according to claim 1, characterized in that: The first clamping block and the second clamping block are both provided with pressure sensors, and the pressure sensors are communicatively connected with the driving motor.

5. The glass cup laser engraving device according to claim 1, characterized in that: The workbench is also provided with a CCD camera, and the CCD camera is used to detect the clamping condition and the engraving condition of the glass.

6. A control method for a glass laser engraving device, applied to a glass laser engraving device according to any one of claims 1 to 5, characterized in that: The control method comprises the following steps: Acquire etching feature data of the laser engraving device under various preset engraving parameters, and construct a dictionary tree search map according to the etching feature data of the laser engraving device under various preset engraving parameters; Acquire real-time engraving parameters of a laser engraving device at a preset time node, and acquire predicted etching feature data of the glass at the current preset time node based on the real-time engraving parameters and in combination with a dictionary tree search graph; The laser engraving equipment is adjusted and optimized based on the predicted etching feature data of the glass at the current preset time node.

7. The control method of glass cup laser engraving equipment according to claim 6, characterized in that: Obtain etching feature data of the laser engraving device under various preset engraving parameters, and construct a dictionary tree search map according to the etching feature data of the laser engraving device under various preset engraving parameters, specifically: Testing the laser engraving device to obtain etching feature data of the laser engraving device under various preset engraving parameters; Constructing a dictionary tree, converting various preset engraving parameters into branch nodes of the dictionary tree, and converting various etching feature data into leaf nodes of the dictionary tree; According to the etching feature data of the laser engraving device under various preset engraving parameters obtained by the test, a directed description relationship between various preset engraving parameters and various etching feature data is established; According to the directed description relationship between various preset engraving parameters and various etching feature data, each branch node and each leaf node are feature-combined to form a dictionary tree search map of various preset engraving parameters and various etching feature data; Among them, the preset engraving parameters include the laser power, moving speed and moving direction of the laser engraving head; the etching feature data includes the etching depth, etching width and etching shape corresponding to the corresponding position node of the glass cup after the laser engraving head acts on the corresponding preset engraving parameters.

8. The control method of glass cup laser engraving equipment according to claim 6, characterized in that: The real-time engraving parameters of the laser engraving device are obtained at a preset time node, and the predicted etching feature data of the glass at the current preset time node is obtained according to the real-time engraving parameters and combined with the dictionary tree search map, specifically: Acquire real-time engraving parameters of a laser engraving device at a preset time node, and import the real-time engraving parameters into the dictionary tree search graph; The real-time engraving parameters are converted into the root node of the dictionary tree, a cosine similarity algorithm is introduced, and the cosine similarity between the root node and each branch node in the dictionary tree search graph is calculated based on the cosine similarity algorithm; Constructing a size sorting table, importing the cosine similarity between the root node and each branch node in the dictionary tree search graph into the size sorting table for size sorting, and sorting to obtain the highest cosine similarity; The branch nodes corresponding to the highest cosine similarity are calibrated, the leaf nodes belonging to the calibrated branch nodes are extracted, and the feature information in the extracted leaf nodes is interpreted to obtain the etching feature data corresponding to the extracted leaf nodes; The etching feature data corresponding to the extracted leaf node is output as the predicted etching feature data of the glass at the preset time node.

9. The control method of glass cup laser engraving equipment according to claim 6, characterized in that: The laser engraving equipment is adjusted and optimized according to the predicted etching feature data of the glass at the current preset time node, specifically: Obtaining a preset engraving process scheme for the glass, and obtaining preset engraving feature data range information of the glass at a current preset time node according to the preset engraving process scheme; Determine the maximum preset engraving feature data and the minimum preset engraving feature data of the glass at the current preset time node according to the preset engraving feature data range information; The maximum preset engraving feature data of the glass at the current preset time node is used as the upper boundary condition, and the minimum preset engraving feature data of the glass at the current preset time node is used as the lower boundary condition; Setting a definition domain according to the upper boundary condition and the lower boundary condition, and determining whether the predicted etching feature data of the glass cup at the current preset time node completely falls within the definition domain; If the predicted etching feature data of the glass cup at the current preset time node completely falls within the definition domain, the real-time engraving parameters of the laser engraving device are not adjusted, and the real-time engraving parameters of the laser engraving device are continuously monitored at the next preset time node; If the predicted etching feature data of the glass cup at the current preset time node partially or completely falls outside the definition domain, the particle swarm optimization algorithm is introduced, and the real-time engraving parameters of the laser engraving device are iteratively optimized based on the particle swarm optimization algorithm, and the real-time engraving parameters of the laser engraving device at the next preset time node are readjusted so that the predicted etching feature data at the next time node falls within the definition domain again.

10. The control method of glass cup laser engraving equipment according to claim 9, characterized in that: If the predicted etching feature data of the glass at the current preset time node partially or completely falls outside the definition domain, the particle swarm optimization algorithm is introduced to iteratively optimize the real-time engraving parameters of the laser engraving device based on the particle swarm optimization algorithm, and the real-time engraving parameters of the laser engraving device at the next preset time node are readjusted so that the predicted etching feature data at the next time node falls within the definition domain again, specifically: Determine the particle population size in the particle swarm optimization algorithm, where each particle corresponds to a set of real-time engraving parameters of the laser engraving device, including the laser power, moving speed, and moving direction of the laser engraving head; Initialize the position and velocity vectors of the particles, where the position vector represents a specific combination of engraving parameters, and the velocity vector represents the adjustment direction and amplitude of the parameters; A fitness function is constructed based on the adaptability of the predicted etching feature data and the definition domain, and the fitness value of each particle is calculated. The fitness value reflects the effectiveness of the engraving parameter combination represented by the particle in making the predicted etching feature data fall into the definition domain. For each particle, compare its current fitness value with the individual historical optimal fitness value. If it is better, update the individual optimal position; at the same time, compare in the entire particle group and update the global optimal position; According to the particle's current position, speed, individual optimal position and global optimal position, continue to update the particle's speed and position vector to generate a new carving parameter combination; The above steps are continuously iterated until the predicted etching feature data completely falls into the definition domain, and the engraving parameters corresponding to the global optimal position are set as the real-time engraving parameters of the next preset time node, thereby ensuring that the predicted etching feature data of the next time node falls into the definition domain again.

Images