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

Abstract

To improve the reliability of electronic component testing. [Solution] The inspection device uses a first measured value Rdc4 of DC resistance measured by the four-terminal method and a second measured value Rdc2 of DC resistance measured by the two-terminal method as explanatory variables, and a first resistance value Rc, which is the value of the resistance component due to the measurement system by the two-terminal method, as the objective variable. The first model f(Rdc4,Rdc2) uses a fourth measured value Ls of inductance as an explanatory variable, and a second model g(Ls), which is the value of the second resistance value Rs0, which is the value of the resistance component due to the object being measured, as the objective variable. The measurement data 50 is input to a learned model 35, and the obtained value is adjusted based on an adjustment value Rtng2 to calculate an estimated value of AC resistance Rse. The adjustment value Rtng2 is calculated according to the difference between the reference AC resistance value and the estimated value g(Ls_av2) of the second resistance value calculated by inputting at least one measurement data 50 into the second model g(Ls).

Description

Technical Field

[0001] The present invention relates to an inspection device, a learned model generation device, an inspection method, a learned model generation method, an inspection program, and a learned model program, and for example, relates to an inspection device for inspecting an inductor element.

Background Art

[0002] Conventionally, an inspection device that measures the electrical characteristics of electronic components such as chip inductors and determines the quality of the electronic components based on the measurement results is known. For example, Patent Document 1 discloses an inspection device that measures the AC resistance and inductance of an inductor element to be inspected, calculates a Q value using the measured values, and determines the quality of the inductor element based on the calculated Q value.

[0003] Patent Document 1 describes that as a method for measuring the AC resistance of an inductor element, the AC resistance is calculated by subtracting an estimated value of the contact resistance of the measurement probe used when measuring by the two-terminal method from the measured value of the AC resistance measured by the two-terminal method. Further, Patent Document 1 describes that an estimated value of the contact resistance is calculated by subtracting the measured value of the DC resistance measured by the four-terminal method from the measured value of the DC resistance measured by the two-terminal method, and the measured value of the AC resistance is corrected using a value obtained by multiplying the estimated value of the contact resistance by a coefficient of 0 or more and 1 or less.

[0004] 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 actual series resistance and the AC resistance of the inductor element is unknown. For example, if the relationship is non-linear, the correction of the measured value of the AC resistance may not be appropriately performed. Further, the inspection device disclosed in Patent Document 1 corrects the measured value of the AC resistance using a value obtained by multiplying the estimated value of the contact resistance by a coefficient in order to avoid overcorrection of the AC resistance. However, if the coefficient is not appropriate, the correction of the AC resistance may not be appropriately performed. [Prior art documents] [Patent Documents]

[0005] [Patent Document 1] Patent No. 6949675 [Overview of the project] [Problems that the invention aims to solve]

[0006] Prior to this application, the inventors of the present invention investigated how to generate a model that calculates an estimated AC resistance from the DC resistance measurement of an inductance element under inspection by machine learning using already measured series resistance and AC resistance values ​​of inductance elements, and how to use this model to determine whether an inductance element is good or bad. As a result of the investigation, the following problems became apparent.

[0007] Generally, when inspecting a DUT (Device Under Test) using an inspection device (automatic measuring instrument), a calibration process is performed before the inspection begins. In the calibration process, first, one reference device (golden device) is selected from among several DUTs. Next, in a low-noise environment, the electrical characteristics of three reference devices (e.g., open reference, short reference, and load reference) are measured manually using a fixture, and these measured values ​​are used as the defined values ​​for the reference devices. If an open reference or short reference with predetermined defined values ​​is used, only the DUT's reference device is measured. Next, the three reference devices with defined values ​​are measured by the inspection device to obtain the measured values ​​for each reference. Then, a correction formula is generated to correct the measured values ​​so that the error between the measured value for each reference and the predetermined defined value is minimized, and this formula is stored in the inspection device. This completes the series of calibration processes. After the inspection device is put into operation, the inspection device corrects the values ​​obtained from measuring the DUT based on the above correction formula and outputs the corrected value as the measured value of the DUT. This makes it possible to reduce measurement errors during inspection.

[0008] As mentioned above, the defined value of the reference device is measured in a low-noise environment, so the measurement error included in the defined value of the reference device is small. However, if the defined value of the reference device contains an error, it will affect all subsequent measurement results of the DUT by the inspection equipment.

[0009] For example, if there is a difference between the training measurement data used in machine learning to generate the above-mentioned trained model and the defined value of the reference device used in the calibration process, the accuracy of the inspection using the trained model may decrease.

[0010] This invention has been made in view of the above-mentioned problems, and aims to improve the reliability of inspection of electronic components. [Means for solving the problem]

[0011] A typical embodiment of the present invention includes a data acquisition unit that acquires measurement data including a first measurement of the DC resistance of an object to be measured measured by a four-terminal method, a second measurement of the DC resistance of the object to be measured measured by a two-terminal method, a third measurement of the AC resistance of the object to be measured measured by a two-terminal method, and a fourth measurement of the inductance of the object to be measured measured by a two-terminal method; a storage unit that stores a learned model for causing a computer to function to calculate the third measurement based on the input first measurement, second measurement, and fourth measurement; and an estimation unit that calculates an estimated value of the third measurement corresponding to the first measurement, second measurement, and fourth measurement acquired by the data acquisition unit, based on the learned model stored in the storage unit. The system includes an adjustment value calculation unit that calculates an adjustment value to adjust the value calculated by the model, wherein the learned model is represented by the sum of a first model, which uses the first measured value and the second measured value as explanatory variables and a first resistance value, which is the value of the resistance component caused by the two-terminal measurement system, as the objective variable, and a second model, which uses the fourth measured value as an explanatory variable and a second resistance value, which is the value of the resistance component caused by the object being measured, as the objective variable, and the adjustment value calculation unit calculates the adjustment value according to the difference between the reference AC resistance value and an estimated value of the second resistance value calculated by inputting the fourth measured value included in at least one of the measurement data into the second model, and the estimation unit adjusts the value calculated by the learned model based on the adjustment value and calculates it as an estimated value of the third measured value. [Effects of the Invention]

[0012] According to the inspection apparatus of the present invention, it is possible to improve the reliability of inspection of electronic components. [Brief explanation of the drawing]

[0013] [Figure 1] This diagram shows the configuration of the inspection system according to Embodiment 1. [Figure 2]It is a diagram showing an example of the configuration of a learned model generation device in the inspection system according to Embodiment 1. [Figure 3] It is a diagram for explaining the deviation between the estimated value of the second resistance value Rs0 of the AC resistance calculated by the model g(Ls) and the defined value (AC resistance Rs) of the DUT in the calibration process. [Figure 4] It is a flowchart showing the flow of generation of a learned model by the learned model generation device according to Embodiment 1. [Figure 5] It is a flowchart showing the flow of machine learning (step S3) according to Embodiment 1. [Figure 6] It is a diagram showing an example of the configuration of a data processing control device in the inspection device according to Embodiment 1. [Figure 7] It is a flowchart showing the flow of inspection by the inspection device according to Embodiment 1. [Figure 8] It is a flowchart showing the flow of adjustment value calculation processing (step S12) according to Embodiment 1. [Figure 9] It is a flowchart showing the flow of inspection (step S13) by the inspection device according to Embodiment 1. [Figure 10] It is a diagram showing an example of the configuration of a learned model generation device in the inspection system according to Embodiment 2. [Figure 11] It is a flowchart showing the flow of machine learning (step S3) according to Embodiment 2. [Figure 12] It is a diagram showing an example of the configuration of a data processing control device in the inspection device according to Embodiment 2. [Figure 13] It is a flowchart showing the flow of adjustment value calculation processing (step S12) in the inspection process according to Embodiment 2.

MODE FOR CARRYING OUT THE INVENTION

[0014] 1. Outline of the Embodiment First, a general overview of a typical embodiment of the invention disclosed in this application will be provided. In the following description, as an example, reference numerals on the drawings corresponding to the components of the invention are indicated in parentheses.

[0015] [1] An inspection apparatus (2) according to a typical embodiment of the present invention includes: a data acquisition unit (21) that acquires measurement data (50) including a first measured value (Rdc4) of the DC resistance of an object to be measured measured by a four-terminal method, a second measured value (Rdc2) of the DC resistance of the object to be measured measured by a two-terminal method, a third measured value (Rs) of the AC resistance of the object to be measured measured by a two-terminal method, and a fourth measured value (Ls) of the inductance of the object to be measured measured by a two-terminal method; a storage unit (22) that stores a learned model (35) for causing a computer to function to calculate the third measured value based on the input first measured value, second measured value, and fourth measured value; an estimation unit (23) that calculates an estimated value of the third measured value corresponding to the first measured value, second measured value, and fourth measured value acquired by the data acquisition unit based on the learned model stored in the storage unit; and the computer The system includes an adjustment value calculation unit (26) that calculates an adjustment value (Rtng2) for adjusting the value calculated by a trained model, wherein the trained model is represented by the sum of a first model (f(Rdc4,Rdc2)) which uses the first measurement value and the second measurement value as explanatory variables and a first resistance value (Rc), which is the value of the resistance component caused by the two-terminal measurement system, as the objective variable, and a second model (g(Ls)) which uses the fourth measurement value as an explanatory variable and a second resistance value (Rs0), which is the value of the resistance component caused by the object being measured, as the objective variable, and the adjustment value calculation unit calculates the adjustment value according to the difference between the value of the AC resistance to be used as a reference and the estimated value of the second resistance value calculated by inputting the fourth measurement value included in at least one of the measurement data into the second model, and the estimation unit adjusts the value calculated by the trained model based on the adjustment value and calculates it as the estimated value of the third measurement value.

[0016] 〔2〕In the inspection apparatus according to 〔1〕 above, the value of the reference AC resistance is the average value (Rs_av2) of the third measurement value included in the plurality of measurement data of the measurement object where the difference between the first measurement value and the second measurement value is smaller than the threshold (|Rdc2−Rdc4|<Rdth). The adjustment value calculation unit may calculate the estimated value (g(Ls_av2)) of the second resistance value by inputting the average value (Ls_av2) of the fourth measurement value included in the plurality of measurement data of the measurement object where the difference between the first measurement value and the second measurement value is smaller than the threshold into the second model, and calculate the difference between the average value of the third measurement value and the estimated value of the second resistance value as the adjustment value (Rtng2 = Rs_av2−g(Ls_av2)).

[0017] 〔3〕In the inspection apparatus according to 〔1〕 above, the value of the reference AC resistance is the defined value (Rs_d2) of the AC resistance of the reference device in the calibration process. The adjustment value calculation unit may calculate the estimated value (g(Ls_av2)) of the second resistance value by inputting the average value (Ls_av2) of the fourth measurement value included in the measurement data of the plurality of measurement objects where the difference between the first measurement value and the second measurement value is smaller than the threshold (|Rdc2−Rdc4|<Rdth) into the second model, and calculate the difference between the defined value and the estimated value of the second resistance value as the adjustment value (Rtng2 = Rs_d2−g(Ls_av2)).

[0018] [4] A typical inspection apparatus (2A) according to another embodiment of the present invention includes: a data acquisition unit (21) that acquires measurement data (50) including a first measured value (Rdc4) of the DC resistance of an object to be measured measured by a four-terminal method, a second measured value (Rdc2) of the DC resistance of the object to be measured measured by a two-terminal method, and a third measured value (Rs) of the AC resistance of the object to be measured measured by a two-terminal method; a storage unit (22) that stores a learned model (35A) for causing a computer to function to calculate the third measured value based on the input first and second measured values; an estimation unit (23A) that calculates an estimated value of the third measured value corresponding to the first and second measured values ​​acquired by the data acquisition unit based on the learned model stored in the storage unit; and a unit for adjusting the value calculated by the learned model. The system includes an adjustment value calculation unit (26A) for calculating an adjustment value, wherein the learned model is represented by the sum of a first model (f(Rdc4,Rdc2)) which uses the first measured value and the second measured value as explanatory variables and a first resistance value which is the value of the resistance component caused by the two-terminal measurement system as the objective variable, and a second model (g(Rdc4)) which uses the first measured value as an explanatory variable and a second resistance value (Rs0) which is the value of the resistance component caused by the object being measured as the objective variable, and the adjustment value calculation unit calculates the adjustment value according to the difference between the value of the AC resistance to be used as a reference and an estimated value of the second resistance value calculated by inputting the first measured value included in at least one of the measurement data into the second model, and the estimation unit adjusts the value calculated by the learned model based on the adjustment value and calculates it as an estimated value of the third measured value.

[0019] 〔5〕In the inspection apparatus according to 〔4〕 above, the value of the reference AC resistance is the average value (Rs_av2) of the third measurement values included in the plurality of measurement data of the measurement object where the difference between the first measurement value and the second measurement value is smaller than the threshold value (|Rdc2−Rdc4|<Rdth). The adjustment value calculation unit may calculate an estimated value (g(Rdc4_av2)) of the second resistance value by inputting the average value (Rdc4_av2) of the first measurement values included in the plurality of measurement data of the measurement object where the difference between the first measurement value and the second measurement value is smaller than the threshold value into the second model, and calculate the difference (Rs_av2−g(Rdc4_av2)) between the average value of the third measurement values and the estimated value of the second resistance value as the adjustment value.

[0020] 〔6〕In the inspection apparatus according to 〔4〕 above, the value of the reference AC resistance is the defined value (Rs_d2) of the AC resistance of the reference device in the calibration process. The adjustment value calculation unit may calculate an estimated value (g(Rdc4_av2)) of the second resistance value by inputting the average value (Rdc4_av2) of the first measurement values included in the measurement data of the plurality of measurement objects where the difference between the first measurement value and the second measurement value is smaller than the threshold value (|Rdc2−Rdc4|<Rdth) into the second model, and calculate the difference (Rs_d2−g(Rdc4_av2)) between the defined value and the estimated value of the second resistance value as the adjustment value. 〔7〕In the inspection apparatus according to any one of 〔1〕 to 〔6〕 above, a correction unit (24) may be further provided that performs a correction process of correcting the third measurement value 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, and outputs the corrected third measurement value as the value of the AC resistance of the measurement object.

[0021] [8] In the inspection apparatus described in [7] above, the correction unit may calculate the resistance component (Rc) caused by the two-terminal measurement system according to the first model based on the first and second measurement values ​​acquired by the data acquisition unit, and as the correction process, the third measurement value may be corrected based on the resistance component caused by the two-terminal measurement system.

[0022] [9] In the inspection apparatus described in [8] above, the correction unit may calculate the error between the estimated value of the third measurement calculated by the estimation unit and the third measurement acquired by the data acquisition unit, and perform the correction process if the error is smaller than a predetermined threshold.

[0023]

[10] A typical embodiment of the present invention provides an inspection method which includes: a first step (S121, S131~S133) of acquiring measurement data (50) including a first measured value (Rdc4) of the DC resistance of the object to be measured measured by a four-terminal method, a second measured value (Rdc2) of the DC resistance of the object to be measured measured by a two-terminal method, a third measured value (Rs) of the AC resistance of the object to be measured measured by a two-terminal method, and a fourth measured value (Ls) of the inductance of the object to be measured measured by a two-terminal method; a second step (S134) of calculating an estimated value of the third measured value corresponding to the first measured value, the second measured value, and the fourth measured value acquired in the first step, based on a trained model (35) for causing a computer to function to estimate the third measured value based on the input first measured value, the second measured value, and the fourth measured value; a third step (S122~S126) of calculating an adjustment value to adjust the value calculated by the trained model; and the calculation of the third measured value corresponding to the first measured value, the second measured value, and the fourth measured value acquired in the second step The method includes a fourth step (S134) of adjusting the estimated value of the 3rd measurement based on the adjustment value, and a fifth step (S135-S138) of correcting the 3rd measurement based on the 1st and 2nd measurement obtained in the 1st step according to the estimated value of the 3rd measurement adjusted in the fourth step, and outputting the corrected 3rd measurement as the value of the AC resistance of the object to be measured. The learned model is represented by the sum of a first model in which the 1st and 2nd measurement values ​​are explanatory variables and the 1st resistance value, which is the value of the resistance component caused by the two-terminal measurement system, is the objective variable, and a second model in which the 4th measurement value is an explanatory variable and the 2nd resistance value, which is the value of the resistance component caused by the object to be measured, is the objective variable. The third step includes a step (S122-S126) of calculating the adjustment value according to the difference between the reference AC resistance value and the estimated value of the 2nd resistance calculated by inputting the 4th measurement included in at least one of the measurement data into the second model.

[0024]

[11] In the inspection method described in

[10] above, the value of the AC resistance used as the reference is the average value (Rs_av2) of the third measurement value included in a plurality of measurement data of the object to be measured in which the difference between the first measurement value and the second measurement value is less than a threshold, and the third step may include the step (S125) of calculating an estimated value of the second resistance value (g(Ls_av2)) by inputting the average value (Ls_av2) of the fourth measurement value included in a plurality of measurement data of the object to be measured in which the difference between the first measurement value and the second measurement value is less than a threshold into the second model, and the step (S126) of calculating the difference between the average value of the third measurement value and the estimated value of the second resistance value as the adjustment value.

