Design assistance method, design assistance device, program, and recording medium

Abstract

A design assistance method according to the present disclosure comprises: first topology optimization in which input functions including a basis function are used, and first weighting factors in the basis function are optimized; first selection in which a select weighting factor is selected from the first weighting factors generated in the process of the first topology optimization; principal component analysis of the select weighting factor; second selection in which a number principal components less than the total number of principal components are selected as select principal components; and second topology optimization in which the sum of the products of the select principal components and second weighting factors associated with the respective select principal components are defined as an input function.

Description

Design support method, design support device, program, and recording medium

[0001] The present disclosure relates to a design support method for a design object, a design support device, a program for executing the design support method, and a recording medium on which the program is recorded.

[0002] Patent Document 1 discloses a rotor design method for a rotating machine. The design method described in Patent Document 1 is a method for designing a rotor including magnets and an iron core, and includes a variable determination step, an image generation step, a mapping step, and an optimization step.

[0003] In the variable determination step, a design domain for the entire rotor is defined, and a first design variable related to the dimensions of the magnets and a second design variable related to the shape of the core are determined independently. In the image generation step, a rotor mesh is generated according to the first and second design variables. In the mapping step, the shape defined in the design domain is mapped onto the mesh. In the optimization step, the first and second design variables are modified in accordance with an evaluation of the mapped shape, thereby optimizing the shape of the core. As an optimization method, for example, a density method, which is one of topology optimization methods, is used.

[0004] Japanese Patent Application Laid-Open No. 2023-136975

[0005] In the design method described in Patent Document 1, it is difficult to extract the characteristic shape of the object to be designed from the overall shape derived by optimization.

[0006] An object of the present disclosure is to provide a design support method, a design support device, a program, and a recording medium that can more easily extract a characteristic shape of a design object from an overall shape derived by optimization.

[0007] A design support method according to the present disclosure includes: performing a first topology optimization using an input function that expresses a shape of a design object using basis functions, and optimizing a plurality of first weighting coefficients in the basis functions so as to maximize or minimize an objective function; performing a first selection that selects some of the plurality of first weighting coefficients generated in the process of the first topology optimization as a plurality of selected weighting coefficients; performing principal component analysis of the plurality of selected weighting coefficients; performing a second selection that selects, from a plurality of principal components generated by the principal component analysis, a number of principal components that is less than the total number of the plurality of principal components as selected principal components, wherein a contribution rate of each selected principal component in the principal component analysis is higher than a contribution rate of principal components excluding the selected principal component; and performing a second topology optimization using, as an input function, a sum of products of the selected principal components and second weighting coefficients corresponding to each selected principal component.

[0008] The program according to the present disclosure is a program for causing a calculation unit to execute the design support method.

[0009] A recording medium according to the present disclosure is a recording medium on which the program is recorded.

[0010] a calculation unit connected to the input unit and the output unit, the calculation unit performing a first topology optimization, a principal component analysis, and a second topology optimization, wherein the calculation unit: receives, via the input unit, an input function used in the first topology optimization and expressing a shape of a design object using basis functions, and an objective function used in the first topology optimization; performs the first topology optimization by optimizing a plurality of first weighting coefficients in the basis functions so as to maximize or minimize the objective function; performs a first selection by selecting some of the plurality of first weighting coefficients generated in the process of the first topology optimization as a plurality of selected weighting coefficients; performs principal component analysis of the plurality of selected weighting coefficients; and performs a second selection by selecting, from a plurality of principal components generated by the principal component analysis, principal components whose number is less than the total number of the plurality of principal components as selected principal components, the second selection being such that a contribution rate of each selected principal component in the principal component analysis is higher than a contribution rate of principal components other than the selected principal component; performing the second topology optimization using, as an input function, a sum of products of the selected principal components and second weighting coefficients corresponding to each selected principal component; and outputting a result of the second topology optimization by the output unit.

[0011] According to the present disclosure, it is possible to provide a design support method, a design support device, a program, and a recording medium that can more easily extract a characteristic shape of an object to be designed from an overall shape derived by optimization.

[0012] FIG. 2 is a perspective view showing an example of a power transmission coil component that is a design object. FIG. 3 is a perspective view showing a magnetic core in the power transmission coil component of FIG. 1. FIG. 4 is a block diagram showing a design support apparatus according to an embodiment of the present disclosure. FIG. 5 is a flowchart of a design support method according to an embodiment of the present disclosure. FIG. 6 is a schematic diagram showing a virtual design domain of a power transmission coil component and a power receiving coil component. FIG. 7 is a graph showing a transition of power supply efficiency in the first optimization process. FIG. 8 is a diagram showing shapes derived by optimization of a power transmission coil component and a power receiving coil component. FIG. 9 is a graph showing singular value distribution of principal component analysis. FIG. 10 is a diagram showing shapes corresponding to first to third principal components of a power transmission coil component and a power receiving coil component.

[0013] <Findings that Form the Basis of the Present Disclosure> Topology optimization is an analytical method for identifying a shape that maximizes (or minimizes) a specific function in the design of a design object such as an industrial product. In topology optimization, an objective is set, such as minimizing weight or maximizing rigidity. This method is performed independently of the designer's ideas by repeatedly performing simulations in a virtual design space in line with the objective. This makes it possible to derive a new shape that is not bound by existing concepts.

[0014] In topology optimization, the degree of freedom of the derived shape increases with the number of variables representing the shape of the design object. Therefore, it is desirable to increase the number of variables so that new shapes can be derived more easily. On the other hand, as the number of variables increases, the complexity of the derived shape increases, making it difficult to extract the characteristic shape of the design object from the derived overall shape. Here, a characteristic shape refers to a shape of the design object that particularly contributes to the objective set in the optimization, and is hereinafter also referred to as a "contributing shape." In particular, the ON / OFF method, which is one topology optimization method, is more likely to derive shapes that are difficult to extract contributing shapes from, such as shapes separated into multiple parts, compared to other density methods.

[0015] During the development process of a design object, designers may modify the shape derived by topology optimization, taking into account factors such as the feasibility of mass production and product durability. However, as mentioned above, if topology optimization derives a shape that is difficult to extract a contributing shape, the designer may end up modifying the contributing shape that should be maintained. In this case, the effect intended by the optimization may not be fully exerted in the product.

[0016] Therefore, the inventors discovered that by performing principal component analysis on the weighting coefficients generated during the topology optimization process and then performing topology optimization again using a new function including some of the principal components as an input function, it is possible to derive a shape whose contributing shape is easier to understand. This makes it easier to extract the contributing shape from the overall shape derived by optimization. Based on this novel finding, the inventors have come up with the following disclosure.

[0017] Hereinafter, embodiments of the present disclosure will be described with reference to the drawings. In the following description, terms indicating specific directions or positions (e.g., terms including "up," "down," "right," and "left") will be used as necessary. However, the use of these terms is intended to facilitate understanding of the present disclosure with reference to the drawings, and the meanings of these terms do not limit the technical scope of the present disclosure. Furthermore, the following description is merely exemplary in nature and is not intended to limit the present disclosure, its applications, or its uses.

