A laser welding process and apparatus for battery casing

Abstract

The present invention relates to the field of welding technology, specifically to a battery shell laser welding process method and device, including the steps of: performing laser fine-tuning processing on the joint surface of the battery shell and obtaining image data; obtaining a lattice structure data set of a microscopic lattice, adjusting the arrangement direction and grain boundary position of the microscopic lattice to optimize the lattice structure data set, and obtaining an optimized grain distribution data set; identifying and processing the microscopic morphology data in the surface state of the joint surface to obtain surface roughness data and perform thermal stress simulation analysis, performing regional division processing on the thermal stress distribution to obtain a thermal stress gradient data set, and analyzing the thermal conductivity difference of the joint surface based on the thermal stress gradient data set to obtain a heat transfer data set; combining the heat transfer data set with the grain distribution data set to obtain the optimal laser welding parameters and control the welding robot to perform welding. The present application reduces welding defects caused by uneven thermal stress during the welding of the battery shell, and improves the consistency of welding quality.

Description

Technical Field

[0001] The present invention relates to the field of welding technology, and in particular to a battery shell laser welding process method and device. Background Art

[0002] With the development of electric vehicles and renewable energy, battery performance and safety have become key research directions. In particular, the production and assembly process of battery casings directly affects the overall performance, stability and service life of batteries.

[0003] The existing battery shell welding process usually uses laser welding and other methods to connect the various parts of the battery shell. During the laser welding process, the welding quality is affected by many factors, especially the selection of welding parameters and the characteristics of the welding joint surface. The surface state of the welding joint surface is a key factor affecting the welding quality. The existing technology generally improves the welding process through numerical simulation, experimental testing and other means, and often does not accurately consider the surface details of the joint surface, thereby affecting the distribution of thermal stress and heat conduction efficiency during welding, resulting in uneven quality of the welded joint, and even cracks, deformation and other problems. Therefore, how to systematically optimize the joint surface characteristics and select appropriate laser welding parameters based on these optimization data is a difficult problem in the current welding process research. Summary of the invention

[0004] The main purpose of the present invention is to provide a battery casing laser welding process method and device, aiming to overcome the technical problem that the existing welding technology does not accurately consider the surface details of the joint surface, thereby affecting the welding effect.

[0005] In order to achieve the above-mentioned invention problem, the present invention proposes a battery shell laser welding process method, the method comprising:

[0006] Performing laser fine-tuning processing on the joint surface of the battery shell to obtain image data of the microscopic lattice of the joint surface after the laser fine-tuning processing;

[0007] Performing diffraction analysis on the image data to obtain a lattice structure data set of the microscopic lattice, wherein the lattice structure data set at least includes arrangement information of crystal grains in a bonding surface and positions of grain boundaries;

[0008] Adjusting the arrangement direction and grain boundary position of the microscopic lattice to optimize the lattice structure data set to obtain an optimized grain distribution data set;

[0009] Acquiring the surface state of the joint surface, identifying and processing the microscopic morphology data in the surface state of the joint surface to obtain surface roughness data, performing thermal stress simulation analysis on the surface roughness data, and obtaining the thermal stress distribution of the joint surface under different welding parameters;

[0010] Performing regional division processing on the thermal stress distribution to obtain a thermal stress gradient data set of each region, and analyzing the thermal conductivity difference of the joint surface based on the thermal stress gradient data set to obtain a heat transfer data set;

[0011] The heat transfer data set is combined with the grain distribution data set to obtain optimal laser welding parameters, and a welding robot is controlled to weld the joint surface of the battery casing according to the optimal laser welding parameters.

[0012] Furthermore, the step of performing laser fine-tuning on the joint surface of the battery casing and obtaining image data of the microscopic lattice of the joint surface after the laser fine-tuning includes:

[0013] Locally irradiating the joint surface with a high-frequency laser pulse to obtain the spot size and energy distribution of the laser beam irradiated on the joint surface;

[0014] Calculating the surface flatness of the laser beam based on the spot size and energy distribution, and adjusting the focal length and pulse frequency of the laser beam based on the surface flatness;

[0015] Collecting a microscopic lattice image of the bonding surface after laser fine-tuning processing, and preprocessing the microscopic lattice image through an adaptive filtering algorithm to obtain an optimized scanning image;

[0016] The boundary positions of the grains in the optimized scanned image are extracted based on an edge detection algorithm, and the grains are identified and calibrated based on the boundary positions to obtain image data of the microscopic lattice of the bonding surface.

[0017] Further, the step of performing diffraction analysis on the image data to obtain a lattice structure data set of the microscopic lattice, wherein the lattice structure data set at least includes arrangement information of grains in the joint surface and positions of grain boundaries, comprises:

[0018] Acquire the angular distribution information and periodic characteristics of the microscopic lattice in the image data based on the diffraction technology to obtain a microscopic lattice distribution map;

[0019] Performing Fourier transform processing on the microscopic lattice distribution map to obtain frequency domain image data;

[0020] Acquire the intensity distribution of different frequency components in the frequency domain image data, identify and calculate the main orientation direction of the grains, and form grain arrangement direction data;

[0021] Performing grain boundary identification processing on the microscopic lattice distribution map to obtain grain boundary position data;

[0022] The grain arrangement direction data and the grain boundary position data are combined based on an image fusion algorithm to form the lattice structure data set.

[0023] Furthermore, the step of adjusting the arrangement direction and grain boundary position of the microscopic lattice to optimize the lattice structure data set to obtain an optimized grain distribution data set includes:

[0024] Extracting the distribution characteristics of each grain in the arrangement direction of the microscopic lattice to obtain a grain distribution characteristic data set;

[0025] Acquiring the relative orientation relationship between the grains in the grain distribution characteristic data set, adjusting the arrangement direction of the grains based on the relative orientation relationship, and obtaining a grain orientation optimization data set;

[0026] Acquire the mutual relationship between the crystal grains and the grain boundaries in the lattice structure data set, and adjust the grain boundary position based on the mutual relationship to obtain an optimized grain boundary position data set;

[0027] The relative angle value between each of the grains and the adjacent grains in the grain direction optimization data set is calculated, and the relative position between the grains is optimized based on the relative angle value and the grain boundary position data set to obtain an optimized grain distribution data set.

[0028] Further, the step of obtaining the surface state of the joint surface, identifying and processing the microscopic morphology data in the surface state of the joint surface to obtain surface roughness data, performing thermal stress simulation analysis on the surface roughness data, and obtaining the thermal stress distribution of the joint surface under different welding parameters includes:

[0029] Performing three-dimensional imaging processing on the bonding surface by using a confocal microscope to obtain surface microscopic morphology data of the bonding surface, wherein the surface microscopic morphology data at least includes surface texture data and height difference data;

[0030] Calculating surface roughness data of the joint surface based on the surface texture data and the height difference data;

[0031] Calculating and extracting statistical characteristic data of the surface roughness data, obtaining the distribution of the surface roughness, and forming statistical data of the surface roughness based on the distribution;

[0032] According to the surface roughness characteristic data, thermal stress simulation is performed on the joint surface by finite element analysis to obtain thermal stress distribution data;

[0033] Based on the thermal stress simulation results, the back propagation algorithm is used to optimize the thermal stress distribution data to obtain the optimized thermal stress data;

[0034] According to the optimized thermal stress data, the influence of different welding parameters on the thermal stress distribution is analyzed to obtain the thermal stress distribution of the joint surface under different welding parameters.

