Method for laser cutting trajectory planning of special-shaped curved surface glass

Abstract

The present application belongs to the technical field of laser processing, and particularly relates to a laser cutting trajectory planning method for special-shaped curved glass. The present application realizes high-precision prediction and compensation of thermal deformation in the laser cutting process by coupling calculation of instantaneous thermal effect and historical thermal cumulative effect, improves the accuracy of the cutting trajectory, avoids profile deviation caused by thermal deformation, and improves the geometric precision. According to the local geometric complexity of the special-shaped curved glass, the region is divided and the differentiated virtual scanning layer and scanning mode are matched, so as to guarantee the dynamic balance of efficiency and precision, ensure the processing quality of high-curvature and thin-wall regions, improve the cutting speed of flat regions, and shorten the processing cycle. By monitoring the temperature during cutting and feeding back the actual thermal history parameters, the model parameters are reversely optimized after cutting is completed, so as to enhance the robustness and stability, and improve the yield and process reliability of the laser cutting of the special-shaped curved glass.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of laser processing, and particularly relates to a laser cutting trajectory planning method for special-shaped curved glass. BACKGROUND

[0002] As an advanced non-contact processing method, laser cutting technology has been widely used in the field of industrial manufacturing due to its high energy density, high precision and high flexibility. Especially in the processing of brittle materials such as glass, laser cutting has become a key technology for realizing complex shape cutting due to its small heat-affected zone and no mechanical stress. Special-shaped curved glass, such as automobile sunroofs and smart phone 3D covers, has complex non-planar geometry, which puts high requirements on the trajectory planning and process control of laser cutting. The core is how to accurately control the laser beam to follow the predetermined contour in three-dimensional space.

[0003] In the existing practice of laser cutting of special-shaped curved glass, the commonly used trajectory planning method is to directly generate a cutting path based on a three-dimensional CAD model, and to control the movement of the laser head through a multi-axis linkage numerical control system. To deal with the problem of thermal deformation generated during processing, some methods introduce a fixed or simple linear model-based compensation amount to preliminarily correct the theoretical path. At the same time, in the scanning strategy, a global unified scanning parameter is often used, such as constant scanning speed and laser power, to simplify the process flow and control complexity.

[0004] However, the above existing technology has certain deficiencies in processing high requirement special-shaped curved glass: (1) The compensation method based on fixed value or simple model cannot accurately cope with the nonlinear and uneven thermal deformation caused by local geometric changes on the complex curved surface, especially in the regions with drastic changes in curvature and thickness, which is easy to cause overcut or undercut. (2) The global unified scanning strategy lacks pertinence and cannot meet the differentiated needs of different regions for precision and efficiency, resulting in low efficiency in flat areas or sacrificing precision in complex areas. (3) The existing method generally ignores the heat accumulation effect caused by the processing path sequence, which makes the local thermal stress too high when cutting dense paths, increasing the risk of material cracking and edge collapse. SUMMARY

[0005] In view of this, in order to solve the problems raised in the background art, a laser cutting trajectory planning method for special-shaped curved glass is proposed.

[0006] The object of the present application can be achieved by the following technical solutions: The present application provides a method for laser cutting trajectory planning of special-shaped curved glass, comprising: S1, taking the to-be-cut special-shaped curved glass as a target special-shaped curved glass, obtaining a three-dimensional digital model of the target special-shaped curved glass, and after identifying the curved surface geometric characteristic parameters, dividing the special-shaped curved surface into key feature regions and non-key feature regions, and generating feature-sensitive sub-regions.

[0007] S2, based on the geometric complexity of each sub-region in the feature-sensitive sub-region and the preset accuracy, dynamically generating corresponding virtual scanning layer parameters, and assigning virtual scanning mode parameters to each sub-region.

[0008] S3, combining the curved surface geometric characteristic parameters and the preset material thermal physical property parameters, generating a basic thermal deformation prediction value of the target special-shaped curved glass, and simultaneously obtaining the time sequence thermal history parameters of the scanned path of the target special-shaped curved glass.

[0009] S4, coupling calculation of the basic thermal deformation prediction value and the time sequence thermal history parameters, generating a space vector compensation field, and correcting the spatial position of the theoretical cutting path according to the space vector compensation field, and generating the actual laser cutting trajectory of the target special-shaped curved glass.

[0010] S5, based on the scanning mode parameters and the actual laser cutting trajectory, controlling the laser cutting equipment to perform cutting operation.

[0011] Compared with the prior art, the present application has the following advantages: 1, the present application realizes high-precision prediction and compensation of thermal deformation in the laser cutting process by coupling calculation of instantaneous thermal effect and historical thermal accumulation effect, fully considers the influence of local geometric characteristics of the material on thermal deformation, is beneficial to improving the accuracy of the cutting trajectory, effectively avoids the contour deviation caused by thermal deformation, and improves the geometric accuracy of the final product.

