An insulating material performance evaluation method and device based on molecular simulation, an electronic device, and a storage medium

Abstract

The application discloses an insulating material performance evaluation method and device based on molecular simulation, electronic equipment and a storage medium, and belongs to the technical field of material performance evaluation. The method comprises the following steps: obtaining a target molecular model of an insulating material to be evaluated, obtaining Fermi energy level, state density function, molecular orbital energy level and electron wave function; determining the target trap depth based on the Fermi energy level, the state density function and the molecular orbital energy level; extracting characteristic parameters representing the attractive and repulsive interaction within the molecule based on the electron wave function; determining the electron state complexity parameter based on the state density function; and finally evaluating the insulating performance based on the target trap depth, the interaction characteristic parameters and the electron state complexity parameter. Through the implementation of the application, the influence of various electronic structure characteristics on the material insulating performance can be comprehensively represented, and the accuracy of the insulating performance evaluation is improved.

Description

Technical Field

[0001] This invention relates to the field of material performance evaluation technology, specifically to a method, apparatus, electronic device, and storage medium for evaluating the performance of insulating materials based on molecular simulation. Background Technology

[0002] With the widespread application of polymer insulating materials in power equipment and electronic packaging, accurately evaluating their insulation performance has become a key issue in material design and performance optimization. However, most existing technologies analyze and evaluate insulation performance based on a single electronic structure parameter. A single parameter can usually only characterize the local electronic structure features of the material or a single action mechanism, making it difficult to comprehensively reflect the influence of the coupling effects of multiple factors such as electron transport behavior, weak intermolecular interactions, and electronic state distribution on the insulation performance of the material. Summary of the Invention

[0003] This invention provides a method, apparatus, electronic device, and storage medium for evaluating the performance of insulating materials based on molecular simulation. This addresses the problem that existing insulation material performance evaluation technologies use a single parameter to evaluate the insulation performance of materials, making it difficult to comprehensively reflect the influence of the coupling effects of multiple factors such as electron transport behavior, weak intermolecular interactions, and electronic state distribution on the insulation performance of materials.

[0004] This invention provides a method for evaluating the performance of insulating materials based on molecular simulation, comprising: Obtain the target molecular model of the insulating material to be evaluated; Based on density functional theory, first-principles electronic structure calculations were performed on the target molecular model to obtain the corresponding Fermi level, density of states function, molecular orbital energy levels and electronic wave functions. Peak analysis of the density of states function is performed to obtain the number of density of states peaks, and the number of density of states peaks is used as a parameter of electronic state complexity. Based on the Fermi level, density of states function, and molecular orbital energy levels, the trap state energy level of each local electronic state in the target molecular model is calculated; Based on the energy difference between the trap state energy level of each local electronic state and the preset equilibrium energy level, the trap state energy level with the largest energy difference is taken as the target trap depth. Based on each electronic wavefunction, the spatial distribution characteristics of electron density are constructed, and weak interaction analysis is performed on the electron density variation characteristics to extract the first characteristic parameter characterizing intramolecular attractive interactions and the second characteristic parameter characterizing intramolecular repulsive interactions. The insulation performance of the insulating material to be evaluated is assessed based on the target trap depth, the first characteristic parameter, the second characteristic parameter, and the electronic state complexity parameter.

[0005] Furthermore, an electronic density field is constructed based on density functional theory; an external potential field is constructed based on a target molecule model. The initial electron density distribution function is determined based on the electron density field. The external potential field and the initial electron density distribution function are substituted into the Kohn–Sham equation as initial conditions for iterative solution to generate several electron wave functions. Each eigenvalue of the Kohn–Sham equation is taken as a molecular orbital energy level; Based on the distribution of molecular orbital energy levels, the energy of each molecular orbital level is divided into occupied and unoccupied molecular orbital levels. The Fermi level is determined based on the occupied and unoccupied molecular orbital energy levels. Energy level broadening is performed on each molecular orbital energy level to generate the density of states function.

[0006] Furthermore, local maxima detection is performed on the density of states function to obtain the peak values ​​of the density of states at each local maximum point; The peak value of the density of states corresponding to each local maximum point is compared with a preset threshold, and the peak value of the density of states that is greater than the preset threshold is taken as the effective peak value. The number of effective peaks is counted to obtain the number of density of states peaks, which is then used as an electronic state complexity parameter to characterize the complexity of electronic state distribution.

[0007] Furthermore, for each local maximum point, the density of states peak value is determined according to the position of the density of states peak value on the energy axis; and the corresponding energy screening interval is constructed based on the reference energy and the preset interval interval. Determine whether each occupied molecular orbital energy level falls within the energy screening interval. If the occupied molecular orbital energy level falls within any energy screening interval, then the occupied molecular orbital energy level is taken as the candidate molecular orbital energy level. Obtain the electronic state corresponding to the candidate molecular orbital energy level as the candidate electronic state. The spatial electron density value of the target molecular model is calculated based on the electronic wave function corresponding to each occupied molecular orbital energy level. For each candidate electronic state, the electron density value is calculated using the corresponding electronic wavefunction, and the ratio of the corresponding electron density value to the spatial electron density value is calculated to obtain the electron density ratio. Candidate electronic states with electron density ratios greater than the density ratio threshold are taken as local electronic states. For each local electronic state, the energy difference between the molecular orbital energy level and the Fermi level corresponding to the local electronic state is calculated to obtain the trap state energy level corresponding to that local electronic state.

[0008] Furthermore, by performing a modulus-squared superposition of the electronic wavefunctions corresponding to all occupied molecular orbital energy levels, the spatial electron density distribution function is obtained; Integrating the space electron density distribution function yields the space electron density value.

[0009] Furthermore, the space of the target molecular model is discretized to obtain several discrete grids; For each discrete grid, the spatial gradient value of the electron density corresponding to the discrete grid is calculated by performing partial derivatives on the spatial electron density distribution function. The electron density value corresponding to the discrete grid is obtained by integrating the spatial electron density distribution function. The ratio of the spatial gradient of electron density to the electron density value is calculated to obtain the reduced density gradient of the discrete grid. Based on the reduced density gradient value of each discrete grid, a reduced density gradient scalar field is constructed. The electron density matrix is ​​obtained by taking the second derivative of the spatial electron density distribution function. The electron density matrix is ​​then decomposed into eigenvalues ​​to obtain eigenvalues ​​that characterize the local curvature variation of the electron density. Based on the eigenvalues, a symbolic piecewise function is constructed. The symbolic piecewise function is then multiplied by the spatial electron density distribution function to obtain the electron interaction type labeling function. Projecting the type labeling function onto a preset isosurface of the approximate density gradient scalar field yields a three-dimensional candidate spatial point set based on the full-space electron density distribution. In the candidate spatial point set, the interaction type regions are divided according to the positive and negative attributes of the labeling function, and the spatial points closest to zero are selected in the positive and negative value regions respectively. The label function value of the spatial point closest to zero in the negative region is used as the first feature parameter, and the label function value of the spatial point closest to zero in the positive region is used as the second feature parameter.

[0010] Furthermore, the target trap depth, first feature parameter, second feature parameter, and electronic state complexity parameter are normalized to obtain the corresponding membership degree. Based on preset weight parameters, the membership degrees are weighted and summed to obtain the score of the insulation material to be evaluated, which is used to characterize the insulation performance.

[0011] Based on the above method embodiments, the present invention provides a corresponding apparatus embodiment.