[0035] Furthermore, the step of dividing the thermal stress distribution into regions to obtain a thermal stress gradient data set of each region, and analyzing the thermal conductivity difference of the joint surface based on the thermal stress gradient data set to obtain a heat transfer data set includes:

[0036] Performing spatial segmentation processing on the thermal stress distribution data based on a threshold segmentation algorithm to obtain a plurality of thermal stress regions;

[0037] Calculating the thermal stress difference between each thermal stress area and an adjacent position to obtain a thermal stress gradient data set for each thermal stress area;

[0038] interpolating the thermal stress gradient data sets between the different thermal stress regions to generate a smooth and continuous thermal stress gradient distribution data set;

[0039] constructing a relationship model between thermal stress and thermal conductivity according to the thermal stress gradient data set, and generating a thermal conductivity difference data set based on the relationship model;

[0040] The thermal conductivity difference data set is substituted into the heat conduction equation for simulation calculation to obtain the heat transfer data set of each area of ​​the joint surface.

[0041] Furthermore, the step of combining the heat transfer data set with the grain distribution data set to obtain optimal laser welding parameters, and controlling a welding robot to weld the joint surface of the battery casing according to the optimal laser welding parameters comprises:

[0042] Standardizing the thermal conductivity data in the heat transfer data set and the grain arrangement information in the grain distribution data set to obtain a standardized heat transfer data set and a standardized grain distribution data set;

[0043] Based on a weighted algorithm, the standardized heat transfer data set and the grain distribution data set are weightedly fused to obtain a fused comprehensive data set;

[0044] Constructing a multidimensional search space for laser welding parameters, and calculating, within the multidimensional search space, the influence value of each set of the laser welding parameters on the joint surface according to the comprehensive data set based on a particle swarm optimization algorithm;

[0045] The influence values ​​are sorted and an optimal laser welding parameter combination is screened out, and a welding robot is controlled to weld the joint surface of the battery casing based on the optimal laser welding parameter combination.

[0046] The present invention also provides a battery housing laser welding device, comprising:

[0047] An adjustment module, used to perform laser fine-tuning processing on the joint surface of the battery shell, and obtain image data of the microscopic lattice of the joint surface after the laser fine-tuning processing;

[0048] An analysis module, configured to perform diffraction analysis on the image data to obtain a lattice structure data set of the microscopic lattice, wherein the lattice structure data set at least includes arrangement information of grains in a bonding surface and positions of grain boundaries;

[0049] An optimization module, used for adjusting the arrangement direction and grain boundary position of the microscopic lattice to optimize the lattice structure data set to obtain an optimized grain distribution data set;

[0050] a processing module, for obtaining the surface state of the joint surface, identifying and processing the microscopic morphology data in the surface state of the joint surface to obtain surface roughness data, performing thermal stress simulation analysis on the surface roughness data, and obtaining the thermal stress distribution of the joint surface under different welding parameters;

[0051] A partitioning module, used to perform a region partitioning process on the thermal stress distribution to obtain a thermal stress gradient data set of each region, and to analyze the thermal conductivity difference of the joint surface based on the thermal stress gradient data set to obtain a heat transfer data set;

[0052] A control module is used to combine the heat transfer data set with the grain distribution data set to obtain optimal laser welding parameters, and control a welding robot to weld the joint surface of the battery casing according to the optimal laser welding parameters.

[0053] The present invention further provides a computer device, which includes a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above method when executing the computer program.

[0054] The present invention also provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the above method is implemented.

[0055] Compared with the prior art, this application has the following beneficial effects:

[0056] The present application proposes a battery shell laser welding process method and device, which obtains image data of the microscopic lattice by laser fine-tuning the joint surface of the battery shell, and then performs diffraction analysis on the lattice structure to obtain grain arrangement information and grain boundary positions, and adjusts the arrangement direction and grain boundary positions of the microscopic lattice, optimizes the grain distribution data set, and combines the surface roughness data of the joint surface to perform thermal stress simulation analysis to obtain the thermal stress distribution under different welding parameters during the welding process. By dividing the area of ​​the thermal stress distribution, a thermal stress gradient data set is obtained, and then based on the thermal conductivity difference analysis, a heat transfer data set is generated to comprehensively evaluate the thermal conduction efficiency during the welding process. Finally, by combining the heat transfer data set with the grain distribution data set, the optimal laser welding parameters are obtained, thereby ensuring the uniform distribution of thermal stress during welding and efficient heat conduction of the joint surface.

[0057] To sum up, by controlling welding parameters, the present application can optimize the quality of welding joints, reduce welding defects caused by uneven thermal stress, avoid the occurrence of cracks and deformation, and significantly improve the consistency and reliability of welding quality. It not only improves the strength and stability of welding joints, but also provides a strong guarantee for the overall performance, stability and service life of the battery. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0059] The structures, proportions, sizes, etc. illustrated in the drawings of this specification are only used to match the contents disclosed in the specification so as to facilitate understanding and reading by persons familiar with this technology. They are not used to limit the conditions under which this application can be implemented, and therefore have no substantive technical significance. Any structural modification, change in proportion or adjustment of size, without affecting the effects and purposes that can be achieved by this application, should still fall within the scope of the technical contents disclosed in this application.

[0060] Figure 1 This is a schematic diagram of the steps of a laser welding process for a battery housing in one embodiment of the present invention;

[0061] Figure 2 It is a schematic block diagram of the structure of a battery casing laser welding device according to one embodiment of the present invention;

[0062] Figure 3 is a schematic block diagram of the structure of a computer device according to an embodiment of the present invention;

[0063] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0064] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0065] It will be appreciated by those skilled in the art that, unless expressly stated, the singular forms "one", "above", and "the" used herein may also include plural forms. It should be further understood that the term "comprising" used in the specification of the present invention refers to the presence of features, integers, steps, operations, elements, modules, and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, modules, components, and / or groups thereof. It should be understood that when an element is said to be "connected" or "coupled" to another element, it may be directly connected or coupled to the other element, or there may be an intermediate element. In addition, the "connection" or "coupling" used herein may include wireless connection or wireless coupling. The term "and / or" used herein includes all or any module and all combinations of one or more associated listed items.

[0066] Those skilled in the art will understand that, unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as those generally understood by those skilled in the art in the field to which the present invention belongs. It should also be understood that terms such as those defined in common dictionaries should be understood to have meanings consistent with the meanings in the context of the prior art, and will not be interpreted with idealized or overly formal meanings unless specifically defined as here.

[0067] Reference Figure 1 The embodiment of the present invention provides a battery housing laser welding process method, comprising the following steps:

[0068] S1: performing laser fine-tuning processing on the joint surface of the battery shell to obtain image data of the microscopic lattice of the joint surface after the laser fine-tuning processing;

[0069] In step S1, the joint surface of the battery shell is laser trimmed, and the joint surface is locally irradiated with a controlled laser beam, so as to make detailed adjustments to the joint surface. When the laser is irradiated to the joint surface, the local temperature of the material surface will change, resulting in slight changes in the physical properties of the material. By adjusting the parameters such as the laser power, pulse frequency, and irradiation time, the laser beam can adjust the surface structure of the joint surface with extremely high precision, change the arrangement of the lattice of the joint surface through a slight thermal effect, and produce corresponding changes in the microscopic lattice structure. After the laser trimming process is completed, the joint surface will undergo a certain degree of microscopic changes, so the lattice structure of the joint surface will also change accordingly. The joint surface after laser trimming is scanned using high-resolution microscopy technology, such as scanning electron microscopy (SEM) or transmission electron microscopy (TEM), to capture the lattice arrangement and grain structure of the joint surface at the microscopic level, including the morphology, size, and position of grain boundaries of the grains. For example, in practical applications, assuming that the battery shell is made of aluminum alloy material, after the joint surface is laser trimmed, the image data of the microscopic lattice of the joint surface can be obtained by scanning electron microscopy. In this process, the electron beam scans the surface of the joint and interacts with the material to generate an image. The arrangement of the grains and the location of the grain boundaries can be clearly seen in the image. For example, the laser trimming process may cause the arrangement direction of the grains on the joint surface to change, the size of the grains may decrease, and the distribution of the grain boundaries may be adjusted, so these image data are acquired and analyzed.

