1. Patent Search
  2. Details

ELECTROMAGNETIC FIELD ANALYSIS METHOD FOR ANISOTROPIC CONDUCTIVE MATERIAL

20170284964 · 2017-10-05

    Inventors

    Cpc classification

    International classification

    Abstract

    An electromagnetic field analysis method for an anisotropic conductive material involves using an analysis grid having a first side and a second side that are orthogonal to each other to analyze an electromagnetic property of an anisotropic conductive material in which conductivity in a first direction is different from conductivity in a second direction. One or both of the first direction and the second direction are parallel to a direction different from either one of the first side and the second side of the analysis grid. One electromagnetic field component located on the first side and extending along the second side is calculated based on electromagnetic field components that are located on a plurality of the second sides surrounding the one electromagnetic field component and that extend along the second sides.

    Claims

    1. An electromagnetic field analysis method for an anisotropic conductive material, wherein an analysis grid having a first side and a second side that are orthogonal to each other is used to analyze an electromagnetic property of an anisotropic conductive material in which conductivity in a first direction is different from conductivity in a second direction, wherein one or both of the first direction and the second direction are parallel to a direction different from either one of the first side and the second side of the analysis grid, and wherein one electromagnetic field component located on the first side and extending along the second side is calculated based on electromagnetic field components that are located on the second sides surrounding the one electromagnetic field component and that extend along the second sides.

    2. The electromagnetic field analysis method according to claim 1, wherein the one electromagnetic field component is calculated as an average value of the electromagnetic field components.

    3. The electromagnetic field analysis method according to claim 1, wherein in a case where the one electromagnetic field component is located on the first side extending along a peripheral edge of the anisotropic conductive material, the one electromagnetic field component is calculated based on any of the electromagnetic field components included in the anisotropic conductive material.

    4. The electromagnetic field analysis method according to one of claims 2, wherein in a case where the one electromagnetic field component is located on the first side extending along a peripheral edge of the anisotropic conductive material, the one electromagnetic field component is calculated based on any of the electromagnetic field components included in the anisotropic conductive material.

    5. The electromagnetic field analysis method according to claim 1, wherein the analysis grid is a cubical analysis grid extending along three orthogonal axes, and wherein the electromagnetic field components comprise four electromagnetic field components respectively located on four of the second sides coupled to the first side at two nodal points of opposite ends of the first side on which the one electromagnetic field component is located.

    6. The electromagnetic field analysis method according to claim 2, wherein the analysis grid is a cubical analysis grid extending along three orthogonal axes, and wherein the electromagnetic field components comprise four electromagnetic field components respectively located on four of the second sides coupled to the first side at two nodal points of opposite ends of the first side on which the one electromagnetic field component is located.

    7. The electromagnetic field analysis method according to claim 3, wherein the analysis grid is a cubical analysis grid extending along three orthogonal axes, and wherein the electromagnetic field components comprise four electromagnetic field components respectively located on four of the second sides coupled to the first side at two nodal points of opposite ends of the first side on which the one electromagnetic field component is located.

    8. The electromagnetic field analysis method according to claim 4, wherein the analysis grid is a cubical analysis grid extending along three orthogonal axes, and wherein the electromagnetic field components comprise four electromagnetic field components respectively located on four of the second sides coupled to the first side at two nodal points of opposite ends of the first side on which the one electromagnetic field component is located.

    9. The electromagnetic field analysis method according to claim 1, wherein the electromagnetic field analysis method uses a finite-difference time-domain method, wherein the analysis grid is a cubical analysis grid extending along three orthogonal axes, wherein a conductivity tensor obtained when conductivity defined in a coordinate system of the three orthogonal axes is rotated about an axis orthogonal to both of the first direction and the second direction is applied to Ampere's expression and is subsequently discretized by using a finite difference method, and wherein when performing calculation repeatedly by applying an iterative method to the discretized expression, an undefined electromagnetic field component included in the discretized expression is calculated as the one electromagnetic field component from the electromagnetic field components.

    10. The electromagnetic field analysis method according to claim 2, wherein the electromagnetic field analysis method uses a finite-difference time-domain method, wherein the analysis grid is a cubical analysis grid extending along three orthogonal axes, wherein a conductivity tensor obtained when conductivity defined in a coordinate system of the three orthogonal axes is rotated about an axis orthogonal to both of the first direction and the second direction is applied to Ampere's expression and is subsequently discretized by using a finite difference method, and wherein when performing calculation repeatedly by applying an iterative method to the discretized expression, an undefined electromagnetic field component included in the discretized expression is calculated as the one electromagnetic field component from the electromagnetic field components.

