Insulating oil molecule lightning impulse performance evaluation and optimization method based on first principle

Abstract

The invention relates to an insulating oil molecule lightning impulse performance evaluation and optimization method based on a first principle, and belongs to the technical field of insulating material performance evaluation. The method comprises the following steps: S1, finely constructing a molecular model based on quantum chemistry calculation; s2, performing dispersion correction and geometric optimization on the insulating oil molecular model; s3, calculating numerical values of the performance evaluation index parameters under the action of different electric fields; and S4, comparing and analyzing the lightning impulse performance of different types of insulating oil molecules by adopting a parallel coordinate graph. The invention creatively provides a micro-scale evaluation and structure optimization method aiming at the lightning impulse performance of the insulating oil, and is expected to break through the limitation of experimental period and cost.

Description

Technical Field

[0001] This invention relates to the field of insulation material performance evaluation technology, and more specifically, to a first-principles-based method for evaluating and optimizing the lightning impact performance of insulating oil molecules. Background Technology

[0002] Oil-immersed transformers are core components of power systems, and their reliability is crucial for the safe and stable operation of the entire system. High-performance insulating oil is a fundamental prerequisite for ensuring the reliable and safe operation of oil-immersed transformers. Traditional mineral oil transformers have advantages such as excellent insulation performance and good economy, but they also have drawbacks such as poor thermal stability, susceptibility to aging, and a low ignition / explosion temperature. With the global shift towards green power grids, traditional mineral insulating oils can no longer meet the development needs of the power grid. In recent years, ester-based insulating oils have become a research hotspot for finding good alternatives to mineral oil as insulating liquids, mainly including natural ester insulating oils and modified ester insulating oils. Compared with mineral oil, natural esters have strong flame retardancy, high thermal stability, and environmentally friendly characteristics such as biodegradability and recyclability. They also exhibit excellent power frequency / DC breakdown characteristics and can delay the aging of insulating paperboard.

[0003] Transformers are highly susceptible to lightning impulse voltages during operation, especially large-capacity, ultra-high-voltage transformers. In the research and application of ester-based insulating oils, their lightning impulse insulation performance is a crucial factor ensuring safe operation in transformers. The differences in lightning breakdown voltage among various natural esters mainly stem from differences in molecular structure and intermolecular interaction energy. Currently, most methods for evaluating the lightning impulse performance of insulating oils rely on traditional testing methods, which are not only costly and time-consuming but also only characterize the test results, failing to analyze performance changes at the microscopic level or the mechanisms underlying performance differences among different insulating oils. Therefore, a highly efficient and accurate method for evaluating the lightning impulse performance of insulating oils and selecting the best-performing insulating oil is lacking.

[0004] This invention aims to propose a first-principles-based method for evaluating and optimizing the lightning impulse performance of insulating oil molecules, specifically addressing the impact of variations in the internal structure and composition of different oils on micro-parameters related to lightning voltage. The method involves meticulously constructing a molecular model of the insulating oil, correcting for electron distribution in the liquid environment, and then using CP2K software for geometric optimization and energy calculation. Finally, the obtained performance evaluation indicators are classified and standardized, parallel coordinate graphs are plotted, and the area under the curve is calculated to achieve optimal selection. This invention is expected to overcome the limitations of traditional methods in terms of experimental cycle and cost, establishing a link between the microstructure and macroscopic properties of insulating materials. Summary of the Invention

[0005] In view of this, the purpose of this method is to propose a first-principles-based method for evaluating and optimizing the lightning impact performance of insulating oil molecules, so as to solve the above-mentioned technical problems.

[0006] To achieve the above objectives, the present invention adopts the following technical solution:

[0007] S1. Fine-scale construction of molecular models based on quantum chemical calculations

[0008] S11: An initial three-dimensional structural model of the insulating oil molecule was constructed using GaussView software. Then, based on a known crystal structure database, the bond lengths, bond angles, and dihedral angles of the molecule were precisely set. To ensure the accuracy of the simulation results, a sufficient vacuum region was reserved around the model to avoid the influence of periodic boundaries.

[0009] S12: Based on the molecular model constructed in S11, an implicit solvation model (IEF-PCM) is used in Gaussian software. The dielectric constant of the solvent is set to a fixed value in the range of 2.0 to 2.5. The single-point energy of the initial molecular model is calculated to simulate its initial actual electron distribution in this liquid environment.

