Comprehensive optimization method and system for thermal conductive plastic performance

Abstract

The invention relates to a comprehensive optimization method and system for thermal conductive plastic performance in the technical field of new materials, and the method comprises the following steps: obtaining initial dispersion state data of an inorganic filler in thermal conductive plastic, and analyzing spatial distribution of filler particles through a scanning electron microscope to obtain characteristic parameters of random distribution of the filler; according to the characteristic parameters of the random distribution of the filler, calculating the heat transfer efficiency among filler particles by adopting a finite element simulation method, and determining the distribution rule of the phonon scattering phenomenon; according to the improved interface bonding strength parameters, the Monte Carlo method is adopted to simulate orientation distribution of filler particles in a matrix, and a preliminary scheme of filler orientation control is obtained; for optimization parameters of complex mold forming, a particle swarm optimization algorithm is adopted to adjust the filler proportion and distribution, and the structure configuration of the efficient three-dimensional heat conduction network is generated; and according to the structure configuration of the efficient three-dimensional heat-conducting network, the improvement degree of the heat-conducting plastic performance is analyzed and verified through a heat-conducting finite element, and final heat transfer efficiency data is obtained.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of new materials, in particular to high-performance composite materials, and more particularly to a method and system for comprehensive optimization of the performance of thermally conductive plastics. BACKGROUND

[0002] High-molecular composite materials are crucial in the field of thermal management, especially in high-performance scenarios such as electronic devices, 5G communication, and new energy vehicles. Thermally conductive plastics are ideal choices due to their lightweight, corrosion resistance, and high processability.

[0003] However, the performance of existing thermally conductive plastics is far from meeting actual needs, limiting their application in high-end heat dissipation modules. Current solutions often rely on high proportions of inorganic fillers to improve thermal conductivity, but this often leads to a significant decrease in material processing fluidity, making it difficult to meet complex mold forming requirements. At the same time, the dispersion state of fillers in the matrix makes it difficult to form an efficient thermal conduction network, and the heat transfer efficiency is low, making it difficult to cope with the heat dissipation needs of high-power devices. The limitations of existing methods lie in the weak interfacial bonding between fillers and the matrix, with fillers particles randomly distributed, making it difficult to form a continuous thermal conduction path. Surface modification can slightly improve interfacial affinity, but the improvement is limited, and the process is complex and costly. The disorder of filler distribution further exacerbates phonon scattering, hindering heat transfer. For example, in 5G base station heat dissipation modules, the high thermal resistance of traditional thermally conductive plastics often causes the chip temperature to exceed the safety threshold, affecting device stability and lifespan. The core challenge lies in two interrelated technical factors. First, the orientation and connection of fillers in the matrix are difficult to control, and randomly stacked filler particles form "thermal islands" that cannot build an efficient three-dimensional thermal conduction network. Second, the presence of interfacial thermal resistance further weakens the heat transfer efficiency, and the lack of strong interfacial interaction between fillers and the matrix leads to severe phonon scattering at the interface. For example, in high-power electronic devices, thermally conductive plastics need to quickly transfer heat from the chip to the heat sink, but disordered fillers and weak interfacial bonding cause heat accumulation, leading to local overheating.

[0004] Therefore, how to optimize filler orientation and interfacial forces to build an efficient three-dimensional thermal conduction network while maintaining material processing fluidity has become a key issue in the development of high-thermal-conductivity plastics. SUMMARY

[0005] The present application provides a method for comprehensive optimization of the performance of thermally conductive plastics, comprising the following steps:

[0006] Obtain the initial dispersion state data of inorganic fillers in the thermally conductive plastic, analyze the spatial distribution of filler particles by scanning electron microscopy, and obtain the characteristic parameters of random distribution of fillers;

[0007] According to the characteristic parameters of the random distribution of the filler, the heat transfer efficiency between the filler particles is calculated by using the finite element simulation method, and the distribution rule of the phonon scattering phenomenon is determined;

[0008] The key area is extracted from the distribution rule of the phonon scattering phenomenon, the interfacial bonding strength between the filler and the matrix is analyzed by using the molecular dynamics simulation, and the quantitative data of the interfacial thermal resistance is obtained;

[0009] According to the quantitative data of the interfacial thermal resistance, the interface modification molecular structure is preset, the filler surface modification process is optimized by chemical bonding simulation, and the improved interfacial bonding strength parameters are obtained;

[0010] According to the improved interfacial bonding strength parameters, the orientation distribution of the filler particles in the matrix is simulated by using the Monte Carlo method, and a preliminary scheme of the filler orientation control is obtained;

[0011] The high-probability orientation path is extracted from the preliminary scheme of the filler orientation control, the influence of the processing fluidity reduction is analyzed in combination with the fluid mechanics model, and the optimization parameters of the complex mold forming are determined;

[0012] According to the optimization parameters of the complex mold forming, the particle swarm optimization algorithm is used to adjust the proportion and distribution of the filler, and the structure configuration of the high-efficiency three-dimensional heat conduction network is generated;

[0013] According to the structure configuration of the high-efficiency three-dimensional heat conduction network, the improvement degree of the heat conduction plastic performance is verified by the heat conduction finite element analysis, and the final heat transfer efficiency data is obtained.

[0014] The application provides a comprehensive optimization system for the performance of heat conduction plastic, mainly comprising:

[0015] The filler dispersion state acquisition module is used for acquiring the initial dispersion state data of the inorganic filler in the heat conduction plastic, analyzing the spatial distribution of the filler particles by using a scanning electron microscope, and obtaining the characteristic parameters of the random distribution of the filler;

[0016] The heat transfer efficiency calculation module is used for calculating the heat transfer efficiency between the filler particles according to the characteristic parameters of the random distribution of the filler, and determining the distribution rule of the phonon scattering phenomenon by using the finite element simulation method;

[0017] The phonon scattering analysis module is used for extracting the key area from the distribution rule of the phonon scattering phenomenon, analyzing the interfacial bonding strength between the filler and the matrix by using the molecular dynamics simulation, and obtaining the quantitative data of the interfacial thermal resistance;

[0018] The interfacial thermal resistance quantification module is used for presetting the interface modification molecular structure according to the quantitative data of the interfacial thermal resistance, optimizing the filler surface modification process by chemical bonding simulation, and obtaining the improved interfacial bonding strength parameters;

[0019] The interface modification and optimization module is used to simulate the orientation distribution of filler particles in the matrix using the Monte Carlo method based on the improved interface and strength parameters, so as to obtain a preliminary scheme for filler orientation control.