[0012] 2, according to the local geometric complexity of the special-shaped curved glass, the present application intelligently divides the region and matches the differentiated virtual scanning layer and scanning mode, which is beneficial to the dynamic balance of efficiency and accuracy in the processing process, ensures the processing quality of key regions such as high curvature and thin wall, and improves the cutting speed of flat regions, so as to shorten the overall processing cycle without sacrificing product quality.

[0013] 3, the present application realizes real-time monitoring of temperature and feedback of actual thermal history parameters during cutting, and collects contour accuracy data after cutting to optimize model parameters in reverse, enhances the robustness and stability of long-term application, and improves the yield and process reliability of laser cutting of special-shaped curved glass. BRIEF DESCRIPTION OF DRAWINGS

[0014] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the description of the embodiments. Obviously, the drawings described in the following description are only some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor based on these drawings.

[0015] Figure 1 The schematic diagram for implementing the method steps of the present application is shown. DETAILED DESCRIPTION

[0016] The technical solutions in the embodiments of the present application will be clearly and completely described in the following with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the protection scope of the present application.

[0017] Embodiment one

[0018] Please refer to Figure 1 As shown in the figure, the present application provides a laser cutting trajectory planning method for special-shaped curved glass, and the specific steps are as follows: S1, the special-shaped curved glass to be cut is recorded as a target special-shaped curved glass, a three-dimensional digital model of the target special-shaped curved glass is obtained, and after identifying the geometric feature parameters of the curved surface, the special-shaped curved surface is divided into a key feature region and a non-key feature region, and a feature sensitive partition is generated.

[0019] In a preferred feasible embodiment of the present application, the specific process of identifying the geometric features of the curved surface includes: discretizing the three-dimensional digital model of the target special-shaped curved glass into a triangular mesh model composed of a plurality of vertices and surface elements, and generating a curvature change rate distribution map of the target special-shaped curved glass according to this.

[0020] It should be noted that the generation of the curvature change rate distribution map of the target special-shaped curved glass is specifically: analyzing the spatial relationship between each vertex and adjacent vertices on the triangular mesh model, calculating the principal curvature of the vertex, further solving the average curvature representing the local bending degree, and calculating the spatial gradient of the average curvature in the vertex neighborhood to obtain the curvature change rate. The curvature change rate values of the plurality of vertices are normalized to form a numerical curvature change rate distribution map, which is recorded as the curvature change rate distribution map of the target special-shaped curved glass. Wherein, the curvature change rate directly reflects the transition speed of the curved surface from gentle to steep.

[0021] The normal thickness value of each of the plurality of vertices is extracted from the three-dimensional digital model and compared with the thickness deviation of adjacent vertices to obtain a thickness gradient vector of the target special-shaped curved glass, and a modulus value thereof is extracted and normalized to obtain a thickness gradient change matrix of the target special-shaped curved glass.

[0022] It should be noted that the thickness gradient vector indicates the direction and rate of the fastest thickness change, and the thickness gradient change matrix represents the material thickness uniformity.

[0023] The physical boundary profile of the target special-shaped curved glass is identified from the three-dimensional digital model, the shortest path distance of any vertex on the curved surface to the nearest boundary along the curved surface is calculated, which is recorded as the geodesic distance, and a spatial attenuation coefficient field of the target special-shaped curved glass is generated based on a preset attenuation function.

[0024] It should be noted that the spatial attenuation coefficient field of the target special-shaped curved glass is specifically: based on the geodesic distance of a plurality of vertices of the target special-shaped curved glass, a spatial attenuation coefficient of the plurality of vertices is generated by a preset attenuation function, such as an exponential attenuation function, so that the spatial attenuation coefficients of all vertices constitute the spatial attenuation coefficient field. Wherein, the spatial attenuation coefficient is the largest at the boundary, and decreases with the increase of distance, so as to quantify the influence range and intensity of the boundary thermal effect.

[0025] Exemplarily, assuming that the exponential attenuation function is , wherein S(P) is the spatial attenuation coefficient of vertex P, the value range is ; d(P) is the geodesic distance (unit: mm) of vertex P to the nearest boundary; k is the attenuation coefficient, i.e. the empirical constant, which is set according to the characteristics of the glass material, for example, taking , which is used to adjust the attenuation speed.

[0026] Taking three vertices on the special-shaped curved glass as an example: vertex A: located on the glass boundary, the geodesic distance , then: , the boundary thermal effect is the strongest, and the attenuation coefficient is the largest.