[0018] <Embodiment> A design support method and a design support device according to an embodiment of the present disclosure will be described. In this embodiment, design objects designed by the design support method are a power transmission coil component and a power receiving coil component used in contactless power transfer. In the following description, the power transmission coil component and the power receiving coil component may be collectively referred to as "coil components."

[0019] An example of the configuration of a power transmission coil component that is an object to be designed will be described with reference to Figures 1 and 2. Figure 1 is a perspective view showing an example of a power transmission coil component that is an object to be designed. Figure 2 is a perspective view showing a magnetic core in the power transmission coil component of Figure 1. For convenience of explanation, an X-Y-Z Cartesian coordinate system is shown in Figures 1 and 2 and Figures 7 and 9 described below, but this coordinate system is intended to facilitate understanding of the present disclosure and does not limit the present disclosure.

[0020] As shown in Fig. 1, the power transmission coil component 1 includes a magnetic core 2 and a power transmission coil 3. As shown in Fig. 2, the magnetic core 2 includes a base 21, a pillar portion 22 extending from the base 21 in the Z direction, a cap portion 23 connected to the pillar portion 22, and a wall portion 24 extending upward from an outer edge portion 211 of the base 21. The Z direction is, for example, the up-down direction. As shown in Fig. 1, the power transmission coil 3 is provided around the pillar portion 22.

[0021] The magnetic core 2 is made of a magnetic material such as a sintered ferrite body or a molded body of a resin containing magnetic powder. In this embodiment, the base 21, the column 22, the cap 23, and the wall 24 are integrally formed. In Figures 1 and 2, the boundary between the base 21 and the wall 24 is indicated by a dashed line.

[0022] 2, the base 21 has a flat plate shape having a thickness along the Z direction. In a plan view seen from the Z direction, the shape of the base 21 is, for example, circular, elliptical, or polygonal. In this embodiment, the shape of the base 21 in a plan view is circular.

[0023] The column portion 22 has a base end portion 221 connected to the base portion 21 and a tip end portion 222 located on the opposite side of the base end portion 221. The column portion 22 has, for example, a circular, elliptical, or polygonal cross section. The area of ​​the cross section may be constant or may vary over the entire length of the column portion 22 along the Z direction. In this embodiment, the column portion 22 is a cylinder having a constant diameter over the entire length.

[0024] The cap portion 23 is connected to the tip portion 222. The cap portion 23 has a flat plate shape with a thickness along the Z direction. In a plan view, the cap portion 23 protrudes outward from the column portion 22. The shape of the cap portion 23 in a plan view is, for example, a circle, an ellipse, or a polygon. In this embodiment, the shape of the cap portion 23 in a plan view is a circle that is smaller than the base portion 21.

[0025] As shown in FIG. 1 , in this embodiment, the base 21 , the cap 23 , the wall 24 , and the power transmission coil 3 are concentrically positioned around an imaginary central axis A1 extending in the Z direction of the column 22 .

[0026] In the present embodiment, the wall portion 24 is provided around the entire periphery of the outer edge portion 211. The height of the wall portion 24 in the Z direction is constant around the entire periphery of the wall portion 24. An upper edge portion 241 of the wall portion 24 is located lower in the Z direction than an upper edge portion 31 of the cap portion 23 and the power transmission coil 3.

[0027] 1 has a cylindrical shape formed around the pillar portion 22. In this embodiment, the power transmission coil 3 is a conductive wire wound around the pillar portion 22.

[0028] The configuration of the magnetic core 2 is not limited to the above configuration, except for the provision of the pillar portion 22 that is surrounded by the power transmission coil 3 when viewed from the Z direction. In other words, the presence or absence and shape of each of the base portion 21, the cap portion 23, and the wall portion 24, as well as the shape of the pillar portion 22, can be determined arbitrarily depending on the electromagnetic properties and the like required of the power transmission coil component 1. Furthermore, the power transmission coil 3 only needs to have a cylindrical shape, and the dimensions of the power transmission coil 3, such as its height in the Z direction and its width in a plan view, can be determined arbitrarily.

[0029] The power receiving coil component (not shown) paired with the power transmitting coil component 1 has the same configuration as the power transmitting coil component 1. That is, the power receiving coil component includes a magnetic core having a pillar portion and a power receiving coil provided around the pillar portion. However, the shapes of the magnetic core and the power receiving coil may be different from the shapes of the magnetic core 2 and the power transmitting coil 3 provided in the power transmitting coil component 1.

[0030] When the power transmitting coil component 1 and the power receiving coil component are arranged facing each other, the coils provided in each coil component are magnetically coupled to each other. Power supplied to the power transmitting coil 3 is transmitted to the power receiving coil via this magnetic coupling. This realizes contactless power transfer between the coil components.

[0031] In the following description, the position of the coil component that maximizes the efficiency of contactless power transfer is referred to as the "normal position" of the coil component. On the other hand, when the coil component is positioned in a position that is shifted from the normal position, it is referred to as "misalignment." For example, the power receiving coil component may be positioned in a position that is shifted in the opposing direction of the coil components (e.g., the Z direction shown in FIGS. 1 and 2 ) and in an intersecting direction that intersects the opposing direction.

[0032] A design support device according to an embodiment of the present disclosure will be described with reference to Fig. 3. Fig. 3 is a block diagram showing the design support device according to an embodiment of the present disclosure. The design support device 100 according to the present embodiment can be used to support the design of a power transmission coil component and a power receiving coil component to which power is supplied from the power transmission coil component.

[0033] 3, the design support device 100 includes an input unit 101, an output unit 102, a storage unit 103, and a calculation unit 104. For example, the design support device 100 is a personal computer (desktop computer, laptop computer), a mobile terminal (smartphone, tablet terminal, etc.), etc.

[0034] The input unit 101 has, for example, one or more human-machine interfaces (HMIs) for inputting information. The HMIs of the input unit 101 are, for example, input interfaces such as a keyboard, a pointing device (mouse, trackball, etc.), a touchpad, or a position input device such as a touch panel display. The HMIs of the input unit 101 may be built into the design support device 100 or may be externally attached. In other words, the input unit 101 may include both an HMI of the design support device 100 itself and an HMI connected to the design support device 100. In this embodiment, the input unit 101 is a keyboard and a mouse. Alternatively, the input unit 101 may be an interface for inputting information from a storage medium provided outside the design support device 100.

[0035] The output unit 102 includes one or more HMIs for outputting information. The HMI of the output unit 102 is, for example, an output interface such as a display, a speaker, or a display device such as a touch panel display. The HMI of the output unit 102 may be built into the design support device 100 or may be externally attached. In other words, the output unit 102 may include both an HMI of the design support device 100 itself and an HMI connected to the design support device 100. In this embodiment, the output unit 102 is a display. Alternatively, the output unit 102 may be an interface for outputting information to a storage medium provided outside the design support device 100.

