Method, device, equipment and medium for selecting liquid impregnant suitable for power capacitor

Abstract

The application discloses a liquid impregnant selection method, device, equipment and medium suitable for power capacitors, relates to the technical field of impregnant selection, and comprises the following steps: constructing a target molecular descriptor feature quantity set; the molecular descriptor is used for characterizing the gas adsorption performance of the liquid impregnant; constructing an adsorption system model of the liquid impregnant when adsorbing a target gas molecule through density functional theory, and determining corresponding micro-adsorption characteristic parameters; the adsorption energy, adsorption distance and charge transfer amount are used for characterizing the gas adsorption performance of the liquid impregnant; normal distribution verification is performed on the molecular descriptors; according to the verification result and by using a preset correlation statistical method, the correlation between the molecular descriptors and the micro-adsorption characteristic parameters is analyzed, so that the target molecular descriptor is determined; and based on the correlation analysis result and the target molecular descriptor, comparative analysis is performed on the liquid impregnants to be selected for the target power capacitor, so that the target liquid impregnant with the strongest gas adsorption performance is screened out.

Description

Technical Field

[0001] This invention relates to the field of impregnating agent selection technology, and particularly to a method, apparatus, equipment and medium for selecting liquid impregnating agents suitable for power capacitors. Background Technology

[0002] Liquid impregnating agents are the core of high-voltage power capacitor insulation systems, and their performance determines the reliability and service life of capacitor insulation. Gas absorption is a key performance characteristic of impregnating agents and is crucial for the safe operation of capacitors. Accurately assessing the gas absorption performance of liquid impregnating agents is of great significance for selecting impregnating agents for high-voltage power capacitors. Currently, the assessment of impregnating agent gas absorption performance largely relies on macroscopic experimental methods. For example, by using a piston to extract insulating oil samples, utilizing the principle of vacuum degassing, and combining pressure detection components to measure the pressure change of the oil sample after degassing, the gas content in the insulating oil is calculated. However, this method does not overcome the inherent limitations of traditional experimental assessment methods, resulting in low efficiency and high cost in gas absorption assessment techniques, and inefficient selection of liquid impregnating agents with strong gas absorption performance.

[0003] In summary, how to achieve efficient evaluation of the gas absorption performance of liquid impregnating agents in order to improve the efficiency of selecting liquid impregnating agents with strong gas absorption performance is an urgent problem to be solved. Summary of the Invention

[0004] In view of this, the purpose of this invention is to provide a method, apparatus, device, and medium for selecting liquid impregnating agents suitable for power capacitors, which can achieve efficient evaluation of the gas-absorbing performance of liquid impregnating agents to improve the efficiency of selecting liquid impregnating agents with strong gas-absorbing performance. The specific solution is as follows: In a first aspect, this application provides a method for selecting a liquid impregnating agent suitable for power capacitors, including: Construct a set of target molecule descriptor features; the set of target molecule descriptor features includes several molecular descriptors characterizing the gas-getter properties of the liquid impregnating agent; An adsorption system model for the adsorption of target gas molecules by a liquid impregnating agent is constructed using density functional theory, and the corresponding microscopic adsorption characteristic parameters are determined through the adsorption system model. The microscopic adsorption characteristic parameters include adsorption energy, adsorption distance, and charge transfer amount, which characterize the gas absorption performance of the liquid impregnating agent. The normal distribution verification of the molecular descriptors is performed. Based on the verification results and using the corresponding preset correlation statistical method, the correlation between the molecular descriptors and the microscopic adsorption characteristic parameters is analyzed, so as to determine the target molecular descriptor based on the obtained correlation analysis results. Based on the correlation analysis results and the target molecule descriptor, a comparative analysis is performed on the candidate liquid impregnating agents for the target power capacitor, so as to select the target liquid impregnating agent with the strongest gas absorption performance from the candidate liquid impregnating agents according to the obtained comparative analysis results.

[0005] Optionally, constructing the target molecule descriptor feature set includes: Based on the principle of minimum energy, the law of charge conservation, density functional theory, and electronic structure theory, several molecular descriptors characterizing the gas absorption performance of liquid impregnating agents are screened from the molecular feature dimension. The molecular descriptors are classified according to the molecular feature dimensions to construct a target molecular descriptor feature set including the molecular descriptors.

[0006] Optionally, the step of constructing an adsorption system model for the adsorption of target gas molecules by the liquid impregnating agent using density functional theory includes: Using density functional theory and the target quantum chemistry module of the target molecule simulation software, an impregnating agent molecular model and a target gas molecule gas molecule model are constructed to build an initial adsorption system model when the liquid impregnating agent adsorbs the target gas molecule. The initial adsorption system model includes multiple adsorption system models with different initial conditions. The initial conditions include the initial distance and action site of the target gas molecule placed in the liquid impregnating agent. Using target density functional theory and target dispersion correction methods, the initial adsorption system model, the impregnating agent molecule model, and the gas molecule model are geometrically optimized to obtain target stable geometric configurations; the target stable geometric configurations include the stable geometric configurations of the adsorption system, the stable geometric configurations of the impregnating agent molecules, and the stable geometric configurations of the gas molecules; The single-point energy of the target stable geometry is calculated based on the target hybrid functional to obtain the corresponding single-point energy value.

[0007] Optionally, determining the corresponding microscopic adsorption characteristic parameters through the adsorption system model includes: Based on the single-point energy value, the adsorption energy of the stable geometry of the adsorption system is determined, so as to determine the target stable geometry of the adsorption system with the minimum adsorption energy. The adsorption distance of the liquid impregnating agent for the target gas molecules is extracted from the stable geometry of the target adsorption system. The charge transfer behavior of the stable geometry of the target adsorption system is quantitatively characterized using target charge analysis to calculate the corresponding charge transfer amount; the charge transfer behavior refers to the charge transfer behavior during the adsorption of the target gas molecules by the liquid impregnating agent. The microscopic adsorption characteristic parameters include the adsorption energy of the stable geometry of the target adsorption system, the adsorption distance, and the charge transfer amount.

[0008] Optionally, the step of performing normal distribution verification on the plurality of molecular descriptors, and analyzing the correlation between the plurality of molecular descriptors and the microscopic adsorption characteristic parameters based on the obtained verification results and using a corresponding preset correlation statistical method, includes: The target test method was used to perform normal distribution verification on the aforementioned molecular descriptors to obtain the corresponding verification results; If the verification results indicate that the molecular descriptor follows a normal distribution, then the correlation between the molecular descriptor and the microscopic adsorption characteristic parameter is analyzed using the Pearson correlation coefficient method. If the verification results indicate that the molecular descriptor does not follow a normal distribution, then the Spearman correlation coefficient method is used to analyze the correlation between the molecular descriptor and the microscopic adsorption characteristic parameters.

[0009] Optionally, the step of analyzing the correlation between the plurality of molecular descriptors and the microscopic adsorption characteristic parameters using a corresponding preset correlation statistical method, so as to determine the target molecular descriptor based on the obtained correlation analysis results, includes: Using corresponding preset correlation statistical methods, the correlation coefficients between the molecular descriptors and the microscopic adsorption characteristic parameters are determined respectively; Based on the absolute value of the correlation coefficient, candidate molecule descriptors that satisfy a preset strong correlation with the adsorption energy, the adsorption distance, and the charge transfer amount are selected. Redundant molecular descriptors are removed from the candidate molecular descriptors, and the removed candidate molecular descriptors are determined as the target molecular descriptors.

