Method for constructing three-dimensional hybrid grid model of electrode microstructure

Abstract

The invention discloses a method for constructing a three-dimensional hybrid grid model of an electrode microstructure, and belongs to the technical field of lithium ion batteries. Comprising the following steps: S1, carrying out image preprocessing on an electrode tomography image set; s2, performing feature matching on each scanned image of the preprocessed image set, and calculating an optimal gray threshold; s3, identifying different phases of the scanning image electrodes by adopting the optimal gray threshold value and segmenting the different phases of the scanning image electrodes; s4, reconstructing a structured grid, performing secondary division on the structured grid at the phase boundary, and correcting an irregular interface through a tetrahedral grid; and S5, performing defect repair to obtain a three-dimensional hybrid grid model. The technical scheme has the beneficial effects that the representation reconstruction of the microstructure in the electrode can be realized, the generated reconstructed electrode hybrid grid model can be directly used for numerical simulation calculation, the electrode reconstruction efficiency is improved, and the reduction degree of the reconstructed electrode structure is higher.

Description

Technical Field

[0001] The invention relates to the technical field of lithium ion batteries, and in particular to a method for constructing a three-dimensional hybrid grid model of an electrode microstructure. Background Art

[0002] In addition to the influence of material properties, the heterogeneity of the internal microstructure of the electrode directly affects the material transport, charge conduction and interface reaction inside the electrode, which in turn affects the electrochemical, thermodynamic and mechanical properties of the electrode, thereby affecting the energy density, cycle life and safety of the battery. Therefore, accurately characterizing the internal microstructure of the electrode is of great significance for studying the influencing factors and performance improvement of electrode performance, and can provide ideas for the design of a new generation of high-performance electrode structures. The reconstruction of the electrode microstructure based on the tomography atlas can accurately describe the internal structure of the porous electrode, and by meshing the reconstructed model, it can be used for numerical simulation to study the influence of the internal structure of the electrode on the electrode performance.

[0003] In the prior art, the process of three-dimensional reconstruction and mesh generation of electrode microstructure based on tomographic atlas is complicated. It is difficult to ensure the accuracy of interface reconstruction using only structured meshes, while the calculation efficiency of unstructured meshes is low. Summary of the invention

[0004] The purpose of the present invention is to provide a method for constructing a three-dimensional hybrid grid model of an electrode microstructure to solve the above technical problems;

[0005] A method for constructing a three-dimensional hybrid grid model of an electrode microstructure, comprising:

[0006] Step S1, performing image preprocessing on the electrode tomography atlas to obtain a preprocessed atlas;

[0007] Step S2, performing feature matching on each scanned image in the preprocessed atlas, calculating the best grayscale threshold for segmenting a single scanned image, and performing statistical analysis to obtain the optimal grayscale threshold;

[0008] Step S3, using the optimal grayscale threshold to respectively identify different phases of the scanned image electrodes and segment them;

[0009] Step S4, reconstructing a structured grid according to the segmentation results of different phases of the electrode, performing secondary division on the structured grid at the phase boundary to obtain an unstructured grid, and correcting the irregular interface by a tetrahedral grid to obtain an electrode hybrid grid;

[0010] Step S5, repairing defects of the electrode hybrid mesh, removing the free active material phase and the pore phase by voxel expansion and erosion technology, and obtaining a three-dimensional hybrid mesh model.

[0011] Preferably, step S1 comprises,

[0012] Step S11, using a median filter to perform smoothing filtering on the electrode tomography atlas;

[0013] Step S12, performing image enhancement on the filtered electrode tomography atlas using histogram equalization to obtain the preprocessed atlas.

[0014] Preferably, the optimal grayscale threshold in step S2 is the optimal threshold for identifying each phase of each scanned image in the preprocessed atlas, and is calculated based on the auxiliary division of various morphological features and the volume ratio of each phase of the electrode material. The optimal grayscale threshold is a weighted comprehensive calculation value of the statistical median, mode and average of the optimal threshold in the preprocessed atlas.

[0015] Preferably, step S3 comprises,

[0016] The grayscale values ​​of the scanned image are segmented, and the grayscale values ​​are grouped using the optimal grayscale threshold, and the coated areas of different grayscale value groups are the areas of each phase of the electrode.

[0017] Preferably, step S4 comprises,

[0018] Step S41, selecting the size of the electrode hybrid grid;

[0019] Step S42, reconstructing the structured grid according to the segmentation results of different phases of the electrode, and assigning corresponding attribute values ​​to each phase for distinction;

[0020] Step S43, performing secondary division on the structured grid at the phase boundary to obtain an unstructured grid;

[0021] Step S44, using a tetrahedral mesh to correct the irregular interface to obtain the electrode hybrid mesh.