[0010] S2. Dispersion Correction and Geometric Optimization of Insulating Oil Molecular Model

[0011] S21: Import the insulating oil molecular structure corrected for liquid environment electron distribution from S12 into Multiwfn software and create a CP2K input file. Configure key calculation parameters in the generated input file, and specify the use of the PBE exchange correlation functional in the DFT section.

[0012] S22: In the density functional theory (DFT) calculation module, a van der Waals interaction correction term is configured. The type of this correction term is set to DFT-D3(BJ), and the parameter file dftd3.dat is called to implement third-generation dispersion correction.

[0013] S23: Based on the input file of S12, the DZVP-MOLOPT-SR-GTH basis set is selected in CP2K software to perform geometry optimization on the molecule, and a specified sub-item is created for each element. The BASIS_SET parameter is set to the corresponding basis set file, the optimization step number is set to 500, the force convergence threshold is 4.5e-4Hartree / Bohr, and the energy convergence threshold is 1.0e-6Hartree.

[0014] S3. Calculate the numerical values ​​of performance evaluation parameters under different electric fields.

[0015] S31: Based on the stable structure after geometric optimization of S23, while keeping the functional and dispersion correction settings unchanged, the basis set is upgraded to TZVP-MOLOPT-SR-GTH for high-precision single-point energy calculation.

[0016] S32: Different self-consistent field (SCF) convergence algorithms are used for different computational tasks. For conventional single-point energy calculations, the orbital transformation (OT) method is used, with SCF_GUESS set to RESTART to activate &OT; when calculating the density of states and molecular surface electrostatic potential, &Diagonalization needs to be activated to switch to the diagonalization method for electronic structure calculations.

[0017] S33: Based on the prerequisite settings of S32, apply 0 to 0.0004 atomic units (1 a.u. = 5.1423 × 10⁻⁴) in the X direction of the insulating oil model. 11 The electric field within the range of V / m is calculated according to formulas (1)(2)(3)(4) to obtain the lightning impact performance evaluation indexes of insulating oil molecules under different electric fields: ionization energy (IP), electron affinity (EA), insulating fluid softness (Softness), excitation energy (E). ex ).

[0018] IP = E(N) + )-E(N) (1)

[0019] EA = E(N) - E(N) - (2)

[0020]

[0021] E ex =E S1 -E S0 (4)

[0022] S4. Comparative analysis of the molecular lightning impact performance of different types of insulating oils using parallel coordinate graphs.

[0023] S41: Based on the four performance evaluation index parameters obtained in S33, they are classified into benefit-type indicators and cost-type indicators, and standardized according to formulas (5) and (6) respectively, so that all parameter values ​​are unified to the [0,1] interval and are positively correlated with lightning impact performance.

[0024]

[0025] S42: Based on the standardized performance evaluation parameters obtained from S41, a parallel coordinate graph is plotted, and the six standardized data points (V1-V6) of each molecule are connected in sequence to form a broken line. The area under the curve of each broken line in the six-dimensional performance space is calculated by the Gaussian area formula (7) to comprehensively evaluate the lightning impact performance and differences of different types of insulating oil molecules.

[0026]

[0027] S43: Based on the area enclosed by the parallel coordinate diagrams of different types of molecules calculated in S42, all evaluated molecules are sorted in descending order. The higher the ranking, the better the molecule's overall performance in multiple key performance indicators and the better its lightning impulse insulation performance, thus achieving the optimal selection of molecules.

[0028] Compared with the prior art, the present invention has the following beneficial effects:

[0029] This invention, based on first-principles calculations, employs molecular simulation technology to perform microscopic-level computational analysis of the lightning impact performance of different types of insulating oils. Compared with traditional experimental methods, it eliminates the need for large quantities of experimental materials and complex equipment, saving both experimental time and cost. Furthermore, it reveals the microscopic mechanisms underlying the adaptation of lightning impact performance to molecular structure. In addition, this method can accurately evaluate material performance and, by comparing the performance differences between different materials, more efficiently select materials with superior properties. Attached Figure Description

[0030] Figure 1 This is an overall flowchart of the present invention;

[0031] Figure 2 The ionization potential, electron affinity, and softness of SO and PFAE under different electric field strengths are given.

[0032] Figure 3 The excitation energies of SO and PFAE under different electric field strengths;

[0033] Figure 4 The waveforms of the negative breakdown voltage and current for SO and PFAE;

[0034] Figure 5 The average breakdown voltage of SO and PFAE is based on 10 measurements;

[0035] Figure 6 Parallel coordinate graphs of the lightning impulse insulation performance of different insulating oil molecules; (a) is 0a.u.; (b) is 0.002au; (c) is 0.004au. Detailed Implementation

[0036] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments. 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.

