A method for processing microstructure by high-speed laser cutting

Abstract

The application provides a method for realizing micro-structure machining through high-speed laser cutting, which comprises the following steps: constructing a material database in advance, dynamically adjusting a scanning speed and laser power by combining an adaptive algorithm to correct an energy distribution model, and fusing a hardness value to iteratively optimize a depth parameter to ensure machining precision; meanwhile, aiming at the stability fluctuation under high-speed scanning, a deviation correction vector is generated by a dynamic matching scheme and a real-time monitoring module to update a scanning path planning, and finally a stable machining path is output. The method significantly improves the precision and stability of micro-structure machining of materials, especially in the high-speed machining scene, through feedback control and data cycle verification, intelligent optimization of the machining path is realized, and efficient and accurate laser machining effect is exhibited.

Description

Technical Field

[0001] This invention belongs to the field of laser processing technology and relates to a method for processing microstructures by high-speed laser cutting. Background Technology

[0002] Microstructure fabrication technology, which alters the microscopic morphology of material surfaces, is an indispensable process in high-end equipment manufacturing. However, current processing methods struggle to balance precision and efficiency, especially when dealing with diverse materials and complex surface morphologies. Insufficient process adaptability often leads to unstable processing results. A prominent challenge in microstructure fabrication lies in controlling the laser scanning speed. High-speed laser movement significantly affects the uniformity of energy distribution, making precise control of surface texture depth extremely difficult. This speed-depth matching problem also triggers another chain reaction: the stability of laser action on different material surfaces fluctuates due to variations in material properties. On harder materials, excessively fast scanning speeds may result in insufficient texture depth, while on softer materials, overcutting may occur.

[0003] Therefore, how to achieve precise matching between scanning speed and texture depth during high-speed laser cutting, and ensure processing stability on various material surfaces, has become a key issue that urgently needs to be addressed. Summary of the Invention

[0004] This invention proposes a method for processing microstructures through high-speed laser cutting, which solves the problems of uneven energy distribution, texture depth deviation and path stability fluctuation caused by the complexity of matching material hardness, thermal conductivity and laser parameters in laser processing.

[0005] In one aspect, this invention provides a method for fabricating microstructures using high-speed laser cutting, comprising:

[0006] By using a pre-established material database, the hardness and thermal conductivity values ​​of the target material are obtained from the database, and the correspondence between these values ​​and laser parameters is processed to obtain the scanning speed range;

[0007] Based on the obtained scanning speed range, real-time laser energy distribution data is acquired. If the data exceeds a preset uniformity threshold, the laser power output is adjusted through a feedback control algorithm to obtain a corrected energy distribution model.

[0008] Texture depth prediction values ​​are extracted from the modified energy distribution model. Considering the differences in the surfaces of various materials, it is determined whether the prediction values ​​conform to the target depth range. If they do not conform, the information processing stage is used to integrate hardness values ​​for iterative calculation to obtain optimized depth parameters.

[0009] After obtaining the optimized depth parameters, the stability fluctuation under high-speed scanning is determined by the correlation analysis between the parameters and the surface texture. If the fluctuation is higher than the threshold, the short-term adjustment requirements of the integrated device are used to obtain a dynamic matching scheme.

[0010] A velocity-depth correspondence table is extracted from the dynamic matching scheme, and a real-time monitoring module is used to process the deviation between the correspondence table and the actual processing data to obtain a deviation correction vector.

[0011] Based on the deviation correction vector, the response change of the material surface is judged. If the response change exceeds the preset range, the scanning path planning is updated to obtain the final processing control sequence.

[0012] In another aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, characterized in that the computer program, when executed by a processor, implements the method described thereon.

[0013] In another aspect, the present invention also provides a computer program product, comprising a computer program, characterized in that the computer program, when executed by a processor, implements the method described herein.

[0014] Compared with existing technologies, this invention pre-constructs a material database, dynamically adjusts the scanning speed and laser power using an adaptive algorithm to correct the energy distribution model, and iteratively optimizes depth parameters using hardness values ​​to ensure processing accuracy. Simultaneously, to address stability fluctuations under high-speed scanning, a deviation correction vector is generated through a dynamic matching scheme and a real-time monitoring module to update the scanning path planning, ultimately outputting a stable processing path. This invention significantly improves the accuracy and stability of processing microstructures, especially in high-speed processing scenarios. Through feedback control and data loop verification, it achieves intelligent optimization of the processing path, demonstrating highly efficient and precise laser processing results. Attached Figure Description

[0015] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 This is a flowchart illustrating the method of the present invention. Detailed Implementation

[0017] The technical solution of the present invention will be clearly and completely described below with reference to the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0018] In one aspect, this invention proposes a method for fabricating microstructures using high-speed laser cutting, comprising the following steps:

[0019] S101, using a pre-established material database, obtain the hardness and thermal conductivity values ​​of the target material from the database, process the correspondence between these values ​​and laser parameters, and obtain the scanning speed range.