[0027] Vertex B: the geodesic distance to the boundary is , then: , the distance increases, the thermal effect weakens, and the attenuation coefficient decreases.

[0028] Vertex C: the geodesic distance to the boundary is , then: , the distance is far, the thermal effect is significantly weakened, and the attenuation coefficient is close.

[0029] The curvature change rate distribution map, the thickness gradient change matrix and the spatial attenuation coefficient field are fused to output the curved surface geometric feature parameters.

[0030] It should be noted that the curved surface geometry characteristic parameter is obtained by weighting and fusing the curvature rate distribution map, the thickness gradient change matrix and the spatial attenuation coefficient field, and for any point P on the curved surface, the curved surface geometry characteristic parameter G(P) is calculated by the following fusion formula: Wherein G(P) is the curved surface geometry characteristic parameter of point P; is the normalized curvature rate at point P; is the normalized thickness gradient modulus at point P; S(P) is the spatial attenuation coefficient at point P. is a preset dimensionless weight coefficient, the sum of which is 1, which respectively defines the weight factor of the curvature, the thickness change and the boundary effect on the laser processing sensitivity. By performing the fusion calculation on all points on the curved surface, the final output of the curved surface geometry characteristic parameter is a composite parameter field that can comprehensively and accurately reflect the processing difficulty and risk level of each local area of the special-shaped curved surface.

[0031] In a preferred feasible embodiment of the present application, the specific process of generating the feature sensitive partition includes: setting key feature area judgment thresholds, including curvature mutation threshold, thin wall thickness threshold and boundary proximity threshold.

[0032] Traverse the curved surface geometry characteristic parameters of the target special-shaped curved glass, mark several curved surface regions where the curvature rate is higher than the curvature mutation threshold or the thickness gradient is lower than the thin wall thickness threshold or the boundary distance is less than the boundary proximity threshold as key feature areas, and mark the remaining several regions as non-key feature areas, thereby obtaining the feature sensitive partition of the target special-shaped curved glass.

[0033] It should be noted that the generation of the feature sensitive partition aims to identify and process isolated points or small regions in the partition map, merge all adjacent units marked as key feature areas into connected clusters, and calculate the area of each cluster. The cluster with an area smaller than the preset minimum size threshold will be regarded as processing noise, and the marking of the internal unit will be reclassified as a non-key feature area. In order to ensure smooth transition of laser scanning parameters between different regions, the boundary line between the key feature area and the non-key feature area is smoothed and optimized, for example, the Laplace smoothing algorithm is used to adjust the boundary vertex position to eliminate sharp corners and jagged edges.

[0034] In a specific example, the preset minimum size threshold can be 5mm², which comprehensively considers the effective processing range corresponding to the device minimum spot diameter of 0.1mm, and the filtering demand for small protrusions, depressions and other non-functional areas in actual production.

[0035] The application realizes high-precision prediction and compensation of thermal deformation in the laser cutting process by coupling calculation of transient thermal effect and historical thermal cumulative effect, fully considers the influence of local geometric characteristics of the material on thermal deformation, is beneficial to improving the accuracy of the cutting track, effectively avoids the profile deviation caused by thermal deformation, and improves the geometric precision of the final product.

[0036] S2, based on the geometric complexity of each sub-region in the feature sensitive partition and the preset accuracy, dynamically generating corresponding virtual scanning layer parameters, and allocating virtual scanning mode parameters for each sub-region.

[0037] In a preferred feasible embodiment of the application, the specific process of dynamically generating corresponding virtual scanning layer parameters includes: extracting the geometric complexity score of each sub-region in the feature sensitive partition, matching it with the preset layer number mapping table, and determining the virtual layer number of each sub-region in the feature sensitive partition.

[0038] Table 1: preset layer number mapping example

[0039]

[0040] The preset accuracy of each sub-region in the feature sensitive partition is extracted, matched with the preset interval compensation coefficient mapping table, and the interlayer interval compensation coefficient of each sub-region in the feature sensitive partition is determined.

[0041] Table 2: preset interval compensation coefficient mapping example

[0042]

[0043] The virtual scanning layer parameters of each sub-region in the feature sensitive partition are generated in combination with the virtual layer number and the interlayer interval compensation coefficient.

[0044] It should be noted that the geometric complexity score is obtained by aggregating the surface geometric feature parameter values of all calculation units in the sub-region, and the average value or the maximum value of the parameter values in the sub-region is usually used as the final geometric complexity score.

[0045] The preset layer number mapping table is a pre-established database or function, which maps different ranges of geometric complexity scores to specific virtual layer numbers. The higher the score, the more complex the geometry or the higher the thermal sensitivity of the sub-region, and the more virtual layer numbers will be allocated to it, and vice versa.

