Training device, substrate coating system, and substrate coating method

Abstract

The present invention provides a training device that improves insulation performance of a substrate after coating.　A training device (1) for obtaining a trained model for inferring the coating condition of a substrate provided with a conductor and an electronic component. The training device (1) comprises: a training data acquisition unit (11) for acquiring first input data and second input data; and a model generation unit (12) for generating a trained model for inferring second input data from the first input data. The first input data includes: substrate information including information on the type and shape of the substrate; conductor information including information on the type and pattern of the conductor; electronic component mounting information including information on the type, shape, and arrangement position of the electronic component; coating material information including information on insulation performance of the coating material; and substrate use environment information including a voltage waveform applied to the conductor on the substrate. The second input data is a coating condition for dielectric breakdown of the coating material to be suppressed or aerial discharge in the substrate to be suppressed.

Description

Learning device, substrate coating system, and substrate coating method

[0001] The present disclosure relates to a learning apparatus, a substrate coating system, and a substrate coating method.

[0002] Conventionally, a substrate processing system has been proposed that derives recommended processing conditions for substrate processing based on a learning model generated by machine learning using a data set including performance data on the quality of substrate processing (see, for example, Patent Document 1).

[0003] International Publication No. 2020 / 105517

[0004] In the substrate processing system shown in Patent Document 1, the insulating performance of the substrate after coating is unknown, and therefore there is a problem that the coating material after coating may break down or aerial discharge may occur on the substrate while the substrate is in use.

[0005] The present disclosure discloses a technique for solving the above-mentioned problems, and aims to provide a learning device, a substrate coating system, and a substrate coating method that improve the insulating performance of a substrate after coating.

[0006] The learning device disclosed herein is a learning device that obtains a trained model for inferring coating conditions in a substrate coating system that coats a substrate having a conductor and electronic components with a coating material, and is equipped with a training data acquisition unit that acquires training data that is a combination of first input data and second input data, and a model generation unit that uses the training data to generate a trained model for inferring the second input data from the first input data, wherein the first input data includes substrate information including information on the type and shape of the substrate, conductor information including information on the type and pattern of the conductor, electronic component mounting information including information on the type, shape and placement position of the electronic components, coating material information including information on the insulating performance of the coating material, and substrate usage environment information including the waveform of the voltage applied to the conductor on the substrate, and the second input data are coating conditions that, on a substrate coated under the conditions indicated by the first input data, prevent dielectric breakdown of the coating material after coating or prevent aerial discharge on the substrate after coating, and the coating conditions are at least one of the coating thickness distribution of the coating material on the substrate and the number of times the coating material is applied to the substrate.

[0007] The learning device disclosed herein is a learning device that obtains a trained model for inferring coating conditions in a substrate coating system that coats a substrate having a conductor and electronic components with a coating material, and includes a training data acquisition unit that acquires training data that is a combination of first input data and second input data, and a model generation unit that uses the training data to generate a trained model for inferring the second input data from the first input data, wherein the first input data includes substrate information including information on the type and shape of the substrate, conductor information including information on the type and pattern of the conductor, electronic component mounting information including information on the type, shape and placement position of the electronic components, coating material information including information on the insulating performance of the coating material, and substrate usage environment information including the waveform of the voltage applied to the conductor on the substrate, and the second input data are coating conditions that, in a substrate coated under the conditions indicated by the first input data, prevent dielectric breakdown of the coating material after coating or prevent aerial discharge in the substrate after coating, and by coating according to these coating conditions, the insulating performance of the substrate after coating can be improved.

[0008] 1 is a block diagram showing the configuration of a learning device in embodiment 1. FIG. 2 is a diagram showing an example of a neural network in embodiment 1. FIG. 3 is a flowchart for explaining the processing of the learning device and the learned model storage unit in embodiment 1. FIG. 4 is a block diagram showing the configuration of a substrate coating system in embodiment 1. FIG. 5 is a flowchart for explaining the processing of the inference device and the coating device in embodiment 1. FIG. 6 is a block diagram showing another configuration of the substrate coating system in embodiment 1. FIG. 7 is a block diagram showing the configuration of a learning device in embodiment 2. FIG. 8 is a block diagram showing the configuration of a substrate coating system in embodiment 3. FIG. 9 is a flowchart for explaining the processing of the inference device and the coating device in embodiment 3. FIG. 9 is a block diagram showing the configuration of a substrate coating system in embodiment 4. FIG. 10 is a flowchart for explaining the processing of the inference device and the coating device in embodiment 4. FIG. 11 is a schematic diagram showing an example of the hardware configuration of the substrate coating systems in embodiments 1 and 2. FIG. 12 is a schematic diagram showing an example of the hardware configuration of the substrate coating system in embodiment 3. FIG. 13 is a schematic diagram showing an example of the hardware configuration of the substrate coating system in embodiment 4.

[0009] Hereinafter, a learning device, a substrate coating system, and a substrate coating method according to embodiments will be described in detail with reference to the drawings. Note that the same reference numerals in the various drawings indicate the same or corresponding parts.

[0010] Embodiment 1. <Learning Phase> Figure 1 is a block diagram showing the configuration of a learning device 1 in embodiment 1. The learning device 1 obtains a trained model for inferring coating conditions in a substrate coating system that coats a substrate having a conductor and an electronic component with a coating material. The learning device 1 includes a training data acquisition unit 11 and a model generation unit 12.

[0011] The training data acquisition unit 11 acquires training data, which is a combination of first input data and second input data. Here, the first input data includes: substrate information, including information on the type and shape of the substrate when coating a coating material on a substrate including conductors and electronic components; conductor information, including information on the type and pattern of the conductors; electronic component mounting information, including information on the type, shape, and placement position of the electronic components; coating material information, including information on the insulating performance of the coating material; and substrate usage environment information, including the waveform of a voltage applied to the conductor on the substrate. The coating material information may include, for example, at least one of the dielectric constant, conductivity, and breakdown field of the coating material after coating on the substrate and drying or curing. The substrate usage environment information may include the substrate usage temperature, substrate usage humidity, substrate usage atmospheric pressure, and voltage application time to the conductor on the substrate. The second input data are coating conditions that suppress dielectric breakdown of the coating material after coating or suppress aerial discharge on the substrate after coating under the conditions specified by the first input data, and the coating conditions are at least one of the coating thickness distribution of the coating material on the substrate and the number of times the coating material is applied to the substrate. The learning data acquisition unit 11 may acquire the first input data and the second input data from a data input device (not shown). The data input device is, for example, a keyboard, a mouse, a keypad, a touch panel, or the like, which is operated by a user.

[0012] The model generation unit 12 learns coating conditions that suppress dielectric breakdown of the coating material after coating or suppress aerial discharge on a substrate coated under the conditions indicated by the first input data, based on the training data that is a combination of the first input data and the second input data output from the training data acquisition unit 11. That is, the model generation unit 12 generates a trained model that infers optimal coating conditions from the training data that is a combination of the first input data and the second input data. Here, the training data is data that associates the first input data and the second input data with each other.

[0013] Information on the waveform of a voltage applied to a conductor on a substrate, included in the substrate use environment information of the first input data acquired by the learning data acquisition unit 11, may be actual measurement data obtained through an experiment. For example, the experimental conditions are first assumed to include experimental coating conditions, substrate information, conductor information, electronic component mounting information, and coating material information. If the coating conditions include a coating thickness distribution of a coating material on a substrate, the experimental coating conditions include an experimental coating thickness distribution, which is the coating thickness distribution of the coating material on the substrate in the test, and if the coating conditions include the number of times the coating material is reapplied to the substrate, the experimental coating number is assumed to include the number of times the coating material is reapplied to the substrate in the test.