[0020] First, the hardness and thermal conductivity values ​​of the target material are retrieved from a pre-established material database. For example, for a certain aluminum alloy, the hardness value is 150 HV and the thermal conductivity is 237 W / m·K. A query mechanism is used to extract these hardness and thermal conductivity values. This query mechanism is a database retrieval system based on SQL statements. By inputting the material name, such as "aluminum alloy," as a keyword, the system automatically matches and extracts the corresponding attribute values, thus obtaining a material attribute dataset. When constructing the query statement, the SELECT command is used to specify the hardness and thermal conductivity value fields, and a WHERE condition is added to filter the target material, thereby avoiding interference from irrelevant data. In the result parsing, the raw data returned by the query is converted into a structured format, such as a JSON object, containing the hardness value, thermal conductivity value, and their unit information. This method can quickly respond to query needs for different alloy materials and supports batch extraction to adapt to large-scale production.

[0021] An adaptive algorithm, based on a machine learning-based neural network model, is input into a material property dataset. This algorithm establishes a correspondence between laser power and focal diameter, using hardness and thermal conductivity as input features. The algorithm calculates the mapping function between power and diameter using a multilayer perceptron and determines whether the correspondence satisfies a melting temperature threshold. If so, a parameter matching model is obtained. Specifically, this process involves threshold comparison and iterative optimization: for example, defining the melting temperature threshold as the melting point of aluminum alloys; then, the algorithm simulates the heat distribution under laser irradiation, using thermal conductivity to calculate the temperature field. If the calculated temperature exceeds the threshold, the power is adjusted from 500W to 400W, and the focal diameter is correspondingly modified from 0.1mm to 0.2mm until the relationship satisfies the threshold condition; finally, a parameter matching model is formed, storing the optimized power-diameter pairs in a relational data table, thus achieving adaptive parameter configuration for different materials; the relational data in the parameter matching model is obtained, and... The processing accuracy control attributes are adaptively adjusted. The adjustment process is achieved through a weighted fusion method: extract the power-diameter relationship data from the model, such as a power of 400W corresponding to a diameter of 0.2mm; then, introduce the accuracy attribute as a weighting factor to calculate the speed range. For example, based on empirical formulas, estimate the initial speed lower limit as 10mm / s and the upper limit as 50mm / s, and fine-tune it to 15-40mm / s according to accuracy requirements to form the calculation basis. Extract the optimization results from the speed range calculation basis, such as selecting the optimal sub-range, such as 20-30mm / s, through a sorting algorithm to obtain the scanning speed range.

[0022] S102, based on the obtained scanning speed range, acquire real-time laser energy distribution data. If the data exceeds a preset uniformity threshold, adjust the laser power output through a feedback control algorithm to obtain a corrected energy distribution model.

[0023] Based on a scanning speed range of 20-30 mm / s, the energy density distribution and focal offset are monitored by an optical sensor to determine whether the data exceeds a preset uniformity threshold. This threshold is defined as the standard deviation of the energy distribution being less than 5%. The judgment process involves two sub-steps: data acquisition and threshold comparison. In the data acquisition stage, a high-speed camera is used to capture a two-dimensional energy map of the laser beam, and the focal position coordinates are recorded simultaneously. In the threshold comparison, the variance of the energy distribution is calculated. If the variance exceeds the threshold of 5.2%, it is determined to be non-uniform, thus obtaining a distribution deviation index. This index quantifies the degree of deviation in percentage form, thereby providing a quantitative basis for subsequent adjustments.