    11. The electromagnetic field analysis method according to claim 3, wherein the electromagnetic field analysis method uses a finite-difference time-domain method, wherein the analysis grid is a cubical analysis grid extending along three orthogonal axes, wherein a conductivity tensor obtained when conductivity defined in a coordinate system of the three orthogonal axes is rotated about an axis orthogonal to both of the first direction and the second direction is applied to Ampere's expression and is subsequently discretized by using a finite difference method, and wherein when performing calculation repeatedly by applying an iterative method to the discretized expression, an undefined electromagnetic field component included in the discretized expression is calculated as the one electromagnetic field component from the electromagnetic field components.

    12. The electromagnetic field analysis method according to claim 4, wherein the electromagnetic field analysis method uses a finite-difference time-domain method, wherein the analysis grid is a cubical analysis grid extending along three orthogonal axes, wherein a conductivity tensor obtained when conductivity defined in a coordinate system of the three orthogonal axes is rotated about an axis orthogonal to both of the first direction and the second direction is applied to Ampere's expression and is subsequently discretized by using a finite difference method, and wherein when performing calculation repeatedly by applying an iterative method to the discretized expression, an undefined electromagnetic field component included in the discretized expression is calculated as the one electromagnetic field component from the electromagnetic field components.

    13. The electromagnetic field analysis method according to claim 1, wherein the anisotropic conductive material is fiber reinforced plastic in which plastic is reinforced with conductive fiber extending in either one of the first direction and the second direction.

    14. The electromagnetic field analysis method according to claim 2, wherein the anisotropic conductive material is fiber reinforced plastic in which plastic is reinforced with conductive fiber extending in either one of the first direction and the second direction.

    15. The electromagnetic field analysis method according to claim 3, wherein the anisotropic conductive material is fiber reinforced plastic in which plastic is reinforced with conductive fiber extending in either one of the first direction and the second direction.

    16. The electromagnetic field analysis method according to claim 4, wherein the anisotropic conductive material is fiber reinforced plastic in which plastic is reinforced with conductive fiber extending in either one of the first direction and the second direction.

    Description

    BRIEF DESCRIPTION OF THE DRAWINGS

    [0017] FIGS. 1A and 1B illustrate calculation methods for determining an undefined electromagnetic field component from a plurality of defined electromagnetic field components; and

    [0018] FIG. 2A illustrates an analytic model in an analytic example, FIG. 2B illustrates an analysis result obtained in accordance with an analysis method in the related art, and FIG. 2C illustrates an analysis result obtained in accordance with an electromagnetic field analysis method according to an implementation.

    DETAILED DESCRIPTION

    [0019] An implementation of the present invention will be described below with reference to the drawings.

    [0020] In an electromagnetic field analysis method for an anisotropic conductive material (simply referred to as “electromagnetic field analysis method” hereinafter) according to this implementation, the electromagnetic property, such as the electric current distribution, of a material having anisotropic conductivity is calculated by using a finite-difference time-domain (FDTD) method. The FDTD method is a numerical analysis method that involves calculating electromagnetic field components along three orthogonal axes by using a cubical analysis grid (Yee grid) in which the respective sides extend along three orthogonal xyz axes.

    [0021] The anisotropic conductive material being analyzed has anisotropic conductivity in which the conductivity in a certain direction differs from that in another direction. In this implementation, the anisotropic conductive material is fiber reinforced plastic (FRP) in which plastic is reinforced with conductive fibers. In such FRP, the conductivity in the fiber direction is higher than that in another direction, so that electric current flows readily in the fiber direction.

    [0022] In the electromagnetic field analysis method according to this implementation, an analysis technique using anisotropic-conductivity tensor representation is employed.

    [0023] This analysis technique will be described in detail below.

    [0024] In view of a state where the object being analyzed is rotated about an axis orthogonal to the fiber direction, the anisotropic conductivity in the direction different from the three orthogonal axes can be derived by using a relational expression of a current density J and an electric field E expressed with expression (1) below and rotating the current density and the electric field about the z-axis by an angle θ, as in expression (2) below. Each underlined part in expression (3) obtained by modifying expression (2) is an off-diagonal element of a conductivity tensor. This off-diagonal element is zero when θ=0.