[0022] Preferably, the division of the structured grid in step S42 is to generate a discretized grid according to the representative volume unit size of the selected electrode hybrid grid;

[0023] The length and width of the representative volume unit are both 5 to 8 times the diameter of the active particle. The size of the structured grid is based on 1 / 30 to 1 / 20 of the diameter of the active particle. The structured grid is generated according to the actual thickness and grid density of each structural area of ​​the electrode.

[0024] The identification results of different phases of the electrode are mapped by pixel-voxel in a one-to-one or many-to-one mapping ratio.

[0025] Preferably, the defect repair process of the electrode hybrid grid in step S5 includes:

[0026] The electrode mixed grid is traversed to find the free active material phase not connected to the current collector and convert the free active material phase into the electrolyte phase, and the free pores not connected to the separator are found and converted into the active material phase.

[0027] Preferably, the method further comprises step S6, calculating electrode microstructure characteristic parameters according to the three-dimensional hybrid grid model, wherein the electrode microstructure characteristic parameters include electrode porosity, electrode specific surface area and electrode tortuosity.

[0028] Preferably, the electrode porosity is calculated as follows:

[0029] The volume of the electrolyte phase grid in the porous electrode region of the three-dimensional hybrid grid model is counted to obtain the pore phase volume of the porous electrode region, and the ratio of the pore phase volume of the porous electrode region to the volume of the porous electrode region is calculated to obtain the electrode porosity.

[0030] Preferably, the calculation method of the electrode specific surface area is:

[0031] Find all active material phase mixed grids and count the volume and total surface area of ​​the active material phase mixed grids, calculate the ratio of the total surface area of ​​the active material phase mixed grids to the total volume and density of the active material phase mixed grids to obtain the electrode reconstructed specific surface area;

[0032] The total surface area of ​​the active material phase mixed grid is the sum of the contact surface areas of the active particle phase grid and the pore phase grid;

[0033] The calculation method of the electrode tortuosity is:

[0034] The electrode tortuosity is obtained by calculating the ratio of the shortest path from the current collector to the interior of the vertical interpore region of the diaphragm to the vertical thickness from the current collector to the diaphragm.

[0035] The beneficial effects of the present invention are: it can realize the reconstruction of the internal microstructure characterization of the electrode, and the generated reconstructed electrode hybrid grid model can be directly used for numerical simulation calculations, thereby improving the efficiency of electrode reconstruction; a voxel erosion algorithm is used to repair free pores and active particle phase defects introduced into the reconstructed electrode due to the quality of the atlas, so that the reconstructed electrode structure has a higher degree of restoration. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 is a step diagram of the method for constructing a three-dimensional hybrid grid model of the present invention;

[0037] Figure 2 is a schematic diagram of step S1 of the present invention;

[0038] Figure 3 is a schematic diagram of step S4 of the present invention;

[0039] Figure 4 It is a step diagram of a method for constructing a three-dimensional hybrid grid model according to a preferred embodiment of the present invention;

[0040] Figure 5 is a step diagram of a defect repair method for reconstructing an electrode hybrid grid structure according to the present invention;

[0041] Figure 6 It is a step diagram of the method for characterizing the specific surface area of ​​electrode reconstruction of the present invention;

[0042] Figure 7 It is a step diagram of the electrode tortuosity characterization method of the present invention. DETAILED DESCRIPTION

[0043] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0044] It should be noted that, in the absence of conflict, the embodiments of the present invention and the features in the embodiments may be combined with each other.

[0045] The present invention will be further described below in conjunction with the accompanying drawings and specific embodiments, but they are not intended to limit the present invention.

[0046] A method for constructing a three-dimensional hybrid mesh model of an electrode microstructure, such as Figure 1 As shown, including

[0047] Step S1, performing image preprocessing on the electrode tomography atlas to obtain a preprocessed atlas;

[0048] Step S2, performing feature matching on each scanned image in the preprocessed atlas, calculating the best grayscale threshold for segmenting a single scanned image, and performing statistical analysis to obtain the optimal grayscale threshold;

[0049] Step S3, using the optimal grayscale threshold to respectively identify and segment different phases of the scanned image electrodes;

[0050] Step S4, reconstructing a structured grid according to the segmentation results of different phases of the electrode, performing secondary division on the structured grid at the phase boundary to obtain an unstructured grid, and correcting the irregular interface by a tetrahedral grid to obtain an electrode hybrid grid;

[0051] Step S5, repairing defects of the electrode hybrid mesh, removing the free active material phase and the pore phase by voxel expansion and erosion technology, and obtaining a three-dimensional hybrid mesh model.