[0025]

[12] In the inspection method described in

[10] above, the value of the AC resistance used as a reference is the defined value of the AC resistance of the reference device in the calibration process (Rs_d2), and the third step may include the step (S125) of calculating an estimated value of the second resistance (g(Ls_av2)) by inputting the average value (Ls_av2) of the fourth measurement values ​​included in the measurement data of a plurality of objects to be measured in which the difference between the first measurement value and the second measurement value is less than a threshold into the second model, and the step (S126) of calculating the difference between the defined value and the estimated value of the second resistance (Rs_d2-g(Ls_av2)) as the adjustment value.

[0026]

[13] A typical embodiment of the present invention provides an inspection method comprising: a first step (S121, S131-S133) of acquiring measurement data (50) including a first measured value (Rdc4) of the DC resistance of the object to be measured measured by a four-terminal method, a second measured value (Rdc2) of the DC resistance of the object to be measured measured by a two-terminal method, and a third measured value (Rs) of the AC resistance of the object to be measured measured by a two-terminal method; a second step (S134) of calculating an estimated value of the third measured value corresponding to the first and second measured values ​​acquired in the first step, based on a trained model (35A) for causing a computer to function to estimate the third measured value based on the input first and second measured values; a third step (S122, S123, S124A-S126A) of calculating an adjustment value (Rtng2) for adjusting the value calculated by the trained model; and a fourth step (S134) of adjusting the estimated value of the third measured value calculated in the second step based on the adjustment value. The third step includes a fifth step (S135~S138) in which the third measurement value is corrected based on the first measurement value and the second measurement value obtained in the first step, according to the estimated value of the third measurement value adjusted in the fourth step, and the corrected third measurement value is output as the value of the AC resistance of the object to be measured. The learned model is represented by the sum of a first model (f(Rdc4,Rdc2)) in which the first measurement value and the second measurement value are explanatory variables and the first resistance value, which is the value of the resistance component due to the two-terminal measurement system, is the objective variable, and a second model (g(Rdc4)) in which the first measurement value is an explanatory variable and the second resistance value, which is the value of the resistance component due to the object to be measured, is the objective variable. The third step is characterized by including a step (S123,S124A~S126A) in which the adjustment value is calculated according to the difference between the reference AC resistance value and the estimated value of the second resistance value calculated by inputting the first measurement value included in at least one of the measurement data into the second model.

[0027]

[14] In the inspection method described in

[13] above, the value of the AC resistance used as the reference is the average value (Rs_av2) of the third measurement value included in a plurality of measurement data of the object to be measured in which the difference between the first measurement value and the second measurement value is less than a threshold, and the third step may include the step (S124A) of calculating an estimated value of the second resistance value (g(Rdc4_av2)) by inputting the average value (Rdc4_av2) of the first measurement value included in a plurality of measurement data of the object to be measured in which the difference between the first measurement value and the second measurement value is less than a threshold into the second model, and the step (S126A) of calculating the difference between the average value of the third measurement value and the estimated value of the second resistance value (Rs_av2-g(Rdc4_av2)) as the adjustment value.

[0028]

[15] In the inspection method described in

[13] above, the value of the AC resistance used as a reference is the defined value of the AC resistance of the reference device in the calibration process (Rs_d2), and the third step may include the step (S124A) of calculating an estimated value of the second resistance (g(Rdc4_av2)) by inputting the average value of the first measurement (Rdc4_av2) included in the measurement data of a plurality of objects to be measured in which the difference between the first measurement and the second measurement is less than a threshold into the second model, and the step (S126) of calculating the difference between the defined value and the estimated value of the second resistance (Rs_d2-g(Rdc4_av2)) as the adjustment value.

[0029]

[16] An inspection program according to a typical embodiment of the present invention causes a computer to execute each step of the inspection method described in any one of

[10] to 15 above. It is characterized by the following.

[0030] 〔17〕The trained model generation device (3) according to a representative embodiment of the present invention acquires learning measurement data (34, 34_1 to 34_n) in which a third measurement value of the AC resistance of the measurement object measured by the two-terminal method is associated with a first measurement value (Rdc4) of the DC resistance of the measurement object measured by the two-terminal method, a second measurement value (Rdc2) of the DC resistance of the measurement object measured by the two-terminal method, and a fourth measurement value (Ls) of the inductance of the measurement object measured by the two-terminal method. A learning measurement data acquisition unit (31), and by machine learning the learning measurement data, a trained model (35) for causing a computer to function so as to calculate the third measurement value based on input data including the first measurement value, the second measurement value, and the fourth measurement value. A trained model generation unit (32) that generates, the trained model is represented by the sum of a first model (f(Rdc4, Rdc2)) having the first measurement value and the second measurement value as explanatory variables and a first resistance (Rc) that is a value of a resistance component caused by a two-terminal measurement system as an objective variable, and a second model (g(Ls)) having the fourth measurement value as an explanatory variable and a second resistance value (Rs0) that is a value of a resistance component caused by the measurement object as an objective variable. The trained model generation unit includes a first model generation unit (321) that generates the first model by machine learning the first measurement value and the second measurement value included in the learning measurement data of the measurement object, and a difference between the first measurement value and the second measurement value is smaller than the threshold (|Rdc2 - Rdc4| < Rdth). A second model generation unit (322) that generates the second model by machine learning the fourth measurement value included in the learning measurement data of the measurement object, and an adjustment value calculation unit (323) that calculates an adjustment value (Rtng1) according to a difference between a reference AC resistance value and an estimated value of the second resistance value calculated by inputting the fourth measurement value included in at least one of the learning measurement data into the second model. The first model generation unit adjusts the parameters of the first model in a state where the parameters of the second model are fixed in a model represented by the sum of the first model, the second model, and the adjustment value.

[0031]

[18] In the trained model generation device described in

[17] above, the reference value of AC resistance is the average value (Rs_av1) of the third measurement values ​​included in the training measurement data of a plurality of the objects to be measured in which the difference between the first measurement value and the second measurement value is less than the threshold, and the adjustment value calculation unit may calculate an estimated value of the second resistance value (g(Ls_av1)) by inputting the average value (Ls_av1) of the fourth measurement values ​​included in the training measurement data of a plurality of the objects to be measured in which the difference between the first measurement value and the second measurement value is less than the threshold into the second model, and calculate the difference between the average value of the third measurement value and the estimated value of the second resistance value (Rs_av1-g(Ls_av1)) as the adjustment value.

[0032]

[19] In the trained model generation device described in

[17] above, the value of the AC resistance used as a reference is the defined value (Rs_d1) of the AC resistance of the reference device in the calibration process, and the adjustment value calculation unit calculates an estimated value (g(Ls_av1)) of the second resistance by inputting the average value (Ls_av1) of the fourth measurement value included in the training measurement data of a plurality of the objects to be measured, the difference between the first measurement value and the second measurement value being less than the threshold, into the second model, and the difference (Rs_d1-g(Ls_av1)) between the defined value and the estimated value of the second resistance is calculated as the adjustment value.

[0033] 〔20〕The trained model generation device (3A) according to a representative embodiment of the present invention acquires learning measurement data (34A, 34A_1 to 34A_n) in which a third measurement value (Rs) of the AC resistance of the measurement object measured by the two-terminal method is associated with a first measurement value (Rdc4) of the DC resistance of the measurement object measured by the four-terminal method and a second measurement value (Rd2) of the DC resistance of the measurement object measured by the two-terminal method. A learning measurement data acquisition unit (31), and a trained model generation unit (32A) that generates a trained model (35A) for causing a computer to function so as to calculate the third measurement value based on input data including the first measurement value and the second measurement value by machine learning the learning measurement data. The trained model is represented by the sum of a first model (f(Rdc4, Rdc2)) having the first measurement value and the second measurement value as explanatory variables and a first resistance (Rc), which is a value of a resistance component caused by a measurement system using the two-terminal method, as an objective variable, and a second model (g(Rdc4)) having the first measurement value as an explanatory variable and a second resistance value (Rs0), which is a value of a resistance component caused by the measurement object, as an objective variable. The trained model generation unit includes a first model generation unit (321) that generates the first model by machine learning the first measurement value and the second measurement value included in the learning measurement data of the measurement object, and a second model generation unit (322A) that generates the second model by machine learning the first measurement value included in the learning measurement data of the measurement object for which the difference between the first measurement value and the second measurement value is smaller than the threshold (|Rdc2 - Rdc4| < Rdth). An adjustment value calculation unit (323A) that calculates an adjustment value (Rtng1) according to the difference between the value of the reference AC resistance and the estimated value of the second resistance value calculated by inputting the first measurement value included in at least one of the learning measurement data into the second model. The first model generation unit adjusts the parameters of the first model in a state where the parameters of the second model are fixed in a model (f(Rdc4, Rdc2)+g(Rdc4)+Rtng1) represented by the sum of the first model, the second model, and the adjustment value.

[0034]

[21] In the trained model generation device described in

[20] above, the reference value of AC resistance is the average value (Rs_av1) of the third measurement values ​​included in the training measurement data of a plurality of the objects to be measured in which the difference between the first measurement value and the second measurement value is less than the threshold, and the adjustment value calculation unit may calculate an estimated value of the second resistance value (g(Rdc4_av1)) by inputting the average value (Rdc4_av1) of the first measurement values ​​included in the training measurement data of a plurality of the objects to be measured in which the difference between the first measurement value and the second measurement value is less than the threshold into the second model, and calculate the difference between the average value of the third measurement value and the estimated value of the second resistance value (Rs_av1-g(Rdc4_av1)) as the adjustment value.

[0035]

[22] In the trained model generation device described in

[20] above, the value of the AC resistance used as a reference is the defined value of the AC resistance of the reference device in the calibration process (Rs_d1), and the adjustment value calculation unit calculates an estimated value of the second resistance (g(Rdc4_av1)) by inputting the average value of the first measurement (Rdc4_av1) included in the training measurement data of a plurality of the objects to be measured, where the difference between the first measurement and the second measurement is less than the threshold, into the second model, and the difference between the defined value and the estimated value of the second resistance (Rs_d1-g(Rdc4_av1)) may be calculated as the adjustment value.

[0036]

[23] A method for generating a trained model according to a typical embodiment of the present invention includes: a first step (S1) of acquiring training measurement data (34, 34_1~34_n) by associating a third measurement of the AC resistance of an object measured by a two-terminal method with a first measurement value (Rdc4) of the DC resistance of an object measured by a four-terminal method, a second measurement value (Rdc2) of the DC resistance of the object measured by a two-terminal method, and a fourth measurement value (Ls) of the inductance of the object measured by a two-terminal method; and a second step (S3, S4) of generating a trained model (35) by machine learning the training measurement data acquired in the first step, thereby causing the computer to function to calculate the third measurement value based on input data including the first measurement value, the second measurement value, and the fourth measurement value, wherein the trained model is a first model (f(Rdc4, Rdc2)) with the first measurement value and the second measurement value as explanatory variables and a first resistance value (Rc), which is the value of the resistance component due to the two-terminal measurement system, as the objective variable. The second step is represented by the sum of the first model and the second model and the adjustment value, and the second step is represented by the sum of the first model and the second model and the adjustment value, and the second step is represented by the sum of the first model and the second model and the adjustment value, and the second step is represented by the sum of the first model and the second model and the adjustment value, and the second step is represented by the sum of the first model and the second model and the adjustment value, and the second step is represented by the sum of the first model and the second model and the adjustment value, and the second step is represented by the sum of the first model and the second model and the adjustment value, and the second step is represented by the sum of the first model and the second model and the adjustment value, and the second step is represented by the sum of the first model and the second model and the adjustment value, and the second step is represented by the sum of the first model and the second model and the adjustment value, and the second step is represented by the sum of the first model and the second model and the adjustment value, and the second step is represented by the sum of the first model and the second model and the adjustment valueThe method is characterized by including a step (S38) of adjusting the parameters of the first model while keeping the parameters of the second model fixed in Rdc2)+g(Ls)+Rtng1).

[0037]

[24] In the method for generating a trained model described in

[23] above, the value of the AC resistance used as a reference is the average value (Rs_av1) of the third measurement values ​​included in the training measurement data of a plurality of the objects to be measured in which the difference between the first measurement value and the second measurement value is less than the threshold, and the fourth step may include the step (S35) of calculating an estimated value of the second resistance value (g(Ls_av1)) by inputting the average value (Ls_av1) of the fourth measurement values ​​included in the training measurement data of a plurality of the objects to be measured in which the difference between the first measurement value and the second measurement value is less than the threshold into the second model, and the step (S35) of calculating the difference between the average value of the third measurement values ​​and the estimated value of the second resistance value as the adjustment value (Rs_av1-g(Ls_av1)).

[0038]

[25] In the method for generating a trained model described in

[23] above, the value of the AC resistance used as a reference is the defined value of the AC resistance of the reference device in the calibration process (Rs_d1), and the fourth step may include the step (S35) of calculating an estimated value of the second resistance (g(Ls_av1)) by inputting the average value (Ls_av1) of the fourth measurement values ​​included in the training measurement data of a plurality of the objects to be measured, the difference between the first measurement value and the second measurement value being less than the threshold, into the second model, and the step (S35) of calculating the difference between the defined value and the estimated value of the second resistance (Rs_d1-g(Ls_av1)) as the adjustment value.

[0039]

[26] A method for generating a trained model according to another representative embodiment of the present invention is a first step (S1) of acquiring training measurement data (34A,34A_1~34A_n) by associating a first measured value (Rdc4) of the DC resistance of the object to be measured measured by the four-terminal method, a second measured value (Rdc2) of the DC resistance of the object to be measured measured by the two-terminal method, with a third measured value (Rs) of the AC resistance of the object to be measured measured by the two-terminal method; and by performing machine learning on the training measurement data acquired in the first step, the first measured value and The method includes a second step (S3, S4) of generating a trained model for causing a computer to function to calculate the third measurement based on input data including the second measurement, wherein the trained model comprises a first model (f(Rdc4, Rdc2)) with the first measurement and the second measurement as explanatory variables and a first resistance value (Rc), which is the value of the resistance component due to the two-terminal measurement system, as the objective variable, and a second model (g(Rdc4, Rdc2)) with the first measurement as an explanatory variable and a second resistance value (Rs0), which is the value of the resistance component due to the object being measured, as the objective variable. 4)) is expressed as a sum of the above, and the second step includes a fourth step (S32A) of generating the second model by machine learning the first measurement value included in the learning measurement data of the object to be measured where the difference between the first measurement value and the second measurement value is less than a threshold, a fifth step (S33, S34A, S35A) of calculating an adjustment value according to the difference between a reference AC resistance value and an estimated value of the second resistance value calculated by inputting the first measurement value included in at least one of the learning measurement data into the second model, and a sixth step (S37A, S38A) of generating the first model by machine learning the first measurement value and the second measurement value included in the learning measurement data of the object to be measured where the difference between the first measurement value and the second measurement value is greater than a threshold, wherein the sixth step includes a step (S38A) of adjusting the parameters of the first model while fixing the parameters of the second model in a model (f(Rdc4, Rdc2) + g(Rdc4) + Rtng1) expressed as a sum of the first model, the second model and the adjustment value.

[0040]

[27] In the method for generating a trained model described in

[26] above, the value of the AC resistance used as a reference is the average value (Rs_av1) of the third measurement values ​​included in the training measurement data of a plurality of the objects to be measured in which the difference between the first measurement value and the second measurement value is less than the threshold, and the fifth step may include the step (S35A) of calculating an estimated value of the second resistance value (g(Rdc4_av1)) by inputting the average value (Rdc4_av1) of the first measurement values ​​included in the training measurement data of a plurality of the objects to be measured in which the difference between the first measurement value and the second measurement value is less than the threshold into the second model, and the step (S35A) of calculating the difference between the average value of the third measurement values ​​and the estimated value of the second resistance value (Rs_av1-g(Rdc4_av1)) as the adjustment value.

[0041]

[28] In the method for generating a trained model described in

[26] above, the value of the AC resistance used as a reference is the defined value of the AC resistance of the reference device in the calibration process (Rs_d1), and the fifth step may include the steps of: (S35A) calculating an estimated value of the second resistance (g(Rc4_av1)) by inputting the average value of the first measurement (Rdc4_av1) included in the training measurement data of a plurality of objects to be measured, the difference between the first measurement and the second measurement being less than the threshold, into the second model; and (S35A) calculating the difference between the defined value and the estimated value of the second resistance (Rs_d1-g(Rdc4_av1)) as the adjustment value.

[0042]

[29] A program for generating a trained model according to a typical embodiment of the present invention is characterized in that it causes a computer to execute each step of the trained model generation method described in any one of

[23] to

[28] above.

[0043] 2. Specific Examples of Embodiments Hereinafter, specific examples of embodiments of the present invention will be described with reference to the figures. In the following description, common components in each embodiment will be denoted by the same reference numerals, and repeated descriptions will be omitted.

[0044] <Embodiment 1> Figure 1 is a diagram showing the configuration of the inspection system 1 according to Embodiment 1.

[0045] The inspection system 1 shown in Figure 1 is a system for inspecting the quality of an object to be inspected (hereinafter also referred to as "DUT"). As shown in Figure 1, the inspection system 1 comprises a trained model generation device 3 that generates a trained model by learning training measurement data based on multiple measurement results of the DUT using machine learning, and an inspection device 2 that performs inspection of the DUT using the generated trained model.