[0020] The orientation distribution simulation module is used to extract high-probability orientation paths from the initial scheme of filler orientation control, and combine it with the fluid dynamics model to analyze the impact of decreased processing fluidity and determine the optimization parameters for complex mold forming.

[0021] The molding parameter optimization module is used to optimize the molding parameters of complex molds. It uses a particle swarm optimization algorithm to adjust the filler ratio and distribution, and generates a structural configuration of a highly efficient three-dimensional heat conduction network.

[0022] The thermal conductivity network verification module is used to verify the degree of improvement in the performance of thermally conductive plastics through finite element analysis of thermal conductivity based on the structural configuration of the high-efficiency three-dimensional thermal conductivity network, and to obtain the final heat transfer efficiency data.

[0023] The technical solutions provided by the embodiments of the present invention may include the following beneficial effects:

[0024] This invention discloses a comprehensive optimization method for the properties of thermally conductive plastics. Addressing key issues such as the dispersion, interfacial bonding, and heat transfer efficiency of inorganic fillers in a matrix, it constructs a highly efficient thermally conductive network through multi-scale simulation and optimization design. First, the invention analyzes the initial dispersion state of the fillers using scanning electron microscopy, extracting randomly distributed characteristic parameters. Then, it combines finite element method (FEM) simulation to quantify the heat transfer efficiency and phonon scattering patterns between fillers, clarifying the influence of interfacial thermal resistance on thermal conductivity. To address the interfacial thermal resistance issue, this invention optimizes the filler surface modification process through molecular dynamics simulation to improve interfacial bonding strength. It also employs the Monte Carlo method to simulate the filler orientation distribution and combines fluid dynamics analysis to determine processing flowability and mold forming parameters for complex molds. Finally, it adjusts the filler ratio and distribution using a particle swarm optimization algorithm to construct a highly efficient three-dimensional thermally conductive network. Finite element analysis of heat conduction confirms a significant improvement in heat transfer efficiency. This invention achieves comprehensive optimization of the properties of thermally conductive plastics through the fusion of multiple methods, providing an innovative solution for the design and fabrication of high-performance thermally conductive composite materials. Attached Figure Description

[0025] Figure 1 This is a flowchart of a comprehensive optimization method for the thermally conductive plastic properties according to the present invention.

[0026] Figure 2 This is a schematic diagram of a comprehensive optimization method for the thermally conductive plastic properties according to the present invention.

[0027] Figure 3 This is another schematic diagram of a comprehensive optimization method for the thermally conductive plastic properties according to the present invention.

[0028] Figure 4This is a schematic diagram of the framework of a comprehensive optimization system for the thermally conductive plastic properties according to the present invention. Detailed Implementation

[0029] To further understand the content of this invention, a detailed description of the invention is provided in conjunction with the accompanying drawings and embodiments. The specific embodiments described herein are for illustrative purposes only and are not intended to limit the invention. It should also be noted that, for ease of description, only the parts relevant to the invention are shown in the accompanying drawings.

[0030] like Figures 1-4 The comprehensive optimization method for the thermally conductive plastic properties in this embodiment specifically includes the following steps:

[0031] Step S101: Obtain the initial dispersion state data of inorganic fillers in thermally conductive plastics, and analyze the spatial distribution of filler particles by scanning electron microscopy to obtain the characteristic parameters of random distribution of fillers.

[0032] Initial dispersion images of inorganic fillers in thermally conductive plastics are acquired using a scanning electron microscope to obtain first image data. The first image data is then denoised using an image processing algorithm to obtain second image data. Contour information of the filler particles is extracted from the second image data to determine particle size and spacing. The spatial distribution density is calculated based on the particle size and spacing to obtain distribution density data. If the distribution density data meets a preset uniformity threshold, random distribution feature parameters are extracted through statistical analysis to obtain a first feature parameter set. For the first feature parameter set, K-means clustering is used to classify spatial uniformity to obtain classification results. Based on the classification results, comprehensive feature parameters for spatial uniformity and random distribution are generated to obtain the final feature parameter set.

[0033] For example, to acquire initial dispersion data of inorganic fillers in thermally conductive plastics using scanning electron microscopy (SEM), the sample must first be prepared and imaged. A thermally conductive plastic sample is selected, assuming it is a polypropylene matrix doped with 30 wt% alumina (Al2O3) filler. The sample is subjected to liquid nitrogen cryogenic fracture, and the cross-section is sputter-coated with gold to enhance conductivity. Using SEM, the sample cross-section is scanned at an accelerating voltage of 15 kV and a magnification of 500x, acquiring 10 images with a resolution of 2048 × 2048 pixels, each covering an area of ​​200 μm × 200 μm. After image grayscale processing, the Otsu thresholding algorithm is used to separate the filler particles from the matrix. A grayscale threshold of 120 is set, and the contours of the filler particles are extracted. The particle center coordinates and area are calculated. It is assumed that 500 particles are identified per image, with an average particle diameter of 5 μm and a standard deviation of 1.2 μm. To characterize the spatial distribution of the packing material, the nearest neighbor distance between particle centers was calculated, yielding an average nearest neighbor distance of 8.5 μm and a standard deviation of 2.3 μm. Further analysis of spatial randomness was performed using Ripley's K function, with a search radius r ranging from 2 μm to 20 μm and a step size of 1 μm. The K(r) value was calculated and compared with a Poisson random distribution model. It was found that K(r) deviated from the random distribution by approximately 15% when r = 10 μm, indicating slight aggregation of the packing material. Based on this, the Voronoi segmentation algorithm was used to divide the image into Voronoi units with particle centers as nodes, and the area of ​​each unit was calculated, with an average value of 85 μm. 2 The standard deviation is 25 μm. 2 This reflects the non-uniformity of the filler distribution. A two-dimensional distribution map of the filler volume fraction was constructed by combining particle area and Voronoi unit area. The volume fraction ranged from 25% to 35%, with an average value of 30.2%, which is consistent with the initial doping ratio of 30 wt%, verifying the accuracy of the analysis. This method, through image processing and spatial statistical algorithms, achieves quantitative characterization of the filler dispersion state, providing data support for optimizing the properties of thermally conductive plastics.