[0052] Specifically, the present invention provides a method for constructing a three-dimensional hybrid grid model of an electrode microstructure, which is suitable for lithium-ion batteries, can take into account both computational complexity and reconstruction accuracy, improve grid availability and convergence, and can be used to generate high-quality grids for various complex electrode microstructures. The present invention includes an image preprocessing step to improve image quality, reduce subsequent phase identification and segmentation errors, and use a hybrid grid to achieve three-dimensional reconstruction of electrodes based on tomographic atlases, which can not only achieve the characterization and reconstruction of the internal microstructure of the electrode, but also the generated reconstructed electrode hybrid grid model can be directly used for numerical simulation calculations, simplifying the electrode reconstruction and grid division process, and improving the efficiency of electrode reconstruction. At the same time, the generated hybrid grid contains complete set topology and attribute information, is easy to convert grid types, and has high compatibility. The voxel erosion algorithm is used to repair the free pores and active particle phase defects introduced by the atlas quality inside the reconstructed electrode, so that the reconstructed electrode structure has a higher degree of restoration and a more accurate characterization of microstructure features.

[0053] Specifically, the image is preprocessed to improve the quality of the atlas, the electrode phases in the image are segmented, and the reconstruction interval is initially constructed. Then the two-dimensional phase recognition results are mapped one by one with the three-dimensional hybrid grid attributes, and finally the internal structural defects of the reconstructed electrode are removed to achieve the reconstruction of the electrode microstructure and the generation of the electrode grid.

[0054] In a preferred embodiment, referring to Figure 2 , step S1 comprises,

[0055] Step S11, using a median filter to smooth the electrode tomography atlas;

[0056] Step S12, using histogram equalization to perform image enhancement on the filtered electrode tomography atlas to obtain a preprocessed atlas.

[0057] Specifically, image preprocessing includes smoothing filtering to remove noise in the image and enhance the contrast of phase boundaries to improve image quality and visibility of details. Smoothing filtering is median filtering to remove noise in the image. Image enhancement processing uses histogram equalization to enhance contrast and highlight particle edges, thereby improving the recognition accuracy of each phase of the electrode.

[0058] In a preferred embodiment, the optimal grayscale threshold in step S2 is the optimal threshold for identifying each phase in each scanned image in the preprocessed atlas. It is calculated based on the auxiliary division of various morphological features and the volume ratio of each phase of the electrode material. The optimal grayscale threshold is a weighted comprehensive calculation value of the statistical median, mode and average of the best threshold in the preprocessed atlas.

[0059] Specifically, step S2 performs feature matching on each electrode tomographic scan image after preprocessing, calculates the optimal grayscale threshold for segmenting each scanned image, and selects the optimal grayscale threshold for electrode phase segmentation in the atlas through statistical analysis.

[0060] Specifically, the optimal threshold for phase segmentation is determined by analyzing the statistical mean, mode, and median of the optimal threshold calculated based on the Otsu algorithm, and selecting the integer value closest to the three. If there are two phases, there is only one optimal threshold; if there are more than three material phases, the optimal threshold number is the number of material phases minus one;

[0061] In a preferred embodiment, step S3 includes:

[0062] The grayscale values ​​of the scanned image are segmented, and the grayscale values ​​are grouped using the optimal grayscale threshold. The coated areas of different grayscale value groups are the areas of each phase of the electrode.

[0063] Specifically, in step S3, phase region classification traverses all pixel points inside the image reconstructed section, determines the gray value based on the optimal threshold, and processes the images in the atlas using the optimal gray threshold, thereby identifying different phases of the electrode respectively; and accurately segmenting each phase based on the identification result.

[0064] In a preferred embodiment, referring to Figure 3 , step S4 comprises,

[0065] Step S41, selecting the size of the electrode mixing grid;

[0066] Step S42, reconstructing a structured grid according to the segmentation results of different phases of the electrode, and assigning corresponding attribute values ​​to each phase for distinction;

[0067] Step S43, performing secondary division on the structured grid at the phase boundary to obtain an unstructured grid;

[0068] Step S44, using a tetrahedral mesh to correct the irregular interface to obtain an electrode hybrid mesh.