[0046] The interlayer spacing compensation coefficient is used to fine-tune the depth interval between layers. It is directly related to the preset global or regional accuracy requirements. Based on the preset tolerance value, a dimensionless compensation coefficient is generated by looking up a table. Higher accuracy requirements correspond to smaller tolerances, thus producing a smaller compensation coefficient to reduce the interlayer spacing.

[0047] The virtual scan layer parameters explicitly define the total number of scan layers required for the sub-region, as well as the precise depth position of each layer in the glass thickness direction, thereby translating macroscopic strategies into executable microscopic operation instructions.

[0048] In a preferred embodiment of the present invention, the virtual scanning mode parameters include laser spot size, power, frequency, and scanning speed.

[0049] The specific process of assigning virtual scanning mode parameters to each sub-region includes: comparing the virtual scanning layer parameters of each sub-region with a preset threshold; if the virtual scanning layer parameters of a sub-region are greater than or equal to the preset threshold, then a fine-cut scanning mode is assigned to that sub-region; otherwise, an efficient scanning mode is assigned to that sub-region.

[0050] For example, the specific mode parameters included in the precision cutting scanning mode are as follows: laser spot size: 0.05-0.1mm, which is a small spot to improve the clarity of the outline; power: 10-20W, which is low power to reduce thermal deformation and avoid glass edge chipping; frequency: 50-100kHz, which is high frequency to achieve fine dot marking and improve the smoothness of cutting; scanning speed: 50-100mm / s, which is slow speed to ensure that energy is applied evenly to the material.

[0051] The high-efficiency scanning mode includes the following parameters: laser spot size: 0.15-0.3mm, which means a large spot with wide coverage and improved efficiency; power: 30-50W, which means high power to accelerate material ablation and shorten processing time; frequency: 10-30kHz, which means low frequency to reduce the number of dots and match high-speed scanning; scanning speed: 200-500mm / s, which means fast speed to improve the overall processing rhythm.

[0052] Based on the local geometric complexity of irregular curved glass, this invention intelligently divides regions and matches differentiated virtual scanning layers and scanning modes, which facilitates a dynamic balance between efficiency and accuracy during processing. This ensures the processing quality of key areas such as high curvature and thin walls, and improves the cutting speed of flat areas, thereby shortening the overall processing cycle without sacrificing product quality.

[0053] S3. Combining the surface geometric feature parameters and the preset material thermophysical property parameters, generate the basic thermal deformation prediction value of the target irregular curved glass, and at the same time obtain the time-series thermal history parameters of the scanned path of the target irregular curved glass in real time.

[0054] In one preferred embodiment of the present application, the specific process of generating the basic thermal deformation prediction value of the target profiled curved glass comprises: inputting the curved geometric feature parameters and the preset material thermal physical property parameters in the finite element simulation of the target profiled curved glass; through the combination of the curved geometric feature parameters and the preset material thermal physical property parameters, the surface of the target profiled curved glass will generate a transient thermal deformation value at a certain point, which is recorded as the basic thermal deformation prediction value of the target profiled curved glass at the point, and the basic thermal deformation prediction value of the target profiled curved glass is generated accordingly.

[0055] S4, coupling calculation of the basic thermal deformation prediction value and the time sequence thermal history parameter, generating a space vector compensation field, and correcting the spatial position of the theoretical cutting path according to the space vector compensation field, to generate the actual laser cutting track of the target profiled curved glass.

[0056] In one preferred embodiment of the present application, the specific process of generating the space vector compensation field comprises: establishing a basic thermal deformation prediction model taking local geometric features as input.

[0057] A historical thermal coupling influence model containing a heat conduction equation is constructed.

[0058] It should be noted that the core of the historical thermal coupling influence model is to solve the non-steady-state heat conduction partial differential equation, so as to simulate the accumulation and diffusion process of heat in the glass matrix.

[0059] The time sequence thermal history parameter is input into the historical thermal coupling influence model, and the historical thermal coupling deformation value is output.

[0060] Specifically, the time sequence thermal history parameter accurately recording the position, processing time and laser energy used at that time of each processed point is taken as the dynamic input of the historical thermal coupling influence model, the model calculates the cumulative temperature rise caused by these historical heat sources on the current point to be processed, and deduces the historical thermal coupling deformation value generated thereby, i.e. the output historical thermal coupling deformation value.

[0061] Exemplarily, the specific form of the historical thermal coupling influence model can be: (1) cumulative temperature rise calculation, i.e. the diffusion of heat in the glass is described based on the non-steady-state heat conduction equation: , wherein T(r,t) is the temperature of position r at time t; is the thermal diffusion coefficient of the glass, such as about for ordinary glass; is the Laplace operator, representing the spatial temperature gradient; is the heat source term, related to the laser energy of the processed point, and the expression is: , n is the number of processed points; 0.6 for the efficiency of laser energy conversion into heat; is the laser energy of the i-th processed point; is the heat affected volume of the point; is the processing time and position of the point, respectively; is a time decay function, such as exponential decay is the decay coefficient.

