WO2023153282A1

WO2023153282A1 - Inspection device, inspection method, trained model generating device, inspection program, and trained model generating program

Info

Publication number: WO2023153282A1
Application number: PCT/JP2023/003157
Authority: WO
Inventors: 昌史小林
Original assignee: 日置電機株式会社
Priority date: 2022-02-08
Filing date: 2023-02-01
Publication date: 2023-08-17
Also published as: JP2023115771A; JP7308314B1

Abstract

The present invention improves reliability of inspection of electronic parts.　An inspection device (2) calculates an estimation value (Rse) of a measured value of alternating current resistance, using a trained model (35) that causes a computer to function so as to calculate a measured value (Rs) of alternating current resistance of a measurement object measured by a two-terminal method, on the basis of a measured value (Rdc4) of direct current resistance of the measurement object measured by a four-terminal method and a measured value (Rdc2) of direct current resistance of the measurement object measured by the two-terminal method.

Description

Inspection device, inspection method, trained model generation device, inspection program, and trained model generation program

The present invention relates to an inspection device, an inspection method, a learned model generation device, an inspection program, and a learned model generation program, and for example, to an inspection device for inspecting inductor elements.

Conventionally, inspection devices are known that measure the electrical characteristics of electronic components such as chip inductors and judge the quality of electronic components based on the measurement results. For example, Patent Document 1 discloses that the AC resistance and inductance of an inductor element to be inspected are measured, the Q value is calculated using the measured values, and the quality of the inductor element is determined based on the calculated Q value. An inspection device is disclosed.

In Patent Document 1, as a method of measuring the AC resistance of an inductor element, an estimated value of the contact resistance of a measuring probe used in the measurement by the two-terminal method is obtained from the measured value of the AC resistance measured by the two-terminal method. Calculating the AC resistance by subtraction is described. Further, in Patent Document 1, an estimated value of contact resistance is calculated by subtracting a measured value of DC resistance measured by a two-terminal method from a measured value of DC resistance measured by a four-terminal method, and an estimated value of contact resistance is multiplied by a coefficient of 0 or more and 1 or less to correct the measured value of AC resistance.

Japanese Patent No. 6949675

The inspection device disclosed in Patent Document 1 corrects the measured value of the AC resistance on the premise that the relationship between the series resistance and the AC resistance in the inductor element is linear. However, the relationship between the series resistance and the AC resistance of the actual inductor element is unknown. For example, if the relationship is non-linear, there is a possibility that the measured value of the AC resistance will not be properly corrected. . In addition, the inspection device disclosed in Patent Document 1 corrects the measured value of AC resistance using a value obtained by multiplying the estimated value of contact resistance by a coefficient in order to avoid overcorrection of AC resistance. If the coefficients are not appropriate, AC resistance correction may not be performed properly.

The present invention has been made in view of the above-described problems, and aims to improve the reliability of inspection of electronic components.

An inspection apparatus according to a representative embodiment of the present invention comprises a first measurement value of the DC resistance of a measurement object measured by a four-terminal method, and a second measurement value of the DC resistance of the measurement object measured by a two-terminal method. A data acquisition unit that acquires a measured value and a third measured value of the AC resistance of the measurement object measured by the two-terminal method, and the third measured value based on the input first measured value and the second measured value a storage unit that stores a trained model for causing a computer to calculate a measured value; and the first measured value that is acquired by the data acquisition unit based on the trained model that is stored in the storage unit. and an estimating unit that calculates an estimated value of the third measured value corresponding to the second measured value.

According to the inspection device according to the present invention, it is possible to improve the reliability of inspection of electronic components.

1 is a diagram showing the configuration of a measurement system including a learned model generation device and an inspection device according to an embodiment; FIG. It is a figure which shows an example of a structure of the learned model production|generation apparatus in the inspection system which concerns on embodiment. 6 is a flow chart showing the flow of generation of a trained model by the trained model generating device according to the embodiment; It is a figure which shows an example of a structure of the data-processing control apparatus in the inspection apparatus which concerns on embodiment. FIG. 5 is a flow chart showing the flow of inspection by the inspection apparatus 2 according to the embodiment.

1. Outline of Embodiment First, an outline of a representative embodiment of the invention disclosed in the present application will be described. In the following description, as an example, reference numerals on the drawings corresponding to constituent elements of the invention are described with parentheses.

[1] An inspection apparatus (2) according to a representative embodiment of the present invention measures a first measurement value (Rdc4) of the DC resistance of an object to be measured (DUT) measured by the four-terminal method, and by the two-terminal method A data acquisition unit (21) for acquiring a second measured value (Rdc2) of the DC resistance of the measured object and a third measured value (Rs) of the AC resistance of the measured object measured by the two-terminal method. a storage unit (22) for storing a trained model (35) for causing a computer to calculate the third measured value based on the input first measured value and the second measured value; calculating an estimated value (Rse) of the third measured value corresponding to the first measured value and the second measured value acquired by the data acquisition unit, based on the trained model stored in the storage unit; and an estimation unit (23).

[2] In the inspection apparatus according to [1] above, the third measurement is performed based on the first measurement value and the second measurement value acquired by the data acquisition unit according to the estimated value of the third measurement value. A correction unit (24) may be further provided for performing a correction process of correcting the measured value and outputting the corrected third measured value as the value of the AC resistance of the object to be measured.

[3] In the inspection apparatus described in [2] above, the learned model includes a first model (g(Rdc4, Rdc2)) representing a resistance component caused by a measurement system using a two-terminal method, and the object to be measured. and a second model (h(Rdc4)) representing a resistance component caused by the first model based on the first measured value and the second measured value obtained by the data obtaining unit. The resistance component (Rc) caused by the measurement system by the two-terminal method is calculated according to the above, and as the correction process, the third measured value is corrected based on the resistance component caused by the measurement system by the two-terminal method. may

[4] In the inspection apparatus described in [3] above, the first model uses the first measured value and the second measured value as explanatory variables, and the resistance component caused by the measurement system according to the two-terminal method. The second model is a regression model in which the first measured value is an explanatory variable and the value of the resistance component caused by the object to be measured is a regression model in which the objective variable is the value of the first The first model and the second model are adjusted by machine-learning measured data for learning (34_1 to 34_n) in which the third measured value is associated with the first measured value and the second measured value. May contain parameters.

[5] In the inspection apparatus according to any one of [2] to [4] above, the correction unit calculates the estimated value of the third measurement value calculated by the estimation unit and the third measurement value obtained by the data acquisition unit. The error (|Rse-Rs|) from the three measured values may be calculated, and the correction process may be performed when the error is smaller than the threshold value (Rth).

