A method and apparatus for modeling self-crosslinking of epoxy resins based on molecular dynamics

Abstract

This application provides a molecular dynamics-based method and apparatus for modeling the self-crosslinking of epoxy resin, relating to the field of molecular dynamics. The method includes: constructing epoxy resin monomer molecules and labeling target atoms in the three-membered ring as reaction atom pairs; constructing an amorphous structure model based on the reaction atom pairs to form a molecular dynamics simulation system; entering a crosslinking cycle after energy minimization and kinetic equilibrium; searching and filtering reaction atom pairs with a first cutoff radius to obtain target reaction atom pairs; applying harmonic oscillator constraints to flexibly bring the reaction atoms closer and construct crosslinking chemical bonds, while simultaneously performing ring-opening operations on epoxy groups; calculating the current crosslinking conversion rate and comparing it with a preset crosslinking conversion rate threshold; if it is less than the threshold, expanding the cutoff radius to continue the crosslinking cycle; if it is greater than or equal to the threshold, performing abnormal coordination atom detection and repair. This application solves the problem that existing epoxy resin crosslinking simulation methods are difficult to realistically reproduce the actual crosslinking process of epoxy resin.

Description

Technical Field

[0001] This application relates to the field of molecular dynamics, and in particular to a method and apparatus for modeling the self-crosslinking of epoxy resins based on molecular dynamics. Background Technology

[0002] Epoxy resin possesses excellent mechanical properties, corrosion resistance, and bonding performance. When combined with carbon fiber, it forms carbon fiber reinforced composites, which combine high strength, high modulus, and lightweight characteristics, making them widely used in high-end manufacturing fields such as aerospace, rail transportation, and new energy equipment. The interfacial integrity between carbon fiber and epoxy resin is crucial to the overall performance of the composite material, as stress transfer in the composite structure depends entirely on interfacial interactions and bonding quality.

[0003] Currently, molecular dynamics simulations have become the mainstream method for studying the interfacial properties of composite materials. Common simulation methods simulate matrix formation by constructing a reaction system of epoxy resin and curing agent to accurately reproduce the actual crosslinking process of epoxy resin. However, most existing epoxy resin crosslinking simulation methods use a fixed, increasing cutoff distance during reaction search. This method ignores the spatial progressiveness of the reaction, potentially leading to unreasonable reaction sequences and network structures. Consequently, the efficiency of crosslinking simulation is low, making it difficult to accurately reproduce the actual crosslinking process of epoxy resin. This affects the accuracy of the predicted interfacial structure and mechanical properties of carbon fiber reinforced composites, and ultimately the reliability of composite material structural design and performance evaluation.

[0004] Therefore, there is an urgent need for a method and apparatus for modeling the self-crosslinking of epoxy resins based on molecular dynamics. Summary of the Invention

[0005] This application provides a molecular dynamics-based method and apparatus for modeling epoxy resin self-crosslinking, which solves the problems of low efficiency and difficulty in realistically reproducing the actual crosslinking process of epoxy resin in existing epoxy resin crosslinking simulation methods.

[0006] The first aspect of this application provides a molecular dynamics-based method for modeling the self-crosslinking of epoxy resin. The method includes: constructing epoxy resin monomer molecules, labeling target atoms in corresponding three-membered rings as reaction atom pairs, and constructing an amorphous structure model based on the reaction atom pairs; setting initial reaction parameters in a preset script, and loading the amorphous structure model through the preset script to form a molecular dynamics simulation system; the molecular dynamics simulation system undergoing energy minimization and kinetic equilibrium processing before entering a crosslinking cycle; searching for reaction atom pairs using a first cutoff radius as a constraint condition, and performing topological filtering on reaction atom pairs that satisfy the constraint condition to obtain target reaction atom pairs; prioritizing the target reaction atom pairs, and applying harmonic oscillator constraints to the target reaction atom pairs according to the ranking results; the harmonic oscillator constraints are used to model the self-crosslinking of epoxy resin. During the molecular dynamics evolution process, the corresponding reactive atoms are flexibly pulled closer to the target bonding distance; after removing the harmonic oscillator constraint, cross-linked chemical bonds are constructed in the target reactive atom pairs, and ring-opening operations of epoxy groups are performed simultaneously; the current cross-linking conversion rate value corresponding to the molecular dynamics simulation system is calculated, and it is determined whether the current cross-linking conversion rate value is greater than or equal to the preset cross-linking conversion rate threshold; if it is confirmed that the current cross-linking conversion rate value is less than the preset cross-linking conversion rate threshold, the first cutoff radius is expanded to the second cutoff radius, and the cross-linking cycle is returned to be executed; if it is confirmed that the current cross-linking conversion rate value is greater than or equal to the preset cross-linking conversion rate threshold, the cross-linking cycle is ended, abnormal coordinating atom detection and repair are performed, and geometric optimization and molecular dynamics equilibrium processing are performed on the repaired molecular dynamics simulation system to obtain the target epoxy resin self-cross-linking network structure model.

[0007] Optionally, an epoxy resin monomer molecule is constructed by labeling the target atoms in the corresponding three-membered ring as reaction atom pairs, and an amorphous structure model is constructed based on the reaction atom pairs. Specifically, this includes: constructing an epoxy resin monomer molecule model based on the chemical structure of the epoxy resin monomer molecule by using standard bond lengths and standard bond angles; in the epoxy resin monomer molecule model, the C atom on the outside of the three-membered ring is labeled as the first reaction atom, and the O atom connected to the C atom is labeled as the second reaction atom; the first reaction atom and the second reaction atom constitute a reaction atom pair; constructing a target unit cell based on the spatial size parameters of the epoxy resin monomer molecule; and filling the epoxy resin monomer molecule model into the target unit cell in a random manner according to a preset density or number, thereby constructing an amorphous structure model containing multiple reaction atom pairs.

[0008] Optionally, the molecular dynamics simulation system is subjected to energy minimization and kinetic equilibrium treatment, specifically including: minimizing the energy of the molecular dynamics simulation system by applying a heat bath temperature control to perform isothermal molecular dynamics simulation; and simulating the motion trajectory of each atom over time by performing time integration on the atomic motion equations in the molecular dynamics simulation system, thereby achieving kinetic equilibrium treatment of the molecular dynamics simulation system.

[0009] Optionally, the reaction atom pairs are searched using a first cutoff radius as a constraint, and topological filtering is performed on the reaction atom pairs that meet the constraint to obtain target reaction atom pairs. Specifically, this includes: calculating the interatomic spacing of the reaction atom pairs based on a preset script, and taking reaction atom pairs with an interatomic spacing smaller than the first cutoff radius as reaction atom pairs that meet the constraint; determining whether the reaction atom pairs that meet the constraint belong to the same molecule or the same crosslinking segment; filtering the reaction atom pairs that belong to the same molecule or the same crosslinking segment, and taking the remaining reaction atom pairs as target reaction atom pairs.

[0010] Optionally, cross-linked chemical bonds are constructed in the target reaction atom pairs, while simultaneously performing an epoxy group ring-opening operation. Specifically, this includes: constructing cross-linked chemical bonds in the target reaction atom pairs and using the Ewald algorithm to ensure uniform charge distribution in the cross-linked structure; calculating the target bonding positions of hydrogen atoms based on the epoxy oxygen atoms connected to the reaction carbon atoms, the CO bonds in the broken epoxy ring, and the dangling atoms generated by the marked ring-opening after the CO bonds are broken, and performing an epoxy group ring-opening operation based on the target bonding positions.

[0011] Optionally, the first cutoff radius is extended to a second cutoff radius, specifically including: determining the adjustment step size of the cutoff radius based on the reaction efficiency of the reaction atom pairs in the crosslinking cycle; and extending the first cutoff radius to a second cutoff radius according to the adjustment step size.

[0012] Optionally, if the current crosslinking conversion rate is confirmed to be greater than or equal to the preset crosslinking conversion rate threshold, the crosslinking cycle is terminated, abnormal coordination atom detection and repair are performed, and geometric optimization and molecular dynamics equilibrium processing are performed on the repaired molecular dynamics simulation system. Specifically, this includes: detecting the coordination number of atoms in the molecular dynamics simulation system, and judging whether there are abnormal coordination atoms that do not conform to the valence bond theory based on the detection results; if there are abnormal coordination atoms, the epoxy group ring-opening operation is performed again, and the molecular dynamics simulation system is repaired by retaining the newly formed crosslinking bonds.

[0013] A second aspect of this application provides a molecular dynamics-based epoxy resin self-crosslinking modeling device, the device comprising an acquisition module and a processing module, wherein, The acquisition module is used to construct epoxy resin monomer molecules to label the target atoms in the corresponding three-membered rings as reaction atom pairs, and to construct an amorphous structure model based on the reaction atom pairs; the initial reaction parameters are set in the preset script, and the amorphous structure model is loaded through the preset script to form a molecular dynamics simulation system; the molecular dynamics simulation system enters the crosslinking cycle after energy minimization and kinetic equilibrium processing.