[0070] S2: performing diffraction analysis on the image data to obtain a lattice structure data set of the microscopic lattice, wherein the lattice structure data set at least includes arrangement information of crystal grains in a bonding surface and positions of grain boundaries;

[0071] In step S2, the detailed structural information of the lattice of the joint surface is further extracted from the microscopic level by diffraction analysis technology. Specifically, diffraction analysis is a method of analyzing the internal structure of a material using wave properties. Through the diffraction phenomenon, accurate structural information of the material lattice can be obtained. In this embodiment, the diffraction analysis mainly uses the microscopic lattice image data of the joint surface, and uses X-ray diffraction (XRD), electron diffraction or other diffraction techniques to analyze the arrangement of the lattice and the distribution of grain boundaries. By irradiating the treated joint surface with rays or electron beams, a specific diffraction pattern is formed. The pattern reflects the arrangement of atoms inside the crystal, and information such as the size, shape, direction and position of the grain boundaries can be obtained from it. Specifically, in the experiment, the diffraction pattern on the surface of the material can be captured by X-ray diffraction (XRD) analysis technology. When the X-ray passes through the joint surface, it interacts with the lattice to generate a series of diffraction peaks. The intensity and angle of these diffraction peaks can reflect the arrangement of the lattice and the grain size. By measuring the diffraction angle and the diffraction peak intensity, combined with the diffraction theory, the crystallographic information of the grains, such as the lattice constant, grain size, grain morphology, etc., can be inverted. The lattice structure data set obtained by diffraction analysis can obtain the arrangement information of the grains in the joint surface and the position of the grain boundaries. The arrangement information of the grains describes the direction and distribution of the grains on the joint surface, which is expressed by parameters such as the orientation angle of the grains, the grain size, and the morphology of the grains. The position of the grain boundary indicates the interface position between the grains. It is worth noting that in this embodiment, the grains refer to the tiny areas with the same crystal structure and direction inside the material, and the grain boundaries are the boundaries between adjacent grains.

[0072] S3: adjusting the arrangement direction and grain boundary position of the microscopic lattice to optimize the lattice structure data set to obtain an optimized grain distribution data set;

[0073] In step S3, the arrangement direction and grain boundary position of the lattice are adjusted to optimize the distribution of the grains, so that the welding process is more stable and a higher quality welded joint is obtained. In actual operation, the arrangement direction and grain boundary position of the grains have a direct impact on the quality of the welded joint. The arrangement direction of the grains will affect the mechanical properties of the material, especially the performance of the material after being subjected to force. For example, if the arrangement direction of the grains is inconsistent, the stress in the welding area may be concentrated at the grain boundary, resulting in reduced joint strength. The grain boundary position determines the connection mode between the grains, and unstable grain boundaries may cause cracks or deformation in the welded joint. Specifically, adjusting the arrangement direction and grain boundary position of the grains can be achieved by controlling the laser welding parameters, optimizing the focal length of the laser, the power of the laser, the welding speed, etc. For example, by changing the scanning mode of the laser or adjusting the frequency of the laser pulse, the direction and arrangement of the grains on the joint surface can be changed. In addition, the control of the temperature field during the welding process will also affect the movement of the grain boundary and the growth of the grains. Appropriate temperature control can optimize the distribution of the grains, making the combination between the grains tighter, thereby improving the welding quality. By optimizing the grain distribution data set, the obtained optimized grain distribution data set will contain information such as the arrangement direction of grains in the joint surface, the size distribution of grains, the position of grain boundaries, etc.

[0074] S4: acquiring the surface state of the joint surface, identifying and processing the microscopic morphology data in the surface state of the joint surface to obtain surface roughness data, performing thermal stress simulation analysis on the surface roughness data, and obtaining the thermal stress distribution of the joint surface under different welding parameters;

[0075] In step S4, a high-precision scanning device (such as a scanning electron microscope, a laser confocal microscope, etc.) is used to scan the joint surface, and the microscopic morphology data of the joint surface is collected. In this way, tiny surface features on the joint surface, such as surface roughness, microcracks, particulate impurities, pores, etc., can be obtained, and they can be identified and processed to obtain surface roughness data. Surface roughness refers to the degree of fluctuation of the microstructure of the material surface, which is quantified by the change in the profile height in the surface morphology scanning result. For example, through mathematical analysis of the scanning data, parameters such as the peak-to-valley height, average roughness, and root mean square roughness of the surface profile can be extracted, and thermal stress simulation analysis can be performed based on these data. The purpose of thermal stress simulation analysis is to predict the thermal stress distribution of each area on the joint surface under different welding parameters. This analysis can be based on modeling the heat transfer process, such as using a finite element analysis method to simulate the thermal stress of the joint surface. During the simulation process, the surface roughness data will be input into the thermal model as one of the factors affecting heat conduction, thereby affecting the distribution of thermal stress. By comparing the thermal stress distribution under different welding parameters (such as laser power, welding speed, focal length, etc.), the thermal stress concentration area during the welding process can be evaluated, thereby avoiding cracks or deformation of the welded joint caused by excessive thermal stress.

[0076] S5: performing regional division processing on the thermal stress distribution to obtain a thermal stress gradient data set of each region, and analyzing the thermal conductivity difference of the joint surface based on the thermal stress gradient data set to obtain a heat transfer data set;

[0077] In step S5, the thermal stress distribution is processed by regional division. The purpose of regional division is to extract different thermal stress regions in the thermal stress distribution diagram and classify them according to the size and distribution characteristics of the thermal stress value. The thermal stress gradient refers to the rate at which the thermal stress in different regions changes with the spatial position during the welding process. The thermal stress values ​​in each divided region are analyzed to calculate the rate of thermal stress change between different regions. The height of the thermal stress gradient can directly reflect the efficiency of heat transfer during the welding process and the temperature change rate of different regions. For example, if the thermal stress gradient of a certain area is large, it means that the temperature in the area changes rapidly, and there may be a large thermal stress concentration, which can easily lead to thermal cracks or deformation on the joint surface. Based on the thermal stress gradient data set, the thermal conductivity difference of the joint surface is analyzed. During the laser welding process, different regions of the joint surface may show different thermal conductivities due to factors such as grain structure, surface roughness, and alloy composition. By analyzing the thermal stress gradient data set, the reasons for the change of thermal stress in different regions can be inferred, and the welding process can be further optimized in combination with the thermal conductivity difference of the material. Specifically, in areas with large thermal stress gradients, heat may not be effectively conducted, resulting in concentrated thermal stress in that area; while in areas with small thermal stress gradients, heat transfer is more uniform and thermal stress is more stable. By analyzing the difference in thermal conductivity, a heat transfer data set can be obtained, which reflects the effect of heat transfer between different areas during the welding process. The heat transfer data set can include information such as heat flux density, temperature distribution, and thermal diffusion rate of each area under different welding parameters.

[0078] S6: combining the heat transfer data set with the grain distribution data set to obtain optimal laser welding parameters, and controlling the welding robot to weld the joint surface of the battery casing according to the optimal laser welding parameters.