[0036] The storage unit 103 is used to store information used by the design support device 100 and information generated by the design support device 100. The storage unit 103 includes one or more storages (non-transitory storage media). The storage may be, for example, a hard disk drive, an optical drive, or a solid-state drive. The storage may also be an internal storage, an external storage, or a network-attached storage (NAS) storage.

[0037] The calculation unit 104 is connected to the input unit 101 and the output unit 102 and can access the storage unit 103. The calculation unit 104 is realized, for example, by a computer system. The computer system includes one or more processors (microprocessors) and one or more internal memories (e.g., RAMs (Random Access Memories)). The one or more processors execute programs to realize various functions of the design support device 100. Here, the programs to be executed are stored, for example, in one or more internal memories or the storage unit 103. The programs may be pre-recorded in the storage unit 103, or may be provided via a telecommunications line such as the Internet, or recorded on a non-transitory recording medium such as a memory card. In this embodiment, the storage unit 103 stores a program P that is executed by the calculation unit 104 to realize the functions of the design support device 100. The calculation unit 104 may also realize functions similar to those achieved by the program P using wired logic such as an ASIC (Application Specific Integrated Circuit).

[0038] A design support method according to an embodiment of the present disclosure will be described with reference to FIG. 4 . FIG. 4 is a flowchart of the design support method according to an embodiment of the present disclosure. For example, the design support method is performed in the design support device 100 by the calculation unit 104 executing a program P stored in the storage unit 103. The design support method shown in FIG. 4 uses topology optimization and dimensional optimization to derive an optimal shape for a coil component. Note that the design of the coil component aims to improve power supply efficiency when misalignment occurs.

[0039] Specifically, the calculation unit 104 mainly executes a first topology optimization (S2), a principal component analysis (S4), and a second topology optimization (S6). Both the first and second topology optimizations are performed using the ON / OFF method.

[0040] First, the calculation unit 104 receives input of design information necessary for design support via the input unit 101 (S1). The design information includes an input function and an objective function for the first topology optimization. In this embodiment, in addition to the input function and the objective function, the design information also includes constraints for the first topology optimization, a threshold value for power supply efficiency used in the first selection described below, and a lower limit value for the cumulative contribution rate used in the second selection described below. Details of the input function, objective function, constraints, threshold value for power supply efficiency, and lower limit value for the cumulative contribution rate will be described later.

[0041] When design information is input via the input unit 101, the calculation unit 104 executes a first topology optimization so as to maximize or minimize an objective function (S2).

[0042] The optimization of the design of the coil components will be described with reference to FIG. 5 . FIG. 5 is a schematic diagram showing a virtual design domain for a power transmitting coil component and a power receiving coil component. As shown in FIG. 5 , this design support method simultaneously performs two optimizations: a first topology optimization of two magnetic cores and a dimensional optimization of two coils. In the following description, the first topology optimization and dimensional optimization may be collectively referred to as the "first optimization."

[0043] A first topology optimization of two magnetic cores will be described. The input function in the first topology optimization is a function for expressing the shapes of the two magnetic cores, which are the design objects, by defining the distribution of material in a virtual design space. In this embodiment, a normalized Gaussian network (NGnet), which is an example of a basis function, is used as the input function. The input function is expressed by the following equation (1):

[0044] In equation (1), x is the input vector, N is the number of Gaussian basis functions, and w ｉis a combination weight of the Gaussian basis function. The combination weight is an example of the "first weighting coefficient" in this disclosure. The first weighting coefficients take continuous values ​​between -1 and +1, independent of each other. FIG. 5 shows a schematic sphere representing the Gaussian basis function in the design space. b in Equation (1) ｉ (x) is defined by the following equations (2) and (3).

[0045] In equations (2) and (3), D is the input vector dimension, μ ｋ is the central vector of Gauss k, Σ ｋ is the covariance matrix of Gaussian k.

[0046] The first topology optimization is performed by adjusting the first weighting coefficients included in the input function (i.e., the coupling weights w of the Gaussian basis functions). ｉ ) to derive the shape of the magnetic core that maximizes the objective function. That is, the calculation unit 104 repeatedly performs trials to maximize the objective function while varying the first weighting coefficients corresponding to each Gaussian basis function. In this embodiment, the number of Gaussian basis functions in the input function is 108. Therefore, the number of variables in the first topology optimization is 108, which is the number of first weighting functions.

[0047] In the coil size optimization, the calculation unit 104 repeatedly performs trials to maximize the objective function while varying predefined size parameters of the transmitting coil and the receiving coil. As shown in Fig. 5, the shape of each coil to be subjected to size optimization is defined as a cylindrical shape centered on the imaginary central axis A1 (see Fig. 1). Furthermore, each coil is defined to have a rectangular cross section shown in Fig. 5 in a cross section including the imaginary central axis (for example, the YZ cross section in Fig. 1).

[0048] The dimensional parameters to be optimized are the width W of the rectangular cross section, the height H of the rectangular cross section, and the x-coordinate C of the center of the rectangular cross section. ｘ and the y coordinate C of the center of the rectangular cross section ｙThese four dimension parameters are set for each of the transmitting coil and the receiving coil. Thus, the number of variables in the coil dimension optimization is eight. Therefore, the number of variables in the entire first optimization is 116, which is the sum of the number of variables in the first topology optimization (108) and the number of variables in the dimension optimization (8).

[0049] The objective function in the first optimization is a value to be maximized or minimized in the optimization. In this embodiment, the objective is to maximize the power supply efficiency when a positional misalignment occurs. The "power supply efficiency" is, for example, the value obtained by dividing the output power from a power receiving device provided with a power receiving coil by the input power to a power transmitting device provided with a power transmitting coil. The objective function is expressed by the following equation (4).

[0050] In equation (4), η ａ (p) is the power supply efficiency when the coil component is in the normal position, and η ｍ (p) is the power supply efficiency when the coil components are shifted from the normal position by a predetermined distance. For example, the distance between the coil components in the normal position is 15 in the Z direction (see FIG. 1 ) and 0 in the transverse direction intersecting the Z direction. On the other hand, the distance between the coil components in the shifted position is, for example, 25 in the Z direction and 10 in the transverse direction intersecting the Z direction.

[0051] The constraint condition in the first optimization is the condition expressed by the following equation (5).

[0052] Here, V Ｌｏａｄ is the output voltage from the receiving coil when the input voltage to the transmitting coil is 100-500V.

[0053] In each trial of the first optimization, 116 variables including 108 first weighting coefficients and 8 dimensional parameters are generated, and an evaluation value and a selection value of the shape are calculated based on the first weighting coefficients and the dimensional parameters. Here, the evaluation value is a value calculated by the objective function in each trial, and in this embodiment, it is the sum of the power supply efficiency at the normal position and the power supply efficiency at the shifted position. The selection value is a value calculated by a selection function included in the objective function, and is used in the first selection described below. In this embodiment, the selection function is the power supply efficiency at the shifted position (η in Equation (4)). ｍ (p)) The selection function may be a part of the objective function or may be the same as the objective function.