[0046] Inspection device 2 is a device that measures the electrical characteristics of the DUT and inspects whether the DUT is good or bad based on the measurement results. For example, 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 that are determined to be good into a state ready for shipment.

[0047] In the following explanation, we will assume that the DUT is an inductor element (for example, a chip inductor element) as an example, but it is not limited to this.

[0048] The inspection device 2 measures the electrical characteristics of the inductor element as a DUT using a trained model, which will be described later. Specifically, as shown in Figure 1, the inspection device 2 includes a data processing control device 10, a first measurement unit 11, a second measurement unit 12, an operation unit 13, an output unit 14, and a transport mechanism 15.

[0049] The first measurement unit 11 is a device that measures the electrical characteristics of the inductor element as a DUT using the four-terminal method. Examples of the first measurement unit 11 include impedance measuring instruments such as resistance meters and LCR meters that are capable of measuring impedance using the four-terminal method.

[0050] The second measurement unit 12 is a device that measures the electrical characteristics of the inductor element as a DUT using the two-terminal method. An example of the second measurement unit 12 is an impedance measuring instrument such as an LCR meter capable of measuring impedance using the two-terminal method.

[0051] The first measurement unit 11 and the second measurement unit 12 can be any device capable of measuring the electrical characteristics of the DUT, such as impedance, and are not limited to the examples described above.

[0052] The first measurement unit 11 measures the DC resistance of the inductor element acting as a DUT using the four-terminal method in response to instructions from the data processing control device 10. For example, the first measurement unit 11 includes a moving mechanism (not shown) for moving probes 61a to 61d, a current output unit and a voltage detection unit (not shown), and a measurement value calculation unit (not shown) for calculating the measurement value based on the detection result.

[0053] For example, when the first measurement unit 11 receives a measurement execution instruction from the data processing control device 10, the first measurement unit 11's moving mechanism brings probes 61a and 61c into contact with one terminal of the inductor element that has been transported to a predetermined measurement position, and probes 61b and 61d into contact with the other terminal of the inductor element. Next, the current output unit of the first measurement unit 11 supplies a DC current to the inductor element via probes 61a and 61b. The voltage detection unit of the first measurement unit 11 detects the voltage value between the inductor terminals when the DC current is supplied to the inductor element via probes 61c and 61d. The measurement value calculation unit of the first measurement unit 11 calculates the 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.

[0054] The second measurement unit 12 measures the DC resistance, AC resistance, and inductance of the inductor element acting as a DUT using the two-terminal method in response to instructions from the data processing control device 10. For example, the second measurement unit 12 includes a moving mechanism (not shown) for moving probes 62a and 62b, a current output unit and a voltage detection unit (not shown), and a measurement value calculation unit (not shown) for calculating the measurement value based on the detection result.

[0055] For example, when the second measurement unit 12 receives a measurement execution instruction from the data processing control device 10, the second measurement unit 12's moving mechanism brings probe 62a into contact with one terminal of the inductor element that has been transported to a predetermined measurement position, and probe 62b into contact with the other terminal of the inductor element. Next, the current output unit of the second measurement unit 12 supplies a DC current to the inductor element via probes 62a and 62b, and the voltage detection unit of the second measurement unit 12 detects the voltage value between both terminals of the inductor element via probes 62a and 62b. The measurement value calculation unit of the second measurement unit 12 calculates the 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.

[0056] Furthermore, with the probes 62a and 62b in contact with both terminals of the inductor element by the moving mechanism of the second measuring unit 12, the current output unit of the second measuring unit 12 supplies AC current to the inductor element via the probes, and the voltage detection unit of the second measuring unit 12 detects the AC voltage value between both terminals of the inductor element via the probes 62a and 62b. The measurement value calculation unit of the second measuring unit 12 calculates the measured value Rs of the AC resistance and the measured value Ls of the inductor element based on the detected AC voltage value (effective voltage), the AC current value (effective current) of the AC current supplied to the inductor element, and the phase difference between the AC voltage and the AC current.

[0057] Furthermore, some functions of the first measurement unit 11 and the second measurement unit 12 may be implemented by the data processing control device 10. For example, the calculations performed by the measurement value calculation units of the first measurement unit 11 and the second measurement unit 12 may be performed by the data processing control device 10.

[0058] The operation unit 13 is an input interface for the user to operate the inspection device 2. Examples of the operation unit 13 include various buttons and touch panels. For example, by operating the operation unit 13, the user can set various inspection conditions for inspecting the inductor element as a DUT in the inspection device 2, and can also instruct the inspection device 2 to execute and stop inspections.

[0059] The output unit 14 is a functional unit for outputting various information such as inspection conditions and inspection results from the inspection device 2. The output unit 14 is a display device equipped with, for example, an LCD (Liquid Crystal Display) or an organic EL. For example, when the user instructs the execution of an inspection of the DUT by operating the operation unit 13, the output unit 14 displays information such as inspection results on the screen in accordance with the control of the data processing control device 10.

[0060] The output unit 14 may also be a display device equipped with a touch panel that performs some of the functions of the operation unit 13. Furthermore, the output unit 14 may include a communication circuit or the like that outputs data such as inspection results to an external source via wired or wireless connection.

[0061] The transport mechanism 15 is a device that transports the inductor elements to be inspected to an appropriate location within the inspection device 2 in accordance with the control of the data processing control device 10. For example, when the first measurement unit 11 performs measurement, the transport mechanism 15 transports the inductor elements to be inspected to a predetermined measurement position by the first measurement unit 11. Also, for example, when the second measurement unit 12 performs measurement, the transport mechanism 15 transports the inductor elements to be inspected to a predetermined measurement position by the second measurement unit 12. Furthermore, the transport mechanism 15 transports the inductor elements that have been determined to be good products after inspection to a location for packaging, and then transports the packaged inductor elements to a predetermined location in the next process.

[0062] The data processing control device 10 is a functional unit that comprehensively controls each functional unit within the inspection device 2 and performs various data processing for the inspection of the DUT. For example, the data processing control device 10 is a program processing unit having a processor such as a CPU, a storage device such as ROM, RAM, or flash memory, and peripheral circuits such as a timer. Examples of program processing units include MCUs and FPGAs.

[0063] 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 determines whether the inductor element under inspection is good or bad based on the calculated index. Here, the index indicating the performance of the inductor element is, for example, the Q value.

[0064] As described above, when the AC resistance of an inductor element is measured using the two-terminal method, the measured value is affected by the resistance component caused by the measurement system using the two-terminal method. Therefore, when the data processing control device 10 of the inspection device 2 according to Embodiment 1 calculates an index (Q value) indicating the performance of the inductor element to be inspected based on the measurement results from the first measurement unit 11 and the second measurement unit 12, it corrects the measured value Rs of the AC resistance measured by the second measurement unit 12 using a pre-generated learned model 35 as needed.

[0065] Before explaining in detail the correction of the measured value Rs of AC resistance performed by the data processing control device 10 using the learned model 35, we will first describe the learned model 35.

[0066] The trained model 35 is a model for estimating the measured value Rs of the AC resistance of the inductor element under test. For example, the trained model 35 is a program generated by machine learning based on a predetermined algorithm. Examples of predetermined algorithms include polynomial regression and multiple regression.

[0067] The trained model 35 according to Embodiment 1 is a function that causes a computer (MPU, etc.) to function to estimate the measured AC resistance Rs (third measurement) of the DUT measured by the two-terminal method, based on input data including the measured DC resistance Rdc4 (first measurement) of the DUT measured by the four-terminal method, the measured DC resistance Rdc2 (second measurement) of the DUT measured by the two-terminal method, and the measured inductance Ls (fourth measurement) of the DUT measured by the two-terminal method.

[0068] In other words, the trained model 35 is a program that causes the information processing device (computer) to function by performing calculations based on predetermined trained parameters on the input measurement data (measured values ​​of DC resistance Rdc4, Rdc2 and measured value of inductance Ls), and outputting a quantitative value (estimated value) of the AC resistance based on the said measurement data.

[0069] For example, the trained model 35 includes a first model representing a first resistance value Rc, which is the value of the resistance component caused by the two-terminal measurement system, and a second model representing a second resistance value Rs0, which is the value of the resistance component caused by the object under test (DUT).

[0070] Here, the first resistance value Rc includes, for example, the resistance component of the transmission line consisting of a cable or probe between the second measuring unit 12 and the DUT, and the resistance component (so-called contact resistance) resulting from the contact state between the probe and the DUT.

[0071] The first model representing the first resistance value Rc is, for example, a regression model in which the measured value Rdc4 (first measurement) of DC resistance obtained by the four-terminal method and the measured value Rdc2 (second measurement) of DC resistance obtained by the two-terminal method are explanatory variables, and the first resistance value Rc of the resistance component caused by the two-terminal measurement system is the dependent variable. In other words, the first model is a function that calculates the first resistance value Rc of the resistance component caused by the two-terminal measurement system from the measured values ​​Rdc4 and Rdc2 of DC resistance.

[0072] The second model representing the second resistance value Rs0 of the resistance component caused by the DUT is, for example, a regression model in which the measured inductance value Ls (fourth measured value) obtained by the two-terminal method is the explanatory variable and the second resistance value Rs0 of the resistance component (AC resistance) caused by the DUT is the dependent variable. In other words, the second model is a function that estimates the second resistance value Rs0 of the resistance component (AC resistance) caused by the DUT from the measured inductance value obtained by the two-terminal method.

[0073] When the first model representing the resistance component (Rc) caused by the two-terminal measurement system is defined as “Rc=f(Rdc4,Rdc2)”, and the second model representing the resistance component (AC resistance Rs0) caused by the DUT is defined as “Rs0=g(Ls)”, the trained model 35, which is a function for calculating the estimated value Rse of the measured AC resistance, can be expressed as Rse=Rc+Rs0=f(Rdc4,Rdc2)+g(Ls).

[0074] In other words, the estimated value Rse of the measured AC resistance is expressed as the sum of the first resistance value Rc, which is the resistance component caused by the two-terminal measurement system determined by the first model (Rc=f(Rdc4,Rdc2)), and the second resistance value Rs0, which is the resistance component (AC resistance) caused by the DUT determined by the second model (Rs0=g(Ls)).

[0075] In the following explanation, the first model (Rc=f(Rdc4,Rdc2)) will be referred to as "model f(Rdc4,Rdc2)", and the second model (Rs0=g(Ls)) will also be referred to as "model g(Ls)".

[0076] Model f(Rdc4,Rdc2) and model g(Ls) include parameters. Parameters are, for example, the coefficients (weighting coefficients) of model f(Rdc4,Rdc2) and model g(Ls). In the following explanation, parameters that have been mechanically adjusted to calculate the estimated value Rse of the measured AC resistance using the training measurement data 34_1~34_n described later as input to the training program (a program based on the predetermined algorithm described above) are also referred to as "trained parameters".

[0077] Next, we will describe the trained model generator 3 that generates the trained model 35.

[0078] Figure 2 shows an example of the configuration of the trained model generation device 3 in the inspection system 1 according to Embodiment 1.

[0079] The trained model generation device 3 is implemented by an information processing device (computer), such as a server or a personal computer (PC), and generates multiple training measurement data 34_1 to 34_n (where n is an integer of 2 or more) according to an installed trained model generation program, and generates a trained model 35 by machine learning the generated training measurement data based on a predetermined algorithm.

[0080] For the sake of explanation, Figure 1 shows the trained model generation device 3 and the inspection device 2 placed side by side. However, the trained model generation device 3 and the inspection device 2 do not necessarily have to be installed in the same location. For example, the inspection device 2 and the trained model generation device 3 may be installed in different locations and connected via a communication network such as a LAN or the Internet. In this case, the inspection device 2 and the trained model generation device 3 may send and receive various data, such as measurement data from the inspection device 2 and the trained model 35, via the communication network.

[0081] Furthermore, during DUT inspection, the inspection device 2 and the trained model generation device 3 do not necessarily need to be electrically connected to each other. For example, before or after inspection, various data such as measurement data and trained models 35 from the inspection device 2 may be exchanged via a storage medium such as a memory card.

[0082] The trained model 35 may be distributed via a network, or it may be written to a computer-readable storage medium such as a memory card and distributed therein.

[0083] The trained model generation device 3 includes, for example, a training 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 realized through the cooperation of hardware resources such as the CPU and memory that constitute the information processing device as the trained model generation device 3, with software (various programs including a trained model generation program) installed on the information processing device.

[0084] The training measurement data acquisition unit 31 is a functional unit that acquires training measurement data 34_1 to 34_n necessary for generating the trained model 35.

[0085] Here, the learning measurement data 34_1 to 34_n are data pairs that associate the measured value Rdc4 (first measurement) of the DC resistance of the DUT measured by the 4-terminal method, the measured value Rdc2 (second measurement) of the DC resistance of the DUT measured by the 2-terminal method, the measured value Ls (fourth measurement) of the inductance of the DUT measured by the 2-terminal method, with the measured value Rs (third measurement) of the AC resistance of the DUT measured by the 2-terminal method. In the following explanation, when the learning measurement data 34_1 to 34_n are not distinguished, they will simply be referred to as "learning measurement data 34".

[0086] The learning measurement data acquisition unit 31 acquires a data pair, for example, via wireless or wired communication (not shown) or a storage medium such as a memory card, which includes data 41 of the measured DC resistance Rdc4 of the inductor element measured by the four-terminal method, data 42 of the measured DC resistance Rdc2 of the inductor element measured by the two-terminal method, data 43 of the measured inductance Ls of the DUT measured by the two-terminal method, and data 43 of the measured AC resistance Rs of the DUT measured by the two-terminal method. The learning measurement data acquisition unit 31 acquires a data pair for each inductor element that is inspected. As the data pair, for example, measurement results of the inductor element obtained in the past by the inspection device 2 or by manual measurement using a fixture may be used.

[0087] The learning measurement data acquisition unit 31 generates a learning measurement data 34 by, for example, associating the measured value of AC resistance Rs included in the acquired data pair with the measured values ​​of DC resistance Rdc4, Rdc2 and inductance Ls included in the data pair, using the measured value as the correct answer. 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 them in the storage unit 33.

[0088] Furthermore, instead of generating the learning measurement data 34 itself, as described above, the learning measurement data acquisition unit 31 may acquire the learning measurement data 34, which has been measured manually and stored in another information processing device, via communication or a storage medium.

[0089] The trained model generation unit 32 is a functional unit that generates a trained model 35 by machine learning multiple training measurement data 34_1 to 34_n acquired by the training measurement data acquisition unit 31.

[0090] The trained model generation unit 32 adjusts the parameters of model f(Rdc4,Rdc2) and model g(Ls) by machine learning the training measurement data 34_1 to 34_n, thereby obtaining trained parameters.

[0091] Specifically, the trained model generation unit 32 generates a trained model 35 (Rse = f(Rdc4, Rdc2) + g(Ls)) by individually performing machine learning on model f(Rdc4, Rdc2) and model g(Ls). That is, the trained model generation unit 32 separately executes the process of adjusting the parameters of model f(Rdc4, Rdc2) using machine learning and the process of adjusting the parameters of model g(Ls) using machine learning. In addition, the trained model generation unit 32 generates an adjustment value Rtng1 to adjust the value estimated by the trained model 35 in order to reduce the influence of errors included in the definition value of the reference device used in the calibration process.

[0092] As shown in FIG. 2, the learned model generation unit 32 includes a first model generation unit 321, a second model generation unit 322, and an adjustment value calculation unit 323.

[0093] The second model generation unit 322 generates a model g(Ls) by performing machine learning on learning measurement data 34 with little variation in contact resistance during measurement by the two-terminal method. For example, the second model generation unit 322 generates a model (Rs = g(Ls)) by performing machine learning on learning measurement data 34 (|Rdc2 - Rdc4| < Rdth) in which the difference between the measured value Rdc4 of the DC resistance and the measured value Rdc2 of the DC resistance is smaller than the threshold value Rdth. Here, the threshold value Rdth is preferably a value as close to "0" as possible.

[0094] For example, without using the inspection device 2, measurement data of the measured value Rdc4 of the DC resistance, the measured value Rdc2 of the DC resistance, the measured value Ls of the inductance, and the measured value Rs of the AC resistance of a plurality of DUTs with different inductance values are obtained by manual operation using a fixture on the desktop. These measurement data are stored in the storage unit 33 of the learned model generation device 3 as learning measurement data 34. At this time, the learning measurement data 34 includes measurement data (|Rdc2 - Rdc4| < Rdth) in which the difference between the measured value Rdc4 of the DC resistance and the measured value Rdc2 of the DC resistance is smaller than the threshold value Rdth.

[0095] At this time, the number of learning measurement data 34 used for machine learning when generating the model g(Ls) may be less than the number of learning measurement data 34 used for machine learning when generating the model f(Rdc4, Rdc2).

[0096] For example, about 10 samples each with different inductances (e.g., L = 1 nH, 5 nH, 10 nH, etc.) are measured, and measurement data in which the difference between the measured value Rdc4 of the DC resistance and the measured value Rdc2 of the DC resistance is smaller than the threshold value Rdth may be stored in the storage unit 33 as learning measurement data 34.

