Elasticity matrix determination method and vibration analysis method for laminated iron core of transformer, and elasticity matrix determination program and vibration analysis program

By determining the transverse elastic modulus in laminated cores using a constitutive equation and genetic algorithm optimization, the method addresses inaccuracies in noise prediction, improving vibration analysis accuracy and reducing noise levels in transformers.

JP2024158560AActive Publication Date: 2024-11-08JFE STEEL CORP
View PDF 3 Cites 0 Cited by

Patent Information

Application Number
JP2023073860
Authority / Receiving Office
JP · JP
Patent Type
Applications
Current Assignee / Owner
Filing Date
2023-04-27
Publication Date
2024-11-08
Estimated Expiration
2043-04-27

AI Technical Summary

Technical Problem

Existing methods for determining the elastic matrix of laminated cores in transformers fail to accurately predict noise levels due to discrepancies between calculated and measured mechanical vibrations, leading to inaccurate noise predictions and increased costs for soundproofing.

Method used

A method for determining the elasticity matrix of a laminated core by calculating the transverse elastic modulus in two planes including the lamination direction, using a constitutive equation and genetic algorithm optimization to improve vibration analysis accuracy.

Benefits of technology

The method reduces discrepancies between measured and calculated vibration characteristics, enhancing the accuracy of vibration analysis and noise prediction in transformers.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 2024158560000001_ABST
    Figure 2024158560000001_ABST
Patent Text Reader

Abstract

To provide an elasticity matrix determination method and vibration analysis method for a laminated iron core of a transformer and an elasticity matrix determination program and vibration analysis program for a laminated iron core of a transformer that contribute design to reduce noise of a transformer.SOLUTION: An elasticity matrix determination method for a laminated iron core of a transformer comprises: a step for calculating a degree of matching between a noise frequency spectrum and an excitation vibration frequency spectrum for each combination of a first assumed value and a second assumed value of modulus of transverse elasticity; and a step for adopting the first assumed value and the second assumed value when the degree of matching comes to a maximized value as moduli of transverse elasticity. The step for calculating the degree of matching includes: a step for calculating the degree of matching for trial points distributed in a prescribed range; a step for setting a limited small area; and a step for calculating the degree of matching for search points located between the trial points within the limited small area.SELECTED DRAWING: Figure 3
Need to check novelty before this filing date? Find Prior Art

Description

[Technical field]

[0001] The present disclosure relates to a method for determining an elasticity matrix of a laminated core of a transformer, a method for vibration analysis, and a program for determining an elasticity matrix and a program for vibration analysis. The present disclosure also relates to a method for calculating, by using a computer, mechanical vibration characteristics of a transformer core manufactured by laminating electromagnetic steel sheets. [Background technology]

[0002] An analysis device is known that calculates each node of a finite element according to the magnetic flux density or a nodal force equivalent to the distortion of each finite element based on a numerical analysis model that represents an electromagnetic component including a magnetic body that generates magnetostriction as a combination of multiple finite elements in a finite element method analysis (see, for example, Patent Document 1). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2014-71689 A Summary of the Invention [Problem to be solved by the invention]

[0004] In transformers, which are constructed by winding a coil around a laminated core made of laminated electromagnetic steel sheets, the vibration of the core caused by magnetostriction generates noise. Structural analysis is required to contribute to the design of transformers to reduce noise.

[0005] Therefore, an object of the present disclosure is to provide an elasticity matrix determination method and vibration analysis method, as well as an elasticity matrix determination program and a vibration analysis program, for a laminated core of a transformer, which contribute to designs that reduce transformer noise. [Means for solving the problem]

[0006] (1) A method for determining an elasticity matrix of a laminated core of a transformer according to an embodiment of the present disclosure determines an elasticity matrix of the laminated core of a transformer. The laminated core is formed by stacking a plurality of electromagnetic steel sheets and has at least a first portion and a second portion. Constitutive equations expressing the relationship between stress and strain in the first portion and the second portion each include a different elasticity matrix. The elasticity matrix includes, as elements of the elasticity matrix, transverse elastic modulus in two planes including the stacking direction of the laminated core. The method for determining an elastic matrix of a laminated core of a transformer includes the steps of: A first step of acquiring a frequency spectrum of excitation noise of the laminated core of the transformer; A second step of acquiring a frequency spectrum of the excitation magnetostriction of the electromagnetic steel sheet; a third step of determining a first assumed value for a transverse elastic modulus in two planes including a lamination direction in the first portion of the laminated core of the transformer, determining a second assumed value for a transverse elastic modulus in two planes including a lamination direction in the second portion of the laminated core of the transformer, and calculating a frequency spectrum of a vibration response function of the laminated core of the transformer when the first assumed value and the second assumed value are applied to the elasticity matrix; a fourth step of calculating a frequency spectrum of excitation vibration of the laminated core of the transformer based on the frequency spectrum data of excitation magnetostriction of the electromagnetic steel sheet acquired in the second step and the frequency spectrum of the vibration response function of the laminated core of the transformer calculated in the third step; a fifth step of calculating a degree of agreement between the frequency spectrum of the noise of the laminated core of the transformer obtained in the first step and the frequency spectrum of the excitation vibration of the laminated core of the transformer calculated in the fourth step; a sixth step of calculating the degree of agreement for each combination of the first assumption value and the second assumption value by repeatedly executing the third step through the fifth step while changing each of the first assumption value and the second assumption value; a seventh step of detecting a maximum value of the degree of agreement calculated for each combination of the first assumed value and the second assumed value in the sixth step, adopting the first assumed value when the degree of agreement is a maximum value as a modulus of transverse elasticity in two planes including the lamination direction of the laminated core in the first portion, and adopting the second assumed value when the degree of agreement is a maximum value as a modulus of transverse elasticity in two planes including the lamination direction of the laminated core in the second portion; Includes. The sixth step is a sixth step of expressing each combination of the first assumption value and the second assumption value corresponding to the degree of agreement as a trial point set in a two-dimensional space with the first assumption value as a first axis and the second assumption value as a second axis, sequentially selecting points distributed within a predetermined range of the two-dimensional space as the trial points and calculating the degree of agreement for the combination of the first assumption value and the second assumption value corresponding to the trial points; a sixth step of setting a limited small region in the two-dimensional space based on the degree of coincidence calculated for a combination of the first assumption value and the second assumption value corresponding to the trial point; a 6-C step of setting search points in the limited small region at intervals narrower than the intervals of the trial points and calculating the degree of agreement for combinations of the first assumption value and the second assumption value corresponding to the search points; Includes.

[0007] (2) In the fifth step of the method for determining the elasticity matrix of a laminated core of a transformer described in (1) above, the degree of agreement may be calculated by calculating the cosine value of the angle between a first vector having elements that are frequency components of the frequency spectrum of the noise of the laminated core of the transformer, and a second vector having elements that are frequency components of the frequency spectrum of the excitation vibration of the laminated core of the transformer.

[0008] (3) A vibration analysis method according to one embodiment of the present disclosure includes a step of performing vibration analysis of a laminated core of a transformer based on a constitutive equation expressing the relationship between stress and strain in each of the first part and the second part, incorporating a transverse elastic modulus determined by the elastic matrix determination method of a laminated core of a transformer described in (1) or (2) above.

[0009] (4) An elasticity matrix determination program according to one embodiment of the present disclosure causes a computer to execute the elasticity matrix determination method described in (1) or (2) above.

[0010] (5) A vibration analysis program according to an embodiment of the present disclosure causes a computer to execute the vibration analysis method described in (3) above. Effect of the Invention

[0011] According to the elasticity matrix determination method and elasticity matrix determination program for a laminated core of a transformer of the present disclosure, the deviation between the measured value and the calculated value of the vibration characteristics can be reduced. Also, according to the vibration analysis method and vibration analysis program for a laminated core of a transformer of the present disclosure, the accuracy of the vibration analysis can be improved. [Brief description of the drawings]