[0010] Optionally, the step of selecting the target liquid impregnating agent with the strongest gas-getting performance from the candidate liquid impregnating agents based on the obtained comparative analysis results further includes: The macroscopic air-absorbing performance of the candidate liquid impregnating agent was tested using an air-absorbing tester to obtain the corresponding experimental results. Cross-validation was performed based on the experimental test results and the comparative analysis results to screen out the target liquid impregnating agent with the strongest air absorption performance from the candidate liquid impregnating agents.

[0011] Secondly, this application provides a liquid impregnating agent selection device suitable for power capacitors, comprising: A set construction module is used to construct a set of target molecule descriptor features; the set of target molecule descriptor features includes several molecular descriptors characterizing the uptake performance of the liquid impregnating agent; The parameter determination module is used to construct an adsorption system model of the liquid impregnating agent adsorbing target gas molecules using density functional theory, and to determine the corresponding microscopic adsorption characteristic parameters through the adsorption system model; the microscopic adsorption characteristic parameters include adsorption energy, adsorption distance and charge transfer amount, which characterize the gas absorption performance of the liquid impregnating agent. The correlation analysis module is used to perform normal distribution verification on the plurality of molecular descriptors, and analyze the correlation between the plurality of molecular descriptors and the microscopic adsorption characteristic parameters based on the obtained verification results and the corresponding preset correlation statistical methods, so as to determine the target molecular descriptor based on the obtained correlation analysis results. The comparative analysis module is used to perform comparative analysis on the candidate liquid impregnating agents for the target power capacitor based on the correlation analysis results and the target molecule descriptor, so as to screen out the target liquid impregnating agent with the strongest gas absorption performance from the candidate liquid impregnating agents according to the obtained comparative analysis results.

[0012] Thirdly, this application provides an electronic device, comprising: Memory, used to store computer programs; A processor is used to execute the computer program to implement the aforementioned method for selecting liquid impregnating agents suitable for power capacitors.

[0013] Fourthly, this application provides a computer-readable storage medium for storing a computer program; wherein, when the computer program is executed by a processor, it implements the aforementioned method for selecting a liquid impregnating agent suitable for power capacitors.

[0014] In this application, a set of target molecule descriptor features is constructed. This set includes several molecular descriptors characterizing the gas-getter performance of the liquid impregnating agent. An adsorption system model is constructed using density functional theory to model the adsorption of target gas molecules by the liquid impregnating agent, and corresponding microscopic adsorption characteristic parameters are determined using this model. These microscopic adsorption characteristic parameters include adsorption energy, adsorption distance, and charge transfer, which characterize the gas-getter performance of the liquid impregnating agent. The molecular descriptors are validated using a normal distribution. Based on the validation results and a pre-defined correlation statistical method, the correlation between the molecular descriptors and the microscopic adsorption characteristic parameters is analyzed to determine the target molecule descriptor based on the obtained correlation analysis results. Based on the correlation analysis results and the target molecule descriptors, a comparative analysis is performed on candidate liquid impregnating agents for the target power capacitor. Based on the comparative analysis results, the target liquid impregnating agent with the strongest gas-getter performance is selected from the candidate liquid impregnating agents. As can be seen from the above, this application first constructs a set of feature quantities containing molecular descriptors characterizing the gas absorption performance of liquid impregnating agents, and then constructs an adsorption system model of the liquid impregnating agent adsorbing target gas molecules through density functional theory to obtain microscopic adsorption characteristic parameters including adsorption energy, adsorption distance, and charge transfer amount. Subsequently, the normal distribution of the aforementioned molecular descriptors is verified, and the correlation between them and the microscopic adsorption characteristic parameters is analyzed through a preset correlation statistical method to determine the target molecular descriptor. Finally, based on the correlation results and the target molecular descriptor, the candidate liquid impregnating agents for the target power capacitor are compared and analyzed to screen out the target liquid impregnating agent with the best gas absorption performance. In this way, through the above-described process of this application, an adsorption system model is constructed using density functional theory to obtain microscopic adsorption characteristic parameters, which can accurately characterize the gas-getter mechanism of liquid impregnating agents at the microscopic level, providing a theoretical basis for performance evaluation. By screening target molecular descriptors through normal distribution verification and correlation statistics methods, key features strongly correlated with gas-getter performance can be effectively extracted, improving the accuracy and efficiency of subsequent evaluation. Based on the screened molecular descriptors, quantitative comparative analysis of the impregnating agents to be selected is performed, revealing the adsorption energy, adsorption distance, and charge transfer laws of different impregnating agents at the molecular level. This allows for a direct differentiation of the gas-getter performance of different materials, achieving efficient and accurate optimal material screening, and thus realizing efficient evaluation of the gas-getter performance of liquid impregnating agents to improve the efficiency of selecting liquid impregnating agents with strong gas-getter performance. Attached Figure Description

[0015] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0016] Figure 1 This application discloses a flowchart of a liquid impregnating agent selection method for power capacitors. Figure 2 This is a flowchart illustrating a liquid impregnating agent selection method for power capacitors disclosed in this application; Figure 3(a) is a schematic diagram of the correlation coefficient between adsorption energy and molecular descriptor disclosed in this application; Figure 3(b) is a schematic diagram of the correlation coefficient between adsorption distance and molecular descriptor disclosed in this application; Figure 3(c) is a schematic diagram of the correlation coefficient between charge transfer and molecular descriptor disclosed in this application; Figure 4 This is a schematic diagram of a liquid impregnating agent selection device for power capacitors disclosed in this application; Figure 5 This is a structural diagram of an electronic device disclosed in this application. Detailed Implementation

[0017] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0018] Gas absorption is a key performance characteristic of impregnating agents and is crucial for the safe operation of capacitors. Accurately assessing the gas absorption performance of liquid impregnating agents is of great significance for selecting impregnating agents for high-voltage power capacitors. Currently, the assessment of impregnating agent gas absorption performance largely relies on macroscopic experimental methods. For example, by extracting insulating oil samples with a piston, utilizing the principle of vacuum degassing, and combining this with a pressure detection component to measure the pressure change after degassing, the gas content in the insulating oil can be calculated. However, this method does not overcome the inherent limitations of traditional experimental assessment methods, resulting in low efficiency, high cost, and inefficient selection of liquid impregnating agents with strong gas absorption performance.

[0019] To overcome the above-mentioned technical problems, this application provides a method for selecting liquid impregnating agents suitable for power capacitors, which can achieve efficient evaluation of the gas absorption performance of liquid impregnating agents to improve the efficiency of selecting liquid impregnating agents with strong gas absorption performance.

[0020] See Figure 1 As shown, this embodiment of the invention discloses a method for selecting a liquid impregnating agent suitable for power capacitors, including: Step S11: Construct a set of target molecule descriptor features; the set of target molecule descriptor features includes several molecular descriptors characterizing the gas absorption performance of the liquid impregnating agent.

[0021] In this embodiment, a set of target molecular descriptor features for characterizing the gas absorption performance of liquid impregnating agents is constructed, which includes several molecular descriptors.