[0014] Next, when the second input data are coating conditions under which aerial discharge in the coated substrate is suppressed, a partial discharge test is performed on an experimental substrate that satisfies these experimental conditions to determine an aerial-discharge applied voltage peak value that has the smallest absolute value among the applied voltage peak values ​​at which aerial discharge occurs in the coated experimental substrate, and a discharge-suppression applied voltage waveform is determined that is a voltage waveform applied to a conductor in the experimental substrate that has an absolute value smaller than the obtained aerial-discharge applied voltage peak value and that suppresses aerial discharge in the coated experimental substrate, and the determined discharge-suppression applied voltage waveform may be used as the applied voltage waveform included in the first input data. The aerial-discharge applied voltage peak value may be determined by performing a partial discharge test on the experimental substrate that satisfies the experimental conditions, gradually increasing the absolute value of the applied voltage peak value of the voltage waveform applied to the conductor in the experimental substrate. Alternatively, when the second input data specifies coating conditions that suppress breakdown of the coating material after coating, a dielectric breakdown test on an experimental substrate that satisfies these experimental conditions may determine a breakdown application voltage peak value that has the smallest absolute value among the applied voltage peak values ​​that cause breakdown of the experimental coating material, which is the coating material of the coated experimental substrate, and determine a breakdown-suppression application voltage waveform that is a voltage waveform applied to a conductor on the experimental substrate that suppresses breakdown of the experimental coating material after coating and whose absolute value is smaller than the obtained breakdown application voltage peak value, and the determined breakdown-suppression application voltage waveform may be used as the applied voltage waveform included in the first input data. The breakdown application voltage peak value may be determined by conducting a dielectric breakdown test on an experimental substrate that satisfies the experimental conditions, gradually increasing the absolute value of the applied voltage peak value of the voltage waveform applied to the conductor on the experimental substrate.

[0015] Information on the waveform of a voltage applied to a conductor on a substrate, included in the substrate use environment information of the first input data acquired by the learning data acquisition unit 11, may be calculated data obtained by analysis. For example, the analysis conditions include analysis coating conditions, substrate information, conductor information, electronic component mounting information, and coating material information. If the coating conditions include a coating thickness distribution of a coating material on a substrate, the analysis coating conditions include an analysis coating thickness distribution, which is the coating thickness distribution of the coating material on the substrate in the analysis, and if the coating conditions include the number of reapplications of a coating material on the substrate, the analysis coating conditions include an analysis reapplication count, which is the number of reapplications of the coating material on the substrate in the analysis.

[0016] Next, when the second input data are coating conditions under which aerial discharge is suppressed on the coated substrate, at least one of finite element analysis and equivalent circuit analysis may be performed on an analysis substrate that satisfies these analysis conditions to determine an aerial-discharge applied voltage peak value that has the smallest absolute value among the applied voltage peak values ​​at which aerial discharge occurs on the coated analysis substrate, and a discharge-suppression applied voltage waveform that is a voltage waveform applied to a conductor on the analysis substrate that has a smaller absolute value than the obtained aerial-discharge applied voltage peak value and suppresses aerial discharge on the coated analysis substrate may be determined, and the determined discharge-suppression applied voltage waveform may be used as the applied voltage waveform included in the first input data. The aerial-discharge applied voltage peak value may be determined by performing an analysis on the analysis substrate that satisfies the analysis conditions, in which the absolute value of the applied voltage peak value of the applied voltage waveform to the conductor is gradually increased. Alternatively, when the second input data specifies coating conditions that suppress breakdown of the coating material after coating, at least one of finite element analysis and equivalent circuit analysis may be performed on an analysis substrate that satisfies these analysis conditions to determine a breakdown voltage peak value that has the smallest absolute value among the voltage peak values ​​that cause breakdown in the analysis coating material of the analysis substrate after coating, and a breakdown-prevention voltage waveform may be determined as a voltage waveform applied to a conductor on the analysis substrate that has a smaller absolute value than the breakdown voltage peak value obtained and that suppresses breakdown of the analysis coating material after coating, and the determined breakdown-prevention voltage waveform may be used as the applied voltage waveform included in the first input data. The breakdown voltage peak value may be determined by performing an analysis on the analysis substrate that satisfies the analysis conditions, gradually increasing the absolute value of the voltage peak value of the voltage waveform applied to the conductor.

[0017] The learning algorithm used by the model generation unit 12 may be a known algorithm such as supervised learning, unsupervised learning, reinforcement learning, or semi-supervised learning. An example will be described in which a neural network is used. The model generation unit 12 learns coating conditions by so-called supervised learning, for example, according to a neural network model. Here, supervised learning refers to a technique in which a set of input and result (label) data is provided to the learning device 1, whereby the learning device learns the features of the learning data and infers the result from the input.

[0018] A neural network is composed of an input layer consisting of multiple neurons, an intermediate layer (hidden layer) consisting of multiple neurons, and an output layer consisting of multiple neurons. The intermediate layer may be one layer or two or more layers. FIG. 2 is a diagram showing an example of a neural network in embodiment 1. For example, in a three-layer neural network as shown in FIG. 2, when multiple inputs are input to the input layer (X1-X3), the values ​​are multiplied by a weight w1 (w11-w16) and input to the intermediate layer (Y1-Y2), and the result is further multiplied by a weight w2 (w21-w26) and output from the output layer (Z1-Z3). This output result varies depending on the values ​​of the weights w1 and w2.

[0019] In the learning device 1 of the first embodiment, the neural network learns coating conditions that suppress dielectric breakdown of the coating material after coating or that suppress aerial discharge on the coated substrate by so-called supervised learning in accordance with learning data created based on a combination of first input data and second input data acquired by the learning data acquisition unit 11. That is, the neural network inputs, into the input layer, first input data that includes substrate information including information on the type and shape of the substrate when coating a substrate having conductors and electronic components with a coating material, conductor information including information on the type and pattern of the conductors, electronic component mounting information including information on the type, shape, and placement position of the electronic components, coating material information including information on the insulating performance of the coating material, and substrate use environment information including the waveform of a voltage applied to the conductor on the substrate, and learns by adjusting the weights w1 and w2 so that the results output from the output layer approach coating conditions that suppress dielectric breakdown of the coating material after coating or that suppress aerial discharge on the coated substrate in a substrate coated under the conditions specified by the first input data.

[0020] The model generation unit 12 generates and outputs a trained model by executing the above-described learning. The trained model storage unit 2 stores the trained model output from the model generation unit 12.

[0021] Next, a description will be given of the learning process performed by the learning device 1. Fig. 3 is a flowchart for explaining the process performed by the learning device 1 and the trained model storage unit 2 in embodiment 1. Step S01 is a training data acquisition step, step S02 is a learning process step, and step S03 is a trained model storage step.

[0022] In step S01, the training data acquiring unit 11 acquires training data and outputs it to the model generating unit 12, and then the process proceeds to step S02. Note that the training data acquiring unit 11 may acquire the first input data and the second input data included in the training data simultaneously, or may acquire the first input data and the second input data at different times as long as the first input data and the second input data are acquired in association with each other.

[0023] In step S02, the model generation unit 12 learns the coating conditions by so-called supervised learning in accordance with the training data created based on the combination of the first input data and the second input data acquired by the training data acquisition unit 11, generates a trained model, and outputs the generated trained model to the trained model storage unit 2, and then proceeds to step S03. In step S03, the trained model storage unit 2 stores the trained model acquired from the model generation unit 12, and the process ends.