[0024] Next, the distribution deviation index is processed using a feedback control algorithm. This algorithm adjusts the laser power output and focus position calibration based on a closed-loop PID control system to determine the power correction value. First, the deviation index is input as an error signal. Then, the correction amount is calculated using proportional, integral, and derivative terms. The proportional term amplifies the error for rapid response, the integral term accumulates historical errors to eliminate steady-state deviation, and the derivative term predicts trends to suppress overshoot. Based on this, the power is adjusted from the initial 400W to 380W, and the focus position offset is calibrated by 0.05mm, ultimately determining a power correction value of -20W, thus achieving dynamic stability. Subsequently, optimization parameters such as the corrected power and position values ​​are extracted from the power correction value. A correspondence is established with the material property dataset to obtain the energy distribution correction framework, describing the mapping relationship between power and energy distribution. The correlation between hardness and thermal conductivity values ​​in the dataset and the correction parameters is fitted using linear regression to form a correction function, thereby optimizing the distribution uniformity. Adaptive adjustments are made to the energy distribution correction framework by incorporating a parameter matching model, which is the previously established power-diameter correspondence model. The corrected energy distribution model is determined, and the framework parameters are updated through iterative fusion to achieve adaptive optimization.

[0025] S103: Extract the predicted texture depth value from the modified energy distribution model. Considering the differences in the surfaces of various materials, determine whether the predicted value conforms to the target depth range. If not, use the information processing step to integrate the hardness value for iterative calculation to obtain the optimized depth parameter.

[0026] Texture depth prediction values ​​are extracted from the modified energy distribution model. The texture depth of the processed surface is predicted by analyzing the energy gradient data in the model, for example, 2.5 mm. At the same time, surface difference features such as the distribution of micro-bumps are obtained, and a surface roughness distribution map is generated to visualize the roughness changes. Based on the surface roughness distribution map, it is determined whether the texture depth prediction value conforms to the target depth range, which is preset to 1.8-3.0 mm. If the prediction value, such as 2.7 mm, is within the range, it is considered to conform. Otherwise, the depth deviation is extracted. For example, the deviation is calculated as the difference between the prediction value and the midpoint of the range, such as 0.3 mm. This judgment process involves image processing technology, where the surface roughness distribution map is generated from three-dimensional surface data acquired by a laser scanner. Specifically, the acquired data is first filtered to remove noise, and then Fourier transform is applied to analyze the frequency components to quantify the roughness index Ra value. This value is then compared with the target depth range. If they do not match, the depth deviation is quantified using a difference formula. Furthermore, hardness values, such as material hardness HV300, are integrated to determine the laser penetration attenuation model. This model describes the attenuation law of laser energy inside the material. The model is based on an exponential decay function. After inputting the deviation and hardness value, the penetration attenuation coefficient is calculated to simulate the energy loss process.

[0027] When processing the depth deviation, it is first standardized as a relative deviation percentage, such as 15%. Then, the hardness value is integrated using a weighted average method to form the parameter basis of the attenuation model. For example, the higher the hardness value, the larger the attenuation coefficient, reflecting the material's absorption characteristics of laser. For the laser penetration attenuation model, a solidification rate calculation is introduced. This calculation considers the cooling rate after the material melts, such as 5 degrees / s, to obtain a depth compensation matrix. This matrix is ​​a two-dimensional array representing the compensation value at different depths. Values ​​are extracted from it, such as a compensation of 0.2 mm, to obtain optimized depth parameters, thereby improving processing accuracy.

[0028] S104 After obtaining the optimized depth parameters, the stability fluctuation level under high-speed scanning is determined by the correlation analysis between the parameters and the surface texture. If the fluctuation level is higher than the threshold, the short-term adjustment requirements of the integrated device are determined to determine the dynamic matching scheme.

[0029] The optimized depth value is obtained from the previously calculated depth parameters. The optimized depth of 1.2 mm is obtained by analyzing the output data of the energy model. Simultaneously, surface texture features, such as microscopic peak distribution, are collected using optical microscopy to quantify surface irregularities. After the above extraction, a correlation analysis is performed between the optimized depth value and the surface texture features to obtain a texture-depth mapping matrix. Correlation analysis establishes a correspondence between depth values ​​and texture features through correlation calculations. Specifically, the optimized depth value is first standardized into a vector (e.g., converting 1.2 mm into a relative depth ratio). Then, the surface texture features are decomposed into multiple dimensional indicators, such as peak height and spacing. These indicators are multiplied by the depth vector using matrix operations to form a multidimensional texture-depth mapping matrix. This matrix is ​​essentially an array representing the mapping between depth and texture. For example, in the analysis, the rows of the matrix correspond to different depth layers, and the columns correspond to texture feature values, thus revealing how depth affects texture distribution.