[0022] It should be noted that liquid impregnating agents are the core of the insulation system of high-voltage power capacitors, and their performance determines the reliability and service life of the capacitor insulation. Gas absorption is a key performance characteristic of impregnating agents and is crucial for the safe operation of capacitors. Accurately assessing the gas absorption performance of liquid impregnating agents is of great significance for the selection of impregnating agents for high-voltage power capacitors. Currently, the assessment of the gas absorption performance of impregnating agents largely relies on macroscopic experimental methods, which have the inherent limitations of traditional experimental assessment methods. Research on the microscopic adsorption mechanism of impregnating agents is still relatively lacking, and it is difficult to fully consider the inherent microscopic characteristics of molecules. Therefore, this application provides a method for selecting liquid impregnating agents suitable for power capacitors, comprising four aspects: 1) Constructing a set of key molecular descriptor characteristic quantities characterizing the gas absorption performance of the impregnating agent: Based on the principle of minimum energy, the law of charge conservation, and electronic structure theory, 11 key molecular descriptors affecting the gas absorption performance of the liquid impregnating agent are selected to construct a set of key molecular descriptor characteristic quantities characterizing the gas absorption performance of the impregnating agent; 2) Calculating numerical values ​​of microscopic physical property parameters characterizing the gas absorption of the impregnating agent molecules: Based on density functional theory, 30 typical impregnating agent molecule models are constructed, and the adsorption energy, adsorption distance, and charge transfer of hydrogen adsorbed by the selected impregnating agent molecules are calculated; 3) Selecting and determining molecular descriptors closely related to the microscopic adsorption performance of the impregnating agent molecules: Performing a normal distribution test on all molecular descriptors, and selecting those that conform to the normal distribution test... Descriptors exhibiting state distributions and those not conforming to normal distributions were analyzed using Pearson and Spearman correlation coefficient methods, respectively, to assess their correlation with adsorption energy, charge transfer, and adsorption distance. This process identified the most strongly correlated key molecular descriptors. Comparative analysis of the gas-adsorption performance of different types of liquid impregnating agents was then conducted using molecular descriptors. Combined with macroscopic-level comparative analysis of liquid impregnating agent gas adsorption performance, this provides theoretical support and experimental evidence for elucidating the gas-adsorption mechanism of different types of impregnating agents. Furthermore, it provides methodological support for rapidly screening impregnating agent molecules with strong gas adsorption properties. Characterizing and analyzing the microscopic gas adsorption performance of impregnating agent molecules through molecular descriptors can be used to rapidly and accurately evaluate the gas adsorption performance of different types of liquid impregnating agents for high-voltage power capacitors, providing methodological support for selecting impregnating agents with strong gas adsorption performance. Figure 2The diagram shows a flowchart of a liquid impregnating agent selection method for power capacitors provided in this application. By extracting the molecular descriptor and microscopic property parameters of the impregnating agent, its gas-getting performance can be quickly evaluated using a mathematical model. This effectively solves the problems of low efficiency and high cost in existing gas-getting performance evaluation techniques, achieving efficient evaluation of the gas-getting performance of impregnating agents. This is of great significance for accelerating high-throughput screening of highly gas-getting liquid impregnating agent molecules.

[0023] Specifically, based on the principle of minimum energy, the law of charge conservation, density functional theory, and electronic structure theory, several molecular descriptors characterizing the gas-getter performance of liquid impregnating agents are screened from the molecular feature dimension. These molecular descriptors are then classified according to the molecular feature dimension to construct a target molecular descriptor feature set including these descriptors. That is, the Hohenberg-Kohn theorem (HK theorem, the foundation of density functional theory) in density functional theory states that for a quantum system with a definite number of particles, the minimum value of its energy functional corresponds to the ground state energy of the system. Adsorption energy quantitatively describes the binding strength between gas molecules and impregnating agent molecules; the larger its absolute value, the easier it is for gas molecules to be stably adsorbed. Adsorption distance reflects the spatial proximity between the gas and the impregnating agent; the shorter the distance, the stronger the van der Waals forces, π-π stacking, and charge transfer, resulting in higher adsorption stability. Charge transfer quantifies the redistribution of electron clouds between molecules; the greater the transfer amount, the stronger the electron donor-acceptor interaction, the easier it is for gas molecules to be chemically fixed, thereby increasing the gas-getter rate. Therefore, based on the principle of minimum energy, the law of charge conservation, density functional theory, and electronic structure theory, several molecular descriptors affecting the gas absorption performance of liquid impregnating agents are selected from the molecular feature dimension. These descriptors are then classified according to the molecular feature dimension, constructing a set of target molecular descriptor feature quantities characterizing the gas absorption performance of the impregnating agent, including these molecular descriptors. In this way, this embodiment constructs a set of target molecular descriptor feature quantities specifically characterizing the gas absorption performance of liquid impregnating agents. This allows for the systematic extraction of key features related to gas absorption performance at the molecular level, providing a standardized and targeted feature foundation for subsequent quantitative analysis and performance evaluation. Using multiple theoretical bases to screen molecular descriptors from the molecular feature dimension ensures that the selected descriptors are highly correlated with the gas absorption mechanism, improving feature effectiveness. Classifying the descriptors according to the molecular feature dimension to construct the set makes the feature system structure clear and the dimensions well-defined, providing standardized and systematic feature support for subsequent adsorption performance analysis.

[0024] Step S12: Construct an adsorption system model for the liquid impregnating agent to adsorb target gas molecules using density functional theory, and determine the corresponding microscopic adsorption characteristic parameters through the adsorption system model; the microscopic adsorption characteristic parameters include adsorption energy, adsorption distance and charge transfer amount, which characterize the gas absorption performance of the liquid impregnating agent.

[0025] In this embodiment, density functional theory is used to construct an adsorption system model for the liquid impregnating agent to adsorb target gas molecules, and microscopic adsorption characteristic parameters, including adsorption energy, adsorption distance, and charge transfer amount, are obtained through this model to characterize the gas absorption performance.

[0026] It should be noted that the process of constructing the adsorption system model for the liquid impregnating agent adsorbing target gas molecules using density functional theory is as follows: Using density functional theory and the target quantum chemistry module of the target molecule simulation software, an impregnating agent molecule model and a target gas molecule model are constructed to build an initial adsorption system model for the liquid impregnating agent adsorbing the target gas molecules. The initial adsorption system model includes multiple adsorption system models with different initial conditions. The initial conditions include the initial distance and action site of the target gas molecules placed on the liquid impregnating agent. Using target density functional theory and target dispersion correction methods, the initial adsorption system model, the impregnating agent molecule model, and the gas molecule model are geometrically optimized to obtain a target stable geometric configuration. The target stable geometric configuration includes the adsorption system stable geometric configuration, the impregnating agent molecule stable geometric configuration, and the gas molecule stable geometric configuration. Based on the target hybrid functional theory, the single-point energy of the target stable geometric configuration is calculated to obtain the corresponding single-point energy value. The target molecule simulation software can be Material Studios software (a materials calculation software, an integrated simulation platform for materials science); the target quantum chemistry module can be the DMol3 module in Material Studio software (a quantum chemistry module for molecular and solid-state calculations based on density functional theory); and the target gas molecule can be hydrogen (H2) molecule. That is, based on density functional theory (DFT) and the quantum chemistry module of the target molecule simulation software, molecular models of the liquid impregnating agent and the target gas molecule are constructed respectively. In order to screen out the lowest energy adsorption configuration, the target gas molecules are placed on the impregnating agent molecule model with different initial distances and action sites to form multiple initial adsorption system models when the liquid impregnating agent adsorbs the target gas molecules. Then, the target density functional, such as the generalized gradient approximation (GGA) GGA-PBE (Perdew-Burke-Ernzerhof) functional, combined with the target dispersion correction method, such as DFT-D2 dispersion correction, is used to perform geometric optimization on various models to obtain the target stable geometric configuration. Finally, the single-point energy value is obtained by using the target hybrid functional, such as the high-precision M062X (Gaussian M06-2X, a function approximation method) hybrid functional.

