Method, superconducting wire, and control system

Abstract

To prevent yield drop of superconducting wire caused by surface abnormality formed in the middle of a production step of the superconducting wire.SOLUTION: A method of the present disclosure includes a step of acquiring surface images of a workpiece in the producing process of a superconducting wire, a step of predicting a current value distribution representing the distribution of critical current values of the superconducting wire, and a step of dividing the workpiece into at least one first portion and at least one second portion. At least one first portion is a portion predicted, in the current value distribution, to have a critical current value below a reference value at the position of that portion in the longitudinal direction. At least one second portion is a portion predicted, in the current value distribution, to have a critical current value equal to or above the reference value at the position of that portion in the longitudinal direction. The method further includes a step of cutting the workpiece so that at least one first portion is separated from the workpiece.SELECTED DRAWING: Figure 16

Description

[Technical Field]

[0001] The present disclosure relates to methods, superconducting wires, and control systems. [Background technology]

[0002] Conventionally, laminated materials manufactured by stacking multiple layers including a superconducting layer have been known. For example, Japanese Patent Application Laid-Open No. 2015-198015 (Patent Document 1) discloses an oxide superconducting thin film wire formed by stacking a metal substrate, a ceramic intermediate layer, an oxide superconducting layer, an Ag protective layer, and a Cu stabilization layer. The Cu stabilization layer is formed on the outer periphery of a laminate formed by stacking a metal substrate, a ceramic intermediate layer, an oxide superconducting layer, and an Ag protective layer in this order. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Laid-Open No. 2015-198015 [Non-patent literature]

[0004] [Non-Patent Document 1] Mark Friesen and Alex Gurevich, "Nonlinear current flow in superconductors with restricted geometries", PHYSICAL REVIEW B, VOLUME 63, 064521 Summary of the Invention [Problem to be solved by the invention]

[0005] During the manufacturing process of superconducting wires such as oxide superconducting thin film wires, anomalies may form on the surface of the workpiece, such as a substrate or laminate. Such surface anomalies may reduce the critical current value of the superconducting wire as a finished product. If surface anomalies remain, the percentage (yield) of superconducting wires having the desired critical current value may decrease.

[0006] The present disclosure has been made to solve the above-mentioned problems, and its purpose is to prevent a decrease in the yield of superconducting wire caused by surface abnormalities formed on the workpiece during the manufacturing process of superconducting wire. [Means for solving the problem]

[0007] The method disclosed herein is implemented in a manufacturing process for superconducting wire. The method includes the steps of acquiring a surface image of a workpiece during the manufacturing process, including a substrate for the superconducting wire; predicting a current value distribution representing the distribution of critical current values ​​of the superconducting wire in the longitudinal direction of the workpiece based on the surface image; and dividing the workpiece into at least one first portion and at least one second portion based on the results of the current value distribution prediction. The at least one first portion is a portion in the current value distribution where the critical current value at that position in the longitudinal direction is predicted to be less than a reference value. The at least one second portion is a portion in the current value distribution where the critical current value at that position in the longitudinal direction is predicted to be equal to or greater than a reference value. The method further includes the step of cutting the workpiece so as to separate at least one first portion from the workpiece.

[0008] The superconducting wire of the present disclosure is produced by carrying out the above-described method. The system disclosed herein includes an imaging device, a cutting device, and a control device. The imaging device captures an image of a workpiece, including a substrate for a superconducting wire, in a manufacturing process of the superconducting wire to generate a surface image of the workpiece. The cutting device is configured to cut the workpiece. The control device executes a prediction process, a division process, and cutting control. The prediction process is a process of predicting a current value distribution representing a distribution of critical current values ​​of the superconducting wire in the longitudinal direction of the workpiece based on the surface image. The division process is a process of dividing the workpiece into at least one first portion and at least one second portion based on the current value distribution. The at least one first portion is a portion where the critical current value at that position in the longitudinal direction is predicted to be less than a reference value in the current value distribution. The at least one second portion is a portion where the critical current value at that position in the longitudinal direction is predicted to be equal to or greater than a reference value in the current value distribution. The cutting control is a process of controlling the cutting device to cut the workpiece so that at least one first portion is separated from the workpiece. [Effects of the Invention]

[0009] According to the present disclosure, it is possible to prevent a decrease in the yield of superconducting wire caused by surface abnormalities formed on the workpiece during the manufacturing process of the superconducting wire. [Brief explanation of the drawings]

[0010] [Figure 1] FIG. 1 is a cross-sectional view of a superconducting wire according to the first embodiment. [Figure 2] FIG. 2 is a diagram for explaining an outline of the manufacturing process of superconducting wire 10. As shown in FIG. [Figure 3] FIG. 3 is a diagram for explaining an outline of the manufacturing process of superconducting wire 10. As shown in FIG. [Figure 4] FIG. 4 is a diagram illustrating an example of a control system according to the first embodiment. [Figure 5] FIG. 5 is a diagram illustrating data stored in the storage 196. As shown in FIG. [Figure 6]FIG. 6 is a diagram illustrating an example of a prediction processing technique. [Figure 7] FIG. 7 is a diagram illustrating an example of a prediction processing technique. [Figure 8] FIG. 8 is a diagram illustrating another example of a prediction processing technique. [Figure 9] FIG. 9 is a diagram illustrating yet another example of a prediction processing technique. [Figure 10] FIG. 10 is a diagram illustrating yet another example of a prediction processing technique. [Figure 11] FIG. 11 is a diagram for specifically explaining the grading process. [Figure 12] FIG. 12 is a diagram illustrating an example of the division process executed in the first embodiment. [Figure 13] FIG. 13 is a diagram illustrating an example of the disconnection control executed in the first embodiment. [Figure 14] FIG. 14 is a diagram for explaining another example of the division process and the disconnection control executed in the first embodiment. [Figure 15] FIG. 15 is a diagram illustrating yet another example of the division process and the disconnection control executed in the first embodiment. [Figure 16] FIG. 16 is a flowchart showing an example of processing executed by the control device 190 in the first embodiment. [Figure 17] FIG. 17 is a flowchart showing an example of the detailed procedure of S110. [Figure 18] FIG. 18 is a flowchart showing another example of the detailed procedure of S110. [Figure 19] FIG. 19 is a flowchart showing yet another example of the detailed procedure of S110. [Figure 20] FIG. 20 is a flowchart showing yet another example of the detailed procedure of S110. [Figure 21] FIG. 21 is a diagram for explaining the processing of the control device 190 in the first modification. [Figure 22]FIG. 22 is a flowchart showing an example of processing executed by the control device 190 in the first modification. [Figure 23] FIG. 23 is a diagram for explaining the control executed by the control device 190 in the second modification. [Figure 24] FIG. 24 is a flowchart showing the processing executed by the control device 190 in the second modification. [Figure 25] FIG. 25 is a diagram for explaining the processing executed by the control device 190 in the third modification. [Figure 26] FIG. 26 is a diagram illustrating an example of a control system according to the fourth modification. [Figure 27] FIG. 27 is a diagram illustrating another example of a control system according to the fourth modification. [Figure 28] FIG. 28 is a diagram showing how the length of an object in the longitudinal direction changes in the manufacturing process of a superconducting wire. [Figure 29] FIG. 29 is a flowchart illustrating an outline of a manufacturing process for a superconducting wire according to the second embodiment. [Figure 30] FIG. 30 is a flowchart illustrating an outline of a manufacturing process for a superconducting wire according to the second embodiment. [Figure 31] FIG. 31 is a flowchart illustrating an outline of a manufacturing process for a superconducting wire according to the second embodiment. [Figure 32] FIG. 32 is a diagram illustrating an example of progress information according to the second embodiment. [Figure 33] FIG. 33 is a flowchart illustrating a process executed by the control device when setting the first predetermined value. [Figure 34] FIG. 34 is a diagram illustrating how the length of an object for each superconducting wire changes in a manufacturing process for manufacturing N superconducting wires. [Figure 35] FIG. 35 is a flowchart outlining a manufacturing process for N superconducting wires in the modification of the second embodiment. [Figure 36] FIG. 36 is a diagram illustrating an example of progress information according to a modification of the second embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0011] [Description of the embodiments of the present disclosure] First, embodiments of the present disclosure will be listed and described.

[0012] (1) A method according to one embodiment of the present disclosure is carried out in a manufacturing process of a superconducting wire. The method includes the steps of acquiring a surface image of a workpiece during the manufacturing process, the workpiece including a substrate for the superconducting wire; predicting a current value distribution representing a distribution of critical current values ​​of the superconducting wire in a longitudinal direction of the workpiece based on the surface image; and dividing the workpiece into at least one defective portion and at least one good portion based on the results of the current value distribution prediction. The at least one defective portion is a portion where the critical current value at that position in the longitudinal direction is predicted to be less than a reference value in the current value distribution. The at least one good portion is a portion where the critical current value at that position in the longitudinal direction is predicted to be equal to or greater than the reference value in the current value distribution. The method further includes the step of cutting the workpiece so as to separate the at least one defective portion from the workpiece.

[0013] With the above configuration, the distribution of critical current values ​​is predicted during the manufacturing process, the workpiece is divided into at least one defective portion and at least one good portion, and the at least one defective portion is separated. As a result, only the at least one good portion remains. By using only this good portion in the next process, it is possible to almost reliably manufacture a desired superconducting wire having a critical current value equal to or greater than a reference value. In other words, the superconducting wire almost reliably has a critical current value equal to or greater than the reference value. Therefore, it is possible to prevent a decrease in the yield of superconducting wire (finished product) due to surface abnormalities formed on the workpiece during the superconducting wire manufacturing process.

[0014] (2) In the method of (1), the at least one good-quality portion may include a plurality of provisionally accepted portions having a length in the longitudinal direction that is less than the reference length. The method may further include joining the plurality of provisionally accepted portions in the longitudinal direction after cutting the workpiece, when the total length of the plurality of provisionally accepted portions in the longitudinal direction is equal to or greater than the reference length.

[0015] From a practical standpoint, each good-quality portion may be required to have a length equal to or greater than the reference length. With the above configuration, even if the at least one good-quality portion includes multiple provisionally accepted portions, each of which has a length less than the reference length, these portions can be joined together. The object produced by joining has a length equal to or greater than the reference length. Therefore, this object can be treated in the same way as a portion of the at least one good-quality portion that originally had a length equal to or greater than the reference length. As a result, practical requirements can be met, and multiple provisionally accepted portions can be effectively utilized to reduce loss of the workpiece.

[0016] (3) The method of (1) or (2) above may further include a step of marking both longitudinal ends of at least one good quality portion.

[0017] With the above configuration, the at least one good portion and the at least one defective portion are effectively distinguished by the markings. This allows an operator to easily recognize that each good portion is located between the marked portions. As a result, it is possible to easily manage the at least one good portion and the at least one defective portion in the manufacturing process.

[0018] (4) In the method of (1) or (2), the at least one good quality portion may include at least one normal quality portion and at least one high quality portion. The at least one normal quality portion is a portion in the current value distribution where a critical current value at that portion in the longitudinal direction is predicted to be less than a threshold value that is equal to or greater than a reference value. The at least one high quality portion is a portion in the current value distribution where a critical current value at that portion in the longitudinal direction is predicted to be equal to or greater than a threshold. The method may further include forming first markings on both ends in the longitudinal direction of the at least one normal quality portion and forming second markings different from the first markings on both ends in the longitudinal direction of the at least one high quality portion.

[0019] With the above configuration, the at least one good-quality portion and the at least one normal-quality portion are effectively distinguished by the markings. This allows an operator to easily recognize that each good-quality portion is located between the first markings and that each normal-quality portion is located between the second markings. As a result, management of the at least one normal-quality portion and the at least one good-quality portion in the manufacturing process can be facilitated. Furthermore, for example, if the normal-quality portion is currently sufficient as a good-quality portion, the at least one normal-quality portion can be used preferentially in the next process for a while, while the at least one good-quality portion can be stored in a storage space. Therefore, even if the required quality of the superconducting wire increases later and at least one good-quality portion is required, the increase in required quality can be appropriately addressed by removing the at least one good-quality portion from the storage space and using it in the next process.

[0020] (5) In any of the methods (1) to (4) above, the manufacturing process may include a step of polishing a substrate. The workpiece may include an unpolished substrate as the substrate before polishing. The surface image may include an image representing the surface of the unpolished substrate.

[0021] With the above configuration, at least one defective portion of the unpolished substrate (including defects as surface abnormalities of the substrate) is separated, leaving only at least one good portion of the unpolished substrate. This allows only this good portion to be used as the object to be processed in the substrate polishing process. As a result, it is possible to prevent a decrease in the yield of superconducting wire due to defects on the surface of the unpolished substrate.

[0022] (6) In any of the methods (1) to (4) above, the manufacturing process may include a process of polishing a substrate. The workpiece may include a polished substrate as the substrate after polishing. The surface image may include an image representing the surface of the polished substrate.

[0023] With the above configuration, at least one defective portion of the polished substrate (including defects on the surface of this substrate) is separated, leaving only at least one good portion of the polished substrate. This allows only this good portion to be used as the object to be processed in the next process after the substrate polishing process. As a result, it is possible to prevent a decrease in the yield of superconducting wire due to defects on the surface of the polished substrate.

[0024] (7) In any of the methods (1) to (4) above, the manufacturing process may include polishing a substrate and forming an intermediate layer on the polished substrate. The workpiece may include the polished substrate and a first laminate including the intermediate layer. The surface image may include an image representing the surface of the intermediate layer as the surface of the first laminate.

[0025] With the above configuration, at least one defective portion of the first laminate (including a formation defect as a surface abnormality of the intermediate layer) is separated, leaving only at least one good portion of the first laminate. This allows only this good portion to be used as the object to be processed in the process following the intermediate layer formation process. As a result, it is possible to prevent a decrease in the yield of superconducting wire due to surface abnormalities of the intermediate layer.

[0026] (8) In any of the methods (1) to (4) above, the manufacturing process may include polishing a substrate, forming an intermediate layer on the polished substrate, and forming a superconducting layer on the intermediate layer. The object to be processed may include a second laminate including the polished substrate, the intermediate layer, and the superconducting layer. The surface image may include an image representing the surface of the superconducting layer as the surface of the second laminate.

[0027] With the above configuration, at least one defective portion of the second laminate (including a formation defect as a surface abnormality of the superconducting layer) is separated, leaving only that at least one good portion. This allows only that good portion to be used as the object to be processed in the process following the superconducting layer formation process. As a result, it is possible to prevent a decrease in the yield of superconducting wire due to a surface abnormality of the superconducting layer.

[0028] (9) In any of the methods (1) to (4) above, the manufacturing process may include polishing a substrate, forming an intermediate layer on the polished substrate, forming a superconducting layer on the intermediate layer, and forming a protective layer on the superconducting layer. The workpiece may include a third laminate material including the polished substrate, the intermediate layer, the superconducting layer, and the protective layer. The surface image may include an image representing the surface of the protective layer as the surface of the third laminate material.

[0029] With the above configuration, at least one defective portion of the third laminate (including a formation defect as a surface abnormality of the protective layer) is separated, leaving only the at least one good portion. This allows only the good portion to be used as the object to be processed in the process following the protective layer formation process. As a result, it is possible to prevent a decrease in the yield of superconducting wire due to the surface abnormality of the protective layer.

[0030] (10) In any of the methods (1) to (4) above, the manufacturing process may include polishing a substrate, forming an intermediate layer on the polished substrate, forming a superconducting layer on the intermediate layer, forming a protective layer on the superconducting layer, and forming a stabilizing layer on the protective layer. The workpiece may include a fourth laminate including the polished substrate, the intermediate layer, the superconducting layer, the protective layer, and the stabilizing layer. The surface image may include an image representing the surface of the stabilizing layer as the surface of the fourth laminate.

