A cross-grid parameter conversion method and system for charge transfer process simulation

Abstract

The application discloses a cross-grid parameter conversion method and system for charge transfer process simulation. The method is based on grid points representing molecular positions, obtaining a source grid and charge transfer parameters, the charge transfer parameters including energy levels representing Marcus transition rate, transfer integral, reorganization energy, or attempt hop rate and charge localization length representing Miller-Abrahams transition rate. If there is no target grid, a regular grid is generated based on the source grid density, and the initial parameters of the target grid are generated by fitting the parameter law. The charge mobility under different directions and field strengths is calculated for the source grid model. The charge mobility of the target grid model is calculated, and the multi-parameter optimization is performed with the source grid model mobility as the target, so that the deviation between the two reaches the preset convergence condition, and the optimized parameters are output to complete the conversion. The application solves the physical mapping distortion problem between cross-scale grids and realizes high-fidelity conversion of parameters.

Description

Technical Field

[0001] This invention relates to the field of molecular-level charge transfer process simulation and multi-scale modeling technology, and in particular to a cross-grid parameter conversion method and system for charge transfer process simulation. Background Technology

[0002] High-precision simulations of charge transfer processes within materials typically require consideration of charge jumping behavior between molecules or fragments, which is usually described using specific mesh models. Mesh models can take various forms. At the microscopic level, in source meshes derived from molecular dynamics simulations, the three-dimensional coordinates of mesh points are usually determined by the molecular centroid coordinates or the equivalent charge transport center, and their mesh spacing and spatial arrangement are not fixed. However, in models suitable for large-scale, efficient computation, regularly arranged target meshes are typically used. Because the source and target mesh models differ in spatial structure and physical mapping, directly applying the same charge transport parameters can easily lead to significant deviations in transport results such as charge mobility. Therefore, a method is needed to achieve parameter consistency conversion between source and target mesh models.

[0003] Specifically, current KMC simulations or master equation methods based on molecular dynamics (MD) packing structures can preserve microscopic structural information such as molecular coordinates, molecular nearest neighbor relationships, molecular energy levels, and intermolecular transfer integrals, resulting in high simulation fidelity. However, their simulation scale is usually limited, making them difficult to use for efficient simulations of larger-scale systems. In contrast, KMC simulations or master equation methods based on regular target mesh models are computationally more efficient and can support the simulation of charge transfer processes in larger-scale systems. However, there is a fundamental difference between the two types of mesh models: the molecular arrangement in the source mesh model is usually irregular. If charge transport parameters such as average energy levels and average transfer integrals extracted from the source mesh model are directly substituted into the target mesh model, it may lead to a significant deviation between the charge mobility calculated by the target mesh model and the results from the source mesh model.

[0004] To address the inconsistency between the source and target mesh models, existing methods often rely on empirical settings. For example, they might directly set the average charge transport parameters extracted from the source mesh as constant parameters between sites in the target mesh, or adjust the charge transport parameters in the target mesh model based on experience. These methods lack a closed-loop correction mechanism that targets the calculation results of the source mesh model. This makes it difficult to ensure consistency between the target and source mesh models in key transport results such as charge mobility, and also fails to meet the accuracy requirements for parameters in new material screening and multi-scale charge transfer process simulations.

[0005] Patent CN116110524A discloses a method and system for predicting the charge mobility of polymer insulating materials. The method includes: obtaining an amorphous molecular dynamics model of the polymer insulating material; calculating the recombination energy and free energy difference parameters for each polymer molecule; calculating the charge transfer integral parameters between each pair of nearest-neighbor polymer molecules; calculating the electron or hole transfer rate between nearest-neighbor molecules based on Marcus theory; and performing KMC simulations or master equation calculations on the polymer insulating molecular dynamics model to obtain the charge mobility of the polymer insulating material. This method primarily focuses on single-scale mobility calculations based on molecular dynamics models and does not involve cross-grid parameter conversion from the source grid model to the target grid model, nor does it provide a parameter consistency conversion mechanism that optimizes and corrects the parameters of the target grid model based on the charge mobility of the source grid model.