[0034] Step S102: Based on the characteristic parameters of the random distribution of the packing, the heat transfer efficiency between the packing particles is calculated using the finite element simulation method to determine the distribution law of phonon scattering phenomenon.

[0035] Using the finite element method (FEM), random distribution data of filler particles is obtained to generate a three-dimensional geometric model, determining the particle spacing and interface contact state. Based on the three-dimensional geometric model, the thermal conductivity distribution of the filler particles is calculated to obtain the interparticle heat transfer efficiency. If the heat transfer efficiency is lower than a preset threshold, the particle spacing and interface contact parameters are adjusted, and the three-dimensional geometric model is regenerated. A phonon scattering model is used to analyze the regenerated three-dimensional geometric model to obtain the spatial distribution data of phonon scattering. Using the Monte Carlo algorithm, the probability density of phonon scattering is calculated based on the spatial distribution data to determine the distribution law. Based on the probability density of phonon scattering, the material properties and size distribution of the filler particles are optimized to generate an optimized three-dimensional geometric model. The optimized three-dimensional geometric model is verified through finite element simulation to obtain the final heat transfer efficiency and phonon scattering distribution law.

[0036] For example, based on the characteristic parameters of the randomly distributed filler particles, the finite element method is used to calculate the heat transfer efficiency between filler particles and determine the distribution law of phonon scattering. This can be achieved through the following specific implementation method. First, a random distribution model of the filler particles is constructed. The Monte Carlo algorithm is used to generate the particle positions. Assuming the filler particles are spherical with a diameter of 0.5 micrometers and a volume fraction of 20%, 1000 particles are randomly distributed within a 10-micrometer × 10-micrometer × 10-micrometer cubic region. The minimum distance between particles is 0.1 micrometers to avoid overlap. Periodic boundary conditions are used to ensure the realism of the simulation. Next, based on the finite element method, a heat transfer model is established using COMSOL Multiphysics software. The material thermal conductivity is set to 1 W / (m·K), the inter-particle contact thermal resistance is 10^-8 m^2·K / W, the ambient temperature is 300K, and a heat flux boundary condition of 1 W / m^2 is applied. Tetrahedral meshes are used, with an average mesh size of 0.05 micrometers to ensure calculation accuracy. The heat transfer efficiency was obtained by calculating the ratio of heat flux density to temperature gradient at the particle interface, using the formula η = q / (k·ΔT), where q is the heat flux density, k is the thermal conductivity, and ΔT is the temperature difference. Simulation results showed an average heat transfer efficiency of 0.85 and a standard deviation of 0.12, indicating that the non-uniform particle distribution led to efficiency fluctuations. To analyze the phonon scattering distribution, a molecular dynamics simulation phonon scattering model was used, assuming a phonon mean free path of 100 nm. Combined with temperature field data from finite element simulation, the phonon scattering rate was calculated using the formula Γ = 1 / (λ·v), where λ is the mean free path and v is the phonon velocity (set to 2000 m / s). The results showed that the phonon scattering rate reached as high as 10^12 s^-1 in densely populated regions, while decreasing to 10^11 s^-1 in sparse regions. Analysis showed that when the particle spacing was less than 0.2 μm, phonon scattering significantly increased, leading to a decrease in heat transfer efficiency of approximately 15%. To optimize computation, the parallel computing framework MPI was used to process large-scale grid data, reducing the computation time from 12 hours to 3 hours. Finally, by comparing the simulation results with different volume fractions (10%, 20%, and 30%), it was found that when the volume fraction increased to 30%, the heat transfer efficiency decreased to 0.75, while the phonon scattering rate increased by 20%, indicating that particle aggregation has a significant impact on heat transfer and phonon scattering.

[0037] Step S103: Extract key regions from the distribution law of phonon scattering phenomenon, and use molecular dynamics simulation to analyze the interfacial bonding strength between the filler and the matrix to obtain quantitative data of interfacial thermal resistance.

[0038] The scattering intensity distribution is obtained from phonon scattering data. Fourier transform is used to perform frequency domain analysis on this distribution to obtain the frequency characteristics of key regions. Based on these frequency characteristics, a density clustering algorithm is used to divide the key regions, identifying core regions with high scattering intensity. For these core regions, a molecular dynamics simulation model is constructed to obtain the atomic interaction forces between the filler and the matrix, and the interfacial bonding strength is calculated. From the interfacial bonding strength data, the heat transfer process is simulated using finite element analysis to obtain a preliminary distribution of interfacial thermal resistance. If the variance of the interfacial thermal resistance distribution exceeds a preset threshold, the molecular dynamics simulation parameters are adjusted, and the interfacial bonding strength is recalculated to obtain an optimized interfacial thermal resistance distribution. The optimized interfacial thermal resistance distribution is statistically analyzed using the Monte Carlo method to obtain quantitative data on the interfacial thermal resistance. Based on the quantitative data, an interpolation algorithm is used to generate a spatial distribution map of the thermal resistance, determining the microscopic laws governing thermal conductivity.