[0097] The second model generation unit 322 calculates the second resistance value Rs0 of the resistance component (alternating current resistance) due to the DUT by inputting the measured value Rs of the alternating current resistance included in the learning measurement data 34, in which the difference between the measured value Rdc4 of the direct current resistance and the measured value Rdc2 of the direct current resistance is smaller than the threshold value Rdth, obtained by the above-described method, into the regression model (g(Ls)). Next, the second model generation unit 322 calculates the difference (error) between the measured value Rs of the alternating current resistance, which is the correct value included in the learning measurement data 34, and the second resistance value Rs0 calculated by the regression model (g(Ls)). Then, the learned model generation unit 32 sequentially updates the parameters of the regression model (Rs0 = g(Ls)) so that the calculated error becomes smaller, for example, by the error backpropagation method, to generate the learned parameters of the model g(Ls) and stores them in the storage unit 33.

[0098] The first model generation unit 321 generates a model f(Rdc4, Rdc2) by machine-learning the learning measurement data 34. Here, in the machine learning by the first model generation unit 321, it is preferable to use the learning measurement data 34 in which the variation in the contact resistance during measurement by the two-terminal method is large. For example, not only the learning measurement data 34 in which the difference between the measured value Rdc4 of the direct current resistance and the measured value Rdc2 of the direct current resistance is smaller than the threshold value (|Rdc2 - Rdc4| < Rdth), but also the learning measurement data 34 in which the difference between the measured value Rdc4 of the direct current resistance and the measured value Rdc2 of the direct current resistance is larger than the threshold value (|Rdc2 - Rdc4| > Rdth) are preferably used for the machine learning by the first model generation unit 321. The first model generation unit 321 generates a model f(Rdc4, Rdc2) by machine-learning the measured value Ls of the inductance included in the learning measurement data 34.

[0099] For example, after calibrating the inspection device 2 using a reference device for the DUT, multiple DUTs are measured using the inspection device 2 to obtain measurement data for DC resistance (Rdc4, Rdc2), inductance (Ls), and AC resistance (Rs). For example, hundreds to thousands of samples with different inductances are measured. These measurement data are stored as training measurement data 34 in the storage unit 33 of the trained model generation device 3.

[0100] The first model generation unit 321 adjusts the parameters of model g(Ls), and then uses the adjusted parameters of model g(Ls) to adjust the parameters of f(Rdc4,Rdc2).

[0101] For example, the first model generation unit 321 preferably adjusts the parameters of f(Rdc4,Rdc2) by machine learning the training measurement data 34 while fixing the learned parameters of model g(Ls) in the model (f(Rdc4,Rdc2)+g(Ls)) which is represented as the sum of model f(Rdc4,Rdc2) and model g(Ls).

[0102] However, the relationship between inductance (Ls) and AC resistance (Rs) when acquiring measurement data used for machine learning of model g(Ls) may differ from the relationship between inductance (Ls) and AC resistance (Rs) when performing inspections using the calibration-processed inspection device 2. For example, if the measurement error included in the measurement data used for machine learning of model g(Ls) differs from the measurement error included in the defined value (AC resistance Rs) of the reference device in the calibration process of inspection device 2, or if the characteristics of the DUT used for machine learning differ from the identification of the DUT during inspection, there is a risk of discrepancies between the estimated AC resistance calculated by inputting the measured inductance value of the DUT measured by the calibration-processed inspection device 2 into model g(Ls) and the measured AC resistance value of the DUT.

[0103] FIG. 3 is a diagram for explaining the deviation between the second resistance value Rs0 of the AC resistance calculated by the model g(Ls) and the defined value (AC resistance Rs) of the DUT in the calibration process.

[0104] In FIG. 3, the horizontal axis represents the value of the inductance (Ls) of the DUT, and the vertical axis represents the AC resistance Rs of the DUT. In FIG. 3, reference numeral 301 indicates the model g(Ls), and reference numeral 302 indicates the defined value of the reference device (the defined value of the load reference).

[0105] For example, as described above, when the measurement error included in the measurement data used for the machine learning of the model g(Ls) is different from the measurement error included in the defined value (AC resistance Rs) of the reference device in the calibration process of the inspection device 2, as shown in FIG. 3, a difference (offset ΔRs) occurs between the estimated value of the second resistance value Rs0 of the AC resistance Rs calculated by the model g(Ls) and the defined value Rs of the reference device, and there is a possibility that the estimation accuracy of the learned model may decrease.

[0106] Therefore, in order to reduce the influence of the offset ΔRs shown in FIG. 3 on the accuracy of the learned model, the learned model generation device 3 according to Embodiment 1 calculates an adjustment value Rtng1 corresponding to the offset ΔRs, and performs machine learning of the model f(Rdc4, Rdc2) using the adjustment value Rtng1.

[0107] Specifically, first, the adjustment value calculation unit 323 calculates the adjustment value Rtng1 according to the difference between the reference AC resistance value and the estimated value of the second resistance value Rs0 calculated by inputting the measured value Ls of the inductance included in at least one piece of learning measurement data 34 into the model g(Ls).

[0108] Here, the reference AC resistance value is the average value Rs_av1 of the measured values Rs of the AC resistance included in the learning measurement data 34 (|Rdc2−Rdc4|<Rdth) of a plurality of DUTs in which the difference between the measured value Rdc4 of the DC resistance and the measured value Rdc2 of the DC resistance is smaller than the threshold value Rdth.

[0109] The adjustment value calculation unit 323 calculates an estimated value g(Ls_av1) of the second resistance value Rs0 by inputting the average value Ls_av1 of the measured inductance values Ls included in the learning measurement data 34 of a plurality of DUTs where the difference between the measured value Rdc4 of the DC resistance and the measured value Rdc2 of the DC resistance is smaller than the threshold value Rdth (|Rdc2 - Rdc4| < Rdth) into the model g(Ls). Next, the difference between the average value Rs_av1 of the measured AC resistance value Rs and the estimated value g(Ls_av1) of the second resistance value Rs0 is calculated as an adjustment value Rtng1 (= Rs_av1 - g(Ls_av1)) and stored in the storage unit 33.

[0110] Alternatively, the reference AC resistance value may be the defined value Rs_d1 of the AC resistance of the reference device in the calibration process. Information on the defined value Rs_d1 of the AC resistance may be stored in advance in the storage unit 33 of the pre-trained model generation device 3 after the calibration process of the inspection system 1 described above.

[0111] In this case, the adjustment value calculation unit 323 calculates an estimated value g(Ls_av1) of the second resistance value Rs0 by inputting the average value Ls_av1 of the measured inductance values Ls included in the learning measurement data 34 of a plurality of DUTs where the difference between the measured value Rdc4 of the DC resistance and the measured value Rdc2 of the DC resistance is smaller than the threshold value Rdth (|Rdc2 - Rdc4| < Rdth) into the model g(Ls). Next, the difference between the defined value Rs_d of the AC resistance and the estimated value of the second resistance value Rs0 is calculated as an adjustment value Rtng1 (= Rs_d1 - g(Ls_av1)) and stored in the storage unit 33.

[0112] The first model generation unit 321 adjusts the parameters of the model f(Rdc4, Rdc2) in a state where the parameters of the already learned model g(Ls) are fixed in the model represented by the sum of the model f(Rdc4, Rdc2), the model g(Ls), and the adjustment value Rtng1.

[0113] Specifically, the first model generation unit 321 calculates an estimated value Rse of the AC resistance measurement by inputting the measured values ​​Rdc4, Rdc2, and Ls of the DC resistance included in the learning measurement data 34 into a regression model (Rse = f(Rdc4, Rdc2) + g(Ls) + Rtng1). Next, the first model generation unit 321 calculates the difference (error) between the calculated estimated value Rse of the AC resistance measurement and the correct value Rs of the AC resistance measurement included in the learning measurement data 34. Next, the first model generation unit 321 sequentially updates the parameters of model f(Rdc4,Rdc2) in the regression model (Rse=f(Rdc4,Rdc2)+g(Ls)+Rtng1) to reduce the calculated error, for example by backpropagation, thereby generating learned parameters for model f(Rdc4,Rdc2) and storing them in the storage unit 33. This allows us to generate a model f(Rdc4,Rdc2) that reduces the effect of the offset ΔRs.

[0114] The trained model generation unit 32 stores a regression model (Rse=f(Rdc4,Rdc2)+g(Ls)) containing the trained parameters adjusted by the method described above as a trained model 35 in the storage unit 33.

[0115] Note that the adjustment value Rtng1 is not included in the model (Rse=f(Rdc4,Rdc2)+g(Ls)) registered as trained model 35. In other words, the adjustment value Rtng1 is only considered when adjusting the parameters of model f(Rdc4,Rdc2).

[0116] The memory unit 33 is a functional unit for storing various data, such as the training measurement data 34_1 to 34_n necessary for generating the trained model 35, and the generated trained model 35.

[0117] The memory unit 33 is configured to be accessible from the outside, for example. For example, by the inspection device 2 communicating with the trained model generation device 3, the inspection device 2 can read and acquire the trained model 35 from the memory unit 33. Also, for example, by the inspection device 2 communicating with the trained model generation device 3, the inspection device 2 can write measurement result data, etc., to the memory unit 33.

[0118] Figure 4 is a flowchart showing the flow of generating a trained model 35 by the trained model generation device 3 according to Embodiment 1.

[0119] As shown in Figure 4, first, in the trained model generation device 3, the training measurement data acquisition unit 31 acquires training measurement data 34_1 to 34_n (step S1). Specifically, as described above, the training measurement data acquisition unit 31 acquires data pairs including the measured values ​​of DC resistance Rdc4 and Rdc2 and the measured value of AC resistance Rs of inductor elements that have been inspected in the past. Based on the acquired data pairs, the training measurement data 34 is generated by assigning the measured value of AC resistance Rs as the correct value to the measured values ​​of DC resistance Rdc4 and Rdc2.

[0120] As described above, the training measurement data 34 includes measurement data of the DUT measured manually using a fixture and measurement data of the DUT measured by the inspection device 2 that has undergone calibration processing. Here, it is preferable that each training measurement data 34 is accompanied by information that allows identification of whether it is measurement data measured manually using a fixture or measurement data measured by the inspection device 2.

[0121] Next, the learned model generation device 3 determines whether the required number of learning measurement data 34 has been generated to generate the learned model 35 (step S2). For example, the number of data of the learning measurement data 34 required to generate the learned model 35 is preset in the learned model generation device 3, and each time the learning measurement data 34 is generated, the learned model generation device 3 increments the generation count by +1. Then, the learned model generation device 3 determines whether the counted generation count has reached the preset number of data, thereby determining whether the required number of learning measurement data 34 has been generated.

[0122] If the required number of learning measurement data 34 has not been generated (step S2: NO), the learned model generation device 3 returns to step S1, obtains a data pair related to the measurement result of a new inductor element, and repeats generating the learning measurement data 34 related to the inductor element (steps S1, S2).

[0123] On the other hand, if the required number of learning measurement data 34 has been generated (step S2: YES), the learned model generation device 3 performs machine learning using the plurality of learning measurement data 34 generated in step S1 (step S3).

[0124] FIG. 5 is a flowchart showing the flow of machine learning (step S3) according to Embodiment 1.

[0125] Here, as an example, the case where the average value Rs_av2 of the measured values Rs of the AC resistance included in the learning measurement data 34 of a plurality of DUTs (|Rdc2 - Rdc4| < Rdth) in which the difference between the measured value Rdc4 of the DC resistance and the measured value Rdc2 of the DC resistance is smaller than the threshold value Rdth is used as the reference value of the AC resistance will be described. <0​​First, the second model generation unit 322 extracts learning measurement data 34 with small variation in DC resistance from the learning measurement data 34_1 to 34_n (step S31). For example, as described above, among the learning measurement data 34 measured manually using the fixture, the learning measurement data 34 where the difference between the measured value Rdc4 of the DC resistance and the measured value Rdc2 of the DC resistance is smaller than the threshold (|Rdc2 - Rdc4| < Rdth) is extracted.

[0127] Next, the second model generation unit 322 generates a model g(Ls) using the learning measurement data 34 extracted in step S31 by the method described above (step S32).

[0128] Next, the adjustment value calculation unit 323 calculates the average value Rs_av1 of the measured value Rs of the AC resistance (step S33). Specifically, the adjustment value calculation unit 323 extracts the learning measurement data 34 where the difference between the measured value Rdc4 of the DC resistance and the measured value Rdc2 of the DC resistance is smaller than the threshold (|Rdc2 - Rdc4| < Rdth) among the learning measurement data 34 measured by the inspection system 1 in which the calibration process has been performed. The adjustment value calculation unit 323 calculates the average value Rs_av1 of the measured value Rs of the AC resistance included in the extracted plurality of learning measurement data 34 (|Rdc2 - Rdc4| < Rdth).

[0129] Next, the adjustment value calculation unit 323 calculates the average value Ls_av1 of the measured value Ls of the inductance (step S34). Specifically, the adjustment value calculation unit 323 calculates the average value Ls_av1 of the measured value Ls of the inductance included in the plurality of learning measurement data 34 (|Rdc2 - Rdc4| < Rdth) used for calculating the average value Rs_av1 of the measured value Rs of the AC resistance in step S33.

[0130] Next, the adjustment value calculation unit 323 calculates the adjustment value Rtng1 (step S35). Specifically, the adjustment value calculation unit 323 generates the adjustment value Rtng1 (=Rs_av1-g(Ls_av1)) by subtracting the value g(Ls_av1), which is obtained by inputting the average value Ls_av1 of the measured inductance Ls calculated in step S34 into the model g(Ls) generated in step S32, from the average value Rs_av1 of the measured AC resistance Rs calculated in step S33.

[0131] Next, the first model generation unit 321 extracts training measurement data 34 with large variations in DC resistance from the training measurement data 34_1 to 34_n (step S36). For example, the first model generation unit 321 extracts at least training measurement data 34 (|Rdc2 - Rdc4| > Rdth) from the training measurement data 34 measured by the calibration inspection system 1 in which calibration processing has been performed, in which the difference between the measured value of DC resistance Rdc4 and the measured value of DC resistance Rdc2 is greater than a threshold.

[0132] Next, the first model generation unit 321 sets a model that takes into account the adjustment value Rtng1 (step S37). Specifically, as described above, the first model generation unit 321 sets a model (Rse = f(Rdc4,Rdc2) + g(Ls) + Rtng1) which is the sum of the model f(Rdc4,Rdc2), the model g(Ls) generated in step S32, and the adjustment value Rtng1 calculated in step S35.

[0133] Next, the first model generation unit 321 uses the model set in step S37 (Rse = f(Rdc4, Rdc2) + g(Ls) + Rtng1) to adjust the parameters of the model f(Rdc4, Rdc2) while fixing the parameters of the already trained model g(Ls).

[0134] Through the above processing flow, the parameters of model g(Ls) and model f(Rdc2,Rdc4) are individually adjusted, and the trained model 35 (Rse=f(Rdc2,Rdc4)+g(Ls)) is generated.

[0135] Next, as shown in Figure 4, the trained model 35 generated in step S3 is registered with the inspection device 2 (step S4). For example, in response to a user's operation of the operation unit 13 of the inspection device 2 or the input device (e.g., touch panel, keyboard, mouse, etc.) of the trained model generation device 3, the trained model generation device 3 transmits the trained model 35 stored in the storage unit 33 to the inspection device 2, and the inspection device 2 stores the received trained model 35 in the storage unit of the data processing control device 10. Note that the registration of the trained model 35 to the inspection device 2 may also be performed using a storage medium such as a memory card, as described above.

[0136] Furthermore, the program for generating a trained model, which causes the computer (information processing device) acting as the trained model generation device 3 to execute each of the above steps (S1 to S4), may be distributed via a network, or it may be written to a computer-readable storage medium such as a memory card and distributed therein.

[0137] Using the method described above, a trained model 35 for testing inductor elements is generated.

[0138] Next, we will explain in detail how to correct the measured value Rs of AC resistance using the trained model 35 from the inspection device 2.

[0139] Similar to the process for acquiring training measurement data 34, calibration is performed in the inspection device 2 before starting the automated inspection of the DUT by the inspection device 2 with the trained model 35 installed. Therefore, the inspection device 2 may also experience a decrease in inspection accuracy due to errors included in the defined values ​​of the reference device. To address this, the inspection device 2 calculates an adjustment value Rtng2, similar to the trained model generation device 3, and uses the adjustment value Rtng2 to adjust the value estimated by the trained model 35.

[0140] Figure 6 shows an example of the configuration of the data processing control device 10 in the inspection device 2 according to Embodiment 1.

[0141] As shown in Figure 6, the data processing control device 10 of the inspection device 2 includes, for example, a data acquisition unit 21, a storage unit 22, an estimation unit 23, a correction unit 24, a determination unit 25, and an adjustment value calculation unit 26. These functional units are realized, for example, in a program processing device as the data processing control device 10, by a CPU executing various calculations according to a program stored in memory and controlling peripheral circuits such as counters.

[0142] The data acquisition unit 21 is a functional unit that acquires various data necessary to calculate the performance index (Q value) of the inductor element under inspection.