[0079] In step S6, by combining these two data sets, the welding process can be finely adjusted. For example, in the heat transfer data set, the temperature rises faster in some areas, which may lead to the concentration of thermal stress, while the grain distribution data set shows that the grains in this area are larger or the grain boundaries are more irregular, which may lead to poor thermal stability in this area during the welding process. After combining these two data sets, the heat treatment process in this area can be optimized by adjusting the welding parameters to avoid thermal cracks or excessive deformation. In actual operation, the combination can be achieved through a data model or an optimization algorithm. By analyzing the relationship between the heat transfer data and the grain distribution data, the welding quality of each area of ​​the joint surface under different welding parameters (such as laser power, welding speed, etc.) can be determined. For example, if the heat transfer efficiency of a certain area is low (i.e., the thermal conductivity is poor), and the grain distribution in this area is relatively uneven, the laser power or welding speed of this area can be adjusted to reduce overheating, thereby avoiding welding defects. On the contrary, if the thermal conductivity of a certain area is high and the grain distribution is uniform, the laser welding parameters can be slightly increased, thereby improving the welding efficiency, and effectively improving the stability and production efficiency of welding. The welding robot welds the joint surface of the battery shell according to the optimal laser welding parameters, which can ensure that the welding quality of the joint surface meets the requirements, reduce welding defects, and improve the performance and reliability of the battery shell.

[0080] In steps S1-S6, by laser fine-tuning the joint surface of the battery shell, the structure of the microscopic lattice of the joint surface can be changed and image data can be formed. Diffraction analysis is performed on the acquired image data, which is a detailed analysis of the lattice of the joint surface by methods such as X-ray diffraction (XRD) or electron diffraction, so as to extract the structural data set of the lattice. Lattice optimization can be performed based on the lattice structure data set. By adjusting the arrangement direction of the grains and the position of the grain boundaries, the distribution of the grains is optimized, so that the joint surface can better cope with thermal stress during welding and reduce the occurrence of welding defects. The microscopic morphology of the joint surface is identified and processed, and the surface roughness data can be extracted to reflect the surface characteristics of the joint surface at different scales. The analysis of the surface roughness data can provide an important basis for the thermal stress simulation of the joint surface. Next, in the thermal stress simulation analysis, based on the surface roughness data, the thermal stress distribution of the joint surface under different welding parameters can be predicted. Based on the thermal stress distribution, the region is divided and processed to obtain the thermal stress gradient data sets of different regions. The thermal stress gradient data set describes the thermal stress changes at different locations on the joint surface. By analyzing this data, it is possible to find areas of thermal stress concentration that may occur during welding. Based on this data, the difference in thermal conductivity of the joint surface in different areas is further analyzed. By combining the heat transfer data set with the grain distribution data set, the distribution of thermal stress, the difference in thermal conductivity, and the optimization of grain arrangement can be comprehensively considered to determine the key parameters such as laser power, welding speed, and focus position that are most suitable for welding the joint surface, and control the welding robot to perform welding according to the optimal laser welding parameters, ultimately achieving high-quality welding of the battery shell joint surface.

[0081] In one embodiment, the step of performing laser fine-tuning on the joint surface of the battery casing and obtaining image data of the microscopic lattice of the joint surface after the laser fine-tuning includes:

[0082] Locally irradiating the joint surface with a high-frequency laser pulse to obtain the spot size and energy distribution of the laser beam irradiated on the joint surface;

[0083] Calculating the surface flatness of the laser beam based on the spot size and energy distribution, and adjusting the focal length and pulse frequency of the laser beam based on the surface flatness;

[0084] Collecting a microscopic lattice image of the bonding surface after laser fine-tuning processing, and preprocessing the microscopic lattice image through an adaptive filtering algorithm to obtain an optimized scanning image;

[0085] The boundary positions of the grains in the optimized scanned image are extracted based on an edge detection algorithm, and the grains are identified and calibrated based on the boundary positions to obtain image data of the microscopic lattice of the bonding surface.

[0086] In the above embodiment, the process of laser fine-tuning treatment starts with the irradiation of high-frequency laser pulses. The laser beam locally irradiates the joint surface of the battery shell with a certain power and frequency, and produces fine microscopic adjustments on the joint surface. Through the irradiation of the laser, the joint surface will undergo slight physical changes. After obtaining the size and energy distribution of the spot, the surface flatness of the joint surface after the laser beam irradiation is calculated. Surface flatness is a parameter that describes the uniformity of the surface microstructure, which can reflect whether the laser irradiation is uniform and whether there are some local irregularities or fluctuations. In actual operation, the calculation of surface flatness is achieved by analyzing the data after the laser beam irradiation. Specifically, the calculation of surface flatness is based on the size and energy distribution of the spot. By analyzing these data, a conclusion is drawn as to whether the surface is flat. The higher the flatness, the better the uniformity of the surface treatment, and the higher the surface quality. According to the calculated surface flatness, the focal length and pulse frequency of the laser beam are adjusted. The focal length is adjusted to optimize the focusing effect of the laser beam, thereby ensuring that the heat generated by the laser pulse on the joint surface can be evenly distributed. The adjustment of pulse frequency is related to the irradiation frequency of the laser. A reasonable pulse frequency helps to ensure that the thermal effect during the laser treatment process is properly controlled and does not cause overheating or local overheating. The image data of the microscopic lattice of the joint surface after laser fine-tuning is collected, and the lattice structure of the joint surface at the microscopic scale is captured through high-precision imaging technology. The image data can provide information such as the morphology, size and distribution of the grains, and the image is pre-processed by an adaptive filtering algorithm. Adaptive filtering is a technology that can dynamically adjust the filtering parameters according to the local features of the image, which can effectively remove the noise in the image while retaining important structural information. Through this filtering process, the details in the image are clearer and the noise is suppressed, thereby obtaining a more accurate scanned image. The edge detection algorithm is used to extract the boundary position of the grain, identify the area with drastic grayscale changes in the image, and thus find the boundary of the grain. In the microstructure of the joint surface, the boundary of the grain is manifested as an area with obvious light and dark contrast. The edge detection algorithm can efficiently identify these boundaries and extract the outline of the grain. The extracted grain boundary information will be used for subsequent grain identification and calibration. Based on the extracted boundary position, each grain in the image is separated and numbered or marked according to the boundary information, thereby obtaining image data of the microscopic lattice of the joint surface.

[0087] In one embodiment, the step of performing diffraction analysis on the image data to obtain a lattice structure data set of the microscopic lattice, wherein the lattice structure data set at least includes arrangement information of grains in the joint surface and positions of grain boundaries, comprises:

[0088] Acquire the angular distribution information and periodic characteristics of the microscopic lattice in the image data based on the diffraction technology to obtain a microscopic lattice distribution map;

[0089] Performing Fourier transform processing on the microscopic lattice distribution map to obtain frequency domain image data;

[0090] Acquire the intensity distribution of different frequency components in the frequency domain image data, identify and calculate the main orientation direction of the grains, and form grain arrangement direction data;

[0091] Performing grain boundary identification processing on the microscopic lattice distribution map to obtain grain boundary position data;

[0092] The grain arrangement direction data and the grain boundary position data are combined based on an image fusion algorithm to form the lattice structure data set.