[0012] The material acquisition module is used to acquire the target molecular model of the insulating material to be evaluated. The electronic structure calculation module is used to perform first-principles electronic structure calculations on the target molecular model based on density functional theory, and to obtain the corresponding Fermi level, density of states function, molecular orbital energy levels and electronic wave functions. The peak analysis module is used to perform peak analysis on the density of states function to obtain the number of density of states peaks, and uses the number of density of states peaks as a parameter of electronic state complexity. The trap state energy level calculation module is used to calculate the trap state energy level of each local electronic state in the target molecular model based on the Fermi level, density of states function, and molecular orbital energy level. The target trap depth acquisition module is used to take the trap state energy level with the largest energy difference as the target trap depth based on the energy difference between the trap state energy level of each local electronic state and the preset equilibrium energy level. The weak interaction analysis module is used to construct the spatial distribution characteristics of electron density based on each electronic wavefunction, and to perform weak interaction analysis on the electron density variation characteristics to extract the first characteristic parameter characterizing intramolecular attractive interactions and the second characteristic parameter characterizing intramolecular repulsive interactions. The evaluation score calculation module is used to evaluate the insulation performance of the insulating material to be evaluated based on the target trap depth, the first feature parameter, the second feature parameter, and the electronic state complexity parameter.

[0013] Based on the above method embodiments, the present invention provides corresponding electronic device embodiments.

[0014] An electronic device is characterized by comprising a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor, when executing the computer program, implements a method for evaluating the performance of insulating materials based on molecular simulation as described in any one of claims 1 to 7.

[0015] Based on the above method embodiments, the present invention provides a corresponding storage medium embodiment.

[0016] A storage medium, characterized in that the storage medium includes a stored computer program, wherein, when the computer program is executed, it controls the device where the storage medium is located to perform a molecular simulation-based method for evaluating the performance of insulating materials as described in any one of claims 1 to 7.

[0017] Compared with the prior art, the present invention has the following beneficial effects: This invention provides a method, apparatus, terminal device, and computationally readable storage medium for evaluating the performance of insulating materials based on molecular simulation. The method first constructs a target molecular model of the insulating material to be evaluated. Then, it identifies the Fermi level, density of states function, molecular orbital energy level, and electronic wavefunction corresponding to the target molecular model. Based on the Fermi level, density of states function, and molecular orbital energy level, it determines the trap state energy level of each local electronic state in the target molecular model and extracts the target trap depth. Furthermore, based on the electronic wavefunction, it constructs the spatial distribution characteristics of the electron density and extracts a first characteristic parameter characterizing intramolecular attractive interactions and a second characteristic parameter characterizing intramolecular repulsive interactions. Based on the density of states function, it determines the electronic state complexity parameter. Finally, it evaluates the insulation performance of the insulating material to be evaluated based on the target trap depth, the first characteristic parameter, the second characteristic parameter, and the electronic state complexity parameter. Compared to existing technologies, this application does not use a single electronic structure parameter to evaluate insulation performance. Instead, it extracts target trap depth (characterizing different insulation mechanisms), a first characteristic parameter (characterizing intramolecular attractive interactions), a second characteristic parameter (characterizing intramolecular repulsive interactions), and an electronic state complexity parameter. The target trap depth characterizes the material's ability to bind electrons, the first and second characteristic parameters characterize the strength of weak intramolecular interactions, and the electronic state complexity parameter characterizes the complexity of electronic state distribution. These parameters are then used to evaluate insulation performance, thus achieving a synergistic evaluation of different insulation mechanisms. Compared to existing methods that analyze insulation performance based on a single independent parameter, this approach more comprehensively reflects the combined influence of electron transport behavior, weak intermolecular interactions, and electronic structure distribution characteristics on the material's insulation performance, improving the accuracy of insulation performance evaluation. Attached Figure Description

[0018] Figure 1 This is a schematic flowchart of a method for evaluating the performance of insulating materials based on molecular simulation, provided in an embodiment of the present invention.

[0019] Figure 2 This is a schematic diagram of the structure of an insulation material performance evaluation device based on molecular simulation provided in an embodiment of the present invention.

[0020] Figure 3 This is a diagram of the reduced density gradient scalar field result of the target molecular model provided in an embodiment of the present invention.

[0021] Figure 4 This is a scatter plot of the target molecular model provided in an embodiment of the present invention. Detailed Implementation

[0022] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this application. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of those different embodiments or examples.

[0023] To address the challenge that existing insulation material performance evaluation techniques struggle to comprehensively reflect the influence of electron transport, weak interactions, and electronic state distribution on insulation performance, this invention provides a molecular simulation-based insulation material performance evaluation method. This method solves the problem that existing insulation material performance evaluation techniques, which rely solely on single-parameter analysis, fail to comprehensively reflect the impact of factors such as electron transport, weak interactions, and electronic state distribution on material insulation performance.

[0024] One embodiment of the present invention provides a method for evaluating the performance of insulating materials based on molecular simulation, comprising at least the following steps: S1. Obtain the target molecular model of the insulating material to be evaluated; Specifically, this embodiment can evaluate the performance of various insulating materials. In this invention, as an illustration, the performance of any one of the following five grafted modified PP types can be evaluated: PP- g -AA、PP- g -MMA, PP- g -MAH, PP- g -ITA, PP-g-NA; wherein, polypropylene (PP) matrix is ​​selected, and acrylic acid (AA) is grafted onto PP to obtain PP- g -AA, grafting methyl ester (MMA) onto PP yields PP- g -MMA, grafting maleic anhydride (MAH) onto PP yields PP- g -MAH, grafting itaconic anhydride (ITA) onto PP yields PP- g -ITA, grafting norbornene (NA) onto PP yields PP- g -NA.

[0025] S2. Based on density functional theory, first-principles electronic structure calculations are performed on the target molecular model to obtain the corresponding Fermi level, density of states function, molecular orbital energy levels and electronic wave functions. Specifically, based on density functional theory, an electronic structure calculation system is established for the target molecular model, and the ground state electronic structure of the system is obtained by solving the Kohn–Sham equation. On this basis, the Fermi level, density of states distribution function, molecular orbital energy level and corresponding electronic wave function of the system are extracted to characterize the electronic structure properties of the target molecular model.

[0026] In a preferred embodiment, based on density functional theory, first-principles electronic structure calculations are performed on the target molecular model to obtain the corresponding Fermi level, density of states function, molecular orbital energy levels, and electronic wavefunctions, including: Constructing the electronic density field based on density functional theory; constructing the external potential field based on the target molecule model; The initial electron density distribution function is determined based on the electron density field. The external potential field and the initial electron density distribution function are substituted into the Kohn–Sham equation as initial conditions for iterative solution to generate several electron wave functions. Each eigenvalue of the Kohn–Sham equation is taken as a molecular orbital energy level; Based on the distribution of molecular orbital energy levels, the energy of each molecular orbital level is divided into occupied and unoccupied molecular orbital levels. The Fermi level is determined based on the occupied and unoccupied molecular orbital energy levels. Energy level broadening is performed on each molecular orbital energy level to generate the density of states function.