[6] The inspection device (2) according to another representative embodiment of the present invention includes a first measurement value (Rdc4) of the DC resistance of the measurement object measured by the four-terminal method, and a data acquisition unit (21) for acquiring a second measured value (Rdc2) of the DC resistance of the object to be measured and a third measured value (Rs) of the AC resistance of the object to be measured measured by the two-terminal method; , a storage unit (22) for storing a first model (35) indicating a correspondence relationship between the first measured value and the second measured value and the third measured value; and the first model stored in the storage unit (22). an estimating unit (23) for calculating an estimated value (Rse) of the third measured value corresponding to the first measured value and the second measured value acquired by the data acquiring unit, based on a model; Calculate the error (|Rse-Rs|) between the estimated value of the third measured value calculated by and the third measured value acquired by the data acquisition unit, and if the error is smaller than the threshold value (Rth), Correction for correcting the third measured value based on the first measured value and the second measured value obtained by the data obtaining unit, and outputting the corrected third measured value as an AC resistance value of the measurement object. and a correction unit (24) that performs processing.

[7] In the inspection apparatus described in [6] above, the first model uses the first measured value and the second measured value as explanatory variables, and the value of the resistance component resulting from the measurement system according to the two-terminal method. and a third model having the first measured value as an explanatory variable and the value of the resistance component caused by the object to be measured as an objective variable, wherein the correction unit includes the correction As a process, based on the first measured value and the second measured value acquired by the data acquisition unit, a resistance component caused by the measurement system by the two-terminal method is calculated according to the second model, and the two-terminal method The third measured value may be corrected based on the resistance component caused by the measurement system, and the corrected third measured value may be output as the value of the AC resistance of the object to be measured.

[8] The trained model generation device (3) according to the representative embodiment of the present invention provides the first measured value (Rdc4) of the DC resistance of the measurement object measured by the four-terminal method, and the Learning measurement data (34_1) in which the third measurement value (Rs) of the AC resistance of the measurement object measured by the two-terminal method is associated with the second measurement value (Rdc2) of the DC resistance of the measurement object that has been measured 34_n), and machine-learning the measured data for learning to obtain the third measured data based on the input data including the first measured value and the second measured value. a trained model generation unit (32) for generating a trained model (35) for causing a computer to function to calculate the measured value.

[9] In the trained model generation device according to [8] above, the trained model uses the first measured value and the second measured value as explanatory variables, and the resistance caused by the measurement system by the two-terminal method. A first regression model (g (Rdc4, Rdc2)) whose objective variable is the value of the component, and a second regression model (g (Rdc4, Rdc2)) whose explanatory variable is the first measured value and whose objective variable is the value of the resistance component caused by the measurement object. and a regression model (h(Rdc4)), wherein the learned model generation unit adjusts the learned parameters of the first regression model and the second regression model by performing machine learning on the learning measurement data. may

[10] An inspection method according to a representative embodiment of the present invention includes a first measurement value (Rdc4) of the DC resistance of a measurement object measured by a four-terminal method, and the measurement object measured by a two-terminal method A first step (S11 to S14) of acquiring a second measured value (Rdc2) of the DC resistance of and a third measured value (Rs) of the AC resistance of the measurement object measured by the two-terminal method, and input said first measured value obtained by said first step based on a trained model (35) for operating a computer to estimate said third measured value based on said first measured value and said second measured value; A second step (S15) of calculating an estimated value (Rse) of the third measured value corresponding to the measured value and the second measured value, and according to the estimated value of the third measured value, in the first step A third step of correcting the third measured value based on the obtained first measured value and the second measured value, and outputting the corrected third measured value as an AC resistance value (Rsr) of the object to be measured. (S17, S18).

[11] An inspection method according to another representative embodiment of the present invention includes a first measurement value (Rdc4) of the DC resistance of the measurement object measured by the four-terminal method, and the measurement measured by the two-terminal method A first step (S11 to S14) of acquiring a second measured value (Rdc2) of the DC resistance of the object and a third measured value (Rs) of the AC resistance of the object measured by the two-terminal method; the first measured value and the second measurement obtained in the first step, based on a first model (35) showing the correspondence relationship between the first measured value and the second measured value and the third measured value; A second step (S15) of calculating an estimated value (Rse) of the third measured value corresponding to the value, and the estimated value of the third measured value calculated in the second step and the obtained in the first step A third step (S16) of calculating an error (|Rse−Rs|) with a third measured value, and if the error is smaller than a threshold value (Rth), the first measured value obtained in the first step and a fourth step (S17, S18) of correcting the third measured value based on the second measured value and outputting the corrected third measured value as the value of the AC resistance of the object to be measured; and a fifth step (S19) of outputting the third measured value obtained in the first step as the value of the AC resistance of the object to be measured when the measured value is greater than a threshold value.

[12] An inspection program according to a representative embodiment of the present invention is characterized by causing a computer to execute each step in the inspection method described in [10] or [11] above.

[13] A trained model generation method according to a representative embodiment of the present invention includes a first measurement value (Rdc4) of the DC resistance of the measurement object measured by the four-terminal method, and Measurement data for learning (34_1 to 34_n) in which the third measurement value (Rs) of the AC resistance of the measurement object measured by the two-terminal method is associated with the second measurement value (Rdc2) of the DC resistance of the measurement object and performing machine learning on the learning measurement data acquired in the first step, based on the input data including the first measurement value and the second measurement value and a second step (S3) of generating a trained model (35) for activating a computer to calculate said third measure.

[14] A trained model generation program according to a representative embodiment of the present invention is characterized by causing a computer to execute each step in the trained model generation method described in [13] above.

2. Specific Examples of Embodiments Specific examples of embodiments of the present invention will be described below with reference to the drawings. In the following description, constituent elements common to each embodiment are denoted by the same reference numerals, and repeated descriptions are omitted.

FIG. 1 is a diagram showing the configuration of an inspection system 1 including a trained model generation device 3 and an inspection device 2 according to an embodiment.

The inspection system 1 shown in FIG. 1 is a system for inspecting the quality of an object to be measured (hereinafter also referred to as "DUT"). As shown in FIG. 1, an inspection system 1 includes a trained model generation device 3 that generates a trained model by learning measurement data for learning based on a plurality of measurement results of a DUT by machine learning, and a trained model that has been generated. and an inspection device 2 for inspecting the DUT using a.

The inspection device 2 is a device that measures the electrical characteristics of the DUT and inspects the quality of the DUT based on the measurement results. For example, the inspection device 2 is a device (a so-called chip taping machine) that inspects the quality of small electronic components (chip components) and packages the chip components determined to be non-defective in a state ready for shipment.

In the present embodiment, the case where the DUT is an inductor element (eg, chip inductor element) will be described as an example, but it is not limited to this.

The inspection device 2 measures the electrical characteristics of the inductor element as the DUT using the learned model described later. Specifically, the inspection apparatus 2 includes a data processing control device 10, a first measurement section 11, a second measurement section 12, an operation section 13, an output section 14, and a transport mechanism 15, as shown in FIG. .

The first measurement unit 11 is a device that measures the electrical characteristics of an inductor element as a DUT by the four-terminal method. As the first measurement unit 11, an impedance measuring instrument such as a resistance meter or an LCR meter capable of measuring impedance by the four-terminal method can be exemplified.