[0037] Example

[0038] like Figure 1As shown in the figure, this embodiment proposes a first-principles-based method for evaluating and optimizing the lightning impulse performance of insulating oil molecules. The specific implementation steps are as follows:

[0039] S1. Fine-scale construction of molecular models based on quantum chemical calculations

[0040] S11: An initial three-dimensional structural model of the insulating oil molecule was constructed using GaussView software. Then, based on a known crystal structure database, the bond lengths, bond angles, and dihedral angles of the molecule were precisely set. To ensure the accuracy of the simulation results, a sufficient vacuum region was reserved around the model to avoid the influence of periodic boundaries.

[0041] S12: Based on the molecular model constructed in S11, an implicit solvation model (IEF-PCM) is used in Gaussian software. The dielectric constant of the solvent is set to a fixed value in the range of 2.0 to 2.5. The single-point energy of the initial molecular model is calculated to simulate its initial actual electron distribution in this liquid environment.

[0042] S2. Dispersion Correction and Geometric Optimization of Insulating Oil Molecular Model

[0043] S21: Import the insulating oil molecular structure corrected for liquid environment electron distribution from S12 into Multiwfn software to create a CP2K input file. Configure key calculation parameters in the generated input file. In the density functional theory (DFT) section, the PBE exchange-correlation functional is used.

[0044] S22: In the density functional theory (DFT) calculation module, a van der Waals interaction correction term is configured. The type of this correction term is set to DFT-D3(BJ), and the parameter file dftd3.dat is called to implement third-generation dispersion correction.

[0045] S23: Based on the input file of S12, the DZVP-MOLOPT-SR-GTH basis set is selected in CP2K software to perform geometry optimization on the molecule, and a specified sub-item is created for each element. Then, the BASIS_SET parameter is set to the corresponding basis set file, the optimization step number is set to 500, the force convergence threshold is 4.5e-4Hartree / Bohr, and the energy convergence threshold is 1.0e-6Hartree.

[0046] S3. Calculate the numerical values ​​of performance evaluation parameters under different electric fields.

[0047] S31: Based on the stable structure after geometric optimization of S23, while keeping the functional and dispersion correction settings unchanged, the basis set is upgraded to TZVP-MOLOPT-SR-GTH for high-precision single-point energy calculation.

[0048] S32: Different self-consistent field (SCF) convergence algorithms are used for different computational tasks. For conventional single-point energy calculations, the orbital transformation (OT) method is used, with SCF_GUESS set to RESTART to activate &OT; when calculating the density of states and molecular surface electrostatic potential, &Diagonalization needs to be activated to switch to the diagonalization method for electronic structure calculations.

[0049] S33: Based on the prerequisite settings of S32, apply 0 to 0.0004 atomic units (1 a.u. = 5.1423 × 10⁻⁴) in the X direction of the insulating oil model. 11 The electric field within the range of V / m is calculated according to formulas (1)(2)(3)(4) to obtain the lightning impact performance evaluation indexes of insulating oil molecules under different electric fields: ionization energy (IP), electron affinity (EA), insulating fluid softness (Softness), excitation energy (E). ex ).

[0050] IP = E(N) + )-E(N) (1)

[0051] EA = E(N) - E(N) - (2)

[0052]

[0053] E ex =E S1 -E S0 (4)

[0054] S4. Comparative analysis of the molecular lightning impact performance of different types of insulating oils using parallel coordinate graphs.

[0055] S41: Based on the four performance evaluation index parameters obtained in S33, they are classified into benefit-type indicators and cost-type indicators, and standardized according to formulas (5) and (6) respectively, so that all parameter values ​​are unified to the [0,1] interval and are positively correlated with lightning impact performance.

[0056]

[0057] S42: Based on the standardized performance evaluation parameters obtained from S41, a parallel coordinate graph is plotted, and the six standardized data points (V1-V6) of each molecule are connected in sequence to form a broken line. The area under the curve of each broken line in the six-dimensional performance space is calculated by the Gaussian area formula (7) to comprehensively evaluate the lightning impact performance and differences of different types of insulating oil molecules.