[0054] For example, the first optimization may be terminated when the 116 variables have converged to a predetermined degree, or when a predetermined number of trials have been reached. At the end of the first optimization, a set of 116 variables, including the 108 first weighting coefficients and the 8 dimension parameters, evaluation values, and selection values ​​has been generated for the number of trials.

[0055] 6 is a graph showing the transition of power supply efficiency during the first optimization process. In the example shown in FIG. 6, the power supply efficiency at the normal position and the offset position generally increases as the trials are repeated for approximately 100 trials from the start of the first optimization. Thereafter, as the trials are repeated, the power supply efficiencies converge to the values ​​at the end of the first optimization, albeit with some fluctuation. The power supply efficiency at the normal position is higher than the power supply efficiency at the offset position in almost all trials.

[0056] 7 is a diagram showing shapes derived by optimization of the power transmitting coil component and the power receiving coil component. The first topology optimization calculates optimal values ​​for the first weighting coefficient 108 and each of the dimensional parameters 8. FIG. 7 shows the shapes of each coil component derived based on these optimal values.

[0057] 4, when the first optimization is completed, the calculation unit 104 executes a first selection in which some of the variables (i.e., the first weighting coefficients and the dimension parameters) generated in the first optimization process are selected as "selected variables" (S3). The selected variables are subjected to principal component analysis, which is executed after the first selection.

[0058] Here, the first selection includes selecting some of the first weighting coefficients generated in the first topology optimization process as "selected weighting coefficients." That is, the variables among the selected variables that correspond to the first weighting coefficients are the selected weighting coefficients. Furthermore, the selected weighting coefficients are subjected to principal component analysis as part of the selected variables.

[0059] In this embodiment, the calculation unit 104 selects, as the selected variables, first weighting coefficients and dimension parameters corresponding to some of the trials of the first optimization. For example, when maximizing the objective function in the first optimization, the calculation unit 104 selects, as the selected variables, variables in trials of all the trials whose selected values ​​are equal to or greater than a predetermined threshold. That is, the calculation unit 104 selects, as the selected weighting coefficients, first weighting coefficients in trials of all the trials whose selected values ​​are equal to or greater than a predetermined threshold. In the following description, trials whose selected values ​​are equal to or greater than a predetermined threshold are also referred to as "trials to be analyzed."

[0060] On the other hand, when minimizing the objective function in the first optimization, the calculation unit 104 selects, as the selected variable, a variable in a trial in which the selected value is equal to or less than a predetermined threshold value among all trials. That is, the calculation unit 104 selects, as the selected weight coefficient, the first weight coefficient in a trial in which the selected value is equal to or less than a predetermined threshold value among all trials.

[0061] In this embodiment, "80%" is input as the threshold value of the selection value (i.e., the power supply efficiency at the offset position) in the input of design information (S1). Therefore, the calculation unit 104 selects, as the selection variables, the first weighting coefficients and dimension parameters in trials of all trials whose selection values ​​are 80% or greater. In FIG. 6, the position corresponding to a power supply efficiency of 80% is indicated by a dashed line. Therefore, referring to FIG. 6, trials in which the power supply efficiency at the offset position is at the same position as the dashed line or above the dashed line are selected as the trials to be analyzed.

[0062] By narrowing down the selection variables to be subjected to the principal component analysis to variables in trials with high selection values, it is possible to analyze only trials with high selection values ​​in the steps after the first selection (S3).

[0063] 4, once the selection variables including the selection weight coefficients are selected, the calculation unit 104 performs principal component analysis on the selection variables (S4). That is, the calculation unit 104 performs principal component analysis on the selection weight coefficients. The number of principal components generated in the principal component analysis is 116, which is the same as the number of the selection variables.

[0064] When the principal component analysis is completed, the calculation unit 104 selects, as selected principal components, a number of principal components that is less than the total number of all principal components from all principal components generated by the principal component analysis (S5). Here, the contribution rate of each selected principal component in the principal component analysis is higher than the contribution rate of each of the principal components that were not selected as selected principal components. In other words, the calculation unit 104 selects, from all principal components, a number of principal components that is less than the total number of all principal components in descending order of contribution rate, and sets these as selected principal components. The number of selected principal components may be one, or may be multiple, as long as it is less than the total number of all principal components.

[0065] For example, the calculation unit 104 selects the minimum number of selected principal components so that the cumulative contribution ratio in the principal component analysis is equal to or greater than a predetermined lower limit. In this embodiment, "70%" is input as the lower limit of the cumulative contribution ratio in the input of design information (S1). Therefore, the calculation unit 104 adds up the principal components in order from the highest principal component, and selects the principal components whose cumulative contribution ratio exceeds 70% as the selected principal components. In this embodiment, the calculation unit 104 selects five principal components, the first to fifth principal components, as the selected principal components.

[0066] The cumulative contribution rate of the first to fifth principal components is 70% or more, meaning that the first to fifth principal components can represent 70% or more of the information contained in the selected variables.

[0067] 8 is a graph showing the singular value distribution of principal component analysis. In the graph shown in Fig. 8, the vertical axis represents the singular values ​​in principal component analysis, and the horizontal axis represents the number of principal components added together in descending order of contribution rate. As shown in Fig. 8, the singular values ​​obtained by adding together the first to fifth principal components are in the range of 1 to 10.

[0068] 4, when the second selection is completed, the calculation unit 104 executes the second topology optimization (S6). The input function in the second topology optimization is the sum of the products of each selected principal component and the second weighting coefficient corresponding to each selected principal component, and is expressed by the following equation (6).

[0069] In equation (6), w ｉ is a second weighting factor different from the first weighting factor, Z ｉ is the selected principal component. The second weighting coefficient w ｉ are independent of each other and take continuous values ​​between -1 and +1.

[0070] The objective function and constraints in the second topology optimization are the same as those in the first topology optimization. In the second topology optimization, the calculator 104 repeats trials to maximize the objective function while varying the second weighting factor. The second topology optimization may be terminated when the second weighting factor has converged to a predetermined degree or when a predetermined number of trials has been performed.

[0071] The number of variables in the second topology optimization is the same as the number of second weighting factors, i.e., the number of selected principal components, which is five in this embodiment. That is, the number of variables in the second topology optimization is significantly reduced compared to the first topology optimization. On the other hand, the input function in the second topology optimization is the sum of the selected principal components, each weighted by the second weighting factors, and therefore retains 70% or more of the information content of the selected variables. Therefore, by performing principal component analysis on the selected variables and performing the second topology optimization using the input function including the selected principal components, it is possible to significantly reduce the number of variables in the optimization while suppressing a reduction in the amount of information.