[0062] (2) Thermal deformation calculation, i.e. based on cumulative temperature rise , is the ambient temperature, and the deformation is derived by the thermal expansion formula: wherein is the historical thermal coupling deformation at position r; is the original characteristic length of the position, such as the distance from the processed point to the point to be processed; is the linear expansion coefficient of the glass, such as about .

[0063] The basic thermal deformation prediction value is superimposed with the historical thermal coupling deformation to generate the total predicted deformation.

[0064] Specifically, the basic thermal deformation prediction value representing the local instantaneous effect is vector added with the historical thermal coupling deformation representing the global cumulative effect, thereby generating the total predicted deformation. The specific process can be expressed by the following formula: wherein is the total predicted deformation vector of the current point to be processed i; is the basic thermal deformation prediction value vector calculated by the basic thermal deformation prediction model, which is affected by the local geometric characteristics of the point and the scanning mode parameters is the historical thermal coupling deformation vector calculated by the historical thermal coupling influence model, which is caused by the cumulative heat of all the processed paths before the point i.

[0065] According to the total predicted deformation, a spatial vector compensation field is constructed.

[0066] It should be noted that according to the total predicted deformation of each point to be processed, a spatial vector compensation field covering the entire cutting path is constructed, and the vector direction of each point in the spatial vector compensation field is opposite to the direction of the total predicted deformation, and the size is equal, which provides accurate compensation basis for subsequent trajectory correction.

[0067] ​In a preferred feasible embodiment of the present application, the specific process of generating the actual laser cutting trajectory of the target special-shaped curved glass comprises: vector superposition of the space vector compensation field and a plurality of theoretical coordinate points of the theoretical cutting path stored in the database to verify whether the offset path points are located in the original feature sensitive partition.

[0068] It should be noted that the theoretical cutting path is a sequence composed of a series of space coordinate points, and the sequence composed of the space coordinate points is denoted as a plurality of theoretical coordinate points.

[0069] The vector superposition is to extract the corresponding total predicted deformation vector from the space vector compensation field for the plurality of theoretical coordinate points, and to correct the space position of the theoretical coordinate points through the vector superposition operation. Specifically, the coordinate vector of the theoretical coordinate point is subtracted by the corresponding total predicted deformation vector in the space vector compensation field to calculate the compensated coordinate point. This process can be accurately described by the following formula: , wherein represents the i-th point coordinate on the corrected actual laser cutting trajectory; is the i-th point coordinate on the theoretical cutting path; is the total predicted deformation vector corresponding to the point obtained from the space vector compensation field.

[0070] After generating the preliminary actual laser cutting trajectory, the key verification link is entered, that is, whether each offset path point is still located in the feature sensitive partition of the original theoretical point. This step is completed by querying the space position of and comparing the partition category to which it belongs with the original partition category of .

[0071] If the offset path point is not located in the original feature sensitive partition, it is determined that the offset amount of the offset path point is greater than the corresponding safety threshold, and a dynamic adjustment instruction of the scanning mode parameter is triggered.

[0072] The actual laser cutting trajectory is regenerated based on the adjusted scanning mode parameter.

[0073] Specifically, when the system detects that any one of the corrected points When the offset across the partition boundary, i.e. the partition where the offset belongs to is not the same as the original partition, it is determined that the offset is too large, which destroys the consistency of the preset strategy. At this time, the system will immediately trigger a dynamic scanning mode parameter dynamic adjustment instruction. The instruction carries the position information of the violation point and the offset details, and is sent back to the scanning strategy module. Based on the instruction, the system will adjust the scanning mode parameters of the violation area, and then the whole thermal deformation prediction and compensation process will be re-executed using the adjusted new parameters, so as to regenerate a new spatial vector compensation field and a new actual laser cutting trajectory based on the adjusted scanning mode parameters. The closed-loop iteration process of "compensation-verification-adjustment-recompensation" will continue until all points on the actual laser cutting trajectory pass the verification, i.e. no cross-partition offset occurs, and finally a stable and self-consistent path is output.

[0074] In a preferred feasible embodiment of the present application, the specific content of the scanning mode parameter dynamic adjustment instruction includes: identifying the specific area where the cross-border offset occurs, and increasing the number of virtual layers of the local area of the specific area that exceeds the boundary.

[0075] It should be noted that increasing the number of virtual layers of the local area of the specific area that exceeds the boundary means that the cutting that may have been planned to be completed with fewer layers is now decomposed into more times and shallower layers of scanning, aiming to reduce the instantaneous thermal shock by reducing the single energy injection.