[0014] The processing module searches for reaction atom pairs with a first cutoff radius as a constraint, and performs topological filtering on the reaction atom pairs that meet the constraint to obtain target reaction atom pairs. It prioritizes the target reaction atom pairs and applies harmonic oscillator constraints to each pair according to the ranking. The harmonic oscillator constraints are used to flexibly pull the corresponding reaction atoms closer to the target bonding distance during molecular dynamics evolution. After removing the harmonic oscillator constraints, cross-linking chemical bonds are constructed in the target reaction atom pairs, and ring-opening operations of epoxy groups are performed. The module calculates the current cross-linking conversion rate value corresponding to the molecular dynamics simulation system and determines whether the current cross-linking conversion rate value is greater than or equal to a preset cross-linking conversion rate threshold. If the current cross-linking conversion rate value is confirmed to be less than the preset cross-linking conversion rate threshold, the first cutoff radius is expanded to a second cutoff radius, and the cross-linking cycle is returned to execution. If the current cross-linking conversion rate value is confirmed to be greater than or equal to the preset cross-linking conversion rate threshold, the cross-linking cycle ends, abnormal coordination atom detection and repair are performed, and geometric optimization and molecular dynamics equilibrium processing are performed on the repaired molecular dynamics simulation system to obtain the target epoxy resin self-cross-linking network structure model.

[0015] A third aspect of this application provides an electronic device including a processor, a memory, a user interface, and a network interface. The memory is used to store instructions, the user interface and the network interface are used to communicate with other devices, and the processor is used to execute the instructions stored in the memory to cause the electronic device to perform the method as described above.

[0016] A fourth aspect of this application provides a non-transitory computer-readable storage medium storing a computer program, the computer program being executed by a processor using any of the methods described above.

[0017] One or more technical solutions provided in the embodiments of this application have at least the following technical effects or advantages: 1. Construct epoxy resin monomer molecules and label target atoms in the three-membered ring as reaction atom pairs. Based on the reaction atom pairs, construct an amorphous structure model and form a molecular dynamics simulation system. After energy minimization and kinetic equilibrium, enter the crosslinking cycle. Search and filter reaction atom pairs with the first cutoff radius to obtain the target reaction atom pairs. Apply harmonic oscillator constraints to flexibly pull the reaction atoms closer and construct crosslinking chemical bonds, while performing ring-opening operations on epoxy groups. Calculate the current crosslinking conversion rate value and compare it with the preset crosslinking conversion rate threshold. If it is less than the threshold, expand the cutoff radius to continue the crosslinking cycle. If it is greater than or equal to the threshold, perform abnormal coordination atom detection and repair. This truly restores the actual crosslinking process of epoxy resin and obtains a target epoxy resin self-crosslinking network structure model with a reasonable structure that can truly characterize the epoxy resin crosslinking process.

[0018] 2. Based on the chemical structure of epoxy resin monomer molecules, a model of the epoxy resin monomer molecule is constructed using standard bond lengths and standard bond angles. In the epoxy resin monomer molecule model, the C atom on the outside of the three-membered ring is marked as the first reactant atom, and the O atom connected to the C atom is marked as the second reactant atom. The first reactant atom and the second reactant atom constitute a reactant atom pair. A target unit cell is constructed based on the spatial size parameters of the epoxy resin monomer molecule. According to a preset density or number, the epoxy resin monomer molecule model is randomly filled into the target unit cell, and an amorphous structure model containing multiple reactant atom pairs is constructed, thereby providing a reasonable initial molecular configuration and reaction site distribution basis for subsequent crosslinking reaction search.

[0019] 3. By applying a heat bath temperature control to the molecular dynamics simulation system to perform isothermal molecular dynamics simulation, the energy of the molecular dynamics simulation system is minimized. By performing time integration on the atomic motion equations in the molecular dynamics simulation system to simulate the motion trajectory of each atom over time, the molecular dynamics simulation system is brought into dynamic equilibrium. This allows the system structure to gradually release the internal stress in the initial configuration and reach a stable thermodynamic state, providing a stable molecular structure basis for the accurate search and execution of subsequent crosslinking reactions. Attached Figure Description

[0020] Figure 1 This is a schematic flowchart of a molecular dynamics-based epoxy resin self-crosslinking modeling method provided in an embodiment of this application; Figure 2 This is a schematic diagram of a cross-linking script flow provided in an embodiment of this application; Figure 3 This is a schematic diagram of an epoxy resin crosslinking reaction mechanism provided in an embodiment of this application; Figure 4 This is a schematic diagram of an epoxy resin all-atom system provided in an embodiment of this application; Figure 5 This is a schematic diagram of a module of an epoxy resin self-crosslinking modeling device based on molecular dynamics provided in an embodiment of this application; Figure 6 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application.

[0021] Explanation of reference numerals in the attached figures: 51, acquisition module; 52, processing module; 601, processor; 602, communication bus; 603, user interface; 604, network interface; 605, memory. Detailed Implementation

[0022] To enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments.

[0023] The terminology used in the following embodiments of this application is for the purpose of describing particular embodiments only and is not intended to be limiting of this application. As used in the specification of this application, the singular expressions “a,” “an,” “the,” “the,” “the,” and “this” are intended to include the plural expressions as well, unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used in this application refers to and includes any or all possible combinations of one or more of the listed items.

[0024] Hereinafter, the terms "first" and "second" are used for descriptive purposes only and should not be construed as implying or suggesting relative importance or implicitly indicating the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature, and in the description of the embodiments of this application, unless otherwise stated, "multiple" means two or more.

[0025] To enable those skilled in the art to better understand the technical solution of the present invention, the present invention will be further described in detail below with reference to the accompanying drawings.

[0026] Please refer to Figure 1 The diagram illustrates a flowchart of a molecular dynamics-based epoxy resin self-crosslinking modeling method provided in this application embodiment. The flowchart mainly includes the following steps: S101 to S108.

[0027] Step S101: Construct epoxy resin monomer molecules to label the target atoms in the corresponding three-membered rings as reaction atom pairs, and construct an amorphous structure model based on the reaction atom pairs.

[0028] Existing methods for simulating epoxy resin crosslinking typically involve introducing curing agent molecules and relying on simple distance search strategies to screen reaction pairs. The reaction sequence is often directly determined by instantaneous spatial distance, lacking comprehensive consideration of the chemical environment and steric hindrance. This easily leads to an unreasonable crosslinking reaction sequence. Furthermore, the use of a fixed step size to expand the reaction distance during the search process results in low search efficiency and is prone to local structural distortions, making it difficult to obtain a crosslinking network model with a reasonable chemical structure and stable stress distribution. To address these shortcomings, this application introduces an intelligent scoring mechanism to optimize the reaction sequence and utilizes a dynamic step size strategy to improve search efficiency. Without introducing complex curing agent molecules, a chemically realistic epoxy resin crosslinking model with relaxed internal stress is constructed. To achieve this model construction, this application first uses MaterialsStudio molecular modeling software to draw the chemical structure of epoxy resin monomer molecules, with bisphenol A diglycidyl ether (DGEBA) as a typical monomer. Since the molecular structure has two three-membered rings, the C atoms on the outer sides of both three-membered rings are named R1, and the O atoms connected to the two C atoms are named R2. Then, using the Modules→AmorphousCell module in the software, molecules of a preset density are randomly placed into periodic bounding boxes to construct an amorphous structure model. The initial density is set to... This allows for volume shrinkage space during the crosslinking process.

[0029] In one possible implementation, step S101 further includes: constructing an epoxy resin monomer molecule model based on the chemical structure of the epoxy resin monomer molecule by using standard bond lengths and standard bond angles; in the epoxy resin monomer molecule model, marking the C atom on the outside of the three-membered ring as the first reactant atom and the O atom connected to the C atom as the second reactant atom; the first reactant atom and the second reactant atom constitute a reactant atom pair; constructing a target unit cell based on the spatial size parameters of the epoxy resin monomer molecule; filling the epoxy resin monomer molecule model into the target unit cell in a random manner according to a preset density or number, and constructing an amorphous structure model containing multiple reactant atom pairs.

[0030] Specifically, a three-dimensional structural model of the epoxy resin monomer molecule is first established based on its chemical structure information. During the construction process, standard bond lengths and bond angles are set according to the chemical bond relationships, and a complete epoxy resin monomer molecule model is generated through molecular modeling. The monomer molecule typically contains two epoxy three-membered ring structures. Each three-membered ring consists of two carbon atoms and one oxygen atom forming a closed structure. The three-membered ring structure exhibits high ring strain due to bond angle compression, thus possessing high reactivity during subsequent crosslinking.

[0031] After constructing the epoxy resin monomer molecular model, potential reaction sites within the monomer molecule are labeled. Specifically, the carbon atom on the outer side of each epoxy three-membered ring participating in the reaction is identified and labeled as the first reactant atom. Simultaneously, the oxygen atom directly bonded to this carbon atom is labeled as the second reactant atom, forming a reactant atom pair. This labeling method enables rapid identification of potential reaction sites during subsequent crosslinking reaction searches and provides a structural basis for the screening and bonding of reactant atom pairs.

[0032] After completing the reaction atom labeling, a target unit cell to accommodate multiple monomer molecules was constructed based on the spatial size parameters of the epoxy resin monomer molecules. The target unit cell serves as the periodic boundary space for subsequent molecular dynamics simulations. Its size is set according to the spatial size of the monomer molecules and the target density of the system to ensure that there is sufficient space within the unit cell to accommodate multiple monomer molecules and to reserve the necessary structural adjustment space for subsequent crosslinking reactions.

