Estimating device, estimating method and estimating program

Abstract

To provide a technology capable of appropriately outputting data on an electronic product that meets a specification required by a user.SOLUTION: An estimating device 20 includes: an obtaining unit 24 that obtains specification data representing the specification of an electronic product; an estimating unit 21 that estimates a design parameter based on the specification data obtained by the obtaining unit using at least one estimation model 40 for estimating the design parameter based on the specification data; and an output unit 24 that outputs output data based on the design parameter estimated by the estimating unit.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present disclosure relates to an estimation device, an estimation method, and an estimation program for estimating design parameters for designing high-frequency components. [Background technology]

[0002] Conventionally, a technology for outputting data of electronic products that satisfy specifications required by a user is known. For example, Japanese Patent Application Laid-Open No. 2002-358449 (Patent Document 1) discloses a trading system that has a user input specifications of an electronic product, extracts data of electronic products that satisfy the specifications required by the user from a database, and outputs the data of the electronic products to a manufacturer. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2002-358449 A Summary of the Invention [Problem to be solved by the invention]

[0004] According to the trading system disclosed in Patent Document 1, it is possible to use a database to output data of electronic products that meet the specifications required by a user, but if the data of an electronic product that meets the specifications required by the user is not stored in the database, it is not possible to extract the data of that electronic product.

[0005] The present disclosure has been made to solve such problems, and its purpose is to provide a technology that can appropriately output data of an electronic product that meets specifications required by a user. [Means for solving the problem]

[0006] An estimation device according to an aspect of the present disclosure estimates design parameters for designing an electronic product. The estimation device includes an acquisition unit that acquires specification data indicating specifications of the electronic product, an estimation unit that estimates the design parameters based on the specification data acquired by the acquisition unit using at least one estimation model for estimating the design parameters based on the specification data, and an output unit that outputs output data based on the design parameters estimated by the estimation unit.

[0007] An estimation method according to another aspect of the present disclosure is a method for estimating design parameters for designing an electronic product. The estimation method includes, as a process executed by a computer, a step of acquiring specification data indicating specifications of the electronic product, a step of estimating the design parameters based on the specification data acquired by the acquiring step using at least one estimation model for estimating the design parameters based on the specification data, and a step of outputting output data based on the estimation result of the design parameters by the estimating step.

[0008] An estimation program according to another aspect of the present disclosure is a program for estimating design parameters for designing an electronic product. The estimation program causes a computer to execute the steps of acquiring specification data indicating specifications of the electronic product, estimating the design parameters based on the specification data acquired by the acquiring step using at least one estimation model for estimating the design parameters based on the specification data, and outputting output data based on the estimation result of the design parameters by the estimating step. Effect of the Invention

[0009] According to the present disclosure, an estimation model can be used to estimate design parameters for designing an electronic product based on specification data indicating the specifications of the electronic product, and output data based on the estimated design parameters can be output, thereby making it possible to appropriately output data for an electronic product that meets the specifications required by a user. [Brief description of the drawings]

[0010] [Figure 1] FIG. 1 is a diagram showing a configuration of an estimation system according to a first embodiment. [Diagram 2] FIG. 1 is a diagram showing an example of a configuration of an electronic product according to a first embodiment. [Diagram 3] FIG. 4 is a diagram showing an example of design parameters of the electronic product according to the first embodiment. [Figure 4] FIG. 4 is a diagram showing an example of S parameters of the electronic product according to the first embodiment. [Diagram 5] FIG. 2 is a diagram for explaining input data and output data in the estimation model according to the first embodiment. [Figure 6] FIG. 2 is a diagram for explaining an estimation range of the estimation model according to the first embodiment. [Figure 7] FIG. 2 is a diagram for explaining data exchange between an estimation device and a user device in the estimation system according to the first embodiment. [Figure 8] 4 is a flowchart relating to processing executed by the estimation device according to the first embodiment. [Figure 9] FIG. 11 is a diagram for explaining a coupling matrix handled by the estimation model according to the second embodiment. [Figure 10] FIG. 11 is a diagram for explaining a coupling matrix handled by the estimation model according to the second embodiment. [Figure 11] FIG. 11 is a diagram for explaining input data and output data in the estimation model according to the second embodiment. [Figure 12] FIG. 13 is a diagram for explaining an estimation range of an estimation model according to the third embodiment. [Figure 13] 13 is a flowchart relating to processing executed by an estimation device according to a third embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0011] Hereinafter, embodiments of the present disclosure will be described in detail with reference to the drawings. In the drawings, the same or corresponding parts are designated by the same reference characters and their description will not be repeated.

[0012] <Embodiment 1> An estimation system 1 according to a first embodiment will be described with reference to FIGS.

[0013] [Configuration of the estimation system] A configuration of an estimation system 1 according to the first embodiment will be described with reference to Fig. 1. Fig. 1 is a diagram showing a configuration of the estimation system 1 according to the first embodiment. As shown in Fig. 1, the estimation system 1 includes a user device 10 and an estimation device 20. In the estimation system 1 according to the first embodiment, the user device 10 acquires specification data indicating specifications of a specific electronic product input by a user and outputs the data to the estimation device 20. The estimation device 20 estimates design parameters for designing an electronic product that satisfies the specifications required by the user, based on the specification data acquired from the user device 10.