[0031] With the above configuration, at least one defective portion of the fourth laminate (including a formation defect as a surface abnormality of the stabilization layer) is separated, leaving only the at least one good portion. This allows only the good portion to be used as the object to be processed in the process from the formation of the stabilization layer to the completion of the superconducting wire. As a result, it is possible to prevent a decrease in the yield of the superconducting wire due to the surface abnormality of the stabilization layer.

[0032] (11) In any of the above methods (1) to (10), the step of predicting the current value distribution may include a step of identifying the type of surface abnormality according to the surface image using a trained model that has learned the relationship between the surface image and the type of surface abnormality of the workpiece; a step of estimating the amount of reduction corresponding to the type of surface abnormality according to the identified type using a predetermined correspondence relationship between the type of surface abnormality and the amount of reduction in the critical current value; and a step of predicting the current value distribution according to the estimated amount of reduction.

[0033] (12) In any of the methods (1) to (10) above, the step of predicting the current value distribution may include the steps of recognizing an anomalous region on the surface of the superconducting layer of the superconducting wire based on the surface image, and calculating an effective width of the superconducting layer in the lateral direction of the superconducting layer at the position of the anomalous region in the longitudinal direction based on the results of the recognition of the anomalous region. The effective width is calculated by subtracting the width of the surface anomaly included in the anomalous region in the lateral direction from the width of the superconducting wire in the lateral direction. The step of predicting the current value distribution may further include the steps of estimating a critical current value at the position based on the effective width, the thickness of the superconducting layer, and the critical current density of the superconducting wire, and predicting the current value distribution based on the estimation result of the critical current value at the position.

[0034] (13) In any of the above methods (1) to (10), the step of predicting the current value distribution may include a step of inferring the electric field distribution according to the surface image using a trained model that has learned the relationship between the surface image and the electric field distribution of the superconducting layer of the superconducting wire, and a step of predicting the current value distribution according to the inferred result of the electric field distribution.

[0035] (14) In any of the above methods (1) to (10), the step of predicting the current value distribution may include a step of inferring the surface roughness distribution according to the surface image using a trained model that has learned the relationship between the surface image and the surface roughness distribution of the workpiece, and a step of predicting the current value distribution according to the inferred result of the surface roughness distribution.

[0036] (15) Any of the above methods (1) to (14) may further include a step of predicting the production volume of superconducting wire having a critical current value equal to or greater than a reference value according to the result of the above classification step, and a step of notifying the user if the predicted production volume falls below a predetermined value.

[0037] If the predicted yield falls below a predetermined value, the required length of the superconducting wire may not be achieved. In this case, it is necessary to prepare additional workpieces and manufacture additional superconducting wire to make up for the shortfall in the superconducting wire. With the above configuration, when the predicted yield of the superconducting wire falls below a predetermined value, the user is notified of this. As a result, when it is predicted during the superconducting wire manufacturing process that the yield may not achieve the required length of the superconducting wire, the user is prompted to prepare additional workpieces. As a result, preparation of the additional workpieces can be started before the superconducting wire is completed, rather than after. As a result, preparation of the additional workpieces can be completed early, and manufacturing of the additional superconducting wire using these workpieces can be started early. Therefore, the additional superconducting wire can be completed early. Therefore, the required length can be achieved early by making up for the shortfall in the superconducting wire originally manufactured using the additional superconducting wire.

[0038] (16) In the method of (15) above, the predetermined value may be equal to or greater than the required length of the superconducting wire.

[0039] If the output falls below the required length value, the output cannot reliably achieve the required length. Alternatively, even if the output has not yet fallen below the required length, the output may fall below the required length value depending on the status of the subsequent manufacturing process. With the above configuration, for example, when the predetermined value is equal to the required length value, a user is notified that the predicted output falls below the required length value. As a result, if it is predicted during the superconducting wire manufacturing process that the required length will not be reliably achieved, the user is prompted to prepare an additional workpiece. In this case, additional workpieces are not unnecessarily prepared. The predetermined value may be greater than the required length by a given value. In this case, the user is notified that the predicted output falls below a value greater than the required length by the given value. As a result, if the output may fall below the required length value depending on the status of the subsequent manufacturing process (slightly before it is determined that the required length value will not be achieved), the user is prompted to prepare an additional workpiece for backup. This allows preparation of the additional workpieces to begin earlier, which in turn allows processing of the additional workpieces to begin earlier, which in turn allows the additional superconducting wire to be completed earlier, which in turn allows the required length to be achieved earlier.

[0040] (17) The method of (16) may further include a step of reading information indicating a degree of progress in a manufacturing process of the superconducting wire from a storage unit that stores the information. The predetermined value may be determined based on the degree of progress.

[0041] With the above configuration, when it is predicted in the early or middle stages of the manufacturing process that the required length will likely not be achieved even if the current predicted value of the output has a large margin relative to the predetermined value, the user is prompted to prepare an additional workpiece as a backup. This allows the preparation of the additional workpiece to be started earlier. As a result, the additional superconducting wire can be completed earlier using this workpiece. Therefore, the required length can be achieved earlier.

[0042] (18) A superconducting wire according to an embodiment of the present disclosure is produced by carrying out any one of the methods (1) to (17) above.

[0043] (19) A control system according to an embodiment of the present disclosure includes an imaging device, a cutting device, and a control device. The imaging device captures an image of a workpiece, including a substrate for a superconducting wire, in a manufacturing process of the superconducting wire to generate a surface image of the workpiece. The cutting device is configured to cut the workpiece. The control device executes a prediction process, a division process, and cutting control. The prediction process is a process of predicting a current value distribution representing a distribution of critical current values ​​of the superconducting wire in the longitudinal direction of the workpiece based on the surface image. The division process is a process of dividing the workpiece into at least one defective portion and at least one good portion based on the current value distribution. The at least one defective portion is a portion where the critical current value at that position in the longitudinal direction is predicted to be less than a reference value in the current value distribution. The at least one good portion is a portion where the critical current value at that position in the longitudinal direction is predicted to be equal to or greater than a reference value in the current value distribution. The cutting control is a process of controlling the cutting device to cut the workpiece so as to separate at least one defective portion from the workpiece.

[0044] (20) The control system of (19) may further include a notification device that notifies a user. The control device may further execute a prediction process for predicting the volume of superconducting wire having a critical current value equal to or greater than a reference value according to the result of the classification process, and notification control for controlling the notification device to notify a user when the predicted volume falls below a predetermined value.

[0045] (21) In the control system of (20) above, the predetermined value may be equal to or greater than the required length of the superconducting wire.

[0046] (22) The control system of (21) may further include a storage unit that stores information indicating a progress level of the manufacturing process of the superconducting wire. The predetermined value may be determined based on the progress level.

[0047] [Details of the embodiments of the present disclosure] Hereinafter, embodiments of the present disclosure will be described in detail with reference to the drawings. The same or corresponding parts in the drawings will be denoted by the same reference numerals, and the description thereof will not be repeated. The embodiments and their modifications may be combined with each other as appropriate.

[0048] <First Embodiment> 1 is a cross-sectional view of a superconducting wire according to Embodiment 1. This superconducting wire is, for example, an oxide superconducting thin film wire.

[0049] Referring to FIG. 1, this cross-sectional view is a plan view of a cut surface of superconducting wire 10 taken along line K1-K2 along the Y-axis direction, viewed from the X-axis direction. The X-axis, Y-axis, and Z-axis are perpendicular to one another. The X-axis direction is the direction in which superconducting wire 10 extends and corresponds to the longitudinal direction of superconducting wire 10. The longitudinal direction may be defined as the direction from the +X-axis to the -X-axis, or may be defined as the direction from the -X-axis to the +X-axis. The Y-axis direction corresponds to the lateral direction of superconducting wire 10. The Z-axis direction is the normal direction to the top surface of superconducting wire 10. Width W is the width of superconducting wire 10 in the Y-axis direction.

[0050] Superconducting wire 10 includes substrate 11, intermediate layer 12, superconducting layer 13, protective layer 14, and stabilizing layer 15. Superconducting wire 10 is manufactured through multiple steps (described in detail below) of laminating intermediate layer 12, superconducting layer 13, protective layer 14, and stabilizing layer 15 in this order on substrate 11 in the Z-axis direction. Superconducting wire 10 may be a relatively long superconducting wire having a length that allows it to be wound up. The laminated material may be a relatively short superconducting wire, or may be a chip-like device.

[0051] Substrate 11 is a substrate for superconducting wire 10 and is, for example, a clad material composed of a stainless steel tape, a copper layer disposed on the tape, and a nickel (Ni) layer disposed on the copper layer. The copper layer and the nickel layer are crystal-oriented. Substrate 11 is not limited to this structure and may be made of Hastelloy (registered trademark). Substrate 11 is polished, for example, by a mirror roll method during the manufacturing process of superconducting wire 10. Substrate 11 before polishing is also referred to as an "unpolished substrate." Substrate 11 after polishing is also referred to as a "polished substrate."

[0052] The intermediate layer 12 is formed on the substrate 11. Specifically, the intermediate layer 12 is formed on the nickel layer of the substrate 11. Although the intermediate layer 12 is shown as a single layer in FIG. 1, it may have a three-layer structure formed by sequentially stacking, for example, a layer of yttrium oxide (YO), a layer of stabilized zirconia (YSZ), and a layer of cerium oxide (CeO). As described above, the nickel layer of the substrate 11 is crystal-oriented, and therefore the intermediate layer 12 thereon is also crystal-oriented. The intermediate layer 12 is formed, for example, by magnetron sputtering. When the substrate 11 is made of Hastelloy or the like, the crystal-oriented intermediate layer 12 is formed, for example, by IBAD (Ion Beam Assisted Deposition).

[0053] Superconducting layer 13 is formed on intermediate layer 12. Superconducting layer 13 is made of a copper oxide superconductor (REBCO:REBaCuOy) containing a rare earth element (RE). The rare earth element is, for example, yttrium (Y), praseodymium (Pr), neodymium (Nd), samarium (Sm), europium (Eu), gadolinium (Gd), holmium (Ho), ytterbium (Yb), or the like. Superconducting layer 13 may be formed, for example, by a metal organic decomposition (MOD) method or a pulsed laser deposition (PLD) method. As described above, since intermediate layer 12 is crystalline oriented, superconducting layer 13 thereon is also crystalline oriented. Specifically, the c-axes of the REBCO crystal grains constituting superconducting layer 13 are aligned along the normal direction to the top surface of superconducting wire 10. The thickness Th is the thickness of the superconducting layer 13 in the Z-axis direction.

[0054] Protective layer 14 is formed on the outer periphery of a laminate in which substrate 11, intermediate layer 12, and superconducting layer 13 are laminated in this order. Protective layer 14 is made of silver (Ag). Protective layer 14 may also be made of a silver alloy. Protective layer 14 may be formed, for example, by sputtering or plating. FIG. 2 illustrates an example in which protective layer 14 is provided so as to cover the entire periphery of a laminate including substrate 11, intermediate layer 12, and superconducting layer 13. Protective layer 14 does not need to cover the entire periphery of the laminate, but it is sufficient that it is formed at least on superconducting layer 13.

[0055] The stabilizing layer 15 is formed on the outer periphery of the protective layer 14. The stabilizing layer 15 is made of copper. The stabilizing layer 15 may also be made of a copper alloy. The stabilizing layer 15 is formed, for example, by a plating method. The thickness of the stabilizing layer 15 is, for example, thicker than the thickness of the protective layer 14.

[0056] The number of intermediate layers included in superconducting wire 10 is not limited to one, but may be two or more (for example, three to five). The number of superconducting layers included in superconducting wire 10 is also not limited to one, but may be two or more. The characteristics of superconducting wire 10 are expressed, for example, by its critical current value. The critical current value is the maximum value of the current flowing through superconducting layer 13 at the critical temperature of superconducting wire 10. The critical temperature is the upper limit of the temperature at which the superconducting state is maintained.

[0057] 2 and 3 are diagrams for explaining an outline of a manufacturing process for superconducting wire 10 according to the first embodiment. At the start of the flowchart in Fig. 2, unpolished substrate Sb1 (see Fig. 3) has already been prepared. Hereinafter, steps will be abbreviated as "S".

[0058] 2, a first inspection process is performed (S1) to inspect the surface of unpolished substrate Sb1 for defects that would be considered unacceptable surface anomalies. An unacceptable surface anomaly is a surface anomaly that would reduce the critical current value of superconducting wire 10 (superconducting layer 13) below a predetermined reference value. This will be described in more detail later.

[0059] If the above defects are present (YES in S1), a first cutting process is performed (S5). After the first cutting process is performed, or if the above defects are not present (NO in S1), a polishing step is performed to polish the unpolished substrate Sb1 (S10). This produces a polished substrate Sb2 (see FIG. 3).

[0060] After S10, a second inspection process is performed to check whether or not there are any defects corresponding to unacceptable surface abnormalities on the surface of the polished substrate Sb2 (S11).

[0061] If such defects are present (YES in S11), a second cutting process is performed (S15). After the second cutting process is performed, or if the defects are not present (NO in S11), an intermediate layer forming process is performed (S20), in which an intermediate layer 12 is formed on the polished substrate Sb2. This produces a laminate Lm1 (see FIG. 3). The laminate Lm1 includes a substrate 11 (polished substrate Sb2) and an intermediate layer 12.

[0062] After S20, a third inspection process is carried out to check whether or not there is an unacceptable surface abnormality on the surface of the laminate Lm1 (the surface of the intermediate layer 12) (S21).

[0063] If such a surface abnormality is present (YES in S21), a third cutting process is performed (S25). After the third cutting process is performed, or if the above-mentioned surface abnormality is not present (NO in S21), a superconducting layer formation step is performed (S30), in which a superconducting layer 13 is formed on the intermediate layer 12. This produces a laminated material Lm2 (see FIG. 3). The laminated material Lm2 includes a substrate 11, an intermediate layer 12, and a superconducting layer 13.

[0064] After S30, a fourth inspection process is carried out to check whether or not there is an unacceptable surface abnormality on the surface of the laminate Lm2 (the surface of the superconducting layer 13) (S31).

[0065] If such a surface abnormality is present (YES in S31), a fourth cutting process is performed (S35). After the fourth cutting process is performed, or if the above-mentioned surface abnormality is not present (NO in S31), a protective layer forming step is performed to form a protective layer 14 on the superconducting layer 13 (S40). This produces a laminated material Lm3 (see FIG. 3). The laminated material Lm3 includes a substrate 11, an intermediate layer 12, a superconducting layer 13, and a protective layer 14.

[0066] After S40, a fifth inspection process is carried out to check whether or not there is an unacceptable surface abnormality on the surface of the laminate material Lm3 (the surface of the protective layer 14) (S41).

[0067] If such a surface abnormality is present (YES in S41), a fifth cutting process is performed (S45). After the fifth cutting process is performed, or if the above-mentioned surface abnormality is not present (NO in S41), a stabilization layer formation step is performed to form a stabilization layer 15 on the protective layer 14 (S50). This produces a laminate Lm4 (see FIG. 3). The laminate Lm4 includes a substrate 11, an intermediate layer 12, a superconducting layer 13, a protective layer 14, and a stabilization layer 15.

[0068] After S50, a sixth inspection process is carried out to check whether or not there is an unacceptable surface abnormality on the surface of the laminate Lm4 (the surface of the stabilization layer 15) (S51).

[0069] If such a surface abnormality is present (YES in S51), a sixth cutting process is performed (S55). After the sixth cutting process is performed, or if the above-mentioned surface abnormality is not present (NO in S51), a predetermined final step (S60) is performed, and then superconducting wire 10 is manufactured as a finished product.

[0070] The above-mentioned first to sixth inspection processes and first to sixth cutting processes will be explained in detail later.