[0033] Subsequently, multiple epoxy resin monomer molecule models are randomly filled into the target unit cell according to a preset density or preset number of molecules. During the filling process, each monomer molecule is distributed in the unit cell space with random orientation and random position, thus forming a three-dimensional structural system with disordered molecular arrangement. Since the filled monomer molecule models all contain first and second reaction atoms, a molecular aggregate structure containing multiple reaction atom pairs will be formed in the target unit cell, ultimately constructing an amorphous structural model for subsequent crosslinking simulation calculations. This amorphous structural model can well simulate the molecular distribution state of the real epoxy resin system before curing and provides a basic structural environment for subsequent crosslinking reaction search, reaction pair screening, and crosslinking bonding process.

[0034] Step S102: Set the initial reaction parameters in the preset script, and load the amorphous structure model through the preset script to form a molecular dynamics simulation system.

[0035] For details, please refer to Figure 2 The document presents a schematic diagram of a crosslinking script flow provided in an embodiment of this application. Initial reaction parameters are set in a Perl script, an amorphous structure model is loaded through the script, and after energy minimization and kinetic equilibrium, the crosslinking cycle begins. The initial reaction parameters include, but are not limited to: target conversion rate, initial cutoff radius, initial step size, maximum cutoff radius, maximum number of iterations at each cutoff radius, as well as geometric optimization parameters and molecular dynamics simulation parameters.

[0036] In one possible implementation, step S102 further includes: performing isothermal molecular dynamics simulation by applying heat bath temperature control to the molecular dynamics simulation system, thereby minimizing the energy of the molecular dynamics simulation system; and performing dynamic equilibrium processing on the molecular dynamics simulation system by performing time integration on the atomic motion equations in the molecular dynamics simulation system to simulate the motion trajectory of each atom over time.

[0037] Specifically, after constructing the amorphous structure model in step S101, the amorphous structure model is loaded using a preset script to form a molecular dynamics simulation system for subsequent crosslinking simulations. Since randomly filled amorphous structure models often exhibit issues such as excessively small local interatomic spacing, bond angle configurations deviating from stable states, and excessively high local potential energy, directly entering the crosslinking cycle can easily lead to inaccurate identification of subsequent reactions or unreasonable local stress concentrations during bonding. Therefore, energy minimization and kinetic equilibrium processing are first performed on the molecular dynamics simulation system to gradually transition the system from its initial configuration to a more thermodynamically stable and structurally more rational initial state.

[0038] When minimizing the energy of a molecular dynamics simulation system, the main objective is to eliminate unreasonable close contacts, local structural distortions, and high-energy configurations resulting from random filling of amorphous structure models. This process can be combined with isothermal molecular dynamics simulations, allowing the system to gradually release locally excessive potential energy and complete initial relaxation under controlled temperature conditions. To ensure the system temperature remains stable near the target temperature, a heat bath temperature control is applied to the molecular dynamics simulation system. In this embodiment, an Andersen heat bath is used for isothermal control. During the simulation, each atom in the system undergoes statistical collisions with the heat bath according to a preset collision probability. When an atom is determined to have collided with the heat bath, its velocity is resampled to conform to a velocity distribution corresponding to the target temperature, thereby maintaining the entire system at the preset temperature. In this way, temperature fluctuations caused by local high-energy release can be suppressed, and the system can gradually complete structural relaxation under isothermal conditions, providing a stable initial state for subsequent dynamic equilibrium.

[0039] In the Andersen hot bath process, each atom has a probability of colliding with the hot bath at each time step. This probability is related to the collision frequency and the integration time step. After an atom collides, its velocity no longer simply continues the value from the previous moment, but is re-determined according to the Maxwell-Boltzmann distribution under the target temperature condition. This gradually makes the velocity statistical characteristics of the atomic population conform to the distribution characteristics of a thermal equilibrium system. The corresponding velocity distribution expression is:

[0040] in, This represents the atomic velocity vector, used to characterize the motion state of atoms in three-dimensional space; The atomic velocity is represented by a value. The probability density corresponding to the time is used to describe the statistical probability of different speeds occurring under target temperature conditions; This represents the atomic mass. Different types of atoms have different masses, and therefore their velocity distribution widths will also be different. This represents the Boltzmann constant, used to establish the relationship between the thermal motion of microscopic particles and macroscopic temperature; This represents the target temperature, which is usually set to a preset isothermal value according to the simulation task. The higher the target temperature, the wider the velocity distribution, indicating that the overall thermal motion of the atoms is more intense. This expression reflects the statistical law of atomic velocities under thermal equilibrium conditions. By resampling the velocities of colliding atoms according to this distribution, the system can be gradually made to approach the equilibrium state at the preset temperature.

[0041] After initial relaxation under isothermal constraints, the molecular dynamics simulation system is further subjected to kinetic equilibrium treatment. The core of kinetic equilibrium lies in tracking the evolution of each atom over time according to the fundamental laws of molecular dynamics, allowing the displacements, velocities, and interactions of the atoms in the system to gradually stabilize over time. In the kinetic equilibrium treatment, equations of motion are established and integrated for each atom in the system, and the atomic positions and velocities are iteratively updated through continuous time steps, thereby reproducing the actual trajectories of the atoms in the interaction potential field. The corresponding atomic equations of motion are:

[0042] in, Indicates the first The net force on each atom is derived from the combined contributions of bond stretching, bond angle interaction, nonbonding interaction, and electrostatic interaction. Indicates the first The mass of one atom; Indicates the first The position vector of each atom is used to describe the coordinate state of the atom in the three-dimensional simulation space; Indicates the first The second derivative of the position of an atom with respect to time corresponds to the acceleration of the atom. It represents the total potential energy of the system, reflecting the overall interaction energy of the system under all current atomic configurations; The total potential energy of the system is relative to the first... The gradient of each atom's position is used to characterize the direction in which the system's potential energy changes most rapidly when that atom's position changes slightly. The equation uses a negative sign because the actual force acting on an atom always tends towards the direction that lowers the system's potential energy, i.e., it moves along the opposite direction of the potential energy gradient. By integrating this equation step-by-step over time, the latest position and velocity of each atom can be obtained at each instant, thus forming a continuous atomic trajectory.

[0043] The dynamic equilibrium process is not simply about letting atoms move randomly, but rather about gradually adjusting the spatial configuration of each atom within the system along a more rational energy evolution path under a force field description. If some atoms experience significant forces due to local repulsion in a previous time step, these atoms will move in directions that reduce local repulsion and the overall potential energy of the system during subsequent integration. As bond lengths, bond angles, and non-bonded interactions gradually harmonize, the overall energy fluctuations, volume fluctuations, and local structural responses of the system will gradually decrease, eventually reaching a relatively stable dynamic equilibrium state. In other words, the integration process of the atomic motion equations effectively completes the dynamic transition from an initial randomly packed configuration to a thermodynamically acceptable configuration.

[0044] To establish a continuous technical chain between the aforementioned energy minimization and kinetic equilibrium treatments, an initial isothermal simulation is performed under controlled thermal bath conditions. This allows the system to quickly eliminate localized unreasonable contacts and stabilize the temperature distribution. Then, the atomic motion equations are continuously integrated onto the initially relaxed system, further adjusting the position and velocity distribution of each atom under a unified force field. After these two stages, the atomic configuration in the molecular dynamics simulation system no longer maintains the high-energy disordered state of the initial random filling but transforms into an initial crosslinking state suitable for performing cutoff radius search, topological filtering, and harmonic oscillator constraint convergence. Consequently, the candidate reaction atom pairs identified in subsequent crosslinking cycles will better conform to real spatial proximity relationships, and the risk of local distortion during bonding will be significantly reduced, thereby improving the chemical rationality and structural stability of the entire epoxy resin self-crosslinking modeling process.

[0045] Step S103: Search for reaction atom pairs with the first cutoff radius as a constraint, and perform topological filtering on the reaction atom pairs that meet the constraint to obtain the target reaction atom pairs.

[0046] Specifically, after the molecular dynamics simulation system enters the crosslinking cycle, a spatial proximity search is performed on the reaction atom pairs in the molecular dynamics simulation system using the first cutoff radius as a spatial constraint condition to identify candidate reaction atom pairs whose interatomic spacing satisfies the first cutoff radius condition.

[0047] After obtaining candidate reaction atom pairs, topological filtering is performed on the candidate reaction atom pairs to exclude atom combinations that do not meet the reaction conditions, thereby retaining reaction atom pairs that satisfy spatial proximity and topological rationality, and the filtered reaction atom pairs are determined as target reaction atom pairs for subsequent crosslinking reaction processing.

[0048] In one possible implementation, step S103 further includes: calculating the interatomic spacing of reaction atom pairs based on a preset script, and taking reaction atom pairs with an interatomic spacing smaller than the first cutoff radius as reaction atom pairs that satisfy the constraint conditions; determining whether reaction atom pairs that satisfy the constraint conditions belong to the same molecule or the same crosslinking segment; filtering reaction atom pairs that belong to the same molecule or the same crosslinking segment, and taking the remaining reaction atom pairs as target reaction atom pairs.

[0049] Specifically, after the molecular dynamics simulation system enters the crosslinking cycle, the pre-defined script first performs a spatial proximity screening around the labeled reaction atom pairs to determine the atomic combinations with reaction potential within the current search range. This screening does not indiscriminately traverse all atoms in the system, but rather calculates the directional distance between the labeled first and second reaction atoms, thereby improving the specificity of reaction site identification and ensuring that subsequent crosslinking searches always revolve around the pre-defined reaction sites in the epoxy resin monomer molecules.