[0006] Therefore, the existing technology still lacks a cross-grid parameter conversion method and system that can convert the microstructure information and charge transport parameter laws in the source grid model to the target grid model and achieve automatic parameter optimization through the target constraint of charge mobility. Summary of the Invention

[0007] The purpose of this invention is to provide a cross-grid parameter conversion method and system for charge transfer process simulation, which is used to achieve parameter consistency conversion from source grid model to regular grid model, so that the deviation between the charge mobility calculated by the regular grid model under different directions and different field strengths and the charge mobility of the source grid model reaches the preset convergence condition, thereby improving the fidelity and reliability of cross-grid charge transfer process simulation.

[0008] The objective of this invention can be achieved through the following technical solutions: A cross-grid parameter transformation method for charge transfer process simulation includes the following steps: S1: Obtain the source grid and charge transport parameters, and construct a source grid model based on the source grid after allocating the charge transport parameters; wherein, the charge transport parameters include energy levels, transfer integrals, and recombination energy used to characterize the Marcus transition rate, and attempted jump rate and charge localization length used to characterize the Miller-Abrahams transition rate. S2: If no target grid exists, a regular grid is generated based on the grid point density of the source grid, where the grid spacing represents the intermolecular distance and the grid point coordinates represent the molecular position; the charge transport parameter law of the source grid is fitted, and the initial charge transport parameters of the target grid are generated based on the law; S3: Calculate the charge mobility of the source mesh model under different directions and different field strengths; S4: Calculate the charge mobility of the target mesh model, and perform multi-parameter optimization with the charge mobility of the source mesh model as the target. Adjust the parameters of the target mesh model through the optimization algorithm. The parameters include at least one of energy disorder, recombination energy, pre-exponential factor and decay coefficient in the transfer integral expression, attempted jump rate and charge localization length, so that the deviation between the charge mobility of the target mesh model and the charge mobility of the source mesh model reaches the preset convergence condition. Output the optimized parameters to complete the cross-mesh parameter conversion.

[0009] Furthermore, in the process of generating a regular mesh in step S2, the method for determining the mesh spacing includes: determining the mesh spacing based on the mesh point density of the regular mesh; wherein the mesh point density is consistent with the source mesh density; For source grids that are multiphase mixed systems, it is necessary to maintain consistent grid spacing across the entire system. The spacing is set based on the equivalent grid spacing calculated from either the average density of the source grid system or the density of a certain phase.

[0010] Furthermore, the grid spacing is calculated as follows: Where a represents the grid spacing, N represents the number of molecules, and V represents the system volume.

[0011] Furthermore, the process of fitting the charge transport parameter regularity of the source grid includes: Based on the molecular energy level data in the source grid, the energy level distribution is fitted using a distribution model including the Gaussian distribution to obtain the energy level distribution law: in, The average energy value. This is the energy disorder parameter; Meanwhile, based on the intermolecular distance in the source grid With transfer integral The data were fitted to the transfer integral using an exponential decay function: in, For the pre-exponential factor of the transfer integral, This is the attenuation coefficient.

[0012] Furthermore, the Marcus transition rate is described by the Marcus equation, and the Miller-Abrahams transition rate is described by the Miller-Abrahams equation. The expression for the Marcus transition rate is: in, For site with site The charge transition rate between them; The integral of inter-site transfer; For recombination energy; The energy level difference between sites; Boltzmann's constant; The system temperature; To reduce Planck's constant; The Miller-Abrahams transition rate expression is: in, For site with site The charge transition rate between them; To test the jumping frequency; The energy level difference between sites; This is the charge localization attenuation coefficient; Spatial distance between sites; Boltzmann's constant; The temperature is the system temperature.

[0013] Furthermore, the charge mobility calculation method includes one of KMC simulation and master equation method, wherein both KMC simulation and master equation method are based on the same inter-site jump rate matrix; When using KMC simulation, let the sum of the rates of all possible jump events be: in, For the first Jump rate for each selectable jump event; Generate two independent random numbers Then the next event Choose from the following formula: in, For the first Jump rate for each selectable jump event; The system time changes are as follows: The charge mobility is obtained by statistically analyzing the cumulative displacement of the charge in the direction of the applied electric field and the total simulation time. When using the master equation method, let For the charge located at the th The probability of each site, For charge from site Jump to site rate, For charge from site j Jump to site i If the rate is such that the site occupancy probability satisfies: Under steady-state conditions: Calculate the average drift velocity along the applied electric field direction based on the steady-state occupancy probability and the jump rate: in, and Sites and site coordinates The unit vector is the direction of the applied electric field. The charge mobility is: in, For the applied electric field strength, The average drift velocity is denoted by in the direction of the applied electric field.