[0071] Fig. 4 is a diagram illustrating an example of a control system according to the first embodiment. Referring to Fig. 4, control system 100 is included in a manufacturing system for manufacturing superconducting wire 10, and includes conveying device 110, lighting device 120, camera (imaging device) 130, and marking device 140. Control system 100 further includes conveying device 150, cutting device 160, joining device 180, display device 185, input device 187, and control device 190.

[0072] The conveying device 110 includes motors 112 and 114 and pulleys 116 and 118. Each of the motors 112 and 114 and the pulleys 116 and 118 rotates about an axis extending in the Y-axis direction.

[0073] A workpiece PS is wound around the rotating shaft of motor 112. The workpiece PS is a material to be processed in the manufacturing process of superconducting wire 10. The workpiece PS is, for example, an unpolished substrate Sb1 in the polishing step (S10), a polished substrate Sb2 in the intermediate layer forming step (S20), a laminated material Lm1 in the superconducting layer forming step (S30), a laminated material Lm2 in the protective layer forming step (S40), a laminated material Lm3 in the stabilizing layer forming step (S50), or a laminated material Lm4 in the final step after completion of these steps.

[0074] A first end and a second end of the workpiece PS are fixed to the rotation shafts of the motors 112 and 114, respectively. As the motors 112 and 114 rotate, the workpiece PS passes through the pulleys 116 and 118 in this order and is wound around the motor 114.

[0075] The lighting device 120 illuminates an imaging position Pim located between the pulleys 116 and 118. While the workpiece PS is moving in the X-axis direction, the camera 130 receives light reflected from the imaging position Pim on the surface of the workpiece PS. The camera 130 captures an image of the workpiece PS using the reflected light, thereby generating a surface image of the workpiece PS. The surface image is displayed on the XY plane as an image representing the surface of the workpiece PS.

[0076] As will be described in detail later, a surface image is generated in each of the first to sixth inspection processes (S1, S11, S21, S31, S41, and S51 in FIG. 1). For example, in the first inspection process, the surface image is an image representing the surface of the unpolished substrate Sb1. In the second inspection process, the surface image is an image representing the surface of the polished substrate Sb2. In the third inspection process, the surface image is an image representing the surface of the intermediate layer 12 as the surface of the laminate Lm1. In the fourth inspection process, the surface image is an image representing the surface of the superconducting layer 13 as the surface of the laminate Lm2. In the fifth inspection process, the surface image is an image representing the surface of the protective layer 14 as the surface of the laminate Lm3. In the sixth inspection process, the surface image is an image representing the surface of the stabilizing layer 15 as the surface of the laminate Lm4. The surface image may be a color image or a grayscale image. As the workpiece PS is moved in the X-axis direction by the motors 112 and 114, the camera 130 automatically generates a surface image of the entire workpiece PS.

[0077] The marking device 140 includes containers 145a and 145b and a marker 147. The containers 145a and 145b contain a first paint and a second paint, respectively. Each of the first paint and the second paint is an erasable paint. The second paint has a different color from the first paint. The marker 147 is configured to mark the workpiece PS with a specified one of the first paint and the second paint. The marking device 140 is movable in any of the X-axis direction, the Y-axis direction, and the Z-axis direction.

[0078] The transport device 150 is, for example, a transport robot that detects the stop of the motors 112, 114, winds up the workpiece PS, and transports the workpiece PS or the like to the first storage space 155, the second storage space 157, the third storage space 159, the cutting device 160, or the joining device 180. The workpiece PS may be transported by a worker instead of the transport device 150.

[0079] The cutting device 160 is configured to cut the workpiece PS transported by the transporting device 150 along a cutting line. The cutting line refers to a line representing the trajectory of a blade (not shown) of the cutting device 160 during a cutting operation. The cutting device 160 is, for example, a high-speed rotary cutter, a gas cutter, or a laser cutter. The joining device 180 includes a robot arm configured to join various objects together using a predetermined joining method such as welding. The display device 185 displays various screens. The display device 185 functions as a notification device that notifies the user by notifying various pieces of information. The input device 187 receives input of various user operations.

[0080] The control device 190 includes a memory 192, a CPU (Central Processing Unit) 194, and a storage 196. The memory 192 includes a ROM (Read Only Memory) and a RAM (Random Access Memory) (neither of which are shown). The ROM stores programs executed by the CPU 194. The RAM functions as a working memory. The CPU 194 executes various types of arithmetic processing in accordance with the programs. The storage 196 is a non-volatile storage device such as an SSD (Solid State Drive) or an HDD (Hard Disk Drive), and stores various types of data. The control device 190 may further include a communication device that communicates with an external server (not shown).

[0081] The control device 190 controls various devices such as the conveying device 110 (motors 112, 114), the lighting device 120, the camera 130, the marking device 140, the transporting device 150, the cutting device 160, and the joining device 180. The control device 190 also functions as a processing device that executes various processes. These processes will be described in detail later.

[0082] 5 is a diagram illustrating data stored in storage 196. Referring to FIG. 5, storage 196 stores trained models 251 to 254, maps 255 and 257, and progress information 258.

[0083] The trained model 251 is generated using autoencoder technology. When the trained model 251 receives as an input image a surface image of the workpiece PS that does not include an abnormal region, it reproduces this surface image as an output image. In this case, the input image and the output image are identical, and each pixel in the differential image between these images has a zero value. On the other hand, when the trained model 251 receives as an input image a surface image that includes an abnormal region, it generates as an output image an image from which the surface abnormality has been removed. In other words, the input image and the output image are different, and some pixels in the differential image between these images that correspond to the location of the abnormal region have non-zero values. Therefore, the output image of the trained model 251 (the above-mentioned some pixels) reflects the presence or absence of an abnormal region in the surface image, as well as its position and shape. Therefore, the control device 190 can use the trained model 251 to recognize the presence or absence of an abnormal region in the surface image, as well as its position and shape.

[0084] The trained model 252 is a model that learns the relationship between the surface image and the type of surface abnormality of the workpiece PS, and is generated using deep metric learning technology. The trained model 252 may be a deep learning model that uses transfer learning. When a surface image showing a surface abnormality is input, the trained model 252 extracts at least one feature value from the surface image. The trained model 252 clusters the positions in the feature space of points that represent the at least one feature value. The control device 190 can identify the type of surface abnormality of the workpiece PS according to the results of this clustering.

[0085] Trained model 253 is a theoretical model that has learned the relationship between the surface image and the electric field distribution of superconducting layer 13 using machine learning techniques such as neural networks or deep learning. Specifically, trained model 253 is a critical state model that has learned the relationship between the type and position of surface anomalies shown in the surface image and the electric field distribution of superconducting layer 13, and is generated, for example, through an experiment using a current flow method (see Non-Patent Document 1). Control device 190 can use trained model 253 to infer the electric field distribution of superconducting layer 13 according to the surface image.

[0086] The trained model 254 is created in advance as a model that has learned the relationship between a surface image and the distribution of surface roughness of the workpiece PS. The surface roughness is represented by an index value defined by a predetermined standard (e.g., JIS B 0601:2001), and this index value is, for example, the arithmetic mean roughness Ra. The index value may also be another index value, such as the maximum height Rz. When a surface image is input, the trained model 254 outputs the distribution of surface roughness in the XY plane of the workpiece PS related to the surface image.

[0087] The map 255 represents a predetermined relationship between the type of surface abnormality of the work-piece PS and the amount of decrease in the critical current value, as will be described in detail later.

[0088] Map 257 represents a predetermined correspondence relationship between the surface roughness of the work-piece PS and the critical current value. For example, after the surface roughness is measured through an experiment, the critical current value Icv of superconducting layer 13 in the coordinate region where the surface roughness is measured is experimentally measured (evaluated). Based on these measurement results, the relationship between the surface roughness and the critical current value Icv is derived. Progress information 258 will be explained in detail later.

[0089] During the manufacturing process of superconducting wire 10 (e.g., polishing step (S10), intermediate layer forming step (S20), superconducting layer forming step (S30), protective layer forming step (S40), or stabilizing layer forming step (S50)), surface abnormalities may be formed on the workpiece PS. For example, if the workpiece PS is a substrate such as an unpolished substrate Sb1 or a polished substrate Sb2, the surface abnormality is a defect formed on the surface of the substrate. If the workpiece PS is a laminate material such as laminate material Lm1, Lm2, Lm3, or Lm4, the surface abnormality is a defect formed on the surface layer of the laminate material. Such surface abnormalities may reduce the critical current value of superconducting wire 10. If surface abnormalities remain, the proportion (yield) of superconducting wire 10 having a desired critical current value may decrease.

[0090] Therefore, the control system 100 according to the first embodiment has a configuration for dealing with such a problem. Specifically, the control device 190 executes prediction processing, grading processing, division processing, and cutting control.

[0091] The prediction process is described below. The prediction process is a process for predicting the critical current value distribution of superconducting wire 10 (finished product) based on a surface image of workpiece PS. This distribution represents the distribution of critical current values ​​of superconducting wire 10 in the X-axis direction and corresponds to, for example, distributions CD1 to CD4 (see FIGS. 7 to 10) described below. The prediction process is based on the assumption that no surface anomalies will be formed in processes subsequent to the process related to workpiece PS. For example, if workpiece PS is unpolished substrate Sb1 and the critical current value distribution is predicted based on the surface image of unpolished substrate Sb1, the prediction process is based on the assumption that no surface anomalies will be formed on the surfaces of polished substrate Sb2 and laminates Lm1 to Lm4 (the surfaces of intermediate layer 12, superconducting layer 13, protective layer 14, and stabilizing layer 15) during the manufacturing process. As described below, the prediction process is performed using a machine learning model such as trained models 251 to 254, a rule-based method, or a combination of these methods.

[0092] 6 and 7 are diagrams illustrating an example of a prediction processing technique. Referring to FIG. 6, map 255 shows a predetermined correspondence relationship between the type of surface anomaly of workpiece PS and the amount of decrease in critical current value Icv. This correspondence relationship is determined in advance by accumulating experimental data based on, for example, a dielectric method or a current-flow method. The type of surface anomaly is identified by control device 190 according to the surface image using trained models 251 and 252. The workpiece PS may be any one of unpolished substrate Sb1, polished substrate Sb2, laminate Lm1, laminate Lm2, laminate Lm3, and laminate Lm4.

[0093] For example, consider a first case in which a certain type of surface anomaly has already formed on the surface of the workpiece PS during a certain process. In the first case, the critical current value Icv (the critical current value of the completed superconducting wire 10) at the location of this surface anomaly in the X-axis direction is predicted to be reduced by d1 compared to the critical current value Icv at a location without such a surface anomaly. Similarly, in the second case, the critical current value Icv at the location of the corresponding type of surface anomaly is predicted to be reduced by d2. In the third case, the critical current value Icv at the location of the corresponding type of surface anomaly is predicted to be reduced by d3. Note that the circle, triangle, and cross symbols in the figure are merely used to illustrate actual surface anomalies, and the shapes of the actual surface anomalies may differ from these symbols.

[0094] Referring to FIG. 7, distribution CD1 represents the critical current value distribution predicted by the prediction process using map 255. Specifically, distribution CD1 represents the predicted result of the correspondence relationship between the X-coordinate position and the critical current value Icv. In this example, surface anomalies DfA to DfC are formed on the surface of the workpiece PS. The surface anomalies DfA to DfC are assumed to be the same as the surface anomalies in the first to third cases (see FIG. 6), respectively. The workpiece PS may be any one of an unpolished substrate Sb1, a polished substrate Sb2, a laminate Lm1, a laminate Lm2, a laminate Lm3, or a laminate Lm4.

[0095] For example, the critical current value Icv in the coordinate region (X coordinate: X1-X2) where the surface anomaly DfA is formed is predicted to be Ic1, which is smaller than Ic0 by d1. Ic0 is the predicted critical current value Icv in the coordinate region where there is no surface anomaly, and is determined in advance through experiments, etc. The critical current value Icv in the coordinate region (X coordinate: X3-X4) where the surface anomaly DfB is formed is predicted to be Ic2, which is smaller than Ic0 by d2. The critical current value Icv in the coordinate region (X coordinate: X5-X6) where the surface anomaly DfC is formed is predicted to be Ic3, which is smaller than Ic0 by d3.

[0096] Fig. 8 is a diagram for explaining another example of a prediction processing technique. Referring to Fig. 8, distribution CD2 represents a critical current value distribution predicted by rule-based prediction processing. In this example, the work-piece PS is a laminate material Lm2, and surface anomalies DfA to DfC are formed on the surface of the laminate material Lm2 (the surface of the superconducting layer 13).

[0097] Control device 190 predicts the critical current value distribution based on the thickness Th of superconducting layer 13 in the Z-axis direction (see FIG. 1), the critical current density Jc, and the effective width of superconducting layer 13 in the Y-axis direction. Specifically, this prediction process corresponds to a process of estimating the critical current value Icv by calculating the product of the thickness Th, the critical current density Jc, and the effective width for each X coordinate, and predicting the critical current value distribution based on the product. Critical current density Jc is the maximum value of the current flowing per unit volume of superconducting layer 13 at the critical temperature, and is determined in advance by experiment or the like. The effective width is calculated by control device 190 by subtracting the width of the surface anomaly included in the anomaly region in the Y-axis direction from the width W.

[0098] For example, the effective width in the coordinate region (X coordinate: X11-X12) containing the surface anomaly DfA is WA. The critical current value Icv in this region is predicted to be Ic11 (=WA×Th×Jc). The effective width in the coordinate region (X coordinate: X13-X14) containing the surface anomaly DfB is WB. The critical current value Icv in this region is predicted to be Ic12 (=WB×Th×Jc). The effective width in the coordinate region (X coordinate: X15-X16) containing the surface anomaly DfC is WC. The critical current value Icv in this region is predicted to be Ic13 (=WC×Th×Jc). The critical current value Icv in the coordinate region without surface anomalies of the superconducting layer 13 is predicted to be Ic10 (=W×Th×Jc). This is because the effective width in this region is equal to the width W.

[0099] Fig. 9 is a diagram for explaining yet another example of a prediction processing technique. Referring to Fig. 9, distribution CD3 represents a critical current value distribution predicted by prediction processing using trained model 253. In this example, too, the workpiece PS is laminate material Lm2, and surface anomalies DfA to DfC are formed on the surface of superconducting layer 13. Section Msc is a coordinate region defined as an evaluation section for critical current value Icv by current flow method.

[0100] The electric field distributions EfA, EfB, and EfC are inferred using trained model 253 as being generated around surface anomalies DfA to DfC, respectively, in the critical state of superconducting layer 13. The greater the electric field strength (potential gradient) in a section in the superconducting state, the greater the potential drop in that section, and therefore the greater the resistance component in that section. The greater this resistance component, the smaller the critical current value Icv tends to be. Therefore, the greater the electric field strength in a section, the smaller the critical current value Icv tends to be.

[0101] The critical current value Icv in each section Msc is estimated according to the inference results of the electric field distribution using a machine learning model that has learned the relationship between the electric field distribution and the critical current value Icv, or using a rule-based method. For example, the critical current value Icv in the coordinate region (X coordinate: X21 to X22) containing the surface anomaly DfA is estimated to be Ic21 according to the electric field distribution EfA. The critical current value Icv in the coordinate region (X coordinate: X22 to X23) containing the surface anomaly DfB is estimated to be Ic22 according to the electric field distribution EfB. The critical current value Icv in the coordinate region (X coordinate: X23 to X24) containing the surface anomaly DfC is estimated to be Ic23 according to the electric field distribution EfC. The critical current value Icv in the coordinate region without surface anomalies is estimated to be Ic20. By estimating the critical current value Icv in each section Msc as described above, the critical current value distribution can be predicted.

[0102] 10 is a diagram for explaining yet another example of a prediction processing technique. Referring to Fig. 10, a distribution CD4 represents a critical current value distribution predicted by a prediction processing based on the surface roughness of the work-piece PS.