[0024] <Utilization Phase> Next, the substrate coating system according to the first embodiment will be described. FIG. 4 is a block diagram showing the configuration of a substrate coating system 100 according to the first embodiment. The substrate coating system 100 applies a coating material to a substrate having a conductor and an electronic component. The substrate coating system 100 includes an inference device 3 and a coating device 4. The inference device 3 includes a data acquisition unit 31 and an inference unit 32. The trained model storage unit 2, the inference device 3, and the coating device 4 may be separate devices connected via a network. The trained model storage unit 2 and the inference device 3 may be built into the coating device 4. The trained model storage unit 2 and the inference device 3 may also reside on a cloud server.

[0025] The data acquisition unit 31 of the inference device 3 acquires, as input data, board information including information on the type and shape of the board when coating a board including conductors and electronic components with a coating material; conductor information including information on the type and pattern of the conductors; electronic component mounting information including information on the type, shape, and placement of the electronic components; coating material information including information on the insulating performance of the coating material; and board use environment information including the waveform of a voltage applied to the conductor on the board. The coating material information may include, for example, at least one of the dielectric constant, conductivity, and breakdown field of the coating material after coating on the board and drying or curing. The board use environment information may include the operating temperature of the board, the operating humidity of the board, the operating atmospheric pressure of the board, and the duration of voltage application to the conductor on the board. The data acquisition unit 31 may acquire the input data from a data input device (not shown). The data input device may be, for example, a keyboard, a mouse, a keypad, a touch panel, or the like, and is operated by a user.

[0026] The inference unit 32 uses the trained model stored in the trained model storage unit 2 to infer coating conditions under the conditions indicated by the input data that suppress dielectric breakdown of the coating material after coating or suppress aerial discharge on the substrate after coating. That is, by inputting the input data acquired by the data acquisition unit 31 into this trained model, the inference unit 32 infers coating conditions under which dielectric breakdown of the coating material after coating or aerial discharge on the substrate after coating is suppressed on a substrate coated under the conditions indicated by the input data. Specifically, the inference unit 32 can infer at least one of the coating thickness distribution of the coating material on the substrate and the number of times the coating material is applied to the substrate, and output the inference result. Note that the coating thickness distribution may be the thickness distribution before, during, or after curing of the coating.

[0027] In the first embodiment, the inference unit 32 is described as outputting coating conditions using a trained model trained by the model generation unit 12, but it may also be configured to acquire a trained model from an external source, such as another learning device, and output coating conditions based on this trained model.

[0028] The coating device 4 coats the substrate with the coating material based on the coating conditions, which are the output of the inference unit 32 and are at least one of the coating thickness distribution of the coating material on the substrate and the number of times the coating material is applied to the substrate. For example, if the output of the inference unit 32 is the coating thickness distribution of the coating material on the substrate, the coating device 4 coats the substrate with the coating material based on the acquired information on the coating thickness distribution of the coating material on the substrate. For example, if the output of the inference unit 32 is the number of times the coating material is applied to the substrate, the coating device 4 applies the coating material to the substrate a specified number of times based on the acquired information on the number of times the coating material is applied to the substrate. For example, if the output of the inference unit 32 is the coating thickness distribution of the coating material on the substrate and the number of times the coating material is applied to the substrate, the coating device 4 applies the coating material to the substrate a specified number of times based on the acquired information on the coating thickness distribution of the coating material on the substrate and the acquired information on the number of times the coating material is applied to the substrate, thereby achieving the specified coating thickness distribution. The coating method in the coating device 4 is not limited to a specific method, and may be, for example, spray coating, dispense coating, slit coating, dip coating, spin coating, or any other known method.

[0029] 5 is a flowchart for explaining the processing of the inference device 3 and the coating device 4 in embodiment 1. Step S11 is a data acquisition step, step S12 is an inference processing step, step S13 is a coating condition output step, and step S14 is a coating step. Steps S12 and S13 are inference steps.

[0030] In step S11, the data acquisition unit 31 acquires as input data the following: substrate information including information on the type and shape of the substrate; conductor information including information on the type and pattern of the conductors; electronic component mounting information including information on the type, shape, and placement of the electronic components; coating material information including information on the insulating performance of the coating material; and substrate use environment information including the waveform of the voltage applied to the conductors on the substrate; and outputs these to the inference unit 32, and the process proceeds to step S12. In step S12, the inference unit 32 inputs the input data acquired from the data acquisition unit 31 into the trained model stored in the trained model storage unit 2 to obtain coating conditions as output data, and the process proceeds to step S13. In step S13, the inference unit 32 outputs the coating conditions as an output of the trained model to the coating device 4, and the process proceeds to step S14. In step S14, the coating device 4 coats the substrate with the coating material based on the coating conditions as output data acquired from the inference unit 32, and the process ends. This prevents dielectric breakdown of the coating material after coating, or prevents aerial discharge on the coated substrate. Furthermore, the number of steps required for teaching or coordinate input in the coating process for the substrate can be reduced, and the insulating performance of the substrate after coating can be improved.

[0031] FIG. 6 is a block diagram showing another configuration of the substrate coating system 100 according to the first embodiment. The substrate coating system 100 shown in FIG. 6 includes a learning device 1, a trained model storage unit 2, an inference device 3, and a coating device 4. In the substrate coating system 100 shown in FIG. 6, the operations of the learning device 1, the trained model storage unit 2, the inference device 3, and the coating device 4 are the same as those of the learning device 1, the trained model storage unit 2, the inference device 3, and the coating device 4 shown in FIGS. 1 and 4. The learning device 1, the trained model storage unit 2, the inference device 3, and the coating device 4 may be separate devices connected via a network. Alternatively, the learning device 1, the trained model storage unit 2, and the inference device 3 may be built into the coating device 4. Furthermore, the learning device 1, the trained model storage unit 2, and the inference device 3 may reside on a cloud server. When the learning device 1 and the inference device 3 are implemented by a single device, the training data acquisition unit 11 may also function as the data acquisition unit 31.

[0032] The model generation unit 12 may learn coating conditions based on training data created for multiple substrate coating systems. The model generation unit 12 may acquire training data from multiple substrate coating systems used in the same area, or may learn coating conditions using training data collected from multiple substrate coating systems operating independently in different areas. Substrate coating systems from which training data is collected may be added or removed from the target system during the process. Furthermore, a learning device that has learned coating conditions for one coating system may be applied to another coating system, and the coating conditions for the other coating system may be re-learned and updated.

[0033] As described above, the learning device 1 according to the first embodiment is a learning device 1 that obtains a trained model for inferring coating conditions in a substrate coating system that coats a substrate having a conductor and an electronic component with a coating material, and includes a training data acquisition unit 11 that acquires training data that is a combination of first input data and second input data, and a model generation unit 12 that uses the training data to generate a trained model for inferring the second input data from the first input data, where the first input data includes substrate information including information on the type and shape of the substrate, conductor information including information on the type and pattern of the conductor, electronic component mounting information including information on the type, shape, and placement position of the electronic component, coating material information including information on the insulating performance of the coating material, and substrate use environment information including the waveform of the voltage applied to the conductor on the substrate, and the second input data are coating conditions that, in a substrate coated under the conditions indicated by the first input data, prevent dielectric breakdown of the coating material after coating or prevent aerial discharge in the substrate after coating, and the coating conditions are at least one of the coating thickness distribution of the coating material on the substrate and the number of times the coating material is applied to the substrate, thereby improving the insulating performance of the substrate after coating.