[0039] For example, key regions can be extracted from the distribution patterns of phonon scattering phenomena. This can be achieved by processing lattice dynamics data using the molecular dynamics simulation software LAMMPS. Assuming the system is a Si / Ge interface, phonon scattering data is calculated from the phonon density states (DOS) within the 10^12 Hz frequency range. A Fast Fourier Transform (FFT) algorithm is used to convert the time-domain vibrational signals to the frequency domain, extracting the high-scattering region of 0.1-1 THz. Specifically, this involves performing an FFT on the atomic vibrational trajectories over 10,000 time steps (1 fs per step) to obtain the phonon distribution map, and then filtering for scattering intensities greater than 0.5 W / m². 2 The K region was designated as the critical region. Next, the interfacial bonding strength between the filler and the matrix was analyzed using molecular dynamics simulations. The filler was set to SiO2 particles with a diameter of 5 nm, the matrix to be PMMA, and the simulation chamber size to be 10 × 10 × 10 nm. 3 It contains approximately 5000 atoms. The SiO2-PMMA interaction was described using the embedded atom method (EAM) potential function. The interfacial binding energy was calculated. A strain rate of 0.01 nm / ps was applied through tensile simulation, and the maximum stress before interfacial fracture was recorded as 2.5 GPa, with a binding energy of [missing value]. Interfacial thermal resistance was quantified using nonequilibrium molecular dynamics (NEMD) calculations. With the heat flux direction set along the z-axis and temperature gradients of 300 K and 350 K applied, the heat flux density was 10^9 W / m² after 100 ps of simulation. 2 The interfacial thermal resistance is calculated as R = ΔT / J = (350-300) / 10^9 = 5 × 10^-8 m 2K / W. The analysis process, combined with the Kapitza thermal resistance model, verifies the negative correlation between thermal resistance and interfacial binding energy; the higher the binding energy, the lower the thermal resistance. Through the above steps, a logical chain is formed from phonon scattering distribution to interfacial thermal resistance. The phonon scattering region guides the selection of the simulation system, and the intensity data supports the rationality of the thermal resistance calculation. The quantified thermal resistance results can be further used to optimize the filler-matrix interface design.

[0040] Step S104: Based on the quantitative data of interfacial thermal resistance, a pre-defined interfacial modification molecular structure is established, and the filler surface modification process is optimized through chemical bonding simulation to obtain improved interfacial bonding strength parameters.

[0041] Molecular structure data related to interfacial thermal resistance are obtained from a pre-defined molecular structure database. Molecular dynamics simulations are used to determine the chemical bonding characteristics of the filler surface. Based on these chemical bonding characteristics, a filler surface modification model is constructed. Molecular interaction parameters of the filler surface modification model are obtained through molecular dynamics simulations. If the molecular interaction parameters meet a pre-defined chemical bonding threshold, response surface methodology is used to optimize the surface modification process, resulting in an optimized set of process parameters. Based on the optimized set of process parameters, chemical bonding strength data of the filler surface is obtained. Improved interfacial bonding strength parameters are determined through molecular dynamics simulations. Based on these interfacial bonding strength parameters, a quantitative model of interfacial thermal resistance is constructed, obtaining correlation data between thermal resistance parameters and bonding strength. A support vector machine algorithm is used to determine the degree of optimization of the interfacial thermal resistance in the correlation data, resulting in improved interfacial thermal resistance parameters. Based on the improved interfacial thermal resistance parameters, the final filler surface modification scheme is determined, yielding optimized interfacial bonding performance data.

[0042] For example, the quantitative data for interfacial thermal resistance are first obtained using molecular dynamics simulation software.

[0043] LAMMPS was used to quantitatively analyze the thermal resistance at the filler-matrix interface. The filler was set as carbon nanotubes (CNTs), and the matrix as epoxy resin. The simulation system contained 1000 carbon atoms and 5000 epoxy resin molecules at a temperature of 300 K. The Tersoff force field was used to describe the interactions between CNT atoms, and the OPLS-AA force field was used to describe the epoxy resin molecules. The interfacial thermal resistance was calculated using a non-equilibrium molecular dynamics method. A heat flux (10^9 W / m^2) was applied, and the temperature difference ΔT (e.g., 5.2 K) across the interface was recorded to calculate the thermal resistance.

[0044] R = ΔT / q, yielding R = 2.5 × 10⁻⁸ m²·K / W. Analysis indicates that the thermal resistance mainly originates from the vibrational mismatch at the interface. Next, a pre-defined interface modification molecular structure was established, using the silane coupling agent KH-550. The filler surface modification process was optimized through chemical bonding simulation. KH-550 molecules (containing Si-OC bonds) were constructed in LAMMPS, with a grafting density of 0.5 nm⁻². The Monte Carlo algorithm was used to simulate the grafting process, randomly selecting carbon atoms on the CNT surface to form covalent bonds with Si atoms of KH-550. The bond energy was set to 80 kcal / mol. After grafting, the interface energy decreased by approximately 15%, indicating that chemical bonding enhanced interface stability. Finally, the optimized interface bonding strength parameters were calculated through tensile simulation. A constant strain rate of 0.001 ps⁻¹ was applied, and the maximum stress at interface fracture was recorded, resulting in a bonding strength increased to 1.8 GPa (compared to 1.2 GPa before modification). Analysis shows that grafting KH-550 reduces interfacial porosity (from 5% to 2%), enhances intermolecular van der Waals forces and covalent bonds, and thus improves interfacial bonding strength. The entire process is coupled with molecular dynamics and Monte Carlo algorithms, combined with energy minimization and stress analysis, forming a complete logical chain from thermal resistance quantification to interface optimization and strength enhancement. Data can be processed automatically using Python scripts, and the output results are used to guide material design.

[0045] Step S105: Based on the improved interface and strength parameters, the Monte Carlo method is used to simulate the orientation distribution of filler particles in the matrix to obtain a preliminary scheme for filler orientation control.

[0046] A Monte Carlo method is used to initialize the random distribution of filler particles in the matrix, generating an initial orientation angle dataset. Based on this initial orientation angle dataset and interfacial bonding strength parameters, the stress state of each filler particle is calculated, resulting in a stress distribution dataset. If the stress state of any filler particle in the stress distribution dataset does not meet a preset strength threshold, the orientation angle of the filler particle is adjusted, generating an optimized orientation angle dataset. Based on the optimized orientation angle dataset, a spatial distribution model of the filler particles is constructed, resulting in a three-dimensional orientation distribution dataset. Using this three-dimensional orientation distribution dataset, the statistical characteristics of the filler particle orientation angles are analyzed, generating an orientation control parameter set. The finite element method is used to verify the influence of the orientation control parameter set on the interfacial bonding strength, obtaining a verification result dataset. Based on the verification result dataset, the orientation control parameters are adjusted to generate the final filler orientation control scheme.