[0014] Furthermore, the process of adjusting the parameters of the target mesh model through the optimization algorithm includes: defining the objective function as the square of the relative error or the absolute error. The optimization process aims to minimize L; The specific formula for calculating mobility error uses relative error: in, The mobility of the target mesh model. The mobility of the source mesh model; When multiple iterations meet the following conditions: the change in the objective function L is less than a preset threshold, the preset maximum number of iterations is reached, and the relative error is less than any of the preset criteria, the iteration converges and the optimized parameters of the target mesh model are output.

[0015] Furthermore, in the multi-parameter optimization process of step S4, the optimization algorithm includes at least one of the following: coordinate search, pattern search, coarse grid search, random search, local positive and negative perturbation, Nelder-Mead simplex method, Powell method, trust region method, gradient descent, quasi-Newton method, Bayesian optimization, genetic algorithm, differential evolution, particle swarm optimization, simulated annealing, CMA-ES, ant colony optimization, multi-objective Pareto optimization, phased refinement, early stopping search, and heuristic hybrid search.

[0016] Furthermore, when the charge mobility calculated by the source mesh model in different directions is anisotropic, the charge mobility of the source mesh model in one transmission direction is selected as the reference direction, and a correction factor is determined according to the ratio of the charge mobility in other transmission directions to the reference direction. The charge transition rate of the target mesh model is corrected using the correction factor to obtain the charge transition rate corresponding to different transmission directions.

[0017] A cross-grid parameter transformation system for charge transfer process simulation includes: The source grid and charge transport parameter acquisition module is used to acquire the source grid and charge transport parameters, and to construct a source grid model based on the source grid after allocating the charge transport parameters; the charge transport parameters include energy levels, transfer integrals, and recombination energy used to characterize the Marcus transition rate, and attempted jump rate and charge localization length used to characterize the Miller-Abrahams transition rate. The target mesh and initial parameter generation module is used to generate a regular mesh based on the mesh point density of the source mesh when a target mesh does not exist, and to construct a target mesh model by combining the target mesh and initial charge transport parameters. The mesh spacing of the regular mesh represents the intermolecular distance, and the mesh point coordinates represent the molecular positions. The target mesh and initial parameter generation module is also used to fit the charge transport parameter law of the source mesh and generate the initial parameters of the target mesh based on the charge transport parameter law. The mesh spacing is determined according to the mesh point density of the regular mesh, and the mesh point density is consistent with the density of the source mesh. If the source mesh belongs to a multiphase mixture system, it is necessary to maintain a consistent mesh spacing throughout the system, which can be set based on the equivalent mesh spacing calculated from any factor in the density calculation of a certain phase of the source mesh. The source mesh model calculation module is used to calculate the source mesh model and obtain the charge mobility of the source mesh model under different directions and different field strengths. The target mesh model calculation and multi-parameter optimization module is used to calculate the charge mobility of the target mesh model and perform multi-parameter optimization with the charge mobility of the source mesh model as the target, so that the deviation between the charge mobility of the target mesh model and the charge mobility of the source mesh model reaches the preset convergence condition. The output module is used to output the optimized parameters and send them to multi-scale device simulation software and material databases.

[0018] Compared with the prior art, the present invention has the following beneficial effects: 1. Precise parameter conversion: A reverse fitting mechanism oriented towards mobility matching was constructed, which solved the problem of difficulty in effectively corresponding the microscopic simulation information in the source mesh with the charge transfer parameters of the target mesh model. It achieved high-fidelity conversion from the source mesh charge transfer parameter law to the target mesh model charge transfer parameters, and improved the accuracy of cross-mesh parameter conversion.

[0019] 2. Ensuring consistency and fidelity of multi-scale simulation results: By using the charge mobility of the source mesh model as the target, a backfit optimization is performed on the target mesh model, ensuring that the deviation between the charge mobility of the target mesh model and the charge mobility of the source mesh model reaches a preset convergence condition. This closed-loop optimization mechanism guarantees the consistency of simulation results for the same material under different mesh models, significantly improving the fidelity of the simulation of molecular-level charge transfer processes.