[0027] Specifically, the spatial positions of each atom in the target molecular model are determined, and the reference electron density distribution of each isolated atom is extracted based on density functional theory. An initial electron density distribution function is constructed through spatial translation and superposition. Based on the spatial coordinates and nuclear charge distribution of each atomic nucleus in the molecule, an external Coulomb potential function is constructed to describe the interaction between the electron and the atomic nucleus. Based on the initially guessed electron density distribution and combined with the external potential field, the Kohn–Sham effective potential term is constructed, and the electron density and single-electron wavefunction are updated through a self-consistent iterative process, causing the system to gradually converge to the ground state solution. After the self-consistent iterative convergence, multiple eigenvalues ​​obtained from solving the Kohn–Sham equation are extracted, where each eigenvalue corresponds to a molecular orbital. Energy levels are used to characterize the energy distribution of electronic states. Solving the Kohn–Sham equation yields the electronic wavefunctions of each molecular orbital, which are used to describe the spatial distribution characteristics of different electronic states in the system. Based on the energy level order from low to high and combined with the electron filling principle, the occupancy of molecular orbital energy levels is determined: orbitals below a certain critical energy are occupied states, and orbitals above that energy are unoccupied states. Based on the electron occupancy probability, an energy boundary point is determined between the highest occupied state and the lowest unoccupied state; this boundary energy is the Fermi level, used to characterize the electron filling boundary of the system. Discrete molecular orbital energy levels are smoothed using Gaussian broadening or equivalent broadening functions, transforming the discrete energy levels into a density of states function, used to characterize the distribution density of electronic states in different energy ranges.

[0028] S3. Perform peak analysis on the density of states function to obtain the number of density of states peaks, and use the number of density of states peaks as the electronic state complexity parameter. Specifically, a density of states function is constructed based on the molecular orbital energy level distribution, and a continuous density of states distribution curve is generated; structural feature analysis is performed on the density of states distribution curve to extract the number of peak features of electronic state distribution, and the number of peak features is used as a characterization parameter of the electronic structure complexity of the target molecular model.

[0029] In a preferred real-time example, peak analysis is performed on the density of states function to obtain the number of density of states peaks. The number of density of states peaks is used as a parameter of electronic state complexity, including: Local maxima detection is performed on the density of states function to obtain the peak values ​​of the density of states at each local maximum point; The peak value of the density of states corresponding to each local maximum point is compared with a preset threshold, and the peak value of the density of states that is greater than the preset threshold is taken as the effective peak value. The number of effective peaks is counted to obtain the number of density of states peaks, which is then used as an electronic state complexity parameter to characterize the complexity of electronic state distribution.

[0030] Specifically, structural feature analysis is performed on the continuous density of states distribution curve along the energy axis to characterize the distribution characteristics of electronic states in different energy ranges; by analyzing the variation trend of the density of states function in the energy dimension, local density enhancement regions in the density of states function are identified, and the corresponding local maxima are determined; the density of states values ​​corresponding to each local maxima are used as candidate peaks, and the candidate peaks are screened based on a preset density threshold. When a candidate peak is greater than the preset threshold, it is determined as a valid peak; the valid peaks are statistically processed to obtain the number of density of states peaks; the number of density of states peaks is used as an electronic state complexity parameter to characterize the complexity of the electronic state distribution of the target molecule model.

[0031] In this invention, the changing trend of electronic state distribution is identified through structural feature analysis along the energy axis, and the local maxima corresponding to the local density enhancement region are further located, thereby transforming the continuous electronic state distribution into a set of candidate peaks with clear physical meaning. Subsequently, a preset density threshold is introduced to screen the candidate peaks, effectively eliminating insignificant peaks caused by numerical fluctuations or weakly contributing electronic states, thereby improving the stability and physical reliability of peak structure identification. On this basis, the effective peaks are statistically analyzed to obtain the number of density of states peaks, realizing a quantitative characterization of the dispersion of electronic state distribution and the degree of energy level splitting in energy space. Finally, the number of peaks is used as an electronic state complexity parameter, so that the originally continuous and high-dimensional electronic state distribution information is compressed into a single comparable structured index, thereby achieving a unified quantitative description of the electronic state complexity of different molecular models while maintaining the key features of electronic structure, and providing global constraints and feature inputs for subsequent electronic state screening and local electronic state analysis.

[0032] S4. Based on the Fermi level, density of states function, and molecular orbital energy levels, calculate the trap state energy level of each local electronic state in the target molecular model; Specifically, for each density of states peak, a corresponding reference energy is determined based on its position on the energy axis, and an energy screening interval set is constructed based on a preset energy interval. The occupied molecular orbital energy level is matched with the energy screening interval set. When it falls into any energy screening interval, its corresponding electronic state is taken as a candidate electronic state obtained based on energy screening. For each candidate electronic state, its spatial distribution characteristics are calculated based on the electronic wavefunction, and its localization degree is judged based on the spatial electron density ratio. When the localization condition is met, it is determined as a local electronic state. For each local electronic state, the energy difference between its corresponding molecular orbital energy level and the Fermi level is calculated to obtain the corresponding trap state energy level.

[0033] In a preferred embodiment, based on the Fermi level, density of states function, and molecular orbital energy levels, the trapped state energy levels of each local electronic state in the target molecular model are calculated, including: For each local maximum point, the peak value of the density of states is determined according to the position of the peak value on the energy axis; and the corresponding energy screening interval is constructed based on the reference energy and the preset interval. Determine whether each occupied molecular orbital energy level falls within the energy screening interval. If the occupied molecular orbital energy level falls within any energy screening interval, then the occupied molecular orbital energy level is taken as the candidate molecular orbital energy level. Obtain the electronic state corresponding to the candidate molecular orbital energy level as the candidate electronic state. The spatial electron density value of the target molecular model is calculated based on the electronic wave function corresponding to each occupied molecular orbital energy level. For each candidate electronic state, the electron density value is calculated using the corresponding electronic wavefunction, and the ratio of the corresponding electron density value to the spatial electron density value is calculated to obtain the electron density ratio. Candidate electronic states with electron density ratios greater than the density ratio threshold are taken as local electronic states. For each local electronic state, the energy difference between the molecular orbital energy level and the Fermi level corresponding to the local electronic state is calculated to obtain the trap state energy level corresponding to that local electronic state.

[0034] Specifically, structural feature analysis is performed on the density of states function along the energy axis to identify local maxima and determine the energy positions corresponding to each local maxima as reference energies. Based on the reference energies and preset energy intervals, multiple energy screening interval sets are constructed on the energy axis to characterize the energy range of possible electronic state distributions. Occupied molecular orbital energy levels are matched with the energy screening interval sets. When an occupied molecular orbital energy level falls into any energy screening interval, the electronic state corresponding to that molecular orbital energy level is determined as a candidate electronic state. Based on the electronic wavefunctions corresponding to each occupied molecular orbital energy level, the spatial electron density distribution function of the target molecular model is calculated to characterize the overall distribution of electrons in space. For each candidate electronic state, the local electron density distribution is calculated based on its corresponding electronic wavefunction, and the ratio of the local electron density to the spatial electron density distribution is calculated to obtain the electron density ratio. When the electron density ratio is greater than a preset density threshold, the corresponding candidate electronic state is determined as a local electronic state. For each local electronic state, the energy difference between its corresponding molecular orbital energy level and the Fermi level is calculated to obtain the trap state energy level of the corresponding local electronic state.

[0035] In this invention, based on the structural feature analysis of the density of states function along the energy axis, local maxima are identified and their corresponding reference energies are determined, thereby transforming the originally continuous electronic state distribution into physically meaningful discrete energy calibration points. Furthermore, an energy screening interval set is constructed based on the reference energy and a preset energy interval, achieving a partitioned characterization of the possible distribution range of electronic states. The occupied state molecular orbital energy levels are matched and screened with this interval set, initially determining candidate electronic states from an energy perspective, thus improving the targeting of electronic state screening. A spatial electron density distribution function is constructed based on the occupied state molecular orbital wavefunction, and a ratio of local electron density to overall spatial electron density is introduced for candidate electronic states, thus improving the spatial distribution. The localization characteristics of electronic states are further constrained by the angle of arrangement, thereby effectively eliminating non-local or weakly contributing electronic states and improving the physical rationality and stability of electronic state identification. On this basis, electronic states that meet the localization conditions are identified as local electronic states, and the energy difference between their corresponding molecular orbital energy levels is calculated using the Fermi level as a unified energy benchmark, thereby obtaining the trap state energy level and realizing the quantitative characterization of the electron binding depth and trapping ability. Overall, this step introduces a multi-layer judgment mechanism of "energy screening - spatial constraint - energy level calibration" to realize the hierarchical analysis of electronic states from continuous distribution to localized energy level characterization, thereby improving the accuracy, physical consistency and characterization ability of electronic state screening and the ability to characterize the differences in the electronic structure of materials.

