Method for morphology transformation and network optimization of a binder

Abstract

The application provides a binder morphology transformation and network structure optimization method, simulates a heat transfer path through finite element analysis, tracks binder molecular structure adjustment through molecular dynamics simulation, accurately determines a morphology transformation threshold, and effectively improves network skeleton stability by iteratively optimizing and adjusting heat distribution and binder distribution, further constructs a three-dimensional virtual model, analyzes connection strength and stress distribution, simulates deformation response under a processing environment, and finally outputs a process parameter set and an optimization scheme. The application realizes significant improvement of overall mechanical properties of electrode materials.

Description

Technical Field

[0001] This invention relates to the field of dry electrode technology, and more particularly to a method for morphological transformation and network structure optimization of binders. Background Technology

[0002] Dry electrode manufacturing, as a crucial technology in the new energy field, holds irreplaceable value in improving battery performance and reducing production costs. This technology achieves electrode material preparation through a solvent-free process, which is not only environmentally friendly but also effectively reduces resource waste and pollution problems caused by traditional wet processes. Its research and application are considered a key path to drive innovation in energy storage technology.

[0003] However, despite its enormous potential, current technological practices still face numerous challenges, requiring in-depth exploration and breakthroughs. Existing methods in dry electrode manufacturing often struggle to balance the uniformity of material mixing with energy control during production, especially in solvent-free environments where the interactions between powder materials are complex, and there is a lack of effective means to control heat transfer and material morphology changes at the microscopic level. This limitation not only affects the overall performance of the electrode materials but also restricts further improvements in production efficiency. Particularly during high-intensity machining, the problems of uneven heat distribution and localized stress differences in the material are particularly prominent, making it difficult to guarantee the quality stability of the final product.

[0004] Focusing on the technical challenges, a core issue in dry electrode manufacturing lies in the unclear heat transfer paths between powder particles caused by localized heat generated under high shear forces. This uncertainty in heat transfer directly affects the physical morphology changes within the material, especially the molecular structure adjustment and interweaving process of the binder after heating. For example, during processing, some areas may experience premature softening or even degradation of the binder due to concentrated heat, while other areas may lack sufficient heat to effectively connect material particles, resulting in an electrode structure with uneven performance. This problem further exacerbates the difficulty of constructing a three-dimensional mesh framework, making it difficult to achieve ideal mechanical and electrical properties for the electrode material.

[0005] Therefore, accurately grasping the transfer patterns of localized heat between powder particles and effectively intervening in the morphological transformation of binders and the formation of network structures through these patterns has become a critical issue that urgently needs to be addressed in dry electrode manufacturing. Solving this problem will directly impact the improvement of electrode material performance and the optimization of production processes. Summary of the Invention

[0006] To address the technical problems mentioned in the background section, the present invention provides a method for morphological transformation and network structure optimization of adhesives, the method comprising: S1, based on finite element analysis, simulates the heat transfer path of powder particles, obtains the initial temperature distribution, refines the mesh in the high shear zone, calculates the heat diffusion rate at the particle contact point, and generates a temperature distribution map. S2 simulates the process of binder molecules adjusting their structure under heat, extracts data from areas of concentrated heat, simulates the softening behavior and intertwining dynamics of molecular chains, and determines the threshold for binder morphological transformation. S3. If the threshold for the morphological transformation of the adhesive exceeds the preset upper limit, adjust the simulation parameters to reduce the local heat input, obtain the list of overheated zones, rerun the finite element analysis to balance the heat distribution, and obtain the optimized temperature distribution map. S4. Extract the connection strength index and interweaving density data between powder particles, construct a virtual model of a three-dimensional mesh skeleton and evaluate its stability. S5. If the stability of the mesh skeleton does not meet the standard, iteratively optimize the adhesive distribution, extract weak connection points and incorporate them into the finite element feedback to generate an enhanced mesh skeleton model. S6. Based on the enhanced mesh skeleton model, stress distribution data is extracted, deformation response under processing conditions is simulated, and the degree of improvement in the mechanical properties of the electrode material is evaluated. S7: Obtain the final set of process parameters, extract key control variables from the mechanical property parameters of the electrode material, and output an optimized scheme for dry electrode manufacturing.

[0007] Furthermore, step S1 includes: Step S11: Based on the preset powder particle model and finite element analysis, simulate and obtain the initial heat transfer path and temperature distribution data between particles, and perform mesh refinement for the high shear force local area to extract the contact point geometric information to calculate the local heat transfer rate distribution. Step S12: Combine the initial temperature data to simulate the heat diffusion process, obtain updated temperature distribution information, identify the location and range of abnormal areas, and generate abnormal distribution markers; Step S13: Readjust the mesh density for abnormal areas to correct the heat diffusion rate, and generate an optimized temperature distribution map and a visualized heat transfer path view.