The second measurement unit 12 is a device that measures the electrical characteristics of an inductor element as a DUT by the two-terminal method. As the second measuring unit 12, an impedance measuring instrument such as an LCR meter capable of measuring impedance by a two-terminal method can be exemplified.

Note that the first measurement unit 11 and the second measurement unit 12 are not limited to the above examples as long as they are devices capable of measuring electrical characteristics such as impedance of the DUT.

The first measurement unit 11 measures the DC resistance of the inductor element as the DUT by the four-terminal method in accordance with the instruction from the data processing control device 10 . For example, the first measurement unit 11 includes a moving mechanism (not shown) that moves the probes 61a to 61d, a current output unit and a voltage detection unit (not shown), and a measurement value calculation that calculates a measurement value based on the detection result. (not shown).

For example, when the first measurement unit 11 receives an instruction to perform measurement from the data processing control device 10, the movement mechanism of the first measurement unit 11 connects one terminal of the inductor element transported to the predetermined measurement position. The

probes

61a and 61c are brought into contact, and the other terminals of the inductor elements are brought into contact with the

probes

61b and 61d. Next, the current output section of the first measuring section 11 supplies a DC current to the inductor element via the

probes

61a and 61b. The voltage detection section of the first measurement section 11 detects the voltage value between the inductor terminals through the

probes

61c and 61d when the DC current is supplied to the inductor element. The measured value calculator of the first measuring unit 11 calculates a measured value Rdc4 of the DC resistance of the inductor element based on the detected voltage value and the current value of the DC current supplied to the inductor element.

The second measurement unit 12 measures the DC resistance, AC resistance, and inductance of the inductor element as the DUT according to the instruction from the data processing control device 10 by the two-terminal method. For example, the second measurement unit 12 includes a moving mechanism (not shown) that moves the probes 62a and 62b, a current output unit and a voltage detection unit (not shown), and a measurement value calculator that calculates a measurement value based on the detection result. (not shown).

For example, when the second measurement unit 12 receives an instruction to perform measurement from the data processing control device 10, the movement mechanism of the second measurement unit 12 connects one terminal of the inductor element transported to the predetermined measurement position. The probe 62a is brought into contact, and the other terminal of the inductor element is brought into contact with the probe 62b. Next, the current output section of the second measuring section 12 supplies a DC current to the inductor element via the probes 62a and 62b, and the voltage detecting section of the second measuring section 12 detects the voltage value between both terminals of the inductor element. are detected via probes 62a and 62b. The measured value calculator of the second measuring unit 12 calculates a measured value Rdc2 of the DC resistance of the inductor element based on the detected voltage value and the current value of the DC current supplied to the inductor element.

In addition, in a state in which the probes 62a and 62b are brought into contact with both terminals of the inductor element by the moving mechanism of the second measuring section 12, the current output section of the second measuring section 12 outputs an alternating current to the inductor element through the probe. The voltage detection section of the second measurement section 12 detects the AC voltage value between both terminals of the inductor element via the probes 62a and 62b. The measured value calculator of the second measuring unit 12 calculates the detected AC voltage value (voltage effective value), the AC current value (current effective value) of the AC current supplied to the inductor element, and the position of the AC voltage and the AC current. A measured value Rs of the AC resistance and a measured value L of the inductance of the inductor element are calculated based on the phase difference.

Some functions of the first measurement unit 11 and the second measurement unit 12 may be realized by the data processing control device 10. For example, the data processing control device 10 may perform the above calculations by the measurement value calculation units of the first measurement unit 11 and the second measurement unit 12 .

The operation unit 13 is an input interface for the user to operate the inspection device 2 . Various buttons, a touch panel, and the like can be exemplified as the operation unit 13 . For example, by operating the operation unit 13, the user sets various inspection conditions and the like for inspecting an inductor element as a DUT in the inspection device 2, and instructs the inspection device 2 to perform and stop inspection and the like. can be done.

The output unit 14 is a functional unit for outputting various information such as inspection conditions and inspection results in the inspection device 2 . The output unit 14 is, for example, a display device equipped with an LCD (Liquid Crystal Display) or an organic EL. For example, the output unit 14 displays information such as inspection results on the screen according to the control by the data processing control device 10 when the user operates the operation unit 13 to instruct execution of inspection of the DUT.

It should be noted that the output unit 14 may be a display device having a touch panel that realizes a part of the functions of the operation unit 13. In addition, the output unit 14 may include a communication circuit or the like for outputting data such as test results to the outside by wire or wirelessly.

The transport mechanism 15 is a device that transports the inductor element to be inspected to an appropriate location within the inspection device 2 under the control of the data processing control device 10 . For example, when performing measurement by the first measurement unit 11 , the transport mechanism 15 transports the inductor element to be inspected to a predetermined measurement position by the first measurement unit 11 . Further, for example, when the second measurement unit 12 performs measurement, the transport mechanism 15 transports the inductor element to be inspected to a predetermined measurement position by the second measurement unit 12 . Further, the transport mechanism 15 transports inductor elements determined to be non-defective among the inductor elements that have been inspected to a position for packaging, and transports the packaged inductor elements to a predetermined location in the next step.

The data processing control device 10 is a functional unit that comprehensively controls each functional unit in the inspection device 2 and performs various data processing for inspection of the DUT. For example, the data processing control device 10 is a program processing device having a processor such as a CPU, storage devices such as ROM, RAM, and flash memory, and peripheral circuits such as timers. Examples of program processing devices include MCUs and FPGAs.

The data processing control device 10 acquires the measurement results from the first measurement unit 11 and the second measurement unit 12, calculates an index indicating the performance of the inductor element based on the acquired measurement results, and based on the calculated index, Determining whether the inductor element to be inspected is good or bad. Here, the index indicating the performance of the inductor element is, for example, the Q value.

As described above, when the AC resistance of an inductor element is measured by the two-terminal method, the measured value is affected by the resistance component caused by the measurement system by the two-terminal method. Therefore, the data processing control device 10 of the inspection device 2 according to the present embodiment provides an index (Q value) indicating the performance of the inductor element to be inspected based on the measurement results of the first measurement unit 11 and the second measurement unit 12. is calculated, if necessary, the measured value Rs of the AC resistance measured by the second measuring unit 12 is corrected using a pre-generated learned model.

Here, before describing in detail the correction of the AC resistance measurement value Rs using the learned model performed by the data processing control device 10, the method of generating the learned model will be described.

FIG. 2 is a diagram showing an example of the configuration of the learned model generation device 3 in the inspection system 1 according to the embodiment.

The trained model generation device 3 is realized by, for example, an information processing device (computer) such as a server or a personal computer (PC), and generates a plurality of learning measurement data 34_1 to 34_n ( n is an integer equal to or greater than 2), and machine-learning the generated measurement data for learning based on a predetermined algorithm to generate a trained model 35 .

For convenience of explanation, the trained model generation device 3 and the inspection device 2 are arranged side by side in FIG. No need. For example, the inspection device 2 and the trained model generation device 3 may be installed at different locations and connected via a communication network such as a LAN or the Internet. In this case, the inspection device 2 and the learned model generation device 3 may transmit and receive various data such as measurement data by the inspection device 2 and the learned model 35 to and from each other via a communication network.