[0012] [Figure 1] 1 is a diagram showing an example of the configuration of a laminated core of a transformer that is to be analyzed by a vibration analysis method according to an embodiment; [Diagram 2] A diagram explaining the stress acting on a laminated core. (A) Normal stress in the lamination direction (Z-axis direction), (B) Normal stress in the direction perpendicular to the lamination direction (X-axis direction and Y-axis direction), (D) Shear stress in the ZX plane, (E) Shear stress in the XY plane, and (F) Shear stress in the YZ plane. [Diagram 3] 1 is a flowchart illustrating an example of a procedure for an elasticity matrix determination method according to an embodiment. [Figure 4] 1 is a block diagram showing an example of the configuration of a vibration analysis device according to an embodiment; [Diagram 5] 1 is a graph showing an example of a frequency spectrum of noise of a transformer. [Figure 6] 1 is a graph showing an example of frequency spectrum data of excitation magnetostriction of electromagnetic steel sheets constituting a laminated core of a transformer. [Figure 7] 1 is a graph showing an example of a frequency spectrum of a vibration response function of a laminated core of a transformer. [Figure 8] 1 is a graph showing an example of a frequency spectrum of excitation vibration of a laminated core of a transformer. [Figure 9] 9 is a graph showing an example of a two-dimensional map of the degree of agreement between the frequency spectrum of excitation vibration and the frequency spectrum of excitation noise of a laminated core of a transformer. [Figure 10] FIG. 10 is a diagram showing an example of a search path performed to obtain the two-dimensional map of FIG. 9. [Figure 11] 1 is a graph showing an example of a frequency spectrum of noise of a transformer according to an embodiment. [Figure 12] 1 is a graph showing an example of frequency spectrum data of excitation magnetostriction of electromagnetic steel sheets constituting a laminated core of a transformer according to an embodiment. [Figure 13] 1 is a graph showing an example of a frequency spectrum of a vibration response function of a laminated core of a transformer according to an embodiment. [Figure 14] 1 is a graph showing an example of a frequency spectrum of excitation vibration of a laminated core of a transformer according to an embodiment. [Figure 15] 1 is a graph showing an example of a two-dimensional map of the degree of agreement between the frequency spectrum of excitation vibration and the frequency spectrum of excitation noise of a laminated core of a transformer according to an embodiment. [Figure 16] 11 is a graph comparing calculated values ​​of the frequency spectrum of excitation vibration of a laminated core of a transformer according to an embodiment with actual measured values ​​of the frequency spectrum of noise obtained by measuring excitation noise of a laminated core of a transformer according to an embodiment. [Figure 17] FIG. 13 is a diagram showing a search trajectory using a GA (genetic algorithm). [Figure 18A] 1 is a graph showing an example of trial points distributed at 0.1 GPa intervals on a two-dimensional map. [Figure 18B] 18B is a graph showing an example of the results of calculating the degree of match at the trial points in FIG. 18A. [Figure 19A] 1 is a graph showing an example of trial points distributed at 0.075 GPa intervals on a two-dimensional map. [Figure 19B] 19B is a graph showing an example of the results of calculating the degree of match at the trial points in FIG. 19A. [Figure 20A] 1 is a graph showing an example of trial points distributed at 0.05 GPa intervals on a two-dimensional map. [Figure 20B] 20B is a graph showing an example of the results of calculating the degree of match at the trial points in FIG. 20A. [Figure 21A] 1 is a graph showing an example of trial points distributed at 0.025 GPa intervals on a two-dimensional map. [Figure 21B] 21B is a graph showing an example of the results of calculating the degree of match at the trial points in FIG. 21A. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0013] Transformers such as power distribution transformers are constructed by winding a coil around a laminated core made of laminated electromagnetic steel sheets. Important performance characteristics of a transformer include, for example, core loss (no-load loss) characteristics, excitation current characteristics, and noise characteristics.

[0014] Distribution transformers are installed in various locations. There is a strong demand for low noise levels, especially for transformers installed in urban areas. In recent years, in consideration of the impact of transformers on the surrounding environment, noise reduction has become increasingly important as a required performance for transformers.

[0015] Grain-oriented magnetic steel sheets, which are often used as the iron core material of transformers, vibrate due to material expansion and contraction caused by excitation. Material expansion and contraction caused by excitation is also called magnetostriction. Vibration caused by magnetostriction is also called magnetostrictive vibration. Magnetostrictive vibration is one of the main causes of noise generated in transformers. Therefore, the noise performance of a transformer can be strongly dependent on the magnetostrictive properties of the magnetic steel sheet used as the iron core material. In order to reduce transformer noise, magnetic steel sheets with low magnetostrictive properties can be used as the iron core material of transformers.

[0016] However, even when electromagnetic steel sheets with excellent magnetostrictive performance are used as the core material, the noise of the transformer is often not sufficiently reduced. The reason why the noise is not sufficiently reduced is thought to be the occurrence of a resonance phenomenon caused by the natural frequency of the transformer core being close to the frequency of the magnetostrictive vibration of the electromagnetic steel sheet. For this reason, when designing and manufacturing a transformer, it is extremely important to calculate the mechanical vibration characteristics, such as the natural frequency of the transformer core, in advance.

[0017] Here, it is considered to analyze the mechanical vibration characteristics of an electromagnetic component including a magnetic material that generates magnetostriction, such as an iron core of a transformer, by the finite element method. For example, based on a numerical analysis model in which the electromagnetic component is expressed by a combination of multiple finite elements in the finite element analysis, each node of the finite element or a nodal force equivalent to the distortion of each finite element according to the magnetic flux density applied to the electromagnetic component can be calculated. In this case, a quasi-static structural analysis is performed using a governing equation of structural analysis consisting of an equation of balance, a constitutive equation showing the relationship between stress and distortion, and an equation of the relationship between displacement and distortion.

[0018] In the above quasi-static structural analysis, the relationship between the stress tensor {σ} and the strain tensor {ε} is expressed by the formula {σ}={D}{ε}, where {·} represents a tensor, and {D} is a tensor that represents the relationship between strain and stress.

[0019] Here, the component representation of the above formula is expressed as formula (1).

number

[0020] In formula (1), {D} is expressed with 81 components. {σ} and {ε} are each expressed with 9 components. Here, the tensor as a physical quantity is a symmetric tensor. By considering the symmetric components, the number of independent components of {σ} and {ε} each becomes 6. As a result, by expressing the tensor relationship as a matrix, such as [σ]=[D][ε] using the symbol [·] representing a matrix, and then expressing the components, it is expressed as formula (2). Here, [D] is also called the elasticity matrix.

number

[0021] In addition, the relationship between the normal stress σi and the normal strain εi, and the relationship between the shear strain τij and the shear strain γij are expressed using the elasticity matrix [D].

[0022] Here, the elastic coefficient of the material itself, such as an electromagnetic steel sheet, constituting the electromagnetic component to be structurally analyzed, can be applied as it is as the elastic matrix [D]. However, it is known that when the elastic coefficient of the material itself is applied as the elastic matrix [D], the difference between the calculation result of the mechanical vibration of the electromagnetic component to be structurally analyzed and the actual measurement result of the mechanical vibration becomes large. Due to the large difference between the calculation result and the measurement result, it has been considered difficult to reflect the result of the structural analysis by the structural analysis program in the design of the electromagnetic component, which is a laminated structure formed by laminating steel sheets.

[0023] In addition, the iron core used in transformers such as distribution transformers is also a laminated structure formed by laminating steel plates. Therefore, even if the noise is predicted based on the results of structural analysis using a structural analysis program, the prediction accuracy is low. The low accuracy of the noise prediction sometimes results in a loud noise in the actually manufactured transformer. If the transformer is noisy, problems such as extra costs for soundproofing measures such as soundproofing walls arise. In order to reduce the noise of the transformer and suppress the occurrence of extra costs, it is possible to change the design of the transformer iron core. However, when there is a large difference between the calculation results and the measurement results of the mechanical vibration, it is not clear how to change the design to reduce the noise. In order to determine a policy for changing the design of the transformer, it is required to improve the prediction accuracy of the noise using structural analysis using a structural analysis program.

[0024] When comparing the mechanical vibration calculation results of the transformer core that is the subject of structural analysis with the results of actually measuring the mechanical vibration, the following is thought to be the cause of the large discrepancy between the calculated and measured values. When the transformer core is a laminated structure made of laminated steel plates, the elasticity matrix [D] of the laminated structure is originally a different value from the elasticity matrix [D] of the parts themselves, such as the electromagnetic steel plates, that make up the laminated core of the transformer. Despite this, the structural analysis is performed by directly substituting the value of the elasticity matrix [D] of the parts themselves, such as the electromagnetic steel plates, which is thought to be the reason for the large discrepancy between the calculated and measured values.

[0025] Therefore, it is necessary to determine the elastic matrix [D] of the laminated structure with high accuracy so that the difference between the calculation result and the measurement result can be reduced. In particular, when performing vibration analysis of a laminated core made of laminated steel plates using a constitutive equation that expresses the relationship between stress and strain in a matrix, it is necessary to determine with high accuracy the transverse elastic modulus in two planes including the lamination direction of the laminated core of the transformer, which is included in the elastic matrix [D] in the constitutive equation.

[0026] The present disclosure aims to provide a method for determining the elasticity matrix of a laminated core of a transformer, which can determine with high accuracy the transverse elastic modulus in two planes including the lamination direction of the laminated core of a transformer, which is included in the elasticity matrix in a constitutive equation that expresses the relationship between stress and strain in a matrix representation.The present disclosure also aims to provide a vibration analysis method that can improve the accuracy of the vibration analysis results by analyzing the vibration of a laminated core made of laminated steel plates using a constitutive equation including the determined elasticity matrix.

[0027] Hereinafter, embodiments of the elasticity matrix determination method for a laminated core and the vibration analysis method for a laminated core according to the present disclosure will be described with reference to the drawings. The drawings are schematic and may differ from the actual product. Furthermore, the following embodiments are illustrative of an apparatus or method for embodying the technical idea of ​​the present invention, and are not intended to limit the configuration to that described below. In other words, the technical idea of ​​the present disclosure can be modified in various ways within the technical scope described in the claims.