[0008] Furthermore, step S11 includes: for the powder particle model, inputting the thermal conductivity of the particles and the initial ambient temperature, solving the heat conduction equation using the Galerkin finite element method, and outputting the initial heat transfer path network and temperature distribution matrix between the particles.

[0009] Furthermore, step S11 also includes: determining the spatial coordinates of particles and the location of contact points within a local region using the Delaunay triangulation algorithm.

[0010] Furthermore, step S12 includes: if there are abnormally high temperature nodes in the updated temperature distribution information that exceed the temperature anomaly threshold, then it is determined that there is heat accumulation in the area, and an abnormal distribution marker is generated by a connected component marking algorithm.

[0011] Furthermore, step S2 includes: Step S21: Extract the boundary data of the heat concentration area from the temperature distribution map, segment and label it using image processing technology, and use molecular dynamics simulation to construct an initial model of the binder molecules to determine its initial stress state under thermal action. Step S22: Simulate the softening behavior of molecular chains under different thermal gradients, obtain displacement and deformation data to identify key inflection points, analyze the interweaving dynamics between molecules, and extract the dynamic distribution of interactions. Step S23: Combining heat distribution data and intermolecular interaction distribution, simulate the evolution process of the overall structure of the binder to determine the critical conditions for morphological transformation, and determine whether to record characteristic information or adjust the heat gradient to re-simulate based on the critical threshold.

[0012] Furthermore, step S21 includes: based on the heat concentration area in the temperature distribution map, performing binarization processing on the image pixels using the maximum inter-class variance method, segmenting the pixels with temperatures higher than the segmentation threshold into high-heat areas, extracting the contour coordinates of the area using an edge detection algorithm, and outputting the boundary range data of the heat concentration area.

[0013] Furthermore, step S5 includes: Step S51: If the stability value of the mesh skeleton does not meet the standard, locate the weak links and obtain the stress distribution information of the key areas, start the iterative optimization process to adjust the adhesive distribution scheme, and obtain the preliminary optimized distribution parameters. Step S52: Integrate the initially optimized distribution parameters into the finite element analysis tool, simulate the stress response of the mesh skeleton under different loads to obtain improvement suggestions, and then update the adhesive distribution scheme and build the enhanced model. If the stability standard is still not met, repeat the adjustment and finite element analysis until the standard is met. Step S53: Generate the corresponding parameter configuration scheme based on the final optimized structural model to determine the stability and reliability of the mesh skeleton in practical applications.

[0014] Furthermore, step S51 includes: inputting the coordinate set and connection matrix of the mesh skeleton nodes, and using principal component analysis algorithm to extract the global stiffness matrix eigenvalues ​​of the skeleton as stability values.

[0015] Furthermore, step S51 also includes: if the deformation gradient of a certain local area exceeds the deformation gradient threshold, then the nodes in that area are marked as weak links, and their specific three-dimensional coordinate range is output.

[0016] The technical solution provided by this invention has the following beneficial effects: This invention discloses a method for optimizing the morphological transformation and network structure of binders. Addressing the issues of heat transfer between powder particles, binder morphological transformation, and network skeleton stability in dry electrode manufacturing, it proposes an integrated optimization scheme. By simulating heat transfer paths through finite element analysis and tracking binder molecular structure adjustments through molecular dynamics simulation, the threshold for morphological transformation is precisely determined. Iterative optimization of heat and binder distribution effectively improves the stability of the network skeleton. Furthermore, this invention constructs a three-dimensional virtual model to analyze connection strength and stress distribution, simulates deformation response under processing conditions, and ultimately outputs a set of process parameters and an optimized scheme. With heat balance and structural stability as its core, this invention achieves a significant improvement in the overall mechanical properties of electrode materials, providing scientific and efficient technical support for dry electrode manufacturing. Attached Figure Description

[0017] Figure 1 This is a flowchart of a method for morphological transformation and network structure optimization of an adhesive according to the present invention. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0019] like Figure 1 As shown in this embodiment, a method for morphological transformation and network structure optimization of an adhesive may specifically include: S1, based on finite element analysis, simulates the heat transfer path of powder particles, obtains the initial temperature distribution, refines the mesh in the high shear zone, calculates the heat diffusion rate at the particle contact point, and generates a temperature distribution map.

[0020] Optionally, this step also includes: Step S11: Based on the preset powder particle model and finite element analysis, simulate and obtain the initial heat transfer path and temperature distribution data between particles, and perform mesh refinement for high shear force local areas to extract contact point geometric information to calculate the local heat transfer rate distribution.

[0021] Using a pre-defined powder particle model, the finite element method (FEM) is employed to simulate the heat transfer process, obtaining the heat transfer paths and initial temperature distribution data between particles. For localized areas under high shear force, a mesh generation operation is performed to generate a refined mesh structure, determining the particle distribution and contact point locations within the local area. Geometric information at the contact points is extracted from the mesh generation results, and the heat diffusion rate at these points is calculated to obtain the heat transfer rate distribution within the local area.