[0014] An "electronic product" includes high-frequency components such as a capacitor, an inductor, a balun, a coupler, an LC filter, a SAW (Surface Acoustic Wave) filter, a BAW (Bulk Acoustic Wave) filter, a dielectric resonator filter, a waveguide filter, or an antenna. An "electronic product" may include components other than high-frequency components. "Design parameters" include the shape or dimensions of at least one component or member that constitutes an electronic product. "Specification data" may be any data that indicates the specifications of an electronic product. For example, if an "electronic product" is a high-frequency component, the "specification data" includes the insertion loss or attenuation of a high-frequency signal in a specific frequency band.

[0015] If the "electronic product" is a capacitor, the "design parameters" include the dimensions (length, width) of the external electrodes, or the distance between multiple external electrodes, etc. As the dimensions of the external electrodes change, the inductance (L value) in the equivalent circuit of the capacitor changes, and as the distance between multiple external electrodes changes, the capacitance (C value) in the equivalent circuit of the capacitor changes, which can result in a change in the S parameters of the capacitor.

[0016] When the "electronic product" is an inductor, the "design parameters" include the number of windings, the distance between multiple windings, etc. As the number of windings changes, the inductance (L value) in the equivalent circuit of the inductor changes, and as the distance between multiple windings changes, the capacitance (C value) in the equivalent circuit of the inductor changes, which can result in a change in the S parameters of the inductor.

[0017] If the "electronic product" is a balun, coupler, or LC filter, the "design parameters" include the dimensions (length, width) of the plate electrode, or the distance between multiple plate electrodes, etc. Depending on the change in the dimensions of the plate electrode, the inductance (L value) in the equivalent circuit of the balun, coupler, or LC filter changes, and depending on the change in the distance between multiple plate electrodes, the capacitance (C value) in the equivalent circuit of the balun, coupler, or LC filter changes, which can result in a change in the S-parameter of the balun, coupler, or LC filter.

[0018] When the "electronic product" is a SAW filter, the "design parameters" include the pitch of an interdigital transducer (IDT) on the surface of the piezoelectric substrate. As the pitch of the interdigital transducer changes, the resonant frequency in the equivalent circuit of the SAW filter changes, and as a result, the S-parameters of the SAW filter may change.

[0019] When the "electronic product" is a BAW filter, the "design parameters" include the thickness of the piezoelectric substrate, or the dimensions (length, width) of the multiple plate electrodes that sandwich the piezoelectric substrate, etc. Depending on the change in the thickness of the piezoelectric substrate or the dimensions (length, width) of the multiple plate electrodes that sandwich the piezoelectric substrate, the resonant frequency in the equivalent circuit of the BAW filter changes, and as a result, the S parameters of the BAW filter may change.

[0020] When the "electronic product" is a dielectric resonator filter, the "design parameters" include the dimensions (length, width) of the dielectric, etc. Depending on the change in the dimensions (length, width) of the dielectric, the resonant frequency in the equivalent circuit of the dielectric resonator filter changes, and as a result, the S-parameters of the dielectric resonator filter may change.

[0021] When the "electronic product" is a waveguide filter, the "design parameters" include the internal dimensions of the space surrounded by the metal, etc. Depending on the change in the internal dimensions of the space surrounded by the metal, the resonant frequency in the equivalent circuit of the waveguide filter may change, and as a result, the S-parameters of the waveguide filter may change.

[0022] If the "electronic product" is an antenna, the "design parameters" include the dimensions (length, width) of the electrodes, etc. Depending on the change in the dimensions (length, width) of the electrodes, the resonant frequency in the equivalent circuit of the antenna may change, and as a result, the S parameters of the antenna may change.

[0023] In addition to the above examples, the S parameters of an electronic product can also change due to changes in the physical properties of the substrate that constitutes the electronic product (for example, the dielectric constant or dielectric tangent of a dielectric, the propagation velocity of a piezoelectric body, etc.).

[0024] The user device 10 is, for example, an information terminal such as a desktop PC (Personal Computer), a laptop PC, or a tablet PC. The user device 10 may also be a mobile terminal such as a smartphone or a smartwatch owned by a user. For example, the user device 10 may be configured to have an application provided by the estimation system 1 installed therein, and to be able to input specification data indicating the specifications of an electronic product by executing the installed application.

[0025] The user device 10 comprises a computing device 11 , a memory 12 , a storage device 13 , a communication device 14 , an input device 15 and a display 16 .

[0026] The arithmetic device 11 is a computing entity (computer) that executes various processes by executing various programs. The arithmetic device 11 is composed of a processor such as a CPU (Central Processing Unit), an MPU (Micro Processing Unit), a TPU (Tensor Processing Unit), or a GPU (Graphics Processing Unit). Note that a processor, which is an example of the arithmetic device 11, has a function of executing various processes by executing a program, but some or all of these functions may be implemented using a dedicated hardware circuit such as an ASIC (Application Specific Integrated Circuit) or an FPGA (Field Programmable Gate Array). The term "processor" is not limited to a processor in the narrow sense that executes processing in a stored program manner such as a CPU, an MPU, a TPU, or a GPU, but may include a hardwired circuit such as an ASIC or an FPGA. The arithmetic device 11 is not limited to a von Neumann type computer such as a CPU or a GPU, and may be composed of a non-von Neumann type computer such as a quantum computer or an optical computer. The arithmetic device 11 as described above can also be read as a processing circuitry that executes a predetermined process. The computing device 11 may be configured as one chip or multiple chips. Furthermore, the processor and related processing circuits may be configured as multiple computers interconnected by wire or wirelessly via a local area network or a wireless network. The processor and related processing circuits may be configured as a cloud computer that performs remote calculations based on input data and outputs the results of the calculations to another device in a remote location.