[0069] Specifically, the construction of the initial reconstructed structured grid model in step S4 requires setting the reconstructed grid size. The thickness of the reconstructed electrode positive current collector, positive electrode porous area, diaphragm, negative electrode porous area, and negative electrode diaphragm are set in sequence according to the electrode thickness direction; the unstructured grid performs secondary refinement and segmentation of the structured grid on the boundary according to the actual shape of the boundary to obtain a better restoration effect. Select the electrode reconstruction grid size, reconstruct the structured grid according to the phase segmentation result, and assign corresponding attribute values ​​to each phase for distinction. For the phase boundary part, the structured grid is subdivided twice, and the irregular interface is accurately characterized using a tetrahedral grid to achieve the construction of the microstructure of each phase in the three-dimensional electrode hybrid grid.

[0070] In a preferred embodiment, the division of the structured grid in step S42 is to generate a discretized grid according to the representative volume unit size of the selected electrode hybrid grid;

[0071] The length and width of the representative volume unit are 5 to 8 times the diameter of the active particle. The size of the structured grid is based on 1 / 30 to 1 / 20 of the diameter of the active particle. The structured grid is generated according to the actual thickness and grid density of each structural area of ​​the electrode.

[0072] The identification results of different phases of the electrode are mapped by pixel-voxel in a one-to-one or many-to-one mapping ratio.

[0073] Specifically, the hybrid grid attribute assignment is performed according to the mapping ratio constructed between the voxel unit size and the pixel size to perform attribute corresponding mapping, so as to realize the hybrid grid unit attribute setting of the porous electrode area and construct the phase structure of each material in the porous electrode hybrid grid.

[0074] In a preferred embodiment, the defect repair process of the electrode hybrid grid in step S5 includes:

[0075] Traverse the electrode mixed grid, find the free active material phase that is not connected to the current collector and convert it into the electrolyte phase, find the free pores that are not connected to the separator and convert them into the active material phase.

[0076] Specifically, step S5 reconstructs the electrode hybrid grid defect repair and is divided into two major modules: identification and repair of free active material phases, and identification and repair of free pore phases. Traverse the reconstructed electrode hybrid grid generated in step S5 to find the free active material phase that is not connected to the current collector, and convert its hybrid grid attributes to the electrolyte phase. Find the free pores that are not connected to the diaphragm, and convert their hybrid grid attributes to the active material phase, so as to achieve defect identification and repair of the porous electrode area structure in the reconstructed electrode hybrid grid. The free active material phase and pore phase are removed by voxel expansion and erosion to complete the construction of a three-dimensional hybrid grid model suitable for the microstructure of lithium-ion battery electrodes. The defects are the free phase inside the reconstructed electrode, the active material particles that are not connected to the current collector, and the free pores coated inside the active material. After identification, homogenized voxel erosion is used to obtain the reconstructed electrode hybrid grid after defect removal.

[0077] In a preferred embodiment, the method further includes step S6, calculating electrode microstructure characteristic parameters according to a three-dimensional hybrid grid model, wherein the electrode microstructure characteristic parameters include electrode porosity, electrode specific surface area and electrode tortuosity.

[0078] Specifically, the three-dimensional hybrid grid model is used for electrochemical numerical simulation, and the characteristic parameters of the internal structure of the reconstructed electrode can be extracted through the reconstruction model. The electrode porosity, electrode tortuosity and electrode specific surface area can be quantitatively characterized. The tortuosity calculated by the three-dimensional A* algorithm has higher geometric accuracy than the tortuosity derived by the MacMullin number formula.

[0079] In a preferred embodiment, the electrode porosity is calculated as follows:

[0080] The volume of the electrolyte phase grid in the porous electrode region of the three-dimensional hybrid grid model is statistically analyzed to obtain the pore phase volume of the porous electrode region, and the electrode porosity is obtained by calculating the ratio of the pore phase volume of the porous electrode region to the volume of the porous electrode region.

[0081] Specifically, the electrode porosity is characterized by the ratio of the pore phase to the volume of the reconstructed porous electrode region. The volume of the pore phase in the reconstructed electrode hybrid grid model is obtained by traversing the properties of the hybrid grid cells of the reconstructed porous electrode region through a traversal algorithm.

[0082] In a preferred embodiment, the method for calculating the specific surface area of ​​the electrode is:

[0083] Find all active material phase mixed grids and count the volume and total surface area of ​​the active material phase mixed grids, calculate the ratio of the total surface area of ​​the active material phase mixed grids to the total volume and density of the active material phase mixed grids to obtain the electrode reconstruction specific surface area;

[0084] The total surface area of ​​the active material phase mixed grid is the sum of the contact surface areas of the active particle phase grid and the pore phase grid;

[0085] The calculation method of electrode tortuosity is:

[0086] The electrode tortuosity is calculated by calculating the ratio of the shortest path from the current collector to the vertical pore region of the separator to the vertical thickness from the current collector to the separator.