[0027] It should be further pointed out that the processing flow for determining the corresponding microscopic adsorption characteristic parameters through the adsorption system model is as follows: Based on the single-point energy value, the adsorption energy of the stable geometry of the adsorption system is determined to identify the target stable geometry of the adsorption system with the minimum adsorption energy; from the target stable geometry of the adsorption system, the adsorption distance of the liquid impregnating agent on the target gas molecules is extracted; using target charge analysis, the charge transfer behavior of the target stable geometry of the adsorption system is quantitatively characterized to calculate the corresponding charge transfer amount; the charge transfer behavior is the charge transfer behavior during the adsorption of the target gas molecules by the liquid impregnating agent; wherein, the microscopic adsorption characteristic parameters include the adsorption energy of the target stable geometry of the adsorption system, the adsorption distance, and the charge transfer amount. That is, the adsorption energy of the stable geometry of the adsorption system is calculated based on the single-point energy value (…). E ads The specific formula for calculating the adsorption energy is as follows: ; in, E Impregnant / Gas The energy of the impregnating agent molecules after adsorbing the target gas molecules. E Impregnant To optimize the energy of the impregnating agent molecules, E Gas The adsorption energy is the individual energy of the target gas molecule. The stable geometric configuration of the target adsorption system at which the adsorption energy is minimized, i.e., when the system energy converges to a minimum, is determined. The adsorption distance D is then statistically extracted from this configuration. Target charge analysis methods, such as Mulliken (Maliken population analysis), are used to quantitatively characterize the charge transfer behavior during the adsorption of the target gas molecules, and the amount of charge transferred during the adsorption process is quantitatively calculated. Q t The specific formula for calculating the amount of charge transfer is as follows: ; in, Q Impregnant / Gas This indicates the charge value carried by the impregnating agent molecules after adsorbing the target gas molecules. Q ImpregnantThis represents the charge value carried by the impregnating agent molecules themselves. This leads to three types of microscopic adsorption characteristic parameters: adsorption energy, adsorption distance, and charge transfer amount, representing the stable geometry of the target adsorption system. In this embodiment, by using density functional theory to set multiple sets of different initial adsorption conditions to construct an adsorption system model, the gas adsorption process can be accurately simulated at the microscopic level. Dispersion correction and geometric optimization are employed to more realistically reflect weak intermolecular interactions, improving the reliability of the model structure. Single-point energy calculations based on hybrid functional theory ensure the accuracy of energy calculations, providing accurate microscopic energy data for subsequent adsorption performance analysis. Quantitative characterization of gas absorption performance using three types of microscopic characteristic parameters—adsorption energy, adsorption distance, and charge transfer—directly reflects the strength of adsorption and the mechanism of action, providing a reliable microscopic basis for evaluating the gas absorption performance of liquid impregnating agents. Selecting target adsorption systems based on minimum adsorption energy ensures that the selected configuration is closest to the actual stable adsorption state, improving the authenticity of the characteristic parameters. Quantitative characterization of charge transfer during the adsorption process using charge analysis accurately describes the microscopic nature of intermolecular interactions, making the obtained microscopic characteristic parameters more accurately reflect the gas absorption performance of the liquid impregnating agent.

[0028] Step S13: Perform normal distribution verification on the plurality of molecular descriptors. Based on the verification results and using the corresponding preset correlation statistical method, analyze the correlation between the plurality of molecular descriptors and the microscopic adsorption characteristic parameters, so as to determine the target molecular descriptor based on the obtained correlation analysis results.

[0029] In this embodiment, the normal distribution verification of the molecular descriptors is first carried out. Then, based on the verification results, a preset correlation statistical method is used to analyze the correlation between the molecular descriptors and the microscopic adsorption characteristic parameters, namely adsorption energy, charge transfer and adsorption distance. Based on the correlation results, the target molecular descriptor with the strongest correlation is selected.

[0030] It should be noted that the process for verifying the normality of the molecular descriptors and analyzing the correlation between the molecular descriptors and the microscopic adsorption characteristic parameters is as follows: The molecular descriptors are verified for normality using a target test method to obtain the corresponding verification results. If the verification results indicate that the molecular descriptors follow a normal distribution, the correlation between the molecular descriptors and the microscopic adsorption characteristic parameters is analyzed using the Pearson correlation coefficient method. If the verification results indicate that the molecular descriptors do not follow a normal distribution, the correlation between the molecular descriptors and the microscopic adsorption characteristic parameters is analyzed using the Spearman correlation coefficient method. That is, a target test method, such as the Shapiro-Wilk test suitable for small sample data, is first used to verify the normality of the molecular descriptors. This method uses a statistic W to measure the degree of deviation of the data from the normal distribution. The W value is between 0 and 1; the closer the value is to 1, the better the normality fit, and vice versa. The specific calculation formula is as follows: ; ; in, x i Characterizing the first i Sample values ​​of each molecule descriptor. a i The number after sorting in ascending order i Each order statistic The sample mean of the molecular descriptors. m i This represents the expected value of the order statistic of the standard normal distribution. After obtaining the validation results, if the characterization follows a normal distribution, the Pearson correlation coefficient method is used to analyze the correlation; if not, the Spearman correlation coefficient method is used to analyze the correlation between the molecular descriptor and the microscopic adsorption characteristic parameters. The specific formula for calculating the Pearson correlation coefficient is as follows: ; Where r is the Pearson correlation coefficient. y i Characterizing the first i Sample values ​​of microscopic adsorption characteristic parameters This represents the sample mean of the microscopic adsorption characteristic parameters. The specific formula for calculating the Spearman rank correlation coefficient is as follows: ; in, r s denoted as Spearman correlation coefficient, d represents the rank difference between the molecular descriptor and the microscopic adsorption characteristic parameter in the i-th sample, and n is the total number of samples.