[0093] In the above embodiment, the microscopic lattice angle distribution information and periodic characteristics in the image data are obtained based on the diffraction technology. The diffraction technology can extract the basic information of the lattice by measuring the diffraction pattern of the laser or other light source irradiated on the joint surface. In this process, after the laser beam is irradiated on the joint surface, the lattice structure will diffract the light in a specific way. These diffraction patterns carry information about the lattice arrangement and periodic characteristics. By calculating the angle distribution information of these diffraction patterns, the distribution characteristics of the lattice in space can be obtained, and the distribution map of the microscopic lattice can be further obtained. The distribution map can be transformed into the representation of the frequency component through Fourier transformation, so that the periodic characteristics of the lattice become more significant. Specifically, the frequency domain image data after Fourier transformation contains the intensity distribution of different frequency components. The frequency component is directly related to the periodic structure of the lattice. The performance of the lattice structure with a longer period in the frequency domain usually corresponds to a lower frequency, while the structure with a shorter period corresponds to a higher frequency. By analyzing the intensity distribution of these frequency components, the main orientation direction of the grains can be identified. The main orientation direction of the grain refers to the arrangement direction of the crystal structure inside the grain, which determines the physical properties of the grain and the overall mechanical properties. At the microscopic scale, the arrangement of the grains is not only affected by the properties of the material itself, but also by external factors such as processing and laser irradiation. Therefore, the main orientation direction of the grain can be extracted through the intensity distribution in the frequency domain image data, and then the microscopic lattice distribution map is processed for grain boundary recognition. The microscopic lattice distribution map is analyzed by image processing technology to detect the edge parts in the image, which usually correspond to the position of the grain boundary. Grain boundary recognition can effectively distinguish the boundaries between different grains, and then obtain the position data of the grain boundary. Based on the image fusion algorithm, the grain arrangement direction data and the grain boundary position data are combined to form a complete lattice structure data set. The lattice structure data set contains the arrangement information of the grains and accurately calibrates the position of the grain boundary.

[0094] In one embodiment, the step of adjusting the arrangement direction and grain boundary position of the microscopic lattice to optimize the lattice structure dataset to obtain an optimized grain distribution dataset includes:

[0095] Extracting the distribution characteristics of each grain in the arrangement direction of the microscopic lattice to obtain a grain distribution characteristic data set;

[0096] Acquiring the relative orientation relationship between the grains in the grain distribution characteristic data set, adjusting the arrangement direction of the grains based on the relative orientation relationship, and obtaining a grain orientation optimization data set;

[0097] Acquire the mutual relationship between the crystal grains and the grain boundaries in the lattice structure data set, and adjust the grain boundary position based on the mutual relationship to obtain an optimized grain boundary position data set;

[0098] The relative angle value between each of the grains and the adjacent grains in the grain direction optimization data set is calculated, and the relative position between the grains is optimized based on the relative angle value and the grain boundary position data set to obtain an optimized grain distribution data set.

[0099] In the above embodiment, the distribution characteristics in the arrangement direction of each grain in the microscopic lattice are extracted, and a grain distribution characteristic data set is formed, the relative orientation relationship between these grains is obtained, and the arrangement direction of the grains is adjusted based on these relative orientation relationships. The relative orientation relationship between grains refers to the orientation angle between adjacent grains and their arrangement rules. For example, in polycrystalline materials, the orientation difference between grains usually affects the mechanical properties of the material, such as strength and toughness. The orientation difference between grains can be calculated by an algorithm, and methods such as orientation matrix, Euler angle or other related methods in crystallography are used to quantify the orientation relationship between grains. Then, based on these relationships, the arrangement direction of the grains is adjusted to optimize their distribution in space. The relationship between the grains and the grain boundaries in the lattice structure data set is obtained, and the position of the grain boundaries is adjusted based on this relationship. The relationship between the grains and the grain boundaries is analyzed by high-resolution imaging or crystallographic methods, including the relative position of the grains, the type of grain boundaries, and the angle relationship between the grains and the grain boundaries. Based on this information, the grain boundary position can be adjusted to reduce the formation of bad grain boundaries, thereby optimizing the overall performance of the material, and finally obtaining an optimized grain boundary position data set. Based on the grain orientation optimization data set and the grain boundary position data set, the relative angle value between each grain and the adjacent grains is calculated, and the relative position between the grains is further optimized. By calculating the relative angle and distance between the grains, combined with the optimized grain orientation and grain boundary position data, the grains can be fine-tuned in space. Specifically, if it is found that the angle deviation between some grains and their adjacent grains is large, it may cause uneven performance of the material. The position or orientation of these grains can be adjusted to reduce the angle difference between the grains, thereby improving the orderliness of the grains and the comprehensive performance of the material, and obtaining a fully optimized grain distribution data set, which contains the optimized position and orientation information of each grain.

[0100] In one embodiment, the steps of obtaining the surface state of the joint surface, identifying and processing the microscopic morphology data in the surface state of the joint surface to obtain surface roughness data, performing thermal stress simulation analysis on the surface roughness data, and obtaining the thermal stress distribution of the joint surface under different welding parameters include:

[0101] Performing three-dimensional imaging processing on the bonding surface by using a confocal microscope to obtain surface microscopic morphology data of the bonding surface, wherein the surface microscopic morphology data at least includes surface texture data and height difference data;

[0102] Calculating surface roughness data of the joint surface based on the surface texture data and the height difference data;

[0103] Calculating and extracting statistical characteristic data of the surface roughness data, obtaining the distribution of the surface roughness, and forming statistical data of the surface roughness based on the distribution;

[0104] According to the surface roughness characteristic data, thermal stress simulation is performed on the joint surface by finite element analysis to obtain thermal stress distribution data;

[0105] Based on the thermal stress simulation results, the back propagation algorithm is used to optimize the thermal stress distribution data to obtain the optimized thermal stress data;

[0106] According to the optimized thermal stress data, the influence of different welding parameters on the thermal stress distribution is analyzed to obtain the thermal stress distribution of the joint surface under different welding parameters.

[0107] In the above embodiment, the joint surface is subjected to three-dimensional imaging processing by a confocal microscope to obtain the surface microscopic morphology data of the joint surface. In this step, the confocal microscope scans the surface of the joint surface, and collects the three-dimensional coordinates of each point to generate a three-dimensional image of the surface of the joint surface. The obtained microscopic morphology data includes at least surface texture data and height difference data. The surface texture data represents the microscopic morphology of the surface of the joint surface, such as fine undulations, uneven areas on the surface, etc. The height difference data reflects the height difference between different areas on the joint surface. The surface roughness data of the joint surface is calculated based on the surface texture data and the height difference data, and the scale and intensity of the surface undulation are extracted by a certain algorithm. For example, common roughness indicators such as root mean square roughness (Rq) and average roughness (Ra) can be used to quantify the microscopic roughness of the surface. The surface roughness data is statistically analyzed, such as the acquisition of distribution, the calculation of features such as mean, standard deviation, and peak value. Based on these statistical feature data, statistical data of surface roughness can be formed, and thermal stress simulation can be performed through finite element analysis. Finite element analysis is a numerical simulation method that solves the thermal stress distribution by establishing a mathematical model of the joint surface and its surrounding area. In this process, the surface roughness characteristic data and welding process parameters (such as welding temperature, speed, heating and cooling rate, etc.) are input, and then the heat distribution, temperature change and thermal stress generated by temperature gradient during welding are solved by finite element analysis. The results of thermal stress simulation can reveal the distribution of thermal stress on the joint surface under different welding parameters, and these data provide a theoretical basis for optimizing the welding process. The thermal stress distribution data is optimized by the back propagation algorithm. Based on the gap between the simulation results and the expected results, the model parameters are continuously adjusted to optimize the results. The error of thermal stress can be calculated and the parameters in the model (such as welding parameters, surface roughness, etc.) can be gradually adjusted to obtain more accurate thermal stress distribution data. By analyzing the optimized thermal stress data, the influence of different welding parameters on thermal stress distribution can be further studied. Different welding parameters (such as welding temperature, welding speed, welding method, etc.) will have different effects on the thermal stress distribution on the joint surface. By analyzing the optimized thermal stress data, the thermal stress distribution of the joint surface under different welding parameters can be obtained, thereby revealing which parameters have a significant impact on the thermal stress and which parameters can effectively control the thermal stress, thereby improving the joint quality.