[0028] (Embodiment) The laminated core 21 to be analyzed in this embodiment is assumed to be a laminated core for a three-phase three-legged transformer used as a power distribution transformer, for example. As shown in FIG. 1, the laminated core 21 may be configured by laminating a plurality of grain-oriented electromagnetic steel sheets 22 having a predetermined thickness, each of which has three legs 22c connected between an upper yoke 22a and a lower yoke 22b. As an example, the grain-oriented electromagnetic steel sheets 22 have a thickness of 0.23 mm. As an example, the number of grain-oriented electromagnetic steel sheets 22 is 333. The laminated grain-oriented electromagnetic steel sheets 22 are fixed by wrapping a glass tape around them. As an example, the dimensions of the upper yoke 22a and the lower yoke 22b are set to a width of 100 mm and a length of 500 mm. As an example, the dimensions of the three legs 22c are set to a width of 100 mm and a length of 300 mm. The three legs 22c are connected between the upper yoke 22a and the lower yoke 22b at intervals of 100 mm. The laminated core 21 to be analyzed in this embodiment is not limited to the laminated core for a three-phase three-legged transformer illustrated in Fig. 1. The sheet thickness of the grain-oriented electromagnetic steel sheets 22, the number of laminations of the grain-oriented electromagnetic steel sheets 22, the dimensions of the upper yoke 22a and the lower yoke 22b, or the dimensions of the three legs 22c are not limited to the above-mentioned examples.

[0029] In order to perform a numerical analysis of the vibration of the laminated core 21 of such a three-phase three-limbed transformer, a constitutive equation showing the relationship between stress and strain is used as the governing equation for elastic structural analysis. The constitutive equation is expressed as the following equation (3) by replacing the laminates with an equivalent homogeneous body and expressing the influence of the laminations with matrix properties. [σ] = [C][ε] (3) [σ] is the stress matrix. [C] is the elasticity matrix (stiffness matrix) as a response function. [C] is the same as [D] above. [ε] is the strain matrix.

[0030] 1, the directions of the laminated core 21 are represented by the X-axis, Y-axis, and Z-axis of an XYZ orthogonal coordinate system. The grain-oriented electromagnetic steel sheets 22 extend along the XY plane and are laminated along the Z-axis direction.

[0031] The stress acting on the laminated core 21 includes a vertical component acting in a vertical direction (compression direction or tension direction) to the laminated core 21, and a shear component acting in the shear direction of the laminated core 21. Normal stresses acting in the Z-axis direction, X-axis direction, and Y-axis direction are illustrated in Figs. 2(A), (B), and (C), respectively. The normal stresses acting in the Z-axis direction, X-axis direction, and Y-axis direction are represented by σz, σx, and σy, respectively. Shear stresses acting in the ZX plane, XY plane, and YZ plane are represented by τzx, τxy, and τyz, respectively. The stress matrix [σ] has normal stress and shear stress as components.

[0032] The strain of the laminated core 21 includes a component that distorts the laminated core 21 in the vertical direction (compression or tension direction) and a component that distorts the laminated core 21 in the shear direction. The vertical strains in the Z-axis, X-axis, and Y-axis directions are represented by εz, εx, and εy, respectively. The shear strains in the ZX plane, XY plane, and YZ plane are represented by γzx, γxy, and γyz, respectively.

[0033] The elasticity matrix [C] has 6 x 6 = 36 elements that specify the relationship between the six components of stress and the six components of strain. The 36 elements are expressed as elastic coefficients Cij (i = 1 to 6, j = 1 to 6). The relationship between stress and strain is expressed by the elasticity matrix as shown in the following equation (4).

number

[0034] Since the laminated core 21 is formed by stacking grain-oriented electromagnetic steel sheets 22, it has mechanical symmetry of the laminated core 21, and also has 180-degree symmetry in the longitudinal direction of the laminated grain-oriented electromagnetic steel sheets 22 and in the direction perpendicular to the longitudinal direction. Therefore, it can be said that the laminated core 21 has orthogonal anisotropy in terms of anisotropy classification. The elastic matrix of an object having orthogonal anisotropy is basically expressed by nine elastic coefficients, C11, C12, C13, C22, C23, C33, C44, C55, and C66, as shown in the following formula (5).

number

[0035] Of these nine elastic coefficients, the elastic coefficients C11, C12, C13, C22, C23 and C33 are calculated based on the longitudinal elastic coefficients Ex, Ey and Ez and the Poisson's ratios νxy, νyx, νyz, νzy, νzx and νxz using the following formulas (6) to (12).

number

[0036] As shown in the following formulas (13), (14), and (15), the elastic coefficient C44 corresponds to the transverse elastic coefficient Gyz of the YZ plane, the elastic coefficient C55 corresponds to the transverse elastic coefficient Gzx of the ZX plane, and the elastic coefficient C66 corresponds to the transverse elastic coefficient Gxy of the XY plane.

number

[0037] Here, Ex, Ey, and Ez respectively represent the X-direction elastic modulus (Young's modulus), the Y-direction elastic modulus (Young's modulus), and the Z-direction elastic modulus (Young's modulus). νxy, νyx, νyz, νzy, νzx, and νxz respectively represent the Poisson's ratio of the XY plane (the ratio of the X-direction longitudinal strain to the Y-direction lateral strain), the Poisson's ratio of the YX plane, the Poisson's ratio of the YZ plane, the Poisson's ratio of the ZY plane, the Poisson's ratio of the ZX plane, and the Poisson's ratio of the XZ plane. The relationship between the elastic modulus and the Poisson's ratio is expressed by the following formula (16), which is called the reciprocity theorem.

number

[0038] By the reciprocity theorem, the Poisson's ratio νyx in the YX plane is expressed using Ex, Ey, and νxy. The Poisson's ratio νzy in the ZY plane is expressed using Ez, Ey, and νyz. The Poisson's ratio νxz in the XZ plane is expressed using Ez, Ex, and νzx.

[0039] In this way, the values ​​of the nine elastic coefficients C11, C12, C13, C22, C23, C33, C44, C55, and C66 that represent the elastic matrix of an object having orthogonal anisotropy are expressed using a total of nine mechanical property values, namely, the longitudinal elastic modulus Ex, Ey, and Ez, the transverse elastic modulus Gyz, Gzx, and Gxy, and the Poisson's ratios νxy, νyz, and νzx. Therefore, determining these nine mechanical property values ​​is equivalent to determining the nine elastic coefficients that represent the elastic matrix. Below, a method for determining the longitudinal elastic modulus, the transverse elastic modulus, and the Poisson's ratio will be described.

[0040] First, for the longitudinal elastic modulus of the laminated core having orthogonal anisotropy, the longitudinal elastic modulus Ex and Ey can be set to values ​​equal to the longitudinal elastic modulus Ex0 and Ey0 of one steel plate. On the other hand, the longitudinal elastic modulus Ez cannot be set to a value equal to the longitudinal elastic modulus Ez0 of one steel plate. This is because there is a small gap between the laminated steel plates. Therefore, in this embodiment, an experiment is performed to obtain the relationship between the load and displacement in the lamination direction of the laminated steel plates, and the longitudinal elastic modulus Ez is set based on the experimental results. In this embodiment, Ez is set to 10 GPa based on the experimental results. Note that the magnitude of the value of the longitudinal elastic modulus in the lamination direction has little effect on the vibration calculation result. Therefore, even if Ez is set to a value equal to the longitudinal elastic modulus Ez0 of one steel plate, the error generated in the vibration calculation result will not be large.

[0041] In addition, regarding the Poisson's ratio of the laminated core having orthogonal anisotropy, the Poisson's ratio νxy in the XY plane can be set to a value equal to the Poisson's ratio νxy0 of one steel plate. On the other hand, the Poisson's ratio νyz in the YZ plane and the Poisson's ratio νzx in the ZX plane cannot be set to the Poisson's ratios νyz0 and νzx0 of one steel plate as they are. This is because, in the laminated core, the mechanical coupling between the strain in the lamination direction and the strain in the direction perpendicular to the lamination direction is considered to be extremely weak. Here, it is extremely difficult to actually measure νyz and νzx. However, from the above consideration, it is expected that νyz and νzx will be extremely small values. Therefore, in this embodiment, νyz and νzx are both set to zero.

[0042] Furthermore, regarding the transverse elastic modulus of the laminated core, the transverse elastic modulus Gxy in the XY plane can be set to a value equal to the transverse elastic modulus Gxy0 of one steel plate. On the other hand, the transverse elastic modulus of two planes including the lamination direction, i.e., the transverse elastic modulus Gzx in the ZX plane and the transverse elastic modulus Gyz in the YZ plane, cannot be set to the transverse elastic modulus Gxz0 and Gyz0 of one steel plate as they are. This is because it is necessary to reflect the influence of slippage between the steel plates that occurs in the X and Y directions perpendicular to the lamination direction at the interface between the laminated steel plates in the laminated steel plate in the transverse elastic modulus Gzx and Gyz. Therefore, in order to improve the analysis accuracy of vibration analysis using a constitutive equation that expresses the relationship between the stress and strain of the laminated core in a matrix representation, it is important to determine with high accuracy the transverse elastic modulus of two planes including the lamination direction of the laminated core of the transformer, i.e., the transverse elastic modulus Gzx in the ZX plane and the transverse elastic modulus Gyz in the YZ plane, which are included in the elastic matrix in the constitutive equation.

[0043] Here, as a method for determining the transverse elastic modulus Gzx in the ZX plane and the transverse elastic modulus Gyz in the YZ plane as values ​​that reflect the effect of slippage between the steel sheets, it is conceivable to actually manufacture a laminated core for a three-phase three-limbed transformer and measure the transverse elastic modulus Gzx and Gyz accurately. However, a method for measuring the transverse elastic modulus of a laminated core has not been established. This is because, while mechanical constants including the transverse elastic modulus can be measured by ultrasonic measurements in the case of metallic materials, in the case of a laminated core, slippage between the steel sheets significantly damps vibrations, making accurate measurement difficult.