[0020] 3. Improve the physical consistency of the model's spatial scale: In the process of generating the target mesh and determining the mesh spacing, the mesh point density of the target mesh is kept consistent with that of the source mesh, thereby effectively avoiding spatial scale distortion caused by manually setting the mesh spacing. This allows the target mesh model to more realistically represent the physical transmission distance in the system corresponding to the source mesh, further improving the reliability of multi-scale simulation. Attached Figure Description

[0021] Figure 1 This is a flowchart of the present invention; Figure 2 This is a block diagram of the present invention. Detailed Implementation

[0022] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments. These embodiments are based on the technical solution of the present invention and provide detailed implementation methods and specific operating procedures. However, the scope of protection of the present invention is not limited to the following embodiments.

[0023] Example 1 This embodiment details a cross-grid parameter transformation method for simulating charge transfer processes, such as... Figure 1 As shown, the specific steps are as follows: S1: Obtain source mesh and charge transport parameters Obtain the source grid and charge transport parameters, and construct a source grid model based on the source grid after assigning the charge transport parameters; wherein, the charge transport parameters include energy levels, transfer integrals, and recombination energy used to characterize the Marcus transition rate, or attempted jump rate and charge localization length used to characterize the Miller-Abrahams transition rate. When charge is transported by holes, the molecular energy levels can be HOMO levels; when charge is transported by electrons, the molecular energy levels can be LUMO levels. The mean energy and energy disorder parameter can be obtained from the molecular energy level distribution; based on the intermolecular distance and corresponding transfer integral data, parameters showing the decay of the transfer integral with intermolecular distance can be fitted.

[0024] Specifically, extracting multiple sets of intermolecular distances absolute value of the transfer integral The pre-exponential factor and attenuation coefficient of the transfer integral were determined using a log-linear fitting method. in, For the pre-exponential factor of the transfer integral, The attenuation coefficient is... This represents the intermolecular distance.

[0025] In this embodiment, a Gaussian energy level distribution is used as an example to describe the degree of energy disorder: in, For energy disorder parameter, This represents the average energy. From this, the source grid and its corresponding charge transport parameters are obtained.

[0026] S2: Generate the target mesh and initial charge transport parameters If no target grid exists, a regular grid is generated based on the source grid density, where the grid spacing represents the intermolecular distance and the grid point coordinates represent the molecular positions; the charge transport parameter pattern of the source grid is fitted, and the initial charge transport parameters of the target grid are generated based on the pattern. In this embodiment, the grid spacing is determined based on the grid point density of the regular grid, and the grid point density is kept consistent with the molecular density in the molecular system obtained from the MD simulation. The grid spacing is calculated as follows: Where a represents the grid spacing, N represents the number of molecules, and V represents the system volume.

[0027] After determining the grid spacing, the grid point coordinates of the target grid are generated according to the rule arrangement, and the skippable site pairs in the target grid are determined according to the molecular nearest neighbor relationship criterion in the source grid.

[0028] Furthermore, the charge transport parameter pattern of the source mesh is fitted, and the initial charge transport parameters of the target mesh are generated based on the pattern.

[0029] Specifically, initial energy level parameters for each point in the target grid are generated based on the energy level distribution pattern in the source grid, and initial transfer integral parameters between adjacent sites in the target grid are generated based on the dependency relationship between intermolecular distance and transfer integral in the source grid. For the site energy levels in the target grid model, a Gaussian distribution is used for modeling and generation. For example, in this embodiment, the target grid site energy levels satisfy: in, The average energy value. This is the energy disorder parameter.

[0030] For the transfer integral between adjacent sites, an exponential decay function is used as the model for charge transport: in, For site with site Transfer integrals between For the pre-exponential factor of the transfer integral, The attenuation coefficient is... This represents the distance between sites.

[0031] Therefore, by combining the above statistical distribution law and distance attenuation law, the charge transport parameter system of the target grid is constructed, thereby obtaining the target grid model for subsequent charge mobility calculation.

[0032] S3: Source Mesh Model Calculation The charge transfer process of the source mesh model obtained in S1 is calculated to obtain the charge mobility of the source mesh model under different directions and different field strengths.

[0033] The charge mobility calculation method includes one of KMC simulation and master equation method. Both KMC simulation and master equation method are based on the same inter-site jump rate matrix.