[0143] 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 using the four-terminal method and stores it in the storage unit 22. The data acquisition unit 21 also acquires, for example, the measured value Rdc2 of the DC resistance of the DUT measured by the second measurement unit 12 using the two-terminal method, the measured value Rs of the AC resistance of the DUT measured by the second measurement unit 12 using the two-terminal method, and the measured value Ls of the inductance of the DUT measured by the second measurement unit 12 using the two-terminal method, and stores them in the storage unit 22 as measurement data 50 of the object to be inspected. In addition, 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.

[0144] The memory unit 22 is a functional unit for storing various data necessary for calculating the performance index (Q value) of the inductor element under inspection, as well as the calculated Q value.

[0145] As described above, the memory 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, the measured value Ls of the inductance of the inductor element, and the learned model 35, which are acquired by the data acquisition unit 21. Further, the memory unit 22 stores, for example, the estimated value Rse of the measured value of the AC resistance, the estimated value Rc of the resistance component due to the measurement system by the two-terminal method, the value Rsr of the AC resistance, the Q value, and the adjustment value Rtng1, which will be described later.

[0146] The adjustment value calculation unit 26 calculates an adjustment value Rtng2 corresponding to the offset ΔRs shown in FIG. 3. Specifically, the adjustment value calculation unit 26 calculates an adjustment value Rtng2 according to the difference between the reference AC resistance value and the estimated value of the second resistance value Rs0 calculated by inputting the measured value Ls of the inductance included in at least one measurement data 50 into the model g(Ls).

[0147] Here, the reference AC resistance value is the average value Rs_av2 of the measured values Rs of the AC resistance included in the measurement data 50 of a plurality of DUTs where the difference between the measured value Rdc4 of the DC resistance and the measured value Rdc2 of the DC resistance is smaller than the threshold value Rdth (|Rdc2 - Rdc4| < Rdth).

[0148] The adjustment value calculation unit 26 calculates the estimated value g(Ls_av2) of the second resistance value Rs0 by inputting the average value Ls_av2 of the measured values Ls of the inductance included in the measurement data 50 of a plurality of DUTs where the difference between the measured value Rdc4 of the DC resistance and the measured value Rdc2 of the DC resistance is smaller than the threshold value Rdth (|Rdc2 - Rdc4| < Rdth) into the model g(Ls). The adjustment value calculation unit 26 calculates the difference between the average value Rs_av2 of the measured values Rs of the AC resistance and the estimated value g(Ls_av2) of the second resistance value Rs0 as the adjustment value Rtng2 (= Rs_av2 - g(Ls_av2)), and stores it in the memory unit 22.

[0149] Alternatively, the reference AC resistance value may be the defined value Rs_d2 of the AC resistance of the reference device in the calibration process of the inspection device 2. Information on the defined value Rs_d2 of the AC resistance may be stored in advance in the storage unit 22 of the data processing control device 10 after the calibration process of the inspection device 2 described above.

[0150] In this case, the adjustment value calculation unit 26 inputs the average value Ls_av2 of the measured values Ls of the inductance included in the measurement data 50 of a plurality of DUTs (|Rdc2−Rdc4|<Rdth) where the difference between the measured value Rdc4 of the DC resistance and the measured value Rdc2 of the DC resistance is smaller than the threshold value Rdth into the model g(Ls), and calculates the estimated value g(Ls_av2) of the second resistance value Rs0. The adjustment value calculation unit 26 calculates the difference between the defined value Rs_d2 of the AC resistance and the estimated value g(Ls_av2) of the second resistance value Rs0 as the adjustment value Rtng2 (=Rs_d2−g(Ls_av2)), and stores it in the storage unit 22.

[0151] The estimation unit 23 is a functional unit that estimates the measured value of the AC resistance of the inductor element to be inspected. The estimation unit 23 calculates an estimated value Rse of the measured value of the AC resistance corresponding to the measured values Rdc4 and Rdc2 of the DC resistance of the inductor element to be inspected, which are acquired by the data acquisition unit 21, based on the learned model 35 stored in the storage unit 22 and the adjustment value Rtng2.

[0152] Specifically, the estimation unit 23 adjusts the value obtained by inputting the measured values Rdc4 and Rdc2 of the DC resistance and the measured value Ls of the inductance of the inductor element to be inspected, which are acquired by the data acquisition unit 21, into the learned model 35 (function) by the adjustment value Rtng2, and stores the adjusted value in the storage unit 22 as the estimated value Rse of the AC resistance.

[0153] For example, the estimation unit 23 stores, in the storage unit 22, as the estimated value Rse of the AC resistance, the value calculated by inputting the measured values Rdc4 and Rdc2 of the DC resistance and the measured value Ls of the inductance included in the measurement data 50 into the model (f(Rdc4,Rdc2)+g(Ls)+Rtng2) obtained by adding the adjustment value Rtng2 to the learned model 35.

[0154] The correction unit 24 is a functional unit for correcting the measured value Rs of AC resistance. The correction unit 24 corrects the measured value Rs of AC resistance based on the measured values ​​Rdc4 and Rdc2 of DC resistance acquired by the data acquisition unit 21, according to the estimated value Rse of the measured value Rs of AC resistance, and performs a correction process to output the corrected measured value Rs of AC resistance as the AC resistance value Rsr of the DUT.

[0155] More specifically, the correction unit 24 first calculates the resistance component Rc caused by the two-terminal measurement system according to the model f(Rdc4,Rdc2) described above, based on the measured values ​​Rdc4 and Rdc2 of the DC resistance of the inductor element under inspection acquired by the data acquisition unit 21. The calculated resistance component Rc data is stored, for example, in the storage unit 22.

[0156] Next, the correction unit 24 corrects the measured value Rs of the AC resistance of the inductor element under inspection based on the resistance component Rc caused by the two-terminal measurement system, according to the estimated value Rse of the measured AC resistance, and outputs the corrected measured value Rs of AC resistance as the value Rsr of the AC resistance of the inductor element under inspection.

[0157] For example, the correction unit 24 calculates the error |Rse-Rs| between the estimated value Rse of the AC resistance measured by the estimation unit 23 and the measured value Rs (third measured value) of the AC resistance acquired by the data acquisition unit 21, and evaluates the error |Rse-Rs|. Specifically, the correction unit 24 compares the error |Rse-Rs| with a threshold value Rth. The threshold value Rth is an arbitrary value set in advance.

[0158] Here, if the error |Rse-Rs| is smaller than the threshold Rth, it can be considered that the accuracy of the AC resistance estimation by the trained model 35 is high for the measurement results of the inductor element under test. In other words, it can be considered that the accuracy of the estimation of the resistance component Rc caused by the two-terminal measurement system is high by the model f(Rdc4,Rdc2) included in the trained model 35.

[0159] Therefore, if the error |Rse-Rs| is smaller than the threshold Rth, the correction unit 24 performs a correction process. Specifically, the correction unit 24 calculates the resistance component Rc caused by the two-terminal measurement system using model f(Rdc4,Rdc2), and corrects the measured value Rs (third measured value) of the AC resistance using the calculated resistance component Rc caused by the two-terminal measurement system.

[0160] For example, the correction unit 24 first inputs (substitutes) the measured values ​​Rdc4 and Rdc2 of the DC resistance of the inductor element under test into the model f(Rdc4, Rdc2) and uses the resulting value as the resistance component Rc caused by the two-terminal measurement system. Next, the correction unit 24 subtracts the resistance component Rc caused by the two-terminal measurement system from the measured value Rs (third measured value) of AC resistance acquired by the data acquisition unit 21 and outputs the resulting value as the AC resistance value Rsr (= Rs - Rc).

[0161] On the other hand, if the error |Rse-Rs| is greater than the threshold Rth, it is considered that the accuracy of the AC resistance estimation by the trained model 35 for the inductor element under test is low. In other words, it is considered that the accuracy of the resistance component Rc caused by the two-terminal measurement system is low, as indicated by model f(Rdc4,Rdc2) included in the trained model 35. In this case, if we were to use model f(Rdc4,Rdc2) to calculate the resistance component Rc caused by the two-terminal measurement system, and then use the calculated resistance component Rc to correct the measured value Rs (third measured value) of the AC resistance, it would result in an incorrect correction, and there is a risk that we would not be able to properly determine the value of the AC resistance Rsr.

[0162] Therefore, if the error |Rse-Rs| is greater than the threshold Rth, the correction unit 24 outputs the measured value Rs of the AC resistance acquired by the data acquisition unit 21 as the AC resistance value Rsr (=Rs) of the inductor element under inspection, without performing any correction processing.

[0163] The determination unit 25 is a functional unit for determining whether the DUT (inductor element) is good or bad. Based on the AC resistance value Rsr of the inductor element to be tested output from the correction unit 24 and the measured inductance value Ls of the inductor element to be tested, the determination unit 25 calculates the Q value (Q = ωL / Rsr), which is an index representing the performance of the inductor element to be tested.

[0164] The determination unit 25 determines whether the inductor element under inspection is good or bad by, for example, comparing the calculated Q value with a predetermined reference value. The determination unit 25 controls the transport mechanism 15 to package the DUTs that it has determined to be good into a ready-to-ship state using a packaging device not shown in the diagram.

[0165] Next, we will explain the inspection process of the DUT using inspection device 2.

[0166] Figure 7 is a flowchart showing the inspection process by the inspection device 2 according to Embodiment 1.

[0167] First, the user performs a calibration process on the inspection device 2 (step S11). In the calibration process, the user selects one reference device (golden device) from among the DUTs under inspection. Next, the user manually measures the electrical characteristics of the reference device using a fixture in a low-noise environment and determines the measured value as the defined value of the reference device. Then, the user uses the inspection device 2 to measure the reference device under multiple criteria (e.g., open criterion, short criterion, and load criterion) and obtains the measured value for each criterion. The inspection device 2 then generates a correction formula to correct the measured value so that the error between the measured value and the previously obtained defined value is small, and stores it in the internal memory of the inspection device 2. This completes the series of calibration processes. At this time, the defined value Rs_d2 of the AC resistance of the reference device used in the calibration process is stored in the storage unit 22 of the data processing control device 10.

[0168] After the calibration process, for example, when the user operates the operation unit 13 of the inspection device 2 to instruct the execution of the inspection of the DUT (inductor element), the data processing control device 10 starts an adjustment value calculation process for obtaining the adjustment value Rtng1 as a preparation process for the inspection of the DUT (step S12).

[0169] FIG. 8 is a flowchart showing the flow of the adjustment value calculation process (step S12) according to Embodiment 1.

[0170] Here, as an example, a case where the average value Rs_av2 of the measured values Rs of the AC resistance included in the measurement data 50 of a plurality of DUTs (|Rdc2−Rdc4|<Rdth) in which the difference between the measured value Rdc4 of the DC resistance and the measured value Rdc2 of the DC resistance is smaller than the threshold value Rdth is used as the reference value of the AC resistance will be described.

[0171] In step S12, first, the data processing control device 10 acquires the measurement data of a plurality of inductor elements to be inspected (step S121). For example, the data processing control device 10 controls the first measurement unit 11 and the second measurement unit 12 to acquire measurement data 50 including the measured values Rdc4 and Rdc2 of the DC resistance, the measured value Rs of the AC resistance, and the measured value Ls of the inductance of the inductor element to be inspected.

[0172] The measured values Rdc4 and Rdc2 of the DC resistance, the measured value Rs of the AC resistance, and the measured value Ls of the inductance acquired by the data processing control device 10 by the above method are stored in the storage unit 22 as the measurement data 50 of the inductor element to be inspected.

[0173] Next, the data processing control device 10 extracts a plurality of measurement data 50 with small variations in contact resistance from the measurement data 50 acquired in step S121 (step S122). Specifically, the adjustment value calculation unit 26 extracts a plurality of measurement data 50 of DUTs (|Rdc2−Rdc4|<Rdth) in which the difference between the measured value Rdc4 of the DC resistance and the measured value Rdc2 of the DC resistance is smaller than the threshold value Rdth from the measurement data 50 stored in the storage unit 22.

[0174] Next, the adjustment value calculation unit 26 calculates the average value Rs_av2 of the measured AC resistance Rs included in the multiple measurement data 50 acquired in step S122 (step S123). The adjustment value calculation unit 26 also calculates the average value Ls_av2 of the measured inductance Ls included in the multiple measurement data 50 acquired in step S122 (step S124).

[0175] Next, the adjustment value calculation unit 26 calculates an estimated value g(Ls_av2) of the second resistance value Rs0 by inputting the average value Ls_av2 of the inductance measurement values ​​Ls calculated in step S124 into the model g(Ls) (step S125).

[0176] Next, the adjustment value calculation unit 26 calculates the adjustment value Rtng2 (=Rs_av2-g(Ls_av2)) by subtracting the estimated value g(Ls_av2) of the second resistance value Rs0 calculated in step S125 from the average value Rs_av2 of the measured AC resistance Rs calculated in step S123 (step S126). Through this processing flow, the adjustment value Rtng2 can be obtained.

[0177] Furthermore, if the reference AC resistance value is the defined value Rs_d2 of the AC resistance of the reference device in the calibration process of the inspection device 2, then in step S123, the adjustment value calculation unit 26 reads the defined value Rs_d2 of the AC resistance of the reference device from the storage unit 22. Then, in step S126, the adjustment value calculation unit 26 calculates the adjustment value Rtng2 (=Rs_d2-g(Ls_av2)) by subtracting the estimated value g(Ls_av2) of the second resistance value Rs0 calculated in step S125 from the defined value Rs_d2 of the AC resistance of the reference device.

[0178] After step S12, the data processing control device 10 starts the inspection process of the inductor element to be inspected (step S13).

[0179] Figure 9 is a flowchart showing the flow of inspection (step S13) by the inspection device 2 according to Embodiment 1.

[0180] In step S13, first, the data processing control device 10 controls the first measurement unit 11 to measure the DC resistance of the inductor element to be inspected using the four-terminal method (step S131). For example, the data processing control device 10 controls the transport mechanism 15 in response 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. Subsequently, the data processing control device 10 controls the first measurement unit 11 to measure the DC resistance of the inductor element to be inspected using the four-terminal method and obtains the measured value of the DC resistance Rdc4.

[0181] Next, the data processing control device 10 controls the second measurement unit 12 to measure the DC resistance of the inductor element to be inspected using the two-terminal method (step S132). 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 measurement unit 12. After that, the data processing control device 10 controls the second measurement unit 12 to measure the DC resistance of the inductor element to be inspected using the two-terminal method and obtains the measured value of the DC resistance Rdc2.

[0182] Next, the data processing control device 10 controls the second measurement unit 12 to measure the measured value of the AC resistance Rs and the measured value of the inductance Ls of the inductor element to be inspected using the two-terminal method (step S133). For example, with the inductor element to be inspected placed in the same measurement position as in step S12, the data processing control device 10 controls the second measurement unit 12 to measure the AC resistance of the inductor element to be inspected and obtains the measured value of the AC resistance Rs and the measured value of the inductance Ls, respectively.

[0183] In steps S131 to S133, the measured values ​​of DC resistance Rdc4 and Rdc2, AC resistance Rs, and inductance Ls acquired by the data processing control device 10 are stored in the storage unit 22 as measurement data 50 of the inductor element to be inspected.

[0184] Next, the data processing control device 10 calculates an estimated value Rse of the measured AC resistance of the inductor element under inspection using the adjustment value Rtng2 calculated in step S12 (step S134). Specifically, the estimation unit 23 inputs the measured DC resistance values ​​Rdc4 and Rdc2 obtained in steps S131 and S132 into the trained model 35 using the method described above, and outputs the value output from the trained model 35 plus the adjustment value Rtng2 calculated in step S12 as the estimated value Rse of the measured AC resistance (=f(Rdc4,Rdc2)+g(Ls)+Rtng2).

[0185] Alternatively, after step S133, the DC resistance of the inductor element under test may be measured using the two-terminal method, and the measured value of DC resistance Rdc2 may be obtained again. In this case, the estimation unit 23 may compare the measured value of DC resistance Rdc2 measured in step S132 with the measured value of DC resistance Rdc2 measured again after step S133, and use the smaller of the measured values ​​of DC resistance Rdc2 to calculate the estimated value of AC resistance Rse in step S134.

[0186] Next, the correction unit 24 determines whether the difference |Rse-Rs| between the estimated value Rse of the measured AC resistance calculated in step S134 and the measured value Rs of the AC resistance obtained in step S133 is smaller than the threshold Rth (step S135).

[0187] If the difference |Rse-Rs| is smaller than the threshold Rth (step S135: YES), the correction unit 24 calculates the resistance component Rc caused by the two-terminal measurement system using model f(Rdc4, Rdc2) (step S136). Specifically, the correction unit 24 obtains the resistance component Rc caused by the two-terminal measurement system by inputting the measured values ​​of DC resistance Rdc4 and Rdc2 obtained in steps S131 and S132 into model f(Rdc4, Rdc2).