[0036] In a preferred embodiment, the spatial electron density value of the target molecular model is calculated based on the electron wavefunction corresponding to each occupied molecular orbital energy level, including: The spatial electron density distribution function is obtained by superimposing the modulus squares of the electronic wavefunctions corresponding to all occupied molecular orbital energy levels; Integrating the space electron density distribution function yields the space electron density value.

[0037] Specifically, the electronic wavefunctions corresponding to all occupied molecular orbital energy levels are extracted, with each occupied molecular orbital energy level corresponding to an electronic wavefunction, used to characterize the probability distribution of electrons in space; the modulus square operation is performed on each electronic wavefunction to obtain the electronic probability density distribution of the corresponding electronic state in space; the electronic probability density distributions corresponding to all occupied states are spatially superimposed to construct the spatial electron density distribution function of the target molecular model, used to characterize the overall distribution of electrons in the system across the entire spatial range; based on the spatial electron density distribution function, the spatial region corresponding to the target molecular model is integrally calculated to obtain the spatial electron density value of the target molecular model.

[0038] S5. Based on the energy difference between the trap state energy level of each local electronic state and the preset equilibrium energy level, the trap state energy level with the largest energy difference is taken as the target trap depth. Specifically, using the Fermi level as the preset equilibrium level, the energy references for the trap state energy levels corresponding to each local electronic state are compared with the preset equilibrium level. An energy difference characterization formula is constructed to reflect the degree of binding of the electronic state relative to the equilibrium level. The specific energy difference characterization formula is as follows: In the formula This represents the i-th energy difference. This represents the trap state energy level of the i-th local electronic state. This indicates the preset equilibrium energy level.

[0039] For each local electronic state, calculate the energy difference between its trap state energy level and the preset equilibrium energy level, and compare and analyze the energy differences of all local electronic states; determine the trap state energy level corresponding to the local electronic state with the largest energy difference as the target trap depth.

[0040] S6. Construct the spatial distribution characteristics of electron density based on each electronic wavefunction, and perform weak interaction analysis on the electron density variation characteristics to extract the first characteristic parameter characterizing intramolecular attractive interactions and the second characteristic parameter characterizing intramolecular repulsive interactions. Specifically, the modulus square operation is performed on each electronic wavefunction and then spatially superimposed to construct the full-space electron density distribution function of the target molecule model, which is used to characterize the distribution state of electrons in space. Based on the electron density distribution function, the space of the target molecule model is discretized, and the electron density gradient distribution corresponding to each spatial position is calculated to characterize the variation characteristics of electron density in space. Based on the electron density distribution and electron density gradient distribution, a reduced density gradient distribution function is constructed to identify weakly interacting spatial regions in the gradually varying electron density region. The second-order partial derivative of the electron density distribution function is calculated to construct the electron density Hessian matrix, and the Hessian matrix is ​​then subjected to special... Eigenvalue decomposition yields eigenvalue parameters characterizing local curvature variations in electron density. Based on these eigenvalue parameters and the electron density distribution function, an electron interaction type labeling function is constructed to distinguish different types of weak interaction regions in space. The electron interaction type labeling function is mapped to a reducible density gradient distribution region, and regions with different signs are classified and analyzed, where negative value regions correspond to attractive interaction regions, and positive value regions correspond to repulsive interaction regions. Feature label values ​​are extracted from the attractive and repulsive interaction regions, respectively, as the first feature parameter characterizing intramolecular attractive interactions and the second feature parameter characterizing intramolecular repulsive interactions.

[0041] In a preferred embodiment, the spatial distribution characteristics of electron density are constructed based on each electron wavefunction, and weak interaction analysis is performed on the electron density variation characteristics to extract a first characteristic parameter characterizing intramolecular attractive interactions and a second characteristic parameter characterizing intramolecular repulsive interactions, including: The space of the target molecular model is discretized to obtain several discrete grids; For each discrete grid, the spatial gradient value of the electron density corresponding to the discrete grid is calculated by performing partial derivatives on the spatial electron density distribution function. The electron density value corresponding to the discrete grid is obtained by integrating the spatial electron density distribution function. The ratio of the spatial gradient of electron density to the electron density value is calculated to obtain the reduced density gradient of the discrete grid. Based on the reduced density gradient value of each discrete grid, a reduced density gradient scalar field is constructed. The electron density matrix is ​​obtained by taking the second derivative of the spatial electron density distribution function. The electron density matrix is ​​then decomposed into eigenvalues ​​to obtain eigenvalues ​​that characterize the local curvature variation of the electron density. Based on the eigenvalues, a symbolic piecewise function is constructed. The symbolic piecewise function is then multiplied by the spatial electron density distribution function to obtain the electron interaction type labeling function. Projecting the type labeling function onto a preset isosurface of the approximate density gradient scalar field yields a three-dimensional candidate spatial point set based on the full-space electron density distribution. In the candidate spatial point set, the interaction type regions are divided according to the positive and negative attributes of the labeling function, and the spatial points closest to zero are selected in the positive and negative value regions respectively. The label function value of the spatial point closest to zero in the negative region is used as the first feature parameter, and the label function value of the spatial point closest to zero in the positive region is used as the second feature parameter.