[0027] The memory 12 includes a volatile storage area (for example, a working area) that temporarily stores program codes or work memory when the arithmetic unit 11 executes various programs. An example of the memory 12 is a volatile memory such as a dynamic random access memory (DRAM) or a static random access memory (SRAM), or a non-volatile memory such as a read only memory (ROM) or a flash memory.

[0028] The storage device 13 stores various programs or various data executed by the arithmetic device 11. The storage device 13 may be one or more non-transitory computer readable media, or one or more computer readable storage media. Examples of the storage device 13 include a hard disk drive (HDD) and a solid state drive (SSD).

[0029] The communication device 14 communicates with the estimating device 20 via a network to transmit and receive data to and from the estimating device 20. For example, the communication device 14 receives output data based on design parameters of an electronic product transmitted by the estimating device 20 via the network. In addition, the communication device 24 transmits specification data required by a user input from the input device 15 to the estimating device 20 via the network.

[0030] The input device 15 is an operation unit for receiving predetermined information input by a user, and includes a mouse, a keyboard, a touch panel, etc. For example, the input device 15 receives specification data required by the user input by the user.

[0031] The display 16 displays a predetermined image according to the control of the computing device 11. For example, the display 16 displays a screen for a user to input specification data, or displays a screen for showing design parameters of an electronic product estimated by the estimation device 20.

[0032] The estimation device 20 is, for example, a cloud-type server device provided by the operating company of the estimation system 1.

[0033] The estimation device 20 includes a calculation device 21 , a memory 22 , a storage device 23 , a communication device 24 , and a data reading device 27 .

[0034] The arithmetic device 21 is a computing entity (computer) that executes various processes by executing various programs, and has the functions of an "estimation unit" and a "determination unit." The arithmetic device 21 is configured with a processor such as a CPU, an MPU, a TPU, or a GPU. Note that a processor, which is an example of the arithmetic device 21, has a function of executing various processes by executing a program, but some or all of these functions may be implemented using a dedicated hardware circuit such as an ASIC or an FPGA. The "processor" is not limited to a processor in the narrow sense that executes processing in a stored program manner, such as a CPU, an MPU, a TPU, or a GPU, but may include a hardwired circuit such as an ASIC or an FPGA. Also, the arithmetic device 21 is not limited to a von Neumann type computer such as a CPU or a GPU, but may be configured with a non-von Neumann type computer such as a quantum computer or an optical computer. The arithmetic device 21 as described above can also be read as a processing circuit that executes a predetermined process. Note that the arithmetic device 21 may be configured with one chip or multiple chips. Furthermore, the processor and associated processing circuitry may be implemented as multiple computers interconnected by wire or wirelessly, such as via a local area network or wireless network. The processor and associated processing circuitry may be implemented as a cloud computer that performs remote calculations based on input data and outputs the results of the calculations to other devices at remote locations.

[0035] The memory 22 includes a volatile storage area (for example, a working area) that temporarily stores program code or a work memory when the arithmetic unit 21 executes various programs. An example of the memory 22 is a volatile memory such as a DRAM or an SRAM, or a non-volatile memory such as a ROM or a flash memory.

[0036] The storage device 23 stores various programs or various data executed by the arithmetic device 21. For example, the storage device 23 stores an estimation program 30 and at least one estimation model 40. The storage device 23 may be one or more non-transitory computer readable media, or may be one or more computer readable storage media. An example of the storage device 23 is an HDD or SSD.

[0037] The estimation program 30 includes a program that specifies a processing procedure for the computing device 11 to use at least one estimation model 40 to estimate design parameters of an electronic product that meets the specifications required by a user based on specification data acquired from the user device 10.

[0038] The estimation model 40 is a surrogate model that estimates design parameters of an electronic product that satisfies the specifications required by the user as a result of simulation based on the specification data acquired from the user device 10.

[0039] The communication device 24 has the functions of an "acquisition unit" and an "output unit." The communication device 24 transmits and receives data to and from the user device 10 by communicating with the user device 10 via a network. For example, the communication device 24 receives specification data required by a user transmitted by the user device 10 via the network. The communication device 24 also transmits output data based on design parameters of an electronic product that satisfies the specifications required by the user estimated by the computing device 11 to the user device 10 via the network.

[0040] The data reading device 27 receives the storage medium 50 and obtains various programs or various data from the storage medium 50. The storage medium 50 includes a CD (Compact Disc), a DVD (Digital Versatile Disc), a USB (Universal Serial Bus) memory, an SD (Secure Digital) card, or the like. The estimation program 30 and the estimation model 40 may be stored in the storage device 23 in advance, or the data reading device 27 may obtain the estimation program 30 and the estimation model 40 stored in the storage medium 50 and store them in the storage device 23.

[0041] In the above-described example of the estimation system 1, the estimation device 20 and the user device 10 are separate devices, but the estimation device 20 may be a standalone device that also has the functions of the user device 10. That is, the estimation device 20 may acquire specification data input by a user from an input device, estimate design parameters based on the specification data using the estimation model 40, and display a screen showing the estimated design parameters of the electronic product on a display.