    [00001] J = σ .Math. .Math. E ( 1 ) [ cos .Math. .Math. θ sin .Math. .Math. θ - sin .Math. .Math. θ cos .Math. .Math. θ ] [ J x J y ] = [ σ p 0 0 σ t ] [ cos .Math. .Math. θ sin .Math. .Math. θ - sin .Math. .Math. θ cos .Math. .Math. θ ] [ E x E y ] ( 2 ) [ J x J y ] = [ σ p .Math. cos 2 .Math. θ + σ t .Math. sin 2 .Math. θ ( σ p - σ t ) .Math. cos .Math. .Math. θsin .Math. .Math. θ _ ( σ p - σ t ) .Math. cos .Math. .Math. θsin .Math. .Math. θ _ σ t .Math. cos 2 .Math. θ + σ p .Math. sin 2 .Math. θ ] [ E x E y ] ( 3 )

    [0025] In this case, σ.sub.p denotes the conductivity in the fiber direction, and σ.sub.t denotes the conductivity in a direction orthogonal to the fiber direction within the x-y plane.

    [0026] The conductivity of the material having anisotropy in any direction within the x-y plane can be expressed with expression (4) below, assuming that components σ.sub.xx, σ.sub.yy, and σ.sub.zz correspond to the respective axial directions and off-diagonal elements mentioned above are denoted by σ.sub.xy, and σ.sub.yx.

    [00002] σ = [ σ xx σ xy 0 σ yx σ yy 0 0 0 σ zz ] ( 4 )

    [0027] In order to directly solve Maxwell's equation in the FDTD method, expression (6) below is obtained by applying expression (4) to Ampere's expression (5) in Maxwell's equation.

    [00003] ∇ × H = .Math. .Math. ∂ E ∂ t + σ .Math. .Math. E ( 5 ) { ∂ H z ∂ y - ∂ H y ∂ z = .Math. .Math. ∂ E x ∂ t + σ xx .Math. E x + σ xy .Math. E y ∂ H x ∂ z - ∂ H z ∂ x = .Math. .Math. ∂ E y ∂ t + σ yx .Math. E x + σ yy .Math. E y ∂ H y ∂ x - ∂ H x ∂ z = .Math. .Math. ∂ E z ∂ t + σ zz .Math. E z ( 6 )

    [0028] In expression (6), each parameter is satisfied at any one point in a three-dimensional space. However, in the definition of a Yee grid in the FDTD method, discretized components E.sub.x, E.sub.y, and E.sub.z exist at different positions in the analysis grid. The electric field in the Yee grid is located in an area corresponding to the sides of the grid, and the components E.sub.x, E.sub.y, and E.sub.z are located on different sides (i.e., sides extending along the respective axes) of each grid. In discretization, the position of the electric field has to be clarified.

    [0029] By denoting the position of the electric field as “|i, j, k” and discretizing expression (6) using a finite difference method, expression (7) to expression (9) below are obtained.

    [00004] E x n .Math. .Math. i + 1 2 , j , k .Math. = k 1 .Math. .Math. x .Math. E x n - 1 .Math. i + 1 2 , j , k + k xy ( E y n .Math. .Math. i + 1 2 , j , k .Math. + E y n - 1 .Math. i + 1 2 , j , k ) + k 2 .Math. .Math. x .Math. H z n - 1 .Math. .Math. i + 1 2 , j + 1 2 , k .Math. - H z n - 1 .Math. i + 1 2 , j - 1 2 , k Δ .Math. .Math. y + k 2 .Math. .Math. x .Math. H y n - 1 .Math. .Math. i + 1 2 , j , k + 1 2 .Math. - H y n - 1 .Math. i + 1 2 , j , k - 1 2 Δ .Math. .Math. z ( 7 ) E y n .Math. .Math. i , j + 1 2 , k .Math. = k 1 .Math. .Math. y .Math. E y n - 1 .Math. i , j + 1 2 , k + k yx ( E x n .Math. .Math. i , j + 1 2 , k .Math. + E x n - 1 .Math. i , j + 1 2 , k ) + k 2 .Math. .Math. y .Math. H x n - 1 .Math. .Math. i , j + 1 2 , k + 1 2 .Math. - H x n - 1 .Math. i , j + 1 2 , k - 1 2 Δ .Math. .Math. z + k 2 .Math. .Math. y .Math. H z n - 1 .Math. .Math. i + 1 2 , j + 1 2 , k .Math. - H z n - 1 .Math. i - 1 2 , j + 1 2 , k Δ .Math. .Math. x ( 8 ) E z n .Math. i , j , k = k 1 .Math. .Math. z .Math. E z n - 1 .Math. i , j , k + k 2 .Math. .Math. z .Math. H y n - 1 .Math. .Math. i + 1 2 , j , k + 1 2 .Math. - H y n - 1 .Math. i - 1 2 , j , k + 1 2 Δ .Math. .Math. x + k 2 .Math. .Math. z .Math. H x n - 1 .Math. .Math. i , j + 1 2 , k + 1 2 .Math. - H x n - 1 .Math. i , j - 1 2 , k + 1 2 Δ .Math. .Math. y ( 9 )