[0188] Next, the correction unit 24 corrects the measured value Rs of AC resistance obtained in step S133 using the resistance component Rc caused by the two-terminal measurement system calculated in step S136, and outputs the corrected value as the AC resistance value Rsr (step S137). Specifically, the correction unit 24 subtracts the resistance component Rc caused by the two-terminal measurement system calculated in step S136 from the measured value Rs of AC resistance obtained in step S133 and outputs the value obtained as the AC resistance value Rsr (=Rs-Rc).

[0189] On the other hand, if the difference |Rse-Rs| is greater than the threshold Rth (step S135: NO), the correction unit 24 outputs the measured value Rs of the AC resistance obtained in step S133 as the AC resistance value Rsr (=Rs) without performing any correction processing (step S138).

[0190] Next, the determination unit 25 calculates the Q value of the inductor element to be inspected based on the AC resistance value Rsr output from the correction unit 24 in step S137 or step S138 and the inductance measurement value Ls obtained in step S13 (step S139). After that, the determination unit 25 determines whether the inductor element to be inspected is good or bad based on the Q value calculated in step S19 (step S140). Inductor elements determined to be good are transported by the transport mechanism 15 and packaged.

[0191] Furthermore, the test program used to cause the computer (information processing device) acting as the data processing control device 10 to execute each of the above steps (S131 to S140) may be distributed via a network, or it may be written to a computer-readable storage medium such as a memory card and distributed therein.

[0192] In the inspection system 1 according to Embodiment 1, the trained model generation device 3 generates a trained model 35, which includes a model f(Rdc4, Rdc2) for calculating a first resistance Rc, which is a resistance component caused by the two-terminal measurement system, and a model g(Ls) for calculating a second resistance Rs0, which is a resistance component caused by the DUT (inductor element), by machine learning multiple training measurement data 34_1 to 34_n, which are generated by labeling the measured value Rdc4 of the DC resistance of the DUT measured by the four-terminal method, the measured value Rdc2 of the DC resistance of the DUT measured by the two-terminal method, the measured value Ls of the inductance of the DUT measured by the two-terminal method, and the measured value Rs of the AC resistance of the DUT measured by the two-terminal method.

[0193] According to this, even if the relationship between the measured DC resistance values ​​Rdc4 and Rdc2 of the inductor element and the measured AC resistance value Rs is nonlinear, a trained model 35 (function) can be obtained that appropriately represents the relationship between the measured DC resistance values ​​Rdc4 and Rdc2, the measured inductance value Ls, and the measured AC resistance value Rs.

[0194] Furthermore, in the trained model 35, model f(Rdc4,Rdc2) is a regression model in which the measured values ​​of DC resistance Rdc4 and Rdc2 are explanatory variables and the value of the resistance component due to the two-terminal measurement system is the dependent variable, and model g(Ls) is a regression model in which the measured value of inductance Ls is the explanatory variable and the value of the resistance component due to the DUT is the dependent variable.

[0195] According to this, the trained model 35 can be represented by a simpler function, thus avoiding the black-boxing of the trained model 35 that is a concern in machine learning. Furthermore, in the regression model (g(Ls)) that represents the value of the resistance component due to the DUT, by using the measured value of inductance Ls as the explanatory variable, it becomes possible to estimate the resistance component due to the DUT with higher accuracy.

[0196] Furthermore, as described above, the trained model generator 3 calculates an adjustment value Rtng1 corresponding to the offset ΔRs (see Figure 3), and performs machine learning on the model f(Rdc4,Rdc2) using the adjustment value Rtng1. That is, the trained model generator 3 calculates an adjustment value Rtng1 corresponding to the offset ΔRs according to the difference between the reference AC resistance value and the estimated value of the second resistance Rs0 of the AC resistance Rs, which is calculated by inputting the measured value Ls of inductance included in at least one training measurement data into the model g(Ls). Then, in the model (f(Rdc4,Rdc2)+g(Ls)+Rtng1), which is the sum of the model f(Rdc4,Rdc2), the model g(Ls), and the adjustment value Rtng1, the trained model generator 3 adjusts the parameters of the model f(Rdc4,Rdc2) while keeping the parameters of the model g(Ls) fixed.

[0197] According to this, even if a difference (offset ΔRs) occurs between the estimated value of the second resistance Rs0 of the AC resistance Rs calculated by inputting measurement data into model g(Ls) and the defined value Rs of the reference device, a trained model 35 can be generated that cancels out that difference. This makes it possible to generate a trained model with higher estimation accuracy.

[0198] Furthermore, by calculating the adjustment value Rtng1 using the average value Rs_av1 of the AC resistance measurement Rs included in the training measurement data 34 of multiple DUTs where the difference between the measured DC resistance Rdc4 and the measured DC resistance Rdc2 is less than the threshold Rdth, it becomes possible to further reduce the error of the model g(Ls).

[0199] Furthermore, the deviation of the estimated AC resistance Rse from the defined value of the reference device as described above may also occur during inspection by the inspection device 2. Therefore, the inspection device 2 according to Embodiment 1 calculates an adjustment value Rtng2 according to the difference between the reference AC resistance value and the estimated value of the second resistance Rs0 of the AC resistance Rs, which is calculated by inputting the measured value Ls of inductance included in at least one measurement data 50 into the model g(Ls). The value calculated by the trained model 35 is adjusted based on the adjustment value Rtng2 and output as the estimated value Rse of the measured AC resistance. According to this, even if the above-mentioned offset ΔRs occurs due to differences in the measurement environment, etc., when measuring the training measurement data 34 used for machine learning of the trained model 35 and when performing the calibration process of the inspection device 2, it is possible to calculate the estimated value Rse of AC resistance in a way that cancels out that difference. This makes it possible to improve the reliability of inspection of electronic components.

[0200] Furthermore, by calculating the adjustment value Rtng2 using the average value Rs_av2 of the AC resistance measurement Rs included in the measurement data 50 of multiple DUTs where the difference between the measured DC resistance Rdc4 and the measured DC resistance Rdc2 is less than the threshold Rdth, it becomes possible to further reduce the error of the model g(Ls).

[0201] Furthermore, by calculating the adjustment value Rtng2 using the defined value of the AC resistance of the reference device in the calibration process as the reference AC resistance value, it becomes possible to correct the contact resistance from the first chip being inspected during inspection.

[0202] <Embodiment 2> Figure 10 shows an example of the configuration of the trained model generation device 3A in the inspection system 1A according to Embodiment 2.

[0203] The inspection system 1A according to Embodiment 2 differs from the inspection system 1 according to Embodiment 1 in that, in the regression model (second model) representing the value Rs0 of the resistance component caused by the DUT as the second model, the DC resistance Rdc4 measured by the four-terminal method is used as the explanatory variable instead of the measured value Ls of inductance, but in other respects it is the same as the inspection system 1 according to Embodiment 1.

[0204] In the trained model generation device 3A, the training measurement data acquisition unit 31A, similar to the training measurement data acquisition unit 31 according to Embodiment 1, acquires a data pair for each inspected inductor element, including data 41 of the measured DC resistance Rdc4 of the inductor element measured by the four-terminal method, data 42 of the measured DC resistance Rdc2 of the inductor element measured by the two-terminal method, and data 43 of the measured AC resistance Rs of the inductor element measured by the two-terminal method, via wireless or wired communication (not shown) or a storage medium such as a memory card.

[0205] The learning measurement data acquisition unit 31A generates a single learning measurement data 34A by associating the measured value of AC resistance Rs included in the acquired data pair with the measured values ​​of DC resistance Rdc4 and Rdc2 included in the data pair, using the measured value as the correct answer. The learning measurement data acquisition unit 31 generates learning measurement data 34A_1 to 34A_n for each measurement result of the inspected inductor element and stores them in the storage unit 33.

[0206] The trained model generation unit 32A generates a trained model 35 (Rse=f(Rdc4,Rdc2)+g(Rdc4)) by individually training the first model (Rc=f(Rdc4,Rdc2)) and the second model (Rs0=g(Rdc4)), similar to the trained model generation unit 32 in Embodiment 1. That is, the trained model generation unit 32A separately executes the process of adjusting the parameters of the first model (Rc=f(Rdc4,Rdc2)) using machine learning and the process of adjusting the parameters of the second model (Rs0=g(Rdc4)) using machine learning. The second model (Rs0=g(Rdc4)) is also referred to as "model g(Rdc4)".

[0207] The learned model generation unit 32A generates a model g(Rdc4) by performing machine learning on learning measurement data 34 with small variations in the measured values of the DC resistance (variations in the contact resistance), and generates f(Rdc4, Rdc2) by performing machine learning on learning measurement data 34 with large variations in the measured values of the DC resistance.

[0208] Specifically, the second model generation unit 322A of the learned model generation unit 32A generates a model (Rs = g(Rdc4)) by performing machine learning on learning measurement data 34 where the difference between the measured value Rdc4 of the DC resistance and the measured value Rdc2 of the DC resistance is smaller than the threshold value Rdth (|Rdc2 - Rdc4| < Rdth). Here, the threshold value Rdth is preferably a value as close to "0" as possible.

[0209] For example, among the learning measurement data 34 measured manually using a fixture on a desktop without using the inspection device 2, the second model generation unit 322A generates a model (Rs = g(Rdc4)) using the measured value Rdc4 of the DC resistance and the measured value Rs of the AC resistance of the measurement data where the difference between the measured value Rdc4 of the DC resistance and the measured value Rdc2 of the DC resistance is smaller than the threshold value Rdth (|Rdc2 - Rdc4| < Rdth).

[0210] Similar to the learned model generation device 3 according to Embodiment 1, the learned model generation device 3A calculates an adjustment value Rtng1 by the adjustment value calculation unit 323A, and the first model generation unit 321 performs machine learning on the model f(Rdc4, Rdc2) using the adjustment value Rtng1.

[0211] Specifically, the adjustment value calculation unit 323A calculates an adjustment value Rtng1 according to the difference between the reference value of the AC resistance and the estimated value of the second resistance value Rs0 calculated by inputting the measured value Rdc4 of the DC resistance included in at least one piece of learning measurement data 34 into the model g(Rdc4).

[0212] Here, the reference value of the AC resistance may be the average value Rs_av1 of the measured values Rs of the AC resistance included in the learning measurement data 34 of a plurality of DUTs (|Rdc2−Rdc4|<Rdth) where the difference between the measured value Rdc4 of the DC resistance and the measured value Rdc2 of the DC resistance is smaller than the threshold value Rdth, similar to Embodiment 1. In this case, the adjustment value calculation unit 323A calculates the estimated value g(Rdc4_av1) of the second resistance value Rs0 by inputting the average value Rdc4_av1 of the measured values Rdc4 of the DC resistance included in the learning measurement data 34 of a plurality of DUTs (|Rdc2−Rdc4|<Rdth) where the difference between the measured value Rdc4 of the DC resistance and the measured value Rdc2 of the DC resistance is smaller than the threshold value Rdth into the model g(Rdc4). Next, the adjustment value calculation unit 323A calculates the difference between the average value Rs_av1 of the measured values Rs of the AC resistance and the estimated value g(Rdc4_av1) of the second resistance value Rs0 as the adjustment value Rtng1 (=Rs_av1−g(Rdc4_av1)), and stores it in the storage unit 33.

[0213] Alternatively, the reference value of the AC resistance may be the defined value Rs_d1 of the AC resistance of the reference device in the calibration process, similar to Embodiment 1. In this case, the adjustment value calculation unit 323A calculates the estimated value g(Ls_av1) of the second resistance value Rs0 by inputting the average value Rdc4_av1 of the measured values Rdc4 of the DC resistance included in the learning measurement data 34 of a plurality of DUTs (|Rdc2−Rdc4|<Rdth) where the difference between the measured value Rdc4 of the DC resistance and the measured value Rdc2 of the DC resistance is smaller than the threshold value Rdth into the model g(Rdc4). Next, the adjustment value calculation unit 323A calculates the difference between the defined value Rs_d1 of the AC resistance and the estimated value of the second resistance value Rs0 as the adjustment value Rtng1 (=Rs_d1−g(Rdc4_av1)), and stores it in the storage unit 33.

[0214] The first model generation unit 321 adjusts the parameters of the model f(Rdc4, Rdc2) in a state where the parameters of the already learned model g(Rdc4) are fixed in the model represented by the sum of the model f(Rdc4, Rdc2), the model g(Rdc4), and the adjustment value Rtng1.

[0215] Specifically, the first model generation unit 321 calculates an estimated value Rse of the AC resistance measurement by inputting the measured values ​​Rdc4 of DC resistance, Rdc2 of DC resistance, and Ls of inductance included in the learning measurement data 34 (for example, learning measurement data 34 where at least |Rdc2 - Rdc4| > Rdth) into a regression model (Rse = f(Rdc4, Rdc2) + g(Rdc4) + Rtng1). Next, the first model generation unit 321 calculates the difference (error) between the calculated estimated value Rse of the AC resistance measurement and the correct value Rs of AC resistance included in the learning measurement data 34. Next, the first model generation unit 321 sequentially updates the parameters of model f(Rdc4,Rdc2) in the regression model (Rse=f(Rdc4,Rdc2)+g(Rdc4)+Rtng1) to reduce the calculated error, for example by backpropagation, generating learned parameters for model f(Rdc4,Rdc) and storing them in the storage unit 33. At this time, the learned parameters of model g(Rdc4) are not updated.

[0216] The trained model generation unit 32A stores a regression model (Rse=f(Rdc4,Rdc2)+g(Rdc4)) containing the trained parameters adjusted by the method described above as a trained model 35 in the storage unit 33.

[0217] Here, similar to the trained model generation device 3 according to Embodiment 1, the adjustment value Rtng1 is not included in the model (Rse=f(Rdc4,Rdc2)+g(Rdc4)) registered as the trained model 35A. In other words, the adjustment value Rtng1 is only considered when adjusting the parameters of the model f(Rdc4,Rdc2).

[0218] The overall flow of the method for generating a trained model according to Embodiment 2 is the same as the method for generating a trained model according to Embodiment 1 (steps S1 to S4). Some processing in the machine learning step (S3) according to Embodiment 2 differs from the machine learning processing according to Embodiment 1.

[0219] FIG. 11 is a flowchart showing the flow of machine learning (step S3) according to Embodiment 2.

[0220] Here, as an example, a case where the average value Rs_av2 of the measured values Rs of the AC resistance included in the learning measurement data 34 (|Rdc2 - Rdc4| < Rdth) of a plurality of DUTs in which the difference between the measured value Rdc4 of the DC resistance and the measured value Rdc2 of the DC resistance is smaller than the threshold value Rdth is used as the reference value of the AC resistance will be described.

[0221] First, the second model generation unit 322 extracts learning measurement data 34 with small variations in DC resistance from the learning measurement data 34_1 to 34_n (step S31). For example, as described above, among the learning measurement data 34 measured manually using a fixture, the learning measurement data 34 (|Rdc2 - Rdc4| < Rdth) in which the difference between the measured value Rdc4 of the DC resistance and the measured value Rdc2 of the DC resistance is smaller than the threshold value is extracted.

[0222] Next, the second model generation unit 322 generates a model g(Rdc4) using the learning measurement data 34 extracted in step S31 by the method described above (step S32A).

[0223] Next, the adjustment value calculation unit 323A calculates the average value Rs_av1 of the measured values Rs of the AC resistance by the same method as the adjustment value calculation unit 323A according to Embodiment 1 (step S33).

[0224] Next, the adjustment value calculation unit 323A calculates the average value Rdc4_av1 of the measured values Rdc4 of the DC resistance (step S34A). Specifically, the adjustment value calculation unit 323A calculates the average value Rdc4_av1 of the measured values Rdc4 of the DC resistance included in the plurality of learning measurement data 34 (|Rdc2 - Rdc4| < Rdth) used for calculating the average value Rs_av1 of the measured values Rs of the AC resistance in step S33.

[0225] Next, the adjustment value calculation unit 323A calculates the adjustment value Rtng1 (step S35A). Specifically, the adjustment value calculation unit 323A generates the adjustment value Rtng1 (=Rs_av1-g(Rdc4_av1)) by subtracting the value g(Rdc4_av1), which is obtained by inputting the average value Rdc4_av1 of the measured DC resistance Rdc4 calculated in step S34A into the model g(Rdc4) generated in step S32, from the average value Rs_av1 of the measured AC resistance Rs calculated in step S33.

[0226] Next, the first model generation unit 321A extracts the learning measurement data 34 with large variations in DC resistance from the learning measurement data 34_1 to 34_n (step S36). For example, the first model generation unit 321A extracts the learning measurement data 34 measured by the calibration inspection system 1 (for example, learning measurement data 34 where at least |Rdc2 - Rdc4| > Rdth).

[0227] Next, the first model generation unit 321A sets a model that takes into account the adjustment value Rtng1 (step S37A). Specifically, as described above, the first model generation unit 321A sets a model (Rse=f(Rdc4,Rdc2)+g(Rdc4)+Rtng1) which is the sum of the model f(Rdc4,Rdc2), the model g(Rdc4) generated in step S32, and the adjustment value Rtng1 calculated in step S35A.

[0228] Next, the first model generation unit 321A adjusts the parameters of model f(Rdc4,Rdc2) using the model (Rse=f(Rdc4,Rdc2)+g(Rdc4)+Rtng1) set in step S37A by the method described above, while fixing the parameters of the already trained model g(Ls).