[0047] For example, the Monte Carlo method is used to simulate the orientation distribution of filler particles in a matrix. First, random particle orientation data is generated. The matrix is ​​assumed to be three-dimensional, and the filler particles are slender ellipsoids with a major axis of 0.1 mm and a minor axis of 0.02 mm. The number of particles is set to 1000. Using a random number generator, initial orientation angles θ (0 to π) and φ (0 to 2π) are assigned to each particle based on a uniform distribution. An interfacial bonding strength parameter is introduced, defined as the binding energy under van der Waals force, set at 0.5 × 10⁻¹⁹ joules. The Monte Carlo method is used to simulate the particle motion in the matrix, iterating 10,000 times. In each iteration, the particle orientation angle is randomly perturbed by ±0.01 radians. The total energy under the new orientation is calculated, including the interfacial bonding energy and the interparticle repulsive energy (set at 0.2 × 10⁻¹⁹ joules when the particle spacing is less than 0.03 mm). If the total energy of the new orientation decreases, the perturbation is accepted; otherwise, it is accepted with probability exp(-ΔE / kT), where k is the Boltzmann constant and T is 298K. After the simulation, the particle orientation distribution is statistically analyzed, and the eigenvalues ​​of the orientation tensor A are calculated (e.g., 0.65, 0.25, 0.1). The direction with the largest eigenvalue is the main orientation direction, indicating that 67% of the particles' long axes deflect less than 15 degrees along this direction. Analyzing the anisotropy of the orientation distribution, a filler orientation control scheme is derived: by applying an external shear flow field (shear rate of 100 s^-1), the main orientation direction is aligned with the flow field direction, and the strength parameter is adjusted to 0.7 × 10^-19 Joules to optimize particle orientation consistency. The final scheme is verified by simulation, and the orientation tensor eigenvalue is increased to 0.72, indicating an improvement of approximately 10% in orientation consistency, which can be used to guide the composite material preparation process.

[0048] Step S106: Extract high-probability orientation paths from the preliminary scheme of filler orientation control, analyze the impact of decreased processing fluidity using a fluid dynamics model, and determine the optimal parameters for complex mold forming.

[0049] High-probability orientation paths are extracted from the filler orientation control scheme to obtain a path dataset. A fluid dynamics model is used to analyze this path dataset, calculating the trend of processing flowability changes and obtaining flowability distribution data. If areas in the flowability distribution data do not meet preset thresholds, the high-probability orientation paths are adjusted to obtain an optimized path set. A spatial model is constructed based on the optimized path set to generate a three-dimensional orientation distribution map. The particle stress state is analyzed using the three-dimensional orientation distribution map to obtain a stress distribution dataset. Mold forming parameters are adjusted based on the stress distribution dataset to obtain an optimized parameter set. The optimized parameter set is verified using finite element analysis to obtain a complex mold forming scheme.

[0050] For example, in the preliminary scheme for filler orientation control, high-probability orientation paths are first extracted through digital image analysis. Image processing techniques based on grayscale gradient algorithms are then used to process the filler distribution image. Assuming the filler is fibrous with a length of 0.1 mm and an image resolution of 1000×1000 pixels, the Sobel operator is used to calculate the gradient direction of the fibers in the image, generating an orientation angle distribution map. Paths with an angle concentration greater than 80% are selected and defined as high-probability orientation paths, outputting a path coordinate dataset (x, y, θ), for example (500, 600, 45°). Next, the impact of decreased processing fluidity is analyzed using a fluid dynamics model. The finite element method (FEM) is used to simulate melt flow, and the Navier-Stokes equations are used to describe the flow behavior. The melt viscosity is set to 100 Pa·s, the mold inlet velocity is 0.05 m / s, the mold geometry is a complex curved surface, and the wall thickness is 0.002 m. The Reynolds number (Re = ρvL / μ, ρ = 1200 kg / m³) is calculated. 3 (L=0.01m), yielding Re≈6, indicating laminar flow. Simulation results show that increasing the filler orientation angle from 45° to 60° increases local flow resistance by 15%, leading to a 7% decrease in flowability. Finally, optimized parameters for complex mold forming were determined. A genetic algorithm (GA) was used to optimize the mold design, setting the objective function to minimize flow resistance and molding time, with constraints of mold temperature (180℃-220℃) and injection pressure (50-100MPa). The algorithm iterated 100 times with a population size of 50, a crossover rate of 0.8, and a mutation rate of 0.1, yielding the optimal parameter combination: temperature 200℃, pressure 80MPa, a 10% reduction in flow resistance, and a 5-second reduction in molding time. This process was implemented using image processing software, CFD simulation tools, and optimization algorithm software, ensuring data interconnection at each step and forming a complete technology chain from orientation extraction to parameter optimization.

[0051] Step S107: For the optimization parameters of complex mold forming, the particle swarm optimization algorithm is used to adjust the filler ratio and distribution to generate a structural configuration of a high-efficiency three-dimensional heat conduction network.

[0052] Initial filler ratio and distribution parameters are obtained, and a preliminary three-dimensional thermal conductivity network structure configuration is generated using a particle swarm optimization (PSO) algorithm. The thermal conductivity of this preliminary configuration is then assessed to determine if it falls below a preset threshold. If so, the filler ratio is adjusted using PSO to obtain optimized parameters. Based on these optimized parameters, the filler distribution is recalculated using PSO to obtain a new distribution scheme. A three-dimensional thermal conductivity network model is constructed based on this new scheme to obtain the network's geometric shape. The connectivity of the network structure is then assessed to determine if it falls below a preset threshold. If so, the network structure is adjusted using mesh generation techniques to obtain an optimized structure. Based on this optimized structure, the final configuration for complex mold forming is generated, resulting in a set of forming parameters. Finally, a final configuration scheme for an efficient three-dimensional thermal conductivity network is generated based on this set of parameters.