[0034] When using KMC simulation, let the sum of the rates of all possible jump events be: in, For the first Jump rate for each selectable jump event; Generate two independent random numbers Then the next event Choose from the following formula: in, For the first Jump rate for each selectable jump event; The system time advance is: The charge mobility is obtained by statistically analyzing the cumulative displacement of the charge in the direction of the applied electric field and the total simulation time. When using the master equation method, let For the charge located at the th The probability of each site, For charge from site Jump to site rate, For charge from site j Jump to site i If the rate is such that the site occupancy probability satisfies: Under steady-state conditions: Calculate the average drift velocity along the applied electric field direction based on the steady-state occupancy probability and the jump rate: in, and Sites and site coordinates Let be the unit vector of the direction of the applied electric field.

[0035] The charge mobility is: in, The applied electric field strength.

[0036] S4: Target Mesh Model Calculation and Multi-parameter Optimization The charge mobility of the target mesh model is calculated, and multi-parameter optimization is performed with the charge mobility of the source mesh model as the target. The parameters of the target mesh model are adjusted by the optimization algorithm. The parameters include energy disorder, recombination energy, pre-exponential factor and decay coefficient in the transfer integral expression, or at least one of jump rate and charge localization length, so that the deviation between the charge mobility of the target mesh model and the charge mobility of the source mesh model reaches the preset convergence condition. The optimized parameters are output to complete the cross-mesh parameter conversion.

[0037] Specifically, the objective function is defined as the square of the relative error or the absolute error: The optimization process aims to minimize L; The specific formula for calculating mobility error uses relative error: in, The mobility of the target mesh model. The mobility of the source mesh model; When multiple iterations meet the following conditions: the change in the objective function L is less than a preset threshold, the preset maximum number of iterations is reached, and the relative error is less than any of the preset criteria, the iteration converges and the optimized parameters of the target mesh model are output.

[0038] The optimization algorithm includes at least one or more of the following: coordinate search, pattern search, coarse grid search, random search, local positive and negative perturbation, Nelder-Mead simplex method, Powell method, trust region method, gradient descent, quasi-Newton method, Bayesian optimization, genetic algorithm, differential evolution, particle swarm optimization, simulated annealing, CMA-ES, ant colony optimization, multi-objective Pareto optimization, phased refinement, early stopping search, and heuristic hybrid search. Example 2 This embodiment, based on Embodiment 1, further illustrates the processing method for anisotropic transmission characteristics.

[0039] If the charge mobility calculated by the source mesh model exhibits anisotropy in different directions, the charge mobility in one transmission direction of the source mesh model can be selected as the reference direction first. Then, a correction factor can be determined based on the ratio of the charge mobility in other transmission directions to the charge mobility in the reference direction. Let the reference direction be... Direction, source mesh model in , , The reference mobility in the directions are respectively , , Then the correction factor can be defined as: The charge transition rate of the target mesh model is corrected using the aforementioned direction correction factor to obtain the charge transition rates corresponding to other transmission directions.

[0040] By using the above method, the parameter conversion of anisotropic transport characteristics is achieved while maintaining the overall structural consistency of the target mesh model. The remaining steps are the same as in Example 1.

[0041] Example 3 This embodiment provides a cross-grid parameter conversion system for simulating charge transfer processes, such as... Figure 2 As shown, the system consists of the following functional modules: The source grid and charge transport parameter acquisition module 1 is used to acquire the source grid and charge transport parameters, and to construct a source grid model based on the source grid after allocating the charge transport parameters; the charge transport parameters include energy levels, transfer integrals, and recombination energy used to characterize the Marcus transition rate, and attempted jump rate and charge localization length used to characterize the Miller-Abrahams transition rate. The target mesh and initial parameter generation module 2 is used to generate a regular mesh based on the mesh point density of the source mesh when a target mesh does not exist, and to construct a target mesh model by combining the target mesh and initial charge transport parameters; wherein, the mesh spacing of the regular mesh represents the intermolecular distance, and the mesh point coordinates represent the molecular positions; the target mesh and initial parameter generation module is also used to fit the charge transport parameter law of the source mesh, and to generate the initial parameters of the target mesh based on the charge transport parameter law; wherein, the mesh spacing is determined according to the mesh point density of the regular mesh, and the mesh point density is consistent with the density of the source mesh; if the source mesh belongs to a multiphase mixture system, it is necessary to maintain a consistent mesh spacing throughout the system, which can be set according to the equivalent mesh spacing calculated from any item in the density calculation of a certain phase of the source mesh; Source mesh model calculation module 3 is used to calculate the source mesh model and obtain the charge mobility of the source mesh model under different directions and different field strengths; The target mesh model calculation and multi-parameter optimization module 4 is used to calculate the charge mobility of the target mesh model and perform multi-parameter optimization with the charge mobility of the source mesh model as the target, so that the deviation between the charge mobility of the target mesh model and the charge mobility of the source mesh model reaches the preset convergence condition. Output module 5 is used to output the optimized parameters and output them to multi-scale device simulation software and material database.