[0044] Another method for determining the modulus of transverse elasticity in two planes including the lamination direction of a laminated core is to measure the natural frequency of the laminated core and estimate the modulus of transverse elasticity from that value. However, measuring the natural frequency of a laminated core of a transformer can also be difficult due to the large vibration damping. It is particularly difficult to measure the natural frequency when the core is large.

[0045] Therefore, in the elastic matrix determination method according to the present embodiment, the transverse elastic modulus Gzx and Gyz in two planes including the lamination direction are determined by executing the procedure shown in FIG. 3. In this embodiment, the laminated core 21 for a three-phase three-legged transformer shown as an example in FIG. 1 is analyzed. In this embodiment, the transverse elastic modulus of the upper yoke 22a and the transverse elastic modulus of the lower yoke 22b constituting the laminated core 21 are the same value. The transverse elastic modulus of the upper yoke 22a and the lower yoke 22b constituting the laminated core 21 is different from the transverse elastic modulus of the leg 22c. In other words, the transverse elastic modulus Gyz and Gzx in two planes including the lamination direction of the transformer laminated core 21 included in the elastic matrix in the constitutive formula differs between the upper yoke 22a and the lower yoke 22b constituting the laminated core 21 and the leg 22c. Therefore, in the elastic matrix determination method according to the present embodiment, the transverse elastic modulus Gyz and Gzx in two planes including a plurality of different lamination directions are determined. By executing the elasticity matrix determination method according to this embodiment, the transverse elastic moduli Gyz and Gzx of the upper yoke 22a and the lower yoke 22b, and the transverse elastic moduli Gyz and Gzx of the leg portion 22c are determined, respectively.

[0046] The upper yoke 22a and the lower yoke 22b are also collectively referred to as a first portion. The leg portion 22c is also referred to as a second portion. That is, the laminated core 21 includes a first portion and a second portion. The transverse elastic modulus Gyz and Gzx of the first portion and the transverse elastic modulus Gyz and Gzx of the second portion are different from each other. By executing the elasticity matrix determination method according to this embodiment, the transverse elastic modulus Gyz and Gzx of the first portion and the transverse elastic modulus Gyz and Gzx of the second portion, which are different from each other, are determined.

[0047] Each step of the elasticity matrix determination method according to this embodiment may be executed by a vibration analysis device 40 illustrated in Fig. 4. The vibration analysis device 40 includes a control unit 42, a storage unit 44, and an interface 46.

[0048] The control unit 42 controls and manages each functional unit constituting the vibration analysis device 40 and the vibration analysis device 40 as a whole. The control unit 42 may be configured to include at least one processor, such as a CPU (Central Processing Unit), in order to control and manage various functions. The control unit 42 may be configured with one processor or multiple processors. The processor constituting the control unit 42 may realize the functions of the vibration analysis device 40 by reading and executing a program stored in the storage unit 44.

[0049] The storage unit 44 may function as a memory that stores various information. The storage unit 44 may store, for example, a program executed in the control unit 42, or data used in processing executed in the control unit 42, or a result of processing. The storage unit 44 may also function as a work memory for the control unit 42. The storage unit 44 may be configured, for example, by a semiconductor memory or the like, but is not limited thereto, and may include any storage device. For example, the storage unit 44 may be configured as an internal memory of a processor used as the control unit 42, or may be configured as a hard disk drive (HDD) accessible from the control unit 42.

[0050] The interface 46 may be configured to include a communication interface for communicating with other devices by wire or wirelessly. The interface 46 may be configured to include an input / output port for inputting and outputting data to and from other devices. The interface 46 transmits and receives necessary data and signals to and from a process computer or a higher-level system. The interface 46 may communicate based on a wired communication standard, or may communicate based on a wireless communication standard. For example, the wireless communication standard may include cellular phone communication standards such as 3G, 4G, and 5G. In addition, for example, the wireless communication standard may include IEEE802.11, Bluetooth (registered trademark), and the like. The interface 46 may support one or more of these communication standards. The interface 46 is not limited to these examples, and may communicate with other devices or input and output data based on various standards.

[0051] In the following, each step of the elasticity matrix determining method according to this embodiment will be described as being executed by the control unit 42 of the vibration analysis device 40. Each step of the elasticity matrix determining method may be executed by another device or may be executed by a human being.

[0052] First, in step S1 (first step) of FIG. 3, the control unit 42 obtains the measurement result of the frequency spectrum of the noise obtained by measuring the excitation noise of the laminated core 21 of the transformer to be subjected to the vibration analysis shown in FIG. 1. Specifically, the excitation noise is measured by setting the excitation frequency of the laminated core 21 to 50 Hz. Since the frequency spectrum of the excitation noise of the transformer is composed of frequency components that are integral multiples of twice the excitation frequency (even multiples of the excitation frequency), the spectrum of the excitation noise at each frequency in 100 Hz pitch within the range of 100 Hz to 1000 Hz is obtained as the frequency spectrum of the noise. At the manufacturing site of the transformer, the excitation noise characteristic of the transformer is widely measured as a quality control item of the shipped product. Therefore, there is no particular difficulty in the measurement. The excitation frequency may be set to 60 Hz. When the excitation frequency is set to 60 Hz, the spectrum of the excitation noise at 120 Hz pitch from 120 Hz to 1200 Hz is obtained.

[0053] The following describes the case where the excitation frequency is set to 50 Hz. An example of the frequency spectrum of the noise acquired when the excitation frequency is set to 50 Hz is shown as a graph in Fig. 5. In the graph of Fig. 5, the horizontal axis represents frequency, and the vertical axis represents the intensity of the excitation noise.

[0054] Next, in step S2 (second step) of FIG. 3, the control unit 42 acquires frequency spectrum data of excitation magnetostriction of the same electromagnetic steel sheet as the grain-oriented electromagnetic steel sheet 22 constituting the laminated core 21 of the transformer to be subjected to vibration analysis shown in FIG. 1. The frequency spectrum data of excitation magnetostriction is acquired by acquiring spectrum data of excitation magnetostriction amplitude at each frequency in a pitch of 100 Hz within a range of 100 Hz to 1000 Hz. The frequency spectrum data of excitation magnetostriction can be acquired by measuring the excitation magnetostriction of the electromagnetic steel sheet, or by acquiring it from a manufacturer of the electromagnetic steel sheet. An example of the acquired frequency spectrum data of excitation magnetostriction is shown as a graph in FIG. 6. In the graph of FIG. 6, the horizontal axis represents frequency, and the vertical axis represents the magnitude of the magnetostriction amplitude. The magnitude of the magnetostriction amplitude at each frequency is expressed in dB by dividing the value by the value of the magnetostriction amplitude when the frequency is 100 Hz. The dB display corresponds to a value obtained by multiplying the common logarithm of the displayed value by 10.

[0055] Next, in step S3 (third step), the control unit 42 assumes transverse elastic moduli Gyz and Gzx, and calculates a frequency spectrum of a vibration response function for the assumed transverse elastic moduli. Specifically, the control unit 42 performs a vibration response analysis on the laminated core 21 of the transformer that is the subject of the vibration analysis using structural analysis software. Of the nine mechanical property values ​​of the laminated core 21, namely, the longitudinal elastic moduli Ex, Ey, and Ez, the transverse elastic moduli Gyz, Gzx, and Gxy, and the Poisson's ratios νxy, νyz, and νzx, the seven values ​​excluding the transverse elastic moduli Gyz and Gzx are set as follows, as described above. Ex=Ex0, Ey=Ey0, Ez=10GPa Gxy=Gxy0 νxy=νxy0, νyz=νzx=0

[0056] The transverse elastic modulus Gyz and Gzx in two planes including the two lamination directions are different for each of the multiple (two) parts constituting the laminated core 21. Specifically, the transverse elastic modulus Gyz and Gzx in the upper yoke 22a and the lower yoke 22b are different from the transverse elastic modulus Gyz and Gzx in the leg 22c. Here, the transverse elastic modulus in the upper yoke 22a and the lower yoke 22b is provisionally set as Gyz=Gzx=G1. The transverse elastic modulus in the leg 22c is provisionally set as Gyz=Gzx=G2. G1 and G2 are values ​​provisionally selected and set from a predetermined range as the transverse elastic modulus, and are also referred to as assumed values ​​of the transverse elastic modulus. G1 corresponds to the transverse elastic modulus Gyz and Gzx in the first part, and is also referred to as the first assumed value. G2 corresponds to the transverse elastic modulus Gyz and Gzx in the second part, and is also referred to as the second assumed value. The control unit 42 changes the combination of the assumed values ​​(G1, G2) and calculates the frequency spectrum of the vibration response function of the laminated core 21 of the transformer for each combination.

[0057] An example of a frequency spectrum of the vibration response function of the laminated iron core 21 of the transformer calculated for a combination of specific values ​​(G1, G2) is shown as a graph in FIG. 7. In the graph of FIG. 7, the horizontal axis represents frequency, and the vertical axis represents the magnitude of the response function. The magnitude of the response function is displayed after being normalized so that the vibration response value is 0 dB when the frequency is 100 Hz. Here, each of the hypothetical values ​​(G1, G2) is actually selected from the range of transverse elastic modulus expected from the structure of the laminated iron core 21 of the transformer. In this embodiment, each of G1 and G2 is selected from the range of 0.05 GPa to 0.5 GPa.