[0103] The control device 190 uses the trained model 254 and the map 257 to estimate the critical current value Icv for each X coordinate according to the surface image. For example, the control device 190 uses the trained model 254 to estimate the surface roughness of the workpiece PS in a coordinate region having X coordinates Xa to Xc to be Ra1 according to the surface image. The control device 190 uses the map 257 to estimate that the critical current value Icv in that coordinate region is Ic31, which corresponds to Ra1. Similarly, the control device 190 estimates that the surface roughness of the workpiece PS in a coordinate region having an X coordinate equal to or greater than Xc is Ra2. The control device 190 uses the map 257 to estimate that the critical current value Icv in that coordinate region is Ic32, which corresponds to Ra2.

[0104] The above-mentioned grading process will now be described. This process corresponds to grading (ranking) each portion of the object PS in accordance with the predicted critical current value distribution.

[0105] FIG. 11 is a diagram for specifically explaining the grading process. Referring to FIG. 11, similar to the example of FIG. 7, surface anomalies DfA to DfC are formed on the surface of the workpiece PS. In this example, the control device 190 predicts that the critical current value distribution is distribution CD1 using the method described in relation to FIG. 7. The control device 190 may also predict the critical current value distribution using the method described in relation to FIG. 8, FIG. 9, or FIG. 10.

[0106] The control device 190 grades each portion of the workpiece PS in accordance with the distribution CD1 and thresholds TH1 and TH2. The thresholds TH1 and TH2 are pre-stored in the memory 192. For example, the control device 190 grades each of the portions 305, 315, and 325 of the workpiece PS to rank "A." This is because, in the distribution CD1, the critical current value Icv at the same X coordinates (Xa to X3, X4 to X5, and X6 to Xb) as the portions 305, 315, and 325 is estimated to be equal to or greater than the threshold TH1. Similarly, the control device 190 grades the portion 310 of the workpiece PS to rank "B." This is because, in the distribution CD1, the critical current value Icv at the same X coordinates (X3 to X4) as the portion 310 is estimated to be less than the threshold TH1 and greater than or equal to the threshold TH2. Similarly, the control device 190 grades the portion 320 of the workpiece PS to rank "C." This is because it is estimated that the critical current value Icv at the same X coordinate (X5 to X6) as that of portion 320 in distribution CD1 is less than threshold value TH2. Control device 190 stores information indicating the positions of portions 305 to 325 in the X-axis direction in memory 192. The number of ranks in the grading process is not limited to three, and may be two, or four or more.

[0107] The critical current value Icv is an index that indicates the quality of the performance of the superconducting wire 10, and the higher the critical current value Icv, the better the performance of the superconducting wire 10. Therefore, the critical current value Icv is required to be equal to or greater than a predetermined reference value. This reference value is, for example, a designated value that depends on the application of the superconducting wire 10 and is stored in memory 192.

[0108] The above-mentioned division process will be described below. The division process is a process for dividing the workpiece PS into at least one defective part and at least one good part. The division process is performed according to the results of the grading process (the results of the prediction of the critical current value distribution). For example, each defective part is a part where the critical current value Icv at the position of that part in the X-axis direction (X coordinate) is predicted to be less than the reference value in the critical current value distribution. Each good part is a part where the critical current value Icv at the position of that part in the X-axis direction (X coordinate) is predicted to be equal to or greater than the reference value in the critical current value distribution.

[0109] FIG. 12 is a diagram illustrating an example of the classification process executed in the first embodiment. Referring to FIG. 12, in this example, the reference value is a threshold value TH2. The control device 190 classifies the portions 305, 310, 315, and 325 having a critical current value Icv equal to or greater than the threshold value TH2 (i.e., having a rank of "A" or "B") as good-quality portions 405 and 415. On the other hand, the control device 190 classifies the portion 320 having a critical current value Icv less than the threshold value TH2 (i.e., having a rank of "C") as a defective portion 410. In this case, each of the surface anomalies DfC corresponds to the unacceptable surface anomaly described above. The control device 190 stores information indicating the positions of the good-quality portions 405 and 415 and the defective portion 410 in the X-axis direction in the memory 192. The lengths of the good-quality portions 405 and 415 in the X-axis direction are denoted as L1 and L3, respectively. The length of the defective portion 410 in the X-axis direction is denoted as L2. The control device 190 determines L1 to L3 according to the surface image, and stores information indicating these lengths in the memory 192. After the division process, the control device 190 controls the transporting device 150 to transport the workpiece PS to the cutting device 160.

[0110] The above-mentioned cutting control will be described below. The cutting control corresponds to controlling the cutting device 160 so that the cutting device 160 cuts the workpiece PS so as to separate at least one defective portion from the workpiece PS. The cutting device 160 performs a cutting operation to cut the workpiece PS in accordance with the cutting control. A series of processes consisting of the cutting control and the cutting operation is also referred to as a "cutting process." This cutting process is performed as each step of the above-mentioned first to sixth cutting processes (S5, S15, S25, S35, S45, or S55 in FIG. 2).

[0111] Fig. 13 is a diagram illustrating an example of cutting control executed in embodiment 1. Referring to Fig. 13, in this example, cutting control corresponds to controlling the cutting device 160 so that the cutting device 160 cuts the workpiece PS so as to separate the defective portion 410 from the workpiece PS. Cutting control corresponds to controlling the position of the blade of the cutting device 160 in the XYZ space so that the workpiece PS is cut along cutting lines A1 and A2, for example.

[0112] According to the prediction process, grading process, classification process, and cutting control described above, the critical current value distribution is predicted during the manufacturing process of superconducting wire 10, workpiece PS is classified into good portions 405 and 415 and defective portion 410, and defective portion 410 is separated from workpiece PS. As a result, only good portions 405 and 415 remain. Therefore, by using only good portions 405 and 415 in the next process, desired superconducting wire 10 having a critical current value Icv equal to or greater than the reference value can be manufactured with certainty. In other words, superconducting wire 10 manufactured in the first embodiment (the finished product obtained after NO in S51 or after S55 in FIG. 2 ) has a critical current value Icv equal to or greater than the reference value with certainty. Therefore, it is possible to prevent a decrease in the yield of superconducting wire 10 (finished product) due to surface anomalies formed during the manufacturing process of superconducting wire 10.

[0113] FIG. 14 is a diagram illustrating another example of the classification process and cutting control executed in the first embodiment. Referring to FIG. 14, in this example, the reference value is a threshold value TH1. The control device 190 classifies the portions 305, 315, and 325 having a critical current value Icv equal to or greater than the threshold value TH1 (i.e., having a rank of "A") as good-quality portions 505, 515, and 525, respectively. On the other hand, the control device 190 classifies the portions 310 and 320 having a critical current value Icv less than the threshold value TH1 (i.e., having a rank of "B" or "C") as defective portions 510 and 520, respectively. The lengths of the good-quality portions 505, 515, and 525 in the X-axis direction are represented as L11, L13, and L15, respectively. The lengths of the defective portions 510 and 520 in the X-axis direction are represented as L12 and L14, respectively. Control device 190 determines L11 to L15 according to the surface image and stores information indicating these lengths in memory 192. Each of L11 and L15 is equal to or greater than reference length Lref. Each of L12, L13, and L14 is less than reference length Lref. Reference length Lref is determined depending on the application of superconducting wire 10 (finished product) and is a length that can be required for superconducting wire 10 from a practical standpoint.

[0114] When superconducting wire 10 is required to have a length equal to or greater than reference length Lref, the length of the good-quality portion may also be required to be equal to or greater than reference length Lref. Therefore, control device 190 may execute cutting control so that not only defective portions 510 and 520 but also good-quality portion 515, which has a length less than reference length Lref, are cut from workpiece PS. This leaves only good-quality portions 505 and 525, each having a length equal to or greater than reference length Lref. As a result, using good-quality portion 505 or 525, a superconducting wire having a length equal to or greater than reference length Lref and a critical current value Icv equal to or greater than a reference value can be produced with certainty. This allows practical requirements to be adequately met. The above-described cutting control corresponds to, for example, controlling the position of the blade of cutting device 160 in the XYZ space so that workpiece PS is cut along cutting lines A11 to A14. Hereinafter, good-quality portions equal to or greater than reference length Lref will also be referred to as "acceptable portions." Good-quality portions shorter than reference length Lref will also be referred to as "provisionally accepted portions."

[0115] After the cutting process, the control device 190 controls the conveying device 150 to convey the good quality parts 505, 525 (acceptable parts) to the first storage space 155, the good quality part 515 (provisionally accepted part) to the second storage space 157, and the defective parts 510, 520 to the third storage space 159. These good quality parts and defective parts may be conveyed by a worker instead of by the conveying device 150. The first storage space 155 is a storage space for the acceptable parts (or the work-pieces PS themselves). The second storage space 157 is a storage space for the provisionally accepted parts. The third storage space 159 is a storage space for the defective parts.

[0116] FIG. 15 is a diagram illustrating yet another example of the division process and cutting control executed in the first embodiment. Referring to FIG. 15, the control device 190 divides the workpiece PS into good quality portions 605, 615, 625, and 635 and defective portions 610, 620, and 630 according to the surface image and the predicted critical current value distribution (different from any of the distributions CD1 to CD4). These good quality portions have critical current values ​​Icv equal to or greater than a reference value. These defective portions have critical current values ​​Icv less than the reference value. The lengths of the good quality portions 605, 615, 625, and 635 in the X-axis direction are denoted as L21, L23, L25, and L27, respectively. The lengths of the defective portions 610, 620, and 630 in the X-axis direction are denoted as L22, L24, and L26, respectively. The control device 190 determines L21 to L27 according to the surface image and stores information indicating these lengths in the memory 192. Each of L21 and L27 is equal to or greater than the reference length Lref. Therefore, the good quality portions 605 and 635 are acceptable portions. Each of L23 and L25 is less than the reference length Lref. Therefore, the good quality portions 615 and 625 are provisionally acceptable portions. Each of L22, L24, and L26 is less than the reference length Lref.

[0117] In this example, the control device 190 performs cutting control so that not only the defective portions 610, 620, and 630 but also the good portions 615 and 625 (provisionally accepted portions) are cut off from the workpiece PS. As a result, only the good portions 605 and 635 (acceptable portions) remain. Each of these portions has a length equal to or greater than the reference length Lref. Therefore, by using the good portion 605 or 635, it is possible to almost reliably manufacture a superconducting wire 10 having a length equal to or greater than the reference length Lref and a critical current value Icv equal to or greater than the reference value. The above-described cutting control corresponds to, for example, controlling the position of the blade of the cutting device 160 in the XYZ space so that the workpiece PS is cut along the cutting lines A21 to A26.

[0118] After the cutting process, the control device 190 controls the conveying device 150 to convey the good parts 605 and 635 to the first storage space 155, the good parts 615 and 625 to the second storage space 157, and the defective parts 610, 620 and 630 to the third storage space 159. These good parts and defective parts may be conveyed by an operator instead of the conveying device 150.

[0119] In the above explanation, an example in which the work-piece PS has multiple defective parts has been mainly described, but when there is only one defective part (for example, when the defective part is located only at one end of the work-piece PS), cutting control is performed so that the single defective part is cut off. In this case, only a single good part corresponding to the parts other than the single defective part remains.

[0120] 16 is a flowchart showing an example of processing executed by control device 190 in embodiment 1. This flowchart is executed in each step of the first to sixth inspection processes described above (S1, S11, S21, S31, S41, or S51 in FIG. 2) during the manufacturing process.

[0121] 16, control device 190 acquires a surface image of work-piece PS from camera 130 (S105). Control device 190 predicts the critical current value distribution of superconducting wire 10 according to the surface image (S110). The procedure of S110 will be described in detail later.

[0122] The control device 190 grades each portion of the workpiece PS according to the result of the prediction of the critical current value distribution (S115). The control device 190 determines whether or not the workpiece PS has at least one defective portion according to the result of this grading process (S120). Each defective portion is a portion having a rank (for example, rank "C") for the critical current value Icv that is less than the reference value.

[0123] If there is at least one defective portion (YES in S120), the control device 190 divides the workpiece PS into at least one good portion and at least one defective portion (S125). Thereafter, the process proceeds to a cutting process such as S5, S15, S25, S35, S45, or S55 in FIG. 2. This cutting process includes the above-mentioned cutting control and cutting operation (i.e., a step in which the cutting device 160 cuts the workpiece PS). If there is no defective portion, i.e., if the workpiece PS consists only of good portions (NO in S120), the process proceeds to the next step such as S10, S20, S30, S40, or S50 in FIG. 2.

[0124] Fig. 17 is a flowchart showing an example of the detailed procedure of S110. This procedure is based on the techniques described with reference to Fig. 6 and Fig. 7. Referring to Fig. 17, S110 includes S111 to S114. These steps will be described below.

[0125] The control device 190 determines whether or not there is an abnormal region on the surface of the workpiece PS according to the surface image using the trained model 251 (S111). If there is no abnormal region (NO in S111), the process proceeds to S114. If there is an abnormal region (YES in S111), the process proceeds to S112.

[0126] The control device 190 uses the trained model 252 to identify the type of surface abnormality of the workpiece PS according to the surface image (S112). The control device 190 uses the map 255 (see FIG. 6) to estimate the amount of decrease in the critical current value Icv corresponding to the identified type (S113). If there are multiple surface abnormalities, the control device 190 executes S112 and S113 for each surface abnormality. The control device 190 estimates the critical current value Icv for each X coordinate according to the estimated amount of decrease, and thereby predicts the critical current value distribution (S114). If there is no abnormal region (NO in S111), the control device 190 predicts the critical current value distribution assuming that the amount of decrease in the critical current value Icv is zero. After S114, the process proceeds to S115 in FIG. 16.

[0127] Fig. 18 is a flowchart showing another example of the detailed procedure of S110. This procedure is based on the technique described in relation to Fig. 8. Referring to Fig. 18, in this example, S110 includes S111a to S114a. These steps will be described below.

[0128] The control device 190 determines whether or not there is an abnormal region on the surface of the laminate Lm2 (the surface of the superconducting layer 13) according to the surface image using the trained model 251 (S111a). If there is no abnormal region (NO in S111a), the process proceeds to S114a. If there is an abnormal region (YES in S111a), the process proceeds to S111b.

[0129] Control device 190 recognizes an anomalous region of superconducting layer 13 (more specifically, its position and shape) according to the surface image (S111b). Control device 190 calculates the effective width of superconducting layer 13 in the Y-axis direction at the position of the anomalous region in the X-axis direction (X coordinate) according to the result of recognizing the anomalous region (S112a). Control device 190 estimates the critical current value Icv at the position of the anomalous region according to the calculated effective width, thickness Th, and critical current density Jc (S113a). If superconducting layer 13 has multiple anomalous regions, control device 190 executes S111b to S113a for each anomalous region. Control device 190 predicts the critical current value distribution according to the estimation result of the critical current value Icv at the position of the anomalous region (and the estimation result of the critical current value Icv at positions in regions other than the anomalous region) (S114a). If there is no abnormal region (NO in S111a), the control device 190 predicts the critical current value distribution assuming that the critical current value Icv is constant. After S114a, the process proceeds to S115 in FIG.

[0130] Fig. 19 is a flowchart showing yet another example of the detailed procedure of S110. This procedure is based on the technique described in relation to Fig. 9. Referring to Fig. 19, S110 includes S111c and S112c. These steps will be described below.

[0131] The control device 190 uses the trained model 253 to infer the electric field distribution of the superconducting layer 13 according to the surface image of the laminate Lm2 (the surface image of the superconducting layer 13) (S111c). The control device 190 estimates the critical current value Icv in each section Msc according to the inferred result of the electric field distribution, and thereby predicts the critical current value distribution (S112c). Thereafter, the process proceeds to S115 in FIG. 16.