[0050] Because molecular dynamics simulations employ periodic boundary conditions, periodic mirror images of atoms exist outside the simulation box boundary corresponding to those inside. Therefore, when calculating the spatial distance between reaction atom pairs, it's insufficient to rely solely on the linear coordinate differences within the current simulation box; otherwise, geometrically close atom pairs that are adjacent across the boundary may be overlooked. To address this, a pre-defined script calculates the positional displacements between candidate atoms in each spatial dimension and performs minimum mirror correction based on the simulation box length in the corresponding dimension. This ensures that the obtained displacements always correspond to the shortest spatial separation path between two atoms under periodic boundary conditions. After determining the minimum mirror displacements in each dimension, the three-dimensional displacement components are synthesized to obtain the interatomic distance between reaction atom pairs. The resulting distances accurately reflect the spatial proximity between reaction sites in the periodic simulation system. The calculation method is as follows: First, in each spatial dimension... Calculate the displacement above:

[0051] in, The first in the reaction atom pair The atom and the first Atoms in spatial dimension The coordinate displacement on the top, For the first Atoms in spatial dimension The coordinates on, For the first Atoms in spatial dimension The coordinate values ​​in that dimension are then calculated. The minimum mirror displacement in that dimension is then calculated using the formula:

[0052] in, The first in the reaction atom pair The atom and the first Atoms in spatial dimension The displacement after minimum mirror correction. To simulate the box in spatial dimension The length of the upper side, This is the floor function. The final distance is:

[0053] in, Indicates the first in the reaction atom pair The atom and the first The interatomic spacing between atoms express, express, This means that after obtaining the interatomic spacing, it is compared with the first cutoff radius corresponding to the current crosslinking cycle. Only when the interatomic spacing... Only when the distance is less than the first cutoff radius is the corresponding reaction atom pair determined as a candidate reaction atom pair satisfying the constraint conditions. This method limits the search range of the current crosslinking cycle to a local spatial neighborhood, ensuring that subsequently selected atom pairs have the geometrical potential to form a crosslinking reaction. Simultaneously, it avoids including atom combinations that are too far apart and difficult to bond in the current cycle into the reaction set, thereby improving the effectiveness and computational efficiency of the crosslinking search. Here, the first cutoff radius represents the spatial distance threshold used to limit the search range of reaction atom pairs in the current crosslinking cycle.

[0054] After selecting candidate reaction atom pairs that meet the constraints, a topological filtering process is further performed on these pairs. The primary purpose of topological filtering is to determine whether the two atoms in a candidate reaction atom pair belong to the same molecule or the same cross-linking segment. If the two atoms are already located in the same topological unit, it indicates that the atom pair belongs to an intramolecular contact or a local contact after an established connection, and is not suitable as a new cross-linking reaction target. For such atom pairs, they are directly filtered from the candidate set to avoid repeatedly establishing unreasonable connections within the same molecule, or introducing additional bonds within an already formed cross-linking segment that would disrupt the chemical rationality of the structure.

[0055] Beyond determining whether atoms belong to the same molecule or crosslinking fragment, topological filtering can further constrain the process by considering the reaction execution status of the current crosslinking cycle. For example, by using an atom-locking mechanism, each atom can participate in at most one crosslinking reaction in the current crosslinking cycle, preventing the same atom from forming connections with multiple atoms simultaneously in a single search, thus avoiding coordination anomalies or excessive crosslinking of local structures. Simultaneously, based on the current crosslinking conversion progress of the system, candidate reaction atom pairs that might cause the overall conversion rate to exceed a preset target can be excluded, ensuring that the crosslinking process progresses gradually within a preset range. Through these multiple topological constraints, atom combinations that satisfy spatial distance conditions but not connection relationship constraints or reaction progress constraints can be further eliminated.

[0056] After topological filtering, the remaining unfiltered reaction atom pairs are identified as target reaction atom pairs. These target reaction atom pairs satisfy both spatial proximity and topological rationality requirements, thus serving as the basis for subsequent priority ranking, harmonic oscillator constraint application, and cross-linking chemical bond construction.

[0057] Step S104: Prioritize the target reaction atom pairs and apply harmonic oscillator constraints to each target reaction atom pair according to the prioritization results.

[0058] Specifically, after obtaining the target reaction atom pairs in step S103, the subsequent convergence process is not performed directly in an arbitrary order. Instead, the target reaction atom pairs are first prioritized to ensure that reaction atom pairs that better match the current chemical environment and spatial conditions enter the crosslinking preparation stage first. This is because if processing is performed solely based on random order or distance, locally less desirable reaction atom pairs may be preferentially converged, causing more likely reaction combinations to lag behind, thus affecting the formation order of the crosslinking network structure and the overall structural stability. Therefore, an intelligent priority score is first calculated for the filtered reaction atom pairs, and then the reaction processing order is determined based on the score results.

[0059] Priority scoring comprehensively considers the spatial proximity between reactant atom pairs, the degree of matching of local chemical environments, steric hindrance, and the iterative stage of the current crosslinking cycle, so that the ranking results more closely reflect the tendency of reaction sites to occur during actual crosslinking. The expression for priority scoring is:

[0060] in, Indicates the first The atom and the first The reaction priority of a target reaction atom pair consisting of 1 atom is indicated by the larger the value, the more preferentially the reaction atom pair should enter the subsequent flexible approach process. This represents the interatomic distance between two atoms in the target reaction atom pair. The smaller the interatomic distance, the higher the spatial proximity, and the larger the corresponding exponential decay term, indicating that the atom pair is more likely to approach and form bonds in the current crosslinking cycle. This represents a chemical environment scoring function used to characterize the degree of matching between two reacting atoms in the current local chemical environment; The steric hindrance score function is used to characterize whether the reacting atom has a strong configurational hindrance to the surrounding space. If the local steric hindrance is small, the score is high, indicating that the reacting atom is more likely to move closer together in space. This indicates the current cycle number. As the cycle number increases, the exponent term will have a phased adjustment effect on the priority, making the reaction execution rhythm more balanced under different crosslinking stages. The principle of this formula is to multiplicatively couple spatial distance factors, chemical environment factors, steric hindrance factors, and cycle stage factors. Only when multiple conditions are favorable together will the target reaction atom pair obtain a higher comprehensive priority, thus entering the priority processing sequence.

[0061] The expression for the chemical environment scoring function is as follows: in, Indicates the first The atom and the first The charge difference between atoms is used to reflect the difference in electrical properties between two reacting atoms in the current local chemical environment; This represents a preset charge difference threshold, used to adjust the sensitivity to chemical environmental influences. When the relationship with the set threshold better matches the expected reaction requirements, the corresponding chemical environment score will increase. This function uses an S-shaped mapping relationship to convert the original charge difference into a smoothly changing score value, thereby avoiding sudden priority changes caused by small perturbations in local charge difference, and enabling chemical environment factors to participate in the overall ranking in a more stable manner.

[0062] After priority scoring is completed, all target reaction atom pairs are sorted according to the scoring results, and harmonic oscillator constraints are applied sequentially based on the sorting results. Harmonic oscillator constraints do not immediately establish real chemical bonds between the reacting atoms. Instead, they first use controlled distance constraints to gradually bring the reacting atoms closer to the target bonding distance during molecular dynamics evolution. This is designed to prevent two atoms from jumping directly from their initial separated state to the ideal bond length state. Otherwise, it could easily lead to simultaneous and drastic perturbations in local bond lengths, bond angles, and non-bonded interactions, resulting in large instantaneous stresses in the system and affecting the stability of the cross-linked structure.

[0063] In this embodiment, the harmonic oscillator constraint is applied in a multi-stage, progressive manner. That is, multiple constraint stages are set for the ordered target reactive atom pairs. In the previous stage, a lower constraint strength guides the reactive atoms to approach each other. In the subsequent stage, the constraint strength is gradually increased, and the constraint equilibrium distance is simultaneously reduced from the current interatomic distance to the target bond length. In this way, the reactive atoms and their surrounding local structure can gradually complete positional adjustments and configurational adaptations during each stage of molecular dynamics evolution, thereby maintaining the continuous response of the surrounding structure while approaching bonding conditions and reducing local stress concentration caused by sudden displacements.

[0064] The potential energy function expression for the harmonic oscillator constraint is: in, This represents the actual distance between atoms at present. The corresponding constraint potential energy; This represents the constraint force constant, used to characterize the pull-back strength of the harmonic oscillator constraint when the interatomic distance deviates from the target value. The larger the value, the more obvious the recovery effect when the interatomic distance deviates from the target distance. This represents the actual distance between particles, that is, the real-time distance between the two reacting atoms in the current target reacting atom pair during the simulation process; This parameter represents the target distance between particles, used to characterize the equilibrium distance that is desired to be maintained in the current constraint stage. During the multi-stage convergence process, this parameter gradually decreases from an initial large distance to the target bonding distance. The principle behind this expression is that the distance deviation between the target reactive atom pairs is converted into a potential energy cost. When the actual distance deviates from the target distance, the system will be driven by the potential energy to cause the atoms to return to the target distance, thereby achieving a smooth, continuous, and controllable distance convergence process.

[0065] Therefore, by first performing priority sorting and then applying harmonic oscillator constraints according to the sorting results, the target reaction atom pairs can enter the closing stage in a more reasonable order and gradually and flexibly approach the target bonding distance during molecular dynamics evolution. This not only improves the chemical rationality of the crosslinking reaction execution order but also provides a stable geometric basis for the subsequent removal of constraints and the formation of crosslinking chemical bonds between corresponding reaction atom pairs.

[0066] In step S105, after removing the harmonic oscillator constraint, cross-linked chemical bonds are constructed in the target reaction atom pairs, while simultaneously performing ring-opening operations on the epoxy groups.