[0087] Specifically, refer to Figure 6 The specific surface area of ​​the electrode is characterized by the ratio of the total surface area of ​​the active material phase to its total volume and density, using the following method:

[0088] Step C1, traverse and reconstruct the electrode voxel grid;

[0089] Step C2, find the active material phase grid and calculate its volume V AM ;

[0090] Step C3, counting the contact surface area S between the active material phase grid and the electrolyte phase grid TOUCH ;

[0091] Step C4, reconstructing the specific surface area RSSA = S TOUCH / (ρ AM ·V AM );

[0092] The total surface area of ​​the active material phase is obtained by calculating the sum of the contact surface areas of the active material phase and the pore phase through a traversal algorithm, and the volume of the active material phase is obtained by calculating the difference between the volume of the reconstructed porous electrode and the volume of the pore phase.

[0093] Reference Figure 7 The electrode tortuosity is characterized by the ratio of the shortest path from the current collector to the vertical pore region between the reconstructed grid model and the thickness of the vertical porous electrode region between the two. The method is:

[0094] Step D1, importing the reconstructed electrode voxel grid;

[0095] Step D2, calculating the vertical thickness D between the current collector and the separator;

[0096] Step D3, dynamic three-dimensional A* algorithm to find the shortest path L from the electrode current collector to the vertical pore area of ​​the diaphragm min ;

[0097] Step D4, calculate the tortuosity τ = L min / D.

[0098] Specifically, for lithium-ion battery electrodes of different materials, this method can perform three-dimensional reconstruction of microstructures and generate hybrid meshes based on their corresponding tomographic atlases. Compared with the traditional three-dimensional reconstruction and meshing process of lithium-ion battery electrodes based on tomographic atlases, the present invention integrates the three-dimensional reconstruction of microstructures with the meshing process, and repairs mesh defects, thereby improving reconstruction efficiency and mesh quality, and can quantitatively characterize electrode microstructure features based on the reconstructed hybrid mesh.

[0099] Taking graphite as an example, a method for constructing a three-dimensional hybrid grid model of electrode microstructure is illustrated. Graphite is a common negative electrode material for lithium-ion batteries. The electrode structure formed by its internal particle distribution affects material transport, lithium ion deintercalation and lithium dendrite growth during charging and discharging.

[0100] In this embodiment, flake graphite obtained by scanning with an X-ray tomography microscope (XTM) is selected, and the image pixel is 325 nm.

[0101] Since the atlas contains a lot of noise points, the rest of the image is first smoothed using a median filter with a window size of 5×5, and then the filtered image is histogram equalized to enhance the quality of the atlas.

[0102] Since the graphite electrode contains only two phases, graphite particles and pores, there is only one optimal threshold. The Otsu algorithm is used to traverse the preprocessed electrode atlas, and the statistical mean of the best threshold is 128.34, the mode is 128, and the median is 128. The gray value of 128 is selected as the optimal threshold for phase segmentation of this group of graphite electrodes.

[0103] Considering that the graphite electrode particle size is between 2 μm and 15 μm, the reconstructed electrode cross-section size is selected to be 71.5×71.5 μm. 2 , phase identification is performed according to the optimal threshold.

[0104] The full electrode scanning thickness of 32.5 μm is selected as the RVE (representative volume element) thickness, so the reconstruction interval size is 71.5×71.5×32.5 μm 3 Considering the accuracy and scale of the reconstructed electrode hybrid grid, the voxel-to-pixel mapping ratio is selected as one-to-one, so the grid is divided into 220×220×100 voxels. The thickness of the negative electrode current collector and the separator are set according to the actual thickness of 12μm and 25μm. After the initial grid division is completed, the pixel-voxel unit attribute mapping is performed, and the defect repair of the generated structure is performed to obtain the graphite electrode reconstructed hybrid grid.

[0105] The microstructural characteristics of the graphite electrode hybrid grid were characterized, and its porosity was found to be 30.29% and the reconstructed specific surface area was 397.83 m 2 / g, and the tortuosity is 2.71.

[0106] Specifically, through preliminary image preprocessing to improve the atlas quality, the optimal threshold of the atlas is selected and the electrode phases are segmented, the structured grid is reconstructed according to the phase segmentation results, the structured grid at the phase boundary is divided into an unstructured grid for a second time, and finally the internal structural defects of the electrode are repaired and reconstructed to achieve the generation of electrode hybrid grid for electrode microstructure reconstruction. Based on the grid model, the three microstructural characteristics of the reconstructed electrode, namely the porosity, tortuosity and reconstructed specific surface area, are characterized.