[0132] Fig. 20 is a flowchart showing yet another example of the detailed procedure of S110. This procedure is based on the technique described in relation to Fig. 10. Referring to Fig. 20, S110 includes S111d and S112d. These steps will be described below.

[0133] The control device 190 uses the trained model 254 to infer the distribution of surface roughness of the workpiece PS in the X-axis direction according to the surface image of the workpiece PS (S111d). The control device 190 uses the map 257 to estimate the critical current value Icv for each X coordinate according to the inferred result of the surface roughness distribution, thereby predicting the critical current value distribution (S112d). Thereafter, the process proceeds to S115 in FIG. 16.

[0134] As described above, according to the first embodiment, at least one defective portion is separated from the workpiece PS during the manufacturing process of superconducting wire 10. As a result, at least one good portion remains. By using only such good portion in the next process, superconducting wire 10 can be manufactured stably with high quality. Therefore, a decrease in the yield of superconducting wire 10 due to surface abnormalities in substrate 11, intermediate layer 12, superconducting layer 13, protective layer 14, or stabilizing layer 15 can be prevented.

[0135] <First Modification of First Embodiment> 15 again, in the first embodiment, after the cutting process, each of the good quality portions 615, 625 (provisionally accepted portions) is transported to the second storage space 157. It is desirable to make effective use of these provisionally accepted portions.

[0136] FIG. 21 is a diagram for explaining the processing of the control device 190 in this modified example 1. Referring to FIG. 21, the workpiece PS is the same as that shown in FIG. 15. When the total length of multiple provisionally accepted portions in the X-axis direction is equal to or greater than the reference length Lref, the control device 190 controls the transporting device 150 to transport these provisionally accepted portions from the second storage space 157 to the joining device 180 after the cutting process. The control device 190 then performs joining control to control the joining device 180 to join these provisionally accepted portions in the X-axis direction. This causes the joining device 180 to perform a joining operation to join these provisionally accepted portions. In this example, since the total length of the good-quality portions 615, 625 in the X-axis direction (L30 = L23 + L25) is equal to or greater than the reference length Lref, these provisionally accepted portions are joined to produce a joined product 655. The above joining control is achieved by controlling the position and orientation of the robot arm of the joining device 180 in the XYZ space.

[0137] With this configuration, even if the work-piece PS includes a plurality of provisionally accepted portions such as the good quality portions 615, 625, the bonded product 655 is produced. Because the bonded product 655 has a length equal to or greater than the reference length Lref, it can be treated in the same way as accepted portions such as the good quality portions 605, 635. As a result, a practical requirement regarding the reference length Lref is satisfied, and the plurality of provisionally accepted portions can be effectively utilized to reduce loss of the work-piece PS (avoiding a situation in which these provisionally accepted portions are discarded).

[0138] 22 is a flowchart showing an example of the processing executed by the control device 190 in this modified example 1. This flowchart starts after the disconnection processing.

[0139] Referring to FIG. 22, the control device 190 reads information indicating the lengths of multiple provisionally accepted portions (e.g., good quality portions 615, 625) in the X-axis direction from the memory 192, and determines whether the sum of these lengths is equal to or greater than the reference length Lref (S205).

[0140] If the sum is less than the reference length (NO in S205), the process ends. If the sum is equal to or greater than the reference length (YES in S205), the control device 190 controls the conveying device 150 to convey a plurality of provisionally accepted portions to the joining device 180 (S210). These provisionally accepted portions are, for example, conveyed once from the cutting device 160 to the second storage space 157 and then conveyed from the second storage space 157 to the joining device 180. These provisionally accepted portions may be conveyed from the cutting device 160 to the joining device 180 without being moved to the second storage space 157. These provisionally accepted portions may be conveyed to the joining device 180 by an operator.

[0141] After S210, the control device 190 controls the joining device 180 to join these provisionally accepted portions. As a result, the joining device 180 joins these provisionally accepted portions (S215). These provisionally accepted portions may be joined by an operator. After S215, the process ends.

[0142] As described above, according to Modification 1, a plurality of provisionally accepted portions can be effectively utilized to reduce loss of the work-piece PS.

[0143] <Second Modification of First Embodiment> From the viewpoint of work efficiency, it is desirable for workers at work sites such as factories to be able to easily visually recognize the quality of each part of the workpiece PS. The control device 190 in Modification 2 has a configuration that meets such a need.

[0144] Fig. 23 is a diagram for explaining the control executed by the control device 190 in this modified example 2. Referring to Fig. 23, the workpiece PS is the same as that shown in Fig. 12. In this example, the above-mentioned division process has been completed, and the workpiece PS has been divided into a good quality portion 405, a defective portion 410, and a good quality portion 415.

[0145] The control device 190 executes first marking control to control the marking device 140 to mark both ends in the X-axis direction of each of at least one good quality portion (e.g., good quality portions 405, 415). This causes the marking device 140 to perform a first marking operation to mark both ends of each good quality portion. In this example, the control device 190 controls the marking device 140 to form marks 455, 460 at both ends of the good quality portion 405 and marks 470, 475 at both ends of the good quality portion 415. For example, after winding up of the workpiece PS by the rotation of the motors 112, 114 is once completed (after generation of the surface image is completed), the control device 190 executes the first marking control by rotating the motors 112, 114 in the reverse direction to bring the good quality portions 405, 415 closer to the marker 147 and controlling the position of the marker 147 in the XYZ space. A series of processes consisting of the first marking control and the first marking operation is also referred to as a "first marking process." In this example, the first marking process uses the first paint described above, but the second paint may also be used.

[0146] According to the first marking control, the good quality portions 405, 415 and the defective portion 410 are effectively distinguished by the first marking operation. This allows the worker to easily recognize that the good quality portion 405 is located between the marks 455, 460 and that the good quality portion 415 is located between the marks 470, 475. As a result, it is possible to easily manage the good quality portions 405, 415 and the defective portion 410 in the manufacturing process.

[0147] The above explanation mainly describes an example in which the workpiece PS has multiple good quality parts, but if there is only one good quality part (for example, the defective part is located at only one end of the workpiece PS), both ends of the single good quality part are marked.

[0148] Fig. 24 is a flowchart showing the processing executed by the control device 190 in this modified example 2. This flowchart differs from the flowchart in Fig. 16 in that S130 has been added, but is otherwise the same as the flowchart in Fig. 16.

[0149] 24, after the sorting process of S125, the control device 190 executes the first marking control described above. As a result, the marking device 140 performs the first marking operation (S130). Thereafter, the process proceeds to the cutting process.

[0150] In the above example, the first marking control and the first marking operation are performed before the cutting process, but they may also be performed after the cutting process. In this case, the control device 190 controls the transport device 150 to pick up at least one good part from the first storage space 155 and transport it to a location near the marker 147, and then controls the position of the marker 147 in the XYZ space to mark both ends of each good part.

[0151] According to this modification 2, it is possible to easily manage the good parts 405, 415 and the defective part 410. As a result, it is possible to improve the work efficiency at the work site.

[0152] <Third Modification of First Embodiment> The control device 190 in this third modification has a different configuration to meet the above-mentioned demands at the work site.

[0153] Fig. 25 is a diagram for explaining the processing executed by the control device 190 in Modification 3. Referring to Fig. 25, in this example, the work-piece PS includes an A rank portion (excellent quality portion) 705, a B rank portion (normal quality portion) 710, and a C rank portion (low quality portion) 715.

[0154] The A-rank portion 705, the B-rank portion 710, and the C-rank portion 715 are portions that have been graded as "A," "B," and "C," respectively, as a result of the grading process described above. Specifically, the A-rank portion 705 is a portion where the critical current value Icv at that portion's position in the X-axis direction (X-coordinate) is predicted to be equal to or greater than the threshold value TH1 in the critical current value distribution. The B-rank portion 710 is a portion where the critical current value Icv at that portion's position in the X-axis direction (X-coordinate) is predicted to be less than the threshold value TH1 and greater than or equal to the threshold value TH2 in the critical current value distribution. The C-rank portion 715 is a portion where the critical current value Icv at that portion's position in the X-axis direction (X-coordinate) is predicted to be less than the threshold value TH2 in the critical current value distribution. In this example, the reference value is the threshold value TH2. Therefore, the A-rank portion 705 and the B-rank portion 710 each correspond to a good-quality portion, and the C-rank portion 715 corresponds to a defective portion.

[0155] The control device 190 executes second marking control to control the marking device 140 to form first markings at both ends of the B rank portion 715 in the X-axis direction and to form second markings at both ends of the A rank portion 705 in the X-axis direction. As a result, the marking device 140 performs a second marking operation to form the first marking and the second marking at both ends of the B rank portion 715 and the A rank portion 705. In this example, the control device 190 controls the marking device 140 to form marks 770 and 775 (first markings) at both ends of the B rank portion 715 and to form marks 755 and 760 (second markings) at both ends of the A rank portion 705. In one example, the control device 190 controls the marking device 140 to form the marks 770 and 775 using the first paint and to form the marks 755 and 760 using the second paint. In this case, the marks 755, 760 differ in color from the marks 770, 775. The control device 190 may execute the second marking control so that the shapes of the marks 755, 760 differ from the shapes of the marks 770, 775. For example, after winding up of the workpiece PS by the rotation of the motors 112, 114 has once been completed (after generation of the surface image has been completed), the control device 190 executes the second marking control by rotating the motors 112, 114 in the reverse direction to bring the A rank portion 705 and the B rank portion 715 closer to the marker 147 and controlling the position of the marker 147 in the XYZ space. A series of processes consisting of the second marking control and the second marking operation is also referred to as the "second marking process."

[0156] According to the second marking control, the B-rank portion 715 and the A-rank portion 705 are effectively distinguished by the second marking operation. This allows the worker to easily recognize that (1) the B-rank portion 715 of the good quality portion is located between the marks 770 and 775, and (2) the A-rank portion 705 of the good quality portion is located between the marks 755 and 760. As a result, it is possible to easily manage the B-rank portion 715 and the A-rank portion 705 in the manufacturing process.

[0157] Furthermore, by appropriately using the A-rank portion 705 and the B-rank portion 715 depending on the required quality of the superconducting wire 10, loss of the workpiece PS can be reduced. For example, as long as the reference value is the threshold value TH2, the B-rank portion 715 is sufficient as a good-quality portion. Therefore, for a while, the B-rank portion 715 can be used preferentially in the next process while the A-rank portion 705 can be stored in the first storage space 155. Even if the required quality subsequently increases and the reference value changes from the threshold value TH2 to the threshold value TH1, requiring the A-rank portion 705 as a good-quality portion, the A-rank portion 705 can be removed from the first storage space 155 and used in the next process. As a result, an increase in the required quality can be appropriately handled. For example, consider a case where a rank "A" is required, but no A-rank portion 705 remains and only the B-rank portion 715 remains. In such a case, the B-rank portion 715 cannot be used in the next process, and therefore may be discarded as a defective portion. This results in partial loss of the workpiece PS. On the other hand, according to the third modification, the loss of the work-piece PS can be reduced by appropriately using the A-rank portion 705 and the B-rank portion 715 as described above.

[0158] The workpiece PS may include two or more A-rank portions, or may include two or more B-rank portions. In other words, at least one good portion of the workpiece PS may include at least one A-rank portion and at least one B-rank portion. In this case, the second marking control corresponds to controlling the marking device 140 to form first markings at both ends of each B-rank portion in the X-axis direction and to form second markings at both ends of each A-rank portion in the X-axis direction.

[0159] The processing procedure of control device 190 in this modified example 3 is the same as the procedure of the flowchart in Fig. 24 except that second marking control and second marking operation are executed instead of the first marking control and first marking operation at S130 of the flowchart in this example, and therefore detailed description will not be repeated.

[0160] According to Modification 3, it is possible to easily manage the A-rank portion 705 and the B-rank portion 715 at the work site. Furthermore, by appropriately using the A-rank portion 705 and the B-rank portion 715 depending on the required quality, it is possible to reduce loss of the work-piece PS.

[0161] <Fourth Modification of First Embodiment> Fig. 26 is a diagram illustrating an example of a control system according to Modification 4. Referring to Fig. 26, control system 100A differs from control system 100 (Fig. 4) in that it includes cutting device 160A instead of cutting device 160, but is otherwise the same as control system 100.

[0162] The cutting device 160A is capable of cutting the workpiece PS in accordance with commands from the control device 190, similar to the cutting device 160, but is installed at a different position from the cutting device 160. In this example, the cutting device 160A is installed above the conveying line of the conveying device 110 in the Z-axis direction, and between the camera 130 and the pulley 118 in the X-axis direction (for example, between the marking device 140 and the pulley 118).

[0163] The control device 190 may perform cutting control by controlling the cutting device 160A. For example, when the control device 190 completes generation of a surface image of the workpiece PS and stops the rotation of the motors 112, 114, the control device 190 rotates the motors 112, 114 in the reverse direction and controls the cutting device 160A (more specifically, the position of its blade) so that at least one defective portion is cut off from the workpiece PS being transported on the transport line. The control device 190 controls the transport device 150 so as to transport the at least one cut-off defective portion to the joining device 180.

[0164] Fig. 27 is a diagram illustrating another example of a control system according to Modification 4. Referring to Fig. 27, control system 100B differs from control system 100 (Fig. 4) in that it includes cutting device 160A and joining device 180A instead of cutting device 160 and joining device 180, and does not include conveying device 150, but is otherwise the same as control system 100.

[0165] The joining device 180A includes a robot arm capable of joining a plurality of provisionally accepted parts in accordance with commands from the control device 190, similar to the joining device 180, but is installed in a different position from the joining device 180. In this example, the joining device 180A is installed above the conveying line of the conveying device 110 in the Z-axis direction, and between the cutting device 160A and the pulley 118 in the X-axis direction.

[0166] The control device 190 may perform the joining control by controlling the joining device 180A. For example, after the cutting control using the cutting device 160A, the control device 190 controls the joining device 180A so as to join the multiple provisionally accepted portions that have been separated due to this control.

[0167] <Fifth Modification of First Embodiment> In the first embodiment, the control device 190 performs a classification process to classify each portion of the workpiece PS into a defective portion or a good portion according to the result of the grading process. However, the grading process is not essential. For example, the control device 190 may perform the classification process according to the predicted result of the critical current value distribution without performing the grading process. Specifically, the control device 190 may perform the classification process only according to whether the critical current value Icv estimated for each portion of the workpiece PS is equal to or greater than a reference value.

[0168] Trained models 251 to 254 and maps 255 and 257 (see FIG. 5) may be stored in the external server described above instead of or in addition to storage 196 of control device 190. Control device 190 may then transmit information indicating the surface image to the external server. In this case, the external server uses at least one of trained models 251 to 254 and maps 255 and 257 to recognize the presence or absence of an abnormal region in the surface image and the position and shape of this region, identify the type and position of a surface abnormality, infer the electric field distribution of superconducting layer 13, and infer the surface roughness distribution of the workpiece PS. The external server predicts the critical current value distribution according to the results of such processing. The external server transmits information indicating the prediction result to control device 190. Control device 190 performs the above-described cutting control according to this information.

[0169] <Embodiment 2> In the second embodiment, the control device 190 predicts the yield of the superconducting wire 10 according to the result of the sorting process on the workpiece PS, and when the predicted yield falls below a predetermined value, notifies the user of the control system 100 of this fact. The configuration and processing of the control system 100 for enabling such processing will be described below. The yield is a value indicating the length in the longitudinal direction of the superconducting wire 10 having a critical current value Icv equal to or greater than the aforementioned reference value.

[0170] Fig. 28 is a diagram showing how the length of the work-piece PS in the longitudinal direction changes in the manufacturing process of superconducting wire 10. Referring to Fig. 28, in this example, unpolished substrate Sb1 has length LS in the longitudinal direction.