[0042] The rest is the same as in Example 1.

[0043] The preferred embodiments of the present invention have been described in detail above. It should be understood that those skilled in the art can make numerous modifications and variations based on the concept of the present invention without creative effort. Therefore, all technical solutions that can be obtained by those skilled in the art based on the concept of the present invention through logical analysis, reasoning, or limited experimentation on the basis of existing technology should be within the scope of protection defined by the claims.

Claims

1. A cross-grid parameter transformation method for simulating charge transfer processes, characterized in that, Includes the following steps: S1: Obtain the source grid and charge transport parameters, and construct a source grid model based on the source grid after allocating the charge transport parameters; wherein, the charge transport parameters include energy levels, transfer integrals, and recombination energy used to characterize the Marcus transition rate, and attempted jump rate and charge localization length used to characterize the Miller-Abrahams transition rate. S2: If no target grid exists, a regular grid is generated based on the grid point density of the source grid, where the grid spacing represents the intermolecular distance and the grid point coordinates represent the molecular position; the charge transport parameter law of the source grid is fitted, and the initial charge transport parameters of the target grid are generated based on the law; S3: Calculate the charge mobility of the source mesh model under different directions and different field strengths; S4: Calculate the charge mobility of the target mesh model, and perform multi-parameter optimization with the charge mobility of the source mesh model as the target. Adjust the parameters of the target mesh model through the optimization algorithm. The parameters include at least one of energy disorder, recombination energy, pre-exponential factor and decay coefficient in the transfer integral expression, attempted jump rate and charge localization length, so that the deviation between the charge mobility of the target mesh model and the charge mobility of the source mesh model reaches the preset convergence condition. Output the optimized parameters to complete the cross-mesh parameter conversion.

2. The cross-mesh parameter conversion method for charge transfer process simulation according to claim 1, characterized in that, In step S2, when generating a regular mesh, the method for determining the mesh spacing includes: determining the mesh spacing based on the mesh point density of the regular mesh; wherein the mesh point density is consistent with the source mesh density; For source grids that are multiphase mixed systems, it is necessary to maintain consistent grid spacing across the entire system. The spacing is set based on the equivalent grid spacing calculated from either the average density of the source grid system or the density of a certain phase.

3. The cross-mesh parameter conversion method for charge transfer process simulation according to claim 1, characterized in that, The grid spacing is calculated as follows: Where a represents the grid spacing, N represents the number of molecules, and V represents the system volume.

4. The cross-mesh parameter conversion method for charge transfer process simulation according to claim 1, characterized in that, The process of fitting the charge transport parameter law of the source grid includes: Based on the molecular energy level data in the source grid, the energy level distribution is fitted using a distribution model including the Gaussian distribution to obtain the energy level distribution law: in, The average energy value. This is the energy disorder parameter; Meanwhile, based on the intermolecular distance in the source grid With transfer integral The data were fitted to the transfer integral using an exponential decay function: in, For the pre-exponential factor of the transfer integral, This is the attenuation coefficient.

5. The cross-mesh parameter conversion method for charge transfer process simulation according to claim 1, characterized in that, The Marcus transition rate is described by the Marcus equation, and the Miller-Abrahams transition rate is described by the Miller-Abrahams equation. The expression for the Marcus transition rate is: in, For site with site The charge transition rate between them; The integral of the transfer between sites; For recombination energy; The energy level difference between sites; Boltzmann's constant; The system temperature; To reduce Planck's constant; The Miller-Abrahams transition rate expression is: in, For site with site The charge transition rate between them; To test the jumping frequency; The energy level difference between sites; This is the charge localization attenuation coefficient; Spatial distance between sites; Boltzmann's constant; The temperature is the system temperature.