[0058]

[0059] S43: Based on the area enclosed by the parallel coordinate diagrams of different types of molecules calculated in S42, all evaluated molecules are sorted in descending order. The higher the ranking, the better the molecule's overall performance in multiple key performance indicators and the better its lightning impulse insulation performance, thus achieving the optimal selection of molecules.

[0060] To verify the proposed first-principles method for evaluating and optimizing the lightning impulse performance of insulating oil molecules, the lightning impulse voltages of SO (soybean oil) and PFAE (palm oil fatty acid esters) were tested. First, the ionization potential, electron affinity, softness, and excitation energy parameters of the two insulating oil molecules were calculated under different electric field strengths, as shown in Tables 1 and 2. Data with electric field strengths of 0, 0.002au, and 0.004au were selected to plot parallel coordinate graphs of the microscopic properties of SO and PFAE molecules, and the area under the curves was calculated to compare the differences in their microscopic property values. Figure 2 and Figure 3 The results show that the potential energy and excitation energy of the PFEA molecule are greater than those of the SO molecule, while the electron affinity and softness of the PFEA molecule are less than those of the SO molecule. Therefore, the lightning impulse breakdown voltage of the PFEA molecule should be higher than that of the SO molecule. Table 3 shows the experimental breakdown voltage values ​​of the SO and PFEA molecules.

[0061] To verify the results of the proposed first-principles-based method for evaluating and optimizing the lightning impulse performance of insulating oil molecules, lightning impulse breakdown voltage tests were conducted on SO and PFAE molecules. Following IEC 60897 guidelines, the lightning impulse breakdown voltage of the insulating oil was determined using a step-up method with needle-plate electrodes. Different oil gap widths were set, specifically 25 mm, 45 mm, 65 mm, 85 mm, and 105 mm. For each oil gap width, the breakdown voltage of the insulating oil was measured ten times to obtain the average breakdown voltage value. A one-minute relaxation time was observed before each voltage application. Once the insulating oil broke down, the oil sample was stirred and allowed to stand for 5 minutes. After ten breakdowns, the insulating oil was replaced with a new sample. The experiments were conducted at 26°C to maintain consistency.

[0062] Experimental results show that the average lightning breakdown voltage of PFAE is higher than that of SO. When the oil gap is small (25 mm), the average breakdown voltage of PFAE is only slightly higher than that of SO, by about 6 kV. However, at a medium oil gap distance (65 mm), the average breakdown voltage of PFAE is 45.61 kV higher than that of SO. When the oil gap is long (105 mm), the average breakdown voltage of PFAE is about 46.29% higher than that of SO. This is consistent with the results obtained from the first-principles-based method for evaluating and optimizing the lightning impulse performance of insulating oil molecules proposed in this invention.

[0063] Table 1. Microscopic physical properties of SO molecules under different electric field intensities

[0064]

[0065] Table 2. Microscopic physical property values ​​of PFAE molecules under different electric field strengths.

[0066]

[0067] Table 3. Experimental breakdown voltage values ​​(kV) for SO and PFAE

[0068]

[0069] The above description is merely one embodiment of the present invention and is not intended to limit the present invention in any way. The embodiments described above are not intended to limit the present invention. Any person skilled in the art can make many possible variations and modifications to the technical solutions of the present invention, or modify them into equivalent embodiments, without departing from the spirit and technical essence of the present invention. Therefore, any simple modifications, equivalent substitutions, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention, without departing from the content of the technical solutions of the present invention, shall still fall within the protection scope of the technical solutions of the present invention.

Claims

1. A method for evaluating and optimizing the lightning impulse performance of insulating oil molecules based on first-principles calculations, characterized in that, The method includes the following steps: S1. Fine construction of molecular models based on quantum chemical calculations; the three-dimensional structure of insulating oil molecules is constructed using the GaussView software front end, the bond lengths, bond angles and dihedral angles are precisely set, and a sufficient vacuum region is pre-set around the model; an implicit solvation model (IEF-PCM) is adopted, and the dielectric constant is set to the value of typical insulating oil to accurately simulate the actual electron distribution of molecules in a liquid environment. S2. Dispersion correction and geometric optimization of the insulating oil molecular model: Based on the input file created by Multiwfn, the CP2K software was used to simulate the model using the Perdew-Burke-Ernzerhof (PBE) functional and the third-generation dispersion correction (D3); the MOLOPT basis set (DZVP-MOLOPT-SR-STH) was used to perform geometric optimization of the insulating oil molecular model. S3. Calculate the values ​​of performance evaluation parameters under different electric fields; use the MOLOPT basis set (TZVP-MOLOPT-SR-GTH) to calculate the ionization energy (IP), electron affinity (EA), softness, and excitation energy (E) of the insulating oil molecular model under different electric field strengths. ex ). S4. The lightning impact performance of different types of insulating oil molecules is compared and analyzed using parallel coordinate graphs. Based on the performance evaluation index parameters obtained in step 3, the four microscopic physical property indicators of insulating oil molecules are plotted into parallel coordinate graphs to comprehensively evaluate the lightning impact performance and differences of different types of insulating oil molecules.