[0058] Next, in step S4 (fourth step), the control unit 42 calculates the frequency spectrum of the excitation vibration of the transformer laminated core 21 based on the frequency spectrum data of the excitation magnetostriction of the electromagnetic steel sheets acquired in step S2 (second step) and the frequency spectrum of the vibration response function of the transformer laminated core 21 calculated in step S3 (third step). Specifically, the control unit 42 calculates the frequency spectrum of the excitation vibration of the transformer laminated core 21 as the product of the frequency spectrum data of the excitation magnetostriction of the electromagnetic steel sheets and the frequency spectrum of the vibration response function of the transformer laminated core 21. When the frequency spectrum is displayed in dB, the control unit 42 calculates the frequency spectrum of the excitation vibration of the transformer laminated core 21 as a sum.

[0059] An example of the frequency spectrum of the excitation vibration of the laminated core 21 of the transformer calculated in this manner is shown as a graph in Fig. 8. In the graph of Fig. 8, the horizontal axis represents frequency, and the vertical axis represents the calculated value of the magnitude of the excitation vibration. The magnitude of the excitation vibration is expressed in dB.

[0060] Next, in step S5 (fifth step), the control unit 42 calculates the degree of agreement between the measurement results of the frequency spectrum of the noise of the transformer laminated core 21 acquired in step S1 (first step) and the calculation results of the frequency spectrum of the excitation vibration of the transformer laminated core 21 calculated in step S4 (fourth step).

[0061] The frequency spectrum data of the noise is measured for each frequency from 100 Hz to 1000 Hz at 100 Hz intervals. Therefore, the number of frequency spectrum data of the noise is 10. Therefore, each frequency component of the frequency spectrum data of the noise is expressed as (S100, S200, S300, . . . , S1000). Furthermore, the frequency spectrum data of the excitation vibration is calculated for each frequency from 100 Hz to 1000 Hz at 100 Hz intervals. Therefore, the number of frequency spectrum data of the excitation vibration is 10. Therefore, each frequency component of the frequency spectrum data of the excitation vibration is expressed as (V100, V200, V300, . . . , V1000). Below, a method of calculating the degree of agreement will be explained based on these notations.

[0062] Each frequency component of the noise frequency spectrum data (S100, S200, S300, . . . , S1000) and each frequency component of the excitation vibration frequency spectrum data (V100, V200, V300, . . . , V1000) are regarded as data points in a multi-dimensional space. In this case, these data points are regarded as data points in a 10-dimensional space.

[0063] Here, the control unit 42 can generally calculate the degree of coincidence between the frequency spectrum data of the noise and the frequency spectrum data of the excitation vibration as the Euclidean distance of data points in a multidimensional space. However, when the degree of coincidence is calculated as the Euclidean distance, it is known that the calculation accuracy of the degree of coincidence decreases due to the presence of an offset in the data. When the frequency spectrum data of the excitation noise and the excitation vibration are expressed in dB values, there is a concern that the degree of freedom in determining where the reference value is to be set may cause an offset in the data, deteriorating the calculation accuracy of the degree of coincidence.

[0064] Therefore, in order to avoid the concern that the calculation accuracy may be deteriorated due to the offset of data, in this embodiment, the control unit 42 calculates the cosine coincidence as the coincidence. The cosine coincidence is calculated as the cosine value of the angle between the multidimensional vectors S and V (10-dimensional space vector in this case) having (S100, S200, S300, . . . , S1000) and (V100, V200, V300, . . . , V1000) as elements, respectively. The multidimensional vector S is also referred to as the first vector. The multidimensional vector V is also referred to as the second vector. The cosine coincidence is calculated by the following formula (17). In formula (17), COSθ represents the cosine coincidence. The symbols with "→" added above the symbols "S" and "V" represent the multidimensional vectors S and V.

number

[0065] The cosine agreement is calculated as the inner product of multidimensional vectors S and V divided by the product of the length of multidimensional vector S and the length of multidimensional vector V. If multidimensional vector S and multidimensional vector V are identical, the cosine agreement is calculated as 1. If multidimensional vector S and multidimensional vector V are orthogonal, the cosine agreement is calculated as 0. If multidimensional vector S and multidimensional vector V are oriented in opposite directions, the cosine agreement is calculated as -1.

[0066] Next, in step S6 (sixth step), the control unit 42 judges whether the mapping of the degree of agreement in a predetermined range is completed by changing the assumed values ​​in the procedure from step S3 to S5. Specifically, the control unit 42 repeatedly executes the procedure from step S3 to S5 while changing the combination of the assumed values ​​(G1, G2), and calculates the degree of agreement for each combination of the assumed values ​​(G1, G2). The control unit 42 maps the degree of agreement calculated for each combination of the assumed values ​​onto a two-dimensional map. Specifically, an example of a calculation result of the degree of agreement between the frequency spectrum of the excitation vibration of the laminated iron core 21 of the transformer and the frequency spectrum of the excitation noise is shown as a two-dimensional map in FIG. 9. In the two-dimensional map in FIG. 9, the vertical axis and the horizontal axis represent G1 and G2, respectively. In FIG. 9, isopleths indicating equal degrees of agreement are drawn. In FIG. 9, of the isopleths that are closed curves, the closed curves with a small diameter correspond to isopleths with a high degree of agreement. The combination of G1 and G2 where the degree of agreement is a local maximum value corresponds to the vicinity of the center of the closed curve. The control unit 42 may map the degree of coincidence for each combination of assumption values ​​onto a two-dimensional map, and may determine that mapping is complete when a predetermined range is filled. Alternatively, the control unit 42 may determine that mapping is complete when the presence of a maximum value is detected in the two-dimensional map onto which the degree of coincidence is mapped.

[0067] If the control unit 42 does not determine that mapping is complete (step S6: NO), it returns to the procedure of step S3 and repeats the procedures from step S3 to S6 until it can determine that mapping is complete. If the control unit 42 determines that mapping is complete (step S6: YES), it proceeds to the procedure of step S7.

[0068] In step S7 (seventh step), the control unit 42 detects a point where the degree of agreement is a maximum value from the two-dimensional map. Then, the control unit 42 adopts a combination of hypothetical values ​​(G1, G2) when the degree of agreement is a maximum value as the transverse elastic modulus Gzx and Gyz in two planes including the lamination direction of each of the multiple (two) parts constituting the laminated core 21. Specifically, the control unit 42 adopts G1 as the transverse elastic modulus Gzx and Gyz of the upper yoke 22a and the lower yoke 22b. That is, the control unit 42 sets Gzx=Gyz=G1 for the upper yoke 22a and the lower yoke 22b. In addition, the control unit 42 adopts G2 as the transverse elastic modulus Gzx and Gyz of the leg 22c. That is, the control unit 42 sets Gzx=Gyz=G2 for the leg 22c. After executing the procedure of step S7, the control unit 42 ends the execution of the procedure of the flowchart in FIG. 3.

[0069] In this embodiment, while repeatedly executing the procedure from step S3 to S6, the control unit 42 ends the mapping of the degree of coincidence and changes two parameters G1 and G2 so that the maximum value of the degree of coincidence can be detected. That is, the control unit 42 needs to perform a two-dimensional numerical search. The number of search points becomes very large due to the two-dimensional search. In addition, the two-dimensional search requires a large calculation load. Therefore, in this embodiment, an optimization search method using a genetic algorithm (GA) is used by utilizing artificial intelligence (AI) technology. Specifically, as an evaluation function in the optimization search method, the degree of coincidence between the frequency spectrum of the noise of the laminated iron core 21 of the transformer obtained in step S1 and the frequency spectrum of the excitation vibration of the laminated iron core 21 of the transformer calculated in step S4 is adopted. A commercially available search software may be used as the GA optimization search tool to be used. Also, a GA optimization search tool created by the inventors may be used as well.

[0070] In this embodiment, the transverse elastic modulus is calculated assuming that the transverse elastic modulus Gyz and Gzx of the laminated core are equal. The value of Gyz and the value of Gzx may differ from each other. However, the inventors have confirmed that the value of Gyz and the value of Gzx may be assumed to be equal in a laminated core of grain-oriented electromagnetic steel sheets, not only for the laminated core illustrated in FIG. 1 but also for laminated cores having various dimensions. Therefore, in this embodiment, the value of the transverse elastic modulus Gyz and the value of Gzx are assumed to be equal to each other.

[0071] In the two-dimensional map of Fig. 9 showing an example of the degree of agreement between the frequency spectrum of the excitation vibration and the frequency spectrum of the excitation noise of the laminated core 21 of the transformer calculated in the sixth step of the method for determining the elastic matrix of the laminated core 21 of the transformer described above, the values ​​of (G1, G2) showing the maximum values ​​may be adopted as the transverse elastic coefficients Gyz and Gzx in two planes including the lamination direction of the transformer. It has been confirmed that the accuracy of the vibration analysis is improved by incorporating Gyz and Gzx adopted based on the two-dimensional map of Fig. 9 into the constitutive equation and performing the vibration analysis in the seventh step. Therefore, the accuracy of the vibration analysis of the transformer is improved by determining the elastic matrix of the laminated core 21 of the transformer by performing the procedures from the first step to the seventh step described above.