[0067] Specifically, the ring-opening operation of the oxygen group is used to break the CO bond in the epoxy three-membered ring, so that the epoxy group changes from a closed ring structure to an open ring structure, and the resulting suspended atoms are hydrogenated and saturated after ring opening.

[0068] In one possible implementation, step S105 further includes: constructing cross-linked chemical bonds in the target reaction atom pairs and using the Ewald algorithm to ensure uniform charge distribution in the cross-linked structure; calculating the target bonding positions of hydrogen atoms based on the epoxy oxygen atoms connected to the reaction carbon atoms, the CO bonds in the broken epoxy ring, and the dangling atoms generated by the marked ring opening after the CO bonds are broken, and performing an epoxy group ring-opening operation based on the target bonding positions.

[0069] For details, please refer to Figure 3 The document presents a schematic diagram of an epoxy resin crosslinking reaction mechanism provided in an embodiment of this application. In step S104, the two reacting atoms in the target reacting atom pair are gradually brought closer to the target bonding distance during the molecular dynamics evolution through the harmonic oscillator constraint. Therefore, when entering step S105, the target reacting atom pair already possesses the geometric conditions for forming actual crosslinking chemical bonds. At this point, the aforementioned harmonic oscillator constraint is removed first, and then actual crosslinking chemical bonds are generated between the target reacting atom pairs, transforming the controlled approach process of the previous stage into a real chemical connection process. By applying constraints first and then removing them and forming bonds, the local bond length abrupt changes, bond angle distortions, and stress concentrations caused by the direct jump of reacting atoms from the initial separation state to the bonding state can be avoided, thus making the formation process of the crosslinked structure smoother.

[0070] After cross-linking chemical bonds are formed between the target reactant atom pairs, the atomic connections in the system change, and the local charge distribution and long-range electrostatic interactions also change accordingly. Since the molecular dynamics simulation system uses periodic boundary conditions, any charged atom in the system interacts not only with other atoms in the current simulation box but also with charges in the periodic mirror image. Therefore, it is necessary to perform stable calculations of the electrostatic energy in the cross-linked structure to ensure that the updated system has a reasonable charge distribution. In this embodiment, the Ewald algorithm is used to sum the electrostatic interactions in the cross-linked structure. The Ewald algorithm decomposes the originally poorly convergent long-range Coulomb summation into a short-range part in real space, a long-range part in reciprocal space, and a self-energy correction part, thereby transforming the difficult-to-handle electrostatic summation problem into a stable and convergent computational form.

[0071] The total electrostatic energy decomposes into: in, This represents the total electrostatic energy of the cross-linked molecular dynamics simulation system; This represents the real space term, used to characterize the contribution of short-range electrostatic interactions between atoms in the real space region; This represents the reciprocal space term, used to characterize the contribution of long-range electrostatic interactions in the reciprocal space of a periodic system; The term represents the self-energy, used to correct the additional energy error generated between each charge and its own Gaussian distribution during the division and summation process. The principle of this decomposition method is to divide the Coulomb interaction, which originally decays slowly in a periodic system, into domains, so that short-range interactions converge rapidly in real space, and long-range interactions accumulate effectively in reciprocal space. Then, the error is corrected by the self-energy term, thereby obtaining a stable and physically meaningful total electrostatic energy result.

[0072] Real space item The expression is: in, Indicates the first The charge value of each atom; Indicates the first The charge value of each atom; Indicates the first The atom and the first The interatomic distance between atoms; Represents an integer vector used to characterize the translation index of a periodic mirror box; This represents the side length of the simulated box; This represents the complementary error function, used to attenuate long-range interactions, making the summation in real space converge faster. This represents the Ewald partitioning parameter, used to adjust the convergence balance between the real space and the reciprocal space. The principle behind this expression is that it preserves only the attenuated short-range electrostatic interactions in the real space, thereby avoiding the inefficient summation of the full Coulomb potential.

[0073] Inverted space terms The expression is: in, This represents the volume of the simulated box; Represents the reciprocal space vector, used to characterize the wave vector component in the reciprocal space; Represents the squared magnitude of a vector in reciprocal space; This represents the decay factor controlled by the Ewald partitioning parameter, used to adjust the convergence rate of the reciprocal space term; The structure factor is used to characterize the distribution of all charges in the system in the reciprocal space. The principle behind this expression is that it maps long-range electrostatic interactions in a periodic system to the reciprocal space and sums them in the frequency domain, thereby efficiently characterizing charge coupling on a large scale.

[0074] structural factors The expression is: in, Indicates the first The charge value of each atom; Indicates the first Position vectors of each atom; This represents the dot product between the reciprocal space vector and the atomic position vector; The imaginary unit is represented by . The principle behind this expression is that the positional distribution and charge distribution of each charged atom are projected together onto the reciprocal space, so that the charge contributions at different spatial locations can participate in the calculation of long-range electrostatic energy in the form of phase superposition.

[0075] Self-ability The expression is: in, Indicates the Ewald splitting parameters; Indicates the first The expression calculates the charge value of each atom. The principle behind this expression is to subtract and correct for the additional energy introduced by each atom during the Ewald splitting process due to its own smooth charge distribution, thus preventing this non-physical contribution from being repeatedly included in the total electrostatic energy. Through the combined processing of the real space term, reciprocal space term, and self-energy term, the charge distribution calculation in the cross-linked structure can be made more uniform and stable, providing a reliable electrostatic environment for subsequent ring opening and structural relaxation.

[0076] After completing the formation of crosslinking chemical bonds and updating the electrostatic environment following crosslinking, a ring-opening operation of the epoxy groups is further performed. This ring-opening operation is not simply deleting the original linkages, but rather involves directional recognition of the epoxy oxygen atoms bonded to the reacting carbon atoms, and breaking the CO bonds in the epoxy ring. This operation is necessary because the three-membered rings in the epoxy resin monomer should transition from a closed to an open state during crosslinking. Only after breaking the corresponding CO bonds does the crosslinking result conform to the chemical structure characteristics of the epoxy groups after the actual reaction. If only new crosslinking bonds are formed without performing the three-membered ring-opening operation, it will lead to abnormal local coordination relationships, thus affecting the chemical rationality of the entire crosslinking network.

[0077] After the CO bond in the epoxy ring breaks, the corresponding atoms form dangling atoms that do not meet the normal valence state. To restore the stability of the local structure after ring opening, the dangling atoms need to be hydrogenated to saturate. However, if hydrogen is added only in a fixed direction, the hydrogen atom positions may deviate from the actual chemical environment, leading to local configurational distortion. Therefore, in this embodiment, instead of directly specifying the hydrogen atom positions, the target bonding positions of the hydrogen atoms are optimized by comprehensively considering the initial configurational orientation of the dangling atoms, the local electrostatic potential distribution, and the van der Waals potential distribution, so that the newly added hydrogen atoms can be placed in spatial positions that better conform to the local chemical environment.

[0078] The expression for the target bonding position vector of a hydrogen atom is: in, The vector representing the target bonding position of a hydrogen atom is used to characterize the spatial coordinates of a hydrogen atom after hydrogen saturation. This represents the position vector of the suspended atom, used to characterize the current position of the atom to be hydrogenated; Indicates the XH bond length, used to define the target bonding distance between the dangling atom and the hydrogen atom to which it is added; This represents the initial direction vector, used to provide a basic directional reference for the arrangement of hydrogen atoms; Represents local electrostatic potential. It indicates the gradient direction of the local electrostatic potential, which is used to reflect the more favorable arrangement direction of hydrogen atoms under the influence of electrostatic effects; This indicates Van der Waals's influence. It represents the gradient direction of the van der Waals potential, which reflects the influence of surrounding atoms on the orientation of hydrogen atoms under spatial repulsion and attraction. This represents the electrostatic potential weighting coefficient, used to adjust the degree of influence of the electrostatic environment on the target bonding direction; The van der Waals potential weighting coefficient is used to adjust the influence of the van der Waals environment on the target bonding direction. The modulus term in the denominator is used to normalize the direction vector, so that it retains only the direction information without changing the final XH bond length. The principle of this expression is that, starting from the position of the suspended atom, the combined effect of the electrostatic potential gradient direction and the van der Waals potential gradient direction is superimposed on the initial direction. Then, through normalization and scaling of the target bond length, the final bonding position of the hydrogen atom is obtained, thus taking into account both local electrical matching and the rationality of spatial steric hindrance.

[0079] Therefore, in step S105, the harmonic oscillator constraint is first removed, and real cross-linking chemical bonds are established between the target reactant atom pairs. Then, the long-range electrostatic environment in the cross-linked structure is updated using the Ewald algorithm. Subsequently, the epoxy oxygen atoms connected to the reactant carbon atoms are identified, and the CO bonds in the epoxy ring are broken. Finally, the dangling atoms generated by ring opening are subjected to hydrogenation saturation based on the target bonding positions. Through the above continuous processing, the local connectivity, charge environment, and post-ring-opening valence state relationships in the molecular dynamics simulation system can simultaneously reach a reasonable state, thereby completing the actual bonding and structural correction process in a single cross-linking cycle and providing a reliable structural basis for subsequent cross-linking conversion rate calculations and the next round of cutoff radius adjustment.

[0080] Step S106: Calculate the current crosslinking conversion rate value corresponding to the molecular dynamics simulation system, and determine whether the current crosslinking conversion rate value is greater than or equal to the preset crosslinking conversion rate threshold.