[0108] In one embodiment, the calculation expression of the above content is: ,in, is the optimized thermal stress distribution data, which is the final thermal stress data after optimization by the back propagation algorithm, and represents the thermal stress distribution on the joint surface under different welding parameters. , , are weight coefficients, which control the influence of each part of the calculation. The weight coefficients are empirical values, which determine the importance of surface roughness to thermal stress simulation, back propagation optimization and error minimization. It represents the integral calculation of the thermal stress distribution of the joint surface under different welding parameters P, considering the surface roughness S and height difference data r. The integral here is the thermal stress calculation of the entire joint surface area Ω. The function F is a function that represents the relationship between heat conduction, stress distribution, roughness and welding parameters. This formula assumes that the heat conduction process is nonlinearly coupled with the roughness data and is regulated by the welding parameters. It indicates that the back propagation algorithm BP optimizes the thermal stress distribution data. The goal of the back propagation algorithm is to minimize the error of thermal stress simulation and optimize the result by adjusting the welding parameter P. N is the upper limit of the number of iterations in the back propagation algorithm, which means that the algorithm gradually improves the prediction accuracy through multiple training and optimization. is the weight matrix in the back propagation algorithm, which indicates the influence of each data point (i.e., thermal stress data) in the i-th iteration. is the thermal stress optimization function in the back propagation algorithm, which means improving the i-th thermal stress data by optimizing the welding parameters (P) . Represents the loss function, which is used to measure the optimized thermal stress data and the target thermal stress data The error between The thermal stress error calculated during the optimization process is the difference between the simulation result and the target result. The formula in this embodiment is used to describe the distribution of thermal stress on the joint surface during welding and its optimization process. The thermal stress distribution on the joint surface is calculated based on factors such as surface roughness data, welding parameters, and heat conduction model. The thermal stress data is optimized by the back propagation algorithm to further reduce the prediction error, thereby providing a more accurate thermal stress distribution result.

[0109] In one embodiment, the step of dividing the thermal stress distribution into regions to obtain a thermal stress gradient data set for each region, and analyzing the thermal conductivity difference of the joint surface based on the thermal stress gradient data set to obtain a heat transfer data set includes:

[0110] Performing spatial segmentation processing on the thermal stress distribution data based on a threshold segmentation algorithm to obtain a plurality of thermal stress regions;

[0111] Calculating the thermal stress difference between each thermal stress area and an adjacent position to obtain a thermal stress gradient data set for each thermal stress area;

[0112] interpolating the thermal stress gradient data sets between the different thermal stress regions to generate a smooth and continuous thermal stress gradient distribution data set;

[0113] constructing a relationship model between thermal stress and thermal conductivity according to the thermal stress gradient data set, and generating a thermal conductivity difference data set based on the relationship model;

[0114] The thermal conductivity difference data set is substituted into the heat conduction equation for simulation calculation to obtain the heat transfer data set of each area of ​​the joint surface.

[0115] In the above embodiment, by analyzing the distribution of thermal stress on the joint surface, it is divided into multiple different regions. A threshold segmentation algorithm can be used to spatially divide the entire thermal stress distribution map according to the numerical characteristics of thermal stress. Specifically, a threshold value of thermal stress is set to divide the entire thermal stress distribution map into regions with higher thermal stress and regions with lower thermal stress. In this way, the thermal stress values ​​in each region are relatively uniform, while there are obvious thermal stress differences between regions. The thermal stress difference between each thermal stress region and its adjacent regions is calculated to obtain a thermal stress gradient data set for each region. The thermal stress gradient refers to the rate at which thermal stress changes with position between two adjacent regions in space. The thermal stress value of the adjacent region can be calculated by differential calculation to obtain the thermal stress change of each region relative to its adjacent region, which is completed based on the thermal stress value on the region boundary, thereby obtaining an accurate thermal stress gradient data set. For example, if the thermal stress of a region is 500 MPa and the thermal stress of its adjacent region is 450 MPa, then the thermal stress difference between them is 50 MPa, so that the thermal stress gradient of the region can be obtained. The thermal stress gradient data between different regions are interpolated so that the thermal stress gradient can be smoother and continuous on the entire joint surface. Since the thermal stress gradient data is usually discrete, in practical applications, the interpolation method can effectively fill the gaps between the data, avoid the calculation errors caused by data discontinuity, and obtain a continuous and smooth thermal stress gradient data set. Based on these data, a relationship model between thermal stress and thermal conductivity is constructed. There is a certain physical relationship between thermal stress and thermal conductivity. Since the thermal conductivity of a material is affected by temperature, stress and the properties of the material itself, the difference in thermal conductivity can be inferred by analyzing the thermal stress gradient data set. For example, if the stress in a certain thermal stress area is high, it may cause the thermal conductivity of the material in that area to change. Based on this relationship, a mathematical model can be established through experimental data or theoretical models to describe the relationship between thermal stress and thermal conductivity. Specifically, this relationship model can be a regression model, a physical model, or an empirical formula obtained by fitting in a variety of ways. Through this model, a thermal conductivity difference data set can be further generated, which will be used to describe the change in thermal conductivity between different regions. Substituting the generated thermal conductivity difference data set into the heat conduction equation for simulation calculation, based on Fourier's law, by introducing the thermal conductivity difference data set into the heat conduction equation, the heat transfer process between various regions on the joint surface can be simulated. The change in thermal conductivity will affect the diffusion speed and path of heat, so substituting the thermal conductivity difference into the heat conduction equation can obtain a more accurate heat transfer data set.

[0116] In one embodiment, the step of combining the heat transfer data set with the grain distribution data set to obtain optimal laser welding parameters, and controlling a welding robot to weld the joint surface of the battery casing according to the optimal laser welding parameters comprises:

[0117] Standardizing the thermal conductivity data in the heat transfer data set and the grain arrangement information in the grain distribution data set to obtain a standardized heat transfer data set and a standardized grain distribution data set;

[0118] Based on a weighted algorithm, the standardized heat transfer data set and the grain distribution data set are weightedly fused to obtain a fused comprehensive data set;

[0119] Constructing a multidimensional search space for laser welding parameters, and calculating, within the multidimensional search space, the influence value of each set of the laser welding parameters on the joint surface according to the comprehensive data set based on a particle swarm optimization algorithm;

[0120] The influence values ​​are sorted and an optimal laser welding parameter combination is screened out, and a welding robot is controlled to weld the joint surface of the battery casing based on the optimal laser welding parameter combination.