[0072] 4 , when the second topology optimization is completed, the calculation unit 104 causes the output unit 102 to output the results of the second topology optimization (S7). At this time, the output unit 102 may output the optimized second weighting coefficient values, or may output the shapes of each coil component derived based on the optimized second weighting coefficients. In this embodiment, the calculation unit 104 displays the shapes of each coil component derived by the second topology optimization on a display, which is an example of the output unit 102.

[0073] The shapes of the coil components derived by the second topology optimization are shown in Fig. 7. The second topology optimization has fewer variables than the first optimization, so the shapes derived by the second topology optimization are simpler than the shapes derived by the first optimization.

[0074] 4 , when or after the result of the second topology optimization is output, the calculation unit 104 receives, from the input unit 101, an instruction regarding the operation of the calculation unit 104 after the output (S8). The operations of the calculation unit 104 after the output include, for example, three operations: reducing the selected principal components and performing the second topology optimization again (Operation 1), increasing the selected principal components and performing the second topology optimization again (Operation 2), and terminating the design support (Operation 3).

[0075] In the case of Operation 1, the calculation unit 104 selects a smaller number of selected principal components than the number of selected principal components selected in the second selection (S91) and performs the second topology optimization again (S6). Operation 1 can be performed, for example, when the designer considers that it is difficult to extract a contributing shape from the shape derived by the second topology optimization. In other words, whether Operation 1 is performed or not depends on whether it is possible to extract a contributing shape from the shape derived by the second topology optimization (S9).

[0076] By reducing the number of selected principal components, the number of variables in the second topology optimization can be reduced, which makes the shape derived by the second topology optimization simpler, making it easier to extract the contributing shapes.

[0077] In the case of Operation 2, the calculation unit 104 selects a larger number of selected principal components than the number of selected principal components selected by the second selection (S101), and performs the second topology optimization again (S6). Operation 2 can be performed, for example, when the designer views the shape derived by the second topology optimization and realizes that a contributing shape can be extracted, but also wants to confirm further derived shapes. In other words, whether Operation 2 is performed or not depends on whether confirmation of other shapes derived by the second topology optimization is desired (S10).

[0078] As the number of selected principal components increases, the cumulative contribution rate achieved by the selected principal components increases. In other words, the amount of information of the selected variables considered in the second topology optimization can be increased. This makes it possible to calculate second weighting coefficients that reflect more information while reducing the number of variables in the second topology optimization. In other words, it is possible to derive shapes of the power transmitting coil component and the power receiving coil component that reflect more information.

[0079] In the case of operation 3, the calculation unit 104 ends the design support provided by the design support device 100.

[0080] The instruction for the operation is given, for example, by the designer selecting an operation option displayed on the output unit 102 (for example, a display) via the input unit 101. For example, the option may be "optimize by reducing the selected principal components," "optimize by increasing the selected principal components," or "end design support."

[0081] Alternatively, the options may not include the content of the operation to be executed by the calculation unit 104. In this embodiment, the calculation unit 104 displays three options via the output unit 102: "Contributing shape cannot be extracted" (option 1), "Contributing shape can be extracted, but further shape derivation is desired" (option 2), and "Contributing shape can be extracted, but further shape derivation is not necessary" (option 3). Furthermore, the calculation unit 104 receives an instruction for an operation by selecting one of options 1 to 3 via the input unit 101.

[0082] When option 1 is selected, the calculation unit 104 executes operation 1. For example, the calculation unit 104 reduces the number of selected principal components by one from the number of selected principal components selected in the second selection (S91), and executes the second topology optimization again (S6).

[0083] When Option 2 is selected, the calculation unit 104 executes Operation 2. For example, the calculation unit 104 increases the number of selected principal components by 1 from the number of selected principal components selected in the second selection (S101), and executes the second topology optimization again (S6).

[0084] If option 3 is selected, the calculation unit 104 ends the design support.

[0085] <Modification> If a threshold value for a selected value (e.g., power supply efficiency at a shifted position) is not input in the input of design information (S1), the calculation unit 104 accepts an input of a threshold value for the selected value via the input unit 101 between the first optimization (S2) and the first selection (S3). At this time, the calculation unit 104 may display, via the output unit 102, a table, graph (see FIG. 6 ), or the like, showing the transition of the selected value during the first optimization process. In this case, the designer can refer to the transition of the selected value during the actual first optimization and determine the threshold value for the selected value in accordance with the transition.

[0086] If the lower limit value of the cumulative contribution rate is not input in the input of design information (S1), the calculation unit 104 accepts input of the lower limit value of the cumulative contribution rate via the input unit 101 between the principal component analysis (S4) and the second selection (S5). At this time, the calculation unit 104 may display, via the output unit 102, a table, a graph, or the like showing the relationship between the number of principal components to be added together and the amount of information reflected in the added principal components. For example, the calculation unit 104 may display, via the output unit 102, a graph showing the singular value distribution (see FIG. 8 ) or a graph showing the cumulative contribution rate. In this case, the designer can refer to the singular values, cumulative contribution rates, and the like corresponding to the principal components to be added together and determine the lower limit value of the cumulative contribution rate based on these values.

[0087] The calculation unit 104 may output shapes corresponding to at least some of the multiple principal components generated by the principal component analysis via the output unit 102. For example, the calculation unit 104 may output the result of the second topology optimization (S7) via the output unit 102 and display a shape corresponding to at least the first principal component on a display. For example, as shown in Fig. 9 , the calculation unit 104 may display shapes corresponding to each selected principal component on a display via the output unit 102. Fig. 9 is a diagram showing shapes corresponding to the first to third principal components of a power transmitting coil component and a power receiving coil component.

[0088] The principal components with higher contribution rates reflect more of the information contained in the selected variables than the principal components with lower contribution rates, and therefore the shapes corresponding to the higher principal components reflect more of the features of the shape derived by the first optimization.

[0089] By outputting the shapes corresponding to the top principal components by the output unit 102, the designer can extract contributing shapes while referring to the shapes corresponding to each principal component along with the shape derived by the second topology optimization. This makes it easier to extract contributing shapes from the overall shape derived by optimization.

[0090] Furthermore, when shapes corresponding to principal components other than the selected principal component are output, the designer can refer to the shapes corresponding to the other principal components and decide whether or not to check the other shapes (S10).

[0091] The shapes corresponding to at least some of the principal components may be output at any timing after the principal component analysis (S4). For example, if the lower limit value of the cumulative contribution rate is input after the principal component analysis, the shapes corresponding to at least some of the principal components may be displayed on the display serving as the output unit 102 before the input.

[0092] The design support method described above can be realized, for example, by using a design support device 100 having an input unit 101, an output unit 102, a storage unit 103, and a calculation unit 104. Therefore, the design support device 100 can more easily extract a contributing shape from an overall shape derived by optimization. Furthermore, the design support method can be realized, for example, by the calculation unit 104 executing a program P. Therefore, the program P can more easily extract a contributing shape from an overall shape derived by optimization.