[0081] Specifically, after the formation of cross-linking chemical bonds and the ring-opening treatment of epoxy groups are completed in step S105, some reaction atom pairs in the molecular dynamics simulation system have been converted into actual cross-linking bonds. Therefore, it is necessary to quantitatively evaluate the current cross-linking progress of the system to determine whether the cross-linking modeling process has achieved the preset goal. To this end, the cross-linking conversion rate is calculated for the molecular dynamics simulation system. By statistically analyzing the relationship between the number of currently formed cross-linking bonds and the number of theoretically reaction atom pairs in the system, the current cross-linking conversion rate value of the molecular dynamics simulation system is obtained.

[0082] During the calculation, the cross-linked chemical bonds already formed in the system after the current cross-linking cycle are first identified, and the corresponding number of cross-linked bonds is counted. Simultaneously, based on the number of reaction atom pairs marked in the initially constructed amorphous structure model, the total number of theoretically viable reaction sites in the system is determined. Based on this, the cross-linking conversion rate of the molecular dynamics simulation system after the current cross-linking cycle is obtained by calculating the ratio of the number of currently formed cross-linked bonds to the number of theoretically viable reaction sites. The cross-linking conversion rate characterizes the overall degree to which reaction sites in the system transform from an unreacted state to a cross-linked connected state; a higher value indicates a higher degree of cross-linking in the system.

[0083] After obtaining the current crosslinking conversion rate value, it is compared with a preset crosslinking conversion rate threshold. If the current crosslinking conversion rate value is less than the preset threshold, it indicates that the current crosslinking cycle has not yet reached the target crosslinking degree, and subsequent crosslinking cycles need to be continued by expanding the reaction search range. If the current crosslinking conversion rate value is greater than or equal to the preset threshold, it indicates that the crosslinking reaction in the system has reached the expected modeling target. At this point, the crosslinking cycle ends, and the subsequent structural repair and relaxation treatment stage begins, thus completing the construction process of the epoxy resin self-crosslinking structure model.

[0084] Step S107: If it is confirmed that the current crosslinking conversion rate value is less than the preset crosslinking conversion rate threshold, the first cutoff radius is expanded to the second cutoff radius, and the crosslinking cycle is returned to be executed.

[0085] In one possible implementation, step S107 further includes: determining an adjustment step size for the cutoff radius based on the reaction efficiency of the reaction atoms in the crosslinking cycle; and expanding the first cutoff radius to a second cutoff radius according to the adjustment step size.

[0086] Specifically, when it is determined in step S106 that the current crosslinking conversion rate value has not yet reached the preset crosslinking conversion rate threshold, it indicates that the effective crosslinking reaction that can be achieved under the first cutoff radius constraint within the current crosslinking cycle is still limited, and it is necessary to expand the reaction search range for the next crosslinking cycle. To avoid the problems of low search efficiency or over-expansion in the later stages caused by expanding the cutoff radius with a fixed step size, in this embodiment, the cutoff radius is not directly adjusted according to a constant increment, but the adjustment step size of the cutoff radius is adaptively determined based on the reaction efficiency of the reaction atoms in the crosslinking cycle, so that the expansion process of the cutoff radius can match the current crosslinking progress state.

[0087] The reaction efficiency of atom pairs reflects the degree to which candidate atom pairs in the current crosslinking cycle are successfully converted into actual crosslinking bonds. When a large number of candidate atom pairs in the current crosslinking cycle can successfully form crosslinking bonds, it indicates that there is still strong reactivity near the existing search range. In this case, the adjustment step size of the cutoff radius can be appropriately increased to accelerate the next round of reaction site search. When the actual conversion effect of candidate atom pairs in the current crosslinking cycle is weak, it indicates that the spatial distribution of the remaining reactive sites in the system is already relatively sparse. If a large step size is still used to expand the cutoff radius, it is easy to introduce too many low-quality candidate atom pairs in subsequent searches. Therefore, it is necessary to reduce the adjustment step size of the cutoff radius to more finely advance the crosslinking search process. Thus, the expansion of the cutoff radius can take into account both the needs of rapid search in the early stage and fine control in the later stage.

[0088] In this embodiment, the adjustment step size can be determined according to the following expression: in, Indicates the first The cutoff radius adjustment step size corresponding to the cross-linking cycle is used to characterize the distance increase when expanding from the current cutoff radius to the next round cutoff radius; This represents the initial step size, which is used as the basic step size parameter for the entire cutoff radius adjustment process. Indicates the first The number of cross-linking bonds successfully formed in each iteration is used to characterize the number of target reaction atom pairs that actually complete the reaction in the current cross-linking cycle. Indicates the first The total number of possible reaction atom pairs in the next iteration is used to characterize the total size of reaction atom pairs included in the candidate set within the current crosslinking cycle; This indicates the reaction efficiency of the reaction atom pairs in the current crosslinking cycle. The higher the value, the higher the proportion of candidate reaction atom pairs that are successfully converted into crosslinking bonds. This represents the hyperbolic tangent function, used to smoothly map the adjustment result corresponding to the reaction efficiency to a finite range, thus avoiding abrupt changes in the adjustment step size. The principle behind this expression is that an initial step size is used as a reference value, and this reference value is nonlinearly amplified or reduced according to the current reaction efficiency, so that the adjustment step size of the cutoff radius can change continuously with the change of the crosslinking reaction state.

[0089] After obtaining the adjustment step size, the first cutoff radius is expanded to a second cutoff radius based on the adjustment step size. The second cutoff radius corresponds to the search range of reaction atom pairs in the next crosslinking cycle. Compared to the first cutoff radius, it covers a larger spatial neighborhood, thereby enabling the identification of more potential reaction sites in the molecular dynamics simulation system. This provides updated search boundaries for subsequent execution of reaction atom pair searches, topological filtering, priority ranking, harmonic oscillator constraint application, and crosslinking bond construction. In this way, the crosslinking cycle can progressively advance the cutoff radius expansion process according to an adaptive step size when the target crosslinking conversion rate is not reached, thereby improving the search efficiency and crosslinking rationality of epoxy resin self-crosslinking modeling.

[0090] Step S108: If it is confirmed that the current crosslinking conversion rate value is greater than or equal to the preset crosslinking conversion rate threshold, the crosslinking cycle ends, abnormal coordination atom detection and repair are performed, and geometric optimization and molecular dynamics equilibrium processing are performed on the repaired molecular dynamics simulation system to obtain the target epoxy resin self-crosslinking network structure model.

[0091] Specifically, when the current crosslinking conversion rate reaches or exceeds the preset crosslinking conversion rate threshold, it indicates that the reaction sites in the molecular dynamics simulation system have completed the expected degree of crosslinking reaction, and the crosslinking cycle ends at this point. Subsequently, abnormal coordination atom detection and repair processing is performed on the molecular dynamics simulation system to identify and correct any abnormal coordination structures that may arise during the crosslinking process, ensuring that the bonding relationships of each atom in the system meet reasonable chemical valence state constraints. After completing the abnormal coordination repair, geometric optimization and molecular dynamics equilibrium processing are performed on the repaired molecular dynamics simulation system to further relax the system structure energy and reach a stable equilibrium state, thereby obtaining a structurally stable and chemically reasonable target epoxy resin self-crosslinking network structure model.

[0092] In one possible implementation, step S108 further includes: detecting the coordination number of atoms in the molecular dynamics simulation system, and determining whether there are abnormally coordinated atoms that do not conform to the valence bond theory based on the detection results; if there are abnormally coordinated atoms, performing the ring-opening operation of the epoxy group again, and repairing the molecular dynamics simulation system by retaining the newly formed crosslinked bonds.

[0093] Specifically, after the crosslinking cycle ends, to ensure that the final epoxy resin self-crosslinking network structure model satisfies valence bond theory and maintains reasonable atomic coordination relationships in its chemical structure, it is necessary to detect the atomic coordination state in the molecular dynamics simulation system. By statistically analyzing the coordination number of each atom in the system and comparing it with the normal coordination number of the corresponding element under valence bond theory, the presence of abnormally coordinated atoms can be identified. When an atomic coordination number is detected that does not meet the requirements of the normal valence state, it indicates that an abnormal or incompletely open ring-opening structure may have occurred during the crosslinking reaction or ring-opening process, and the corresponding structure needs to be repaired.

[0094] After detecting anomalous coordinating atoms, the ring-opening operation of the epoxy group is performed again to repair the anomalous coordination structure. During this process, the already formed cross-linked chemical bonds are preserved, and only the original CO bonds that did not participate in the cross-linking pathway are broken, allowing the three-membered ring structure to complete ring-opening according to the normal reaction pathway, thereby eliminating the anomalous coordination state. For example, in some cases, the epoxy group may only form cross-linked connections but not completely break the original CO bonds in the three-membered ring; in this case, the oxygen atom may exhibit a three-coordinate structure. By re-performing the ring-opening operation and breaking the CO bonds in the non-cross-linked pathways, the corresponding oxygen atom can be restored to a coordination state consistent with valence bond theory, thereby achieving structural repair.

[0095] After correcting aberrant coordination, the charge distribution of atoms within the system needs to be updated to ensure the accuracy of electrostatic interaction calculations during subsequent structure optimization. In this embodiment, the force field calculation module in MaterialsStudio is invoked to redistribute partial charges to all atoms in the molecular dynamics simulation system based on the charge distribution scheme constructed within the selected force field. This process ensures a uniform update of the local charge redistribution caused by crosslinking and ring-opening reactions, thereby guaranteeing that the overall charge distribution of the system is consistent with the new molecular topology.