In addition, the inspection device 2 and the learned model generation device 3 do not have to be electrically connected to each other during inspection of the DUT. For example, before or after inspection, measurement data by the inspection device 2 and various data such as the learned model 35 may be exchanged via a storage medium such as a memory card.

In the present embodiment, the learned model 35 is a model for estimating the measured value Rs of the AC resistance of the inductor element to be inspected. The trained model 35 generated by the trained model generation device 3 may be distributed via a network, or may be stored in a computer-readable storage medium (non-transitory computer readable medium) such as a memory card. It may be written in and circulated.

The trained model generation device 3 has, for example, a learning measurement data acquisition unit 31, a trained model generation unit 32, and a storage unit 33 as functional blocks for generating a trained model 35. Each of these functional blocks is composed of hardware resources such as a CPU and a memory that constitute an information processing device as the trained model generation device 3, and software installed in the information processing device (including a trained model generation program). It is realized by collaborating with various programs).

The learning measurement data acquisition unit 31 is a functional unit that acquires the learning measurement data 34_1 to 34_n necessary for generating the trained model 35.

Here, the measured data for learning 34_1 to 34_n are the measured value Rdc4 (first measured value) of the DC resistance of the DUT measured by the four-terminal method and the measured value Rdc2 (first measured value) of the DC resistance of the DUT measured by the two-terminal method. 2 measured value) and the measured value Rs (third measured value) of the AC resistance of the DUT measured by the two-terminal method.
In the following description, when the learning measurement data 34_1 to 34_n are not distinguished, they are simply referred to as "learning measurement data 34".

The learning measurement data acquisition unit 31 acquires the data 41 of the measured value Rdc4 of the DC resistance of the inductor element measured by the four-terminal method, for example, via wireless or wired communication (not shown) or a storage medium such as a memory card. , a data pair including data 42 of the measured value Rdc2 of the DC resistance of the inductor element measured by the two-terminal method and data 43 of the measured value Rs of the AC resistance measured by the two-terminal method. The learning measurement data acquisition unit 31 acquires a data pair for each inspected inductor element. As data pairs, for example, measurement results of inductor elements inspected by the inspection apparatus 2 or the like in the past may be used.

For example, the learning measurement data acquisition unit 31 associates the DC resistance measurement values Rdc4 and Rdc2 included in the acquired data pair with the AC resistance measurement value Rs included in the data pair as a correct value, thereby obtaining one Generate measurement data for learning 34 . The learning measurement data acquisition unit 31 generates learning measurement data 34_1 to 34_n for each measurement result of the inspected inductor element, and stores the learning measurement data 34_1 to 34_n in the storage unit 33 .

As described above, instead of generating the learning measurement data 34 by itself, the learning measurement data acquiring unit 31 receives the learning measurement data 34 generated by another information processing device or the like via communication or a storage medium. may be obtained by

The trained model generation unit 32 is a functional unit that generates a trained model 35 by performing machine learning on a plurality of learning measurement data 34_1 to 34_n acquired by the learning measurement data acquisition unit 31.

The trained model 35 is a program generated by machine learning based on a predetermined algorithm. Examples of the predetermined algorithm include polynomial regression, multiple regression, and the like.

In this embodiment, the trained model 35 includes a measured value Rdc4 (first measured value) of the DC resistance of the DUT measured by the four-terminal method and a measured value Rdc2 (second measured value) of the DC resistance of the DUT measured by the two-terminal method. A function for causing a computer (MPU, etc.) to estimate the measured value Rs of the AC resistance of the DUT measured by the two-terminal method, based on the input data including the measured value).

In other words, the learned model 35 performs calculations based on predetermined learned parameters on the input measurement data (measured values Rdc4 and Rdc2 of the DC resistance), and quantifies the AC resistance based on the measurement data. It is a program for causing an information processing device (computer) to function so as to output the calculated value (estimated value).

For example, the learned model 35 includes a model representing the resistance component Rc caused by the measurement system by the two-terminal method and a model representing the resistance component caused by the object to be measured (DUT).

Here, the resistance component caused by the measurement system by the two-terminal method includes, for example, the resistance component of the line composed of cables, probes, etc. existing between the second measuring section 12 and the DUT, and the contact between the probe and the DUT. and a resistance component (so-called contact resistance) caused by the state.

The model (first model) representing the resistance component Rc caused by the measurement system by the two-terminal method is, for example, the measured value Rdc4 (first measured value) of the DC resistance by the four-terminal method and the measured value of the DC resistance by the two-terminal method. Rdc2 (second measured value) is an explanatory variable, and the value of the resistance component caused by the measurement system by the two-terminal method is a regression model as an objective variable. In other words, the model representing the resistance component (Rc) caused by the measurement system by the two-terminal method is a function for calculating the resistance component Rc caused by the measurement system by the two-terminal method from the measured values Rdc4 and Rdc2 of the DC resistance. .

A model (second model) representing the resistance component caused by the DUT has, for example, the DC resistance measured value Rdc4 (first measured value) by the four-terminal method as an explanatory variable, and the value of the resistance component caused by the DUT as the objective variable. It is a regression model with In other words, the model representing the resistance component caused by the DUT is a function that estimates the value of the AC resistance caused by the DUT from the measured value Rdc4 of the DC resistance by the four-terminal method.

When the model representing the resistance component (Rc) due to the measurement system by the two-terminal method is g (Rdc4, Rdc2) and the model representing the resistance component due to the DUT is h (Rdc4), the measured value of the AC resistance is estimated. The learned model 35 as a function for calculating the value Rse can be expressed as, for example, Rse=g(Rdc4, Rdc2)+h(Rdc4).

That is, the estimated value Rse of the measured value of the AC resistance is the resistance component Rc due to the measurement system by the two-terminal method obtained by the model g (Rdc4, Rdc2) and the resistance due to the DUT obtained by the model h (Rdc4). It is represented by the sum of the components.

Model g (Rdc4, Rdc2) and model h (Rdc4) contain learned parameters. The learned parameters are mechanically adjusted so as to calculate the estimated value Rse of the AC resistance measurement value using the learning measurement data 34_1 to 34_n as inputs to the learning program (the program based on the predetermined algorithm). are the coefficients of model g(Rdc4, Rdc2) and model h(Rdc4) obtained by

The learned model generation unit 32 adjusts the learned parameters of the model g (Rdc4, Rdc2) and the model h (Rdc4) by performing machine learning on the learning measurement data 34_1 to 34_n.

For example, the trained model generation unit 32 first generates an estimated value Rse of the measured value of the AC resistance calculated by inputting the learning measurement data 34_1 to 34_n into the regression model, and the measured value of the AC resistance which is the correct value. A difference (error) from the value Rs is calculated. Next, the learned model generation unit 32 generates the regression model by sequentially updating the learned parameters (coefficients) of the regression model so that the calculated error becomes smaller, for example, by the error back propagation method, It is stored in the storage unit 33 as a trained model 35 .