[0031] It should be further noted that the processing flow for determining the target molecule descriptor based on the obtained correlation analysis results is as follows: Using a corresponding preset correlation statistical method, the correlation coefficients between the several molecule descriptors and the microscopic adsorption characteristic parameters are determined respectively; based on the absolute value of the correlation coefficients, candidate molecule descriptors that satisfy preset strong correlations with the adsorption energy, the adsorption distance, and the charge transfer amount are selected; redundant molecule descriptors are removed from the candidate molecule descriptors, and the removed candidate molecule descriptors are determined as the target molecule descriptors. That is, by using a preset correlation statistical method, the correlation coefficients between each molecule descriptor and the microscopic adsorption characteristic parameters are calculated; based on the absolute value of the correlation coefficients, candidate molecule descriptors with strong correlations with the adsorption energy, the adsorption distance, and the charge transfer amount are selected; then redundant descriptors are removed to obtain the target molecule descriptor. In this way, this embodiment performs normal distribution verification on molecular descriptors and adaptively selects either Pearson or Spearman correlation coefficient methods based on the normal distribution verification results. This adapts to different data distribution characteristics, improves the accuracy and applicability of correlation analysis, and provides a reliable basis for subsequent screening of effective molecular descriptors. By using a preset correlation statistical method to associate molecular descriptors with microscopic adsorption characteristic parameters, key descriptors strongly correlated with uptake performance can be accurately screened, providing efficient and accurate feature basis for subsequent evaluation of impregnating agent performance. Screening strongly correlated descriptors based on the absolute value of the correlation coefficient can accurately retain effective features closely related to uptake performance and eliminate redundant molecular descriptors, thereby reducing feature dimensionality, reducing information overlap, and improving the efficiency and accuracy of subsequent evaluation of liquid impregnating agent performance.

[0032] Step S14: Based on the correlation analysis results and the target molecule descriptor, a comparative analysis is performed on the candidate liquid impregnating agents for the target power capacitor, so as to select the target liquid impregnating agent with the strongest gas absorption performance from the candidate liquid impregnating agents according to the obtained comparative analysis results.

[0033] In this embodiment, based on the correlation analysis results and the target molecular descriptor, a comparative analysis is conducted on the candidate liquid impregnating agents for the target power capacitor. Specifically, based on the correlation between different molecular descriptors and microscopic physical properties such as adsorption energy, adsorption distance, and charge transfer, the numerical differences of various impregnating agent molecular descriptors are compared and analyzed. Based on the comparative analysis results, the target liquid impregnating agent with the strongest gas absorption performance is selected to determine the optimal liquid impregnating agent for the high-voltage power capacitor.

[0034] It should be noted that, while conducting comparative analysis, this embodiment can also further compare and analyze the macroscopic gas absorption results of different liquid impregnating agents through macroscopic gas absorption performance tests. The process is as follows: Macroscopic gas absorption performance tests are performed on the candidate liquid impregnating agents using a gas absorption performance tester to obtain the corresponding test results; cross-validation is then performed based on the test results and the comparative analysis results to screen out the target liquid impregnating agent with the strongest gas absorption performance from the candidate liquid impregnating agents. That is, according to the GB / T 11142 2025 standard "Determination of Gas Emission Properties of Insulating Oil under Electric Field and Ionization", a macroscopic gas absorption performance test is conducted on the candidate liquid impregnating agents using a gas absorption performance tester to obtain the test results, and then the test results are cross-validated with the microscopic comparative analysis results to screen out the target liquid impregnating agent with the strongest gas absorption performance. In this way, this embodiment uses correlation analysis results and target molecule descriptors as evaluation criteria, which can quantitatively compare the gas-getting performance of impregnating agents at the microscopic level, and reveal the adsorption energy, adsorption distance and charge transfer laws of different impregnating agents at the molecular level. Through quantitative comparison and screening of multiple candidate materials, the optimal liquid impregnating agent with the best gas-getting performance can be determined objectively and efficiently, improving the scientificity and accuracy of material selection. By combining macroscopic experimental tests and microscopic analysis results for cross-validation, both actual performance and molecular interaction mechanisms can be taken into account, improving the reliability and persuasiveness of the material selection results, and ensuring that the screened impregnating agents meet both theoretical excellence and practical usability.

[0035] As can be seen from the above, the embodiments of this application first construct a set of feature quantities containing molecular descriptors characterizing the gas absorption performance of liquid impregnating agents, and then construct an adsorption system model of the liquid impregnating agent adsorbing target gas molecules through density functional theory to obtain microscopic adsorption feature parameters including adsorption energy, adsorption distance, and charge transfer amount. Subsequently, the normal distribution verification of the several molecular descriptors is performed, and the correlation between them and the microscopic adsorption feature parameters is analyzed by a preset correlation statistical method to determine the target molecular descriptor. Finally, based on the correlation results and the target molecular descriptor, the candidate liquid impregnating agents for the target power capacitor are compared and analyzed to screen out the target liquid impregnating agent with the best gas absorption performance. In this way, through the above-described process of the embodiments of this application, an adsorption system model is constructed using density functional theory to obtain microscopic adsorption characteristic parameters, which can accurately characterize the gas-getter mechanism of the liquid impregnating agent at the microscopic level, providing a theoretical basis for performance evaluation; by screening target molecular descriptors through normal distribution verification and correlation statistics methods, key features strongly correlated with gas-getter performance can be effectively extracted, improving the accuracy and efficiency of subsequent evaluation; based on the screened molecular descriptors, quantitative comparative analysis of the impregnating agents to be selected is performed, revealing the adsorption energy, adsorption distance, and charge transfer laws of different impregnating agents at the molecular level, which can intuitively distinguish the gas-getter performance of different materials, achieving efficient and accurate optimal material screening, and thus realizing efficient evaluation of the gas-getter performance of liquid impregnating agents to improve the efficiency of selecting liquid impregnating agents with strong gas-getter performance.

[0036] As can be seen from the previous embodiment, this application discloses a method for selecting liquid impregnating agents suitable for power capacitors, which can achieve efficient evaluation of the gas-absorbing performance of liquid impregnating agents to improve the efficiency of selecting liquid impregnating agents with strong gas-absorbing performance. Next, the method for selecting liquid impregnating agents suitable for power capacitors will be described in detail.

[0037] First, a set of key molecular descriptor features characterizing the uptake performance of the impregnating agent is constructed. The Hohenberg-Kohn theorem states that for a quantum system with a definite number of particles, the minimum value of its energy functional corresponds to the ground state energy of the system. Adsorption energy quantitatively describes the binding strength between gas molecules and impregnating agent molecules; the larger the absolute value, the more easily the gas molecules are stably adsorbed. Adsorption distance reflects the spatial proximity between the gas and the impregnating agent; the shorter the distance, the stronger the van der Waals forces, π-π stacking, and charge transfer interactions, resulting in higher adsorption stability. Charge transfer quantifies the redistribution of electron clouds between molecules; the greater the amount of charge transfer, the stronger the electron donor-acceptor interaction, the easier it is for gas molecules to be chemically fixed, thereby increasing the uptake rate. Eleven key molecular descriptors affecting the getter performance of impregnating agents were selected, as shown in Table 1. These are: number of aromatic bonds (nAromBond), number of double bonds (nBondsD2), number of carbon atoms (nC), molecular weight (MW), Kier third-order shape index (Kier3), Petitjean shape index (Petitjean), number of oxygen atoms (nO), number of ester groups (nCOO), Wiener polarity number (WPOL), topological polarity charge correlation descriptor (R_TpiPCTPC), and 7th-order atom type center topological symmetry descriptor (AATS7p). Based on these, a set of key molecular descriptor feature quantities characterizing the getter performance of impregnating agents was constructed.