[0030] The scan trajectory offset is calculated based on the texture depth mapping matrix, and the degree of stability fluctuation is determined by the scan trajectory offset. Specifically, calculating the offset involves extracting gradient information from the matrix, such as taking the derivative of the matrix elements to obtain the trajectory deviation, for example, an offset of 0.15mm. Then, the degree of stability fluctuation is evaluated based on the offset. The degree of fluctuation is quantified by statistically analyzing the variance of the offset. For example, a variance exceeding a preset value indicates large fluctuation. The entire process ensures the continuity of trajectory adjustment. If the degree of stability fluctuation is higher than the fluctuation threshold, a short-term adjustment is extracted. This adjustment is obtained by selecting the peak portion from the fluctuation data, such as adjusting by 0.1mm, for temporary correction. A dynamic matching sequence is determined based on the short-term adjustment, that is, the adjustment is serialized to form a dynamic path matching the laser scan.

[0031] S105, extract the speed-depth correspondence table from the dynamic matching scheme, and use the real-time monitoring module to process the deviation between the correspondence table and the actual processing data to obtain the deviation correction vector.

[0032] Real-time processing data during laser scanning is acquired, and feature extraction is performed on the real-time processing data to obtain the geometric dimensions of the molten pool. Image processing algorithms are used to extract the width and depth indicators of the molten pool from the data. The molten pool boundary is identified using edge detection methods, and the width and depth are calculated. These dimensions are quantified using pixel conversion formulas to reflect real-time changes in the molten pool morphology. The current energy density distribution value is calculated based on the molten pool geometric dimensions, and the energy distribution difference value is obtained by comparing the current energy density distribution value with a velocity-depth correspondence table. Specifically, the energy density distribution value is calculated by inputting the molten pool dimensions into an energy model. For example, energy density E = power / (velocity × width). In the example, if the power is 1000W, the velocity is 500mm / s, and the width is 2.5mm, then E is approximately 0.8J / mm². Then, it is compared with a preset velocity-depth correspondence table. This table maps the energy standards of velocity and depth based on historical data. For example, the standard E at the corresponding velocity in the table is 0.7J / mm², and the difference value is 0.1J / mm². This comparison reveals the degree of energy deviation. The correspondence table can be generated through experimental calibration, covering various materials, to ensure that the difference value captures processing inhomogeneities.

[0033] The expansion gradient of the heat-affected zone is extracted by the energy distribution difference value, and the expansion gradient is classified to obtain a thermal accumulation state label. Specifically, extracting the expansion gradient involves calculating the thermal diffusivity from the difference value. For example, gradient G = difference value / time interval. In the example, if the difference value is 0.1 J / mm² and the interval is 0.1 s, then G is 1.0 J / mm²·s. Then, the classification process uses a threshold model to divide G into low, medium, and high labels. For example, if G > 0.8, it is marked as a high accumulation state. This label is obtained by training a machine learning classifier to adapt to different heat-affected zone scenarios, thereby quantifying the thermal accumulation risk. The system determines whether the heat accumulation status label is in an overheated state. If the heat accumulation status label is in an overheated state, the spot overlap rate adjustment amplitude is calculated based on the heat accumulation status label. A deviation correction vector is generated through the spot overlap rate adjustment amplitude. Specifically, the overheating judgment is based on the label threshold, such as a high accumulation label indicating overheating. Then, the adjustment amplitude A is calculated as label intensity × coefficient. For example, if the intensity is 1.2 and the coefficient is 0.5, then A is 0.6%. The generated vector is mapped to the scanning path for correction. For example, the vector [0.6, -0.3] represents the adjustment in the x / y direction. This process ensures processing stability.

[0034] S106. Based on the deviation correction vector, determine the response change of the material surface. If the response change exceeds the preset range, update the scanning path planning to obtain the final processing control sequence.

[0035] First, a deviation correction vector is obtained. This vector, usually derived from the aforementioned output, represents the deviation adjustment in the laser scanning path. By integrating sensor data, a vector representing the offset in the x and y directions is formed. The latent heat of phase change is obtained by analytically calculating the deviation correction vector. This analytical calculation involves decomposing the vector into heat transfer components and using a thermodynamic model to evaluate the energy absorption during the material's phase transition. For example, during laser melting, the vector offset is input into the latent heat estimation formula, considering the material's transition from solid to liquid. The process includes analyzing how the vector components affect the local temperature gradient, first extracting the vector's magnitude as a heat input index, and then deriving the latent heat value by combining the material's heat capacity parameters. This calculation ensures the capture of phase transition dynamics and avoids direct numerical calculations, estimating energy changes through logical reasoning.