[0042] [Electronic product parameters] Parameters in the electronic product 100 according to the first embodiment will be described with reference to Fig. 2. Fig. 2 is a diagram showing an example of the configuration of the electronic product 100 according to the first embodiment. Fig. 2 shows an LC filter constituted by a plurality of inductors and a plurality of capacitors as the electronic product 100.

[0043] 2, electronic product 100, which is an LC filter, has, as design parameters, plate electrode dimensions L1 and L2 and distances C1, C2, and C3 between the multiple plate electrodes. In the following, the plate electrode dimension is also referred to as the "L dimension," and the distance between the multiple plate electrodes is also referred to as the "C dimension."

[0044] As described above, the inductance (L value) of the electronic product 100 changes in response to a change in the L dimensions L1 and L2, and the capacitance (C value) of the electronic product 100 changes in response to a change in the C dimensions C1, C2, and C3, which may result in a change in the S parameters of the electronic product 100. Therefore, in the configuration of the electronic product 100 as shown in FIG. 2, a user can manufacture an electronic product 100 that has a desired insertion loss or attenuation of a high-frequency signal in a desired frequency band by setting each of the L dimensions L1 and L2 and the C dimensions C1, C2, and C3 to an arbitrary value.

[0045] [Create an estimation model] An example of creating the estimation model 40 according to the first embodiment will be described with reference to Fig. 3 to Fig. 6. First, the creator of the estimation model 40 prepares learning data to be used for training the estimation model 40. As described below, the estimation model 40 according to the first embodiment is trained by supervised learning.

[0046] Fig. 3 is a diagram showing an example of design parameters of the electronic product 100 according to the embodiment 1. As shown in Fig. 3, the creator sets lower limits, median values, and upper limits for each of the L dimensions L1, L2 and C dimensions C1, C2, C3 of the electronic product 100 illustrated in Fig. 2. In the example of Figure 3, the creator sets the lower limit of C dimension C1 to "X11", the median of C dimension C1 to "X12", and the upper limit of C dimension C1 to "X13", the lower limit of C dimension C2 to "X21", the median of C dimension C2 to "X22", and the upper limit of C dimension C2 to "X23", the lower limit of C dimension C3 to "X31", the median of C dimension C3 to "X32", and the upper limit of C dimension C3 to "X33", the lower limit of L dimension L1 to "Y11", the median of L dimension L1 to "Y12", and the upper limit of L dimension L1 to "Y13", the lower limit of L dimension L2 to "Y21", the median of L dimension L2 to "Y22", and the upper limit of L dimension L2 to "Y23". When these L dimensions L1, L2 and C dimensions C1, C2, C3 are combined, electronic products 100 having 243 different performances can be manufactured.

[0047] The creator may generate learning data based on design parameters of previously designed electronic products 100. For example, the creator may set each of the L dimensions L1, L2 and C dimensions C1, C2, C3 of previously designed electronic products 100 as a median value, and set the lower limit and upper limit values ​​using each median value.

[0048] Fig. 4 is a diagram showing an example of S parameters of the electronic product 100 according to the embodiment 1. As shown in Fig. 4, the creator can obtain 243 S parameters from 243 types of electronic products 100 that can be configured by combinations of the L dimensions L1, L2 and the C dimensions C1, C2, C3.

[0049] In this way, the creator prepares 243 sets of learning data, each set consisting of design parameters (L dimensions L1, L2 and C dimensions C1, C2, C3) and S parameters.

[0050] Fig. 5 is a diagram for explaining input data and output data in the estimation model 40 according to embodiment 1. As shown in Fig. 4, the estimation model 40 is configured by a neural network including an input layer, an intermediate layer (hidden layer), and an output layer.

[0051] In the learning phase, design parameters included in previously prepared learning data are input to the estimation model 40. For example, in the case of the example shown in Fig. 2 and Fig. 3 above, a set of design parameters consisting of L dimensions L1, L2 and C dimensions C1, C2, C3 selected from 243 design parameters is input to the estimation model 40.

[0052] The design parameters input to the estimation model 40 are first input to the input layer and multiplied by a predetermined weight. The values ​​output from the input layer are input to the intermediate layer and further multiplied by a predetermined weight. The values ​​output from the intermediate layer are output as the estimation results of the estimation model 40 via the output layer. The estimation model 40 according to the first embodiment estimates and outputs the S-parameters of an electronic product based on the input design parameters of the electronic product.

[0053] Next, the estimation model 40 acquires the S-parameters that are paired with the input design parameters as the correct answer data from among the 243 S-parameters included in the learning data, and calculates the degree of agreement between the S-parameters estimated by the estimation model 40 and the S-parameters that are the correct answer data. Based on the calculated degree of agreement, the estimation model 40 adjusts the weights in the neural network so that the degree of agreement between the S-parameters estimated by the estimation model 40 and the S-parameters that are the correct answer data exceeds a reference value. By repeating the above-mentioned training, the estimation model 40 becomes able to estimate the S-parameters with high accuracy based on the input design parameters.

[0054] Once the estimation model 40 becomes capable of estimating the S-parameters based on the design parameters, the creator optimizes the estimation model 40 to estimate the design parameters based on the S-parameters using an inverse analysis technique. That is, by repeatedly learning the correlation between the design parameters and the S-parameters, the estimation model 40 becomes capable of estimating the design parameters for obtaining the input S-parameters based on the correlation, even when the S-parameters are input with the input and output reversed.

[0055] The learned estimation model 40 trained as described above can, when specification data such as the attenuation amount of a high-frequency signal in a specific frequency band is input, estimate design parameters for obtaining S-parameters that satisfy the specification data.