[0107] Compared with the existing 3D reconstruction technology in the battery field, this embodiment not only integrates the microstructure 3D reconstruction with the mesh generation process, simplifies the electrode 3D reconstruction and hybrid mesh generation process suitable for lithium-ion batteries, and improves the reconstruction efficiency, but also the hybrid mesh improves the computing efficiency and reduces the demand for computing device space and performance.

[0108] At the same time, unstructured grids are used to accurately characterize the irregular microscopic shapes of lithium-ion battery electrodes, which improves the simulation accuracy. The accuracy of the reconstructed structure after image preprocessing defect repair is further improved. The generated hybrid grid has higher quality, more complete information, and higher grid compatibility. The generated hybrid grid can be directly used for numerical simulation calculations and can more accurately characterize the microstructural characteristics of the electrode.

[0109] Reference Figure 4 The embodiment of the present invention provides a method for constructing a three-dimensional hybrid grid model of an electrode microstructure. The method includes:

[0110] Step A1, early image preprocessing;

[0111] Step A2, calculating the optimal threshold of the atlas;

[0112] Step A3, phase identification and segmentation;

[0113] Step A4, initial structured grid division and secondary unstructured grid division in reconstruction interval;

[0114] Step A5, reconstructing the internal structure of the electrode to repair defects;

[0115] Finally, the reconstruction of electrode microstructure and generation of electrode grid are realized. The specific process is as follows:

[0116] First, the pixel size, scanning thickness, scanning area size, number of phases and other parameters of the electrode tomography atlas and related images were obtained. The atlas preprocessing used median filtering to smooth the noise inside the image, and the filter window size was determined according to the atlas quality. At the same time, histogram equalization was used to enhance the contrast of pixels to highlight the edges of particles, thereby obtaining a preprocessed electrode atlas with higher quality and more recognizable features.

[0117] The number of optimal thresholds is determined according to the number of electrode phases. The Otsu algorithm is used to traverse the preprocessed electrode atlas, calculate and count the optimal threshold groups for each phase segmentation of each image, and then analyze the statistical mean, mode and median of each group of optimal thresholds, and select the integer value closest to these three as the optimal threshold of the group.

[0118] Determine the position and interface size of the reconstructed electrode area, and intercept the cross section of the reconstructed interval in the atlas according to the starting point and size. Then, divide the gray value interval of each phase of the electrode with the optimal threshold as the boundary, and determine the phase interval to which the gray value of pixels at different positions belongs, so as to realize the identification of the electrode phase.

[0119] According to the diameter of the particles in the electrode scanning image, the required interval size containing more than 30 particles is estimated, and the size of the reconstruction interval and the size of the structured grid unit are determined. The thickness of the reconstructed electrode positive current collector, positive porous area, diaphragm, negative porous area, and negative diaphragm are set in sequence according to the electrode thickness direction. The cross-sectional size of the reconstructed electrode is the same as the cross-sectional size of the selected phase identification. Then, according to the shape and size of the active particles, the appropriate parameters of the unstructured tetrahedral grid size are set, and the secondary division is into a mixed grid.

[0120] The correspondence between the pixel size and the hybrid grid side length is determined. A hybrid grid unit and a pixel unit group of corresponding size are constructed, and the phase identification results of the two-dimensional pixel unit group are mapped to the corresponding hybrid grid, thereby realizing the setting of the hybrid grid unit properties of the porous electrode area, completing the construction of the internal microstructure of the electrode, and generating the original reconstructed electrode hybrid grid.

[0121] Reference Figure 5 ,The method of defect repair is,

[0122] Step B1, traverse and reconstruct the electrode voxel grid;

[0123] Step B2, finding the free active material phase grid that is not connected to the current collector;

[0124] Step B3, converting the free active material phase grid into an electrolyte phase grid;

[0125] Step B4, finding a free pore phase grid that is not connected to the diaphragm;

[0126] Step B5, converting the free pore phase grid into an active material phase grid;

[0127] This is to achieve the identification and repair of structural defects in the porous electrode area within the reconstructed electrode hybrid grid.

[0128] Starting from the electrode current collector, find the first layer of active particle mixed grid units in contact with the current collector in the thickness direction, and then continue to find the active particle mixed grid units in contact with its adjacent surface, and expand outward in sequence until all active particle grid units connected to the current collector are found. The active particle grid properties that are not connected to the current collector are converted to the electrolyte phase to achieve the identification and repair of the free active particle phase mixed grid.