[0022] Step S12: Combine the initial temperature data to simulate the heat diffusion process, obtain updated temperature distribution information, identify the location and range of abnormal areas, and generate abnormal distribution markers.

[0023] Based on the heat transfer rate distribution and combined with the initial temperature distribution data, the heat diffusion process in a local area is simulated to obtain updated temperature distribution information. If outliers are found in the updated temperature distribution information, the location and extent of the abnormal area are determined by comparing it with a preset temperature anomaly threshold range, and an anomaly distribution marker is generated.

[0024] Step S13: Readjust the mesh density for abnormal areas to correct the heat diffusion rate, and generate an optimized temperature distribution map and a visualized heat transfer path view.

[0025] By identifying anomalous distributions, the mesh density is readjusted for anomalous regions, and the corrected heat diffusion rate is calculated to obtain an optimized temperature distribution map. Finite element analysis tools are then used to visualize the optimized temperature distribution map, generating an intuitive view of heat transfer paths and temperature distribution.

[0026] Specifically, for the powder particle model, the input parameters are a thermal conductivity of 0.5 W / m Kelvin and an initial ambient temperature of 25.0°C. The Galerkin finite element method is used to solve the heat conduction equation, outputting the initial heat transfer path network and temperature distribution matrix between particles. This process provides accurate baseline data for subsequent local thermodynamic evolution.

[0027] For example, in a localized region where powder is subjected to high shear forces, particle friction is intense. An adaptive meshing operation is performed on this region, refining the original coarse mesh with a side length of 2.0 mm into a tetrahedral micromesh of 0.1 mm. The Delaunay triangulation algorithm is used to determine the spatial coordinates of particles and the locations of contact points within this localized region. Mesh refinement significantly improves the geometric accuracy of the high-stress localized region.

[0028] In one possible implementation, geometric information such as the contact area and radius of curvature of the contact points is extracted from the refined mesh. These geometric features are input into a Fourier heat conduction model, calculating the heat diffusion rate at the contact points to be 15.5 joules per second. Combined with initial temperature distribution data, an implicit backward Euler time integral algorithm is used to simulate the heat diffusion process in a local region, outputting updated temperature distribution information after 0.5 seconds, thereby accurately capturing transient thermal shock effects.

[0029] Specifically, if the updated temperature distribution information contains abnormally high-temperature nodes exceeding the preset temperature anomaly threshold of 85.0℃, then heat accumulation is determined to exist in that area. Anomaly distribution markers are generated using a connected component labeling algorithm. For the marked areas, the mesh density is further increased, the mesh size is adjusted to 0.02 mm, and the corrected heat diffusion rate is recalculated, ultimately resulting in an optimized temperature distribution map that eliminates numerical oscillations. This effectively avoids local temperature distortion caused by excessively large mesh sizes.

[0030] For example, a volume rendering algorithm is used to visualize the optimized temperature distribution map, mapping the temperature gradient to a pseudo-color layer from cool to warm tones, generating an intuitive heat transfer path and temperature distribution view, which facilitates an intuitive assessment of the risk of thermal damage during the powder mixing process.

[0031] S2 simulates and tracks the thermal structural adjustment process of binder molecules, extracts data from areas of concentrated heat, simulates the softening behavior and interweaving dynamics of molecular chains, and determines the threshold for binder morphological transformation.

[0032] Optionally, this step also includes: Step S21: Extract the boundary data of the heat concentration area from the temperature distribution map, segment and label it using image processing technology, and use molecular dynamics simulation to construct an initial model of the binder molecules to determine its initial stress state under thermal action.

[0033] Distribution data of heat concentration areas were extracted from the temperature distribution map. Image processing techniques were used to segment and label the high-heat areas in the map, obtaining the boundary range data of the heat concentration areas. Based on the boundary range data of the heat concentration areas, molecular dynamics simulation methods were used to construct an initial structural model of the binder molecules and determine the initial force state of the molecules under the action of heat.

[0034] Step S22: Simulate the softening behavior of molecular chains under different thermal gradients, obtain displacement and deformation data to identify key inflection points, analyze the interweaving dynamics between molecules, and extract the dynamic distribution of interactions.

[0035] Based on the initial force state of molecules under heat, the softening behavior of molecular chains under different heat gradients is simulated to obtain displacement and deformation data of the molecular chains and identify key inflection points in the softening process. By analyzing the data from these key inflection points, the dynamic changes in interweaving between molecular chains are analyzed, and the location and intensity information of the interweaving points are extracted to obtain the dynamic distribution of intermolecular interactions.

[0036] Step S23: Combining heat distribution data and intermolecular interaction distribution, simulate the evolution process of the overall structure of the binder to determine the critical conditions for morphological transformation, and determine whether to record characteristic information or adjust the heat gradient to re-simulate based on the critical threshold.