[0036] The response changes on the material surface are generated based on the latent heat of phase transition. This generation process is achieved by simulating a surface thermal response model. In the laser processing scenario, the latent heat is mapped to changes in surface tension, and the response changes are manifested as surface ripples or oxide layer formation. Following the aforementioned calculations, the analysis further examines how latent heat induces adjustments in the surface microstructure, thus forming a continuous response description. It is determined whether the response changes exceed a preset range. For example, if the surface ripple amplitude exceeds a threshold, it is considered excessive. If the response changes exceed the preset range, the response changes are processed to obtain a stress release rate. This processing includes applying a stress relaxation algorithm, calculating the release rate through the gradient of the response changes, and considering the influence of the material's elastic modulus. Based on the stress release rate, a state estimation is performed to obtain the trajectory deflection angle. This estimation uses a kinematics model, inputting the release rate into the deflection prediction to obtain the angle value. The trajectory deflection angle is used to update the scanning path planning, and the scanning path planning is analyzed to obtain the final processing control sequence.

[0037] In another aspect, the present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-described method.

[0038] In another aspect, the present invention also proposes a computer program product, comprising a computer program, characterized in that the computer program implements the above-described method when executed by a processor.

[0039] In particular, according to some embodiments of this disclosure, the processes described above can be implemented as computer software programs. For example, some embodiments of this disclosure include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowchart. In such embodiments, the computer program can be downloaded and installed from a network via a communication device, or installed from a storage device, or installed from a ROM. When the computer program is executed by a processing device, it performs the functions defined above in the methods of some embodiments of this disclosure.

[0040] It should be noted that, in some embodiments of this disclosure, the computer-readable medium may be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. A computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In some embodiments of this disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In some embodiments of this disclosure, a computer-readable signal medium may include a task data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated task data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium can be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to: wires, optical fibers, RF (radio frequency), etc., or any suitable combination thereof.

[0041] In some implementations, clients and servers can communicate using any currently known or future-developed network protocol, such as HTTP (Hypertext Transfer Protocol), and can interconnect with digital task data communication (e.g., communication networks) of any form or medium. Examples of communication networks include local area networks (“LANs”), wide area networks (“WANs”), the Internet (e.g., the Internet), and end-to-end networks (e.g., ad hoc end-to-end networks), as well as any currently known or future-developed networks.

[0042] The aforementioned computer-readable medium may be included in the aforementioned electronic device; or it may exist independently and not assembled into the electronic device. The aforementioned computer-readable medium carries one or more programs that, when executed by the electronic device, cause the electronic device to: determine the network connection status of the switch production line management application in response to detecting a query operation on a production collaboration document in the switch production line management application; replace the webpage entry information corresponding to the production collaboration document with target entry file information and load target webpage resource information in response to determining that the network connection status of the switch production line management application indicates an offline state, so as to display the webpage of the production collaboration document offline in the switch production line management application, wherein the target entry file information is the file information of the entry file corresponding to the webpage of the production collaboration document downloaded in advance, and the target webpage resource information is the resource information corresponding to the webpage stored locally; in response to determining that the network connection status of the switch production line management application indicates an online state and that the webpage resource information corresponding to the production collaboration document is not stored locally, download the webpage resource information of the webpage from the production line document server, wherein the webpage resource information includes an entry file and resource information; display the webpage of the production collaboration document in the switch production line management application according to the webpage resource information, and store the webpage resource information in a local database.

[0043] Computer program code for performing operations of some embodiments of this disclosure can be written in one or more programming languages ​​or a combination thereof, including product-oriented programming languages ​​such as Java, Smalltalk, and C++, and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0044] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for fabricating microstructures using high-speed laser cutting, characterized in that, include: By using a pre-established material database, the hardness and thermal conductivity values ​​of the target material are obtained from the database, and the correspondence between these values ​​and laser parameters is processed to obtain the scanning speed range; Based on the obtained scanning speed range, real-time laser energy distribution data is acquired. If the data exceeds a preset uniformity threshold, the laser power output is adjusted through a feedback control algorithm to obtain a corrected energy distribution model. Texture depth prediction values ​​are extracted from the modified energy distribution model. Considering the differences in the surfaces of various materials, it is determined whether the prediction values ​​conform to the target depth range. If they do not conform, the information processing stage is used to integrate hardness values ​​for iterative calculation to obtain optimized depth parameters. After obtaining the optimized depth parameters, the stability fluctuation under high-speed scanning is determined by the correlation analysis between the parameters and the surface texture. If the fluctuation is higher than the threshold, the short-term adjustment requirements of the integrated device are used to obtain a dynamic matching scheme. A velocity-depth correspondence table is extracted from the dynamic matching scheme, and a real-time monitoring module is used to process the deviation between the correspondence table and the actual processing data to obtain a deviation correction vector. Based on the deviation correction vector, the response change of the material surface is judged. If the response change exceeds the preset range, the scanning path planning is updated to obtain the final processing control sequence.