[0121] In the above embodiment, the heat transfer data set includes thermal conductivity data related to heat conduction, and the grain distribution data set includes the arrangement information of grains. The two types of data are standardized to eliminate the dimensional differences and data deviations in the original data, so that the two can be comprehensively analyzed under the same benchmark. Based on the weighted algorithm, the standardized heat transfer data set and the grain distribution data set are weighted and fused to form a comprehensive data set, that is, different weight values ​​are given to each type of data according to its importance. For example, heat transfer data may have a more direct impact on the thermal effect in the welding process in some cases, and grain distribution may affect the material properties after welding, so it is necessary to give a reasonable weight according to actual needs. In practical applications, these weight values ​​can be automatically learned and optimized through historical data analysis, machine learning and other methods. The multidimensional search space of laser welding parameters is constructed to find the optimal solution among multiple laser welding parameter combinations. The parameters of laser welding include at least multiple variables such as laser power, welding speed, focal length, spot size, etc. Different combinations of these parameters will affect the quality, strength and size of the heat-affected zone of the welded joint. In a systematic way, all possible parameter combinations are considered to find the best laser welding parameter combination in a wide parameter space. The construction of a multidimensional search space is not only to list all parameter ranges, but also to explore effectively in the space through a reasonable search algorithm so as to find the optimal solution quickly and accurately. In this embodiment, a particle swarm optimization algorithm (PSO) is used to optimize the laser welding parameters in a multidimensional search space, by distributing a group of "particles" (i.e., candidate solutions) in the solution space, and adjusting the positions of the particles according to the information exchange between the particles and their respective search capabilities, and gradually approaching the optimal solution. Each particle in this process represents a combination of laser welding parameters, and the position of each particle represents the corresponding welding parameter value. Each time the particle position is updated, the influence value of the parameter combination represented by the particle on the welded joint is calculated. The influence value is calculated by a comprehensive data set, reflecting the welding effect of the parameter combination under the action of heat transfer and grain distribution. Specifically, in the particle swarm optimization process, an objective function is defined, which calculates the pros and cons of the thermal influence of each particle on the joint surface during the welding process, such as the strength of the welded joint, the size of the heat-affected zone, and other indicators according to the position of each particle. The particles are adjusted according to the feedback of the objective function, and the particles will continue to move toward a better parameter combination during the optimization process, and finally the optimal laser welding parameter combination is obtained. Through multiple iterations and updates of particle positions, an optimal solution can be obtained, which is the best combination of laser welding parameters. After the optimal combination of laser welding parameters is finally selected, the welding robot can accurately weld the joint surface of the battery shell according to this optimal combination. By controlling parameters such as laser power and welding speed, the stability of the welding process and the quality of the welded joint are ensured.

[0122] Reference Figure 2 , a battery shell laser welding device, comprising:

[0123] An adjustment module 100 is used to perform laser fine-tuning on the joint surface of the battery shell and obtain image data of the microscopic lattice of the joint surface after the laser fine-tuning;

[0124] An analysis module 200 is used to perform diffraction analysis on the image data to obtain a lattice structure data set of the microscopic lattice, wherein the lattice structure data set at least includes arrangement information of grains in a bonding surface and positions of grain boundaries;

[0125] An optimization module 300 is used to adjust the arrangement direction of the microscopic lattice and the position of the grain boundary to optimize the lattice structure data set to obtain an optimized grain distribution data set;

[0126] The processing module 400 is used to obtain the surface state of the joint surface, identify and process the microscopic morphology data in the surface state of the joint surface to obtain surface roughness data, perform thermal stress simulation analysis on the surface roughness data, and obtain the thermal stress distribution of the joint surface under different welding parameters;

[0127] A division module 500 is used to perform a region division process on the thermal stress distribution to obtain a thermal stress gradient data set of each region, and analyze the thermal conductivity difference of the joint surface based on the thermal stress gradient data set to obtain a heat transfer data set;

[0128] The control module 600 is used to combine the heat transfer data set with the grain distribution data set to obtain optimal laser welding parameters, and control the welding robot to weld the battery shell joint surface according to the optimal laser welding parameters.

[0129] In this embodiment, by laser fine-tuning the joint surface of the battery shell, the structure of the microscopic lattice of the joint surface can be changed and image data can be formed. Diffraction analysis is performed on the acquired image data, which is a detailed analysis of the lattice of the joint surface by methods such as X-ray diffraction (XRD) or electron diffraction, so as to extract the structural data set of the lattice. Lattice optimization can be performed based on the lattice structure data set. By adjusting the arrangement direction of the grains and the position of the grain boundaries, the distribution of the grains is optimized, so that the joint surface can better cope with thermal stress during welding and reduce the occurrence of welding defects. The microscopic morphology of the joint surface is identified and processed, and surface roughness data can be extracted to reflect the surface characteristics of the joint surface at different scales. The analysis of surface roughness data can provide an important basis for the thermal stress simulation of the joint surface. Next, in the thermal stress simulation analysis, based on the surface roughness data, the thermal stress distribution of the joint surface under different welding parameters can be predicted. Based on the thermal stress distribution, the region division processing is performed to obtain the thermal stress gradient data set of different regions. The thermal stress gradient data set describes the changes in thermal stress at different positions of the joint surface. By analyzing these data, the thermal stress concentration area that may appear during the welding process can be found. Based on these data, we further analyze the differences in thermal conductivity in different areas of the joint surface. By combining the heat transfer data set with the grain distribution data set, we can comprehensively consider the distribution of thermal stress, the difference in thermal conductivity, and the optimization of grain arrangement, so as to determine the key parameters such as laser power, welding speed, and focal position that are most suitable for welding the joint surface. We can also control the welding robot to perform welding according to the optimal laser welding parameters, and finally achieve high-quality welding of the battery shell joint surface.

[0130] Reference Figure 3 In an embodiment of the present application, a computer device is also provided. The computer device may be a server, and its internal structure may be as follows: Figure 3 As shown. The computer device includes a processor, a memory, a network interface and a database connected through a system bus. Among them, the processor designed by the computer is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store data such as a battery housing laser welding process method database. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a battery housing laser welding process method is implemented.

[0131] An embodiment of the present application also provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, a battery shell laser welding process method is implemented, including the following steps: performing laser fine-tuning processing on the joint surface of the battery shell to obtain image data of the microscopic lattice of the joint surface after the laser fine-tuning processing; performing diffraction analysis on the image data to obtain a lattice structure data set of the microscopic lattice, wherein the lattice structure data set at least includes arrangement information and grain boundary positions of the grains in the joint surface; adjusting the arrangement direction and grain boundary positions of the microscopic lattice to optimize the lattice structure data set to obtain an optimized grain distribution data set; obtaining the The surface state of the joint surface is identified, the microscopic morphology data in the surface state of the joint surface is identified and processed to obtain surface roughness data, the surface roughness data is subjected to thermal stress simulation analysis to obtain the thermal stress distribution of the joint surface under different welding parameters; the thermal stress distribution is divided into regions to obtain a thermal stress gradient data set of each region, the thermal conductivity difference of the joint surface is analyzed based on the thermal stress gradient data set to obtain a heat transfer data set; the heat transfer data set is combined with the grain distribution data set to obtain optimal laser welding parameters, and the welding robot is controlled to weld the joint surface of the battery shell according to the optimal laser welding parameters.