[0056] Fig. 6 is a diagram for explaining the estimation range of the estimation model 40 according to the embodiment 1. For example, as shown in Fig. 6, if a creator creates an estimation model 40 for each of a plurality of products A to D having different arrangements, numbers, values, etc. of inductors or capacitors in the procedure as described in Fig. 3 to Fig. 5, a plurality of estimation models 40A to 40D according to characteristics such as center frequency, bandwidth, and attenuation can be obtained.

[0057] That is, since the products A to D have different circuit configurations, the ranges of characteristics such as center frequency, bandwidth, and attenuation are determined. Therefore, the estimation models 40A to 40D created for the products A to D also have different estimation ranges.

[0058] [Utilization of estimation models] An example of utilization of the estimation model 40 according to the first embodiment will be described with reference to Fig. 7 and Fig. 8. Fig. 7 is a diagram for explaining data exchange between the estimation device 20 and the user device 10 in the estimation system 1 according to the first embodiment.

[0059] 7, the estimation device 20 acquires specification data of an electronic product input by a user from the user device 10. Based on the specification data acquired from the user device 10, the estimation device 20 estimates S parameters that satisfy the specification data and design parameters of the electronic product that obtain the S parameters. The estimation device 20 generates output data based on the estimated design parameters and S parameters, and outputs the output data to the user device 10.

[0060] The contents of the process executed by the estimation device 20 will be specifically described with reference to Fig. 8. Fig. 8 is a flowchart relating to the process executed by the estimation device 20 according to the first embodiment. The arithmetic device 21 of the estimation device 20 executes the estimation program 30 to execute the process of the flowchart shown in Fig. 8. In the figure, "S" is used as an abbreviation for "STEP."

[0061] 8, the estimation device 20 determines whether or not it has acquired specification data from the user device 10 (S1). If the estimation device 20 has not acquired specification data (NO in S1), it ends this process. On the other hand, if the estimation device 20 has acquired specification data (YES in S1), it inputs the specification data to each of the multiple estimation models 40 corresponding to each of the multiple electronic products (S2).

[0062] The estimation device 20 determines whether or not design parameters and S-parameters that satisfy the design criteria can be estimated by any of the multiple estimation models 40 (S3). The design criteria include design rules such as the outline size of an electronic product or rules for forming a wiring pattern.

[0063] The estimation device 20 executes the process of S4 when design parameters and S parameters that satisfy the design criteria cannot be estimated by any of the multiple estimation models 40 (NO in S3). For example, the estimation device 20 executes the process of S4 when design parameters and S parameters cannot be estimated based on the specification data acquired in S1 by using any of the multiple estimation models 40, or when the design parameters estimated by any of the multiple estimation models 40 do not satisfy the design criteria.

[0064] In the process of S4, the estimating device 20 outputs to the outside unmanufacturable data indicating that the electronic product cannot be manufactured (S4). The unmanufacturable data is an example of "output data". For example, the estimating device 20 uses the unmanufacturable data to notify the user, the sales department system, the manufacturing department system, or the like that there are no design parameters that satisfy the specification data requested by the user and the design criteria. Thereafter, the estimating device 20 ends this process.

[0065] On the other hand, if the estimation device 20 can estimate design parameters and S-parameters that satisfy the design criteria by using any one of the multiple estimation models 40 (YES in S3), the estimation device 20 estimates the design parameters and S-parameters (S5) using one of the multiple estimation models 40. The estimation device 20 stores the estimated design parameters and S-parameters in the storage device 23 (S6).

[0066] Next, the estimation device 20 outputs the estimation data including the design parameters and the S parameters to the outside (S7). The estimation data is an example of "output data". For example, the estimation device 20 uses the estimation data to notify a user, a sales department system, a manufacturing department system, or the like of design parameters that satisfy the specification data required by the user and the design criteria, and S parameters of an electronic product manufactured based on the design parameters. Note that the estimation data is not limited to both the design parameters and the S parameters, and may include only one of the data.

[0067] Next, the estimation device 20 outputs manufacturability data indicating that the electronic product can be manufactured or that the electronic product is encouraged to be manufactured (S8). The manufacturability data is an example of "output data". For example, the estimation device 20 uses the manufacturability data to notify a user, a sales department system, a manufacturing department system, or the like that an electronic product that meets specification data required by a user can be manufactured or that an electronic product will be ordered.

[0068] Next, the estimating device 20 outputs cost data indicating the cost for manufacturing the electronic product to the outside (S9). The cost data is an example of "output data". For example, the estimating device 20 calculates the cost for manufacturing the electronic product based on the estimated design parameters, and notifies the calculated cost to a user, a sales department system, a manufacturing department system, or the like, using the cost data. Thereafter, the estimating device 20 ends this process.

[0069] As described above, the estimation device 20 according to the first embodiment estimates design parameters for designing an electronic product based on specification data indicating specifications required by a user, using the estimation model 40, and outputs output data based on the estimated design parameters. In this way, the estimation device 20 can output data indicating design parameters, S parameters, manufacturability, ordering, costs, etc., for electronic products that satisfy the specifications required by the user to the user, a sales department system, a manufacturing department system, etc.

[0070] <Embodiment 2> An estimation device 20 according to the second embodiment will be described with reference to Fig. 9 to Fig. 11. Only the parts of the estimation device 20 according to the second embodiment that are different from the estimation device 20 according to the first embodiment will be described below.