[0038] Table 1 Summary of Molecular Descriptors

[0039] Then, the numerical values ​​of the microscopic characteristic parameters characterizing the gas-absorbing properties of the impregnating agent molecules are calculated. This study focuses on 30 common insulating oil molecules, including monoesters, diesters, triesters, tetraesters, olefins, and benzyltoluene. Examples include isooctyl oleate (OE), isooctyl linoleate (LOE), isooctyl stearate (SE), isooctyl trans-4-methoxycinnamate (OMCE), isobutyl laurate (LBE), isooctyl palmitate (PE), methyl oleate (MO), methyl linoleate (ML), methyl linolenate (MLn), methyl tung oil (MELO), ethyl oleate (EO), ethyl linoleate (EL), ethyl linolenate (ELn), isononyl isononanoate (ININ), neopentyl glycol dioleate (NGDO), didecyl phthalate (DTDP), di-n-octyl adipate (DNOA), diisononyl adipate (DINA), diisooctyl adipate (DIOA), diisooctyl azelate (DIOZ), diisooctyl maleate (DIOF), diisooctyl sebacate (DIOS), and 1,2,4-diethyl oleate (DIOS). The following 30 impregnating agents were selected for their H2 molecule adsorption: trinonyl tricresyl ester (TNM), trimethylolpropane trioleate (TMPTO), methyl tung oil diethyl maleate (TOMEC), pentaerythritol tetranonyl ester (M-PETC), pentaerythritol tetraoleate (PETO), triacontane hexaene (SQC), monobenzyltoluene (MBT), and dibenzyltoluene (DBT). Adsorption system models for H2 molecules were constructed using density functional theory. The models were then analyzed using DMol in Material Studio software. 3 The module constructs an adsorption model of impregnating agent molecules and hydrogen. The adsorption system is geometrically optimized using a generalized gradient approximation GGA-PBE functional combined with DFT-D2 dispersion correction, and its single-point energy is calculated using a high-precision M062X hybrid functional. After the system energy converges to a minimum, the adsorption distance of the impregnating agent molecules to hydrogen is statistically recorded. D To screen for the lowest energy adsorption configuration, hydrogen molecules were placed on the impregnating agent molecule model at different initial distances and interaction sites, and the adsorption energy was calculated. E ads The charge transfer behavior during H2 adsorption was quantitatively characterized using the Mulliken charge analysis method, and the amount of charge transferred was calculated. Q t (as shown in Table 2).

[0040] Table 2 Calculation results of adsorption energy, adsorption distance and charge transfer of impregnating agent molecules

[0041] Subsequently, molecular descriptors closely related to the microscopic adsorption properties of the impregnating agent molecules were selected. The Shapiro-Wilk test, suitable for small sample data, was used to verify the normality of all molecular descriptors. This method uses the statistic W to measure the degree of deviation of the data from a normal distribution. The W value ranges from 0 to 1; the closer the value is to 1, the better the normality fit, and vice versa. For descriptors that follow a normal distribution, the Pearson correlation coefficient was used to analyze their correlation with the dependent variable; for descriptors that do not meet the normal distribution, the Spearman rank correlation coefficient was used. Table 3 shows the calculated correlation coefficients. Figure 3 shows a schematic diagram of the correlation coefficients between a microscopic adsorption characteristic parameter and a molecular descriptor provided in this application. Specifically, Figure 3(a) shows the correlation coefficient between adsorption energy and a molecular descriptor; Figure 3(b) shows the correlation coefficient between adsorption distance and a molecular descriptor; and Figure 3(c) shows the correlation coefficient between charge transfer and a molecular descriptor. Analysis shows that R_TpiPCTPC, AATS7p, and nAromBond are significantly negatively correlated with adsorption energy, while nO and nCOOH are positively correlated. Among these, R_TpiPCTPC and nAromBond, and nO and nCOOH, are highly redundant; therefore, R_TpiPCTPC, AATS7p, and nO are selected as key molecular descriptors for adsorption energy. R_TpiPCTPC shows a significant negative correlation with adsorption distance; therefore, it is selected as a key molecular descriptor for adsorption distance. nO and R_TpiPCTPC are positively and negatively correlated with charge transfer values, respectively; therefore, they are selected as key descriptors describing the strength of charge transfer.

[0042] Table 3. Correlation coefficients between adsorption energy, adsorption distance, charge transfer, and molecular descriptor

[0043] Finally, molecular descriptors were used to conduct a comparative analysis of the getter performance of different types of liquid impregnating agents. Based on the correlation between different molecular descriptors and microscopic physical parameters such as adsorption energy, adsorption distance, and charge transfer, the numerical differences of molecular descriptors of various impregnating agents were compared and analyzed to efficiently and accurately screen the types of impregnating agents with excellent getter performance. Taking the getter performance of isooctyl linoleate (LOE) and pentaerythritol tetranonanoate (M-PETC) as examples, the values ​​of three key molecular descriptors, R_TpiPCTPC, AATS7p, and nO, as well as the corresponding microscopic physical parameters such as adsorption energy, adsorption distance, and charge transfer, are listed for LOE and M-PETC respectively. At the same time, according to the GB / T 11142-2025 standard "Determination of Gas Emission Property of Insulating Oil under Electric Field and Ionization", the macroscopic getter performance of LOE and M-PETC was tested using an getter performance tester, and the results are shown in Table 4.

[0044] Table 4. Molecular descriptors and macroscopic updraft values ​​for LOE and M-PETC

[0045] The AATS7p values ​​of the molecular descriptors LOE and M-PETC show little difference, having a relatively small impact on adsorption energy. R_TpiPCTPC and nO show significant differences, with correlation coefficients of -0.665 and 0.484, respectively, indicating a negative and positive correlation. Furthermore, LOE's R_TpiPCTPC value is greater than M-PETC's, while its nO value is less than M-PETC's; therefore, LOE's adsorption energy is lower than M-PETC's. R_TpiPCTPC is a key molecular descriptor characterizing adsorption distance, showing a negative correlation with it. LOE's R_TpiPCTPC value is greater than M-PETC's; therefore, LOE's adsorption distance should be less than M-PETC's. Similarly, nO and R_TpiPCTPC are key descriptors describing charge transfer strength, showing a positive and negative correlation with charge transfer, respectively; therefore, the absolute value of LOE's charge transfer should be greater than M-PETC's. The data in the table show that the trends in adsorption energy, adsorption distance, and charge transfer characterized by the molecular descriptor are consistent with theoretical calculations; the getter properties of impregnating agent molecules are affected by adsorption energy, adsorption distance, and charge transfer. Generally speaking, the larger the absolute values ​​of adsorption energy and charge transfer, and the shorter the adsorption distance, the better the macroscopic getter properties of liquid impregnating agent molecules. LOE has higher adsorption energy and charge transfer absolute values ​​than M-PETC, and its adsorption distance is shorter than that of M-PETC. Therefore, theoretically, the getter properties of LOE should be less than those of M-PETC, and LOE has superior getter properties. After conducting macroscopic getter property tests on LOE and M-PETC, it was found that the getter property value of LOE is significantly smaller than that of M-PETC, which is consistent with the evaluation results using the method of this application.