[0132] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media provided in this application and used in the embodiments may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double-speed data rate SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0133] The above description is only a preferred embodiment of the present invention, and does not limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made by using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A battery shell laser welding process, characterized in that: include: Performing laser fine-tuning processing on the joint surface of the battery shell to obtain image data of the microscopic lattice of the joint surface after the laser fine-tuning processing; Performing diffraction analysis on the image data to obtain a lattice structure data set of the microscopic lattice, wherein the lattice structure data set at least includes arrangement information of crystal grains in a bonding surface and positions of grain boundaries; Adjusting the arrangement direction and grain boundary position of the microscopic lattice to optimize the lattice structure data set to obtain an optimized grain distribution data set; Acquiring the surface state of the joint surface, identifying and processing the microscopic morphology data in the surface state of the joint surface to obtain surface roughness data, performing thermal stress simulation analysis on the surface roughness data, and obtaining the thermal stress distribution of the joint surface under different welding parameters; Performing regional division processing on the thermal stress distribution to obtain a thermal stress gradient data set of each region, and analyzing the thermal conductivity difference of the joint surface based on the thermal stress gradient data set to obtain a heat transfer data set; Combining the heat transfer data set with the grain distribution data set to obtain optimal laser welding parameters, and controlling a welding robot to weld the joint surface of the battery casing according to the optimal laser welding parameters; The step of adjusting the arrangement direction and grain boundary position of the microscopic lattice to optimize the lattice structure data set to obtain an optimized grain distribution data set includes: Extracting the distribution characteristics of each grain in the arrangement direction of the microscopic lattice to obtain a grain distribution characteristic data set; Acquiring the relative orientation relationship between the grains in the grain distribution characteristic data set, adjusting the arrangement direction of the grains based on the relative orientation relationship, and obtaining a grain orientation optimization data set; Acquire the mutual relationship between the crystal grains and the grain boundaries in the lattice structure data set, and adjust the grain boundary position based on the mutual relationship to obtain an optimized grain boundary position data set; Calculating the relative angle value between each of the grains and adjacent grains in the grain direction optimization data set, optimizing the relative position between the grains based on the relative angle value and the grain boundary position data set, and obtaining an optimized grain distribution data set; The step of dividing the thermal stress distribution into regions to obtain a thermal stress gradient data set for each region, and analyzing the thermal conductivity difference of the joint surface based on the thermal stress gradient data set to obtain a heat transfer data set includes: Performing spatial segmentation processing on the thermal stress distribution based on a threshold segmentation algorithm to obtain a plurality of thermal stress regions; Calculating the thermal stress difference between each thermal stress area and an adjacent position to obtain a thermal stress gradient data set for each thermal stress area; interpolating the thermal stress gradient data sets between the different thermal stress regions to generate a smooth and continuous thermal stress gradient distribution data set; constructing a relationship model between thermal stress and thermal conductivity according to the thermal stress gradient data set, and generating a thermal conductivity difference data set based on the relationship model; The thermal conductivity difference data set is substituted into the heat conduction equation for simulation calculation to obtain the heat transfer data set of each area of ​​the joint surface.

2. The battery housing laser welding process according to claim 1, characterized in that: The step of performing laser fine-tuning on the joint surface of the battery casing and obtaining image data of the microscopic lattice of the joint surface after the laser fine-tuning includes: Locally irradiating the joint surface with a high-frequency laser pulse to obtain the spot size and energy distribution of the laser beam irradiated on the joint surface; Calculating the surface flatness of the laser beam based on the spot size and energy distribution, and adjusting the focal length and pulse frequency of the laser beam based on the surface flatness; Collecting a microscopic lattice image of the bonding surface after laser fine-tuning processing, and preprocessing the microscopic lattice image through an adaptive filtering algorithm to obtain an optimized scanning image; The boundary positions of the grains in the optimized scanned image are extracted based on an edge detection algorithm, and the grains are identified and calibrated based on the boundary positions to obtain image data of the microscopic lattice of the bonding surface.

3. The battery housing laser welding process according to claim 1, characterized in that: The step of performing diffraction analysis on the image data to obtain a lattice structure data set of the microscopic lattice, wherein the lattice structure data set at least includes arrangement information of grains in the joint surface and positions of grain boundaries, comprises: Acquire the angular distribution information and periodic characteristics of the microscopic lattice in the image data based on the diffraction technology to obtain a microscopic lattice distribution map; Performing Fourier transform processing on the microscopic lattice distribution map to obtain frequency domain image data; Acquire the intensity distribution of different frequency components in the frequency domain image data, identify and calculate the main orientation direction of the grains, and form grain arrangement direction data; Performing grain boundary identification processing on the microscopic lattice distribution map to obtain grain boundary position data; The grain arrangement direction data and the grain boundary position data are combined based on an image fusion algorithm to form the lattice structure data set.

4. The battery housing laser welding process according to claim 1, characterized in that: The step of obtaining the surface state of the joint surface, identifying and processing the microscopic morphology data in the surface state of the joint surface to obtain surface roughness data, performing thermal stress simulation analysis on the surface roughness data to obtain the thermal stress distribution of the joint surface under different welding parameters includes: Performing three-dimensional imaging processing on the bonding surface by using a confocal microscope to obtain surface microscopic morphology data of the bonding surface, wherein the surface microscopic morphology data at least includes surface texture data and height difference data; Calculating surface roughness data of the joint surface based on the surface texture data and the height difference data; Calculating and extracting statistical characteristic data of the surface roughness data, obtaining the distribution of the surface roughness, and forming statistical data of the surface roughness based on the distribution; According to the surface roughness characteristic data, thermal stress simulation is performed on the joint surface by finite element analysis to obtain thermal stress distribution data; Based on the thermal stress simulation results, the back propagation algorithm is used to optimize the thermal stress distribution data to obtain the optimized thermal stress data; According to the optimized thermal stress data, the influence of different welding parameters on the thermal stress distribution is analyzed to obtain the thermal stress distribution of the joint surface under different welding parameters.

5. The battery housing laser welding process according to claim 1, characterized in that: The step of combining the heat transfer data set with the grain distribution data set to obtain optimal laser welding parameters, and controlling a welding robot to weld the joint surface of the battery casing according to the optimal laser welding parameters comprises: Standardizing the thermal conductivity data in the heat transfer data set and the grain arrangement information in the grain distribution data set to obtain a standardized heat transfer data set and a standardized grain distribution data set; Based on a weighted algorithm, the standardized heat transfer data set and the grain distribution data set are weightedly fused to obtain a fused comprehensive data set; Constructing a multidimensional search space for laser welding parameters, and calculating, within the multidimensional search space, the influence value of each set of the laser welding parameters on the joint surface according to the comprehensive data set based on a particle swarm optimization algorithm; The influence values ​​are sorted and an optimal laser welding parameter combination is screened out, and a welding robot is controlled to weld the joint surface of the battery casing based on the optimal laser welding parameter combination.

6. A battery shell laser welding device, applied to the battery shell laser welding process method according to claim 1, characterized in that: include: An adjustment module, used for performing laser fine-tuning processing on the joint surface of the battery shell, and obtaining image data of the microscopic lattice of the joint surface after the laser fine-tuning processing; An analysis module, configured to perform diffraction analysis on the image data to obtain a lattice structure data set of the microscopic lattice, wherein the lattice structure data set at least includes arrangement information of grains in a bonding surface and positions of grain boundaries; An optimization module, used for adjusting the arrangement direction and grain boundary position of the microscopic lattice to optimize the lattice structure data set to obtain an optimized grain distribution data set; a processing module, for obtaining the surface state of the joint surface, identifying and processing the microscopic morphology data in the surface state of the joint surface to obtain surface roughness data, performing thermal stress simulation analysis on the surface roughness data, and obtaining the thermal stress distribution of the joint surface under different welding parameters; A partitioning module, used to perform a region partitioning process on the thermal stress distribution to obtain a thermal stress gradient data set of each region, and to analyze the thermal conductivity difference of the joint surface based on the thermal stress gradient data set to obtain a heat transfer data set; A control module is used to combine the heat transfer data set with the grain distribution data set to obtain optimal laser welding parameters, and control a welding robot to weld the joint surface of the battery casing according to the optimal laser welding parameters.

7. A computer device, characterized in that: It includes a processor, a memory, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, it implements the battery casing laser welding process method as described in any one of claims 1 to 5.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the battery casing laser welding process method as described in any one of claims 1 to 5 are implemented.

Images