[0053] For example, for optimizing parameters in complex mold forming, the filler ratio and distribution are first adjusted using the Particle Swarm Optimization (PSO) algorithm. The initial particle swarm size is 50 particles, and the search space is the filler ratio (0.1 to 0.5) and the distribution uniformity (0 to 1). Each particle represents a filler ratio and distribution configuration. The velocity update formula is v_i(t+1)=0.8*v_i(t)+1.5*rand()*(pbest_i-x_i(t))+2.0*rand()*(gbest-x_i(t)), where the inertia weight of 0.8 controls the search stability, and the learning factors of 1.5 and 2.0 balance the local and global search. The objective function is to maximize the thermal conductivity. The initial filler ratio is set to 0.3, and the distribution uniformity is set to 0.5. The heat flux density of the heat conduction network is calculated through finite element analysis (FEA). It is assumed that the mold material is an aluminum-based composite material, the filler is carbon fiber, and the initial thermal conductivity is 150W / (m·K). After 100 iterations, the PSO converged to the optimal filler ratio of 0.42, distribution uniformity of 0.78, and thermal conductivity increased to 180 W / (m·K), a 20% improvement compared to the initial value. Next, based on the optimal filler ratio and distribution, a three-dimensional heat conduction network structure was generated using topology optimization. The structure was designed using ANSYS software, with a mesh density of 1 million tetrahedral elements, boundary conditions of a mold surface temperature of 100℃, an ambient temperature of 25℃, and a heat flux constraint of 500 W / m². 2 The structure was optimized using the gradient descent method, reducing the material volume by 10% while maintaining a thermal conductivity of at least 95%, ultimately generating a high thermal conductivity network. Verification results showed a 15% improvement in heat flux distribution uniformity and an 8°C reduction in temperature gradient. The entire process was automated through an algorithm. Data analysis indicated that optimizing the filler ratio and distribution uniformity significantly improved thermal conductivity, while topology optimization further reduced material waste, ensuring a balance between mold forming efficiency and thermal management performance.

[0054] Step S108: Based on the structural configuration of the high-efficiency three-dimensional heat-conducting network, the degree of improvement in the performance of the thermally conductive plastic is verified by finite element analysis of heat conduction, and the final heat transfer efficiency data is obtained.

[0055] The process involves: obtaining initial structural parameters for a three-dimensional heat-conducting network, including node distribution and connection methods; generating an initial three-dimensional heat-conducting network model using a preset geometric modeling tool to determine initial heat flux distribution data; adjusting the initial structural parameters using a particle swarm optimization algorithm to obtain an optimized three-dimensional heat-conducting network model; performing heat conduction simulation on the optimized three-dimensional heat-conducting network model using finite element analysis software to obtain heat flux distribution data and material thermal conductivity; if the uniformity of the heat flux distribution data does not reach a preset uniformity threshold, iteratively adjusting the structural parameters of the optimized three-dimensional heat-conducting network model and re-performing heat conduction simulation using finite element analysis software to obtain new heat flux distribution data; calculating the heat transfer efficiency of the thermally conductive plastic based on the heat flux distribution data and determining whether the performance improvement of the heat transfer efficiency reaches a preset performance threshold; determining a heat conduction path optimization scheme based on the heat transfer efficiency, generating performance improvement data for the thermally conductive plastic, and obtaining the final heat transfer efficiency result.

[0056] For example, based on the structural configuration of a high-efficiency three-dimensional thermally conductive network, a three-dimensional network model of the thermally conductive plastic is first generated. Computer-aided design software is then used to construct a thermally conductive network composed of carbon fibers and graphene. The carbon fibers have a diameter of 7 micrometers and a length of 100 micrometers, randomly distributed within the polymer matrix. The graphene sheets have a thickness of 2 nanometers and an area ratio of 5%. A heat conduction model is established using finite element analysis software (such as ANSYS), setting the thermal conductivity of the polymer matrix to 0.2 W / m·K, and the carbon fibers...

[0057] The thermal conductivity is 500 W / m·K, while that of graphene is 3000 W / m·K. A finite element method (FEM) was used to generate a mesh of 1 million tetrahedral elements to ensure computational accuracy. Boundary conditions were applied, assuming a surface temperature of 100°C on one side and 25°C on the other, with an ambient thermal convection coefficient of 10 W / m. 2 K. Steady-state heat conduction analysis based on Fourier's law of heat conduction. The heat flux density and temperature distribution were calculated, yielding an equivalent thermal conductivity of 15 W / m·K for the thermally conductive network, a 75-fold improvement compared to the 0.2 W / m·K of the pure polymer. To verify the performance improvement, a comparative experiment was designed to simulate a pure polymer sample without a thermally conductive network, calculating its heat transfer efficiency to be 2.5%, while the heat transfer efficiency of the three-dimensional thermally conductive network sample reached 85%. The efficiency improvement was attributed to the formation of efficient thermal channels in the thermally conductive filler through heat flux density integration (Q=∫q·dA). During the analysis, an iterative convergence algorithm was used to ensure the residual was less than 10^-6, verifying the reliability of the results. Furthermore, a temperature gradient cloud map was extracted using a post-processing module to confirm the contribution of the uniformity of the thermally conductive network to the heat transfer efficiency, demonstrating that the high thermal conductivity of the graphene sheets significantly reduced local thermal resistance. Finally, a performance report was generated based on the heat transfer efficiency data, showcasing the potential of thermally conductive plastics in thermal management applications, such as heat dissipation in electronic devices, where the improved efficiency could reduce device operating temperature by approximately 30°C and extend service life.

[0058] This invention provides a comprehensive optimization system for the properties of thermally conductive plastics, mainly comprising:

[0059] The filler dispersion state acquisition module is used to acquire the initial dispersion state data of inorganic fillers in thermally conductive plastics, and to analyze the spatial distribution of filler particles by scanning electron microscopy to obtain the characteristic parameters of the random distribution of fillers.

[0060] The heat transfer efficiency calculation module is used to calculate the heat transfer efficiency between packing particles using the finite element simulation method based on the characteristic parameters of the randomly distributed packing, and to determine the distribution law of phonon scattering phenomenon.

[0061] The phonon scattering analysis module is used to extract key regions from the distribution pattern of phonon scattering phenomena, and to analyze the interfacial bonding strength between the filler and the matrix using molecular dynamics simulation to obtain quantitative data on interfacial thermal resistance.

[0062] The interface thermal resistance quantification module is used to quantify the interface thermal resistance data, preset the interface modification molecular structure, optimize the filler surface modification process through chemical bonding simulation, and obtain the improved interface bonding strength parameters.