The storage unit 33 is a functional unit for storing various data such as the learning measurement data 34_1 to 34_n necessary for generating the trained model 35 and the generated trained model 35.

The storage unit 33 is configured to be accessible from the outside, for example. For example, the inspection device 2 can read and acquire the learned model 35 from the storage unit 33 by communicating with the trained model generation device 3 . Further, for example, the inspection device 2 can write the data of the measurement results and the like to the storage unit 33 by communicating with the trained model generation device 3 .

FIG. 3 is a flow chart showing the flow of generation of the learned model 35 by the trained model generation device 3 according to the embodiment.

As shown in FIG. 3, first, in the trained model generation device 3, the learning measurement data acquisition unit 31 acquires learning measurement data 34_1 to 34_n (step S1). Specifically, as described above, the learning measurement data acquisition unit 31 acquires a data pair including the DC resistance measurement values Rdc4 and Rdc2 and the AC resistance measurement value Rs of inductor elements tested in the past, Learning measurement data 34 is generated by adding the AC resistance measurement value Rs as a correct value to the DC resistance measurement values Rdc4 and Rdc2 based on the acquired data pair.

Next, the trained model generation device 3 determines whether or not the necessary number of learning measurement data 34 to generate the trained model 35 has been generated (step S2). For example, the trained model generation device 3 is preset with the number of learning measurement data 34 required to generate a trained model 35, and the trained model generation device 3 generates the learning measurement data 34 is generated, the number of generations is incremented by +1. Then, the trained model generation device 3 determines whether or not the number of times of generation that has been counted has reached the number of data set in advance, thereby determining whether or not the required number of measurement data for learning 34 has been generated. do.

If the required number of learning measurement data 34 has not been generated (step S2: NO), the trained model generating device 3 returns to step S1 to obtain data pairs related to new inductor element measurement results. Then, generation of the learning measurement data 34 regarding the inductor element is repeated (steps S1 and S2).

On the other hand, when the required number of learning measurement data 34 has been generated (step S2: YES), the trained model generation device 3 performs learning using the plurality of learning measurement data 34 generated in step S1. A finished model 35 is generated (step S3). Specifically, the learned model generation unit 32 performs machine learning based on a predetermined algorithm on the plurality of learning measurement data 34_1 to 34_n created in step S1 by the method described above, thereby generating the learned model 35 to create

Then, the learned model generation unit 32 stores the learned model 35 in the storage unit 33 when sufficient accuracy is obtained for the learned model 35 .

Next, the trained model 35 generated in step S3 is registered in the inspection device 2 (step S4). For example, the learned model generation device 3 stores the The stored learned model 35 is transmitted to the inspection device 2 , and the learned model 35 received by the inspection device 2 is stored in the storage unit in the data processing control device 10 . Note that the registration of the learned model 35 in the inspection apparatus 2 may be performed using a storage medium such as a memory card, as described above.

The learned model generation program for causing the computer (information processing device) as the trained model generation device 3 to execute the above-described steps (S1 to S4) may be distributed via a network. However, it may be distributed by being written in a computer-readable storage medium (non-transitory computer readable medium) such as a memory card.

The learned model 35 for inspecting the inductor element is generated by the method described above.

Next, the correction of the AC resistance measurement value Rs using the learned model 35 by the inspection device 2 will be described in detail.

FIG. 4 is a diagram showing an example of the configuration of the data processing control device 10 in the inspection device 2 according to the embodiment.

As shown in FIG. 4, the data processing control device 10 of the inspection device 2 has, for example, a data acquisition unit 21, a storage unit 22, an estimation unit 23, a correction unit 24, and a determination unit 25. These functional units are implemented by, for example, a program processing device as the data processing control device 10, in which the CPU executes various calculations according to programs stored in a memory and controls peripheral circuits such as counters. be.

The data acquisition unit 21 is a functional unit that acquires various data necessary for calculating an index (Q value) indicating the performance of the inductor element to be inspected.

The data acquisition unit 21 acquires, for example, the measured value Rdc4 of the DC resistance of the DUT measured by the first measurement unit 11 by the four-terminal method, and stores it in the storage unit 22 . The data acquisition unit 21 obtains, for example, a measured value Rdc2 of the DC resistance of the DUT measured by the second measuring unit 12 by the two-terminal method and a measured value Rs of the AC resistance of the DUT measured by the second measuring unit 12 by the two-terminal method. , and the measured value L of the inductance of the DUT measured by the second measurement unit 12 by the two-terminal method, and stored in the storage unit 22 as measurement data 50 to be inspected. Further, the data acquisition unit 21 acquires, for example, the trained model 35 generated by the trained model generation device 3 and stores it in the storage unit 22 .

The storage unit 22 is a functional unit for storing various data necessary for calculating an index (Q value) indicating the performance of the inductor element to be inspected, the calculated Q value, and the like.

As described above, the storage unit 22 stores the measured values Rdc4 and Rdc2 of the DC resistance of the inductor element, the measured value Rs of the AC resistance of the inductor element, and the measured value L of the inductance of the inductor element, which are acquired by the data acquiring unit 21. , and the trained model 35 are stored respectively. Further, the storage unit 22 stores, for example, an estimated value Rse of a measured value of AC resistance, an estimated value of a resistance component Rc caused by a measurement system using a two-terminal method, an AC resistance value Rsr, and a Q value, which will be described later. remembered.

The estimation unit 23 is a functional unit that estimates the measured value of the AC resistance of the inductor element to be inspected. Based on the learned model 35 stored in the storage unit 22, the estimation unit 23 calculates the measured values of the AC resistance corresponding to the measured values Rdc4 and Rdc2 of the DC resistance of the inductor element to be inspected acquired by the data acquisition unit 21. Calculate the estimated value Rse. Specifically, the estimation unit 23 inputs (substitutes) the measured values Rdc4 and Rdc2 of the DC resistance of the inductor element to be inspected acquired by the data acquisition unit 21 into the learned model 35 (function). The value is stored in the storage unit 22 as an estimated value Rse of the measured AC resistance.

The correction unit 24 is a functional unit for correcting the measured value Rs of the AC resistance.
The correction unit 24 corrects the measured value Rs of the AC resistance based on the measured values Rdc4 and Rdc2 of the DC resistance acquired by the data acquisition unit 21 according to the estimated value Rse of the measured value of the AC resistance, and obtains the corrected AC resistance A correction process is performed to output the measured value Rs of the DUT as the value Rsr of the AC resistance of the DUT.

More specifically, first, based on the measured values Rdc4 and Rdc2 of the DC resistance of the inductor element to be inspected acquired by the data acquisition unit 21, the correction unit 24 calculates 2 A resistance component Rc due to a measurement system based on the terminal method is calculated. Data of the calculated resistance component Rc is stored in the storage unit 22, for example.