[0096] After charge redistribution, the repaired molecular dynamics simulation system underwent geometric optimization to gradually converge to a stable configuration on the potential energy surface. During geometric optimization, atomic positions were continuously adjusted to reduce the total potential energy of the system, while the net force on each atom was monitored. When the system reaches a stable state, the net force on each atom tends to be relatively small. Whether the structure has reached a stable convergent state can be evaluated using the root-mean-square force, the expression of which is:

[0097] in, It represents the root mean square value of the net force on all atoms in the system and is used to measure whether the structure has reached a stable equilibrium state. This represents the total number of atoms in the system; Indicates the first The net force acting on each atom; This represents the preset convergence accuracy threshold. When the root-mean-square force is less than the convergence accuracy threshold, it indicates that the system architecture has essentially reached a stable state. The principle behind this expression is that by averaging the squares of the forces acting on all atoms and then taking the square root, the average level of the residual forces within the system can be reflected as a whole. When this value is sufficiently small, it indicates that the system is in a stable configuration near the potential minimum.

[0098] After geometric optimization, molecular dynamics equilibrium treatment was further applied to the molecular dynamics simulation system, allowing the system to continue evolving under target temperature conditions and release residual stress, thereby further stabilizing the crosslinked network structure. Through the above-mentioned abnormal coordination detection and repair, charge redistribution, geometric optimization, and molecular dynamics equilibrium treatment, a target epoxy resin self-crosslinked network structure model with stable structure, reasonable coordination relationship, and satisfying valence bond theory constraints was finally obtained. Please refer to [reference needed]. Figure 4 The document presents a schematic diagram of an epoxy resin all-atom system provided in an embodiment of this application.

[0099] Please refer to Figure 5 This illustration shows a schematic diagram of a molecular dynamics-based epoxy resin self-crosslinking modeling device provided in an embodiment of this application. The device includes an acquisition module 51 and a processing module 52, wherein... The acquisition module 51 is used to construct epoxy resin monomer molecules to mark the target atoms in the corresponding three-membered rings as reaction atom pairs, and to construct an amorphous structure model based on the reaction atom pairs; the initial reaction parameters are set in the preset script, and the amorphous structure model is loaded through the preset script to form a molecular dynamics simulation system; the molecular dynamics simulation system enters the crosslinking cycle after energy minimization and kinetic equilibrium processing.

[0100] Processing module 52 is used to search for reaction atom pairs with a first cutoff radius as a constraint, and perform topological filtering on the reaction atom pairs that meet the constraint to obtain target reaction atom pairs; prioritize the target reaction atom pairs, and apply harmonic oscillator constraints to the target reaction atom pairs according to the ranking results; the harmonic oscillator constraints are used to flexibly pull the corresponding reaction atoms closer to the target bonding distance during molecular dynamics evolution; after removing the harmonic oscillator constraints, cross-linked chemical bonds are constructed in the target reaction atom pairs, and ring-opening operations of epoxy groups are performed simultaneously; the current cross-linking conversion rate value corresponding to the molecular dynamics simulation system is calculated, and it is determined whether the current cross-linking conversion rate value is greater than or equal to the preset cross-linking conversion rate threshold; if it is confirmed that the current cross-linking conversion rate value is less than the preset cross-linking conversion rate threshold, the first cutoff radius is expanded to the second cutoff radius, and the cross-linking cycle is returned; if it is confirmed that the current cross-linking conversion rate value is greater than or equal to the preset cross-linking conversion rate threshold, the cross-linking cycle is ended, abnormal coordination atom detection and repair are performed, and geometric optimization and molecular dynamics equilibrium processing are performed on the repaired molecular dynamics simulation system to obtain the target epoxy resin self-cross-linking network structure model.

[0101] In one possible implementation, the acquisition module 51 is used to construct epoxy resin monomer molecules to label target atoms in the corresponding three-membered rings as reaction atom pairs, and to construct an amorphous structure model based on the reaction atom pairs. Specifically, this includes: constructing an epoxy resin monomer molecule model based on the chemical structure of the epoxy resin monomer molecule by using standard bond lengths and standard bond angles; in the epoxy resin monomer molecule model, labeling the C atoms on the outside of the three-membered rings as first reaction atoms, and labeling the O atoms connected to the C atoms as second reaction atoms; the first reaction atoms and the second reaction atoms constitute reaction atom pairs; constructing a target unit cell based on the spatial size parameters of the epoxy resin monomer molecule; and filling the epoxy resin monomer molecule model into the target unit cell in a random manner according to a preset density or quantity, and constructing an amorphous structure model containing multiple reaction atom pairs.

[0102] In one possible implementation, the acquisition module 51 is used to perform energy minimization and kinetic equilibrium processing on the molecular dynamics simulation system, specifically including: performing energy minimization processing on the molecular dynamics simulation system by applying a heat bath temperature control to perform isothermal molecular dynamics simulation; and performing kinetic equilibrium processing on the molecular dynamics simulation system by performing time integration on the atomic motion equations in the molecular dynamics simulation system to simulate the motion trajectory of each atom over time.

[0103] In one possible implementation, the processing module 52 is used to search for reaction atom pairs with a first cutoff radius as a constraint, and to perform topological filtering on the reaction atom pairs that meet the constraint to obtain target reaction atom pairs. Specifically, this includes: calculating the interatomic spacing of the reaction atom pairs based on a preset script, and taking reaction atom pairs with an interatomic spacing smaller than the first cutoff radius as reaction atom pairs that meet the constraint; determining whether the reaction atom pairs that meet the constraint belong to the same molecule or the same crosslinking segment; filtering the reaction atom pairs that belong to the same molecule or the same crosslinking segment, and taking the remaining reaction atom pairs as target reaction atom pairs.

[0104] In one possible implementation, the processing module 52 is used to construct cross-linked chemical bonds in the target reaction atom pairs while performing an epoxy group ring-opening operation. Specifically, it includes: constructing cross-linked chemical bonds in the target reaction atom pairs and using the Ewald algorithm to ensure uniform charge distribution in the cross-linked structure; calculating the target bonding position of hydrogen atoms based on the epoxy oxygen atoms connected to the reaction carbon atoms, the CO bonds in the broken epoxy ring, and the dangling atoms generated by the marked ring-opening after the CO bonds are broken, and performing an epoxy group ring-opening operation based on the target bonding position.

[0105] In one possible implementation, the processing module 52 is used to expand the first cutoff radius to a second cutoff radius, specifically including: determining an adjustment step size for the cutoff radius based on the reaction efficiency of the reaction atom pairs in the crosslinking cycle; and expanding the first cutoff radius to the second cutoff radius according to the adjustment step size.

[0106] In one possible implementation, the processing module 52 is used to terminate the crosslinking cycle if it is confirmed that the current crosslinking conversion rate value is greater than or equal to the preset crosslinking conversion rate threshold, perform abnormal coordination atom detection and repair, and perform geometric optimization and molecular dynamics equilibrium processing on the repaired molecular dynamics simulation system. Specifically, it includes: detecting the coordination number of atoms in the molecular dynamics simulation system, and judging whether there are abnormal coordination atoms that do not conform to the valence bond theory based on the detection results; if there are abnormal coordination atoms, performing the epoxy group ring-opening operation again, and repairing the molecular dynamics simulation system by retaining the newly formed crosslinking bonds.

[0107] It should be noted that the above embodiments of the apparatus are only illustrated by the division of the above functional modules. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. In addition, the apparatus and method embodiments provided above belong to the same concept, and the specific implementation process can be found in the method embodiments, which will not be repeated here.

[0108] This application also provides an electronic device. (See reference...) Figure 6 , Figure 6 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. The electronic device may include: at least one processor 601, at least one communication bus 602, a user interface 603, at least one network interface 604, and a memory 605.

[0109] The communication bus 602 is used to enable communication between these components.

[0110] The user interface 603 may include a display screen and a camera. Optionally, the user interface 603 may also include a standard wired interface and a wireless interface.

[0111] The network interface 604 may optionally include a standard wired interface or a wireless interface (such as a Wi-Fi interface).

[0112] The processor 601 may include one or more processing cores. The processor 601 connects to various parts of the server using various interfaces and lines, and performs various server functions and processes data by running or executing instructions, programs, code sets, or instruction sets stored in the memory 605, and by calling data stored in the memory 605. Optionally, the processor 601 may be implemented using at least one hardware form of Digital Signal Processing (DSP), Field-Programmable Gate Array (FPGA), or Programmable Logic Array (PLA). The processor 601 may integrate one or a combination of several of the following: Central Processing Unit (CPU), Graphics Processing Unit (GPU), and modem. The CPU primarily handles the operating system, user interface, and applications; the GPU is responsible for rendering and drawing the content required for display; and the modem handles wireless communication. It is understood that the modem may also not be integrated into the processor 601 and may be implemented as a separate chip.

[0113] The memory 605 may include random access memory (RAM) or read-only memory. Optionally, the memory 605 may include a non-transitory computer-readable storage medium. The memory 605 may be used to store instructions, programs, code, code sets, or instruction sets. The memory 605 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as touch function, sound playback function, image playback function, etc.), instructions for implementing the above-described method embodiments, etc.; the data storage area may store data involved in the above-described method embodiments, etc. Optionally, the memory 605 may also be at least one storage device located remotely from the aforementioned processor 601. (Refer to...) Figure 6 The memory 605, which serves as a computer storage medium, may include an operating system, a network communication module, a user interface module, and a molecular dynamics-based epoxy resin self-crosslinking modeling application.