[0046] Accordingly, see Figure 4 As shown in the illustration, this application also provides a liquid impregnating agent selection device suitable for power capacitors, comprising: The set construction module 11 is used to construct a set of target molecule descriptor features; the set of target molecule descriptor features includes several molecular descriptors characterizing the gas absorption performance of the liquid impregnating agent; The parameter determination module 12 is used to construct an adsorption system model of the liquid impregnating agent adsorbing target gas molecules using density functional theory, and to determine the corresponding microscopic adsorption characteristic parameters through the adsorption system model; the microscopic adsorption characteristic parameters include adsorption energy, adsorption distance and charge transfer amount, which characterize the gas absorption performance of the liquid impregnating agent. The correlation analysis module 13 is used to perform normal distribution verification on the plurality of molecular descriptors, and analyze the correlation between the plurality of molecular descriptors and the microscopic adsorption characteristic parameters based on the obtained verification results and using the corresponding preset correlation statistical methods, so as to determine the target molecular descriptor based on the obtained correlation analysis results. The comparative analysis module 14 is used to perform comparative analysis on the candidate liquid impregnating agents for the target power capacitor based on the correlation analysis results and the target molecule descriptor, so as to screen out the target liquid impregnating agent with the strongest gas absorption performance from the candidate liquid impregnating agents according to the obtained comparative analysis results.

[0047] In some specific embodiments, the collection construction module 11 may specifically include: The first descriptor screening unit is used to screen out several molecular descriptors characterizing the gas absorption performance of the liquid impregnating agent from the molecular feature dimension, based on the principle of minimum energy, the law of charge conservation, the density functional theory, and the electronic structure theory. The descriptor classification unit is used to classify the plurality of molecular descriptors according to the molecular feature dimensions, so as to construct a target molecular descriptor feature set including the plurality of molecular descriptors.

[0048] In some specific embodiments, the parameter determination module 12 may specifically include: The model building unit is used to construct an impregnating agent molecular model and a target gas molecule model using density functional theory and the target quantum chemistry module of the target molecule simulation software, so as to construct an initial adsorption system model when the liquid impregnating agent adsorbs the target gas molecule; the initial adsorption system model includes multiple adsorption system models with different initial conditions; the initial conditions include the initial distance and action site of the target gas molecule placed in the liquid impregnating agent; The geometry optimization unit is used to perform geometry optimization on the initial adsorption system model, the impregnating agent molecule model, and the gas molecule model using the target density functional and the target dispersion correction method, respectively, to obtain the target stable geometry configuration; the target stable geometry configuration includes the stable geometry configuration of the adsorption system, the stable geometry configuration of the impregnating agent molecule, and the stable geometry configuration of the gas molecule. A single-point energy calculation unit is used to perform single-point energy calculation on the target stable geometry based on the target hybrid functional to obtain the corresponding single-point energy value.

[0049] In some specific embodiments, the parameter determination module 12 may specifically include: An adsorption energy determination unit is used to determine the adsorption energy of the stable geometry of the adsorption system based on the single-point energy value, so as to determine the target stable geometry of the adsorption system with the minimum adsorption energy. The distance extraction unit is used to extract the adsorption distance of the liquid impregnating agent on the target gas molecules from the stable geometry of the target adsorption system. The quantitative characterization unit is used to quantitatively characterize the charge transfer behavior of the stable geometry of the target adsorption system using the target charge analysis method, so as to calculate the corresponding charge transfer amount; the charge transfer behavior is the charge transfer behavior during the adsorption of the target gas molecules by the liquid impregnating agent; The microscopic adsorption characteristic parameters include the adsorption energy of the stable geometry of the target adsorption system, the adsorption distance, and the charge transfer amount.

[0050] In some specific embodiments, the correlation analysis module 13 may specifically include: The descriptor verification unit is used to perform normal distribution verification on the plurality of molecular descriptors using the target verification method to obtain the corresponding verification results; The first correlation analysis unit is used to analyze the correlation between the molecular descriptor and the microscopic adsorption characteristic parameter using the Pearson correlation coefficient method if the verification result indicates that the molecular descriptor follows a normal distribution. The second correlation analysis unit is used to analyze the correlation between the molecular descriptor and the microscopic adsorption characteristic parameter using the Spearman correlation coefficient method if the verification result indicates that the molecular descriptor does not follow a normal distribution.

[0051] In some specific embodiments, the correlation analysis module 13 may specifically include: The coefficient determination unit is used to determine the correlation coefficients between the plurality of molecular descriptors and the microscopic adsorption characteristic parameters using corresponding preset correlation statistical methods. The second descriptor filtering unit is used to filter out candidate molecule descriptors that satisfy a preset strong correlation with the adsorption energy, the adsorption distance and the charge transfer amount, respectively, based on the absolute value of the correlation coefficient. The descriptor determination unit is used to remove redundant molecular descriptors from the candidate molecular descriptors and determine the removed candidate molecular descriptors as target molecular descriptors.

[0052] In some specific embodiments, the comparison analysis module 14 may further include: The experimental testing unit is used to conduct macroscopic air absorption performance tests on the liquid impregnating agent to be selected using an air absorption tester, so as to obtain the corresponding experimental test results. The cross-validation unit is used to perform cross-validation based on the experimental test results and the comparative analysis results, so as to screen out the target liquid impregnating agent with the strongest air absorption performance from the candidate liquid impregnating agents.

[0053] Furthermore, embodiments of this application also disclose an electronic device, Figure 5 This is a structural diagram of an electronic device 20 according to an exemplary embodiment. The content of the diagram should not be construed as limiting the scope of this application. The electronic device 20 may specifically include: at least one processor 21, at least one memory 22, a power supply 23, a communication interface 24, an input / output interface 25, and a communication bus 26. The memory 22 stores a computer program, which is loaded and executed by the processor 21 to implement the relevant steps in the liquid impregnating agent selection method for power capacitors disclosed in any of the foregoing embodiments. Alternatively, the electronic device 20 in this embodiment may specifically be an electronic computer.

[0054] In this embodiment, the power supply 23 is used to provide operating voltage for each hardware device on the electronic device 20; the communication interface 24 can create a data transmission channel between the electronic device 20 and external devices, and the communication protocol it follows can be any communication protocol applicable to the technical solution of this application, and is not specifically limited here; the input / output interface 25 is used to acquire external input data or output data to the outside world, and its specific interface type can be selected according to specific application needs, and is not specifically limited here.

[0055] In addition, the memory 22, as a carrier for resource storage, can be a read-only memory, random access memory, disk or optical disk, etc. The resources stored thereon can include operating system 221, computer program 222, etc., and the storage method can be temporary storage or permanent storage.

[0056] The operating system 221 is used to manage and control the various hardware devices on the electronic device 20 and the computer program 222, which may be Windows Server, Netware, Unix, Linux, etc. In addition to including a computer program capable of performing the liquid impregnating agent selection method for power capacitors disclosed in any of the foregoing embodiments, the computer program 222 may further include computer programs capable of performing other specific tasks.

[0057] Furthermore, this application also discloses a computer-readable storage medium for storing a computer program; wherein, when the computer program is executed by a processor, it implements the aforementioned method for selecting a liquid impregnating agent suitable for power capacitors. Specific steps of this method can be found in the corresponding content disclosed in the foregoing embodiments, and will not be repeated here.

[0058] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the apparatus disclosed in the embodiments, since it corresponds to the method disclosed in the embodiments, the description is relatively simple; relevant parts can be referred to in the method section.

[0059] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0060] The steps of the methods or algorithms described in conjunction with the embodiments disclosed herein can be implemented directly by hardware, a software module executed by a processor, or a combination of both. The software module can be located in random access memory (RAM), main memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium known in the art.