Next, the correction unit 24 corrects the measured value Rs of the AC resistance of the inductor element to be inspected based on the resistance component Rc caused by the measurement system using the two-terminal method according to the estimated value Rse of the measured value of the AC resistance. Then, the corrected AC resistance measurement value Rs is output as the AC resistance value Rsr of the inductor element to be inspected.

For example, the correction unit 24 determines the error |Rse−Rs between the estimated value Rse of the AC resistance measurement value calculated by the estimation unit 23 and the AC resistance measurement value Rs (third measurement value) acquired by the data acquisition unit 21. | is calculated, and the error |Rse-Rs| is evaluated. Specifically, the correction unit 24 compares the error |Rse-Rs| with the threshold value Rth. The threshold Rth is an arbitrary preset value.

Here, if the error |Rse-Rs| is smaller than the threshold value Rth, it can be considered that the AC resistance estimation accuracy of the learned model 35 is high with respect to the measurement result of the inductor element to be inspected. That is, it is considered that the model g (Rdc4, Rdc2) included in the trained model 35 has high estimation accuracy of the resistance component Rc due to the measurement system by the two-terminal method.

Therefore, when the error |Rse-Rs| is smaller than the threshold value Rth, the correction unit 24 performs correction processing. Specifically, the correction unit 24 calculates the resistance component Rc caused by the measurement system by the two-terminal method using the model g (Rdc4, Rdc2), and calculates the calculated resistance component Rc caused by the measurement system by the two-terminal method. is used to correct the measured value Rs (third measured value) of the AC resistance.

For example, the correction unit 24 first inputs (substitutes) the measured values Rdc4 and Rdc2 of the DC resistance of the inductor element to be inspected into the model g (Rdc4 and Rdc2) to measure the values obtained by the two-terminal method. It is assumed that the resistance component caused by the system is Rc. Next, the correction unit 24 calculates the value obtained by subtracting the resistance component Rc caused by the measurement system by the two-terminal method from the AC resistance measurement value Rs (third measurement value) obtained by the data obtaining unit 21. , as an AC resistance value Rsr (=Rs-Rc).

On the other hand, if the error |Rse-Rs| is larger than the threshold value Rth, it is considered that the accuracy of AC resistance estimation by the learned model 35 is low for the inductor element to be inspected. That is, it is considered that the estimation accuracy of the resistance component Rc due to the measurement system by the two-terminal method by the model g (Rdc4, Rdc2) included in the trained model 35 is low.
In this case, it is assumed that the resistance component Rc caused by the measurement system by the two-terminal method is calculated using the model g (Rdc4, Rdc2), and the calculated resistance component Rc caused by the measurement system by the two-terminal method is used to If the resistance measurement value Rs (third measurement value) is corrected, the correction may be erroneous, and the AC resistance value Rsr may not be obtained appropriately.

Therefore, when the error |Rse−Rs| is larger than the threshold value Rth, the correction unit 24 converts the measured value Rs of the AC resistance acquired by the data acquisition unit 21 into the AC value of the inductor element to be inspected without performing the correction process. Output as a resistance value Rsr (=Rs).

The judgment unit 25 is a functional unit for judging whether the DUT (inductor element) is good or bad.
The determination unit 25 expresses the performance of the inductor element to be inspected based on the AC resistance value Rsr of the inductor element to be inspected and the measured inductance value L of the inductor element to be inspected, which are output from the correction unit 24. A Q value (Q=ωL/Rsr), which is an index, is calculated.

The judging unit 25 judges the quality of the inductor element to be inspected by, for example, comparing the calculated Q value with a predetermined reference value. The determining unit 25 controls the transport mechanism 15 to package the DUT determined as a non-defective product into a state ready for shipment by a packaging device (not shown).

Next, the flow of DUT inspection processing by the inspection device 2 will be described.

FIG. 5 is a flow chart showing the flow of inspection by the inspection device 2 according to the embodiment.

For example, when the user operates the operation unit 13 of the inspection device 2 to instruct execution of inspection of a DUT (inductor element), the data processing control device 10 starts inspection of the inductor element to be inspected.

First, the data processing control device 10 controls the first measuring unit 11 to measure the DC resistance of the inductor element to be inspected by the four-terminal method (step S11). For example, the data processing control device 10 controls the transport mechanism 15 according to an instruction signal from the operation unit 13 to transport the inductor element to be inspected to a predetermined measurement position in the first measurement unit 11 . After that, the data processing control device 10 controls the first measuring unit 11 to measure the DC resistance of the inductor element to be inspected by the four-terminal method, and acquires the measured value Rdc4 of the DC resistance.

Next, the data processing control device 10 controls the second measuring unit 12 to measure the DC resistance of the inductor element to be inspected by the two-terminal method (step S12). For example, the data processing control device 10 controls the transport mechanism 15 to transport the inductor element to be inspected to a predetermined measurement position in the second measuring section 12 . After that, the data processing control device 10 controls the second measuring section 12 to measure the DC resistance of the inductor element to be inspected by the two-terminal method, and acquires the measured value Rdc2 of the DC resistance.

Next, the data processing control device 10 controls the second measuring unit 12 to measure the measured value Rs of the AC resistance and the measured value L of the inductance of the inductor element to be inspected by the two-terminal method (step S13). . For example, in a state in which the inductor element to be inspected is placed at the same measurement position as in step S12, the data processing control device 10 controls the second measuring unit 12 to measure the AC resistance of the inductor element to be inspected, A measurement value Rs of AC resistance and a measurement value L of inductance are obtained respectively.

Next, the data processing control device 10 controls the second measuring section 12 to measure the DC resistance of the inductor element to be inspected by the two-terminal method in the same manner as in step S12 (step S14).

The measured values Rdc4 and Rdc2 of the DC resistance, the measured value Rs of the AC resistance, and the measured value L of the inductance acquired by the data processing control device 10 in steps S11 to S14 are stored as the measured data 50 of the inductor element to be inspected. 22.

Next, the data processing control device 10 calculates an estimated value Rse of the measured AC resistance of the inductor element to be inspected based on the measurement data 50 of the inductor element to be inspected (step S15). Specifically, the estimating unit 23 inputs the measured values Rdc4 and Rdc2 of the direct current resistance acquired in steps S11 and S12 (S14) to the learned model 35 by the method described above, so that from the learned model 35 An estimated value Rse of the measured value of the output AC resistance is obtained.

At this time, the estimating unit 23 compares, for example, the measured value Rdc2 of the DC resistance measured in step S12 with the measured value Rdc2 of the DC resistance measured in step S14, and uses the smaller measured value Rdc2 of the DC resistance. An estimated value Rse of the measured value of AC resistance may be calculated.

Next, the correction unit 24 determines whether or not the difference |Rse-Rs| (step S16).

If the difference |Rse−Rs| is smaller than the threshold value Rth (step S16: YES), the correction unit 24 uses the model g (Rdc4, Rdc2) to calculate the resistance caused by the measurement system by the two-terminal method. A component Rc is calculated (step S17). Specifically, the correction unit 24 inputs the measured values Rdc4 and Rdc2 of the direct current resistance obtained in steps S11 and S12 (S14) to the model g (Rdc4, Rdc2), so that the measurement system using the two-terminal method A resulting resistance component Rc is obtained.