[0071] In the estimation device 20 according to the second embodiment, the estimation model 40 may be configured to learn the correlation between the design parameters and the coupling matrix. Figures 9 and 10 are diagrams for explaining the coupling matrix handled by the estimation model 40 according to the second embodiment. The coupling matrix is ​​a matrix including a plurality of elements obtained from an equivalent circuit of an electronic product corresponding to the S-parameters, and is a matrix that aggregates various parameters that can be extracted from the equivalent circuit.

[0072] 9, for example, the internal configuration of the electronic product 100 includes four resonators corresponding to points 1 to 4. Furthermore, a coupled topology can be configured with six points including an external point S on the power source side (Source) and an external point L on the load side (Load), and the wiring connecting these points.

[0073] Here, the resonant frequency of the first resonator corresponding to point 1 is defined as "M 11 The resonant frequency of the second resonator corresponding to point 2 is expressed as "M 22 The resonant frequency of the third resonator corresponding to point 3 is expressed as "M 33 The resonant frequency of the fourth resonator corresponding to point 4 is expressed as "M 44 The degree of coupling (coupling coefficient) between point 1 and point 2 is expressed as "M 12 The degree of coupling (coupling coefficient) between point 1 and point 4 is expressed as "M 14 The degree of coupling (coupling coefficient) between point 2 and point 3 is expressed as "M 23 The degree of coupling (coupling coefficient) between point 3 and point 4 is expressed as "M 34 The degree of coupling between point 1 and point S (external Q value) is expressed as "M S1The degree of coupling between point 4 and point L (external Q value) is expressed as "M 4L The degree of coupling between point S and point L (input / output direct coupling) is expressed as "M SL " is expressed as

[0074] The above-mentioned M 11 , M 22 , M 33 , M 44 , M 12 , M 23 , M 34 , M 14 , M S1 , M 4L , and M. SL As an element of the matrix, a single coupling matrix can be generated.

[0075] Furthermore, since one connection matrix can be generated from one electronic product, 243 connection matrices can be generated from 243 electronic products combining L dimensions L1, L2 and C dimensions C1, C2, C3 as shown in Figure 3. For example, in Figure 10, 243 M 11 , M 22 , M 33 , M 44 , M 12 , M 23 , M 34 , M 14 , M S1 , M 4L , and M. SL It is shown that the following can be obtained. In addition, the horizontal axis of each graph shown in Fig. 10 represents sample numbers 1 to 243, and the vertical axis represents values ​​such as resonance frequency or coupling coefficient. Also, some data are omitted.

[0076] The S-parameters obtained from a single electronic product are expressed as insertion loss or attenuation depending on the frequency, so the amount of data becomes enormous. However, if the equivalent circuit of the electronic product is expressed using the above-mentioned coupling matrix instead of the S-parameters, the M 11 , M 22 , M 33 , M 44 , M 12 , M23 , M 34 , M 14 , M S1 , M 4L , and M. SL Since the electronic product can be identified based on only the image, the amount of data handled by the estimation device 20 (estimation model 40) can be reduced.

[0077] Fig. 11 is a diagram for explaining input data and output data in the estimation model 40 according to the second embodiment. As shown in Fig. 11, in the learning phase, design parameters included in learning data prepared in advance are input to the estimation model 40. For example, in the case of the example shown in Figs. 2 and 3 described above, a set of design parameters consisting of L dimensions L1, L2 and C dimensions C1, C2, C3 selected from 243 design parameters is input to the estimation model 40.

[0078] The design parameters input to the estimation model 40 are first input to the input layer and multiplied by a predetermined weight. The values ​​output from the input layer are input to the intermediate layer and further multiplied by a predetermined weight. The values ​​output from the intermediate layer are output as the estimation result of the estimation model 40 via the output layer. The estimation model 40 according to the second embodiment estimates and outputs the connection matrix of the electronic product based on the input design parameters of the electronic product.

[0079] Next, the estimation model 40 acquires the coupling matrix set with the input design parameters as the correct answer data from among the 243 coupling matrices included in the learning data, and calculates the degree of agreement between the coupling matrix estimated by the estimation model 40 and the coupling matrix that is the correct answer data. Based on the calculated degree of agreement, the estimation model 40 adjusts the weights in the neural network so that the degree of agreement between the coupling matrix estimated by the estimation model 40 and the coupling matrix that is the correct answer data exceeds a reference value. By repeating the above-mentioned training, the estimation model 40 becomes able to estimate the coupling matrix with high accuracy based on the input design parameters. Note that one engagement matrix is ​​associated with one S-parameter, and the estimation model 40 finally converts the estimated coupling matrix into the S-parameter corresponding to the coupling matrix and outputs it.

[0080] When the estimation model 40 becomes capable of estimating the coupling matrix or S-parameters based on the design parameters, the creator optimizes the estimation model 40 using an inverse analysis technique to estimate the design parameters based on the S-parameters corresponding to the coupling matrix. That is, by repeatedly learning the correlation between the design parameters and the S-parameters corresponding to the coupling matrix, the estimation model 40 becomes capable of estimating the design parameters for obtaining the input S-parameters based on the correlation, even when the S-parameters are input with the input and output reversed.

[0081] In this way, by training the estimation model 40 using the connection matrix, the creator can minimize the amount of learning data, and therefore can create the estimation model 40 in a shorter time than if the estimation model 40 were trained using S parameters.