[0037] By using dynamic distribution data of intermolecular interactions, combined with distribution data of heat concentration areas, the change process of the overall structure of the binder molecules is simulated to determine the critical conditions for morphological transformation. If the critical conditions for morphological transformation meet the preset critical threshold, the corresponding heat distribution and molecular chain state data are recorded to obtain the final morphological transformation characteristic information; if the preset critical threshold is not met, the heat gradient parameters are adjusted, and the molecular chain softening behavior is re-simulated.

[0038] Specifically, for the heat concentration areas in the temperature distribution map, the Otsu's method is used to binarize the image pixels, segmenting pixels with temperatures above a threshold (e.g., 85.5℃) into high-heat regions. An edge detection algorithm is then used to extract the contour coordinates of these regions, outputting the boundary range data of the heat concentration areas. This processing method can accurately locate the positions of localized hot spots generated by high shear friction between powder particles, providing accurate spatial boundaries for subsequent microstructure evolution analysis.

[0039] For example, based on the aforementioned boundary range data, an initial structural model of the polymer binder is constructed using molecular dynamics simulations. The degree of polymerization of the binder is set to 50, and the van der Waals interactions between molecules are calculated by introducing the Leonard-Jones potential function to determine the force state of the molecules under the initial thermal field. Based on this, a thermal gradient is applied at a rate of 2.0 degrees per picosecond to simulate the softening behavior of the molecular chains. By calculating the mean square displacement data of the molecular chains, their nonlinear growth trend with increasing temperature is monitored. When the slope of the mean square displacement curve changes abruptly, that temperature point is identified as the critical inflection point in the molecular chain softening process, i.e., the glass transition temperature.

[0040] In one possible implementation, molecular trajectory data before and after key inflection points are extracted, and the dynamic changes in interweaving between molecular chains are analyzed using a radial distribution function. When the distance between two molecular chain segments is less than 0.35 nanometers, it is determined to be an effective interweaving point, and the density of interweaving points per unit volume is statistically analyzed as intensity information. Combined with the distribution data of the heat concentration region, the overall structural change of the binder molecules from the glassy state to the elastic state is evaluated. The critical condition for morphological transformation is set as the interweaving point density decreasing to 0.4 times the initial value. If the current simulation results show that the interweaving point density is 0.35 times the initial value, satisfying the preset critical density threshold, the heat distribution and molecular chain conformation data at this time are recorded, and the final morphological transformation characteristic information is output. If the interweaving point density is higher than 0.4 times the initial value, the heat gradient parameter is adjusted in steps of 0.5 degrees per picosecond, and the simulation process of molecular chain softening behavior is repeated until the melt spreading state of the binder in the gaps between powder particles is accurately captured.

[0041] S3. If the threshold for the morphological transformation of the adhesive exceeds the preset upper limit, adjust the simulation parameters to reduce the local heat input, obtain the list of overheated zones, rerun the finite element analysis to balance the heat distribution, and obtain the optimized temperature distribution map.

[0042] Specifically, by extracting records exceeding a preset upper limit from the threshold data of morphological transformation, an initial list of overheated regions is determined. A pre-established heat distribution model is used to analyze the local heat distribution, resulting in a preliminary heat distribution map. If significant heat concentration areas exist in the preliminary heat distribution map, the simulation parameters are adjusted to reduce the heat input, and the adjusted heat distribution data is obtained. The finite element analysis is then rerun to balance the heat distribution in each region, determining the optimized heat distribution result. Temperature distribution information is extracted from the optimized heat distribution result to generate a corresponding temperature distribution map. It is then determined whether any regions still exceed the morphological transformation threshold. If so, local parameters are adjusted again for these regions to obtain the final optimized temperature distribution map. The relevant data is saved to a preset database, completing the heat distribution optimization process.

[0043] In one possible implementation, for the threshold data of morphological transformation of adhesive molecules, records where the local kinetic energy or structural deformation of the molecular chains exceeds a preset upper limit are extracted. The critical temperature threshold for morphological transformation is set to 450.5 Kelvin. When a local temperature of 465.0 Kelvin is detected in a certain micro-region, that spatial coordinate region is marked as an overheated state, and these coordinate data are summarized to determine an initial list of overheated regions.

[0044] For example, the list of overheated regions is used as input data into a pre-established heat distribution model based on Fourier's law of heat conduction. This model employs spatial gridding technology to divide the overall structure of the binder molecules into tiny units with a side length of 2.0 nanometers. It analyzes the heat flux and temperature gradient within each unit, outputting a preliminary heat distribution map. This map allows for precise location of specific spatial locations where heat is excessively concentrated, providing data support for subsequent heat dissipation.