[0171] Assume that unpolished substrate Sb1 is divided into defective portion 805 and good portion 807 by the division process. Control device 190 recognizes that defective portion 805 has length Lp1 in the longitudinal direction based on the surface image of unpolished substrate Sb1. Control device 190 stores the value of length Lp1 in memory 192. Value Ln1 corresponds to the difference between the length of good portion 807 in the longitudinal direction and length LS, and is equal to the value of length Lp1. Control device 190 predicts the yield of superconducting wire 10 before the polishing process based on the results of the division process for unpolished substrate Sb1. For example, control device 190 calculates the difference between length LS and value Ln1 as the predicted yield of superconducting wire 10 under the assumption that no surface abnormalities will be formed in subsequent processes (from the polishing process to the final process). The control device 190 may recognize the length of the good portion 807 in the longitudinal direction according to the surface image of the unpolished substrate Sb1, and determine the value of the recognized length as an expected value of the yield. The defective portion 805 is separated from the unpolished substrate Sb1 by a first cutting process. Thereafter, a polished substrate Sb2 is produced by a polishing process.

[0172] Assume that ground substrate Sb2 is divided into defective portion 815 and good portion 817 by the division process. Control device 190 recognizes that defective portion 815 has length Lp2 in the longitudinal direction according to the surface image of ground substrate Sb2. Control device 190 calculates cumulative value Ln2 of lengths Lp1 and Lp2 and stores the cumulative value Ln2 in memory 192. Control device 190 predicts the yield of superconducting wire 10 before the intermediate layer forming step according to the result of the division process on ground substrate Sb2. For example, control device 190 calculates the difference between the value of length LS and cumulative value Ln2 as the predicted yield of superconducting wire 10 under the assumption that no surface abnormalities will be formed in the subsequent steps (from the intermediate layer forming step to the final step). Control device 190 may recognize the length of good portion 817 in the longitudinal direction according to the surface image of ground substrate Sb2 and determine the value of the recognized length as the predicted yield. The defective portion 815 is cut off from the polished substrate Sb2 by the second cutting process, after which the laminate Lm1 is produced by the intermediate layer forming step.

[0173] Assume that laminated material Lm1 is divided into defective portion 825 and good portion 827 by the division process. Control device 190 recognizes that defective portion 825 has length Lp3 in the longitudinal direction based on the surface image of laminated material Lm1. Control device 190 calculates cumulative total Ln3 (=Ln2+Lp3) of lengths Lp1 to Lp3 and stores the total in memory 192. Control device 190 predicts the yield of superconducting wire 10 before the superconducting layer forming step based on the result of the division process for laminated material Lm1. For example, control device 190 calculates the difference between the value of length LS and cumulative total Ln3 as the predicted yield of superconducting wire 10 under the assumption that no surface abnormalities will be formed in the subsequent steps (from the superconducting layer forming step to the final step). The control device 190 may recognize the length of the good portion 827 in the longitudinal direction according to the surface image of the laminated material Lm1, and determine the value of the recognized length as an expected value of the production volume. The defective portion 825 is cut off from the laminated material Lm1 by a third cutting process. Thereafter, the laminated material Lm2 is produced by the superconducting layer forming process.

[0174] The laminated material Lm2 is divided into a defective portion 835 and a good portion 837 by the division process. The control device 190 recognizes that the defective portion 835 has a length Lp4 in the longitudinal direction according to the surface image of the laminated material Lm2. The control device 190 calculates a cumulative value Ln4 (=Ln3+Lp4) of the lengths Lp1 to Lp4 and stores the cumulative value Ln4 in memory 192. The control device 190 predicts the yield of the superconducting wire 10 before the protective layer forming step according to the result of the division process for the laminated material Lm2. For example, the control device 190 calculates the difference value between the value of the length LS and the cumulative value Ln4 as the predicted value of the yield of the superconducting wire 10 under the assumption that no surface abnormalities will be formed in the subsequent steps (from the protective layer forming step to the final step). The control device 190 may recognize the length of the good portion 837 in the longitudinal direction according to the surface image of the laminated material Lm2 and determine the value of the recognized length as the predicted value of the yield. The defective portion 835 is cut off from the laminated material Lm2 by the fourth cutting process. Thereafter, the laminated material Lm3 is produced by the protective layer forming process.

[0175] The laminated material Lm3 is divided into a defective portion 845 and a good portion 847 by the division process. The control device 190 recognizes that the defective portion 845 has a length Lp5 in the longitudinal direction according to the surface image of the laminated material Lm3. The control device 190 calculates a cumulative value Ln5 (=Ln4+Lp5) of the lengths Lp1 to Lp5 and stores the cumulative value Ln5 in memory 192. The control device 190 predicts the yield of the superconducting wire 10 before the stabilization layer forming step according to the result of the division process for the laminated material Lm3. For example, the control device 190 calculates the difference value between the value of the length LS and the cumulative value Ln5 as the predicted value of the yield of the superconducting wire 10 under the assumption that no surface abnormalities will be formed in the subsequent steps (from the stabilization layer forming step to the final step). The control device 190 may recognize the length of the good portion 847 in the longitudinal direction according to the surface image of the laminated material Lm3 and determine the value of the recognized length as the predicted value of the yield. The defective portion 845 is cut off from the laminated material Lm3 by the fifth cutting process, after which the laminated material Lm4 is produced by the stabilizing layer forming process.

[0176] The laminated material Lm4 is assumed to be divided into a defective portion 855 and a good portion 857 by the division process. The control device 190 recognizes that the defective portion 855 has a length Lp6 in the longitudinal direction according to the surface image of the laminated material Lm4. The control device 190 calculates a cumulative value Ln6 (=Ln5+Lp6) of lengths Lp1 to Lp6 and stores the cumulative value Ln6 in memory 192. The control device 190 predicts the yield of the superconducting wire 10 before the final process according to the result of the division process for the laminated material Lm4. For example, the control device 190 calculates the difference value between the value of the length LS and the cumulative value Ln6 as the predicted value of the yield of the superconducting wire 10 under the assumption that no surface abnormalities will be formed in the subsequent final process. The control device 190 may recognize the length of the good portion 857 in the longitudinal direction according to the surface image of the laminated material Lm4 and determine the value of the recognized length as the predicted value of the yield. The defective portion 855 is cut off from the laminated material Lm4 by the sixth cutting process. Thereafter, the superconducting wire 10 is manufactured by the final process. In this example, it is assumed that no defective portion due to surface abnormalities occurs in the final process. Hereinafter, the cumulative value of the lengths Lp1 to Lpx recognized up to the x-th processing step among the six processing steps from the polishing step to the final process is also referred to as the "cumulative value Lnx" (1≦x≦6). For example, when x=1, the cumulative value Lnx corresponds to the value Ln1. When x=2, the cumulative value Lnx corresponds to the cumulative value Ln2. When predicting the yield of the superconducting wire 10 after the x-th processing step is performed, the control device 190 calculates the difference value between the value of the length LS and the cumulative value Lnx as the predicted value of the yield of the superconducting wire 10 under the assumption that no surface abnormalities will be formed in the subsequent steps.

[0177] The yield of superconducting wire 10 is the value of length LA in the longitudinal direction. If the yield of superconducting wire 10 is less than a predetermined required length of superconducting wire 10, there is a possibility that superconducting wire 10 will have few practical uses. Therefore, it is preferable that the yield of superconducting wire 10 is equal to or greater than the required length. The required length is equal to, for example, reference length Lref (see FIG. 14 ). Information indicating the required length is pre-stored in memory 192 or storage 196. The difference between length LS and the required length corresponds to the allowable upper limit of the length of all defective portions in the longitudinal direction (e.g., all defective portions 805, 815, 825, 835, 845, and 855). This upper limit is also referred to as the “first allowable upper limit.” The cumulative total Lnx exceeding the first allowable upper limit corresponds to the predicted yield of superconducting wire 10 being lower than the required length.

[0178] During the manufacturing process, the predicted yield of superconducting wire 10 may fall below the required length. In this case, even if superconducting wire 10 (first superconducting wire) is completed without any surface abnormalities in all subsequent processes, the yield may not reliably achieve the required length. As a result, it is necessary to prepare an additional workpiece PS, which is any one of unpolished substrate Sb1, polished substrate Sb2, and laminated material Lm1, Lm2, Lm3, or Lm4, and to manufacture another superconducting wire (second superconducting wire) using the prepared workpiece PS. This superconducting wire must then be used to make up for the shortfall in superconducting wire 10 (for example, by joining this wire to superconducting wire 10).

[0179] If it is discovered after the completion of superconducting wire 10 that the volume does not satisfy the required length and preparation of additional workpieces PS is started at that time, it will be difficult to complete the second superconducting wire early. This is because preparation of additional workpieces PS can take a long lead time of several months, making it impossible to start manufacturing the second superconducting wire early.

[0180] Therefore, when the predicted volume falls below the first predetermined value, the control device 190 executes notification control to notify the user of this. The notification control is, for example, a process of controlling the display device 185 (see FIG. 4) to display a screen notifying the user as described above, but may also be executed by voice. The first predetermined value is appropriately determined in advance based on the requested length, and is, for example, the value of the requested length itself. The first predetermined value may be smaller than the value of the requested length, but is preferably equal to or greater than the value of the requested length, as described below.

[0181] For example, when the first predetermined value is equal to the value of the required length, notification control is executed when the predicted value of the output falls below the value of the required length. Therefore, when it is predicted during the manufacturing process of superconducting wire 10 that the output will not reliably achieve the value of the required length, the user is prompted to prepare an additional work-piece PS, which is any one of unpolished substrate Sb1 to laminate material Lm4. This allows preparation of the additional work-piece PS to be started before, rather than after, completion of superconducting wire 10. As a result, preparation of the additional work-piece PS is completed early, thereby shortening the time interval between completion of superconducting wire 10 and start of manufacturing the second superconducting wire. Therefore, manufacturing of the second superconducting wire can be started early, thereby completing the second superconducting wire early. Therefore, the second superconducting wire can be used to supplement the shortfall in superconducting wire 10, thereby achieving the required length early. Furthermore, since preparation of the additional work-piece PS is started when it is predicted that the output will not reliably achieve the value of the required length, this additional work-piece is not prepared unnecessarily.

[0182] The first predetermined value may be greater than the required length by a given value. In this case, the user is prompted to prepare an additional workpiece PS, any one of the unpolished substrate Sb1 to laminate Lm4, some time before the predicted output falls below the required length (it is determined that the output will not achieve the required length). This allows the preparation of the additional workpiece PS to begin earlier, and therefore allows processing of the additional workpiece to begin earlier. As a result, the second superconducting wire can be completed earlier using this workpiece PS, and the required length can be achieved earlier.

[0183] In the above, each workpiece PS is divided into one defective portion and one good portion, but the number of divided defective portions or the number of divided good portions may be two or more. In this case, the predicted output is the difference between the value of the length LS and the cumulative total of the lengths recognized up to the present time for the two or more defective portions of the workpiece PS. Alternatively, the predicted output may be calculated as the cumulative total of the lengths in the longitudinal direction of the two or more good portions of the workpiece PS.

[0184] 29, 30, and 31 are flowcharts illustrating an outline of a manufacturing process for superconducting wire 10 according to the second embodiment. This flowchart differs from the flowchart of the first embodiment (see FIG. 2) in that S2a to S2c, S3, S4, S12a to S12c, S13, S14, S22a to S22c, S23, S24, S32a to S32c, S33, S34, S42a to S42c, S43, S44, S52a to S52c, S53, and S54 are added. In other respects, the flowcharts of FIGS. 29 to 31 are the same as the flowchart of FIG. 2. Therefore, detailed description will not be repeated.

[0185] 29, if the first inspection process finds that there is a defect on the surface of unpolished substrate Sb1 (YES in S1), control device 190 recognizes the length of at least one defective portion of unpolished substrate Sb1 in the longitudinal direction according to the surface image of unpolished substrate Sb1 (S2a). Control device 190 calculates a cumulative value Lnx of each length recognized up to the present time (S2b) and predicts the yield of superconducting wire 10 according to cumulative value Lnx (S2c). Control device 190 determines whether the predicted yield is below a first predetermined value (S3). In this example, x=1, and control device 190 determines whether the difference between length LS and value Ln1 is below the first predetermined value. If the predicted yield is below the first predetermined value (YES in S3), control device 190 executes notification control (S4). If the predicted yield is equal to or greater than the first predetermined value (NO in S3), or after S4, a first cutting process (S5) is performed, followed by a polishing process (S10) and a second inspection process (S11).

[0186] If the second inspection process finds that there is a defect on the surface of the ground substrate Sb2 (YES in S11), the control device 190 recognizes the length of at least one defective portion of the ground substrate Sb2 in the longitudinal direction based on the surface image of the ground substrate Sb2 (S12a). The control device 190 calculates a cumulative value Lnx of each length recognized up to the current time point (S12b) and predicts the yield of the superconducting wire 10 based on the cumulative value Lnx (S12c). The control device 190 determines whether the predicted yield is below a first predetermined value (S13). In this example, x=2, and the control device 190 determines whether the difference between the value of the length LS and the cumulative value Ln2 is below the first predetermined value. If the predicted yield is below the first predetermined value (YES in S13), the control device 190 executes notification control (S14). If the predicted yield is equal to or greater than the first predetermined value (NO in S13), or after S14, a second cutting process (S15) is performed, followed by an intermediate layer forming process (S20) and a third inspection process (S21 in FIG. 30).

[0187] 30, if the third inspection process finds a defect on the surface of the laminated material Lm1 (YES in S21), the control device 190 recognizes the length of at least one defective portion of the laminated material Lm1 in the longitudinal direction according to the surface image of the laminated material Lm1 (S22a). The control device 190 calculates a cumulative value Lnx of each length recognized up to the current time point (S22b) and predicts the production volume of the superconducting wire 10 according to the cumulative value Lnx (S22c). The control device 190 determines whether the predicted production volume is below a first predetermined value (S23). In this example, x=3, and the control device 190 determines whether the difference between the value of the length LS and the cumulative value Ln3 is below the first predetermined value. If the predicted production volume is below the first predetermined value (YES in S23), the control device 190 executes notification control (S24). If the predicted yield is equal to or greater than the first predetermined value (NO in S23), or after S24, a third cutting process (S25) is performed, followed by a superconducting layer forming process (S30) and a fourth inspection process (S31).

[0188] If the fourth inspection process finds a defect on the surface of the laminated material Lm2 (YES in S31), the control device 190 recognizes the length of at least one defective portion of the laminated material Lm2 in the longitudinal direction based on the surface image of the laminated material Lm2 (S32a). The control device 190 calculates a cumulative value Lnx of each length recognized up to the current time point (S32b) and predicts the production volume of the superconducting wire 10 based on the cumulative value Lnx (S32c). The control device 190 determines whether the predicted production volume is below a first predetermined value (S33). In this example, x=4, and the control device 190 determines whether the difference between the value of the length LS and the cumulative value Ln4 is below the first predetermined value. If the predicted production volume is below the first predetermined value (YES in S33), the control device 190 executes notification control (S34). If the predicted yield is equal to or greater than the first predetermined value (NO in S33), or after S34, a fourth cutting process (S35) is performed, followed by a protective layer forming process (S40) and a fifth inspection process (S41 in FIG. 31).

[0189] 31, if the fifth inspection process finds a defect on the surface of the laminated material Lm3 (YES in S41), the control device 190 recognizes the length of at least one defective portion of the laminated material Lm3 in the longitudinal direction according to the surface image of the laminated material Lm3 (S42a). The control device 190 calculates a cumulative value Lnx of each length recognized up to the current time point (S42b) and predicts the production volume of the superconducting wire 10 according to the cumulative value Lnx (S42c). The control device 190 determines whether the predicted production volume is below a first predetermined value (S43). In this example, x=5, and the control device 190 determines whether the difference between the value of the length LS and the cumulative value Ln5 is below the first predetermined value. If the predicted production volume is below the first predetermined value (YES in S43), the control device 190 executes notification control (S44). If the predicted yield is equal to or greater than the first predetermined value (NO in S43), or after S44, a fifth cutting process (S45) is performed, followed by a stabilizing layer forming process (S50) and a sixth inspection process (S51).