[0076] Switching the scanning mode of the local area of the specific area that exceeds the boundary to the fine cutting scanning mode.

[0077] Updating the boundary buffer zone parameters of the feature-sensitive partition.

[0078] Feeding the adjusted number of virtual layers, the switched fine cutting scanning mode and the boundary buffer zone parameters back to the spatial vector compensation field generation step for iterative optimization.

[0079] It should be noted that the iterative optimization is an iterative optimization process based on a new conservative strategy for the local area of the specific area that exceeds the boundary. The thermal deformation will be recalculated using these adjusted parameters until the generated compensation path is stable within the partition.

[0080] S5, based on the scanning mode parameters and the actual laser cutting trajectory, controlling the laser cutting equipment to perform cutting operation.

[0081] In a preferred feasible embodiment of the present application, the specific content of the control of the laser cutting equipment to perform cutting operation includes: generating a sequence of spatial positions of laser focal points according to the actual laser cutting trajectory.

[0082] convert the scan pattern parameters into a set of device control instructions.

[0083] Real-time monitoring of the temperature distribution field of the cutting area, when the temperature exceeds the preset safety threshold, triggering the cooling intervention mechanism and recording the actual thermal history parameters, while feeding back to the timing thermal history parameter library for updating.

[0084] It should be noted that, at the same time of triggering the cooling intervention mechanism, record all relevant data of this event, including the precise space-time coordinates of the overheating point, peak temperature, cooling intervention time and intensity, etc., which together constitute the actual thermal history parameters. The actual thermal history parameters will be transmitted back to the core database, to update and calibrate the timing thermal history parameter library used in the planning stage, so as to provide more realistic input data for subsequent planning tasks.

[0085] Collect the curved surface contour accuracy data after cutting, and calculate the deviation distribution diagram of the actual contour and the theoretical model.

[0086] It should be noted that the specific way to obtain the deviation distribution diagram is: using a high-precision three-dimensional optical scanner to accurately measure the cutting edge of the cut completed special-shaped curved glass workpiece, collecting its three-dimensional space coordinates, so as to obtain the curved surface contour accuracy data representing the physical form of the final product. Align the curved surface contour accuracy data containing the actual size with the original theoretical three-dimensional model data. After alignment, the normal distance between each point on the actual contour and the corresponding point on the theoretical model is calculated, so as to quantify the processing error, and the error value obtained is visualized and rendered on the three-dimensional model to generate the deviation distribution diagram.

[0087] Optimize the feature sensitive partition judgment threshold based on the deviation distribution diagram.

[0088] It should be noted that the specific process of optimizing the feature sensitive partition judgment threshold is: depth analysis of the deviation distribution diagram, correlation analysis of the error significant area and the original curved surface geometric feature parameters of the area. For example, if it is found that the deviation value of a flat curved surface divided into a non-key area exceeds the allowable range, it is determined that the initial set feature sensitive partition judgment threshold is too loose. Accordingly, the judgment threshold is automatically adjusted and optimized, for example, the sensitivity of the curvature mutation or thickness gradient is improved, so as to more accurately identify such areas as key feature areas in future processing tasks.

[0089] Update the thermal deformation prediction model parameters in the space vector compensation field.

[0090] It should be noted that the specific process of updating the thermal deformation prediction model parameter in the space vector compensation field is that the deviation value is regarded as the residual error between the model prediction and the actual physical deformation, the physical coefficients or weight parameters in the thermal deformation prediction model are fine-tuned through an optimization algorithm such as back propagation or least square method, so that the model can be closer to the actual deformation amount in the next prediction, thereby completing the updating of the thermal deformation prediction model parameter.

[0091] Exemplarily, the thermal deformation prediction model takes the curved surface geometry characteristic parameter, the material thermal physical characteristic parameter and the scanning mode parameter as input, calculates the thermal deformation prediction value through a function relationship containing physical coefficients, and the specific form is as follows: , wherein is a basic thermal deformation prediction value; is a normalized curvature change rate, reflecting the influence of the curved surface bending degree on the thermal deformation, and the value is 0-1; is a normalized thickness gradient module, reflecting the influence of the thickness uniformity on the thermal deformation, and the value is 0-1; is a spatial attenuation coefficient, reflecting the influence of the boundary effect on the thermal deformation, and the value is 0-1; is a laser power; is a scanning speed; are physical coefficients in the model respectively, and respectively represent the contribution degree of the curvature, the thickness, the boundary, the power and the speed to the thermal deformation.