[0045] It should be noted that if there are significant heat concentration areas with heat flux greater than 50.0 W / m² in the initial heat distribution map, the heat input is reduced by adjusting the heat bath parameters in the molecular dynamics simulation. The damping coefficient in the temperature control algorithm is adjusted from 100.0 femtoseconds to 50.0 femtoseconds to obtain the adjusted thermal boundary condition data. Based on the adjusted data, the heat conduction process within the binder matrix is ​​recalculated using a finite element analysis algorithm to balance the heat distribution in each region, outputting the optimized heat distribution results, effectively preventing irreversible degradation of molecular chains due to localized overheating.

[0046] In one possible implementation, temperature distribution information for each node is extracted from the optimized heat distribution results, and a smooth temperature distribution map is generated using a bilinear interpolation algorithm. This temperature distribution map is then used to determine if any regions still exceed 455.0 Kelvin. If so, the thermal conductivity parameters are fine-tuned again for these local meshes, and iterative calculations are performed to obtain the final optimized temperature distribution map. After obtaining the final optimized temperature distribution map, the relevant data, including node coordinates and temperature values, is saved to a pre-defined relational database, completing the optimization process for the adhesive heat distribution.

[0047] S4. Extract the connection strength index and interweaving density data between powder particles, construct a virtual model of a three-dimensional mesh skeleton, and evaluate its stability.

[0048] Specifically, by analyzing the temperature distribution map, the distribution data of the bonding strength between powder particles is obtained, and the range of interparticle forces is determined. Using molecular dynamics simulation results, interweaving density data is extracted to construct a virtual model of a three-dimensional network framework, generating the spatial structure information of the framework.

[0049] In one possible implementation, the corresponding temperature gradient value is extracted as input data for each pixel node in the temperature distribution map. A mean-clustering algorithm is then used to classify the temperature gradients, mapping them to obtain the distribution data of the connection strength between powder particles.

[0050] Specifically, when the local temperature reaches 150.5 degrees Celsius, the particle surface melts or the binder solidifies, at which point the basic value for the bonding strength is set to 45.0 MPa. By calculating the distance between the center points of adjacent particles, if the distance is less than 0.8 micrometers, it is determined that they are within the force range, thus outputting a precise matrix of the interparticle force range.

[0051] For example, molecular dynamics simulations are used to process the aforementioned force range matrix. The initial coordinates and potential energy parameters of the particles are input, and a classical potential function model is used to simulate the attraction and repulsion processes between particles. After 2000 time steps of simulation, the number of effectively connected particle pairs per unit volume is extracted as the interweaving density data.

[0052] Specifically, if the number of effective connection pairs per cubic micrometer reaches 120, the quantification value of the inter-particle interweaving degree is set to 0.85, which directly reflects the bonding state at the microscopic level.

[0053] In one possible implementation, a spatial triangulation algorithm is used to construct a virtual model of a three-dimensional mesh skeleton based on the extracted interweaving density data. The input is the centroid coordinates of each particle, and the output is the spatial structure information of the skeleton, including nodes and connected edges.

[0054] S5. If the stability of the mesh skeleton does not meet the standard, the adhesive distribution is iteratively optimized, weak connection points are extracted and incorporated into the finite element feedback, and an enhanced mesh skeleton model is generated.

[0055] Optionally, this step also includes: Step S51: If the stability value of the mesh skeleton does not meet the standard, locate the weak links and obtain the stress distribution information of the key areas, start the iterative optimization process to adjust the adhesive distribution scheme, and obtain the preliminary optimized distribution parameters.

[0056] By extracting stability values ​​from the initial data of the mesh skeleton, and using automated tools to compare them with a preset stability threshold, it is determined whether the stability requirements are met. If the stability value is lower than the stability threshold, the weak links in the mesh skeleton are located, stress distribution information of key areas is obtained from the evaluation data, the specific location of the weak links is determined, an iterative optimization process is initiated, the adhesive distribution scheme is adjusted, and preliminary optimized distribution parameters are obtained.

[0057] Step S52: Integrate the initially optimized distribution parameters into the finite element analysis tool, simulate the stress response of the mesh skeleton under different loads to obtain improvement suggestions, and then update the adhesive distribution scheme and build an enhanced model. If the stability standard is still not met, repeat the adjustment and finite element analysis until the standard is met.

[0058] Based on the initially optimized distribution parameters, finite element analysis (FEM) tools are incorporated to simulate the stress response of the mesh skeleton under different loads. Improvement suggestions from the feedback mechanism are obtained, the binder distribution scheme is updated, and a reinforced mesh skeleton model is constructed. The model is then assessed to determine if it meets the preset stability criteria. If the reinforced mesh skeleton model still does not meet the criteria, new weak link data are extracted from the feedback mechanism, and adjustment cycles and FEM analyses are repeatedly executed to obtain the final optimized structural model.

[0059] Step S53: Generate the corresponding parameter configuration scheme based on the final optimized structural model to determine the stability and reliability of the mesh skeleton in practical applications.