[0093] According to the design support method, a first topology optimization is first performed. The first topology optimization can derive a complex magnetic core shape because it uses a large number of variables (e.g., the number of first weighting coefficients) to represent the magnetic core shape. This makes it easier to derive a new magnetic core shape that is not bound by existing design concepts.

[0094] By performing principal component analysis on the first weighting coefficients generated in the first topology optimization process, the information on the first weighting coefficients, i.e., the information on the shape of the magnetic core, can be compressed into the principal components with the highest contribution rates.

[0095] Some principal components are selected from all principal components as selected principal components (second selection (S5)), and a second topology optimization is performed using a function including the selected principal components as an input function. Here, the number of second weighting coefficients corresponding to the selected principal components is smaller than the number of first weighting coefficients. Therefore, in the second topology optimization, optimization can be performed using fewer variables while utilizing the information on the first weighting coefficients generated in the first topology optimization process. This makes it possible to provide a design support method that makes it easy to derive a new magnetic core shape and to more easily extract a contributing shape from the overall shape derived by optimization.

[0096] According to the design support method, the selected variables to be subjected to principal component analysis are selected based on the selected values ​​from all first weighting coefficients. Therefore, the selected variables can be limited to first weighting coefficients that are advantageous for maximizing or minimizing the objective function. For example, by selecting the first weighting coefficients in trials whose selected values ​​are equal to or greater than a threshold as the selected weighting coefficients, the first weighting coefficients in trials whose selected values ​​are low can be excluded from the principal component analysis. This makes it easier for the contributing shapes to be reflected in the results of the second topology optimization, thereby more reliably deriving a shape that maximizes the objective function in the second topology optimization.

[0097] According to the design support method, the number of selected principal components is the minimum number that makes the cumulative contribution ratio in the principal component analysis equal to or greater than a predetermined lower limit, thereby reducing the number of variables in the second topology optimization while suppressing a decrease in the amount of information of the selected variables used in the second topology optimization.

[0098] According to the design support method, the calculation unit 104 terminates the design support, increases the number of selected principal components and performs the second topology optimization again, or decreases the number of selected principal components and performs the second topology optimization again, based on an instruction input via the input unit 101.

[0099] For example, if it is difficult to extract a contributing shape from the shape derived by the second topology optimization, the selected principal components are reduced and only the second topology optimization is performed again. In this case, since there is no need to redo the entire design support, the time required to derive a shape by the second topology optimization can be reduced. Therefore, compared to redoing the entire design support, a simpler shape can be easily derived.

[0100] For example, if it is desired to derive a further shape, the selected principal components are increased and only the second topology optimization is performed again. In this case, since it is not necessary to redo the entire design support, the time required to derive a shape by the second topology optimization can be reduced. Therefore, compared to redoing the entire design support, it is possible to derive a further shape more easily. By visually comparing the shape derived by the first second topology optimization with the further shape, it is possible to more easily derive a contributing shape.

[0101] The principal components with higher contribution rates reflect more of the information contained in the selected variables than the principal components with lower contribution rates, and therefore the shapes corresponding to the higher principal components reflect more of the features of the shape derived by the first optimization.

[0102] By outputting the shapes corresponding to the top principal components by the output unit 102, the designer can extract contributing shapes while referring to the shapes corresponding to each principal component along with the shape derived by the second topology optimization. This makes it easier to extract contributing shapes from the overall shape derived by optimization.

[0103] The ON / OFF method, which is one topology optimization method, tends to derive shapes that are difficult to extract contributing shapes from, such as shapes separated into multiple parts, compared to other methods such as the density method. Therefore, it is difficult to extract contributing shapes from the derived shapes.

[0104] According to the design support method, the second topology optimization can be performed using fewer variables while utilizing information on the first weighting coefficients generated in the first topology optimization process, thereby providing a design support method that can easily derive a new magnetic core shape and more easily extract a contributing shape from the overall shape derived by optimization.

[0105] According to the design support method, since the design object is a coil component used for contactless power supply, it is possible to provide a design support method for coil components that can more easily extract the characteristic shapes of each coil component from the overall shape derived by optimization.

[0106] The design support device 100 is realized by utilizing the calculation unit 104. That is, the design support method executed by the design support device 100 can be realized by the calculation unit 104 executing the program P. The program P is a computer program for causing the calculation unit 104 to execute the design support method. The program P makes it possible to more easily extract a contributing shape from the overall shape derived by optimization.

[0107] The program P for causing the calculation unit 104 to execute the design support method may be provided in a recording medium. The recording medium makes it possible to more easily extract the contributing shape from the overall shape derived by optimization.

[0108] The present disclosure is not limited to the above-described embodiment and can be embodied in various other forms. For example, in the above description, the design object is a power transmitting coil component and a power receiving coil component used in contactless power transfer, but the present disclosure is not limited to this. For example, the design object may be only one of the power transmitting coil component and the power receiving coil component. Furthermore, the design object may be only one of the magnetic core and the coil in each coil component.

[0109] For example, the design object may be a heat sink provided on a heat-generating component. When providing design support for a heat sink, a Gaussian basis function representing the shape of the heat sink may be used as an input function for the first topology optimization. The objectives of the first and second topology optimizations may be, for example, minimizing the weight of the heat sink and / or minimizing the average temperature of the heat-generating component. A constraint in the first and second topology optimizations may be, for example, that the temperature of the heat-generating component be within the rated temperature range of the heat-generating component when the heat-generating component is heated with a predetermined amount of heat.

[0110] Furthermore, for example, the design object may be the shape of a printed circuit board or shield provided in an electronic component, the shape of an electromagnet contact in a relay, the shape of an arm in a robot, the shape of a cuff for a blood pressure monitor, or the like.

[0111] Although the first optimization described above involves performing dimensional optimization of the coil along with the first topology optimization of the magnetic core, the present disclosure is not limited thereto. For example, the first optimization may involve performing only the first topology optimization of the magnetic core. In this case, the coil may be defined as having a predetermined shape in the first optimization.

[0112] Although the ON / OFF method is used in the first and second topology optimizations described above, the present disclosure is not limited to this. For example, the density method may be used in the first and second topology optimizations.

[0113] Although a normalized Gaussian function network is used as the input function for the first and second topology optimizations in the above description, the present disclosure is not limited to this. For example, a Gaussian function, a trigonometric function, or the like may be used as the input function.

[0114] Furthermore, although the analysis target trial is described above as a trial in which the selected value (i.e., the power supply efficiency at the shifted position) is 80% or more, the present disclosure is not limited to this. For example, the analysis target trial may be a trial in which the selected value is 50% or more of the maximum selected value among all trials, or a trial in which the selected value is 50% or more of the average selected value of the final generation. Furthermore, when a genetic algorithm is used in the first optimization, the analysis target trial may be a trial in which the selected value is 50% or more of the average selected value of the convergent generation. Here, the convergent generation refers to the nth generation when the difference between the average selected value of the nth generation and the average selected value of the (n-1)th generation becomes 20% or less of the average selected value of the nth generation for the first time.