Next, the correction unit 24 corrects the measured value Rs of the AC resistance obtained in step S13 using the resistance component Rc due to the measurement system by the two-terminal method calculated in step S17, and converts the corrected value to the AC resistance is output as a value Rsr (step S18). Specifically, the correction unit 24 subtracts the resistance component Rc due to the measurement system by the two-terminal method calculated in step S17 from the measured value Rs of the AC resistance obtained in step S13, and converts the value obtained by subtracting the value to the AC resistance. is output as the value Rsr (=Rs-Rc).

On the other hand, if the difference |Rse-Rs| is output as a value Rsr (=Rs) of (step S19).

Next, the determination unit 25 determines the Q of the inductor element to be inspected based on the AC resistance value Rsr output from the correction unit 24 in step S18 or step S19 and the measured inductance value L acquired in step S13. A value is calculated (step S20). After that, the judging section 25 judges whether the inductor element to be inspected is good or bad based on the Q value calculated in step S20. Inductor elements determined to be non-defective are transported by transport mechanism 15 and packaged.

The inspection program for causing the computer (information processing device) as the data processing control device 10 to execute the steps (S11 to S21) described above may be distributed via a network, or may be stored in a memory card. It may be distributed by being written on a computer-readable storage medium (non-transitory computer readable medium) such as.

As described above, in the inspection system 1 according to the present embodiment, the learned model generation device 3 includes the model g (Rdc4, Rdc2) for calculating the resistance component caused by the measurement system by the two-terminal method, and the DUT (inductor element). A trained model 35 including a model h (Rdc4) for calculating a resistance component caused by Then, a plurality of learning measurement data 34_1 to 34_n generated by labeling the measured value Rs of the AC resistance of the DUT measured by the two-terminal method are generated by machine learning.

According to this, even if the relationship between the measured values Rdc4, Rdc2 of the DC resistance of the inductor element and the measured value Rs of the AC resistance is non-linear, the relationship between the measured values Rdc4, Rdc2 of the DC resistance and the measured value Rs of the AC resistance A trained model 35 (function) that appropriately represents the relationship can be obtained.

In the trained model 35, the model g (Rdc4, Rdc2) is a regression model with the measured values Rdc4 and Rdc2 of the DC resistance as explanatory variables and the value of the resistance component resulting from the measurement system by the two-terminal method as the objective variable. and the model h(Rdc4) is a regression model with the measured value Rdc4 of the DC resistance as an explanatory variable and the value of the resistance component caused by the DUT as an objective variable. The model g (Rdc4, Rdc2) and the model h (Rdc4) are adjusted by machine learning a plurality of learning measurement data 34_1 to 34_n in which the measured DC resistance values Rdc4 and Rdc2 are associated with the measured AC resistance values Rs. contains the learned parameters (coefficients).

According to this, the trained model 35 can be represented by a simpler function, so it is possible to avoid black boxing of the trained model 35, which is a concern in machine learning.

Further, in the inspection system 1 according to the present embodiment, the inspection device 2 estimates the measured value of the AC resistance of the DUT using the learned model 35 generated by the trained model generation device 3, and the estimation result , the model g (Rdc4, Rdc2) included in the learned model 35 is used to calculate the resistance component Rc due to the measurement system by the two-terminal method. Then, the inspection apparatus 2 corrects the measured value Rs of the AC resistance based on the calculated resistance component Rc caused by the measurement system by the two-terminal method, and outputs the corrected value as the value of the AC resistance of the DUT.

According to this, using the resistance component Rc due to the measurement system by the two-terminal method calculated based on the highly accurate model g (Rdc4, Rdc2) generated by machine learning the past measurement results of the inductor element, Since the measured value Rs of the AC resistance is corrected, the value of the AC resistance can be obtained more accurately than the conventional method.

In addition, the inspection device 2 calculates the error |Rse-Rs| When |Rse−Rs| is smaller than the threshold value Rth, output the AC resistance measured value Rs corrected based on the resistance component Rc caused by the measurement system by the two-terminal method as the AC resistance value Rsr (=Rse−Rc). do. On the other hand, when the error |Rse−Rs| is larger than the threshold value Rth, the inspection device 2 outputs the actually measured AC resistance value Rs as the AC resistance value Rsr (=Rs).

According to this, instead of uniformly correcting the measured value of AC resistance for all the inductor elements to be inspected, the error between the measured value of AC resistance and the estimated value based on the learned model 35 is Correction of the measured values of AC resistance is performed only for small inductor elements, that is, inductor elements for which it is considered appropriate to apply the learned model 35 (model g(Rdc4, Rdc2)). As a result, overcorrection can be prevented, and a more accurate AC resistance value can be obtained.

Thus, according to the learned model generation device 3 and the inspection device 2 according to the present embodiment, it is possible to improve the reliability of inspection of electronic components.

<<Expansion of Embodiment>>
The invention made by the inventor of the present application has been specifically described above based on the embodiment, but the invention is not limited to it, and it goes without saying that various modifications can be made without departing from the gist of the invention. .

For example, the model (learned model 35) used for inspection by the inspection apparatus 2 may be a function indicating the correspondence between the measured values Rdc4 and Rdc2 of the direct current resistance and the measured value Rs of the alternating current resistance. It may be a model generated by For example, the inspection device 2 inspects the inductor element by the same method as above using models including models g (Rdc4, Rdc2) and h (Rdc4) whose coefficients are adjusted by a method other than machine learning. good too.

Further, in the above-described embodiment, the inspection apparatus 2 integrates the components such as the data processing control device 10, the first measurement unit 11, the second measurement unit 12, the operation unit 13, the output unit 14, and the transport mechanism 15. Although the case where it is the apparatus which carried out was illustrated as an example, some components which comprise the inspection apparatus 2 may be comprised separately from other components. For example, the data processing control device 10, the operation unit 13, and the output unit 14 are realized by a first device (for example, an information processing device such as a PC), and the first measurement unit 11, the second measurement unit 12, and the transport mechanism 15 may be implemented by a separate second device different from the first device. In this case, the first device and the second device may be connected via a wired or wireless network.

The above-mentioned flowchart is an example for explaining the operation, and is not limited to this. That is, the steps shown in each diagram of the flowchart are specific examples, and the flow is not limited to this flow. For example, the order of some processes may be changed, other processes may be inserted between each process, and some processes may be performed in parallel.