2. The method for evaluating and optimizing the lightning impulse performance of insulating oil molecules based on first-principles calculations according to claim 1, characterized in that, Step S1 specifically involves: An initial three-dimensional structural model of the insulating oil molecule was constructed using GaussView software. Then, based on a known crystal structure database, the bond lengths, bond angles, and dihedral angles of the molecule were precisely set to ensure a realistic chemical spatial configuration. To ensure the accuracy of the simulation results, a sufficient vacuum region was reserved around the model to avoid the influence of periodic boundaries on the simulation results. To simulate the real environment of molecules in liquid insulating oil, an implicit solvation model (IEF-PCM) was used in Gaussian software. The dielectric constant of the solvent was set to a fixed value in the range of 2.0 to 2.

5. Single-point energy calculations were performed on the initial molecular model to obtain its initial electron distribution in this liquid environment, providing a more realistic starting point for subsequent geometric optimization.

3. The method for evaluating and optimizing the lightning impulse performance of insulating oil molecules based on first-principles calculations according to claim 1, characterized in that, Step S2 specifically involves: The insulating oil molecular structure corrected for electron distribution in the liquid environment from the first step is imported into Multiwfn software to create a CP2K input file. Key calculation parameters are configured in the generated input file. Within the density functional theory (DFT) calculation framework, the PBE exchange-correlated functional is used, and the DFT-D3 (BJ) method is introduced for dispersion correction. This is specifically achieved by specifying the parameter file dftd3.dat. In the geometry optimization stage, the DZVP-MOLOPT-SR-GTH basis set was selected, and corresponding basis set files were specified for all elements in the system. The convergence threshold for force was set to 4.5e-4Hartree / Bohr, and the convergence threshold for energy was set to 1.0e-6Hartree.

4. The method for evaluating and optimizing the lightning impulse performance of insulating oil molecules based on first-principles calculations according to claim 1, characterized in that, Step S3 specifically involves: Based on the optimized stable structure, while keeping the functional settings unchanged, the basis set is upgraded to TZVP-MOLOPT-SR-GTH for high-precision single-point energy calculations to extract electronic properties. Simultaneously, different self-consistent field (SCF) convergence algorithms are employed for different computational tasks. For routine single-point energy calculations, the orbital transformation (OT) method is used; when calculating the density of states and molecular surface electrostatic potential, the method is switched to diagonalization. Finally, by applying 0 to 0.0004 atomic units (1 a.u. = 5.1423 × 10⁻⁴) in the X direction of the insulating oil model, 11 The electric field within the range of V / m is calculated according to formulas (1)(2)(3)(4) to obtain the lightning impact performance evaluation indexes of insulating oil molecules under different electric fields: ionization energy (IP), electron affinity (EA), insulating fluid softness (Softness), excitation energy (E). ex ). IP=E(N + )-E(N) (1) EA=E(N)-E(N - ) (2) AND ex =And S1 -AND S0 (4)。 5. The method for evaluating and optimizing the lightning impulse performance of insulating oil molecules based on first-principles calculations according to claim 1, characterized in that, Step S4 specifically involves: To achieve comparative analysis and structural optimization of the lightning impact performance of different types of insulating oil molecules, the four indicators obtained from S3 were classified into benefit-type indicators and cost-type indicators, and standardized according to formulas (5) and (6) respectively, so that all parameter values ​​were unified to the [0,1] interval and positively correlated with lightning impact performance. Then, a parallel coordinate graph was plotted, and the six standardized data points (V1-V6) of each molecule were connected in sequence to form a broken line, and the area under the curve of each broken line in the six-dimensional performance space was calculated by the Gaussian area formula (7). The final performance evaluation criterion is: the larger the area enclosed by the parallel coordinate graph, the better the molecule's overall performance across multiple key performance indicators, and the better its lightning impulse insulation performance. Optimal molecule selection can be achieved by arranging all evaluated molecules in descending order based on the area value.

Images