[0034] Second Embodiment. Figure 7 is a block diagram showing the configuration of a learning device 1a according to a second embodiment. Comparing the learning device 1a according to the second embodiment shown in Figure 7 with the learning device 1 according to the first embodiment shown in Figure 1, a calculation unit 13 has been added. The other configuration of the learning device 1a according to the second embodiment is the same as that of the learning device 1 according to the first embodiment. Furthermore, the trained model generated by the learning device 1a is stored in the trained model storage unit 2 and is used in the substrate coating system 100 in the utilization phase, as in the first embodiment. Below, differences between the operation of the learning device 1a according to the second embodiment and the operation of the learning device 1 according to the first embodiment will be described.

[0035] The calculation unit 13 acquires substrate information including information on the type and shape of the substrate, conductor information including information on the type and pattern of the conductors, electronic component mounting information including information on the type, shape, and placement position of the electronic components, and coating material information including information on the insulating performance of the coating material, and sets these as part of the first input data. The calculation unit 13 further acquires analysis coating conditions as second input data. The coating conditions are at least one of the coating thickness distribution of the coating material on the substrate and the number of reapplications of the coating material on the substrate. If the coating conditions include the coating thickness distribution of the coating material on the substrate, the analysis coating conditions include an analysis coating thickness distribution, which is the coating thickness distribution of the coating material on the substrate in the analysis, and if the coating conditions include the number of reapplications of the coating material on the substrate, the analysis coating conditions include an analysis reapplication number, which is the number of reapplications of the coating material on the substrate in the analysis. The coating material information may include, for example, at least one of the dielectric constant, conductivity, and breakdown field of the coating material when it is coated on the substrate and then dried or cured, and the substrate usage environment information may include the substrate usage temperature, substrate usage humidity, substrate usage atmospheric pressure, and the duration of voltage application to the conductor in the substrate, as in embodiment 1.

[0036] Next, the calculation unit 13 uses the acquired information as analysis conditions, and when the second input data are coating conditions under which aerial discharge is suppressed on the coated substrate, performs at least one of finite element analysis and equivalent circuit analysis on the analysis substrate, which is a substrate that satisfies these analysis conditions, to determine an aerial-discharge applied voltage peak value that has the smallest absolute value among the applied voltage peak values ​​at which aerial discharge occurs on the coated analysis substrate, determines a discharge-suppression applied voltage waveform that is a voltage waveform applied to the conductor on the analysis substrate, whose absolute value is smaller than the obtained aerial-discharge applied voltage peak value, and which suppresses aerial discharge on the coated analysis substrate, and includes the determined discharge-suppression applied voltage waveform as the applied voltage waveform in the first input data. The aerial-discharge applied voltage peak value may be determined by performing an analysis on the analysis substrate, which is a substrate that satisfies the analysis conditions, in which the absolute value of the applied voltage peak value of the applied voltage waveform to the conductor is gradually increased. Alternatively, the calculation unit 13 uses the acquired information as analysis conditions, and when the second input data are coating conditions that suppress dielectric breakdown of the coating material after coating, performs at least one of finite element analysis and equivalent circuit analysis on the analysis substrate, which is a substrate that satisfies these analysis conditions, to determine the smallest absolute breakdown voltage peak value among the applied voltage peak values ​​that cause dielectric breakdown of the analysis coating material, which is the coating material of the analysis substrate after coating, and determines a breakdown-prevention applied voltage waveform, which is a voltage waveform applied to a conductor on the analysis substrate whose absolute value is smaller than the obtained breakdown voltage peak value and that suppresses dielectric breakdown of the analysis coating material after coating, and includes the determined breakdown-prevention applied voltage waveform as the applied voltage waveform in the first input data. The breakdown voltage peak value may also be determined by performing an analysis on the analysis substrate, which is a substrate that satisfies the analysis conditions, in which the absolute value of the applied voltage peak value of the applied voltage waveform to the conductor is gradually increased.

[0037] The calculation unit 13 outputs the first input data including the applied voltage waveform obtained in this manner and the acquired second input data to the learning data acquisition unit 11. The processing of the learning device 1a after the learning data acquisition unit 11 acquires the learning data, which is a combination of the first input data and the second input data, is the same as the processing of the learning device 1 according to the first embodiment.

[0038] As described above, the learning device 1a according to the second embodiment acquires, as analysis conditions, the analysis coating conditions, substrate information, conductor information, electronic component mounting information, and coating material information, determines an applied voltage waveform from the information on the analysis conditions, outputs the substrate information, conductor information, electronic component mounting information, coating material information, and applied voltage waveform as first input data to the learning data acquisition unit 11, and includes a calculation unit 13 that outputs the analysis coating conditions as second input data to the learning data acquisition unit 11. The analysis coating conditions include an analysis coating thickness distribution that is the coating thickness distribution of the coating material on the substrate in the analysis if the coating conditions include a coating thickness distribution of the coating material on the substrate, and an analysis recoating count that is the number of recoats of the coating material on the substrate in the analysis if the coating conditions include a coating thickness distribution of the coating material on the substrate, and the calculation unit 13 performs at least one of finite element analysis and equivalent circuit analysis on the analysis substrate, which is a substrate that satisfies the analysis conditions, when the second input data are coating conditions that suppress aerial discharge on the substrate after coating. a discharge-suppression applied voltage waveform is determined, which is a voltage waveform applied to a conductor on the analysis substrate after coating, the absolute value of which is smaller than that of the aerial-discharge applied voltage waveform, and which suppresses aerial discharge on the analysis substrate after coating; and the discharge-suppression applied voltage waveform is used as the applied voltage waveform. When the second input data is a coating condition under which dielectric breakdown of the coating material after coating is suppressed, at least one of finite element analysis and equivalent circuit analysis is performed on the analysis substrate to determine a breakdown applied voltage waveform which is smallest in absolute value among the applied voltage peak values ​​that suppress dielectric breakdown of the analysis coating material, which is the coating material of the analysis substrate after coating; and a breakdown-suppression applied voltage waveform is determined, which is a voltage waveform applied to a conductor on the analysis substrate after coating, the absolute value of which is smaller than that of the breakdown applied voltage peak value, and which suppresses dielectric breakdown of the analysis coating material after coating; and the breakdown-suppression applied voltage waveform is used as the applied voltage waveform. This makes it possible to obtain a trained model with higher inference accuracy.

[0039] Embodiment 3. Figure 8 is a block diagram showing the configuration of a substrate coating system 100a according to embodiment 3. Compared to the substrate coating system 100 according to embodiment 1 shown in Figure 4, the substrate coating system 100a according to embodiment 3 shown in Figure 8 additionally includes a database device 5. The other configuration of the substrate coating system 100a according to embodiment 3 is the same as that of the substrate coating system 100 according to embodiment 1.

[0040] The database device 5 stores the following information when coating a substrate having a conductor and electronic components with a coating material: substrate information including information on the type and shape of the substrate; conductor information including information on the type and pattern of the conductor; electronic component mounting information including information on the type, shape, and placement position of the electronic components; coating material information including information on the insulating performance of the coating material; and substrate use environment information including the waveform of a voltage applied to the conductor on the substrate. The coating material information may include, for example, the dielectric constant, conductivity, and breakdown field of the coating material after coating on the substrate and drying or curing. The substrate use environment information may include the substrate use temperature, substrate use humidity, substrate use atmospheric pressure, and voltage application time to the conductor on the substrate. The data acquisition unit 31 acquires the substrate information, conductor information, electronic component mounting information, coating material information, and substrate use environment information from the database device 5.