[0115] In the above description, the selected principal components in the second selection (S5) are selected so that the cumulative contribution ratio is equal to or greater than a lower limit. However, the present disclosure is not limited to this. For example, the selected principal components may be the minimum number of principal components that result in a singular value of 10 or less. Furthermore, for example, if the difference between a first singular value realized by n principal components and a second singular value realized by n-1 principal components is 30% or less of the second singular value, the selected principal components may be the n principal components. Furthermore, for example, the number of selected principal components may be the minimum number of principal components that results in a slope of the derivative smaller than a predetermined value when the approximation curve of the singular values ​​is differentiated in the graph of the singular value distribution (see FIG. 8 ).

[0116] Although the number of selected principal components is determined based on the cumulative contribution rate or the singular value in the above description, the present disclosure is not limited to this. For example, in the input of design conditions (S1), an arbitrary number of selected principal components may be input via the input unit 101.

[0117] Any of the various embodiments or modifications described above can be combined appropriately to achieve the effects of each. In addition, combinations of embodiments, combinations of examples, or combinations of embodiments and examples are possible, and combinations of features from different embodiments or examples are also possible.

[0118] Although the present disclosure has been fully described in connection with the preferred embodiments with reference to the accompanying drawings, various changes and modifications will be apparent to those skilled in the art, and such changes and modifications are to be understood as being included within the scope of the present disclosure as defined by the appended claims unless they depart therefrom.

[0119] Various embodiments of the present disclosure have been described in detail above with reference to the drawings. Finally, various aspects of the present disclosure will be described. Note that in the following description, reference numerals will also be used as examples.

[0120] According to a first aspect of the present disclosure, there is provided a design support method including: performing a first topology optimization (S2) using an input function that expresses a shape of a design object using basis functions, and optimizing a plurality of first weight coefficients in the basis functions so as to maximize or minimize an objective function; performing a first selection (S3) of selecting some of the plurality of first weight coefficients generated in the process of the first topology optimization (S2) as a plurality of selected weight coefficients; performing principal component analysis (S4) of the plurality of selected weight coefficients; performing a second selection (S5) of selecting, from a plurality of principal components generated by the principal component analysis (S4), principal components in a number less than the total number of the plurality of principal components as selected principal components, wherein a contribution rate of each selected principal component in the principal component analysis (S4) is higher than a contribution rate of principal components excluding the selected principal component; and performing a second topology optimization (S6) using, as an input function, a sum of products of the selected principal components and second weight coefficients corresponding to each selected principal component.

[0121] According to a second aspect of the present disclosure, there is provided a design support method according to the first aspect, wherein in the first selection (S3), when maximizing the objective function in the first topology optimization (S2), the plurality of first weighting coefficients generated when a value of a selection function included in the objective function is equal to or greater than a predetermined threshold are selected as the selected weighting coefficients, and when minimizing the objective function in the first topology optimization (S2), the plurality of first weighting coefficients generated when a value of a selection function included in the objective function is equal to or less than a predetermined threshold are selected as the selected weighting coefficients.

[0122] According to a third aspect of the present disclosure, there is provided the design support method according to the first or second aspect, wherein the number of the selected principal components is the minimum number that makes a cumulative contribution rate in the principal component analysis (S4) equal to or greater than a predetermined lower limit value.

[0123] According to a fourth aspect of the present disclosure, there is provided a design support method according to any one of the first to third aspects, which is executed by a calculation unit (104) connected to an input unit (101), wherein the calculation unit (104) receives, via the input unit (101) after outputting (S7) a result of the second topology optimization, an input (S8) of an instruction regarding the operation of the calculation unit (104) after the output, and based on the instruction, terminates the design support, increases the number of selected principal components (S101) and executes the second topology optimization (S6) again, or decreases the number of selected principal components (S91) and executes the second topology optimization (S6) again.

[0124] According to a fifth aspect of the present disclosure, there is provided a design support method according to any one of the first to fourth aspects, which is executed by a calculation unit (104) connected to an output unit (102), and which includes the calculation unit (104) outputting shapes corresponding to at least some of the plurality of principal components via the output unit (102).

[0125] According to a sixth aspect of the present disclosure, there is provided a design support method according to any one of the first to fifth aspects, wherein the first topology optimization (S2) and the second topology optimization (S6) are performed using an ON / OFF method.

[0126] According to a seventh aspect of the present disclosure, there is provided a design support method according to any one of the first to sixth aspects, in which the design object is at least one of a power transmission coil component (1) and a power receiving coil component used in contactless power supply.

[0127] According to an eighth aspect of the present disclosure, there is provided a program (P) for causing a calculation unit (104) to execute the design support method according to any one of the first to seventh aspects.

[0128] According to a ninth aspect of the present disclosure, there is provided a recording medium on which the program (P) according to the eighth aspect is recorded.

[0129] According to a tenth aspect of the present disclosure, there is provided an apparatus for performing a first topology optimization (S2), a principal component analysis (S4), and a second topology optimization (S6), the apparatus comprising: an input unit (101); an output unit (102); and a calculation unit (104) connected to the input unit (101) and the output unit (102) and configured to perform a first topology optimization (S2), a principal component analysis (S4), and a second topology optimization (S6), wherein the calculation unit (104) receives, via the input unit (101), an input (S1) of an input function used in the first topology optimization (S2) that expresses a shape of a design object using basis functions, and an objective function used in the first topology optimization (S2); performs the first topology optimization (S2) to optimize a plurality of first weighting coefficients in the basis functions so as to maximize or minimize the objective function; and performs a first selection (S3) to select some of the plurality of first weighting coefficients generated in the process of the first topology optimization (S2) as a plurality of selected weighting coefficients. a second selection (S5) for selecting, from the plurality of principal components generated by the principal component analysis (S4), principal components whose number is less than the total number of the plurality of principal components as selected principal components, wherein the contribution rate of each selected principal component in the principal component analysis (S4) is higher than the contribution rates of principal components other than the selected principal component; a second topology optimization (S6) for using, as an input function, the sum of products of the selected principal components and second weighting coefficients corresponding to each selected principal component; and a design support device (100) for outputting (S7) a result of the second topology optimization by the output unit (102).

[0130] According to an eleventh aspect of the present disclosure, there is provided a design support device (100) according to a tenth aspect, wherein in the first selection (S3), when maximizing the objective function in the first topology optimization (S2), the calculation unit (104) selects as the selected weight coefficients the plurality of first weight coefficients generated when a value of a selection function included in the objective function is equal to or greater than a predetermined threshold, and when minimizing the objective function in the first topology optimization (S2), the calculation unit (104) selects as the selected weight coefficients the plurality of first weight coefficients generated when a value of a selection function included in the objective function is equal to or less than a predetermined threshold.