[0042] Specifically, the spatial region corresponding to the target molecule model is discretized to establish a three-dimensional spatial grid structure, and each spatial grid is used as a local calculation unit for the electron density variation characteristics. For each discrete grid, the partial derivative of the spatial electron density distribution function with respect to the spatial coordinate direction is calculated to obtain the spatial gradient value of the electron density of the corresponding discrete grid, which is used to characterize the rate of change of electron density at that spatial location. Based on the spatial electron density distribution function, the spatial region of the corresponding discrete grid is integrally calculated to obtain the electron density value of the corresponding discrete grid, which is used to characterize the electron distribution intensity in that spatial region. The normalized ratio of the spatial gradient value of the electron density to the electron density value of the discrete grid is calculated to obtain the reduced density gradient value of the corresponding discrete grid, which characterizes the weak interaction characteristics in the region of slowly varying electron density. Based on the reduced density gradient values ​​corresponding to all discrete grids, a reduced density gradient scalar field is constructed within the spatial range of the target molecule model to describe the spatial distribution of the electron density variation characteristics, as illustrated below. Figure 3 As shown, when the target molecular model is PP, the corresponding reduced density gradient scalar field result is shown in the figure. Figure 3 (a) When the target molecule model is PP- g -AA, the corresponding reduced density gradient scalar field result is shown in the figure. Figure 3 (b) When the target molecule model is PP- g -MMA, the corresponding reduced density gradient scalar field result is shown in the figure. Figure 3 (c) When the target molecule model is PP- g -MAH, the corresponding reduced density gradient scalar field result is shown in the figure. Figure 3 (d) When the target molecule model is PP- g -ITA, the corresponding reduced density gradient scalar field result is shown in the figure. Figure 3 (e) When the target molecule model is PP- g -NA, the corresponding reduced density gradient scalar field result is shown in the figure. Figure 3(f); Further, the second-order partial derivative of the spatial electron density distribution function is calculated to construct the electron density Hessian matrix, and the electron density Hessian matrix is ​​decomposed into eigenvalues ​​to obtain multiple eigenvalues ​​characterizing the local curvature variation of electron density; a symbolic piecewise function is constructed based on the preset eigenvalues, and the symbolic piecewise function is combined with the spatial electron density distribution function to obtain an electron interaction type labeling function, which is used to distinguish between attractive and repulsive interactions in different spatial regions; the electron interaction type labeling function is mapped to a preset isosurface corresponding to the reduced density gradient scalar field, in the electron density... Spatial location points satisfying preset constraints are extracted from the slowly varying electron density region to obtain a three-dimensional candidate spatial point set based on the full-space electron density distribution. The electron interaction type labeling function is used to characterize the type of weak interaction at different spatial locations, and the reduced density gradient scalar field is used to characterize the distribution characteristics of weak interactions in regions with slowly changing electron density. A mapped scatter plot is constructed based on the electron interaction type labeling values ​​and reduced density gradient values ​​corresponding to each spatial location in the three-dimensional candidate spatial point set to characterize the distribution differences between attractive and repulsive interaction regions in different polymer systems. Schematic, when the target molecule model is PP, the corresponding scatter plot is shown below. Figure 4 (a) When the target molecule model is PP- g -AA, the corresponding scatter plot is as follows Figure 4 (b) When the target molecule model is PP- g -MMA, the corresponding scatter plot is as follows Figure 4 (c) When the target molecule model is PP- g -MAH, the corresponding scatter plot is as follows Figure 4 (d) When the target molecule model is PP- g -ITA, the corresponding scatter plot is as follows Figure 4 (e) When the target molecule model is PP- g -NA, the corresponding scatter plot is shown below. Figure 4(f); In the set of three-dimensional candidate spatial points, the candidate spatial points are divided into interaction types according to the positive and negative attributes of the electronic interaction type labeling function. The region with a negative labeling function corresponds to the attractive interaction region, and the region with a positive labeling function corresponds to the repulsive interaction region. In the attractive interaction region and the repulsive interaction region, the spatial point with the electronic interaction type labeling function closest to zero is selected to characterize the electronic interaction characteristics in the critical region of weak interaction. The electronic interaction type labeling function value corresponding to the spatial point closest to zero in the negative region is used as the first characteristic parameter characterizing the weak attractive interaction in the molecule, and the electronic interaction type labeling function value corresponding to the spatial point closest to zero in the positive region is used as the second characteristic parameter characterizing the weak repulsive interaction in the molecule.

[0043] In this invention, a spatial electron density distribution function is constructed based on the electron wavefunction. By discretizing the target molecular model space into a grid, the spatial variation characteristics of electron density are locally analyzed, enabling quantitative characterization of the electron distribution state in different spatial regions. Furthermore, by calculating the first-order partial derivative of the electron density distribution function, the spatial gradient value of electron density is obtained. This gradient is then combined with the electron density value to construct a reduced density gradient scalar field, thereby identifying regions where electron density changes slowly. Since weak intermolecular interactions typically correspond to spatial regions with low electron density and slow gradient changes, this step effectively locates potential weak interaction regions, improving the spatial resolution of weak interaction identification. By taking the second-order derivative of the electron density distribution function and constructing the electron density Hessian matrix, the local curvature variation characteristics of electron density are further obtained, reflecting the aggregation and diffusion trends of electrons in local space. After constructing an electron interaction type labeling function based on eigenvalues, attractive and repulsive interactions can be further distinguished, allowing different types of weak interactions to be classified and characterized within a unified electron density framework. Subsequently, by mapping the electron interaction type labeling function to a normalized density gradient isosurface, a set of candidate spatial points with practical physical meaning can be extracted in regions with gradually varying electron density. This avoids interference from high-density covalent regions on weak interaction analysis, improving the accuracy and stability of weak interaction identification results. By selecting spatial points closest to zero in both positive and negative regions, the critical states of weak attraction and repulsion can be quantitatively extracted. Negative regions correspond to electron-dominated attractive interactions, while positive regions correspond to electron-dominated repulsive interactions. Therefore, the extracted first and second feature parameters can reflect the strength and spatial distribution characteristics of different types of weak interactions within the molecule, respectively. Overall, this step achieves a layer-by-layer analysis process from electron wavefunction, electron density distribution, electron density variation characteristics to weak interaction type determination, improving the analytical capability for intramolecular microscopic electronic interaction behavior and providing more comprehensive and physically meaningful electronic structure feature parameters for subsequent insulation material performance evaluation.

[0044] S8. Evaluate the insulation performance of the insulating material to be evaluated based on the target trap depth, the first characteristic parameter, the second characteristic parameter, and the electronic state complexity parameter. Specifically, the target trap depth, first feature parameter, second feature parameter, and electronic state complexity parameter are extracted. The target trap depth, first feature parameter, second feature parameter, and electronic state complexity parameter are normalized respectively to obtain the membership degree of the target trap depth, the membership degree of the first feature parameter, the membership degree of the second feature parameter, and the membership degree of the electronic state complexity parameter. Based on the preset weight parameters, the membership degree is weighted and summed to obtain the score of the insulation material to be evaluated for characterizing the insulation performance.

[0045] In a preferred embodiment, the insulation performance of the insulating material to be evaluated is assessed based on the target trap depth, a first characteristic parameter, a second characteristic parameter, and an electronic state complexity parameter, including: The target trap depth, first feature parameter, second feature parameter, and electronic state complexity parameter are normalized to obtain the corresponding membership degree. Based on preset weight parameters, the membership degrees are weighted and summed to obtain the score of the insulation material to be evaluated, which is used to characterize the insulation performance.

[0046] Specifically, the target trap depth is normalized using a membership function that prioritizes larger values ​​to characterize the trend of stronger electron binding ability and better insulation performance of the material as the trap depth increases; as the trap depth parameter increases, its corresponding membership value gradually increases; the first characteristic parameter is normalized using a membership function that prioritizes smaller values ​​to characterize the trend of weakened local electron interactions and reduced carrier migration tendency within the material as the degree of local electron aggregation within the molecule decreases; the second characteristic parameter is normalized using a membership function that prioritizes smaller values ​​to characterize the trend of reduced electronic perturbation and improved insulation stability within the material as weak repulsive interactions within the molecule decrease; the electronic complexity state parameter is directly assigned (illustratively, when the number of peaks is 1). μ When 4=1, and the number of peaks is 3, μ (4=0), illustratively, the calculation results of each membership degree of the target molecular model are shown in Table 1: Table 1 Each membership degree is weighted and summed with preset parameters (illustratively, the weight parameters are [0.6, 0.2, 0.2, 0.1]) to obtain the score of the insulation material to be evaluated, which is used to characterize the insulation performance. The target molecular model insulation performance score results are shown in Table 2. Table 2 In this invention, membership parameters are weighted and summed with preset weight parameters to achieve a comprehensive fusion of multi-dimensional electronic structure parameters such as trap depth, weak interaction characteristics, and electronic state complexity, thereby obtaining a unified score for the insulating material to be evaluated. Compared with methods that evaluate insulation performance based on only a single parameter, this step can comprehensively reflect the influence of the coupling effects of multiple factors such as electron binding ability, electron localization characteristics, intramolecular weak interactions, and electronic state distribution complexity on the insulation performance of the material, improving the accuracy, comprehensiveness, and physical consistency of the insulation material performance evaluation results.