DESCRIPTION OF SYMBOLS 1... Inspection system, 2... Inspection apparatus, 3... Trained model generation apparatus, 10... Data processing control apparatus, 11... First measurement unit, 12... Second measurement unit, 13... Operation unit, 14... Output unit, 15 Conveyance mechanism 21 Data acquisition unit 22 Storage unit 23 Estimation unit 24 Correction unit 25 Judgment unit 31 Measurement data acquisition unit for learning 32 Trained model generation unit 33 Storage part, 34, 34_1 to 34_n ... learning measurement data, 35 ... learned model, 50 ... measurement data, Rc ... resistance component caused by the measurement system by the two-terminal method, Rdc2 ... measured value of DC resistance by the two-terminal method, Rdc4: measured value of DC resistance by 4-terminal method, Rs: measured value of AC resistance by 2-terminal method, Rse: estimated value of measured value of AC resistance by 2-terminal method, Rsr: value of AC resistance, Rth: threshold.

Claims

A first measurement value of the DC resistance of the measurement object measured by the four-terminal method, a second measurement value of the DC resistance of the measurement object measured by the two-terminal method, and the measurement object measured by the two-terminal method a data acquisition unit for acquiring a third measurement of AC resistance;
a storage unit that stores a trained model for causing a computer to calculate the third measured value based on the input first measured value and the second measured value;
an estimating unit that calculates an estimated value of the third measured value corresponding to the first measured value and the second measured value acquired by the data acquisition unit, based on the trained model stored in the storage unit; , an inspection device.
In the inspection device according to claim 1,
According to the estimated value of the third measured value, correcting the third measured value based on the first measured value and the second measured value acquired by the data acquisition unit, and obtaining the corrected third measured value as the An inspection apparatus further comprising a correction unit that performs a correction process for outputting an AC resistance value of an object to be measured.
In the inspection device according to claim 2,
The trained model includes a first model representing a resistance component caused by a measurement system using a two-terminal method, and a second model representing a resistance component caused by the object to be measured,
The correction unit calculates a resistance component caused by the measurement system by the two-terminal method according to the first model based on the first measurement value and the second measurement value acquired by the data acquisition unit, and As the correction process, the inspection apparatus corrects the third measured value based on a resistance component caused by the measurement system according to the two-terminal method.
In the inspection device according to claim 3,
The first model is a regression model in which the first measured value and the second measured value are explanatory variables and the value of the resistance component caused by the measurement system is the objective variable,
The second model is a regression model in which the first measured value is an explanatory variable and the value of the resistance component caused by the measurement object is an objective variable,
The first model and the second model include learned parameters adjusted by machine learning a pair of data pairs in which the third measurement value is associated with the first measurement value and the second measurement value. inspection equipment.
In the inspection device according to any one of claims 1 to 4,
The correction unit calculates an error between the estimated value of the third measurement value calculated by the estimation unit and the third measurement value acquired by the data acquisition unit, and if the error is smaller than a threshold, the correction Inspection equipment for processing.
A first measurement value of the DC resistance of the measurement object measured by the four-terminal method, a second measurement value of the DC resistance of the measurement object measured by the two-terminal method, and the measurement object measured by the two-terminal method a data acquisition unit for acquiring a third measurement of AC resistance;
a storage unit that stores a first model indicating a correspondence relationship between the first measured value, the second measured value, and the third measured value;
an estimating unit that calculates an estimated value of the third measured value corresponding to the first measured value and the second measured value acquired by the data acquisition unit, based on the first model stored in the storage unit; ,
Calculate an error between the estimated value of the third measured value calculated by the estimating unit and the third measured value acquired by the data acquisition unit, and if the error is smaller than a threshold, acquire by the data acquisition unit a correction unit that performs a correction process of correcting the third measured value based on the first measured value and the second measured value and outputting the corrected third measured value as the value of the AC resistance of the measurement object; an inspection device.
In the inspection device according to claim 6,
The first model includes the first measured value and the second measured value as explanatory variables, and a second model in which the value of the resistance component resulting from the measurement system according to the two-terminal method is the objective variable, and the first measurement a third model in which the value is an explanatory variable and the value of the resistance component caused by the object to be measured is an objective variable,
As the correction process, the correction unit corrects a resistance component caused by the measurement system by the two-terminal method according to the second model based on the first measurement value and the second measurement value acquired by the data acquisition unit. an inspection device that calculates and corrects the third measured value based on the resistance component caused by the measurement system according to the two-terminal method, and outputs the corrected third measured value as the value of the AC resistance of the object to be measured.
In addition to the first measured value of the DC resistance of the measurement object measured by the four-terminal method and the second measured value of the DC resistance of the measurement object measured by the two-terminal method, the a learning measurement data acquisition unit that acquires learning measurement data associated with the third measurement value of the AC resistance;
A trained model for causing a computer to calculate the third measured value based on input data including the first measured value and the second measured value by machine learning the learning measured data. A trained model generating device, comprising: a trained model generation unit for generating.
In the trained model generation device according to claim 8,
The trained model includes a first regression model having the first measured value and the second measured value as explanatory variables and a resistance component value resulting from a measurement system according to a two-terminal method as an objective variable; a second regression model in which the measured value is an explanatory variable and the value of the resistance component caused by the object to be measured is an objective variable;
A learned model generating device, wherein the learned model generation unit adjusts the learned parameters of the first regression model and the second regression model by performing machine learning on the learning measurement data.
A first measurement value of the DC resistance of the measurement object measured by the four-terminal method, a second measurement value of the DC resistance of the measurement object measured by the two-terminal method, and the measurement object measured by the two-terminal method a first step of obtaining a third measurement of AC resistance;
said first measurement obtained by said first step, based on a trained model for operating a computer to estimate said third measurement based on said input first and second measurements; a second step of calculating an estimate of the third measurement corresponding to the value and the second measurement;
According to the estimated value of the third measured value, the third measured value is corrected based on the first measured value and the second measured value obtained in the first step, and the corrected third measured value is the and a third step of outputting the AC resistance value of the object to be measured.
A first measurement value of the DC resistance of the measurement object measured by the four-terminal method, a second measurement value of the DC resistance of the measurement object measured by the two-terminal method, and the measurement object measured by the two-terminal method a first step of obtaining a third measurement of AC resistance;
Corresponding to the first measured value and the second measured value obtained in the first step based on a first model showing a correspondence relationship between the first measured value and the second measured value and the third measured value a second step of calculating an estimate of the third measurement of
a third step of calculating an error between the estimated value of the third measured value calculated in the second step and the third measured value obtained in the first step;
when the error is smaller than a threshold, correcting the third measured value based on the first measured value and the second measured value obtained in the first step; A fourth step of outputting as the value of the AC resistance of
and a fifth step of outputting the third measurement value obtained in the first step as a value of the AC resistance of the measurement object when the error is greater than the threshold value.
An inspection program that causes a computer to execute each step in the inspection method according to claim 10 or 11.
In addition to the first measured value of the DC resistance of the measurement object measured by the four-terminal method and the second measured value of the DC resistance of the measurement object measured by the two-terminal method, the A first step of acquiring learning measurement data associated with a third measurement value of AC resistance;
A computer functions to calculate the third measured value based on input data including the first measured value and the second measured value by performing machine learning on the learning measured data acquired in the first step. and a second step of generating a trained model for generating a trained model.
A trained model generation program that causes a computer to execute each step in the trained model generation method according to claim 13.