[0190] If the sixth inspection process finds a defect on the surface of the laminated material Lm4 (YES in S51), the control device 190 recognizes the length of at least one defective portion of the laminated material Lm4 in the longitudinal direction according to the surface image of the laminated material Lm4 (S52a). The control device 190 calculates a cumulative value Lnx of each length recognized up to the current time point (S52b) and predicts the production volume of the superconducting wire 10 according to the cumulative value Lnx (S52c). The control device 190 determines whether the predicted production volume is below a first predetermined value (S53). In this example, x=6, and the control device 190 determines whether the difference between the value of the length LS and the cumulative value Ln6 is below the first predetermined value. If the predicted production volume is below the first predetermined value (YES in S53), the control device 190 executes notification control (S54). If the predicted yield is equal to or greater than the first predetermined value (NO in S53), or after S54, a sixth cutting process (S55) is performed. Thereafter, a final process (S60) is performed on the laminated material Lm4, and superconducting wire 10 is completed.

[0191] In the above, in any of the six processing steps from the polishing step to the final step, the control device 190 recognizes the length of at least one defective portion of the workpiece PS in that step according to the surface image of the workpiece, and executes notification control according to the recognition results ((1) S2a to S4, (2) S12a to S14, (3) S22a to S24, (4) S32a to S34, (5) S42a to S44, (6) S52a to S54).

[0192] Alternatively, the control device 190 may recognize the length of at least one defective portion of the workpiece PS in at least one of the six processing steps according to the surface image of the workpiece, and execute notification control according to the recognition result. In other words, the control device 190 may execute at least one of (1) to (6) above, rather than executing all of (1) to (6). Even in this case, the control device 190 executes notification control when the difference between the value of the length LS and the cumulative value of the lengths recognized up to the present time (the value of the length of the entire defective portion) falls below the first predetermined value. A process consisting of at least one of S2a, S12a, S22a, S32a, S42a, and S52a is also referred to as a "recognition process."

[0193] The first predetermined value may be determined based on the progress information 258 (see FIG. 5) stored in the storage 196. This point will be explained below.

[0194] FIG. 32 is a diagram illustrating an example of progress information 258 according to the second embodiment. Referring to FIG. 32, the progress information 258 indicates the progress of the manufacturing process according to whether each of the six processing steps from the polishing step to the final step has been performed. In this example, the progress dp is defined as the percentage (%) of the number of steps that have been performed among the six processing steps. Specifically, if x steps out of the six steps have been performed, the progress dp is (x / 6×100)%.

[0195] For example, if three processing steps from the polishing step to the superconducting layer forming step have been completed, and three processing steps from the protective layer forming step to the final step have not been completed, the progress rate dp is 50%. Information indicating the implementation status (completed / not completed) of each step is determined by a user operation using the input device 187 (see FIG. 4).

[0196] The control device 190 reads out the progress information 258 and sets the first predetermined value based on the progress dp indicated by the read-out progress information 258. The lower the progress dp, the higher the first predetermined value is set. In this example, the first predetermined value is set as the sum of the requested length of the superconducting wire 10 and the product of the first allowable upper limit and (1-progress dp / 100). In one example, when the progress dp is 50%, the first predetermined value is set as the sum of the requested length and half the first allowable upper limit. As a result, for example, when the expected value of the output is already below the above-mentioned sum when the laminated material Lm2 is produced after the superconducting layer forming step (in other words, when the cumulative value Lnx has already exceeded half the first allowable upper limit), the control device 190 performs notification control, assuming that it is highly likely that the requested length of the superconducting wire 10 will not be achieved.

[0197] By setting the first predetermined value based on the degree of progress dp as described above, even if the currently predicted volume of superconducting wire 10 has a large margin relative to the desired length, if it is predicted in the early to middle stages of the manufacturing process that the volume is likely to be unable to achieve the desired length, the user is prompted to prepare an additional workpiece PS as a backup. As a result, preparation of the additional workpiece PS begins earlier, and the second superconducting wire can be completed earlier. Therefore, by using the second superconducting wire to supplement the shortfall in superconducting wire 10, the desired length can be achieved earlier.

[0198] 33 is a flowchart illustrating the process executed by the control device 190 when setting the first predetermined value. This flowchart is started after S2c, S12c, S22c, S32c, S42c, or S52c.

[0199] 33, the control device 190 reads the progress degree information 258. The control device 190 sets the first predetermined value in accordance with the progress degree dp indicated by the progress degree information 258. Thereafter, the process proceeds to S3, S13, S23, S33, S43, or S53.

[0200] The step of reading progress information 258 is performed by control device 190 as S2d if performed after S2c, as S12d if performed after S12c, as S22d if performed after S22c, as S32d if performed after S32c, as S42d if performed after S42c, or as S52d if performed after S52c.

[0201] The step of setting the first predetermined value is performed by the control device 190 as S2e if performed after S2d, as S12e if performed after S12d, as S22e if performed after S22d, as S32e if performed after S32d, as S42e if performed after S42d, or as S52e if performed after S52d.

[0202] As described above, according to the second embodiment, when it is predicted during the manufacturing process of superconducting wire 10 that the output will not reliably achieve the required length value or that there is a high possibility that the required length value will not be achieved, the user is notified of this fact. This prompts the user to prepare an additional workpiece PS, which is any one of unpolished substrate Sb1 to laminated material Lm4. As a result, preparation of the additional workpiece PS can be started before, rather than after, completion of superconducting wire 10. Therefore, preparation of the additional workpiece PS can be completed early, thereby shortening the time interval between completion of superconducting wire 10 and start of manufacturing the second superconducting wire. Therefore, manufacturing of the second superconducting wire can be started early, thereby allowing the second superconducting wire to be completed early. As a result, the required length can be achieved early by using the second superconducting wire to make up for the shortfall in superconducting wire 10.

[0203] <Modification of the second embodiment> In this modification, the control device 190 predicts the yield of N (≧2) superconducting wires 10 according to the results of the sorting process for the workpiece PS during the manufacturing process of these superconducting wires, and if the predicted yield falls below a predetermined value, notifies the user of the control system 100. In this modification, the yield is the aggregate value of the lengths in the longitudinal direction of the N superconducting wires 10, each having a critical current value Icv equal to or greater than a reference value (hereinafter also referred to as "total length"). This point will be explained in detail below.

[0204] FIG. 34 is a diagram showing how the length of the object PS for processing a superconducting wire changes for each superconducting wire 10 in the manufacturing process for manufacturing N superconducting wires 10. In FIG.

[0205] 34, in this example, unpolished substrates Sb1-1, ..., Sb1-N are prepared to produce superconducting wires 10-1, ..., 10-N in this order. Unpolished substrate Sb1-k (k = 1, 2, ..., N) has length LS-k in the longitudinal direction. As will be described below, to produce superconducting wire 10-k, unpolished substrate Sb1-k is used to produce polished substrate Sb2-k and laminated materials Lm1-k, Lm2-k, Lm3-k, and Lm4-k in this order.

[0206] A series of steps for manufacturing superconducting wire 10-k using unpolished substrate Sb1-k is also referred to as the “kth manufacturing step.” Each of the first to Nth manufacturing steps includes a polishing step, an intermediate layer forming step, a superconducting layer forming step, a protective layer forming step, a stabilizing layer forming step, and a final step for superconducting wire 10 manufactured in the corresponding manufacturing step.

[0207] Unpolished substrate Sb1-k corresponds to unpolished substrate Sb1 for superconducting wire 10-k. Unpolished substrate Sb1-k is divided into defective portion 805-k and good portion 807-k by a division process. Control device 190 recognizes that defective portion 805-k has length Lp1-k in the longitudinal direction based on the surface image of unpolished substrate Sb1-k. Control device 190 stores the value of length Lp1-k in memory 192. Defective portion 805-k is separated by a first cutting process performed on unpolished substrate Sb1-k. Thereafter, unpolished substrate Sb1-k is polished in a polishing process to produce polished substrate Sb2-k.

[0208] The ground substrate Sb2-k corresponds to the ground substrate Sb2 for the superconducting wire 10-k. The ground substrate Sb2-k is divided into a defective portion 815-k and a good portion 817-k by a division process. The control device 190 recognizes that the defective portion 815-k has a length Lp2-k in the longitudinal direction based on the surface image of the ground substrate Sb2-k. The control device 190 calculates the sum of the lengths Lp1-k and Lp2-k and stores it in the memory 192. The defective portion 815-k is cut off by a second cutting process of the ground substrate Sb2-k. Thereafter, an intermediate layer is formed on the ground substrate Sb2-k by an intermediate layer forming process, thereby producing the laminate Lm1-k.

[0209] The laminated material Lm1-k corresponds to the laminated material Lm1 for the superconducting wire 10-k. The laminated material Lm1-k is divided into a defective portion 825-k and a good portion 827-k by a division process. The control device 190 recognizes that the defective portion 825-k has a length Lp3-k in the longitudinal direction according to the surface image of the laminated material Lm1-k. The control device 190 calculates the sum of the lengths Lp1-k to Lp3-k and stores it in the memory 192. The defective portion 825 is cut off from the laminated material Lm1-k by a third cutting process. Thereafter, a superconducting layer is formed on the intermediate layer of the laminated material Lm1-k by a superconducting layer forming process, thereby producing the laminated material Lm2-k.

[0210] The laminated material Lm2-k corresponds to the laminated material Lm2 for the superconducting wire 10-k. The laminated material Lm2-k is divided into a defective portion 835-k and a good portion 837-k by a division process. The control device 190 recognizes that the defective portion 835-k has a length Lp4-k in the longitudinal direction based on the surface image of the laminated material Lm2-k. The control device 190 calculates the sum of the lengths Lp1-k to Lp4-k and stores the sum in the memory 192. The defective portion 835-k is cut off by a fourth cutting process for the laminated material Lm2-k. Thereafter, a protective layer is formed on the superconducting layer of the laminated material Lm2-k by a protective layer forming process, thereby producing the laminated material Lm3-k.

[0211] The laminated material Lm3-k corresponds to the laminated material Lm3 for the superconducting wire 10-k. The laminated material Lm3-k is divided into a defective portion 845-k and a good portion 847-k by a division process. The control device 190 recognizes that the defective portion 845-k has a length Lp5-k in the longitudinal direction based on the surface image of the laminated material Lm3-k. The control device 190 calculates the sum of the lengths Lp1-k to Lp5-k and stores the sum in memory 192. The defective portion 845 is cut off by a fifth cutting process for the laminated material Lm3-k. Thereafter, a stabilizing layer is formed on the protective layer of the laminated material Lm3-k by a stabilizing layer forming process, thereby producing the laminated material Lm4-k.

[0212] The laminated material Lm4-k corresponds to the laminated material Lm4 for the superconducting wire 10-k. The laminated material Lm4-k is divided into a defective portion 855-k and a good portion 857-k by a division process. The control device 190 recognizes that the defective portion 855-k has a length Lp6-k in the longitudinal direction according to the surface image of the laminated material Lm4-k. For each value of k, the control device 190 calculates the sum (ΣLpx-k: 1≦x≦6) of the lengths Lp1-k to Lp6-k and stores the calculated sum in the memory 192. The defective portion 855-k is cut off by a sixth cutting process for the laminated material Lm4-k. Thereafter, the final process is performed to produce the superconducting wire 10-k.

[0213] The control device 190 executes the above-mentioned recognition process for each of the first to Nth manufacturing processes. Specifically, the recognition process is a process for recognizing, for each of at least one processing step from the polishing step to the final step of the manufacturing process, the longitudinal length of at least one defective portion of the workpiece PS processed in that step, according to a surface image of the workpiece. In this example, the control device 190 recognizes lengths Lp1-k, Lp2-k, ... Lp6-k in the kth manufacturing process for each value of k.

[0214] If the yield of superconducting wires 10-1, ..., 10-N, i.e., the total length of these superconducting wires in the longitudinal direction ((LA-1) + ... (LA-N)), is less than a predetermined total required length, there is a possibility that these superconducting wires will have few practical uses. Therefore, it is preferable that the yield of these superconducting wires be equal to or greater than the total required length. The total required length is at least the length necessary as the total length of these superconducting wires in the longitudinal direction, and is determined in advance as appropriate from a practical perspective. Information indicating the value of the total required length is stored in advance in memory 192 or storage 196.

[0215] The difference between the total length ((LS-1) + (LS-N)) of unpolished substrates Sb1-1, ... Sb1-N and the total required length corresponds to the allowable upper limit of the longitudinal length of all defective portions (e.g., all defective portions 805-1, 815-1, 825-1, ... 845-N, 855-N) that occur in the first through Nth manufacturing steps. This upper limit is also referred to as the "second allowable upper limit." If it is predicted during the manufacturing process that the output of superconducting wire 10-1, ... 10-N will fall below the total required length, the total required length cannot be reliably achieved even if no surface abnormalities are caused in the workpiece PS in all subsequent steps.

[0216] If it is discovered after the completion of superconducting wire 10-N that the volume of these superconducting wires does not meet the total required length and preparation of additional processing objects PS is then started, it will be difficult to complete the additional superconducting wires promptly due to the long lead time required for such preparation.

[0217] Therefore, the control device 190 predicts the yield of the superconducting wires 10-1, ..., 10-N according to the results of the sorting process for the workpiece PS, and executes notification control to notify the user when the predicted yield falls below a second predetermined value. The notification control is, for example, a process of controlling the display device 185 (see FIG. 4) to display a screen notifying the user as described above, but may also be executed by voice. For example, in the x (1≦x≦6)th processing step of the kth manufacturing process, the control device 190 assumes that the difference between the total length value of the lengths LS-1, ..., LS-N and the cumulative value LnX of the lengths Lp1-1, Lp1-2, Lp1-3, ..., Lp1-k, ..., Lpx-k is the predicted yield value, and executes notification control when this predicted value falls below a second predetermined value. The second predetermined value is appropriately determined in advance based on the total required length, and is, for example, the total required length value itself. The predicted value falling below the second predetermined value corresponds to the cumulative value LnX exceeding the second allowable upper limit. The second predetermined value may be smaller than the total requested length, but is preferably equal to or greater than the total requested length, as described below.

[0218] For example, when the second predetermined value is equal to the total required length, notification control is executed when the predicted value of the production volume (the difference value) falls below the total required length. Therefore, when it is predicted during the manufacturing process of superconducting wires 10-1, ..., 10-N that the production volume will not reliably achieve the total required length, the user is prompted to prepare an additional workpiece PS, any one of unpolished substrate Sb1 to laminated material Lm4. This allows preparation of the additional workpiece PS to begin before, rather than after, completion of superconducting wire 10-N. As a result, preparation of the additional workpiece PS can be completed early, thereby shortening the time interval between the completion of superconducting wire 10-N and the start of manufacturing the additional superconducting wire. Therefore, manufacturing of the additional superconducting wire can be started early, allowing the superconducting wire to be completed early. Therefore, the required length can be achieved early by using the additional superconducting wire to make up for the shortfall in superconducting wires 10-1, ..., 10-N. Furthermore, if the total required length cannot be reliably achieved, preparation of additional workpieces PS is started, so that additional workpieces are not prepared unnecessarily. The number of additional workpieces PS may be two or more.

[0219] The second predetermined value may be greater than the total required length by a given value. In this case, the notification control is executed some time before the predicted value of the output falls below the total required length (before it is determined that the total required length cannot be achieved). As a result, preparation of additional workpieces PS can be started earlier, and additional superconducting wire can be manufactured earlier. Therefore, the total required length can be achieved earlier.