[0131] According to a twelfth aspect of the present disclosure, there is provided the design support device (100) according to the tenth or eleventh aspect, wherein the number of the selected principal components is the minimum number that makes the cumulative contribution rate in the principal component analysis (S4) equal to or greater than a predetermined lower limit value.

[0132] According to a thirteenth aspect of the present disclosure, there is provided a design support device (100) according to any one of the tenth to twelfth aspects, wherein the calculation unit (104) receives, via the input unit (101), an input (S8) of an instruction regarding the operation of the calculation unit (104) after outputting (S7) a result of the second topology optimization, and based on the instruction, terminates the design support, increases the number of selected principal components (S101) and executes the second topology optimization (S6) again, or decreases the number of selected principal components (S91) and executes the second topology optimization (S6) again.

[0133] According to a fourteenth aspect of the present disclosure, there is provided the design support device (100) according to any one of the tenth to thirteenth aspects, wherein the calculation unit (104) outputs shapes corresponding to at least some of the plurality of principal components via the output unit (102).

[0134] According to a fifteenth aspect of the present disclosure, there is provided a design support device (100) according to any one of the tenth to fourteenth aspects, wherein the calculation unit (104) performs the first topology optimization (S2) and the second topology optimization (S6) using an ON / OFF method.

[0135] According to a sixteenth aspect of the present disclosure, there is provided a design support device (100) according to any one of the tenth to fifteenth aspects, in which the design object is at least one of a power transmission coil component (1) and a power receiving coil component used for contactless power supply.

[0136] INDUSTRIAL APPLICABILITY The present disclosure is useful for designing a design object using topology optimization, since it makes it possible to more easily extract the characteristic shape of the design object from the overall shape derived by optimization.

[0137] REFERENCE SIGNS LIST 1 Power transmission coil component 2 Magnetic core 3 Power transmission coil 31 Upper edge portion 21 Base portion 211 Outer edge portion 22 Column portion 221 Base end portion 222 Tip portion 23 Cap portion 24 Wall portion 241 Upper edge portion 100 Design support device 101 Input portion 102 Output portion 103 Storage portion 104 Calculation portion

Claims

1. A design support method comprising: performing a first topology optimization using an input function that represents the shape of a design object using basis functions, and optimizing a plurality of first weighting coefficients in the basis functions so as to maximize or minimize an objective function; performing a first selection that selects some of the plurality of first weighting coefficients generated in the process of the first topology optimization as a plurality of selected weighting coefficients; performing principal component analysis of the plurality of selected weighting coefficients; performing a second selection that selects, from a plurality of principal components generated by the principal component analysis, a number of principal components that is less than the total number of the plurality of principal components as selected principal components, wherein the contribution rate of each selected principal component in the principal component analysis is higher than the contribution rates of principal components excluding the selected principal component; and performing a second topology optimization using, as an input function, the sum of the products of the selected principal components and second weighting coefficients corresponding to each selected principal component.

2. The design support method according to claim 1, wherein in the first selection, when maximizing the objective function in the first topology optimization, the plurality of first weighting coefficients generated when a value of a selection function included in the objective function is equal to or greater than a predetermined threshold are selected as the selected weighting coefficients, and when minimizing the objective function in the first topology optimization, the plurality of first weighting coefficients generated when a value of a selection function included in the objective function is equal to or less than a predetermined threshold are selected as the selected weighting coefficients.

3. The design support method according to claim 1 or 2, wherein the number of selected principal components is the minimum number that makes the cumulative contribution rate in the principal component analysis equal to or greater than a predetermined lower limit value.

4. A design support method according to any one of claims 1 to 3, which is executed by a calculation unit connected to an input unit, wherein the calculation unit receives, via the input unit, an instruction regarding the operation of the calculation unit after outputting the result of the second topology optimization, and based on the instruction, terminates the design support, increases the number of selected principal components and executes the second topology optimization again, or decreases the number of selected principal components and executes the second topology optimization again.

5. A design support method according to any one of claims 1 to 4, which is executed by a calculation unit connected to an output unit, and which includes outputting shapes corresponding to at least some of the plurality of principal components by the output unit.

6. The design support method according to any one of claims 1 to 5, wherein the first topology optimization and the second topology optimization are performed by an ON / OFF method.

7. A design support method according to any one of claims 1 to 6, wherein the object to be designed is at least one of a power transmission coil component and a power receiving coil component used in contactless power transfer.

8. A program for causing a computing unit to execute the design support method according to any one of claims 1 to 7.

9. A recording medium on which the program according to claim 8 is recorded.

10. A system comprising: an input unit; an output unit; and a calculation unit connected to the input unit and the output unit, the system performing a first topology optimization, a principal component analysis, and a second topology optimization, wherein the calculation unit: receives, via the input unit, an input function used in the first topology optimization and expressing a shape of a design object using basis functions, and an objective function used in the first topology optimization; performs the first topology optimization by optimizing a plurality of first weight coefficients in the basis functions so as to maximize or minimize the objective function; performs a first selection by selecting some of the plurality of first weight coefficients generated in the process of the first topology optimization as a plurality of selected weight coefficients; performs principal component analysis of the plurality of selected weight coefficients; and performs a second selection by selecting, from a plurality of principal components generated by the principal component analysis, principal components whose number is less than the total number of the plurality of principal components as selected principal components, wherein the contribution rate of each selected principal component in the principal component analysis is higher than the contribution rates of principal components other than the selected principal component; performing the second topology optimization using a sum of products of the selected principal components and second weighting coefficients corresponding to each selected principal component as an input function; and outputting a result of the second topology optimization by the output unit.

11. The design support device according to claim 10, wherein the calculation unit, in the first selection, when maximizing the objective function in the first topology optimization, selects as the selected weight coefficients the plurality of first weight coefficients generated when a value of a selection function included in the objective function is equal to or greater than a predetermined threshold, and when minimizing the objective function in the first topology optimization, selects as the selected weight coefficients the plurality of first weight coefficients generated when a value of a selection function included in the objective function is equal to or less than a predetermined threshold.

12. The design support system according to claim 10 or 11, wherein the number of selected principal components is the minimum number that makes the cumulative contribution rate in the principal component analysis equal to or greater than a predetermined lower limit value.

13. The design support device according to any one of claims 10 to 12, wherein the calculation unit receives, via the input unit, an input of an instruction regarding the operation of the calculation unit after outputting the result of the second topology optimization, and based on the instruction, terminates the design support, increases the number of selected principal components and executes the second topology optimization again, or decreases the number of selected principal components and executes the second topology optimization again.

14. A design support device according to any one of claims 10 to 13, wherein said calculation unit outputs shapes corresponding to at least some of said plurality of principal components by said output unit.

15. The design support device according to any one of claims 10 to 14, wherein the calculation unit performs the first topology optimization and the second topology optimization using an ON / OFF method.

16. A design support device according to any one of claims 10 to 15, wherein the object to be designed is at least one of a power transmission coil component and a power receiving coil component used in contactless power transfer.

Images