[0072] Here, a search using a GA optimization search tool, which is executed to obtain the two-dimensional map of FIG. 9 showing the degree of agreement between the frequency spectrum of excitation vibration and the frequency spectrum of excitation noise, is performed along the trajectory shown in the graph of FIG. 10. The vertical and horizontal axes of FIG. 10 correspond to G1 and G2. One circle plotted on the graph of FIG. 10 corresponds to one search point. The number of search points plotted in FIG. 10 is approximately 800.

[0073] In the sixth step, the calculation of the degree of agreement at each search point is repeated. The degree of agreement at each search point is calculated by executing the procedure from the third step to the fifth step. The vibration response analysis executed as the third step of the procedure for calculating the degree of agreement is executed, for example, using structural analysis software. The time required to execute the vibration response analysis using structural analysis software depends on the performance of the computer used, but even if a high-performance computer generally used for numerical calculation is used, about two hours are required per search point. As shown in the two-dimensional map of FIG. 9, at least about 800 search points are required to obtain a two-dimensional map with a data density sufficient to determine the maximum point of the degree of agreement between the frequency spectrum of the excitation vibration and the frequency spectrum of the excitation noise. Therefore, the time required for the vibration response analysis to obtain the result of FIG. 9 is 2 hours x 800 (points) = 1600 hours = about 67 days. In other words, if a search is simply executed using a GA optimization search tool, it will take more than two months to obtain the result of FIG. 9.

[0074] Therefore, the inventors have studied ways to shorten the time required to calculate the two-dimensional map of the degree of agreement in the above-mentioned sixth step, and have found that the time required to calculate the two-dimensional map of the degree of agreement can be shortened by including the following procedure in the sixth step of the elasticity matrix determination method according to the present disclosure.

[0075] Specifically, in a sixth step, the control unit 42 of the vibration analysis device 40 executes a sixth step, in which, when each combination of the first assumption value G1 and the second assumption value G2 for which the degree of similarity is to be calculated is expressed as a trial point for trial searching, points distributed in a two-dimensional space with the first assumption value G1 as the first axis and the second assumption value G2 as the second axis are sequentially selected as trial points, and the calculation of the degree of similarity is repeated by performing processes from the third step to the fifth step on the selected trial points.

[0076] The control unit 42 executes step 6-B, which sets a limited small area within the area containing the trial points distributed in the two-dimensional space based on the degree of similarity calculated in step 6-A, as the target for more detailed search.

[0077] The control unit 42 sets search points between trial points within the limited small area set in step 6-B and performs a search, repeats the calculation of the degree of match by performing processes from step 3 to step 5 at each search point, and performs step 6-C to calculate the combination (G1, G2) in which the degree of match is maximized within the limited small area.

[0078] As a method for setting search points within a limited small region and performing a search in step 6-C, a search method may be used that utilizes AI (artificial intelligence) technology and employs an optimization search method using GA (genetic algorithm), with the degree of agreement between the frequency spectrum of the noise of the transformer laminated core 21 obtained in step 1 and the frequency spectrum of the excitation vibration of the transformer laminated core 21 calculated in step 4 as the evaluation function.

[0079] As described above, by setting search points within the limited small region, the total number of trial points and search points required to calculate the combination of (G1, G2) at which the degree of match is maximized is significantly reduced. As a result, the time required to calculate the two-dimensional map of the degree of match and calculate the combination of (G1, G2) at which the degree of match is maximized is significantly reduced, thereby achieving high speed.

[0080] Finally, the values ​​of (G1, G2) when the degree of agreement is maximized are incorporated into the constitutive equation as the transverse elastic moduli Gyz and Gzx in the two planes including the lamination direction of the transformer, and the vibration analysis in the seventh step is performed, thereby improving the accuracy of the vibration analysis and achieving a significant reduction in the required time.

[0081] As described above, according to the method for determining the elasticity matrix of the laminated core 21 of the transformer according to this embodiment, when performing vibration analysis of the laminated core 21 made of laminated steel plates using a constitutive equation that expresses the relationship between stress and strain in a matrix, the transverse elastic modulus Gyz and Gzx in two planes including a plurality (two pieces) of different lamination directions of the laminated core 21 of the transformer, which are included in the elasticity matrix in the constitutive equation, are optimally determined. As a result, the deviation between the measured value and the calculated value of the vibration characteristics is suppressed.

[0082] (Example) Below, verification of the effects obtained by implementing the elasticity matrix determination method and vibration analysis method for a laminated iron core of a transformer according to this embodiment will be described as an example.

[0083] First, grain-oriented electromagnetic steel sheets with a thickness of 0.23 mm were prepared. Next, the prepared grain-oriented electromagnetic steel sheets were laminated to prepare a laminated core 21 for a three-phase three-legged transformer to be subjected to vibration analysis, as illustrated in FIG. 1. The steel sheet lamination thickness of the laminated core 21 was set to 100 mm. The upper yoke 22a and the lower yoke 22b had dimensions of width 100 mm x length 500 mm. The three legs 22c had dimensions of width 100 mm and core window length 300 mm. The legs 22c were connected to the upper yoke 22a and the lower yoke 22b with an interval of 100 mm. Then, an IV wire was wound around each of the three legs 22c of the prepared laminated core 21 to form an excitation coil.

[0084] A 50 Hz three-phase current was passed through the coil, and the power supply voltage was adjusted so that the core magnetic flux density was exactly 1.7 T. In this state, the excitation noise was measured using a sound level meter, similar to the procedure of step S1 in Fig. 3, to obtain the noise frequency spectrum. An example of the measurement results of the noise frequency spectrum is shown in Fig. 11. The vertical and horizontal axes of the graph in Fig. 11 are explained in the same way as the graph in Fig. 5.

[0085] Next, the magnetostriction frequency spectrum was obtained by measuring the magnetostriction of the prepared grain-oriented electrical steel sheet using a magnetostriction measuring device when the sheet was excited at an excitation frequency of 50 Hz and a magnetic flux density of 1.5 T, 1.6 T, and 1.7 T. An example of the measurement result of the frequency spectrum of magnetostriction is shown in Fig. 12. In the graph of Fig. 12, the horizontal axis represents frequency, and the vertical axis represents the magnitude of magnetostriction.

[0086] Next, a vibration response analysis was performed on the laminated core 21 of the transformer, which was the subject of the vibration analysis, using structural analysis software. Here, seven of the nine mechanical property values, namely, the longitudinal elastic moduli Ex, Ey, and Ez, the transverse elastic moduli Gyz, Gzx, and Gxy, and the Poisson's ratios νxy, νyz, and νzx, of the laminated core 21, were set as follows: Ex=132GPa, Ey=220GPa, Ez=10GPa Gxy=116GPa νxy=0.37, νyz=νzx=0 Here, x corresponds to the rolling direction of the steel sheet, y corresponds to the direction perpendicular to x, and z corresponds to the lamination direction of the steel sheet.

[0087] As for the transverse elastic modulus Gyz and Gzx in the two planes including the remaining two lamination directions, the transverse elastic modulus Gyz and Gzx in the upper yoke 22a and the lower yoke 22b constituting the laminated core 21 are different from the transverse elastic modulus Gyz and Gzx in the leg 22c. Therefore, the transverse elastic modulus Gyz=Gzx=G1 of the upper yoke 22a and the lower yoke 22b constituting the laminated core 21 is set, and the transverse elastic modulus Gyz=Gzx=G2 of the leg 22c is set (each of G1 and G2 is an assumed value selected from a predetermined range), and the frequency spectrum of the vibration response function of the laminated core 21 of the transformer for the combination of the assumed values ​​(G1, G2) was calculated. Here, the predetermined range in which each of the assumed values ​​(G1, G2) is selected is actually the range of the transverse elastic modulus expected from the structure of the laminated core 21 of the transformer, and in this embodiment, it is in the range from 0.05 GPa to 0.4 GPa.

[0088] For example, the calculation result of the frequency spectrum of the vibration response function of the laminated core 21 when the lateral elastic modulus of the upper yoke 22a and the lower yoke 22b constituting the laminated core 21 is Gyz=Gzx=G1=0.2 GPa and the lateral elastic modulus of the leg portion 22c is Gyz=Gzx=G2=0.2 GPa is shown in Fig. 13. The explanation of the vertical and horizontal axes of the graph in Fig. 13 is the same as that of the graph in Fig. 7.

[0089] Next, similarly to the procedure of step S4 in Fig. 3, the frequency spectrum of the excitation vibration of the laminated core 21 of the transformer was calculated based on the frequency spectrum data of the excitation magnetostriction of the electromagnetic steel sheet acquired in step S2 in Fig. 3 and the frequency spectrum of the vibration response function of the laminated core 21 of the transformer calculated in step S3 in Fig. 3. For example, Fig. 14 shows the calculation result of the frequency spectrum of the excitation vibration of the laminated core 21 when the transverse elastic modulus Gyz=Gzx=G1=0.2GPa of the upper yoke 22a and the lower yoke 22b constituting the laminated core 21 and the transverse elastic modulus Gyz=Gzx=G2=0.2GPa of the leg portion 22c. The explanation of the vertical and horizontal axes of the graph in Fig. 14 is the same as that of the graph in Fig. 8.