[0047] Based on the above method embodiments, the present invention provides corresponding apparatus embodiments.

[0048] A molecular simulation-based device for evaluating the performance of insulating materials, characterized in that it comprises: The material acquisition module is used to acquire the target molecular model of the insulating material to be evaluated. The electronic structure calculation module is used to perform first-principles electronic structure calculations on the target molecular model based on density functional theory, and to obtain the corresponding Fermi level, density of states function, molecular orbital energy levels and electronic wave functions. The peak analysis module is used to perform peak analysis on the density of states function to obtain the number of density of states peaks, and uses the number of density of states peaks as a parameter of electronic state complexity. The trap state energy level calculation module is used to calculate the trap state energy level of each local electronic state in the target molecular model based on the Fermi level, density of states function, and molecular orbital energy level. The target trap depth acquisition module is used to take the trap state energy level with the largest energy difference as the target trap depth based on the energy difference between the trap state energy level of each local electronic state and the preset equilibrium energy level. The weak interaction analysis module is used to construct the spatial distribution characteristics of electron density based on each electronic wavefunction, and to perform weak interaction analysis on the electron density variation characteristics to extract the first characteristic parameter characterizing intramolecular attractive interactions and the second characteristic parameter characterizing intramolecular repulsive interactions. The evaluation score calculation module is used to evaluate the insulation performance of the insulating material to be evaluated based on the target trap depth, the first feature parameter, the second feature parameter, and the electronic state complexity parameter.

[0049] In a preferred embodiment, based on density functional theory, first-principles electronic structure calculations are performed on the target molecular model to obtain the corresponding Fermi level, density of states function, molecular orbital energy levels, and electronic wavefunctions, including: Constructing the electronic density field based on density functional theory; constructing the external potential field based on the target molecule model; The initial electron density distribution function is determined based on the electron density field. The external potential field and the initial electron density distribution function are substituted into the Kohn–Sham equation as initial conditions for iterative solution to generate several electron wave functions. Each eigenvalue of the Kohn–Sham equation is taken as a molecular orbital energy level; Based on the distribution of molecular orbital energy levels, the energy of each molecular orbital level is divided into occupied and unoccupied molecular orbital levels. The Fermi level is determined based on the occupied and unoccupied molecular orbital energy levels. Energy level broadening is performed on each molecular orbital energy level to generate the density of states function.

[0050] Specifically, the spatial positions of each atom in the target molecular model are determined, and the reference electron density distribution of each isolated atom is extracted based on density functional theory. An initial electron density distribution function is constructed through spatial translation and superposition. Based on the spatial coordinates and nuclear charge distribution of each atomic nucleus in the molecule, an external Coulomb potential function is constructed to describe the interaction between the electron and the atomic nucleus. Based on the initially guessed electron density distribution and combined with the external potential field, the Kohn–Sham effective potential term is constructed, and the electron density and single-electron wavefunction are updated through a self-consistent iterative process, causing the system to gradually converge to the ground state solution. After the self-consistent iterative convergence, multiple eigenvalues ​​obtained from solving the Kohn–Sham equation are extracted, where each eigenvalue corresponds to a molecular orbital. Energy levels are used to characterize the energy distribution of electronic states. Solving the Kohn–Sham equation yields the electronic wavefunctions of each molecular orbital, which are used to describe the spatial distribution characteristics of different electronic states in the system. Based on the energy level order from low to high and combined with the electron filling principle, the occupancy of molecular orbital energy levels is determined: orbitals below a certain critical energy are occupied states, and orbitals above that energy are unoccupied states. Based on the electron occupancy probability, an energy boundary point is determined between the highest occupied state and the lowest unoccupied state; this boundary energy is the Fermi level, used to characterize the electron filling boundary of the system. Discrete molecular orbital energy levels are smoothed using Gaussian broadening or equivalent broadening functions, transforming the discrete energy levels into a density of states function, used to characterize the distribution density of electronic states in different energy ranges.

[0051] In a preferred real-time example, peak analysis is performed on the density of states function to obtain the number of density of states peaks. The number of density of states peaks is used as a parameter of electronic state complexity, including: Local maxima detection is performed on the density of states function to obtain the peak values ​​of the density of states at each local maximum point; The peak value of the density of states corresponding to each local maximum point is compared with a preset threshold, and the peak value of the density of states that is greater than the preset threshold is taken as the effective peak value. The number of effective peaks is counted to obtain the number of density of states peaks, which is then used as an electronic state complexity parameter to characterize the complexity of electronic state distribution.

[0052] Specifically, structural feature analysis is performed on the continuous density of states distribution curve along the energy axis to characterize the distribution characteristics of electronic states in different energy ranges; by analyzing the variation trend of the density of states function in the energy dimension, local density enhancement regions in the density of states function are identified, and the corresponding local maxima are determined; the density of states values ​​corresponding to each local maxima are used as candidate peaks, and the candidate peaks are screened based on a preset density threshold. When a candidate peak is greater than the preset threshold, it is determined as a valid peak; the valid peaks are statistically processed to obtain the number of density of states peaks; the number of density of states peaks is used as an electronic state complexity parameter to characterize the complexity of the electronic state distribution of the target molecule model.

[0053] In this invention, the changing trend of electronic state distribution is identified through structural feature analysis along the energy axis, and the local maxima corresponding to the local density enhancement region are further located, thereby transforming the continuous electronic state distribution into a set of candidate peaks with clear physical meaning. Subsequently, a preset density threshold is introduced to screen the candidate peaks, effectively eliminating insignificant peaks caused by numerical fluctuations or weakly contributing electronic states, thereby improving the stability and physical reliability of peak structure identification. On this basis, the effective peaks are statistically analyzed to obtain the number of density of states peaks, realizing a quantitative characterization of the dispersion of electronic state distribution and the degree of energy level splitting in energy space. Finally, the number of peaks is used as an electronic state complexity parameter, so that the originally continuous and high-dimensional electronic state distribution information is compressed into a single comparable structured index, thereby achieving a unified quantitative description of the electronic state complexity of different molecular models while maintaining the key features of electronic structure, and providing global constraints and feature inputs for subsequent electronic state screening and local electronic state analysis.

[0054] In a preferred embodiment, based on the Fermi level, density of states function, and molecular orbital energy levels, the trapped state energy levels of each local electronic state in the target molecular model are calculated, including: For each local maximum point, the peak value of the density of states is determined according to the position of the peak value on the energy axis; and the corresponding energy screening interval is constructed based on the reference energy and the preset interval. Determine whether each occupied molecular orbital energy level falls within the energy screening interval. If the occupied molecular orbital energy level falls within any energy screening interval, then the occupied molecular orbital energy level is taken as the candidate molecular orbital energy level. Obtain the electronic state corresponding to the candidate molecular orbital energy level as the candidate electronic state. The spatial electron density value of the target molecular model is calculated based on the electronic wave function corresponding to each occupied molecular orbital energy level. For each candidate electronic state, the electron density value is calculated using the corresponding electronic wave function, and the ratio of the corresponding electron density value to the spatial electron density value is calculated to obtain the electron density ratio. Candidate electronic states with an electron density ratio greater than the density ratio threshold are taken as local electronic states. For each local electronic state, the energy difference between the molecular orbital energy level and the Fermi level corresponding to the local electronic state is calculated to obtain the trap state energy level corresponding to that local electronic state.