[0060] In one possible implementation, for the initial data of the mesh skeleton, the coordinate set of the skeleton nodes and the connection matrix are input, and the principal component analysis algorithm is used to extract the eigenvalues ​​of the global stiffness matrix of the skeleton as stability values.

[0061] Specifically, a preset stability threshold of 0.85 is set. When the extracted stability value is only 0.72, it is determined that the requirement is not met. At this time, the nodal force vectors in the evaluation data are extracted, and the stress distribution information of the key area is obtained by calculating the deformation gradient between adjacent nodes.

[0062] For example, if the deformation gradient of a local area exceeds the deformation gradient threshold of 0.05, the nodes in that area are marked as weak links, and their specific three-dimensional coordinate range is output.

[0063] In one possible implementation, an iterative optimization process is initiated based on the three-dimensional coordinate range of the aforementioned weak links. The current binder volume fraction distribution matrix is ​​input, and a gradient descent algorithm is used to locally increase the binder concentration in the weak regions, yielding initially optimized distribution parameters.

[0064] Specifically, the volume fraction of the adhesive in the weak areas was adjusted from 12% to 18%. Subsequently, the preliminarily optimized distribution parameters were input into the finite element analysis tool, the externally applied uniaxial compressive load was set to 50 MPa, and a linear elastic constitutive model was used to simulate the stress response of the mesh skeleton.

[0065] For example, the equivalent stress value of each element is output by the finite element solver. If it is found that the peak equivalent stress of the original weak area has decreased from 85 MPa to 60 MPa, but a new stress concentration appears in the adjacent area, this phenomenon is output as an improvement suggestion in the feedback mechanism.

[0066] In one possible implementation, the adhesive distribution scheme is updated based on the stress concentration data of adjacent regions output by the feedback mechanism, and the enhanced mesh skeleton model is reconstructed using an adaptive mesh refinement algorithm.

[0067] Specifically, if the enhanced model still has a global maximum equivalent stress higher than the standard value of 70 MPa under a 50 MPa load, then new high-stress node data are extracted, and the cycle of binder concentration allocation and finite element solution is executed again.

[0068] For example, after four iterations, the global maximum equivalent stress drops to 65 MPa. At this point, the iteration stops, and the final optimized structural model is output. Based on this model, a parameter configuration scheme containing the precise proportions of the binder in each region is generated, and the load-bearing limit data of the mesh skeleton in practical applications is determined.

[0069] S6. Based on the enhanced mesh skeleton model, stress distribution data is extracted, deformation response under processing conditions is simulated, and the degree of improvement in the mechanical properties of the electrode material is evaluated.

[0070] Specifically, based on the enhanced mesh skeleton model, mechanical property data of the electrode material is obtained. Considering the structural characteristics of the enhanced model, the distribution range of overall parameters is determined. Finite element analysis is used to extract stress distribution data, and detailed characteristics of stress concentration regions are obtained under different loading conditions in the processing environment. Key points of deformation response are acquired, and the degree of matching between the deformation trend and the actual environment is determined by combining simulation parameters of the processing environment. Based on the degree of matching of the deformation trend, quantitative data of performance improvement is obtained. Through the simulation analysis module, the specific range of improvement is determined, and core factors affecting mechanical performance are extracted. Preset thresholds are used for comparison. If the core factors exceed the threshold, the boundary conditions of the simulation analysis are adjusted to obtain optimized performance data. The correlation mapping between deformation response and stress distribution is obtained, and the final evaluation range is determined through the data extraction module. If the final evaluation range does not meet the preset standard, the processing environment parameters are reloaded through the information processing stage, and combined with the structural characteristics of the mesh skeleton, corrected mechanical property parameters are obtained.

[0071] In one possible implementation, for the constructed electrode material mesh skeleton model, initial basic data such as porosity and Young's modulus are input to determine the distribution range of the overall parameters.

[0072] For example, the porosity distribution range is set to 30% to 50%, and the skeleton wire diameter is set to 2 micrometers to 5 micrometers. Based on the above parameter range, an implicit dynamic finite element analysis algorithm is used to simulate the stress state of the electrode sheet under the rolling processing environment. After applying a uniformly distributed load of 100 MPa in the vertical direction, the stress distribution data inside the mesh skeleton is extracted, the stress concentration area at the skeleton intersection node is located, and the detailed characteristics of the maximum principal stress reaching 150 MPa in this area are obtained.

[0073] For example, from the characteristics of the stress concentration areas mentioned above, the skeleton intersection nodes are extracted as key points of deformation response. Combined with simulation parameters of the actual rolling processing environment, the root mean square error algorithm is used to calculate the deviation between the simulated displacement and the actual physical deformation of the key points. When the calculated displacement deviation is less than 0.01 mm, it is determined that the deformation trend has a high degree of matching with the actual processing environment. Based on this matching degree, the performance improvement is quantified through the simulation analysis module, determining that the overall yield strength improvement is within the specific range of 10% to 15%.