[0114] exist Figure 6 In the illustrated electronic device, the user interface 603 is primarily used to provide an input interface for the user and acquire user input data; while the processor 601 can be used to call the epoxy resin self-crosslinking modeling application based on molecular dynamics stored in the memory 605. When executed by one or more processors 601, the electronic device performs one or more of the methods described in the above embodiments. It should be noted that, for the foregoing method embodiments, for the sake of simplicity, they are all described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, because according to this application, some steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also understand that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily essential to this application.

[0115] This application also provides a non-transitory computer-readable storage medium storing instructions. When executed by one or more processors, these instructions cause an electronic device to perform one or more of the methods described in the above embodiments.

[0116] In the above embodiments, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0117] In the various embodiments provided in this application, it should be understood that the disclosed apparatus can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some service interface; the indirect coupling or communication connection between apparatuses or units may be electrical or other forms.

[0118] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0119] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0120] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage device (CMD). Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a memory and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned memory includes various media capable of storing program code, such as USB flash drives, portable hard drives, magnetic disks, or optical disks.

[0121] The above description is merely an exemplary embodiment disclosed in this application and should not be construed as limiting the scope of this application. Any equivalent changes and modifications made in accordance with the teachings of this application shall still fall within the scope of this application.

[0122] This application is intended to cover any variations, uses, or adaptations disclosed herein that follow the general principles disclosed herein and include common knowledge or customary technical means in the art that are not described in this application.

Claims

1. A molecular dynamics-based method for modeling the self-crosslinking of epoxy resins, characterized in that, The method includes: Epoxy resin monomer molecules are constructed by labeling target atoms in the corresponding three-membered rings as reaction atom pairs, and an amorphous structure model is constructed based on the reaction atom pairs; Initial reaction parameters are set in a preset script, and the amorphous structure model is loaded through the preset script to form a molecular dynamics simulation system; the molecular dynamics simulation system enters a crosslinking cycle after energy minimization and kinetic equilibrium processing. The reaction atom pairs are searched with the first cutoff radius as a constraint, and topological filtering is performed on the reaction atom pairs that satisfy the constraint to obtain the target reaction atom pairs; The target reaction atom pairs are prioritized, and harmonic oscillator constraints are applied to each target reaction atom pair according to the prioritization results; the harmonic oscillator constraints are used to flexibly pull the corresponding reaction atoms closer to the target bonding distance during molecular dynamics evolution; After removing the harmonic oscillator constraint, cross-linked chemical bonds are constructed in the target reaction atom pairs, while simultaneously performing an epoxy group ring-opening operation; Calculate the current crosslinking conversion rate value corresponding to the molecular dynamics simulation system, and determine whether the current crosslinking conversion rate value is greater than or equal to a preset crosslinking conversion rate threshold; If it is confirmed that the current crosslinking conversion rate value is less than the preset crosslinking conversion rate threshold, then the first cutoff radius is expanded to the second cutoff radius, and the crosslinking cycle is returned to be executed. If the current crosslinking conversion rate value is confirmed to be greater than or equal to the preset crosslinking conversion rate threshold, the crosslinking cycle is terminated, abnormal coordination atom detection and repair are performed, and geometric optimization and molecular dynamics equilibrium processing are performed on the repaired molecular dynamics simulation system to obtain the target epoxy resin self-crosslinking network structure model.

2. The method according to claim 1, characterized in that, The construction of epoxy resin monomer molecules involves labeling target atoms in corresponding three-membered rings as reaction atom pairs, and constructing an amorphous structure model based on these reaction atom pairs, specifically including: Based on the chemical structure of the epoxy resin monomer molecule, an epoxy resin monomer molecule model was constructed using standard bond lengths and standard bond angles. In the epoxy resin monomer molecular model, the C atom on the outside of the three-membered ring is marked as the first reactive atom, and the O atom connected to the C atom is marked as the second reactive atom; the first reactive atom and the second reactive atom constitute the reactive atom pair; Construct the target unit cell based on the spatial size parameters of the epoxy resin monomer molecules; According to a preset density or quantity, the epoxy resin monomer molecular model is randomly filled into the target unit cell to construct the amorphous structure model containing multiple reaction atom pairs.

3. The method according to claim 1, characterized in that, The molecular dynamics simulation system is subjected to energy minimization and kinetic equilibrium treatment, specifically including: By applying heat bath temperature control to the molecular dynamics simulation system to perform isothermal molecular dynamics simulation, the energy of the molecular dynamics simulation system is minimized. By integrating the atomic motion equations in the molecular dynamics simulation system over time to simulate the trajectory of each atom over time, the molecular dynamics simulation system is subjected to dynamic equilibrium treatment.

4. The method according to claim 1, characterized in that, The process of searching for reaction atom pairs with a first cutoff radius as a constraint, and performing topological filtering on the reaction atom pairs that satisfy the constraint to obtain the target reaction atom pairs, specifically includes: The atomic spacing of the reaction atom pairs is calculated based on the preset script, and the reaction atom pairs with atomic spacing smaller than the first cutoff radius are taken as the reaction atom pairs that satisfy the constraint conditions; Determine whether the reaction atom pairs that satisfy the constraints belong to the same molecule or the same crosslinking segment; The reaction atom pairs belonging to the same molecule or the same crosslinking segment are filtered out, and the remaining reaction atom pairs are taken as the target reaction atom pairs.

5. The method according to claim 1, characterized in that, The construction of cross-linked chemical bonds in the target reaction atom pairs, while simultaneously performing ring-opening operations on epoxy groups, specifically includes: Cross-linked chemical bonds were constructed in the target reaction atom pairs, and the Ewald algorithm was used to ensure uniform charge distribution in the cross-linked structure. The target bonding position of the hydrogen atom is calculated based on the epoxy oxygen atom bonded to the reacting carbon atom, the CO bond in the broken epoxy ring, and the dangling atom generated by the marked ring opening after the CO bond is broken, and the ring opening operation of the epoxy group is performed based on the target bonding position.

6. The method according to claim 1, characterized in that, The step of extending the first cutoff radius to the second cutoff radius specifically includes: The adjustment step size of the cutoff radius is determined based on the reaction efficiency of the reactant atom pairs in the crosslinking cycle; The first cutoff radius is expanded to the second cutoff radius according to the adjustment step size.

7. The method according to claim 1, characterized in that, If the current crosslinking conversion rate value is confirmed to be greater than or equal to the preset crosslinking conversion rate threshold, the crosslinking cycle is terminated, abnormal coordination atom detection and repair are performed, and geometric optimization and molecular dynamics equilibrium processing are performed on the repaired molecular dynamics simulation system, specifically including: The coordination number of atoms in the molecular dynamics simulation system is detected, and the presence of anomalous coordination atoms that do not conform to valence bond theory is determined based on the detection results. If the anomalous coordinating atoms are present, the ring-opening operation of the epoxy group is performed again, and the molecular dynamics simulation system is repaired by preserving the newly formed crosslinked bonds.

8. A molecular dynamics-based epoxy resin self-crosslinking modeling device, characterized in that, The device includes an acquisition module and a processing module, wherein, The acquisition module is used to construct epoxy resin monomer molecules to mark the target atoms in the corresponding three-membered rings as reaction atom pairs, and to construct an amorphous structure model based on the reaction atom pairs; to set initial reaction parameters in a preset script, and to load the amorphous structure model through the preset script to form a molecular dynamics simulation system; the molecular dynamics simulation system enters the crosslinking cycle after energy minimization and kinetic equilibrium processing; The processing module is used to search for the reaction atom pairs with a first cutoff radius as a constraint, and perform topological filtering on the reaction atom pairs that satisfy the constraint to obtain target reaction atom pairs; prioritize the target reaction atom pairs, and apply harmonic oscillator constraints to the target reaction atom pairs according to the ranking results; the harmonic oscillator constraints are used to flexibly pull the corresponding reaction atoms closer to the target bonding distance during molecular dynamics evolution; the oxygen group ring-opening operation is used to perform hydrogen saturation treatment on the generated dangling atoms; calculate the current crosslinking conversion rate value corresponding to the molecular dynamics simulation system, and determine whether the current crosslinking conversion rate value is greater than or equal to a preset crosslinking conversion rate threshold; if it is confirmed that the current crosslinking conversion rate value is less than the preset crosslinking conversion rate threshold, then the first cutoff radius is expanded to a second cutoff radius, and the crosslinking cycle is returned to be executed; if it is confirmed that the current crosslinking conversion rate value is greater than or equal to the preset crosslinking conversion rate threshold, then the crosslinking cycle is ended, abnormal coordination atom detection and repair are performed, and geometric optimization and molecular dynamics equilibrium processing are performed on the repaired molecular dynamics simulation system to obtain the target epoxy resin self-crosslinking network structure model.

9. An electronic device, characterized in that, The device includes a processor, a communication bus, a user interface, a network interface, and a memory. The memory is used to store instructions. The user interface and the network interface are both used to communicate with other devices. The communication bus is used to enable communication between the components within the electronic device. The processor is used to execute the instructions stored in the memory to cause the electronic device to perform the method as described in any one of claims 1-7.

10. A non-transitory computer-readable storage medium, characterized in that, The non-transitory computer-readable storage medium stores instructions that, when executed, perform the method as described in any one of claims 1 to 7.

Images