[0229] Through the above processing flow, the parameters of model g(Rdc4) and model f(Rdc2,Rdc4) are individually adjusted, and the trained model 35A(Rse=f(Rdc2,Rdc4)+g(Rdc4)) is generated.

[0230] Next, the correction of the measured value Rs of the AC resistance using the learned model 35A by the inspection device 2A according to the second embodiment will be described in detail.

[0231] FIG. 12 is a diagram showing an example of the configuration of the data processing control device 10A in the inspection device 2A according to the second embodiment.

[0232] As shown in FIG. 12, the data processing control device 10A of the inspection device 2A stores the learned model 35A (Rse = f(Rdc2, Rdc4) + g(Rdc4)) in the storage unit 22. Similar to the data processing control device 10 according to the first embodiment, the data processing control device 10A generates an adjustment value Rtng2 by the adjustment value calculation unit 26A.

[0233] Specifically, the adjustment value calculation unit 26A calculates an adjustment value Rtng1 according to the difference between the reference AC resistance value and the estimated value of the second resistance value Rs0 calculated by inputting the measured value Rdc4 of the DC resistance included in at least one measurement data 50 into the model g(Rdc4).

[0234] Here, the reference AC resistance value is the average value Rs_av2 of the measured values Rs of the AC resistance included in the measurement data 50 of a plurality of DUTs (|Rdc2 - Rdc4| < Rdth) where the difference between the measured value Rdc4 of the DC resistance and the measured value Rdc2 of the DC resistance is smaller than the threshold value Rdth.

[0235] The adjustment value calculation unit 26A calculates an estimated value g(Rdc4_av2) of the second resistance value Rs0 by inputting the average value Rdc4_av2 of the measured values Rdc4 of the DC resistance included in the measurement data 50 of a plurality of DUTs (|Rdc2 - Rdc4| < Rdth) where the difference between the measured value Rdc4 of the DC resistance and the measured value Rdc2 of the DC resistance is smaller than the threshold value Rdth into the model g(Rdc4). The adjustment value calculation unit 26A calculates the difference between the average value Rs_av2 of the measured values Rs of the AC resistance and the estimated value g(Rdc4_av2) of the second resistance value Rs0 as an adjustment value Rtng2 (= Rs_av2 - g(Rdc4_av2)), and stores it in the storage unit 22.

[0236] Alternatively, the reference AC resistance value may be the defined value Rs_d2 of the AC resistance of the reference device in the calibration process of the inspection device 2. Information on the defined value Rs_d2 of the AC resistance may be stored in advance in the storage unit 22 of the data processing control device 10 after the calibration process of the inspection device 2 described above.

[0237] In this case, the adjustment value calculation unit has the average value Rdc4_av2 of the measured values Rdc4 of the DC resistance included in the measurement data 50 of a plurality of DUTs (|Rdc2−Rdc4|<Rdth) where the difference between the measured value Rdc4 of the DC resistance and the measured value Rdc2 of the DC resistance is smaller than the threshold value Rdth input into the model g(Rdc4), thereby calculating an estimated value g(Rdc4_av2) of the second resistance value Rs0. The adjustment value calculation unit calculates the difference between the defined value Rs_d2 of the AC resistance and the estimated value g(Rdc4_av2) of the second resistance value Rs0 as an adjustment value Rtng2 (=Rs_d2−g(Rdc).

[0238] [[ID=**10**]] [[ID=**11**]]The estimation unit 23A, similar to the estimation unit 23 according to the first embodiment, inputs 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 35A (function), adjusts the obtained value by the adjustment value Rtng2, and stores the adjusted value in the storage unit 22 as an estimated value Rse of the AC resistance. [[ID=**12**]] [[ID=**13**]]

[0239] [[ID=**14**]] [[ID=**15**]]For example, the estimation unit 23A stores, in the storage unit 22, as an estimated value Rse of the AC resistance, a value calculated by inputting the measured values Rdc4 and Rdc2 of the DC resistance included in the measurement data 50 into a model (f(Rdc4, Rdc2)+g(Rdc4)+Rtng2) obtained by adding the adjustment value Rtng2 to the learned model 35A. [[ID=**16**]] [[ID=**17**]]

[0240] [[ID=**18**]] [[ID=**19**]]Next, the flow of the inspection process of the DUT by the inspection device 2A according to the second embodiment will be described. [[ID=**20**]] [[ID=**21**]]

[0241] [[ID=**22**]] It should be noted that there seems to be an incomplete formula in the translation of line 7. The original text "算出し、記憶部22に記憶する。" is translated as "算出し、記憶部22に記憶する。" which is not a complete English sentence. It should be something like "calculates it and stores it in the storage unit 22." The same issue exists in the translation of line 11 and 15. I've translated it as accurately as possible based on the given rules, but these parts might need further review for a more complete and accurate translation.The overall flow of the inspection process by the inspection device 2A according to Embodiment 2 is the same as the inspection process (Steps S11 to S13) according to Embodiment 1. Some processes in the adjustment value calculation process (Step S12) according to Embodiment 2 are different from the adjustment value calculation process according to Embodiment 1.

[0242] FIG. 13 is a flowchart showing the flow of the adjustment value calculation process (Step S12) in the inspection process according to Embodiment 2.

[0243] Here, as an example, a case where the average value Rs_av2 of the measured values Rs of the AC resistance included in the measurement data 50 of a plurality of DUTs (|Rdc2 - Rdc4| < Rdth) in which the difference between the measured value Rdc4 of the DC resistance and the measured value Rdc2 of the DC resistance is smaller than the threshold value Rdth is used as the reference value of the AC resistance will be described.

[0244] In Step S12, first, the data processing control device 10A acquires the measurement data of a plurality of inductor elements to be inspected (Step S121). For example, the data processing control device 10 controls the first measurement unit 11 and the second measurement unit 12 to acquire the measurement data 50 including the measured values Rdc4, Rdc2 of the DC resistance and the measured value Rs of the AC resistance of the inductor element to be inspected.

[0245] The measured values Rdc4, Rdc2 of the DC resistance and the measured value Rs of the AC resistance acquired by the data processing control device 10A by the above method are stored in the storage unit 22 as the measurement data 50 of the inductor element to be inspected.

[0246] Next, the data processing control device 10A extracts a plurality of measurement data 50 with small variations in contact resistance from the measurement data 50 acquired in Step S121 (Step S122). Specifically, the adjustment value calculation unit 26A extracts a plurality of measurement data 50 of DUTs (|Rdc2 - Rdc4| < Rdth) in which the difference between the measured value Rdc4 of the DC resistance and the measured value Rdc2 of the DC resistance is smaller than the threshold value Rdth from the measurement data 50 stored in the storage unit 22.

[0247] Next, the adjustment value calculation unit 26A calculates the average value Rs_av2 of the measured AC resistance Rs included in the multiple measurement data 50 acquired in step S122 (step S123). The adjustment value calculation unit 26A also calculates the average value Rdc4_av2 of the DC resistance Rdc4 included in the multiple measurement data 50 acquired in step S122 (step S124A).

[0248] Next, the adjustment value calculation unit 26A calculates an estimated value g(Rdc4_av2) of the second resistance value Rs0 by inputting the average value Rdc4_av2 of the DC resistance Rdc4 calculated in step S124 into the model g(Rdc4) (step S125A).

[0249] Next, the adjustment value calculation unit 26A calculates the adjustment value Rtng2 (=Rs_av2-g(Rdc4_av2)) by subtracting the estimated value g(Rdc4_av2) of the second resistance value Rs0 calculated in step S125A from the average value Rs_av2 of the measured AC resistance Rs calculated in step S123 (step S126A). Through this processing flow, the adjustment value Rtng2 can be obtained.

[0250] Here, if the reference AC resistance value is the defined value Rs_d2 of the AC resistance of the reference device in the calibration process of the inspection device 2A, then in step S123, the adjustment value calculation unit 26A reads out the defined value Rs_d2 of the AC resistance of the reference device from the storage unit 22. Then, in step S126A, the adjustment value calculation unit 26A calculates the adjustment value Rtng2 (=Rs_d2-g(Rdc4_av2)) by subtracting the estimated value g(Rdc4_av2) of the second resistance value Rs0 calculated in step S125A from the defined value Rs_d2 of the AC resistance of the reference device.

[0251] The inspection process by the data processing control device 10A (step S13) is the same as the inspection process by the data processing control device 10 according to Embodiment 1, so its explanation will be omitted.

[0252] As described above, the learned model generation device 3A according to Embodiment 2 generates the model g(Rdc4) by performing machine learning on the learning measurement data 34 where the difference between the measured value Rdc4 of the DC resistance and the measured value Rdc2 of the DC resistance is smaller than the threshold (|Rdc2−Rdc4|<Rdth), and generates the model f(Rdc2, Rdc4) by performing machine learning on the learning measurement data 34.

[0253] According to this, similar to the learned model generation device 3 according to Embodiment 1, it is possible to obtain more accurate models g(Rdc4) and f(Rdc2, Rdc4).

[0254] Further, the learned model generation device 3A calculates an adjustment value Rtng1 corresponding to the offset ΔRs according to the difference between the reference AC resistance value and the estimated value of the second resistance value Rs0 of the AC resistance Rs calculated by inputting the measured value Rdc4 of the DC resistance included in at least one learning measurement data into the model g(Rdc4). Then, in the model (f(Rdc4, Rdc2)+g(Rdc4)+Rtng1) represented by the sum of the model f(Rdc4, Rdc2), the model g(Rdc4), and the adjustment value Rtng1, the parameters of the model f(Rdc4, Rdc2) are adjusted while the parameters of the model g(Rdc4) are fixed.

[0255] According to this, even when there is a difference (offset ΔRs) between the estimated value of the second resistance value Rs0 of the AC resistance Rs calculated by inputting the measurement data into the model g(Rdc4) and the defined value Rs of the reference device, the learned model 35A can be generated so as to cancel that difference. Thereby, a learned model with higher estimation accuracy can be generated.

[0256] Furthermore, the deviation of the estimated AC resistance Rse by the trained model 35A from the defined value of the reference device described above may similarly occur during inspection by the inspection device 2A. Therefore, the inspection device 2A according to Embodiment 2 calculates an adjustment value Rtng2 according to the difference between the reference AC resistance value and the estimated value of the second resistance value Rs0 of the AC resistance Rs, which is calculated by inputting the measured value Rdc4 of the DC resistance included in at least one measurement data 50 into model g(Rdc4). The value calculated by the trained model 35A is then adjusted based on the adjustment value Rtng2 and output as the estimated value Rse of the measured AC resistance. According to this, it is possible to improve the reliability of inspection of electronic components, similar to the inspection device 2 according to Embodiment 1.

[0257] <<Extension of the Embodiment>> Although the invention made by the present inventor has been specifically described above based on embodiments, it goes without saying that the present invention is not limited thereto and can be modified in various ways without departing from its essence.

[0258] For example, the model (trained model 35) used by the inspection device 2 for inspection only needs to be a function that shows the correspondence between the measured values ​​of DC resistance Rdc4 and Rdc2 and the measured value of AC resistance Rs, and may be a model generated by a method other than machine learning. For example, the inspection device 2 may use a model that includes a model f(Rdc4, Rdc2) whose coefficients have been adjusted by a method other than machine learning, and a model g(Ls) (or model g(Rdc4)), and perform inspection of the inductor element in the same manner as described above.

[0259] Furthermore, in the above embodiment, the example given was that the inspection device 2(2A) is an integrated device comprising components such as the data processing control device 10(10A), the first measurement unit 11, the second measurement unit 12, the operation unit 13, the output unit 14, and the transport mechanism 15. However, some components constituting the inspection device 2(2A) may be configured separately from other components. For example, the data processing control device 10(10A), the operation unit 13, and the output unit 14 may be implemented by a first device (for example, an information processing device such as a PC), while the first measurement unit 11, the second measurement unit 12, and the transport mechanism 15 may be implemented by a 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.

[0260] The flowchart described above is merely an example illustrating the operation, and is not limited to it. In other words, the steps shown in each diagram of the flowchart are specific examples and are not limited to this flow. For example, the order of some processes may be changed, other processes may be inserted between each process, or some processes may be performed in parallel. [Explanation of Symbols]

[0261] 1,1A…Inspection system, 2,2A…Inspection device, 3,3A…Trained model generation device, 10,10A…Data processing control device, 11…First measurement unit, 12…Second measurement unit, 13…Operation unit, 14…Output unit, 15…Transport mechanism, 21…Data acquisition unit, 22…Storage unit, 23…Estimation unit, 24…Correction unit, 25…Determination unit, 26,26A…Adjustment value calculation unit, 31,31A…Training measurement data acquisition unit, 32,32A…Trained model generation unit, 33…Storage unit, 34,34_1~34_n,34A,34A_1~34A_n…Training measurement data, 35,35A…Trained model, 50 ...Measurement data, 321...First model generation unit, 322,322A...Second model generation unit, 323,323A...Adjustment value calculation unit, Rc...Resistance component due to the measurement system using the 2-terminal method, Rdc2...Measured value of DC resistance using the 2-terminal method, Rdc4...Measured value of DC resistance using the 4-terminal method, Rs...Measured value of AC resistance using the 2-terminal method, Rse...Estimated value of the measured value of AC resistance using the 2-terminal method, Rs0...Value of the resistance component due to the DUT (second resistance value), Rs_d1,Rs_d2...Defined value of AC resistance, Rsr...Value of AC resistance, Rtng1,Rtng2...Adjustment value, Rth,Rdth...Threshold value.

Claims

1. A data acquisition unit that acquires measurement data including a first measured value of the DC resistance of the object to be measured measured by the four-terminal method, a second measured value of the DC resistance of the object to be measured measured by the two-terminal method, a third measured value of the AC resistance of the object to be measured measured by the two-terminal method, and a fourth measured value of the inductance of the object to be measured measured by the two-terminal method. A storage unit that stores a trained model for causing a computer to function to calculate the third measurement based on the input first measurement, second measurement, and fourth measurement; An estimation unit calculates estimated values ​​of the third measurement corresponding to the first measurement, second measurement, and fourth measurement acquired by the data acquisition unit, based on the trained model stored in the storage unit. The system includes an adjustment value calculation unit that calculates an adjustment value to adjust the value calculated by the trained model, The trained model is represented by the sum of a first model, in which the first and second measured values ​​are explanatory variables and the first resistance value, which is the value of the resistance component caused by the two-terminal measurement system, is the objective variable, and a second model, in which the fourth measured value is an explanatory variable and the second resistance value, which is the value of the resistance component caused by the object being measured, is the objective variable. The adjustment value calculation unit calculates the adjustment value according to the difference between the reference AC resistance value and the estimated value of the second resistance value calculated by inputting the fourth measurement value included in at least one of the measurement data into the second model. The estimation unit adjusts the value calculated by the trained model based on the adjustment value and calculates it as the estimated value of the third measurement. Inspection device.

2. In the inspection apparatus according to claim 1, The aforementioned reference value of AC resistance is the average value of the third measurement value included in a plurality of measurement data of the object to be measured in which the difference between the first measurement value and the second measurement value is less than a threshold value. The adjustment value calculation unit calculates an estimated value of the second resistance value by inputting the average value of the fourth measurement value included in a plurality of measurement data of the object to be measured, where the difference between the first measurement value and the second measurement value is less than the threshold, into the second model, and calculates the difference between the average value of the third measurement value and the estimated value of the second resistance value as the adjustment value. Inspection device.

3. In the inspection apparatus according to claim 1, The aforementioned reference value for AC resistance is the defined value for the AC resistance of the reference device in the calibration process, The adjustment value calculation unit calculates an estimated value of the second resistance value by inputting the average value of the fourth measurement value included in the measurement data of a plurality of objects to be measured, where the difference between the first measurement value and the second measurement value is less than a threshold, into the second model, and calculates the difference between the defined value and the estimated value of the second resistance value as the adjustment value. Inspection device.

4. A data acquisition unit that acquires measurement data including a first measured value of the DC resistance of the object to be measured measured by the four-terminal method, a second measured value of the DC resistance of the object to be measured measured by the two-terminal method, and a third measured value of the AC resistance of the object to be measured measured by the two-terminal method. A storage unit that stores a trained model for causing a computer to function to calculate a third measurement based on the input first and second measurement values, An estimation unit calculates an estimated value of the third measurement corresponding to the first and second measurement obtained by the data acquisition unit, based on the trained model stored in the storage unit. The system includes an adjustment value calculation unit that calculates an adjustment value to adjust the value calculated by the trained model, The trained model is represented by the sum of a first model, in which the first measured value and the second measured value are explanatory variables and the first resistance value, which is the value of the resistance component caused by the two-terminal measurement system, is the objective variable, and a second model, in which the first measured value is the explanatory variable and the second resistance value, which is the value of the resistance component caused by the object being measured, is the objective variable. The adjustment value calculation unit calculates the adjustment value according to the difference between the reference AC resistance value and the estimated value of the second resistance value calculated by inputting the first measurement value included in at least one of the measurement data into the second model. The estimation unit adjusts the value calculated by the trained model based on the adjustment value and calculates it as the estimated value of the third measurement. Inspection device.