[0055] Specifically, structural feature analysis is performed on the density of states function along the energy axis to identify local maxima and determine the energy positions corresponding to each local maxima as reference energies. Based on the reference energies and preset energy intervals, multiple energy screening interval sets are constructed on the energy axis to characterize the energy range of possible electronic state distributions. Occupied molecular orbital energy levels are matched with the energy screening interval sets. When an occupied molecular orbital energy level falls into any energy screening interval, the electronic state corresponding to that molecular orbital energy level is determined as a candidate electronic state. Based on the electronic wavefunctions corresponding to each occupied molecular orbital energy level, the spatial electron density distribution function of the target molecular model is calculated to characterize the overall distribution of electrons in space. For each candidate electronic state, the local electron density distribution is calculated based on its corresponding electronic wavefunction, and the ratio of the local electron density to the spatial electron density distribution is calculated to obtain the electron density ratio. When the electron density ratio is greater than a preset density threshold, the corresponding candidate electronic state is determined as a local electronic state. For each local electronic state, the energy difference between its corresponding molecular orbital energy level and the Fermi level is calculated to obtain the trap state energy level of the corresponding local electronic state.

[0056] In this invention, based on the structural feature analysis of the density of states function along the energy axis, local maxima are identified and their corresponding reference energies are determined, thereby transforming the originally continuous electronic state distribution into physically meaningful discrete energy calibration points. Furthermore, an energy screening interval set is constructed based on the reference energy and a preset energy interval, achieving a partitioned characterization of the possible distribution range of electronic states. The occupied state molecular orbital energy levels are matched and screened with this interval set, initially determining candidate electronic states from an energy perspective, thus improving the targeting of electronic state screening. A spatial electron density distribution function is constructed based on the occupied state molecular orbital wavefunction, and a ratio of local electron density to overall spatial electron density is introduced for candidate electronic states, thus improving the spatial distribution. The localization characteristics of electronic states are further constrained by the angle of arrangement, thereby effectively eliminating non-local or weakly contributing electronic states and improving the physical rationality and stability of electronic state identification. On this basis, electronic states that meet the localization conditions are identified as local electronic states, and the energy difference between their corresponding molecular orbital energy levels is calculated using the Fermi level as a unified energy benchmark, thereby obtaining the trap state energy level and realizing the quantitative characterization of the electron binding depth and trapping ability. Overall, this step introduces a multi-layer judgment mechanism of "energy screening - spatial constraint - energy level calibration" to realize the hierarchical analysis of electronic states from continuous distribution to localized energy level characterization, thereby improving the accuracy, physical consistency and characterization ability of electronic state screening and the ability to characterize the differences in the electronic structure of materials.

[0057] In a preferred embodiment, the spatial electron density value of the target molecular model is calculated based on the electron wavefunction corresponding to each occupied molecular orbital energy level, including: The spatial electron density distribution function is obtained by superimposing the modulus squares of the electronic wavefunctions corresponding to all occupied molecular orbital energy levels; Integrating the space electron density distribution function yields the space electron density value.

[0058] Specifically, the electronic wavefunctions corresponding to all occupied molecular orbital energy levels are extracted, with each occupied molecular orbital energy level corresponding to an electronic wavefunction, used to characterize the probability distribution of electrons in space; the modulus square operation is performed on each electronic wavefunction to obtain the electronic probability density distribution of the corresponding electronic state in space; the electronic probability density distributions corresponding to all occupied states are spatially superimposed to construct the spatial electron density distribution function of the target molecular model, used to characterize the overall distribution of electrons in the system across the entire spatial range; based on the spatial electron density distribution function, the spatial region corresponding to the target molecular model is integrally calculated to obtain the spatial electron density value of the target molecular model.

[0059] In a preferred embodiment, the spatial electron density value of the target molecular model is calculated based on the electron wavefunction corresponding to each occupied molecular orbital energy level, including: The spatial electron density distribution function is obtained by superimposing the modulus squares of the electronic wavefunctions corresponding to all occupied molecular orbital energy levels; Integrating the space electron density distribution function yields the space electron density value.

[0060] Specifically, the electronic wavefunctions corresponding to all occupied molecular orbital energy levels are extracted, with each occupied molecular orbital energy level corresponding to an electronic wavefunction, used to characterize the probability distribution of electrons in space; the modulus square operation is performed on each electronic wavefunction to obtain the electronic probability density distribution of the corresponding electronic state in space; the electronic probability density distributions corresponding to all occupied states are spatially superimposed to construct the spatial electron density distribution function of the target molecular model, used to characterize the overall distribution of electrons in the system across the entire spatial range; based on the spatial electron density distribution function, the spatial region corresponding to the target molecular model is integrally calculated to obtain the spatial electron density value of the target molecular model.

[0061] In a preferred embodiment, the insulation performance of the insulating material to be evaluated is assessed based on the target trap depth, a first characteristic parameter, a second characteristic parameter, and an electronic state complexity parameter, including: The target trap depth, first feature parameter, second feature parameter, and electronic state complexity parameter are normalized to obtain the corresponding membership degree. Based on preset weight parameters, the membership degrees are weighted and summed to obtain the score of the insulation material to be evaluated, which is used to characterize the insulation performance.

[0062] Specifically, the target trap depth is normalized using a membership function that prioritizes larger values ​​to characterize the trend of stronger electron binding ability and better insulation performance of the material as the trap depth increases; as the trap depth parameter increases, its corresponding membership value gradually increases; the first characteristic parameter is normalized using a membership function that prioritizes smaller values ​​to characterize the trend of weakened local electron interactions and reduced carrier migration tendency within the material as the degree of local electron aggregation within the molecule decreases; the second characteristic parameter is normalized using a membership function that prioritizes smaller values ​​to characterize the trend of reduced electronic perturbation and improved insulation stability within the material as weak repulsive interactions within the molecule decrease; the electronic complexity state parameter is directly assigned (illustratively, when the number of peaks is 1). μ When 4=1, and the number of peaks is 3, μ (4=0); The membership degree is weighted and summed with the preset parameters (illustratively, the weight parameters are [0.6,0.2,0.2,0.1]) to obtain the score of the insulation material to be evaluated, which is used to characterize the insulation performance.

[0063] In this invention, membership parameters are weighted and summed with preset weight parameters to achieve a comprehensive fusion of multi-dimensional electronic structure parameters such as trap depth, weak interaction characteristics, and electronic state complexity, thereby obtaining a unified score for the insulating material to be evaluated. Compared with methods that evaluate insulation performance based on only a single parameter, this step can comprehensively reflect the influence of the coupling effects of multiple factors such as electron binding ability, electron localization characteristics, intramolecular weak interactions, and electronic state distribution complexity on the insulation performance of the material, improving the accuracy, comprehensiveness, and physical consistency of the insulation material performance evaluation results.

[0064] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A method for evaluating the performance of insulating materials based on molecular simulation, characterized in that, include: Obtain the target molecular model of the insulating material to be evaluated; Based on density functional theory, first-principles electronic structure calculations were performed on the target molecular model to obtain the corresponding Fermi level, density of states function, molecular orbital energy levels and electronic wave functions. Peak analysis of the density of states function is performed to obtain the number of density of states peaks, and the number of density of states peaks is used as a parameter of electronic state complexity. Based on the Fermi level, density of states function, and molecular orbital energy levels, the trap state energy level of each local electronic state in the target molecular model is calculated; Based on the energy difference between the trap state energy level of each local electronic state and the preset equilibrium energy level, the trap state energy level with the largest energy difference is taken as the target trap depth. Based on each electronic wavefunction, the spatial distribution characteristics of electron density are constructed, and weak interaction analysis is performed on the electron density variation characteristics to extract the first characteristic parameter characterizing intramolecular attractive interactions and the second characteristic parameter characterizing intramolecular repulsive interactions. The insulation performance of the insulating material to be evaluated is assessed based on the target trap depth, the first characteristic parameter, the second characteristic parameter, and the electronic state complexity parameter.