[0041] 9 is a flowchart illustrating the processing of the inference device 3 and the coating device 4 in the third embodiment. Comparing the flowchart of the third embodiment shown in FIG. 9 with the flowchart of the first embodiment shown in FIG. 5 , step S11, which is a data acquisition step, is replaced with step S11a. In step S11a, the data acquisition unit 31 acquires, as input data from the database device 5, board information including information on the type and shape of the board, conductor information including information on the type and pattern of the conductor, electronic component mounting information including information on the type, shape, and placement position of the electronic components, coating material information including information on the insulating performance of the coating material, and board use environment information including the waveform of the voltage applied to the conductor on the board, and outputs these to the inference unit 32. The processing from step S12 to step S14 is the same as in the first embodiment.

[0042] As described above, the substrate coating system 100a according to embodiment 3 includes a database device 5 that stores substrate information, conductor information, electronic component mounting information, coating material information, and substrate usage environment information, and the data acquisition unit 31 acquires the substrate information, conductor information, electronic component mounting information, coating material information, and substrate usage environment information from the database device 5, thereby reducing the labor required for teaching or coordinate input during the substrate coating process and improving the insulation performance of the substrate after coating.

[0043] The learning device 1, the trained model storage unit 2, the inference device 3, and the database device 5 may be connected to the coating device 4 via a network, for example, and may be separate devices from the coating device 4. The learning device 1, the trained model storage unit 2, the inference device 3, and the database device 5 may be built into the coating device 4. Furthermore, the learning device 1, the trained model storage unit 2, the inference device 3, and the database device 5 may exist on a cloud server.

[0044] Fourth Embodiment Figure 10 is a block diagram showing the configuration of a substrate coating system 100b according to a fourth embodiment. Compared to the substrate coating system 100 according to the first embodiment shown in Figure 4, the substrate coating system 100b according to the fourth embodiment shown in Figure 10 has an additional scanning device 6. The other configuration of the substrate coating system 100b according to the fourth embodiment is the same as that of the substrate coating system 100 according to the first embodiment.

[0045] The scanning device 6 performs a three-dimensional scan of the appearance of a board having conductors and electronic components, and generates board information including information on the type and shape of the board, conductor information including information on the type and pattern of the conductors, and electronic component mounting information including information on the type, shape, and placement position of the electronic components. The data acquisition unit 31 acquires coating material information and board use environment information, and also acquires board information, conductor information, and electronic component mounting information from the scanning device 6.

[0046] Fig. 11 is a flowchart for explaining the processing of the inference device 3 and the coating device 4 in embodiment 4. Comparing the flowchart of embodiment 4 shown in Fig. 11 with the flowchart of embodiment 1 shown in Fig. 5, step S10, which is a scan step, has been added, and step S11, which is a data acquisition step, has been changed to step S11b.

[0047] In step S10, the scanning device 6 performs a three-dimensional scan of the appearance of the board including the conductors and electronic components, generating board information including information on the type and shape of the board, conductor information including information on the type and pattern of the conductors, and electronic component mounting information including information on the type, shape, and placement position of the electronic components, and then proceeds to step S11b. In step S11b, the data acquisition unit 31 acquires coating material information including information on the insulating performance of the coating material and board use environment information including the waveform of the voltage applied to the conductors on the board, and also acquires the board information, conductor information, and electronic component mounting information from the scanning device 6. The acquired board information, conductor information, electronic component information, coating material information, and board use environment information are output as input data to the inference unit 32, and then proceeds to step S12. The processing from step S12 to step S14 is the same as in embodiment 1.

[0048] As described above, the substrate coating system 100b according to the fourth embodiment includes a scanning device 6 that performs a three-dimensional scan of the appearance of a substrate having conductors and electronic components and generates substrate information, conductor information, and electronic component mounting information, and the data acquisition unit 31 acquires the substrate information, conductor information, and electronic component mounting information from the scanning device 6. This reduces the labor required for teaching or coordinate input during the substrate coating process and improves the insulating performance of the substrate after coating.

[0049] The learning device 1, the trained model storage unit 2, the inference device 3, and the scanning device 6 may be connected to the coating device 4 via a network, for example, and may be separate devices from the coating device 4. The learning device 1, the trained model storage unit 2, the inference device 3, and the scanning device 6 may also be built into the coating device 4. Furthermore, the learning device 1, the trained model storage unit 2, and the inference device 3 may exist on a cloud server.

[0050] FIG. 12 is a schematic diagram showing an example of the hardware configuration of the substrate coating system 100 according to the first and second embodiments. The calculation unit 13, the learning data acquisition unit 11, the model generation unit 12, the data acquisition unit 31, and the inference unit 32 are implemented by a processor 201, such as a CPU (Central Processing Unit), that executes programs stored in a memory 202. The memory 202 is also used as a temporary storage device for each process executed by the processor 201. Multiple processing circuits may cooperate to execute the above functions. Furthermore, the above functions may be implemented by dedicated hardware. When the above functions are implemented by dedicated hardware, the dedicated hardware may be, for example, a single circuit, a composite circuit, a programmed processor, an ASIC, an FPGA, or a combination thereof. The above functions may also be implemented by a combination of dedicated hardware and software, or a combination of dedicated hardware and firmware. The memory 202 may be, for example, a non-volatile or volatile semiconductor memory such as a RAM, a ROM, a flash memory, or an EPROM, a magnetic disk, an optical disk, or a combination thereof. The trained model storage unit 2 is realized by a memory 202. The processor 201, the memory 202, and the coating device 4 are connected to a bus 203.

[0051] Figure 13 is a schematic diagram showing an example of the hardware configuration of a substrate coating system 100a according to embodiment 3. Comparing the schematic diagram showing an example of the hardware configuration of the substrate coating system 100a according to embodiment 3 shown in Figure 13 with the schematic diagram showing an example of the hardware configuration of the substrate coating system 100 according to embodiments 1 and 2 shown in Figure 12, a database device 5 has been added. The other components in the hardware configuration of the substrate coating system 100a according to embodiment 3 are the same as the hardware configuration of the substrate coating system 100 according to embodiments 1 and 2. The database device 5 is connected to a bus 203. The database device 5 may be realized by a memory 202.

[0052] Figure 14 is a schematic diagram showing an example of the hardware configuration of a substrate coating system 100b according to embodiment 4. Comparing the schematic diagram showing an example of the hardware configuration of the substrate coating system 100b according to embodiment 4 shown in Figure 14 with the schematic diagram showing an example of the hardware configuration of the substrate coating system 100 according to embodiments 1 and 2 shown in Figure 12, a scanning device 6 has been added. The other components of the hardware configuration of the substrate coating system 100b according to embodiment 4 are the same as the hardware configuration of the substrate coating system 100 according to embodiments 1 and 2. The scanning device 6 is connected to a bus 203.

[0053] Although various exemplary embodiments are described in this disclosure, the various features, aspects, and functions described in one or more embodiments are not limited to the application of a particular embodiment, but may be applied to the embodiments alone or in various combinations. Therefore, countless variations not illustrated are contemplated within the scope of the technology disclosed in this specification. For example, this includes cases where at least one component is modified, added, or omitted, or where at least one component is extracted and combined with components of another embodiment.