[0063] The interface modification and optimization module is used to simulate the orientation distribution of filler particles in the matrix using the Monte Carlo method based on the improved interface and strength parameters, so as to obtain a preliminary scheme for filler orientation control.

[0064] The orientation distribution simulation module is used to extract high-probability orientation paths from the initial scheme of filler orientation control, and combine it with the fluid dynamics model to analyze the impact of decreased processing fluidity and determine the optimization parameters for complex mold forming.

[0065] The molding parameter optimization module is used to optimize the molding parameters of complex molds. It uses a particle swarm optimization algorithm to adjust the filler ratio and distribution, and generates a structural configuration of a highly efficient three-dimensional heat conduction network.

[0066] The thermal conductivity network verification module is used to verify the degree of improvement in the performance of thermally conductive plastics through finite element analysis of thermal conductivity based on the structural configuration of the high-efficiency three-dimensional thermal conductivity network, and to obtain the final heat transfer efficiency data.

[0067] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A method for comprehensive optimization of the properties of thermally conductive plastics, characterized in that, The method includes the following steps: Step S101: Obtain the initial dispersion state data of inorganic fillers in thermally conductive plastics, and analyze the spatial distribution of filler particles by scanning electron microscopy to obtain the characteristic parameters of random distribution of fillers. Step S102: Based on the characteristic parameters of the random distribution of the packing, the heat transfer efficiency between the packing particles is calculated using the finite element simulation method to determine the distribution law of phonon scattering phenomenon. Step S103: Extract key regions from the distribution law of phonon scattering phenomenon, and use molecular dynamics simulation to analyze the interfacial bonding strength between the filler and the matrix to obtain quantitative data of interfacial thermal resistance. Step S104: Based on the quantitative data of interfacial thermal resistance, a pre-defined interfacial modification molecular structure is established, and the filler surface modification process is optimized through chemical bonding simulation to obtain improved interfacial bonding strength parameters. Step S105: Based on the improved interface and strength parameters, the Monte Carlo method is used to simulate the orientation distribution of filler particles in the matrix to obtain a preliminary scheme for filler orientation control. Step S106: Extract high-probability orientation paths from the preliminary scheme of filler orientation control, analyze the impact of decreased processing fluidity using a fluid dynamics model, and determine the optimal parameters for complex mold forming. Step S107: For the optimization parameters of complex mold forming, the particle swarm optimization algorithm is used to adjust the filler ratio and distribution to generate a structural configuration of a high-efficiency three-dimensional heat conduction network. Step S108: Based on the structural configuration of the high-efficiency three-dimensional heat-conducting network, the degree of improvement in the performance of the thermally conductive plastic is verified by finite element analysis of heat conduction, and the final heat transfer efficiency data is obtained.

2. The method for comprehensively optimizing the properties of thermally conductive plastics according to claim 1, characterized in that, Step S101 includes: The initial dispersion image of the inorganic filler in the thermally conductive plastic was obtained by scanning electron microscopy, thus obtaining the first image data; The first image data is denoised using an image processing algorithm to obtain the second image data. The contour information of the filler particles is extracted from the second image data to determine the particle size and particle spacing; The spatial distribution density is calculated based on the particle size and particle spacing to obtain the distribution density data; If the distribution density data meets the preset uniformity threshold, then the feature parameters of the random distribution are extracted through statistical analysis to obtain the first feature parameter set; For the first set of feature parameters, the K-means clustering algorithm is used to classify spatial uniformity to obtain the classification result; Based on the classification results, comprehensive feature parameters of spatial uniformity and random distribution are generated to obtain the final feature parameter set.

3. The method for comprehensive optimization of the thermally conductive plastic properties according to claim 1, characterized in that, Step S102 includes: By using the finite element method, random distribution data of filler particles are obtained, a three-dimensional geometric model is generated, and the particle spacing and interface contact state are determined. Based on the three-dimensional geometric model, the thermal conductivity distribution of the filler particles is calculated to obtain the interparticle heat transfer efficiency. If the heat transfer efficiency is lower than the preset threshold, the particle spacing and interface contact parameters are adjusted, and the three-dimensional geometric model is regenerated. The spatial distribution data of phonon scattering is obtained by analyzing the regenerated three-dimensional geometric model using a phonon scattering model. Using the Monte Carlo algorithm, the probability density of phonon scattering is calculated based on the spatial distribution data of phonon scattering, and the distribution pattern is determined. Based on the probability density of phonon scattering, the material properties and size distribution of the filler particles are optimized to generate an optimized three-dimensional geometric model; The optimized three-dimensional geometric model was verified through finite element simulation, and the final heat transfer efficiency and phonon scattering distribution were obtained.

4. A method for comprehensively optimizing the properties of thermally conductive plastics according to any one of claims 1-3, characterized in that, Step S103 includes: The scattering intensity distribution is obtained from phonon scattering data, and the frequency domain analysis of the scattering intensity distribution is performed using the Fourier transform method to obtain the frequency characteristics of the key region. Based on the frequency characteristics, a density clustering algorithm is used to divide the key regions and determine the core regions with high scattering intensity; For the core region, a molecular dynamics simulation model was constructed to obtain the atomic interaction forces between the filler and the matrix and to calculate the interfacial bonding strength. From the interface bonding strength data, the heat transfer process is simulated using the finite element analysis method to obtain the preliminary distribution of interface thermal resistance. If the variance of the interface thermal resistance distribution is greater than a preset threshold, the molecular dynamics simulation parameters are adjusted, the interface bonding strength is recalculated, and the optimized interface thermal resistance distribution is obtained. The optimized interface thermal resistance distribution was statistically analyzed using the Monte Carlo method to obtain quantitative data on the interface thermal resistance. Based on the quantified interface thermal resistance data, an interpolation algorithm is used to generate a spatial distribution map of thermal resistance, thereby determining the microscopic laws governing thermal conductivity.