2. The method for evaluating the performance of insulating materials based on molecular simulation as described in claim 1, characterized in that, Based on density functional theory, first-principles electronic structure calculations are performed on the target molecular model to obtain the corresponding Fermi level, density of states function, molecular orbital energy levels, and electronic wavefunctions, including: Constructing the electronic density field based on density functional theory; constructing the external potential field based on the target molecule model; The initial electron density distribution function is determined based on the electron density field. The external potential field and the initial electron density distribution function are substituted into the Kohn–Sham equation as initial conditions for iterative solution to generate several electron wave functions. Each eigenvalue of the Kohn–Sham equation is taken as a molecular orbital energy level; Based on the distribution of molecular orbital energy levels, the energy of each molecular orbital level is divided into occupied and unoccupied molecular orbital levels. The Fermi level is determined based on the occupied and unoccupied molecular orbital energy levels. Energy level broadening is performed on each molecular orbital energy level to generate the density of states function.

3. The method for evaluating the performance of insulating materials based on molecular simulation as described in claim 2, characterized in that, Peak analysis is performed on the density of states function to obtain the number of density of states peaks. The number of density of states peaks is used as a parameter of electronic state complexity, including: Local maxima detection is performed on the density of states function to obtain the peak values ​​of the density of states at each local maximum point; The peak value of the density of states corresponding to each local maximum point is compared with a preset threshold, and the peak value of the density of states that is greater than the preset threshold is taken as the effective peak value. The number of effective peaks is counted to obtain the number of density of states peaks, which is then used as an electronic state complexity parameter to characterize the complexity of electronic state distribution.

4. The method for evaluating the performance of insulating materials based on molecular simulation as described in claim 3, characterized in that, Based on the Fermi level, density of states function, and molecular orbital energy levels, the trapped state energy levels of each local electronic state in the target molecular model are calculated, including: For each local maximum point, the peak value of the density of states is determined according to the position of the peak value on the energy axis; and the corresponding energy screening interval is constructed based on the reference energy and the preset interval. Determine whether each occupied molecular orbital energy level falls within the energy screening interval. If the occupied molecular orbital energy level falls within any energy screening interval, then the occupied molecular orbital energy level is taken as the candidate molecular orbital energy level. Obtain the electronic state corresponding to the candidate molecular orbital energy level as the candidate electronic state. The spatial electron density value of the target molecular model is calculated based on the electronic wave function corresponding to each occupied molecular orbital energy level. For each candidate electronic state, the electron density value is calculated using the corresponding electronic wavefunction, and the ratio of the corresponding electron density value to the spatial electron density value is calculated to obtain the electron density ratio. Candidate electronic states with electron density ratios greater than the density ratio threshold are taken as local electronic states. For each local electronic state, the energy difference between the molecular orbital energy level and the Fermi level corresponding to the local electronic state is calculated to obtain the trap state energy level corresponding to that local electronic state.

5. The method for evaluating the performance of insulating materials based on molecular simulation as described in claim 4, characterized in that, The spatial electron density value of the target molecular model is calculated based on the electron wavefunction corresponding to each occupied molecular orbital energy level, including: The spatial electron density distribution function is obtained by superimposing the modulus squares of the electronic wavefunctions corresponding to all occupied molecular orbital energy levels; Integrating the space electron density distribution function yields the space electron density value.

6. The method for evaluating the performance of insulating materials based on molecular simulation as described in claim 5, characterized in that, Based on each electronic wavefunction, the spatial distribution characteristics of electron density are constructed, and weak interaction analysis is performed on the electron density variation characteristics to extract the first characteristic parameter characterizing intramolecular attractive interactions and the second characteristic parameter characterizing intramolecular repulsive interactions, including: The space of the target molecular model is discretized to obtain several discrete grids; For each discrete grid, the spatial gradient value of the electron density corresponding to the discrete grid is calculated by performing partial derivatives on the spatial electron density distribution function. The electron density value corresponding to the discrete grid is obtained by integrating the spatial electron density distribution function. The ratio of the spatial gradient of electron density to the electron density value is calculated to obtain the reduced density gradient of the discrete grid. Based on the reduced density gradient value of each discrete grid, a reduced density gradient scalar field is constructed. The electron density matrix is ​​obtained by taking the second derivative of the spatial electron density distribution function. The electron density matrix is ​​then decomposed into eigenvalues ​​to obtain eigenvalues ​​that characterize the local curvature variation of the electron density. Based on the eigenvalues, a symbolic piecewise function is constructed. The symbolic piecewise function is then multiplied by the spatial electron density distribution function to obtain the electron interaction type labeling function. Projecting the type labeling function onto a preset isosurface of the approximate density gradient scalar field yields a three-dimensional candidate spatial point set based on the full-space electron density distribution. In the candidate spatial point set, the interaction type regions are divided according to the positive and negative attributes of the labeling function, and the spatial points closest to zero are selected in the positive and negative value regions respectively. The label function value of the spatial point closest to zero in the negative region is used as the first feature parameter, and the label function value of the spatial point closest to zero in the positive region is used as the second feature parameter.

7. The method for evaluating the performance of insulating materials based on molecular simulation as described in claim 6, characterized in that, Based on the target trap depth, the first characteristic parameter, the second characteristic parameter, and the electronic state complexity parameter, the insulation performance of the insulating material to be evaluated is assessed, including: The target trap depth, first feature parameter, second feature parameter, and electronic state complexity parameter are normalized to obtain the corresponding membership degree. Based on preset weight parameters, the membership degrees are weighted and summed to obtain the score of the insulation material to be evaluated, which is used to characterize the insulation performance.

8. A device for evaluating the performance of insulating materials based on molecular simulation, characterized in that, include: The material acquisition module is used to acquire the target molecular model of the insulating material to be evaluated. The material acquisition module is used to perform first-principles electronic structure calculations on the target molecular model based on density functional theory, and to obtain the corresponding Fermi level, density of states function, molecular orbital energy levels and electronic wave functions. The peak analysis module is used to perform peak analysis on the density of states function to obtain the number of density of states peaks, and uses the number of density of states peaks as a parameter of electronic state complexity. The trap state energy level calculation module is used to calculate the trap state energy level of each local electronic state in the target molecular model based on the Fermi level, density of states function, and molecular orbital energy level. The target trap depth acquisition module is used to take the trap state energy level with the largest energy difference as the target trap depth based on the energy difference between the trap state energy level of each local electronic state and the preset equilibrium energy level. The weak interaction analysis module is used to construct the spatial distribution characteristics of electron density based on each electronic wavefunction, and to perform weak interaction analysis on the electron density variation characteristics to extract the first characteristic parameter characterizing intramolecular attractive interactions and the second characteristic parameter characterizing intramolecular repulsive interactions. The evaluation score calculation module is used to evaluate the insulation performance of the insulating material to be evaluated based on the target trap depth, the first feature parameter, the second feature parameter, and the electronic state complexity parameter.

9. An electronic device, characterized in that, The device includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor, when executing the computer program, implements a molecular simulation-based method for evaluating the performance of insulating materials as described in any one of claims 1 to 7.

10. A storage medium, characterized in that, The storage medium includes a stored computer program, wherein, when the computer program is executed, it controls the device containing the storage medium to perform a molecular simulation-based method for evaluating the performance of insulating materials as described in any one of claims 1 to 7.

Images