[0054] The above describes in detail preferred embodiments, but the present invention is not limited to the above-described embodiments, and various modifications and substitutions can be made to the above-described embodiments without departing from the scope of the claims.

[0055] Various aspects of the present disclosure are summarized below as appendices.

[0056] (Supplementary Note 1) A learning device for obtaining a trained model for inferring coating conditions in a substrate coating system for coating a substrate having a conductor and an electronic component with a coating material, the device comprising: a training data acquisition unit for acquiring training data which is a combination of first input data and second input data; and a model generation unit for generating a trained model for inferring the second input data from the first input data using the training data, wherein the first input data includes substrate information including information on the type and shape of the substrate, conductor information including information on the type and pattern of the conductor, electronic component mounting information including information on the type, shape and placement position of the electronic component, coating material information including information on the insulating performance of the coating material, and substrate use environment information including an applied voltage waveform to the conductor on the substrate, and the second input data are coating conditions under which, in the substrate coated under the conditions indicated by the first input data, dielectric breakdown of the coating material after coating is suppressed or aerial discharge on the substrate after coating is suppressed, The learning device is characterized in that the coating conditions are at least one of the coating thickness distribution of the coating material on the substrate and the number of times the coating material is reapplied to the substrate. (Appendix 2) The learning device described in Appendix 1 is characterized in that the coating material information includes at least one of the dielectric constant of the coating material, the conductivity of the coating material, and the dielectric breakdown field of the coating material. (Appendix 3) The learning device described in Appendix 1 or 2 is characterized in that the substrate usage environment information further includes at least one of the usage temperature of the substrate, the usage humidity of the substrate, the usage atmospheric pressure of the substrate, and the time of voltage application to the conductor on the substrate. (Appendix 4) The experimental conditions include experimental coating conditions, the substrate information, the conductor information, the electronic component mounting information, and the coating material information, and the experimental coating conditions include an experimental coating thickness distribution, which is the coating thickness distribution of the coating material on the substrate in a test, if the coating conditions include the coating thickness distribution of the coating material on the substrate,When the coating conditions include the number of times the coating material is recoated onto the substrate, the coating conditions include an experimental number of recoats, which is the number of times the coating material is recoated onto the substrate in the test; when the second input data are coating conditions under which aerial discharge on the substrate after coating is suppressed, a partial discharge test is carried out on an experimental substrate, which is a substrate that satisfies the experimental conditions, to determine an aerial discharge applied voltage peak value which has the smallest absolute value among the applied voltage peak values ​​at which aerial discharge occurs on the experimental substrate after coating, and a discharge suppression applied voltage waveform is determined which is a voltage waveform applied to a conductor on the experimental substrate, the absolute value of which is smaller than the aerial discharge applied voltage peak value, and which suppresses aerial discharge on the experimental substrate after coating, and the discharge suppression applied voltage waveform is set to the applied voltage waveform included in the first input data; when the second input data are coating conditions under which dielectric breakdown of the coating material after coating is suppressed, A learning device described in any one of appendices 1 to 3, characterized in that a breakdown applied voltage peak value is determined which has the smallest absolute value among the applied voltage peak values ​​at which breakdown occurs in the experimental coating material, which is the coating material of the experimental substrate after coating, through a breakdown test on the experimental substrate, and a breakdown prevention applied voltage waveform is determined which is a voltage waveform applied to a conductor on the experimental substrate whose absolute voltage peak value is smaller than the breakdown applied voltage peak value and which prevents breakdown of the experimental coating material after coating, and the breakdown prevention applied voltage waveform is used as the applied voltage waveform included in the first input data. (Supplementary Note 5) The analysis conditions include analysis coating conditions, the substrate information, the conductor information, the electronic component mounting information, and the coating material information, and the analysis coating conditions include, if the coating conditions include a coating thickness distribution of the coating material on the substrate, an analysis coating thickness distribution which is the coating thickness distribution of the coating material on the substrate in the analysis, and if the coating conditions include a number of times the coating material is reapplied on the substrate, an analysis number of times the coating material is reapplied on the substrate in the analysis,When the second input data are coating conditions under which aerial discharge is suppressed on the substrate after coating, by performing at least one of finite element analysis and equivalent circuit analysis on an analysis substrate which is a substrate that satisfies the analysis conditions, an aerial discharge applied voltage peak value having the smallest absolute value among the applied voltage peak values ​​at which aerial discharge occurs on the analysis substrate after coating is determined, and a discharge suppression applied voltage waveform is determined, which is a voltage waveform applied to a conductor on the analysis substrate whose absolute value is smaller than the aerial discharge applied voltage peak value and which suppresses aerial discharge on the analysis substrate after coating, and the discharge suppression applied voltage waveform is set to the applied voltage waveform included in the first input data, and when the second input data are coating conditions under which dielectric breakdown of the coating material after coating is suppressed, A learning device described in any one of appendices 1 to 3, characterized in that by performing at least one of finite element analysis and equivalent circuit analysis on the analysis substrate, a breakdown applied voltage peak value is determined which has the smallest absolute value among the applied voltage peak values ​​at which dielectric breakdown occurs in the analysis coating material, which is the coating material of the analysis substrate after coating, and a breakdown-prevention applied voltage waveform is determined which is a voltage waveform applied to a conductor on the analysis substrate whose absolute value is smaller than the breakdown applied voltage peak value and which prevents dielectric breakdown in the analysis coating material after coating, and the breakdown-prevention applied voltage waveform is used as the applied voltage waveform included in the first input data. (Supplementary Note 6) The present invention includes a calculation unit that acquires, as analysis conditions, analysis coating conditions, the substrate information, the conductor information, the electronic component mounting information, and the coating material information, calculates the applied voltage waveform from the information on the analysis conditions, outputs the substrate information, the conductor information, the electronic component mounting information, the coating material information, and the applied voltage waveform to the learning data acquisition unit as the first input data, and outputs the analysis coating conditions to the learning data acquisition unit as the second input data, wherein the analysis coating conditions areWhen the coating conditions include a coating thickness distribution of the coating material on the substrate, the second input data includes an analyzed coating thickness distribution, which is the coating thickness distribution of the coating material on the substrate in the analysis; and when the coating conditions include a number of times the coating material is recoated on the substrate, the second input data includes an analyzed number of times the coating material is recoated on the substrate in the analysis; and when the second input data are coating conditions under which aerial discharge on the substrate after coating is suppressed, the calculation unit performs at least one of finite element analysis and equivalent circuit analysis on an analysis substrate, which is a substrate that satisfies the analysis conditions, to determine an aerial discharge applied voltage peak value having the smallest absolute value among the applied voltage peak values ​​at which aerial discharge occurs on the analysis substrate after coating, and determines a discharge suppression applied voltage waveform, which is a voltage waveform applied to a conductor on the analysis substrate whose absolute value is smaller than the aerial discharge applied voltage peak value and which suppresses aerial discharge on the analysis substrate after coating, and sets the discharge suppression applied voltage waveform as the applied voltage waveform; and when the second input data are coating conditions under which dielectric breakdown of the coating material is suppressed on the coating material after coating, the calculation unit performs at least one of finite element analysis and equivalent circuit analysis on an analysis substrate, which is a substrate that satisfies the analysis conditions, to determine an aerial discharge applied voltage peak value having the smallest absolute value among the applied voltage peak values ​​at which aerial discharge occurs on the analysis substrate after coating, and determines a discharge suppression applied voltage waveform, which is a voltage waveform applied to a conductor on the analysis substrate The learning device according to any one of Supplementary Notes 1 to 3, characterized in that by performing at least one of finite element analysis and equivalent circuit analysis on the analysis substrate, a dielectric breakdown applied voltage peak value having the smallest absolute value among applied voltage peak values ​​at which dielectric breakdown occurs in the analysis coating material that is the coating material of the analysis substrate after coating is determined, a breakdown prevention applied voltage waveform is determined, which is a voltage waveform to be applied to a conductor on the analysis substrate, the absolute value of which is smaller than the dielectric breakdown applied voltage peak value and which prevents dielectric breakdown in the analysis coating material after coating, and the breakdown prevention applied voltage waveform is used as the applied voltage waveform. (Supplementary Note 7) A substrate coating system that coats a substrate having a conductor and an electronic component with a coating material, comprising:a data acquisition unit that acquires as input data substrate information including information on the type and shape of the substrate, conductor information including information on the type and pattern of the conductors, electronic component mounting information including information on the type, shape and placement position of the electronic components, coating material information including information on the insulating performance of the coating material, and substrate use environment information including the waveform of a voltage applied to the conductors on the substrate; an inference unit that outputs output data from the input data using a trained model for inferring output data which is coating conditions under which, in the substrate coated under conditions specified by the input data, dielectric breakdown of the coating material after coating is suppressed or aerial discharge in the substrate after coating is suppressed; and a coating device that coats the substrate with the coating material based on the output data, wherein the coating conditions are at least one of a coating thickness distribution of the coating material on the substrate and the number of times the coating material is applied to the substrate. (Supplementary Note 8) The substrate coating system of Supplementary Note 7, further comprising a database device that stores the substrate information, the conductor information, the electronic component mounting information, the coating material information, and the substrate use environment information, wherein the data acquisition unit acquires the substrate information, the conductor information, the electronic component mounting information, the coating material information, and the substrate use environment information from the database device. (Supplementary Note 9) The substrate coating system of Supplementary Note 7, further comprising a scanning device that performs a three-dimensional scan of the appearance of the substrate including the conductors and the electronic components, and generates the substrate information, the conductor information, and the electronic component mounting information, wherein the data acquisition unit acquires the substrate information, the conductor information, and the electronic component mounting information from the scanning device. (Supplementary Note 10) The substrate coating system of any one of Supplements 7 to 9, further comprising: the coating material information includes at least one of the dielectric constant of the coating material, the conductivity of the coating material, and the dielectric breakdown field of the coating material. (Supplementary Note 11)A substrate coating system described in any one of Appendices 7 to 10, characterized in that the substrate usage environment information further includes at least one of the usage temperature of the substrate, the usage humidity of the substrate, the usage atmospheric pressure of the substrate, and the voltage application time to the conductor in the substrate. (Supplementary Note 12) A substrate coating method in a substrate coating system that coats a substrate having a conductor and an electronic component with a coating material, comprising: a data acquisition step of acquiring, as input data, substrate information including information on the type and shape of the substrate, conductor information including information on the type and pattern of the conductor, electronic component mounting information including information on the type, shape, and placement position of the electronic component, coating material information including information on the insulating performance of the coating material, and substrate use environment information including an applied voltage waveform to the conductor on the substrate; an inference step of outputting output data from the input data using a trained model for inferring, from the input data, coating conditions under which dielectric breakdown of the coating material after coating is suppressed or aerial discharge on the substrate after coating is suppressed in the substrate coated under conditions indicated by the input data; and a coating step of coating the substrate with the coating material based on the output data, wherein the coating conditions are at least one of a coating thickness distribution of the coating material on the substrate and the number of times the coating material is applied to the substrate. (Supplementary Note 13) The substrate coating method according to Supplementary Note 12, characterized in that the data acquisition step acquires the substrate information, the conductor information, the electronic component mounting information, the coating material information, and the substrate use environment information from a database device. (Supplementary Note 14) The substrate coating method according to Supplementary Note 12, further comprising a scanning step of three-dimensionally scanning the appearance of the substrate comprising the conductors and the electronic components to generate the substrate information, the conductor information, and the electronic component mounting information, and the data acquisition step acquires the substrate information, the conductor information, and the electronic component mounting information generated in the scanning step.