5. In the inspection apparatus according to claim 4, The aforementioned reference value of AC resistance is the average value of the third measurement value included in a plurality of measurement data of the object to be measured in which the difference between the first measurement value and the second measurement value is less than a threshold value. The adjustment value calculation unit calculates an estimated value of the second resistance value by inputting the average value of the first measurement values ​​included in a plurality of measurement data of the object to be measured, where the difference between the first measurement value and the second measurement value is less than the threshold, into the second model, and calculates the difference between the average value of the third measurement values ​​and the estimated value of the second resistance value as the adjustment value. Inspection device.

6. In the inspection apparatus according to claim 4, The aforementioned reference value for AC resistance is the defined value for the AC resistance of the reference device in the calibration process, The adjustment value calculation unit calculates an estimated value of the second resistance value by inputting the average value of the first measurement values ​​included in the measurement data of a plurality of objects to be measured, where the difference between the first measurement value and the second measurement value is less than a threshold, into the second model, and calculates the difference between the defined value and the estimated value of the second resistance value as the adjustment value. Inspection device.

7. In the inspection apparatus according to claim 1 or 4, The system further includes a correction unit that performs a correction process to correct the third measurement value based on the first and second measurement values ​​acquired by the data acquisition unit, according to the estimated value of the third measurement value, and outputs the corrected third measurement value as the value of the AC resistance of the object being measured. Inspection device.

8. In the inspection apparatus according to claim 7, The correction unit calculates the resistance component caused by the two-terminal measurement system according to the first model based on the first and second measurement values ​​acquired by the data acquisition unit, and as a correction process, corrects the third measurement value based on the resistance component caused by the two-terminal measurement system. Inspection device.

9. In the inspection apparatus according to claim 8, The correction unit calculates the error between the estimated value of the third measurement calculated by the estimation unit and the third measurement acquired by the data acquisition unit, and performs the correction process if the error is smaller than a predetermined threshold. Inspection device.

10. A first step of acquiring measurement data including a first measured value of the DC resistance of the object to be measured measured by the four-terminal method, a second measured value of the DC resistance of the object to be measured measured by the two-terminal method, a third measured value of the AC resistance of the object to be measured measured by the two-terminal method, and a fourth measured value of the inductance of the object to be measured measured by the two-terminal method. A second step of calculating an estimated value of the third measurement corresponding to the first measurement, second measurement, and fourth measurement obtained in the first step, based on a trained model that causes the computer to function to estimate the third measurement based on the input first measurement, second measurement, and fourth measurement; A third step involves calculating adjustment values ​​to adjust the values ​​calculated by the aforementioned trained model, A fourth step involves adjusting the estimated value of the third measurement calculated in the second step based on the adjustment value, The fifth step includes correcting the third measurement value based on the first and second measurement values ​​obtained in the first step, according to the estimated value of the third measurement value adjusted in the fourth step, and outputting the corrected third measurement value as the value of the AC resistance of the object to be measured. The trained model is represented by the sum of a first model, in which the first and second measured values ​​are explanatory variables and the first resistance value, which is the value of the resistance component caused by the two-terminal measurement system, is the objective variable, and a second model, in which the fourth measured value is an explanatory variable and the second resistance value, which is the value of the resistance component caused by the object being measured, is the objective variable. The third step includes calculating the adjustment value according to the difference between the reference AC resistance value and an estimated value of the second resistance value calculated by inputting the fourth measurement value included in at least one of the measurement data into the second model. Testing method.

11. In the inspection method described in claim 10, The aforementioned reference value of AC resistance is the average value of the third measurement value included in a plurality of measurement data of the object to be measured in which the difference between the first measurement value and the second measurement value is less than a threshold value. The third step includes the steps of calculating an estimated value of the second resistance by inputting the average value of the fourth measurement included in a plurality of measurement data of the object to be measured, where the difference between the first measurement and the second measurement is less than the threshold, into the second model, and calculating the difference between the average value of the third measurement and the estimated value of the second resistance as the adjustment value. Testing method.

12. In the inspection method described in claim 10, The aforementioned reference value for AC resistance is the defined value for the AC resistance of the reference device in the calibration process, The third step includes the steps of calculating an estimated value of the second resistance by inputting the average value of the fourth measurement values ​​included in the measurement data of a plurality of objects to be measured, where the difference between the first measurement value and the second measurement value is less than a threshold, into the second model, and calculating the difference between the defined value and the estimated value of the second resistance as the adjustment value. Testing method.

13. A first step of acquiring measurement data including a first measured value of the DC resistance of the object to be measured measured by the four-terminal method, a second measured value of the DC resistance of the object to be measured measured by the two-terminal method, and a third measured value of the AC resistance of the object to be measured measured by the two-terminal method. A second step of calculating an estimated value of the third measurement corresponding to the first and second measurements obtained in the first step, based on a trained model that causes the computer to function to estimate the third measurement based on the input first and second measurements, A third step involves calculating adjustment values ​​to adjust the values ​​calculated by the aforementioned trained model, A fourth step involves adjusting the estimated value of the third measurement calculated in the second step based on the adjustment value, The fifth step includes correcting the third measurement value based on the first and second measurement values ​​obtained in the first step, according to the estimated value of the third measurement value adjusted in the fourth step, and outputting the corrected third measurement value as the value of the AC resistance of the object to be measured. The trained model is represented by the sum of a first model, in which the first measured value and the second measured value are explanatory variables and the first resistance value, which is the value of the resistance component caused by the two-terminal measurement system, is the objective variable, and a second model, in which the first measured value is the explanatory variable and the second resistance value, which is the value of the resistance component caused by the object being measured, is the objective variable. The third step includes calculating the adjustment value according to the difference between a reference AC resistance value and an estimated second resistance value calculated by inputting the first measurement value included in at least one of the measurement data into the second model. Testing method.

14. In the inspection method described in claim 13, The aforementioned reference value of AC resistance is the average value of the third measurement value included in a plurality of measurement data of the object to be measured in which the difference between the first measurement value and the second measurement value is less than a threshold value. The third step includes the steps of: calculating an estimated value of the second resistance value by inputting the average value of the first measurement value included in a plurality of measurement data of the object to be measured, where the difference between the first measurement value and the second measurement value is less than the threshold, into the second model; and calculating the difference between the average value of the third measurement value and the estimated value of the second resistance value as the adjustment value. Testing method.

15. In the inspection method described in claim 13, The aforementioned reference value for AC resistance is the defined value for the AC resistance of the reference device in the calibration process, The third step includes the steps of calculating an estimated value of the second resistance value by inputting the average value of the first measurements included in the measurement data of a plurality of objects to be measured, where the difference between the first measurement value and the second measurement value is less than a threshold, into the second model, and calculating the difference between the defined value and the estimated value of the second resistance value as the adjustment value. Testing method.

16. Cause the computer to perform each step of the inspection method according to claim 10 or 13. Testing program.

17. A learning measurement data acquisition unit acquires learning measurement data by associating a first measured value of the DC resistance of the object to be measured measured by the four-terminal method, a second measured value of the DC resistance of the object to be measured measured by the two-terminal method, a fourth measured value of the inductance of the object to be measured measured by the two-terminal method, with a third measured value of the AC resistance of the object to be measured measured by the two-terminal method. The system includes a trained model generation unit that generates a trained model for causing a computer to function to calculate the third measurement based on input data including the first measurement, second measurement, and fourth measurement, by machine learning the aforementioned training measurement data. The trained model is represented by the sum of a first model, in which the first measurement value and the second measurement value are explanatory variables and the first resistance, which is the value of the resistance component caused by the two-terminal measurement system, is the objective variable, and a second model, in which the fourth measurement value is an explanatory variable and the second resistance value, which is the value of the resistance component caused by the object being measured, is the objective variable. The aforementioned trained model generation unit, A first model generation unit generates a first model by machine learning the first measurement value and the second measurement value included in the learning measurement data of the object to be measured, A second model generation unit generates the second model by machine learning the fourth measurement value included in the training measurement data of the object to be measured, where the difference between the first measurement value and the second measurement value is less than a threshold, It includes an adjustment value calculation unit that calculates an adjustment value corresponding to the difference between a reference AC resistance value and an estimated value of the second resistance value calculated by inputting the fourth measurement value included in at least one of the learning measurement data into the second model, The first model generation unit adjusts the parameters of the first model while fixing the parameters of the second model in a model represented by the sum of the first model, the second model, and the adjustment value. A pre-trained model generator.

18. In the trained model generation device described in claim 17, The aforementioned reference value of AC resistance is the average value of the third measurement value included in the learning measurement data of a plurality of the objects to be measured, wherein the difference between the first measurement value and the second measurement value is less than the threshold value. The adjustment value calculation unit calculates an estimated value of the second resistance value by inputting the average value of the fourth measurement value included in the learning measurement data of a plurality of the objects to be measured, where the difference between the first measurement value and the second measurement value is less than the threshold, into the second model, and calculates the difference between the average value of the third measurement value and the estimated value of the second resistance value as the adjustment value. A pre-trained model generator.

19. In the trained model generation device described in claim 17, The aforementioned reference value for AC resistance is the defined value for the AC resistance of the reference device in the calibration process, The adjustment value calculation unit calculates an estimated value of the second resistance value by inputting the average value of the fourth measurement value included in the learning measurement data of a plurality of the objects to be measured, where the difference between the first measurement value and the second measurement value is less than the threshold, into the second model, and calculates the difference between the defined value and the estimated value of the second resistance value as the adjustment value. A pre-trained model generator.

20. A learning measurement data acquisition unit acquires learning measurement data by associating a first measured value of the DC resistance of the object to be measured, measured by the four-terminal method, a second measured value of the DC resistance of the object to be measured, measured by the two-terminal method, with a third measured value of the AC resistance of the object to be measured, also measured by the two-terminal method. The system includes a trained model generation unit that generates a trained model for causing a computer to function to calculate a third measurement based on input data including the first and second measurement values ​​by machine learning the aforementioned training measurement data, The trained model is represented by the sum of a first model, in which the first measured value and the second measured value are explanatory variables and the first resistance, which is the value of the resistance component caused by the two-terminal measurement system, is the objective variable, and a second model, in which the first measured value is the explanatory variable and the second resistance value, which is the value of the resistance component caused by the object being measured, is the objective variable. The aforementioned trained model generation unit, A first model generation unit generates a first model by machine learning the first and second measurement values ​​included in the learning measurement data of the object to be measured, where the difference between the first measurement value and the second measurement value is greater than a threshold, A second model generation unit generates the second model by machine learning the first measurement value included in the training measurement data of the object to be measured, where the difference between the first measurement value and the second measurement value is less than the threshold, It includes an adjustment value calculation unit that calculates an adjustment value corresponding to the difference between a reference AC resistance value and an estimated second resistance value calculated by inputting the first measurement value included in at least one of the learning measurement data into the second model, The first model generation unit adjusts the parameters of the first model while fixing the parameters of the second model in a model represented by the sum of the first model, the second model, and the adjustment value. A pre-trained model generator.

21. In the trained model generation device described in claim 20, The aforementioned reference value of AC resistance is the average value of the third measurement value included in the learning measurement data of a plurality of the objects to be measured, wherein the difference between the first measurement value and the second measurement value is less than the threshold value. The adjustment value calculation unit calculates an estimated value of the second resistance value by inputting the average value of the first measurement values ​​included in the learning measurement data of a plurality of the objects to be measured, where the difference between the first measurement value and the second measurement value is less than the threshold, into the second model, and calculates the difference between the average value of the third measurement value and the estimated value of the second resistance value as the adjustment value. A pre-trained model generator.

22. In the trained model generation device described in claim 20, The aforementioned reference value for AC resistance is the defined value for the AC resistance of the reference device in the calibration process, The adjustment value calculation unit calculates an estimated value of the second resistance value by inputting the average value of the first measurement values ​​included in the learning measurement data of a plurality of the objects to be measured, where the difference between the first measurement value and the second measurement value is less than the threshold, into the second model, and calculates the difference between the defined value and the estimated value of the second resistance value as the adjustment value. A pre-trained model generator.

23. The first step is to acquire learning measurement data by associating a first measured value of the DC resistance of the object to be measured measured by the four-terminal method, a second measured value of the DC resistance of the object to be measured measured by the two-terminal method, a fourth measured value of the inductance of the object to be measured measured by the two-terminal method, with a third measured value of the AC resistance of the object to be measured measured by the two-terminal method. The process includes a second step of generating a trained model by machine learning the training measurement data acquired in the first step, causing the computer to function to calculate the third measurement based on input data including the first measurement, the second measurement, and the fourth measurement, The trained model is represented by the sum of a first model, in which the first and second measured values ​​are explanatory variables and the first resistance value, which is the value of the resistance component caused by the two-terminal measurement system, is the objective variable, and a second model, in which the fourth measured value is an explanatory variable and the second resistance value, which is the value of the resistance component caused by the object being measured, is the objective variable. The second step described above is: A third step is to generate the second model by machine learning the fourth measurement value included in the training measurement data of the object to be measured, where the difference between the first measurement value and the second measurement value is less than a threshold, A fourth step of calculating an adjustment value corresponding to the difference between a reference AC resistance value and an estimated value of the second resistance value calculated by inputting the fourth measurement value included in at least one of the learning measurement data into the second model, The fifth step includes generating the first model by machine learning the first and second measurement values ​​included in the training measurement data of the object to be measured, where the difference between the first measurement value and the second measurement value is greater than a threshold, The fifth step includes adjusting the parameters of the first model while fixing the parameters of the second model in a model represented by the sum of the first model, the second model, and the adjustment value. Method for generating pre-trained models.

24. In the method for generating a trained model according to claim 23, The aforementioned reference value of AC resistance is the average value of the third measurement value included in the learning measurement data of a plurality of the objects to be measured, wherein the difference between the first measurement value and the second measurement value is less than the threshold value. The fourth step includes the steps of: calculating an estimated second resistance value by inputting the average of the fourth measurements included in the learning measurement data of a plurality of objects to be measured, where the difference between the first measurement value and the second measurement value is less than the threshold, into the second model; and calculating the difference between the average of the third measurement values ​​and the estimated second resistance value as the adjustment value. Method for generating pre-trained models.

25. In the method for generating a trained model according to claim 23, The aforementioned reference value for AC resistance is the defined value for the AC resistance of the reference device in the calibration process, The fourth step includes the steps of: calculating an estimated value of the second resistance value by inputting the average value of the fourth measurement values ​​included in the learning measurement data of a plurality of objects to be measured, where the difference between the first measurement value and the second measurement value is less than the threshold, into the second model; and calculating the difference between the defined value and the estimated value of the second resistance value as the adjustment value. Method for generating pre-trained models.

26. The first step is to acquire learning measurement data by associating a first measured value of the DC resistance of the object to be measured, measured by the four-terminal method, a second measured value of the DC resistance of the object to be measured, measured by the two-terminal method, with a third measured value of the AC resistance of the object to be measured, measured by the two-terminal method. The second step includes, by machine learning the training measurement data acquired in the first step, generating a trained model that causes a computer to function to calculate a third measurement based on input data including the first and second measurement values, The trained model is represented by the sum of a first model, in which the first measured value and the second measured value are explanatory variables and the first resistance value, which is the value of the resistance component caused by the two-terminal measurement system, is the objective variable, and a second model, in which the first measured value is the explanatory variable and the second resistance value, which is the value of the resistance component caused by the object being measured, is the objective variable. The second step described above is: A fourth step is to generate the second model by machine learning the first measurement value included in the training measurement data of the object to be measured, where the difference between the first measurement value and the second measurement value is less than a threshold, A fifth step of calculating an adjustment value corresponding to the difference between a reference AC resistance value and an estimated second resistance value calculated by inputting the first measurement value included in at least one of the learning measurement data into the second model, The sixth step includes generating the first model by machine learning the first and second measurement values ​​included in the training measurement data of the object to be measured, where the difference between the first measurement value and the second measurement value is greater than a threshold, The sixth step includes adjusting the parameters of the first model while fixing the parameters of the second model in a model represented by the sum of the first model, the second model, and the adjustment value. Method for generating pre-trained models.

27. In the method for generating a trained model according to claim 26, The aforementioned reference value of AC resistance is the average value of the third measurement value included in the learning measurement data of a plurality of the objects to be measured, wherein the difference between the first measurement value and the second measurement value is less than the threshold value. The fifth step includes: calculating an estimated value of the second resistance value by inputting the average value of the first measurements included in the learning measurement data of a plurality of objects to be measured, where the difference between the first measurement value and the second measurement value is less than the threshold, into the second model; and calculating the difference between the average value of the third measurement value and the estimated value of the second resistance value as the adjustment value. Method for generating pre-trained models.

28. In the method for generating a trained model according to claim 26, The aforementioned reference value for AC resistance is the defined value for the AC resistance of the reference device in the calibration process, The fifth step includes: calculating an estimated value of the second resistance value by inputting the average value of the first measurements included in the learning measurement data of a plurality of objects to be measured, where the difference between the first measurement value and the second measurement value is less than the threshold, into the second model; and calculating the difference between the defined value and the estimated value of the second resistance value as the adjustment value. Method for generating pre-trained models.

29. The computer is made to perform each step in the trained model generation method described in claim 23 or 26. A program for generating pre-trained models.

Images