[0129] Figure 5 The repair process starts with the electrode current collector, finds the first layer of electrolyte mixed grid units in contact with the diaphragm in the thickness direction, and then continues to find the electrolyte mixed grid units in contact with its adjacent surface, and expands outward in sequence until all electrolyte grid units (and pore phases) connected to the diaphragm are found. The pore phase grid properties that are not connected to the current collector are converted to active particle phases to achieve the identification and repair of free active pore phase mixed grids.

[0130] The present invention is applicable to lithium-ion batteries and provides a method for constructing a three-dimensional hybrid grid model of an electrode microstructure, including preliminary image preprocessing to improve atlas quality, atlas optimal threshold selection and electrode phase segmentation, reconstruction of a structured grid according to the phase segmentation result, secondary division of the structured grid at the phase boundary into an unstructured grid, and finally repairing and reconstructing the internal structural defects of the electrode to achieve electrode microstructure reconstruction and electrode hybrid grid generation, and characterization of three microstructural features of the reconstructed electrode porosity, tortuosity, and reconstructed specific surface area based on the grid model.

[0131] The method proposed in the present invention not only integrates the microstructure three-dimensional reconstruction and grid generation process, simplifies the three-dimensional reconstruction of electrodes suitable for lithium-ion batteries and the hybrid grid generation process, and improves the reconstruction efficiency, the hybrid grid improves the computing efficiency and reduces the demand for computing device space and performance. At the same time, the unstructured grid is used to accurately depict the microscopic irregular shape of the lithium-ion battery electrode, thereby improving the simulation accuracy. The accuracy of the reconstructed structure after image preprocessing defect repair is further improved. The generated hybrid grid has higher quality, contains complete information, and has high grid compatibility. The generated hybrid grid can be directly used for numerical simulation calculations and can more accurately characterize the electrode microstructure characteristics.

[0132] The hybrid grid is used to achieve the three-dimensional reconstruction of the electrode based on the tomographic atlas, and the internal microstructure characterization and reconstruction of the electrode is realized. The generated reconstructed electrode hybrid grid model can be directly used for numerical simulation calculations, simplifying the electrode reconstruction and meshing process and improving the efficiency of electrode reconstruction. Structured grids are used inside the phase to reduce the amount of calculation, and unstructured grids are used at the interface between the phases for accurate description.

[0133] At the same time, the generated hybrid grid contains complete set topology and attribute information, which is easy to convert grid types and has high compatibility. The voxel erosion algorithm is used to repair the free pores and active particle phase defects introduced by the atlas quality inside the reconstructed electrode, so that the reconstructed electrode structure has a higher degree of restoration and more accurate characterization of microstructure features.

[0134] The image is preprocessed in the early stage to improve the quality of the atlas, segment the electrode phases in the image, and initially construct the reconstruction interval. Then the two-dimensional phase recognition results are mapped one by one with the three-dimensional hybrid grid attributes, and finally the internal structural defects of the reconstructed electrode are removed to achieve the reconstruction of the electrode microstructure and the generation of the electrode grid.

[0135] In summary, the present application provides a method for constructing a three-dimensional hybrid grid model of an electrode microstructure. First, the microstructure atlas is phase identified and segmented, and an initial structured grid model is established based on the phase segmentation results. The structured grid at the phase boundary is divided into an unstructured grid for a second time, and the grid quality is improved by a grid correction method. The grid contains the structural position distribution of each region of the electrode and the phase attribute information. Finally, the specific surface area, tortuosity, porosity and other parameters of the reconstructed electrode are calculated based on the generated grid model. A structured grid is used inside the phase to reduce the amount of calculation, and an unstructured grid is used to accurately describe the interface between the phases. Finally, the grid defects are repaired to improve the reconstruction efficiency and grid quality. For the reconstructed grid, this method characterizes the characteristics of the reconstructed electrode microstructure, providing an effective tool for electrode structure and performance research and electrochemical simulation.

[0136] The above description is only a preferred embodiment of the present invention, and does not limit the implementation mode and protection scope of the present invention. For those skilled in the art, it should be aware that all solutions obtained by equivalent substitutions and obvious changes made using the description and illustrations of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for constructing a three-dimensional hybrid grid model of an electrode microstructure, characterized in that: include, Step S1, performing image preprocessing on the electrode tomography atlas to obtain a preprocessed atlas; Step S2, performing feature matching on each scanned image in the preprocessed atlas, calculating the best grayscale threshold for segmenting a single scanned image, and performing statistical analysis to obtain the optimal grayscale threshold; Step S3, using the optimal grayscale threshold to respectively identify different phases of the scanned image electrodes and segment them; Step S4, reconstructing a structured grid according to the segmentation results of different phases of the electrode, performing secondary division on the structured grid at the phase boundary to obtain an unstructured grid, and correcting the irregular interface by a tetrahedral grid to obtain an electrode hybrid grid; Step S5, repairing defects of the electrode hybrid mesh, removing the free active material phase and the pore phase by voxel expansion and erosion technology, and obtaining a three-dimensional hybrid mesh model.