[0057] 1, 1a Learning device, 2 Learned model storage unit, 3 Inference device, 4 Coating device, 5 Database device, 6 Scanning device, 11 Learning data acquisition unit, 12 Model generation unit, 13 Calculation unit, 31 Data acquisition unit, 32 Inference unit, 100, 100a, 100b Substrate coating system, 201 Processor, 202 Memory, 203 Bus.

Claims

1. A learning device for obtaining a trained model for inferring coating conditions in a substrate coating system that coats a substrate having a conductor and electronic components with a coating material, the learning device comprising: a training data acquisition unit that acquires training data which is a combination of first input data and second input data; and a model generation unit that uses the training data to generate a trained model for inferring the second input data from the first input data, wherein the first input data includes substrate information including information on the type and shape of the substrate, conductor information including information on the type and pattern of the conductor, electronic component mounting information including information on the type, shape and placement position of the electronic components, coating material information including information on the insulating performance of the coating material, and substrate use environment information including the waveform of a voltage applied to the conductor on the substrate, and the second input data are coating conditions that, on the substrate coated under the conditions indicated by the first input data, prevent dielectric breakdown of the coating material after coating or prevent aerial discharge on the substrate after coating, and the coating conditions are at least one of the coating thickness distribution of the coating material on the substrate and the number of times the coating material is applied to the substrate.

2. The learning device according to claim 1, wherein the coating material information includes at least one of the dielectric constant of the coating material, the conductivity of the coating material, and the dielectric breakdown field of the coating material.

3. A learning device as described in claim 1 or 2, characterized in that the substrate usage environment information further includes at least one of the operating temperature of the substrate, the operating humidity of the substrate, the operating atmospheric pressure of the substrate, and the time of voltage application to the conductor on the substrate.

4. The experimental conditions include experimental coating conditions, the substrate information, the conductor information, the electronic component mounting information, and the coating material information, and the experimental coating conditions include, if the coating conditions include a coating thickness distribution of the coating material on the substrate, an experimental coating thickness distribution which is the coating thickness distribution of the coating material on the substrate in the test, and if the coating conditions include a number of times the coating material is reapplied to the substrate, an experimental number of reapplications which is the number of times the coating material is reapplied to the substrate in the test, and when the second input data are coating conditions which suppress aerial discharge on the substrate after coating, a partial discharge test on an experimental substrate that satisfies the experimental conditions, to determine an aerial discharge applied voltage peak value that has the smallest absolute value among the applied voltage peak values ​​that cause aerial discharge on the experimental substrate after coating, to determine a discharge-suppressing applied voltage waveform that is a voltage waveform applied to a conductor on the experimental substrate that has an absolute voltage peak value that is smaller than the aerial discharge applied voltage peak value and that suppresses aerial discharge on the experimental substrate after coating, and to determine the discharge-suppressing applied voltage waveform as the applied voltage waveform included in the first input data, and when the second input data are coating conditions that suppress dielectric breakdown of the coating material after coating, A learning device described in any one of claims 1 to 3, characterized in that a breakdown applied voltage peak value is determined which has the smallest absolute value among the applied voltage peak values ​​at which breakdown occurs in the experimental coating material, which is the coating material of the experimental substrate after coating, through a breakdown test on the experimental substrate, and a breakdown prevention applied voltage waveform is determined which is a voltage waveform applied to a conductor on the experimental substrate whose absolute value is smaller than the breakdown applied voltage peak value and which prevents breakdown of the experimental coating material after coating, and the breakdown prevention applied voltage waveform is used as the applied voltage waveform included in the first input data.