[0220] 35 is a flowchart for explaining an outline of a manufacturing process for superconducting wire 10-1, ..., 10-N in this modified example. This flowchart starts in a state where k=1. The value of k is determined by a user operation using input device 187, for example.

[0221] 35, first, a manufacturing step (kth manufacturing step) of superconducting wire 10-k is performed (S310). S310 includes steps S1 to S60 (see FIGS. 29 to 31) that are performed to manufacture superconducting wire 10-k. In S310, unpolished substrate Sb1, polished substrate Sb2, laminated materials Lm1, Lm2, Lm3, and Lm4, cumulative value Lnx, and the first predetermined value described in steps S1 to S60 of the second embodiment are replaced with unpolished substrate Sb1-k, polished substrate Sb2-k, laminated materials Lm1-k, Lm2-k, Lm3-k, and Lm4-k, cumulative value LnX, and the second predetermined value, respectively. S310 may further include S2d, S2e, S12d, S12e, S22d, S22e, S32d, S32e, S42d, S42e, S52d, and S52e (all of which are shown in FIG. 33 ) of the second embodiment. In this case, in these steps, the first predetermined value is replaced with a second predetermined value. Control device 190, for example, executes a recognition process on superconducting wire 10-k, and executes notification control when a predicted value of the volume of superconducting wire 10-1, ..., 10-N (a difference value between the total length value of lengths LS-1, ..., LS-N and a cumulative value LnX of each length recognized up to the present time) falls below the second predetermined value.

[0222] After S310, control device 190 receives input of information indicating whether the number of superconducting wires 10 to be manufactured (k) has reached the planned number of superconducting wires 10 to be manufactured (N), and determines whether the number of superconducting wires 10 to be manufactured has reached the planned number of superconducting wires 10 to be manufactured (S315). This information is input by a user operation using input device 187.

[0223] If the number of products to be manufactured has not yet reached the planned number of products to be manufactured (NO in S315), the control device 190 counts up k (S320). Then, the process returns to S310. As a result, S310 to S320 are repeated, and the first manufacturing process, the second manufacturing process, ... are carried out. If the number of products to be manufactured has reached the planned number of products to be manufactured (YES in S315), the control device 190 considers the first to Nth manufacturing processes to be complete and ends the process. While S310 is being repeated, the control device 190 executes the recognition process for each of the first to Nth manufacturing processes.

[0224] In the above, control device 190 executes the recognition process for each of the first to Nth manufacturing processes and executes notification control in accordance with the recognition results. Alternatively, control device 190 may execute the recognition process for at least one of the manufacturing processes, rather than for all of the manufacturing processes, and execute notification control in accordance with the recognition results. Even in this case, when the predicted value of the volume of superconducting wire 10-1, ..., 10-N (the difference between the total length value of lengths LS-1, ..., LS-N and the cumulative value of each length recognized up to the present time) falls below the second predetermined value, control device 190 executes notification control.

[0225] The second predetermined value may be determined based on the progress information 258 (see FIG. 5). This point will be explained below.

[0226] Fig. 36 is a diagram showing an example of progress information 258 in this modified example. Referring to Fig. 36, progress information 258 indicates the progress of each of the first to Nth manufacturing processes, depending on whether or not each of the six processing steps from the polishing step to the final step in the manufacturing process has been carried out.

[0227] In this example, the degree of progress DP is determined as the percentage (%) of the number of processes that have been performed among 6×N processing processes for manufacturing superconducting wires 10-1, ..., 10-N. Specifically, when m processes have been performed among 6×N processing processes, the degree of progress DP is (m / 6N×100)%. In the following description, N=2h (h is a natural number), but N may also be an odd number.

[0228] For example, if 3×N processing steps from the polishing step in the first manufacturing process to the superconducting layer forming step in the h-th (=N / 2) manufacturing process have been completed, and 3×N processing steps from the protective layer forming step in the h-th manufacturing process to the final step in the N-th manufacturing process have not been completed, the degree of progress DP is 50%. Information indicating the implementation status (implemented / not implemented) of each processing step is determined by user operation using input device 187 (see FIG. 4).

[0229] Control device 190 reads out progress information 258 and sets the second predetermined value based on progress DP indicated by the read out progress information 258. The lower the progress DP, the higher the second predetermined value is set. In this example, the second predetermined value is set as the sum of the total required length of superconducting wire 10-1, ..., 10-N, the multiplication of the second allowable upper limit value and (1 - progress DP / 100). In one example, when progress DP is 50%, the second predetermined value is set as the sum of the total required length and half of the second allowable upper limit value. As a result, for example, when laminated material Lm2-h is produced after the superconducting layer forming step in the h-th manufacturing step, if the expected value of the output has already fallen below the above sum (in other words, if cumulative value LnX has already exceeded half of the second allowable upper limit value), control device 190 performs notification control, assuming that it is highly likely that the total required length will not be achieved.

[0230] By setting the second predetermined value based on the degree of progress DP as described above, when it is predicted that the expected volume of superconducting wire 10-1, ..., 10-N currently has a large margin relative to the total required length, the user is prompted to prepare an additional workpiece PS, any one of unpolished substrate Sb1 to laminate Lm4, as a backup when it is highly likely that the volume will not achieve the total required length in the early to middle stages of the manufacturing process. This allows preparation of the additional workpiece PS to begin earlier, so that the additional superconducting wire can be completed earlier. As a result, the additional superconducting wire can be used to supplement the shortfall in superconducting wire 10-1, ..., 10-N, thereby achieving the total required length earlier.

[0231] As described above, according to this modification, when it is predicted during the manufacturing process of superconducting wire 10-1, ..., 10-N that the total required length cannot be reliably achieved or that there is a high possibility that the total required length will not be achieved, the user is prompted to prepare an additional workpiece PS, any one of unpolished substrate Sb1 to laminate Lm4. This allows the preparation of the additional workpiece PS to begin before, rather than after, the completion of superconducting wire 10-N. As a result, the preparation of the additional workpiece PS is completed early, thereby shortening the time interval between the completion of superconducting wire 10-N and the start of manufacturing the additional superconducting wire. Therefore, manufacturing of the additional superconducting wire can be started early, allowing the superconducting wire to be completed early. Therefore, the total required length can be achieved early by using the additional superconducting wire to supplement the shortfall in the overall length of superconducting wire 10-1, ..., 10-N.

[0232] The embodiments disclosed herein should be considered to be illustrative in all respects and not restrictive. The scope of the present invention is defined by the claims, not by the above description, and is intended to include all modifications within the meaning and scope of the claims. [Explanation of symbols]

[0233] 10, 10-1, 10-k, 10-N superconducting wire 11 Circuit Board 12 Middle Class 13 Superconducting layer 14 Protective layer 15 Stabilization Layer 100 Control System 110 Conveyor 112,114 Motor 116,118 Pulley 120 Lighting equipment 130 Camera 140 Marking device 145a,145b Container 147 Marker 150 Transporting Device 155 First Storage Space 157 Second Storage Space 159 Third Storage Space 160,160A cutting device 180,180A joining equipment 190 Control Device 192 memory 196 Storage 251,252,253,254 trained models 255,257 maps 258 Progress Information 305,310,315,320,325 part 405,415,505,515,525,605,615,625,635,807,807-1,807-k,807-N,817,817-1,817-k,817-N,827 ,827-1,827-k,827-N,837,837-1,837-k,837-N,847,847-1,847-k,847-N,857,857-1,857-k,857-N good quality part 410,510,520,610,620,630,805,805-1,805-k,805-N,815,815-1,815-k,815-N,825,825-1 ,825-k,825-N,835,835-1,835-k,835-N,845,845-1,845-k,845-N,855,855-1,855-k,855-N defective part 455,460,470,475,755,760,770,775 marks 655 Joints 705 A rank part 710 B rank part 715 C rank part A1,A2,A11,A14,A21,A26 cutting line CD1,CD2,CD3,CD4 distribution DfA,DfB,DfC surface abnormality EfA, EfB, EfC electric field distribution Icv critical current value Jc critical current density Ln1 value Ln2, Ln3, Ln4, Ln5, Ln6, Lnx, LnX cumulative value Lm1,Lm1-1,Lm1-k,Lm1-N,Lm2,Lm2-1,Lm2-k,Lm2-N,Lm3,Lm3-1,Lm3-k,Lm3-N,Lm4,Lm4-1,Lm4-k,Lm4-N Laminated material Lref, LA, LS, Lp1, Lp2, Lp3, Lp4, Lp5, Lp6 length MSc Section PS processing object Pim imaging position Ra surface roughness Sb1,Sb1-1,Sb1-k,Sb1-N Unpolished substrate Sb2, Sb2-1, Sb2-k, Sb2-N polished substrate TH1, TH2 threshold Th Thickness W width

Claims

1. A method carried out in a manufacturing process of a superconducting wire, the method comprising: acquiring a surface image of an object to be processed in the manufacturing process, the object including a substrate for the superconducting wire; predicting a current value distribution representing a distribution of critical current values ​​of the superconducting wire in a longitudinal direction of the object according to the surface image; and dividing the workpiece into at least one first portion and at least one second portion according to a result of the current value distribution prediction; the at least one first portion is a portion in the current value distribution where the critical current value at the position of the portion in the longitudinal direction is predicted to be less than a reference value, the at least one second portion is a portion in the current value distribution where the critical current value at the position of the portion in the longitudinal direction is predicted to be equal to or greater than the reference value, The method further comprises: cutting the workpiece such that the at least one first portion is separated from the workpiece.

2. the at least one second portion includes a plurality of third portions each having a length in the longitudinal direction that is less than the reference length, The method further comprises:

2. The method of claim 1, further comprising joining the plurality of third portions in the longitudinal direction after cutting the workpiece if the sum of the lengths of the plurality of third portions in the longitudinal direction is equal to or greater than the reference length.

3. The method of claim 1 , further comprising the step of marking both longitudinal ends of the at least one second portion.

4. the at least one second portion includes at least one fourth portion and at least one fifth portion; the at least one fourth portion is a portion in the current value distribution where the critical current value at the position of the portion in the longitudinal direction is predicted to be less than a threshold value that is equal to or greater than the reference value, the at least one fifth portion is a portion in the current value distribution where the critical current value at the position of the portion in the longitudinal direction is predicted to be equal to or greater than the threshold value, The method further comprises:

2. The method of claim 1, further comprising forming a first marking on both ends of the at least one fourth portion in the longitudinal direction, and forming a second marking different from the first marking on both ends of the at least one fifth portion in the longitudinal direction.

5. the manufacturing process includes polishing the substrate; the object to be processed includes an unpolished substrate as the substrate before polishing, The method of claim 1 , wherein the surface image comprises an image representing a surface of the unpolished substrate.

6. the manufacturing process includes polishing the substrate; the object to be processed includes a polished substrate as the substrate after polishing, The method of claim 1 , wherein the surface image comprises an image representing a surface of the polished substrate.

7. The manufacturing process includes: polishing the substrate; forming an intermediate layer on the polished substrate; the workpiece includes the substrate after polishing and a first laminate including the intermediate layer; The method of claim 1 , wherein the surface image includes an image representing the surface of the intermediate layer as the surface of the first laminate.

8. The manufacturing process includes: polishing the substrate; forming an intermediate layer on the polished substrate; forming a superconducting layer on the intermediate layer; the object to be processed includes a second laminated material including the polished substrate, the intermediate layer, and the superconducting layer; The method of claim 1 , wherein the surface image includes an image representing the surface of the superconducting layer as the surface of the second laminate.

9. The manufacturing process includes: polishing the substrate; forming an intermediate layer on the polished substrate; forming a superconducting layer on the intermediate layer; forming a protective layer on the superconducting layer; the object to be processed includes a third laminated material including the polished substrate, the intermediate layer, the superconducting layer, and the protective layer; The method of claim 1 , wherein the surface image includes an image representing the surface of the protective layer as the surface of the third laminate.

10. The manufacturing process includes: polishing the substrate; forming an intermediate layer on the polished substrate; forming a superconducting layer on the intermediate layer; forming a protective layer on the superconducting layer; forming a stabilizing layer on the protective layer; the object to be processed includes a fourth laminated material including the polished substrate, the intermediate layer, the superconducting layer, the protective layer, and the stabilizing layer; The method of claim 1 , wherein the surface image includes an image representing the surface of the stabilization layer as the surface of the fourth laminate.

11. The step of predicting the current value distribution includes: identifying the type of surface abnormality according to the surface image using a trained model that has trained a relationship between the surface image and the type of surface abnormality of the object to be processed; estimating the amount of decrease corresponding to the surface anomaly type according to the identified type using a predetermined correspondence relationship between the type of the surface anomaly and the amount of decrease in the critical current value; and predicting the current value distribution according to the estimated amount of decrease.

12. The step of predicting the current value distribution includes: Recognizing an abnormal region on the surface of the superconducting layer of the superconducting wire according to the surface image; calculating an effective width of the superconducting layer in a short-side direction of the superconducting layer at the position of the anomalous region in the longitudinal direction according to the result of the recognition of the anomalous region, wherein the effective width is calculated by subtracting a width of a surface anomaly included in the anomalous region in the short-side direction from a width of the superconducting wire in the short-side direction; The step of predicting the current value distribution further includes: estimating the critical current value at the position according to the effective width, the thickness of the superconducting layer, and the critical current density of the superconducting wire; and predicting the current value distribution according to the estimated critical current value at the location.

13. The step of predicting the current value distribution includes: inferring the electric field distribution according to the surface image using a trained model that has trained a relationship between the surface image and the electric field distribution of the superconducting layer of the superconducting wire; and predicting the current value distribution according to the inferred result of the electric field distribution.

14. The step of predicting the current value distribution includes: Inferring the distribution of surface roughness according to the surface image using a trained model that has trained the relationship between the surface image and the distribution of surface roughness of the object to be machined; and predicting the current value distribution according to the inferred result of the surface roughness distribution.

15. a step of predicting the output of the superconducting wire having the critical current value equal to or greater than the reference value according to the result of the classifying step; 2. The method of claim 1, further comprising the step of: providing an indication if the predicted volume falls below a predetermined value.

16. The method of claim 15 , wherein the predetermined value is equal to or greater than the required length of the superconducting wire.

17. The method further includes reading information indicating a progress of the manufacturing process of the superconducting wire from a storage unit that stores the information, The method of claim 16 , wherein the predetermined value is determined based on the progress.

18. A superconducting wire manufactured by carrying out the method according to any one of claims 1 to 17.

19. an imaging device that captures an image of an object to be processed in a manufacturing process of the superconducting wire, the object including a substrate for the superconducting wire, and generates a surface image of the object; a cutting device configured to cut the workpiece; a control device that executes prediction processing, division processing, and cutting control; the prediction process is a process of predicting a current value distribution representing a distribution of critical current values ​​of the superconducting wire in a longitudinal direction of the object, based on the surface image; the division process is a process of dividing the object into at least one first portion and at least one second portion according to the current value distribution, the at least one first portion is a portion in the current value distribution where the critical current value at the position of the portion in the longitudinal direction is predicted to be less than a reference value, the at least one second portion is a portion in the current value distribution where the critical current value at the position of the portion in the longitudinal direction is predicted to be equal to or greater than the reference value, The cutting control is to control the cutting device to cut the workpiece so as to separate the at least one first portion from the workpiece.

20. Further comprising an alarm device for notifying a user, The control device further a prediction process for predicting the output of the superconducting wire having the critical current value equal to or greater than the reference value according to the result of the classification process; 20. The control system according to claim 19, further comprising: a notification control for controlling the notification device to notify a user that the predicted volume falls below a predetermined value.

21. The control system according to claim 20 , wherein the predetermined value is equal to or greater than the required length of the superconducting wire.

22. a storage unit that stores information indicating a progress of the manufacturing process of the superconducting wire; The control system according to claim 21 , wherein the predetermined value is determined based on the progress.

Images