[0090] Next, similarly to the procedure of step S5 in Fig. 3, the degree of agreement was calculated between the frequency spectrum of the noise of the transformer laminated core 21 acquired in step S1 in Fig. 3 and the frequency spectrum of the excitation vibration of the transformer laminated core 21 calculated in step S4 in Fig. 3. Here, cosine agreement (Cos similarity) was calculated as the degree of agreement between the frequency spectrum of the noise of the transformer laminated core 21 and the frequency spectrum of the excitation vibration of the transformer laminated core 21.

[0091] Next, similarly to the procedure of step S6 in Fig. 3, the values ​​of the transverse elastic coefficients Gyz=Gzx=G1 of the upper yoke 22a and the lower yoke 22b and the transverse elastic coefficients Gyz=Gzx=G2 of the leg 22c were changed within the above-mentioned predetermined ranges to change the combination of the assumed values ​​(G1, G2), and the procedure of steps S3 to S5 in Fig. 3 was repeated to create a two-dimensional map showing the degree of agreement between the frequency spectrum of the excitation vibration and the frequency spectrum of the excitation noise. Fig. 15 shows an example of the two-dimensional map created in this embodiment. The vertical and horizontal axes of the graph in Fig. 15 correspond to G1 and G2.

[0092] In step S6 of Fig. 3, since there are two parameters (G1, G2), it becomes necessary to perform a two-dimensional numerical search. Therefore, the number of search points becomes very large and a great deal of effort is required. Therefore, in this embodiment, a search method is adopted in which artificial intelligence (AI) technology is utilized, an optimization search method using a genetic algorithm (GA) is used, and the degree of agreement between the frequency spectrum of the noise of the laminated core 21 of the transformer obtained in step S1 of Fig. 3 and the frequency spectrum of the excitation vibration of the laminated core 21 of the transformer calculated in step S4 of Fig. 3 is used as an evaluation function.

[0093] 15, the points where the degree of agreement is at a maximum are the transverse elastic modulus Gyz=Gzx=G1 of the upper yoke 22a and the lower yoke 22b constituting the laminated core 21 being 0.25 GPa, and the transverse elastic modulus Gyz=Gzx=G2 of the leg 22c being 0.15 GPa. In this manner, the combination of (G1, G2) where the degree of agreement is at a maximum is determined. The determined combination of specific values, (G1, G2), is adopted as the transverse elastic modulus Gzx and Gyz of the upper yoke 22a and the lower yoke 22b, and the transverse elastic modulus Gzx and Gyz of the leg 22c, respectively.

[0094] The frequency spectrum of the excitation vibration of the laminated core 21 calculated for the combination (G1, G2) where the transverse elastic modulus Gyz = Gzx = G1 = 0.25 GPa of the upper yoke 22a and the lower yoke 22b constituting the laminated core 21 and the transverse elastic modulus Gyz = Gzx = G2 = 0.15 GPa of the leg 22c was determined as the final calculated value of the frequency spectrum of the excitation vibration of the laminated core 21 of the transformer.

[0095] 16 is a graph comparing the calculated value of the frequency spectrum of the excitation vibration of the laminated core of the transformer in the embodiment with the measured value of the frequency spectrum of the noise obtained by measuring the excitation noise of the laminated core of the transformer. The tendency of the calculated value of the frequency spectrum matches very well with the tendency of the measured value. Therefore, it was confirmed that the transverse elastic modulus Gzx and Gyz in two planes including the lamination direction of the upper yoke 22a and the lower yoke 22b of the laminated core 21 of the transformer, which are included in the elasticity matrix in the constitutive formula, and the transverse elastic modulus Gzx and Gyz in two planes including the lamination direction of the legs 22c of the laminated core 21 of the transformer, were calculated with high accuracy.

[0096] Fig. 17 shows the trajectory of search points that were searched using an optimization search method using a genetic algorithm (GA) to obtain the two-dimensional map of the degree of match shown in Fig. 15. The total number of search points shown in Fig. 17 is 832. In this case, it took two months (eight weeks) to complete the search.

[0097] Therefore, we considered reducing the total number of search points to increase the speed. First, as shown in FIG. 18A, points distributed at equal intervals of 0.1 GPa for each of G1 and G2 on a two-dimensional map are set as trial points and selected sequentially. In the 6-A step, the calculation of the degree of agreement is repeated by performing the processes from the 3rd step to the 5th step on the selected trial points. The two-dimensional map of the degree of agreement obtained by performing the 6-A step on the trial points shown in FIG. 18A is shown in FIG. 18B. In the two-dimensional map in FIG. 18B, the correct answer points with G1 being 0.25 GPa and G2 being 0.15 GPa are not included in the area where the degree of agreement is 0.9 or more. Therefore, it can be seen that even if 0.1 GPa is set as the interval between the trial points, the interval is too coarse.

[0098] Next, as shown in FIG. 19A, points equally distributed at 0.075 GPa intervals for each of G1 and G2 on a two-dimensional map are set as trial points and selected sequentially. In the 6-A step, the calculation of the degree of agreement is repeated by performing the processes from the 3rd step to the 5th step on the selected trial points. The two-dimensional map of the degree of agreement obtained by performing the 6-A step on the trial points shown in FIG. 19A is shown in FIG. 19B. In the two-dimensional map of FIG. 19B, the correct answer points where G1 is 0.25 GPa and G2 is 0.15 GPa are not included in the area where the degree of agreement is 0.9 or more. Therefore, it can be seen that even if 0.075 GPa is set as the interval between trial points, the interval is too coarse.

[0099] Next, as shown in FIG. 20A, points equally distributed at 0.05 GPa intervals for each of G1 and G2 on a two-dimensional map are set as trial points and selected sequentially. In step 6-A, the selected trial points are subjected to the processing from step 3 to step 5 to repeatedly calculate the degree of agreement. FIG. 20B shows a two-dimensional map of the degree of agreement obtained by performing step 6-A on the trial points shown in FIG. 20A. In the two-dimensional map of FIG. 20B, the correct answer points where G1 is 0.25 GPa and G2 is 0.15 GPa are included in the area where the degree of agreement is 0.9 or more.

[0100] If the correct point is included in the area where the degree of agreement is 0.9 or more, in step 6-B, the area in FIG. 20B where the degree of agreement is 0.9 or more is designated as a limited small area to be searched by an optimization search method using GA (genetic algorithm). In step 6-C, in the limited small area, a search is performed using an optimization search method using GA (genetic algorithm) with the degree of agreement between the frequency spectrum of the noise of the laminated iron core of the transformer obtained in step 1 and the frequency spectrum of the excitation vibration of the laminated iron core of the transformer calculated in step 4 as an evaluation function, and the maximum point of the degree of agreement is calculated. As a result of performing steps up to step 6-C, it was confirmed that the value of (G1, G2) when the degree of agreement is maximum is (0.25GPa, 0.15GPa).

[0101] The number of trial points, which were evenly distributed at 0.05 GPa intervals and used to calculate the degree of agreement in step 6-A, was 8 x 8 = 64. In addition, the number of search points when an optimization search was performed using GA (genetic algorithm) for the limited small region in step 6-C was 100. Therefore, the total number of trial points and search points was 164. As a result, the time required to perform the search was reduced to about two weeks, which is about one-quarter of the time required in the case of 832 search points shown in Figure 17.

[0102] Furthermore, as shown in FIG. 21A, points equally distributed at 0.025 GPa intervals for each of G1 and G2 on the two-dimensional map are set as trial points and selected sequentially. In the 6-A step, the calculation of the degree of agreement is repeated by performing the processes from the 3rd step to the 5th step on the selected trial points. The two-dimensional map of the degree of agreement obtained by performing the 6-A step on the trial points shown in FIG. 21A is shown in FIG. 21B. In the two-dimensional map of FIG. 21B, the correct answer points where G1 is 0.25 GPa and G2 is 0.15 GPa are included in the area where the degree of agreement is 0.9 or more.

[0103] If the correct point is included in the area where the degree of agreement is 0.9 or more, in step 6-B, the area in FIG. 21B where the degree of agreement is 0.9 or more is designated as a limited small area to be searched by an optimization search method using GA (genetic algorithm). In step 6-C, in the limited small area, a search is performed using an optimization search method using GA (genetic algorithm) with the degree of agreement between the frequency spectrum of the noise of the laminated iron core of the transformer obtained in step 1 and the frequency spectrum of the excitation vibration of the laminated iron core of the transformer calculated in step 4 as an evaluation function, and the maximum point of the degree of agreement is calculated. As a result of performing steps up to step 6-C, it was confirmed that the value of (G1, G2) when the degree of agreement is maximum is (0.25GPa, 0.15GPa).

[0104] The number of trial points, which were evenly distributed at 0.025 GPa intervals and used to calculate the degree of match in step 6-A, was 15 x 15 = 225 points. In step 6-C, the number of search points was 50 when an optimization search was performed using a genetic algorithm (GA) for the limited small region. Therefore, the total number of trial points and search points was 275 points. As a result, the time required to perform the search was reduced to about three weeks, which is about one-third of the time required in the case of 832 search points shown in FIG. 17.

[0105] From the above results, it was found that in order to significantly reduce the time required to calculate the degree of match of each search point in step 6 and calculate a two-dimensional map of the degree of match, it is extremely effective to change the process of step 6 to a process including steps 6-A, 6-B, and 6-C.