[0082] <Embodiment 3> An estimation device 20 according to a third embodiment will be described with reference to Fig. 12 and Fig. 13. Only the parts of the estimation device 20 according to the third embodiment that are different from the estimation devices 20 according to the first and second embodiments will be described below.

[0083] Fig. 12 is a diagram for explaining the estimation range of the estimation model 40 according to the third embodiment. As shown in Fig. 12, the estimation range of each of the estimation models 40A to 40D is determined in advance according to characteristics such as the center frequency, the bandwidth, and the attenuation amount. For this reason, depending on the specification data requested by the user, the estimation device 20 may not be able to estimate the design parameters based on the specification data using any of the estimation models 40A to 40D.

[0084] Therefore, the estimation device 20 according to the third embodiment is configured to extend the estimation range of the estimation model 40 and estimate the design parameters using the extended estimation model 40 when the design parameters cannot be estimated based on the specification data requested by the user.

[0085] For example, consider the coupling matrix shown in Figure 9. The resonant frequency (M 11 , M 22 , M 33 , M 44 ) is increased by X times, the length of the wavelength corresponding to the frequency is approximately 1 / X times larger, so it is expected that the L dimensions L1, L2 and C dimensions C1, C2, C3 will also be approximately 1 / X times larger.

[0086] Therefore, the estimation device 20 expands the estimation range of the estimation model 40 by multiplying the frequency characteristics, such as the center frequency, the bandwidth, and the amount of attenuation, allowed by the estimation model 40 by X. Furthermore, the estimation device 20 estimates design parameters by inputting the specification data acquired from the user device 10 to the expanded estimation model 40, and outputs values ​​obtained by multiplying the estimated design parameters by 1 / X as estimation results.

[0087] Fig. 13 is a flowchart related to the processing executed by the estimation device 20 according to the third embodiment. The calculation device 21 of the estimation device 20 executes the estimation program 30 to execute the processing of the flowchart shown in Fig. 13. In the drawing, "S" is used as an abbreviation for "STEP." In Fig. 13, the same step numbers are given to processing similar to that shown in Fig. 8, and the description thereof may be omitted.

[0088] 13, when the estimation device 20 cannot estimate design parameters and S-parameters that satisfy the design criteria by any of the multiple estimation models 40 (NO in S3), the estimation device 20 expands the estimation range of each of the multiple estimation models 40 (S3A). For example, the estimation device 20 expands the estimation range of the estimation model 40 by multiplying by X the frequency characteristics such as the center frequency, bandwidth, and attenuation allowed by the estimation model 40.

[0089] The estimation device 20 determines whether or not the design parameters and S-parameters that satisfy the design criteria can be estimated by any of the multiple extended estimation models 40 (S3B).

[0090] If the estimation device 20 cannot estimate design parameters and S-parameters that satisfy the design criteria using any of the multiple estimation models 40 (NO in S3B), it outputs to the outside unmanufacturable data indicating that the electronic product cannot be manufactured (S4).

[0091] On the other hand, if the estimation device 20 can estimate design parameters and S-parameters that satisfy the design criteria using any of the multiple estimation models 40 (YES in S3B), it estimates the design parameters and S-parameters by inputting the specification data acquired in S1 into the extended estimation model 40, and outputs values ​​that are 1 / X times the estimated design parameters as estimation results.

[0092] As described above, the estimation device 20 according to the third embodiment extends the estimation range of the estimation model 40 and estimates design parameters using the extended estimation model 40, and is therefore able to estimate design parameters for a wide range of specification data requested by the user.

[0093] <Aspects> (Item 1) An estimation device (20) according to one embodiment includes an acquisition unit (24) that acquires specification data indicating specifications of an electronic product, an estimation unit (21) that estimates design parameters based on the specification data acquired by the acquisition unit using at least one estimation model (40) for estimating design parameters based on the specification data, and an output unit (24) that outputs output data based on the design parameters estimated by the estimation unit.

[0094] (Clause 2) In the estimation device described in paragraph 1, the estimation unit further estimates S-parameters of the electronic product based on the specification data acquired by the acquisition unit using at least one estimation model, and the output unit outputs output data based on the S-parameters estimated by the estimation unit.

[0095] (3) In the estimation device according to the first or second aspect, the at least one estimation model includes a plurality of estimation models according to characteristics of the electronic product. The estimation unit estimates the design parameters by using one of the plurality of estimation models.

[0096] (4) In the estimation device according to any one of paragraphs 1 to 3, at least one estimation model is trained to estimate S-parameters that satisfy specification data based on design parameters, and is further optimized to estimate the design parameters based on the S-parameters.

[0097] (5) In the estimation device according to any one of paragraphs 1 to 3, at least one estimation model is trained to estimate a coupling matrix based on design parameters, and is further optimized to estimate the design parameters based on S-parameters corresponding to the coupling matrix. The coupling matrix includes a plurality of elements obtained from an equivalent circuit of the electronic product.

[0098] (Item 6) In the estimation device according to any one of items 1 to 5, the output data includes at least one of the design parameters and the S parameters estimated by the estimation unit.

[0099] (Item 7) The estimation device according to any one of items 1 to 6, further comprising a determination unit (21) that determines whether the estimation unit can estimate design parameters that satisfy design criteria for designing an electronic product.