5. The analysis conditions include analysis coating conditions, the substrate information, the conductor information, the electronic component mounting information, and the coating material information, and the analysis coating conditions include, if the coating conditions include a coating thickness distribution of the coating material on the substrate, an analysis coating thickness distribution which is the coating thickness distribution of the coating material on the substrate in the analysis, and if the coating conditions include a number of times the coating material is reapplied to the substrate, an analysis number of reapplications which is the number of times the coating material is reapplied to the substrate in the analysis, and when the second input data are coating conditions which suppress aerial discharge on the substrate after coating, By performing at least one of finite element analysis and equivalent circuit analysis on an analysis substrate that is a substrate that satisfies the analysis conditions, an aerial discharge applied voltage peak value having the smallest absolute value among the applied voltage peak values ​​at which aerial discharge occurs on the analysis substrate after coating is determined, and a discharge suppression applied voltage waveform is determined, which is a voltage waveform applied to a conductor on the analysis substrate whose absolute value is smaller than the aerial discharge applied voltage peak value and at which aerial discharge is suppressed on the analysis substrate after coating, and the discharge suppression applied voltage waveform is set as the applied voltage waveform included in the first input data, and when the second input data are coating conditions at which dielectric breakdown of the coating material after coating is suppressed, The learning device described in any one of claims 1 to 3, characterized in that by performing at least one of finite element analysis and equivalent circuit analysis on the analysis substrate, a breakdown applied voltage peak value is determined which has the smallest absolute value among the applied voltage peak values ​​at which dielectric breakdown occurs in the analysis coating material, which is the coating material of the analysis substrate after coating, and a breakdown-prevention applied voltage waveform is determined which is a voltage waveform applied to a conductor on the analysis substrate whose absolute value is smaller than the breakdown applied voltage peak value and which prevents dielectric breakdown in the analysis coating material after coating, and the breakdown-prevention applied voltage waveform is used as the applied voltage waveform included in the first input data.

6. A calculation unit is provided which acquires, as analysis conditions, analysis coating conditions, the substrate information, the conductor information, the electronic component mounting information, and the coating material information, calculates the applied voltage waveform from the information on the analysis conditions, outputs the substrate information, the conductor information, the electronic component mounting information, the coating material information, and the applied voltage waveform to the learning data acquisition unit as the first input data, and outputs the analysis coating conditions to the learning data acquisition unit as the second input data, wherein the analysis coating conditions include, if the coating conditions include a coating thickness distribution of the coating material on the substrate, an analysis coating thickness distribution which is the coating thickness distribution of the coating material on the substrate in the analysis, and if the coating conditions include a number of reapplications of the coating material on the substrate, an analysis number of reapplications which is the number of reapplications of the coating material on the substrate in the analysis, and the calculation unit, when the second input data are coating conditions which suppress aerial discharge on the substrate after coating, By performing at least one of finite element analysis and equivalent circuit analysis on an analysis substrate that is a substrate that satisfies the analysis conditions, an aerial discharge applied voltage peak value having the smallest absolute value among the applied voltage peak values ​​at which aerial discharge occurs on the analysis substrate after coating is determined, and a discharge-suppression applied voltage waveform is determined, which is a voltage waveform applied to a conductor on the analysis substrate whose absolute value is smaller than the aerial discharge applied voltage peak value and which suppresses aerial discharge on the analysis substrate after coating, and the discharge-suppression applied voltage waveform is used as the applied voltage waveform, and when the second input data are coating conditions at which dielectric breakdown of the coating material after coating is suppressed,The learning device described in any one of claims 1 to 3, characterized in that at least one of finite element analysis and equivalent circuit analysis is performed on the analysis substrate to determine a breakdown applied voltage peak value which has the smallest absolute value among the applied voltage peak values ​​at which breakdown occurs in the analysis coating material, which is the coating material of the analysis substrate after coating, and to determine a breakdown-prevention applied voltage waveform which is a voltage waveform applied to a conductor on the analysis substrate whose absolute value is smaller than the breakdown applied voltage peak value and which prevents breakdown in the analysis coating material after coating, and to use the breakdown-prevention applied voltage waveform as the applied voltage waveform.

7. A substrate coating system for coating a substrate having a conductor and electronic components with a coating material, comprising: a data acquisition unit that acquires as input data substrate information including information on the type and shape of the substrate, conductor information including information on the type and pattern of the conductor, electronic component mounting information including information on the type, shape and position of the electronic components, coating material information including information on the insulating performance of the coating material, and substrate use environment information including the waveform of a voltage applied to the conductor on the substrate; an inference unit that outputs output data from the input data using a trained model for inferring output data which are coating conditions under which, in the substrate coated under the conditions indicated by the input data, dielectric breakdown of the coating material after coating is suppressed or aerial discharge on the substrate after coating is suppressed; and a coating device that coats the substrate with the coating material based on the output data, wherein the coating conditions are at least one of a coating thickness distribution of the coating material on the substrate and the number of times the coating material is applied to the substrate.

8. A substrate coating system as described in claim 7, further comprising a database device that stores the substrate information, the conductor information, the electronic component mounting information, the coating material information, and the substrate use environment information, and the data acquisition unit acquires the substrate information, the conductor information, the electronic component mounting information, the coating material information, and the substrate use environment information from the database device.

9. A substrate coating system as described in claim 7, characterized in that it comprises a scanning device that performs a three-dimensional scan of the appearance of the substrate having the conductors and the electronic components and generates the substrate information, the conductor information and the electronic component mounting information, and the data acquisition unit acquires the substrate information, the conductor information and the electronic component mounting information from the scanning device.

10. A substrate coating system as described in any one of claims 7 to 9, characterized in that the coating material information includes at least one of the dielectric constant of the coating material, the conductivity of the coating material, and the dielectric breakdown field of the coating material.

11. A substrate coating system as described in any one of claims 7 to 10, characterized in that the substrate usage environment information further includes at least one of the usage temperature of the substrate, the usage humidity of the substrate, the usage atmospheric pressure of the substrate, and the time of voltage application to the conductor on the substrate.

12. A substrate coating method in a substrate coating system that coats a substrate having a conductor and electronic components with a coating material, comprising: a data acquisition step of acquiring as input data substrate information including information on the type and shape of the substrate, conductor information including information on the type and pattern of the conductor, electronic component mounting information including information on the type, shape and position of the electronic components, coating material information including information on the insulating performance of the coating material, and substrate use environment information including the waveform of a voltage applied to the conductor on the substrate; an inference step of outputting output data from the input data using a trained model for inferring output data that is a coating condition under the conditions indicated by the input data, such that dielectric breakdown of the coating material after coating is suppressed or aerial discharge on the substrate after coating is suppressed; and a coating step of coating the substrate with the coating material based on the output data, wherein the coating condition is at least one of a coating thickness distribution of the coating material on the substrate and the number of times the coating material is applied to the substrate.

13. A substrate coating method as described in claim 12, characterized in that the data acquisition step acquires the substrate information, the conductor information, the electronic component mounting information, the coating material information, and the substrate use environment information from a database device.

14. A substrate coating method as described in claim 12, further comprising a scanning step of three-dimensionally scanning the appearance of the substrate having the conductor and the electronic component to generate the substrate information, the conductor information, and the electronic component mounting information, wherein the data acquisition step acquires the substrate information, the conductor information, and the electronic component mounting information generated in the scanning step.

Images