[0074] It should be noted that local strain rate is extracted as the core factor influencing mechanical performance from the specific range of the improvement range. The extracted local strain rate is compared with a preset strain rate threshold of 0.02. If the strain rate of a node reaches 0.03, exceeding the strain rate threshold, the boundary condition adjustment mechanism is triggered. By reducing the virtual loading rate or changing the bottom surface fixed constraint conditions, the finite element method is re-solved to obtain optimized local stress and strain data. A support vector regression algorithm is used to establish a nonlinear correlation mapping between the optimized deformation response and stress distribution, and the final structural strength assessment range is calculated. If the assessment range does not reach the preset 12% standard, the initial rolling pressure parameter is reloaded through the information processing stage, lowered to 80 MPa, and the calculation is re-iterated based on the topological characteristics of the mesh skeleton to obtain the corrected mechanical performance parameters.

[0075] S7: Obtain the final set of process parameters, extract key control variables from the mechanical property parameters of the electrode material, and output an optimized scheme for dry electrode manufacturing.

[0076] Specifically, based on the performance improvement assessment, a preliminary set of process parameters is constructed. A data filtering module is used to compare the distribution range of key control variables with preset process thresholds. If a variable exceeds the threshold, the parameter weights are adjusted through information processing to determine the optimized parameter set. For the optimized parameter set, the core control points in the dry electrode manufacturing process are identified. A data mapping tool is used to obtain a preliminary framework for the manufacturing scheme. From this preliminary framework, constraints related to the manufacturing process are extracted. An information processing module performs condition matching to determine the applicability of the scheme description, obtains dynamic adjustment factors in the manufacturing process, and uses data analysis tools to determine the final structure of the optimized scheme. This generates a complete process description for dry electrode manufacturing. An information integration step yields detailed results analysis, and an output template for the manufacturing scheme is constructed. Finally, a data storage module determines the storage format of the final scheme.

[0077] In one possible implementation, a preliminary set of process parameters for dry electrode manufacturing is constructed based on previously acquired electrode material deformation response data. This set includes fundamental parameters such as rolling pressure, rolling speed, and binder mixing ratio. Principal component analysis is then used to filter the data in the preliminary set, extracting the key control variables that have the greatest impact on the stability of the mesh skeleton structure.

[0078] For example, the roller pressing pressure is identified as a key control variable, with its distribution range set to 15.0 MPa to 25.0 MPa. The extracted distribution range is compared with a preset process threshold. If the currently set roller pressing pressure reaches 26.0 MPa, exceeding the preset pressure upper limit threshold, the information processing step is triggered. At this point, the parameter weights are reallocated using the analytic hierarchy process (AHP), decreasing the weight coefficient of roller pressing pressure from 0.6 to 0.4 while increasing the weight coefficient of roller pressing speed from 0.3 to 0.5, thereby determining the optimized parameter set.

[0079] For example, for the optimized parameter set, core control points in the dry electrode manufacturing process are extracted, such as the degree of powder fiberization and the uniformity of electrode thickness. Support vector regression is used as a data mapping tool to associate the core control points with equipment operating parameters, generating a preliminary framework for the manufacturing scheme. Constraints related to the actual processing environment are extracted from this preliminary framework, such as ambient humidity below 5.0% and maximum equipment operating power limits. An information processing module matches these constraints with the preliminary framework to determine the applicability of the scheme in the current workshop environment. If a match is successful, dynamic adjustment factors in the manufacturing process are introduced, such as the temperature rise rate during equipment operation. A time series analysis algorithm is used to process the temperature rise rate, determining that when the temperature rise rate exceeds 2.0 degrees Celsius per minute, the roller speed is automatically reduced, thus determining the final structure of the optimized scheme.

[0080] It should be noted that, based on the final structure of the optimized scheme, all the aforementioned control variables and dynamic adjustment rules are integrated to generate a complete process description for dry electrode manufacturing. This description details the parameter settings for each stage, from powder mixing to final roll forming. A standardized manufacturing scheme output template is constructed based on the detailed results analysis. This template includes fields such as parameter names, set values, weighting coefficients, and dynamic adjustment trigger conditions. The completed output template is converted into a specific storage format using a data storage module and persistently saved using a relational database structure, ensuring that subsequent processing stages can accurately read and execute the manufacturing scheme.

[0081] The above-disclosed embodiments are merely preferred embodiments of the present invention and should not be construed as limiting the scope of the invention. Those skilled in the art will understand that implementing all or part of the above-described embodiments and making equivalent changes in accordance with the claims of the present invention are still within the scope of the invention.