[0092] The present application enhances the robustness and long-term application stability by monitoring the temperature in real time during the cutting process and feeding back the actual thermal history parameter, and collecting the contour precision data after the cutting is completed to optimize the model parameter in reverse, and improves the yield rate and process reliability of the special-shaped curved surface glass laser cutting.

[0093] Embodiment two

[0094] In the second embodiment of the present application, in combination with the above-mentioned special-shaped curved surface glass laser cutting track planning method, the present application provides the following technical scheme, a storage medium, the storage medium stores a computer program, and the computer program is executed by a processor to realize the above-mentioned special-shaped curved surface glass laser cutting track planning method.

[0095] Those skilled in the art will appreciate that the logic and steps of a flowchart depicted herein, and elsewhere described herein, such as can be viewed as a table of sequenced data entries for performing a logical function, can be embodied as executable instructions stored in any computer-readable medium for execution by an instruction execution system, apparatus, or device, such as a computer-based system, processor-containing system, or other system that can fetch the instructions from the instruction execution system, apparatus, or device and execute the instructions. In the context of this specification, a "computer-readable medium" can be any means that can contain, store, communicate, propagate or transport the program for use by or in connection with the instruction execution system, apparatus, or device. The computer-readable medium can be, for example but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, device, propagation medium, or computer memory.

[0096] More specific examples (a non-exhaustive list) of the computer-readable medium include the following: an electrical connection (electronic) having one or more wires, a portable computer diskette (magnetic), a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber (optical), and a portable compact disc read-only memory (CDROM). Additionally, the computer-readable medium can also be paper or another suitable medium upon which the program is printed, as the program can be electronically captured, for example via an optical scanner, then compiled, interpreted, or otherwise processed, and stored in a computer memory in a form that can be later executed by a computer. In some embodiments, the computer-readable medium can be a computer program product that can be traded or sold, for example, a floppy disk, a CD-ROM, an optical disk, and the like.

[0097] It should be understood that aspects of the application can be implemented in hardware, software, firmware or combinations thereof. In the above embodiments, various steps or methods can be implemented in software or firmware that is stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, any of the following technologies, known in the art, or combinations thereof, can be used: a discrete logic circuit having logic gates for implementing logic functions upon data signals, an application specific integrated circuit having appropriate combinational logic gates, a programmable gate array (PGA), a field programmable gate array (FPGA), and / or the like.

[0098] The foregoing is merely illustrative of the principles of this application and various modifications can be made by persons skilled in the art without departing from the scope and nature of the application as defined by the appended claims.

Claims

1. A method for planning the trajectory of laser cutting of irregular curved glass, characterized in that: The method comprises the following steps: S1, a to-be-cut irregular curved glass is recorded as a target irregular curved glass, a three-dimensional digital model of the target irregular curved glass is obtained, and after identifying the curved surface geometric characteristic parameters, the irregular curved surface is divided into a key characteristic region and a non-key characteristic region, and a characteristic sensitive partition is generated; S2, based on the geometric complexity of each sub-region in the characteristic sensitive partition and the preset accuracy, corresponding virtual scanning layer parameters are dynamically generated, and virtual scanning mode parameters are allocated to each sub-region; The specific process of dynamically generating corresponding virtual scanning layer parameters comprises: extracting the geometric complexity score of each sub-region in the characteristic sensitive partition, matching it with a preset number of layers mapping table, and determining the number of virtual layers of each sub-region in the characteristic sensitive partition; extracting the preset accuracy of each sub-region in the characteristic sensitive partition, matching it with a preset interval compensation coefficient mapping table, and determining the interlayer interval compensation coefficient of each sub-region in the characteristic sensitive partition; combining the number of virtual layers and the interlayer interval compensation coefficient, the virtual scanning layer parameters of each sub-region in the characteristic sensitive partition are generated; S3, combining the curved surface geometric characteristic parameters and the preset material thermal physical characteristic parameters, a basic thermal deformation prediction value of the target irregular curved glass is generated, and a time sequence thermal history parameter of a scanned path of the target irregular curved glass is obtained in real time; The specific process of generating the basic thermal deformation prediction value of the target irregular curved glass comprises: inputting the curved surface geometric characteristic parameters and the preset material thermal physical characteristic parameters in the finite element simulation of the target irregular curved glass, and through the combination of the curved surface geometric characteristic parameters and the preset material thermal physical characteristic parameters, the surface of the target irregular curved glass will generate a transient thermal deformation at a certain point, which is recorded as the basic thermal deformation prediction value of the target irregular curved glass at the point, and the basic thermal deformation prediction value of the target irregular curved glass is generated accordingly; S4, coupling calculation is performed on the basic thermal deformation prediction value and the time sequence thermal history parameter, a space vector compensation field is generated, and the theoretical cutting path is corrected in space position offset according to the space vector compensation field, and an actual laser cutting track of the target irregular curved glass is generated; S5, based on the scanning mode parameters and the actual laser cutting track, a laser cutting device is controlled to perform a cutting operation. 2.The method of claim 1, wherein: The specific process of identifying the curved surface geometric characteristics comprises: discretizing the three-dimensional digital model of the target irregular curved glass into a triangular mesh model composed of a plurality of vertices and face elements, and generating a curvature change rate distribution map of the target irregular curved glass accordingly; extracting the normal thickness value of each vertex in the plurality of vertices from the three-dimensional digital model, and comparing the thickness deviation of the adjacent vertices to obtain a thickness gradient vector of the target irregular curved glass, extracting the modulus value and normalizing to obtain a thickness gradient change matrix of the target irregular curved glass; identifying the physical boundary contour of the target irregular curved glass from the three-dimensional digital model, calculating the shortest path distance of any vertex on the curved surface to the nearest boundary along the curved surface, recorded as the geodesic distance, and generating a space attenuation coefficient field of the target irregular curved glass through a preset attenuation function accordingly; Fusing the curvature change rate distribution map, the thickness gradient change matrix and the spatial attenuation coefficient field, the surface geometry characteristic parameter is outputted.