    [0030] In expression (7) and expression (8), there is a mixture of electric fields at different positions. Specifically, in a Yee grid, there is a component E.sub.x (referred to as “undefined component E.sub.x” hereinafter) located on a side where the component E.sub.x is not defined (i.e., the side where a component E.sub.y is defined along the y-axis), and a component E.sub.y (referred to as “undefined component E.sub.y” hereinafter) located on a side where the component E.sub.y is not defined (i.e., the side where the component E.sub.x is defined along the x-axis). Therefore, the calculations cannot be performed using the expressions in these states.

    [0031] With respect to this problem, the analysis technique in the related art simultaneously uses expression (7) and expression (8), assuming that these electric fields at different positions are identical to each other, as in expression (10) below.

    [00005] { E x n .Math. i , j + 1 2 , k = E x n .Math. i + 1 2 , j , k E y n .Math. i + 1 2 , j , k = E y n i , j + 1 2 , k ( 10 )

    [0032] However, defining these electric fields at different positions as being identical to each other leads to reduced approximation accuracy.

    [0033] As illustrated in FIG. 1A, the electromagnetic field analysis method according to this implementation does not use the assumption based on expression (10) above but approximates the value of the undefined component E.sub.x (indicated by a solid arrow in FIG. 1A) as an average value of four components E.sub.x (indicated by voided arrows in FIG. 1A), which are components E.sub.x that are located on sides where the components E.sub.x are defined (referred to as “defined components E.sub.x” hereinafter) and that are located on four sides surrounding the undefined component E.sub.x. Furthermore, the same applies to the component E.sub.y. Specifically, as illustrated in FIG. 1B, similar to the case of the component E.sub.x, the value of the undefined component E.sub.y is approximated as an average value of four components E.sub.y, which are components E.sub.y that are located on sides where the components E.sub.y are defined (referred to as “defined components E.sub.y” hereinafter) and that are located on four sides surrounding the undefined component E.sub.y.

    [0034] In detail, the undefined component E.sub.x and the undefined component E.sub.y are calculated using expression (11) and expression (12) below.

    [00006] E x n .Math. i , j + 1 2 , k = E x n .Math. i + 1 2 , j , k .Math. + E x n .Math. i + 1 2 , j + 1 , k .Math. + E x n .Math. i - 1 2 , j , k .Math. + E x n .Math. i - 1 2 , j + 1 , k 4 ( 11 ) E y n .Math. i + 1 2 , j , k = E y n .Math. i , j + 1 2 , k .Math. + E y n .Math. i + 1 , j + 1 2 , k .Math. + E y n .Math. i , j - 1 2 , k .Math. + E y n .Math. i + 1 , j - 1 2 , k 4 ( 12 )

    [0035] However, if the undefined component E.sub.x is located on a side extending along a peripheral edge of the object being analyzed (i.e., the boundary with the peripheral space), only two defined components E.sub.x (within the boundary), included in the object being analyzed, of the four defined components E.sub.x to be averaged out are used to calculate the undefined component E.sub.x, and the values of the two remaining defined components E.sub.x included in the peripheral space are not used.

    [0036] For instance, in FIG. 1A, in a case where the right analysis grid is included in the object being analyzed and the left analysis grid belongs to the peripheral space, only two defined components E.sub.x of i+1/2 are used for calculating the undefined component E.sub.x.

    [0037] The same applies to the undefined component E.sub.y.

    [0038] The undefined component E.sub.x may be calculated based on a plurality of surrounding defined components E.sub.x, and the method of interpolating the undefined component E.sub.x from the positions and the number of defined components E.sub.x and from a plurality of defined components E.sub.x is not limited to that described above.

    [0039] If expression (11) and expression (12) above are substituted into expression (7) and expression (8), there would be eight unknowns for E.sub.x and E.sub.y. By applying them to various locations on the analysis grids and simultaneously applying them in the entire analytic space to configure a matrix, an analysis is possible.