Claims

1. A method for morphological transformation and network structure optimization of an adhesive, characterized in that, The method includes: S1, based on finite element analysis, simulates the heat transfer path of powder particles, obtains the initial temperature distribution, refines the mesh in the high shear zone, calculates the heat diffusion rate at the particle contact point, and generates a temperature distribution map. S2 simulates the process of binder molecules adjusting their structure under heat, extracts data from areas of concentrated heat, simulates the softening behavior and intertwining dynamics of molecular chains, and determines the threshold for binder morphological transformation. S3. If the threshold for the morphological transformation of the adhesive exceeds the preset upper limit, adjust the simulation parameters to reduce the local heat input, obtain the list of overheated zones, rerun the finite element analysis to balance the heat distribution, and obtain the optimized temperature distribution map. S4. Extract the connection strength index and interweaving density data between powder particles, construct a virtual model of a three-dimensional mesh skeleton and evaluate its stability. S5. If the stability of the mesh skeleton does not meet the standard, the adhesive distribution is iteratively optimized, weak connection points are extracted and incorporated into the finite element feedback to generate an enhanced mesh skeleton model. S6. Based on the enhanced mesh skeleton model, stress distribution data is extracted, deformation response under processing conditions is simulated, and the degree of improvement in the mechanical properties of the electrode material is evaluated. S7: Obtain the final set of process parameters, extract key control variables from the mechanical property parameters of the electrode material, and output an optimized scheme for dry electrode manufacturing.

2. The method according to claim 1, characterized in that, Step S1 includes: Step S11: Based on the preset powder particle model and finite element analysis, simulate and obtain the initial heat transfer path and temperature distribution data between particles, and perform mesh refinement for the high shear force local area to extract the contact point geometric information to calculate the local heat transfer rate distribution. Step S12: Combine the initial temperature data to simulate the heat diffusion process, obtain updated temperature distribution information, identify the location and range of abnormal areas, and generate abnormal distribution markers; Step S13: Readjust the mesh density for abnormal areas to correct the heat diffusion rate, and generate an optimized temperature distribution map and a visualized heat transfer path view.

3. The method according to claim 2, characterized in that, Step S11 includes: for the powder particle model, inputting the thermal conductivity of the particles and the initial ambient temperature, solving the heat conduction equation using the Galerkin finite element method, and outputting the initial heat transfer path network and temperature distribution matrix between the particles.

4. The method according to claim 3, characterized in that, Step S11 further includes: determining the spatial coordinates of particles and the location of contact points within a local area using the Delaunay triangulation algorithm.

5. The method according to claim 2, characterized in that, Step S12 includes: if there are abnormally high temperature nodes in the updated temperature distribution information that exceed the temperature anomaly threshold, it is determined that there is heat accumulation in the area, and an abnormal distribution marker is generated by the connected component marking algorithm.

6. The method according to claim 1, characterized in that, Step S2 includes: Step S21: Extract the boundary data of the heat concentration area from the temperature distribution map, segment and label it using image processing technology, and use molecular dynamics simulation to construct an initial model of the binder molecules to determine its initial stress state under thermal action. Step S22: Simulate the softening behavior of molecular chains under different thermal gradients, obtain displacement and deformation data to identify key inflection points, analyze the interweaving dynamics between molecules, and extract the dynamic distribution of interactions. Step S23: Combining heat distribution data and intermolecular interaction distribution, simulate the evolution process of the overall structure of the binder to determine the critical conditions for morphological transformation, and determine whether to record characteristic information or adjust the heat gradient to re-simulate based on the critical threshold.

7. The method according to claim 6, characterized in that, Step S21 includes: based on the heat concentration area in the temperature distribution map, performing binarization processing on the image pixels using the maximum inter-class variance method, segmenting the pixels with temperatures higher than the segmentation threshold into high-heat areas, extracting the contour coordinates of the area using an edge detection algorithm, and outputting the boundary range data of the heat concentration area.

8. The method according to claim 1, characterized in that, Step S5 includes: Step S51: If the stability value of the mesh skeleton does not meet the standard, locate the weak links and obtain the stress distribution information of the key areas, start the iterative optimization process to adjust the adhesive distribution scheme, and obtain the preliminary optimized distribution parameters. Step S52: Integrate the initially optimized distribution parameters into the finite element analysis tool, simulate the stress response of the mesh skeleton under different loads to obtain improvement suggestions, and then update the adhesive distribution scheme and build the enhanced model. If the stability standard is still not met, repeat the adjustment and finite element analysis until the standard is met. Step S53: Generate the corresponding parameter configuration scheme based on the final optimized structural model to determine the stability and reliability of the mesh skeleton in practical applications.

9. The method according to claim 8, characterized in that, Step S51 includes: inputting the coordinate set and connection matrix of the mesh skeleton nodes, and using principal component analysis algorithm to extract the global stiffness matrix eigenvalues ​​of the skeleton as stability values.

10. The method according to claim 9, characterized in that, Step S51 further includes: if the deformation gradient of a local region exceeds the deformation gradient threshold, then the nodes in that region are marked as weak links, and their specific three-dimensional coordinate range is output.

Images