2. The method as described in claim 1, characterized in that, The specific steps for obtaining the scan speed range include: The hardness and thermal conductivity values ​​of the target material are obtained from a pre-established material database. A query mechanism is used to process the extraction of the hardness and thermal conductivity values ​​to obtain a material property dataset. The material property dataset is used to establish a correspondence between laser power value and focal diameter value. It is then determined whether the correspondence meets the melting temperature threshold. If it does, a parameter matching model is obtained. The relationship data in the parameter matching model is obtained, and the scanning speed range is obtained by adapting and adjusting it in combination with the processing accuracy control attributes.

3. The method as described in claim 1, characterized in that, The specific steps to obtain the corrected energy distribution model include: Based on the scanning speed range, real-time laser energy distribution data and focal position data are obtained, and it is determined whether the data exceeds the preset uniformity threshold to obtain the distribution deviation index. The distribution deviation index is processed by a feedback control algorithm to adjust the laser power output and focus position calibration, thereby obtaining a power correction value. Optimization parameters are extracted from the power correction values, and corresponding relationships are established with the material property dataset to obtain the energy distribution correction framework. The energy distribution correction framework is then adapted and adjusted, incorporated into the parameter matching model, and the corrected energy distribution model is determined.

4. The method as described in claim 1, characterized in that, The specific steps to obtain the optimized depth parameters include: The texture depth prediction value is extracted based on the modified energy distribution model, and the surface roughness distribution map is obtained by acquiring surface difference features. The texture depth prediction value is then judged to be consistent with the target depth range based on the surface roughness distribution map. If the texture depth prediction value is inconsistent, the depth deviation is extracted. The depth deviation is processed and the hardness value is fused to determine the laser penetration attenuation model. A solidification rate is introduced to calculate the depth compensation matrix for the laser penetration attenuation model. The optimized depth parameters are obtained by extracting the values ​​from the depth compensation matrix.

5. The method as described in claim 1, characterized in that, The specific steps to obtain a dynamic matching scheme include: A texture depth mapping matrix is ​​obtained by performing correlation analysis between optimized depth parameters and surface texture features; The scan trajectory offset is calculated based on the texture depth mapping matrix, and the degree of stability fluctuation is determined by the scan trajectory offset; if the degree of stability fluctuation is higher than the fluctuation threshold, the short-term adjustment amount is extracted to obtain the dynamic matching sequence.

6. The method as described in claim 1, characterized in that, The specific steps to obtain the deviation correction vector include: Real-time processing data during laser scanning is acquired, and feature extraction is performed on the real-time processing data to obtain the geometric dimensions of the molten pool; Calculate the current energy density distribution value based on the geometry of the molten pool, and obtain the energy distribution difference value by comparing the current energy density distribution value with the velocity-depth correspondence table. The thermally affected zone expansion gradient is extracted by the energy distribution difference value, and the thermally affected zone expansion gradient is classified to obtain a thermal accumulation state label. If the thermal accumulation state label is an overheated state, the spot overlap rate adjustment range is calculated based on the thermal accumulation state label, and a deviation correction vector is generated by the spot overlap rate adjustment range.

7. The method as described in claim 1, characterized in that, The specific steps to obtain the final processing control sequence include: Obtain the deviation correction vector, perform analytical calculation on the deviation correction vector to obtain the latent heat value of phase change, and generate the response change of the material surface based on the latent heat value of phase change; If the response change exceeds the preset range, the response change is processed to obtain the stress release rate, and the trajectory deflection angle is obtained by state estimation based on the stress release rate. The scanning path plan is updated by the trajectory deflection angle, and the scanning path plan is parsed to obtain the final processing control sequence.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the method as described in any one of claims 1 to 7.

9. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the method as described in any one of claims 1 to 7.

Images