[0106] However, in step 6-A, when the values ​​of G1 and G2 corresponding to the trial points are set as points distributed at equal intervals within a predetermined range that is the range of the transverse elastic modulus expected from the structure of the laminated core 21 of the actual transformer, by setting the trial points at coarse intervals, for example, 0.075 GPa or more, the correct point where G1 is 0.25 GPa and G2 is 0.15 GPa is not included in the region where the degree of agreement is 0.9 or more, as shown in Figures 18B and 19B, and a correct result is not obtained. Therefore, it is necessary to set the trial points at fine intervals, for example, 0.05 GPa or less.

[0107] On the other hand, even if the trial points are set at finer intervals such as 0.025 GPa, the calculation accuracy of the value of (G1, G2) when the degree of agreement is maximized does not improve. Instead, the time required for the 6-A step increases, which does not lead to a reduction in the time required for the entire 6th step. The reason why the calculation accuracy does not improve even if the intervals are too fine is thought to be that many trial points are set in the range away from the correct answers of G1 and G2 within the specified range in which G1 and G2 are selected, which is wasted. Therefore, in this embodiment, when the division interval is set to 0.05 GPa, it is optimal for realizing both the improvement of the calculation accuracy of G1 and G2 and the speed-up of shortening the time required for the entire 6th step.

[0108] Although the embodiments of the present disclosure have been described based on the drawings and examples, it should be noted that those skilled in the art can make various modifications or alterations based on the present disclosure. Therefore, it should be noted that these modifications or alterations are included in the scope of the present disclosure. For example, the functions included in each component or each step can be rearranged so as not to be logically inconsistent, and multiple components or steps can be combined into one or divided. The embodiments of the present disclosure can also be realized as a program executed by a processor included in the device or a storage medium on which a program is recorded. It should be understood that these are also included in the scope of the present disclosure.

[0109] For example, the method of evaluating the degree of agreement between the frequency spectrum of the noise of the transformer laminated core 21 and the frequency spectrum of the excitation vibration of the transformer laminated core 21 is not limited to calculating the cosine agreement. It is also possible to adopt Euclidean distance as a method of evaluating the degree of agreement by using a dB display standard that does not include an offset. In addition, other methods of evaluating the degree of agreement can also be adopted as long as they essentially quantitatively indicate the degree of agreement between the frequency spectrum data of the excitation noise and the excitation vibration.

[0110] In addition, in the above embodiment, the vibration analysis of a three-phase, three-limbed transformer has been described, but the present invention is not limited to this and can also be applied to the vibration analysis of a laminated core in a three-phase, five-limbed transformer or other transformers.

[0111] In addition, in the elastic matrix determination method according to this embodiment, the transverse elastic moduli of the multiple (two) parts constituting the laminated core 21 are different. As a specific example, a case has been described in which the transverse elastic moduli of the upper yoke 22a and lower yoke 22b (first part) constituting the laminated core 21 are different from the transverse elastic moduli of the leg 22c (second part). However, it goes without saying that this also includes the case in which the transverse elastic moduli of the upper yoke 22a and lower yoke 22b constituting the laminated core 21 are the same as the transverse elastic moduli of the leg 22c.

[0112] In addition, in the present embodiment, a configuration has been described in which the laminated core 21 having a first portion and a second portion is the analysis target. The laminated core 21 may have three or more portions. In other words, it can be said that the laminated core 21 has at least a first portion and a second portion. When the laminated core 21 has three or more portions, three or more assumed values ​​of the modulus of transverse elasticity are assumed, such as a third assumed value in addition to the first assumed value and the second assumed value. The maximum value of the degree of agreement is detected in a three- or more-dimensional map.

[0113] In the above embodiment, the trial points are distributed at equal intervals within a predetermined range, but they may be distributed at irregular intervals. Also, in the above embodiment, the trial points are distributed so that the intervals of the first assumption values ​​G1 and the intervals of the second assumption values ​​G2 match, but they may be distributed so that the intervals of the first assumption values ​​G1 and the intervals of the second assumption values ​​G2 differ.

[0114] The positions at which the trial points are set may be determined based on the results of analyzing other transformers using the elasticity matrix determination method or vibration analysis method according to the present embodiment. For example, the positions at which the trial points are set may be determined based on the tendency of the distribution of the degree of agreement grasped by statistically processing the analysis results of other transformers in advance. In this way, the trial points can be set so that the correct point at which the degree of agreement is maximized is found with a small number of trial points. Note that, since the analysis results of other transformers have been processed in advance, the time required to set the trial points based on the analysis results of other transformers is not long.

[0115] The search points may be set between the trial points. The search points may be set at intervals closer than the intervals between the trial points. The search points may be set at positions outside the grid on which the trial points are arranged. The search points may be set at positions independent of the grid on which the trial points are arranged.

[0116] The elasticity matrix determination method for the transformer laminated core 21 may be realized as an elasticity matrix determination program executed by a computer or processor constituting the control unit 42 of the vibration analysis device 40. The vibration analysis method for the transformer laminated core 21 may be realized as a vibration analysis program executed by a computer or processor constituting the control unit 42 of the vibration analysis device 40. The elasticity matrix determination program or the vibration analysis program may be stored in a non-transitory computer-readable medium. [Explanation of symbols]

[0117] 21 Laminated core 22 Grain-oriented electromagnetic steel sheet (22a: upper yoke, 22b: lower yoke, 22c: leg) 40 vibration analysis device (42: control unit, 44: memory unit, 46: interface)

Claims

1. 1. A method for determining an elasticity matrix of a laminated transformer core, comprising the steps of: The laminated core is configured by laminating a plurality of electromagnetic steel sheets, and includes at least a first portion and a second portion, a constitutive equation expressing a relationship between stress and strain in each of the first portion and the second portion includes a different elastic matrix; the elastic matrix includes, as elements of the elastic matrix, transverse elastic moduli in two planes including a lamination direction of the laminated core, A first step of acquiring a frequency spectrum of excitation noise of the laminated core of the transformer; A second step of acquiring a frequency spectrum of the excitation magnetostriction of the electromagnetic steel sheet; a third step of determining a first assumed value for a transverse elastic modulus in two planes including a lamination direction in the first portion of the laminated core of the transformer, determining a second assumed value for a transverse elastic modulus in two planes including a lamination direction in the second portion of the laminated core of the transformer, and calculating a frequency spectrum of a vibration response function of the laminated core of the transformer when the first assumed value and the second assumed value are applied to the elasticity matrix; a fourth step of calculating a frequency spectrum of excitation vibration of the laminated core of the transformer based on the frequency spectrum data of excitation magnetostriction of the electromagnetic steel sheet acquired in the second step and the frequency spectrum of the vibration response function of the laminated core of the transformer calculated in the third step; a fifth step of calculating a degree of agreement between a frequency spectrum of the noise of the laminated core of the transformer acquired in the first step and a frequency spectrum of the excitation vibration of the laminated core of the transformer calculated in the fourth step; a sixth step of repeatedly executing the third step through the fifth step by changing the first assumption value and the second assumption value, respectively, to calculate the degree of agreement for each combination of the first assumption value and the second assumption value; a seventh step of detecting a maximum value of the degree of agreement calculated for each combination of the first assumed value and the second assumed value in the sixth step, adopting the first assumed value when the degree of agreement is a maximum value as a modulus of transverse elasticity in two planes including the lamination direction of the laminated core in the first portion, and adopting the second assumed value when the degree of agreement is a maximum value as a modulus of transverse elasticity in two planes including the lamination direction of the laminated core in the second portion; Including, The sixth step is a sixth step of expressing each combination of the first assumption value and the second assumption value corresponding to the degree of agreement as a trial point set in a two-dimensional space with the first assumption value as a first axis and the second assumption value as a second axis, sequentially selecting points distributed within a predetermined range of the two-dimensional space as the trial points and calculating the degree of agreement for the combination of the first assumption value and the second assumption value corresponding to the trial points; a sixth-B step of setting a limited small region in the two-dimensional space based on the degree of coincidence calculated for a combination of the first assumption value and the second assumption value corresponding to the trial point; a 6-C step of setting search points located between the trial points in the limited small region and calculating the degree of agreement for a combination of the first assumption value and the second assumption value corresponding to the search points; A method for determining the elasticity matrix of a laminated core of a transformer, comprising:

2. 2. The method for determining an elasticity matrix of a transformer laminated core according to claim 1, wherein in the fifth step, the degree of agreement is calculated by calculating a cosine value of an angle between a first vector having elements that are frequency components of a frequency spectrum of noise of the transformer laminated core, and a second vector having elements that are frequency components of a frequency spectrum of excitation vibration of the transformer laminated core.

3. 3. A vibration analysis method comprising: performing a vibration analysis of the laminated core of a transformer based on a constitutive equation expressing a relationship between stress and strain in each of the first portion and the second portion, the constitutive equation incorporating a transverse elastic modulus determined by the method for determining an elastic matrix of a laminated core of a transformer according to claim 1 or 2.

4. An elasticity matrix determining program that causes a computer to execute the elasticity matrix determining method according to claim 1 or 2.

5. A vibration analysis program that causes a computer to execute the vibration analysis method according to claim 3.

Citation Information

Patent Citations

  • Analyzer, analysis method, and computer program

    JP2014071689A

  • Elastic matrix determination method of laminated iron core of transformer and vibration analysis method

    JP2021135285A

  • Elastic matrix determination method and vibration analysis method for laminate core of transformer

    WO2023013208A1