    [0040] However, in this implementation, an iterative method, as typified by the Gauss-Seidel method, is applied as a more efficient method, instead of simultaneously using expression (7) and expression (8), such that the calculation is performed repeatedly until the rates of change of the components E.sub.x and E.sub.y become sufficiently small (lower than or equal to a predetermined threshold value). Then, during this repeated calculation process, the undefined components E.sub.x and E.sub.y are interpolated, as appropriate, with a plurality of components E.sub.x and E.sub.y by using expression (11) and expression (12).

    [0041] Consequently, the electromagnetic property of the anisotropic conductive material can be analyzed.

    Analytic Example

    [0042] Next, an analytic example using the electromagnetic field analysis method according to this implementation will be described.

    [0043] In this analytic example, the electromagnetic field analysis method according to this implementation is compared with the above-described analysis method in the related art to study the validity of the two methods.

    Analytic Model

    [0044] An analytic model used for performing the comparison is a unidirectional-material flat plate, as illustrated in FIG. 2A. This flat plate has a size of 300 mm by 300 mm by 10 mm, and the number of mesh is 21×21×1.

    Analysis Conditions

    [0045] A metal rod defined by a single line on a grid is coupled to an object being analyzed and an absorption boundary so as to define the metal rod as an electrode. An applicator electrode and a return electrode are coupled to opposite ends of the flat plate on a diagonal line thereof that is parallel to the fiber direction.

    [0046] The applied current has a peak value of 3 kA in a waveform of component A for an aircraft lightning test regulated under SAE ARP-5416.

    [0047] Although the current density is calculated in this analytic example, the current density to be determined in the electromagnetic field analysis method according to this implementation has to be calculated in view of the effect of an off-diagonal element of a conductivity tensor, as indicated in expression (13) and expression (14) below.


    J.sub.x=σ.sub.xxE.sub.x+σ.sub.xyE.sub.y  (13)


    J.sub.y=σ.sub.yxE.sub.x+σ.sub.yyE.sub.y  (14)

    [0048] Because the electric field in a Yee grid is located in an area corresponding to the sides of each grid, the current density, which is the calculation result described above, is a value at each side of the grid. The current density in each cell is calculated as an average value of current densities surrounding the cell.

    Analysis Result

    [0049] FIGS. 2B and 2C illustrate results of this analytic example and illustrate the current density after 6.4 μs at which the applied current reaches the peak value. Specifically, FIG. 2B corresponds to the analysis method in the related art, whereas FIG. 2C corresponds to the electromagnetic field analysis method according to this implementation.

    [0050] Each analysis result indicates the electric current distribution displayed by filling in the cells formed by the analysis grid using color gradation. In order to display the direction in which the electric current flows in each cell, the vector directions formed by components J.sub.x and J.sub.y in expression (12) and expression (13) are indicated by arrows.

    [0051] As illustrated in FIG. 2B, in the analysis method in the related art, the current direction in each cell is parallel to the fiber direction, but current concentration caused by anisotropy cannot be calculated.

    [0052] In contrast, as illustrated in FIG. 2C, in the electromagnetic field analysis method according to this implementation, strong electric current distribution is calculated on the diagonal line of the flat plate in a manner such as to couple the two electrodes to each other. Thus, it can be considered that electric current distribution according to the anisotropic conductivity in the fiber direction is reproduced.

    Effects

    [0053] Accordingly, in the electromagnetic field analysis method according to this implementation, an undefined component E.sub.x located on a side extending along the y-axis of the analysis grid is calculated based on a plurality of defined components E.sub.x surrounding the undefined component E.sub.x and located on sides extending along the x-axis, and an undefined component E.sub.y is similarly calculated based on a plurality of defined components E.sub.y.

    [0054] In other words, the undefined components E.sub.x and E.sub.y are calculated by interpolation based on the plurality of surrounding defined components E.sub.x and E.sub.y.

    [0055] Therefore, as compared with the related art in which it is simply defined that the electric fields at different positions are identical to each other, an electromagnetic field analysis of an anisotropic conductive material having anisotropic conductivity can be performed with high accuracy.

    Modifications

    [0056] An implementation to which the present invention is applicable is not limited to the above implementation, and modifications are permissible, as appropriate, so long as they do not depart from the scope of the invention.

    [0057] For instance, although the anisotropic conductive material to be analyzed is fiber reinforced plastic in the above implementation, the anisotropic conductive material according to the implementation of the present invention is not particularly limited so long as the conductivity in a predetermined first direction is different from that in another direction, and does not have to be fiber reinforced plastic.