2. The method for constructing a three-dimensional hybrid grid model of an electrode microstructure according to claim 1, characterized in that: Step S1 comprises, Step S11, using a median filter to perform smoothing filtering on the electrode tomography atlas; Step S12, performing image enhancement on the filtered electrode tomography atlas using histogram equalization to obtain the preprocessed atlas.

3. The method for constructing a three-dimensional hybrid grid model of an electrode microstructure according to claim 1, characterized in that: The optimal grayscale threshold in step S2 is the optimal threshold for identifying each phase of each scanned image in the preprocessed atlas. It is calculated based on the auxiliary division of various morphological features and the volume ratio of each phase of the electrode material. The optimal grayscale threshold is a weighted comprehensive calculation value of the statistical median, mode and average of the optimal threshold in the preprocessed atlas.

4. The method for constructing a three-dimensional hybrid grid model of an electrode microstructure according to claim 1, characterized in that: Step S3 comprises, The grayscale values ​​of the scanned image are segmented, and the grayscale values ​​are grouped using the optimal grayscale threshold, and the coated areas of different grayscale value groups are the areas of each phase of the electrode.

5. The method for constructing a three-dimensional hybrid grid model of an electrode microstructure according to claim 1, characterized in that: Step S4 comprises, Step S41, selecting the size of the electrode hybrid grid; Step S42, reconstructing the structured grid according to the segmentation results of different phases of the electrode, and assigning corresponding attribute values ​​to each phase for distinction; Step S43, performing secondary division on the structured grid at the phase boundary to obtain an unstructured grid; Step S44, using a tetrahedral mesh to correct the irregular interface to obtain the electrode hybrid mesh.

6. The method for constructing a three-dimensional hybrid grid model of an electrode microstructure according to claim 5, characterized in that: The division of the structured grid in step S42 is to generate a discretized grid according to the representative volume unit size of the selected electrode hybrid grid; The length and width of the representative volume unit are both 5 to 8 times the diameter of the active particle. The size of the structured grid is based on 1 / 30 to 1 / 20 of the diameter of the active particle. The structured grid is generated according to the actual thickness and grid density of each structural area of ​​the electrode. The identification results of different phases of the electrode are mapped by pixel-voxel in a one-to-one or many-to-one mapping ratio.

7. The method for constructing a three-dimensional hybrid grid model of an electrode microstructure according to claim 1, characterized in that: The defect repair process of the electrode hybrid grid in step S5 includes: The electrode mixed grid is traversed to find the free active material phase not connected to the current collector and convert the free active material phase into the electrolyte phase, and the free pores not connected to the separator are found and converted into the active material phase.

8. The method for constructing a three-dimensional hybrid grid model of an electrode microstructure according to claim 1, characterized in that: The method further includes step S6, calculating electrode microstructure characteristic parameters according to the three-dimensional hybrid grid model, wherein the electrode microstructure characteristic parameters include electrode porosity, electrode specific surface area and electrode tortuosity.

9. The method for constructing a three-dimensional hybrid grid model of an electrode microstructure according to claim 8, characterized in that: The electrode porosity is calculated as follows: The volume of the electrolyte phase grid in the porous electrode region of the three-dimensional hybrid grid model is counted to obtain the pore phase volume of the porous electrode region, and the ratio of the pore phase volume of the porous electrode region to the volume of the porous electrode region is calculated to obtain the electrode porosity.

10. The method for constructing a three-dimensional hybrid grid model of an electrode microstructure according to claim 8, characterized in that: The calculation method of the electrode specific surface area is: Find all active material phase mixed grids and count the volume and total surface area of ​​the active material phase mixed grids, calculate the ratio of the total surface area of ​​the active material phase mixed grids to the total volume and density of the active material phase mixed grids to obtain the electrode reconstructed specific surface area; The total surface area of ​​the active material phase mixed grid is the sum of the contact surface areas of the active particle phase grid and the pore phase grid; The calculation method of the electrode tortuosity is: The electrode tortuosity is obtained by calculating the ratio of the shortest path from the current collector to the interior of the vertical interpore region of the diaphragm to the vertical thickness from the current collector to the diaphragm.

Images