[0061] Finally, it should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0062] The technical solutions provided in this application have been described in detail above. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this application. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A method for selecting a liquid impregnating agent suitable for power capacitors, characterized in that, include: Construct a set of target molecule descriptor features; the set of target molecule descriptor features includes several molecular descriptors characterizing the gas-getter properties of the liquid impregnating agent; Using density functional theory, an adsorption system model is constructed for the adsorption of target gas molecules by liquid impregnating agent, and the corresponding microscopic adsorption characteristic parameters are determined by the adsorption system model. The microscopic adsorption characteristic parameters include adsorption energy, adsorption distance, and charge transfer amount, which characterize the gas absorption performance of the liquid impregnating agent. The normal distribution verification of the molecular descriptors is performed. Based on the verification results and using the corresponding preset correlation statistical method, the correlation between the molecular descriptors and the microscopic adsorption characteristic parameters is analyzed, so as to determine the target molecular descriptor based on the obtained correlation analysis results. Based on the correlation analysis results and the target molecule descriptor, a comparative analysis is performed on the candidate liquid impregnating agents for the target power capacitor, so as to select the target liquid impregnating agent with the strongest gas absorption performance from the candidate liquid impregnating agents according to the obtained comparative analysis results.

2. The method for selecting a liquid impregnating agent for power capacitors according to claim 1, characterized in that, The construction of the target molecule descriptor feature set includes: Based on the principle of minimum energy, the law of charge conservation, density functional theory, and electronic structure theory, several molecular descriptors characterizing the gas absorption performance of liquid impregnating agents are screened from the molecular feature dimension. The molecular descriptors are classified according to the molecular feature dimensions to construct a target molecular descriptor feature set including the molecular descriptors.

3. The method for selecting a liquid impregnating agent for power capacitors according to claim 1, characterized in that, The adsorption system model for the adsorption of target gas molecules by the liquid impregnating agent, constructed using density functional theory, includes: Using density functional theory and the target quantum chemistry module of the target molecule simulation software, an impregnating agent molecular model and a target gas molecule gas molecule model are constructed to build an initial adsorption system model when the liquid impregnating agent adsorbs the target gas molecule. The initial adsorption system model includes multiple adsorption system models with different initial conditions. The initial conditions include the initial distance and action site of the target gas molecule placed in the liquid impregnating agent. Using target density functional theory and target dispersion correction methods, the initial adsorption system model, the impregnating agent molecule model, and the gas molecule model are geometrically optimized to obtain target stable geometric configurations; the target stable geometric configurations include the stable geometric configurations of the adsorption system, the stable geometric configurations of the impregnating agent molecules, and the stable geometric configurations of the gas molecules; The single-point energy of the target stable geometry is calculated based on the target hybrid functional to obtain the corresponding single-point energy value.

4. The method for selecting a liquid impregnating agent suitable for power capacitors according to claim 3, characterized in that, The determination of the corresponding microscopic adsorption characteristic parameters through the adsorption system model includes: Based on the single-point energy value, the adsorption energy of the stable geometry of the adsorption system is determined, so as to determine the target stable geometry of the adsorption system with the minimum adsorption energy. The adsorption distance of the liquid impregnating agent for the target gas molecules is extracted from the stable geometry of the target adsorption system. The charge transfer behavior of the stable geometry of the target adsorption system is quantitatively characterized using target charge analysis to calculate the corresponding charge transfer amount; the charge transfer behavior refers to the charge transfer behavior during the adsorption of the target gas molecules by the liquid impregnating agent. The microscopic adsorption characteristic parameters include the adsorption energy of the stable geometry of the target adsorption system, the adsorption distance, and the charge transfer amount.

5. The method for selecting a liquid impregnating agent for power capacitors according to claim 1, characterized in that, The step of performing normal distribution verification on the plurality of molecular descriptors, and analyzing the correlation between the plurality of molecular descriptors and the microscopic adsorption characteristic parameters based on the obtained verification results and using a corresponding preset correlation statistical method, includes: The target test method was used to perform normal distribution verification on the aforementioned molecular descriptors to obtain the corresponding verification results; If the verification results indicate that the molecular descriptor follows a normal distribution, then the correlation between the molecular descriptor and the microscopic adsorption characteristic parameter is analyzed using the Pearson correlation coefficient method. If the verification results indicate that the molecular descriptor does not follow a normal distribution, then the Spearman correlation coefficient method is used to analyze the correlation between the molecular descriptor and the microscopic adsorption characteristic parameters.

6. The method for selecting a liquid impregnating agent for power capacitors according to claim 1, characterized in that, The step of analyzing the correlation between the plurality of molecular descriptors and the microscopic adsorption characteristic parameters using a corresponding preset correlation statistical method, and determining the target molecular descriptor based on the obtained correlation analysis results, includes: Using corresponding preset correlation statistical methods, the correlation coefficients between the molecular descriptors and the microscopic adsorption characteristic parameters are determined respectively; Based on the absolute value of the correlation coefficient, candidate molecule descriptors that satisfy a preset strong correlation with the adsorption energy, the adsorption distance, and the charge transfer amount are selected. Redundant molecular descriptors are removed from the candidate molecular descriptors, and the removed candidate molecular descriptors are determined as the target molecular descriptors.

7. The method for selecting a liquid impregnating agent for power capacitors according to claim 1, characterized in that, The step of selecting the target liquid impregnating agent with the strongest air-getter performance from the candidate liquid impregnating agents based on the obtained comparative analysis results also includes: The macroscopic air-absorbing performance of the candidate liquid impregnating agent was tested using an air-absorbing tester to obtain the corresponding experimental results. Cross-validation was performed based on the experimental test results and the comparative analysis results to screen out the target liquid impregnating agent with the strongest air absorption performance from the candidate liquid impregnating agents.

8. A liquid impregnating agent selection device suitable for power capacitors, characterized in that, include: A set construction module is used to construct a set of target molecule descriptor features; the set of target molecule descriptor features includes several molecular descriptors characterizing the uptake performance of the liquid impregnating agent; The parameter determination module is used to construct an adsorption system model of the liquid impregnating agent adsorbing target gas molecules using density functional theory, and to determine the corresponding microscopic adsorption characteristic parameters through the adsorption system model; the microscopic adsorption characteristic parameters include adsorption energy, adsorption distance and charge transfer amount, which characterize the gas absorption performance of the liquid impregnating agent. The correlation analysis module is used to perform normal distribution verification on the plurality of molecular descriptors, and analyze the correlation between the plurality of molecular descriptors and the microscopic adsorption characteristic parameters based on the obtained verification results and the corresponding preset correlation statistical methods, so as to determine the target molecular descriptor based on the obtained correlation analysis results. The comparative analysis module is used to perform comparative analysis on the candidate liquid impregnating agents for the target power capacitor based on the correlation analysis results and the target molecule descriptor, so as to screen out the target liquid impregnating agent with the strongest gas absorption performance from the candidate liquid impregnating agents according to the obtained comparative analysis results.

9. An electronic device, characterized in that, include: Memory, used to store computer programs; A processor for executing the computer program to implement the liquid impregnating agent selection method for power capacitors as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, Used to store a computer program; wherein, when the computer program is executed by a processor, it implements the liquid impregnating agent selection method for power capacitors as described in any one of claims 1 to 7.

Images