[0100] (Clause 8) In the estimation device described in clause 7, if the estimation unit is able to estimate design parameters that satisfy the design criteria, the output data includes at least one of data indicating that the electronic product is manufacturable, data for encouraging the manufacture of the electronic product, and data indicating the cost of manufacturing the electronic product.

[0101] (Item 9) In the estimation device according to item 7 or 8, if the estimation unit is unable to estimate design parameters that satisfy the design criteria, the output data includes data indicating that the electronic product cannot be manufactured.

[0102] (Clause 10) In the estimation device described in any one of clauses 7 to 9, when the estimation unit cannot estimate design parameters that satisfy the design criteria, the estimation unit extends an estimation range of at least one estimation model, and estimates the design parameters using the extended at least one estimation model.

[0103] (Item 11) In the estimation device according to any one of items 1 to 10, at least one estimation model is a surrogate model that estimates design parameters as a result of a simulation based on specification data.

[0104] (Item 12) An estimation method according to one embodiment includes, as processing executed by a computer (21), a step (S1) of acquiring specification data indicating specifications of an electronic product, a step (S5) of estimating design parameters based on the specification data acquired by the acquiring step using at least one estimation model (40) for estimating design parameters based on the specification data, and steps (S7 to S9) of outputting output data based on the estimation results of the design parameters by the estimating step.

[0105] (Item 13) An estimation program according to one embodiment causes a computer (21) to execute the steps of: acquiring specification data indicating specifications of an electronic product (S1); estimating design parameters based on the specification data acquired by the acquiring step using at least one estimation model (40) for estimating design parameters based on the specification data (S5); and outputting output data based on the estimation results of the design parameters by the estimating step (S7 to S9).

[0106] Although a number of embodiments and modifications have been described above, the features of each of these embodiments and modifications can be appropriately combined to the extent that no contradiction occurs.

[0107] The embodiments disclosed herein should be considered to be illustrative and not restrictive in all respects. The scope of the present disclosure is defined by the claims, not by the description of the embodiments described above, and is intended to include all modifications within the meaning and scope of the claims. [Explanation of symbols]

[0108] 1 estimation system, 10 user device, 11, 21 computing device, 12, 22 memory, 13, 23 storage device, 14, 24 communication device, 15 input device, 16 display, 20 estimation device, 27 data reading device, 30 estimation program, 40, 40A, 40D, D estimation model, 50 storage medium, 100 electronic product.

Claims

1. An estimation device for estimating design parameters for designing an electronic product, comprising: an acquisition unit that acquires specification data indicating specifications of the electronic product; an estimation unit that estimates the design parameters based on the specification data acquired by the acquisition unit, using at least one estimation model for estimating the design parameters based on the specification data; and an output unit that outputs output data based on the design parameters estimated by the estimation unit.

2. The estimation unit further estimates S-parameters of the electronic product based on the specification data acquired by the acquisition unit using the at least one estimation model; The estimation device according to claim 1 , wherein the output unit outputs the output data based on the S-parameters estimated by the estimation unit.

3. the at least one estimation model includes a plurality of estimation models according to characteristics of the electronic product; The estimation device according to claim 1 , wherein the estimation unit estimates the design parameters by using one estimation model among the plurality of estimation models.

4. The estimation device according to claim 1 or 2, wherein the at least one estimation model is trained to estimate S-parameters that satisfy the specification data based on the design parameters, and is further optimized to estimate the design parameters based on the S-parameters.

5. the at least one estimation model is trained to estimate a coupling matrix based on the design parameters and is further optimized to estimate the design parameters based on S-parameters corresponding to the coupling matrix; The estimating apparatus according to claim 1 or 2, wherein the coupling matrix includes a plurality of elements obtained from an equivalent circuit of the electronic product.

6. The estimation device according to claim 1 or 2, wherein the output data includes at least one of the design parameters and S-parameters estimated by the estimation unit.

7. The estimation device according to claim 1 , further comprising a determination unit that determines whether the estimation unit is capable of estimating the design parameters that satisfy a design criterion for designing the electronic product.

8. 8. The estimation apparatus of claim 7, wherein if the estimator is able to estimate the design parameters that satisfy the design criteria, the output data includes at least one of data indicating that the electronic product is manufacturable, data for facilitating the manufacture of the electronic product, and data indicative of a cost to manufacture the electronic product.

9. The estimating apparatus of claim 7 , wherein if the estimator is unable to estimate the design parameters that satisfy the design criteria, the output data includes data indicating that the electronic product is not manufacturable.

10. 8. The estimation device according to claim 7, wherein, when the design parameters that satisfy the design criteria cannot be estimated, the estimation unit extends an estimation range of the at least one estimation model, and estimates the design parameters using the extended at least one estimation model.

11. 3. The estimation device according to claim 1, wherein the at least one estimation model is a surrogate model that estimates the design parameters as a result of a simulation based on the specification data.

12. 1. An estimation method for estimating design parameters for designing an electronic product, comprising: The process that the computer executes is as follows: obtaining specification data indicative of a specification of the electronic product; estimating the design parameters based on the specification data acquired by the acquiring step using at least one estimation model for estimating the design parameters based on the specification data; and outputting output data based on a result of the estimation of the design parameters by the estimating step.

13. An estimation program for estimating design parameters for designing an electronic product, comprising: On the computer, obtaining specification data indicative of a specification of the electronic product; estimating the design parameters based on the specification data acquired by the acquiring step using at least one estimation model for estimating the design parameters based on the specification data; and outputting output data based on the estimation result of the design parameters by the estimating step.

Images