6. The cross-mesh parameter conversion method for charge transfer process simulation according to claim 1, characterized in that, The charge mobility calculation method includes one of KMC simulation and master equation method, wherein both KMC simulation and master equation method are based on the same inter-site jump rate matrix; When using KMC simulation, let the sum of the rates of all possible jump events be: in, For the first Jump rate for each selectable jump event; Generate two independent random numbers Then the next event Choose from the following formula: in, For the first Jump rate for each selectable jump event; The system time changes are as follows: The charge mobility is obtained by statistically analyzing the cumulative displacement of the charge in the direction of the applied electric field and the total simulation time. When using the master equation method, let For the charge located at the th The probability of each site, For charge from site Jump to site rate, For charge from site j Jump to site i If the rate is such that the site occupancy probability satisfies: Under steady-state conditions: Calculate the average drift velocity along the applied electric field direction based on the steady-state occupancy probability and the jump rate: in, and They are the sites and site coordinates The unit vector of the direction of the applied electric field; The charge mobility is: in, For the applied electric field strength, The average drift velocity is denoted by in the direction of the applied electric field.

7. The cross-mesh parameter conversion method for charge transfer process simulation according to claim 1, characterized in that, The process of adjusting the parameters of the target mesh model through optimization algorithms includes: defining the objective function as the squared relative error or the absolute error. The optimization process aims to minimize L; The specific formula for calculating mobility error uses relative error: in, The mobility of the target mesh model. The mobility of the source mesh model; When multiple iterations meet the following conditions: the change in the objective function L is less than a preset threshold, the preset maximum number of iterations is reached, and the relative error is less than any of the preset criteria, the iteration converges and the optimized parameters of the target mesh model are output.

8. The cross-mesh parameter conversion method for charge transfer process simulation according to claim 1, characterized in that, In the multi-parameter optimization process in step S4, the optimization algorithm includes at least one of the following: coordinate search, pattern search, coarse grid search, random search, local positive and negative perturbation, Nelder-Mead simplex method, Powell method, trust region method, gradient descent, quasi-Newton method, Bayesian optimization, genetic algorithm, differential evolution, particle swarm optimization, simulated annealing, CMA-ES, ant colony optimization, multi-objective Pareto optimization, phased refinement, early stopping search, and heuristic hybrid search.

9. The cross-mesh parameter conversion method for charge transfer process simulation according to claim 1, characterized in that, When the charge mobility calculated by the source mesh model in different directions is anisotropic, the charge mobility of the source mesh model in one transmission direction is selected as the reference direction, and a correction factor is determined according to the ratio of the charge mobility in other transmission directions to the reference direction. The charge transition rate of the target mesh model is corrected using the correction factor to obtain the charge transition rate corresponding to different transmission directions.

10. A cross-grid parameter conversion system for simulating charge transfer processes, characterized in that, include: The source grid and charge transport parameter acquisition module is used to acquire the source grid and charge transport parameters, and to construct a source grid model based on the source grid after allocating the charge transport parameters; the charge transport parameters include energy levels, transfer integrals, and recombination energy used to characterize the Marcus transition rate, and attempted jump rate and charge localization length used to characterize the Miller-Abrahams transition rate. The target mesh and initial parameter generation module is used to generate a regular mesh based on the mesh point density of the source mesh when a target mesh does not exist, and to construct a target mesh model by combining the target mesh and initial charge transport parameters. The mesh spacing of the regular mesh represents the intermolecular distance, and the mesh point coordinates represent the molecular positions. The target mesh and initial parameter generation module is also used to fit the charge transport parameter law of the source mesh and generate the initial parameters of the target mesh based on the charge transport parameter law. The mesh spacing is determined according to the mesh point density of the regular mesh, and the mesh point density is consistent with the density of the source mesh. If the source mesh belongs to a multiphase mixture system, it is necessary to maintain a consistent mesh spacing throughout the system, which can be set based on the equivalent mesh spacing calculated from any factor in the density calculation of a certain phase of the source mesh. The source mesh model calculation module is used to calculate the source mesh model and obtain the charge mobility of the source mesh model under different directions and different field strengths. The target mesh model calculation and multi-parameter optimization module is used to calculate the charge mobility of the target mesh model and perform multi-parameter optimization with the charge mobility of the source mesh model as the target, so that the deviation between the charge mobility of the target mesh model and the charge mobility of the source mesh model reaches the preset convergence condition. The output module is used to output the optimized parameters and send them to multi-scale device simulation software and material databases.

Images