5. A method for comprehensively optimizing the properties of thermally conductive plastics according to any one of claims 1-3, characterized in that, Step S104 includes: Obtain molecular structure data related to interfacial thermal resistance from a pre-defined molecular structure database; Molecular dynamics simulations were used to determine the chemical bonding characteristics of the filler surface; Based on the aforementioned chemical bonding characteristics, a filler surface modification model was constructed; Molecular interaction parameters of the filler surface modification model were obtained through molecular dynamics simulations. If the molecular interaction parameters meet the preset chemical bonding threshold, the surface modification process is optimized using response surface methodology to obtain an optimized set of process parameters. Based on the optimized set of process parameters, obtain the chemical bonding strength data of the filler surface; Molecular dynamics simulations were used to determine the improved interfacial bonding strength parameters. Based on the interface bonding strength parameters, a quantitative model of interface thermal resistance is constructed to obtain the correlation data between thermal resistance parameters and bonding strength. The support vector machine algorithm is used to determine the degree of optimization of the interface thermal resistance of the associated data, and the improved interface thermal resistance parameters are obtained. Based on the improved interfacial thermal resistance parameters, the final filler surface modification scheme is determined, and optimized interfacial bonding performance data are obtained.

6. A method for comprehensively optimizing the properties of thermally conductive plastics according to any one of claims 1-3, characterized in that, Step S105 includes: The Monte Carlo method was used to initialize the random distribution of filler particles in the matrix and generate an initial orientation angle dataset. Based on the initial orientation angle dataset, combined with the interface and strength parameters, the stress state of each filler particle is calculated to obtain the stress distribution dataset. If the stress state of filler particles in the stress distribution dataset does not meet the preset strength threshold, the orientation angle of the filler particles is adjusted to generate an optimized orientation angle dataset. Based on the optimized orientation angle dataset, a spatial distribution model of filler particles is constructed to obtain a three-dimensional orientation distribution dataset. Using the aforementioned three-dimensional orientation distribution dataset, the statistical characteristics of the orientation angles of the filler particles are analyzed to generate a set of orientation control parameters. The influence of the orientation control parameter set on the interfacial bonding strength was verified using the finite element analysis method, and the verification result dataset was obtained. Based on the verification result dataset, the orientation control parameters are adjusted to generate the final filler orientation control scheme.

7. A method for comprehensively optimizing the properties of thermally conductive plastics according to any one of claims 1-3, characterized in that, Step S106 includes: High-probability orientation paths are extracted from the packing orientation control scheme to obtain a path dataset; The path dataset was analyzed using a fluid dynamics model to calculate the trend of processing flowability changes and obtain flowability distribution data. If there are regions in the liquidity distribution data that do not meet the preset threshold, then the high-probability orientation path is adjusted to obtain an optimized path set; A spatial model is constructed based on the optimized path set, and a three-dimensional orientation distribution map is generated. The stress state of the particles is analyzed by the three-dimensional orientation distribution map to obtain a stress distribution dataset; Adjust the mold forming parameters based on the stress distribution dataset to obtain an optimized parameter set; The optimized parameter set was verified using the finite element analysis method, and a complex mold forming scheme was obtained.

8. A method for comprehensively optimizing the properties of thermally conductive plastics according to any one of claims 1-3, characterized in that, Step S107 includes: The initial filler ratio and distribution parameters are obtained, and a preliminary three-dimensional heat conduction network structure configuration is generated using a particle swarm optimization algorithm. Determine whether the thermal conductivity of the initial three-dimensional thermally conductive network structure configuration is lower than a preset threshold. If it is lower than the preset threshold, adjust the filler ratio through a particle swarm optimization algorithm to obtain the optimized filler ratio parameters. Based on the optimized packing ratio parameters, the packing distribution is recalculated using the particle swarm optimization algorithm to obtain a new packing distribution scheme. Based on the new packing distribution scheme, a three-dimensional heat conduction network model is constructed to obtain the geometric shape of the network structure. Determine whether the connectivity of the network structure is lower than a preset threshold. If it is lower than the preset threshold, adjust the network structure using mesh partitioning technology to obtain an optimized network structure. Based on the optimized network structure, the final structural configuration for complex mold forming is generated, and a set of forming parameters is obtained. Based on the set of molding parameters, a final configuration scheme for a high-efficiency three-dimensional heat conduction network is generated.

9. A comprehensive optimization system for the properties of thermally conductive plastics, characterized in that, This system is used to implement a comprehensive optimization method for the properties of thermally conductive plastics as described in any one of claims 1-8, the system comprising: The filler dispersion state acquisition module is used to acquire the initial dispersion state data of inorganic fillers in thermally conductive plastics, and to analyze the spatial distribution of filler particles by scanning electron microscopy to obtain the characteristic parameters of the random distribution of fillers. The heat transfer efficiency calculation module is used to calculate the heat transfer efficiency between packing particles using the finite element simulation method based on the characteristic parameters of the randomly distributed packing, and to determine the distribution law of phonon scattering phenomenon. The phonon scattering analysis module is used to extract key regions from the distribution pattern of phonon scattering phenomena, and to analyze the interfacial bonding strength between the filler and the matrix using molecular dynamics simulation to obtain quantitative data on interfacial thermal resistance. The interface thermal resistance quantification module is used to quantify the interface thermal resistance data, preset the interface modification molecular structure, optimize the filler surface modification process through chemical bonding simulation, and obtain the improved interface bonding strength parameters. The interface modification and optimization module is used to simulate the orientation distribution of filler particles in the matrix using the Monte Carlo method based on the improved interface and strength parameters, so as to obtain a preliminary scheme for filler orientation control. The orientation distribution simulation module is used to extract high-probability orientation paths from the initial scheme of filler orientation control, and combine it with the fluid dynamics model to analyze the impact of decreased processing fluidity and determine the optimization parameters for complex mold forming. The molding parameter optimization module is used to optimize the molding parameters of complex molds. It uses a particle swarm optimization algorithm to adjust the filler ratio and distribution, and generates a structural configuration of a highly efficient three-dimensional heat conduction network. The thermal conductivity network verification module is used to verify the degree of improvement in the performance of thermally conductive plastics through finite element analysis of thermal conductivity based on the structural configuration of the high-efficiency three-dimensional thermal conductivity network, and to obtain the final heat transfer efficiency data.

Images