3. The method of claim 2, wherein: The specific process of generating the feature-sensitive partition includes: Setting a key feature area judgment threshold, including a curvature mutation threshold, a thin-wall thickness threshold and a boundary proximity threshold; Traversing the surface geometry characteristic parameters of the target special-shaped curved glass, marking several surface areas with a curvature change rate higher than the curvature mutation threshold or a thickness gradient lower than the thin-wall thickness threshold or a boundary distance smaller than the boundary proximity threshold as key feature areas, and marking the remaining several areas as non-key feature areas, so as to obtain the feature-sensitive partition of the target special-shaped curved glass.

4. The method of claim 1, wherein: The virtual scanning mode parameters include laser spot size, power, frequency and scanning speed. The specific process of allocating virtual scanning mode parameters to each sub-area includes: comparing the virtual scanning layer parameters of each sub-area with preset thresholds respectively, if the virtual scanning layer parameter of a certain sub-area is greater than or equal to the preset threshold, assigning the sub-area with fine cutting scanning mode, otherwise assigning the sub-area with efficient scanning mode.

5. The method of claim 1, wherein: The specific process of generating the spatial vector compensation field includes: Establishing a basic thermal deformation prediction model with local geometric features as input; Building a historical thermal coupling influence model containing a heat conduction equation; Inputting the time-series thermal history parameters into the historical thermal coupling influence model to output the historical thermal coupling deformation; Superimposing the basic thermal deformation prediction value and the historical thermal coupling deformation to generate the total predicted deformation; Building a spatial vector compensation field according to the total predicted deformation.

6. The method of claim 5, wherein: The specific process of generating the actual laser cutting trajectory of the target special-shaped curved glass includes: Vector superimposing the spatial vector compensation field and the theoretical coordinate points of the theoretical cutting path stored in the database to verify whether the offset path points are located within the original feature-sensitive partition; If a certain path point after offset is not located within the original feature-sensitive partition, it is determined that the offset amount of the path point after offset is greater than the corresponding safety threshold, and a scanning mode parameter dynamic adjustment instruction is triggered; Re-generating the actual laser cutting trajectory based on the adjusted scanning mode parameters.

7. The method of claim 6, wherein: The specific content of the scanning mode parameter dynamic adjustment instruction includes: Identifying the specific area where the cross-border offset occurs, increasing the number of virtual layers of the local area that exceeds the boundary in the specific area; Switching the scanning mode of the local area that exceeds the boundary in the specific area to fine cutting scanning mode; Updating the boundary buffer zone parameters of the feature-sensitive partition; Feeding the adjusted number of virtual layers, the switched fine cutting scanning mode and the boundary buffer zone parameters back to the spatial vector compensation field generation step for iterative optimization. 8.The method of claim 6, wherein: The specific content of controlling the laser cutting equipment to perform cutting operation includes: Generating a laser focal point spatial position sequence according to the actual laser cutting trajectory; Converting the scanning mode parameters into a device control instruction set; Real-time monitoring the temperature distribution field of the cutting area, triggering a cooling intervention mechanism when the temperature exceeds the preset safety threshold and recording the actual thermal history parameters, while feeding them back to the time-series thermal history parameter library for updating; Collect the profile accuracy data of the curved surface after cutting is completed, and calculate the deviation distribution diagram of the actual profile and the theoretical model; Optimize the feature sensitive partition judgment threshold based on the deviation distribution diagram; Update the thermal deformation prediction model parameters in the space vector compensation field.

Images