Information processing device, glass plate manufacturing device, information processing method, and glass plate manufacturing method

Abstract

The invention provides an information processing device, a glass plate manufacturing device, an information processing method and a glass plate manufacturing method, and provides an information processing device capable of realizing quality maintenance of plate thickness of a manufactured glass plate by rapidly performing temperature adjustment and the like when the plate thickness of the manufactured glass plate deviates from a constant range. The information processing device includes a determination unit that inputs information on manufacturing conditions during manufacturing of a glass plate to a learning completion model for machine learning, with respect to the glass plate that is manufactured while flowing in a predetermined flow direction. An output result from the learned model is acquired as information relating to an estimation result of the thickness of the glass plate.

Description

Technical Field

[0001] The present disclosure relates to an information processing device, a glass plate manufacturing device, an information processing method, and a glass plate manufacturing method. Background Art

[0002] The glass sheet is produced by the float process or the like (see, for example, Patent Document 1).

[0003] In the production of a glass plate, maintaining the plate thickness of the produced glass plate within a constant range is an important quality control.

[0004] Conventionally, the thickness of the glass sheet is constantly monitored downstream of a glass sheet manufacturing apparatus, and the temperature in a molten metal tank is adjusted based on the thickness to maintain the thickness within a constant range.

[0005] Prior art literature

[0006] Patent Literature

[0007] Patent Document 1: Japanese Patent Application Publication No. 2016-98161

[0008] Problems to be solved by the invention

[0009] However, in the prior art, the thickness of the glass sheet is monitored in the downstream of the glass sheet manufacturing device while the temperature of the glass sheet has dropped, and the temperature in the molten metal tank upstream thereof is adjusted, so there is a large time difference (for example, 30 minutes, etc.) between them. In the prior art, due to such a time difference, when the temperature adjustment is incomplete for some reason, the correction may be greatly delayed, and the reduction of the yield may continue for a long time.

[0010] In addition, not limited to the float process, the same problem also exists in the melt process and the like. Summary of the invention

[0011] The present disclosure was completed in consideration of such a situation, and the subject is to provide an information processing device, a glass plate manufacturing device, an information processing method and a glass plate manufacturing method that can maintain the quality of the thickness of the glass plate after manufacturing by quickly adjusting the temperature when the thickness of the glass plate after manufacturing deviates from a constant range.

[0012] Means for solving problems

[0013] One scheme is an information processing device, which includes a determination unit, which inputs information on manufacturing conditions during the manufacturing of a glass plate that flows in a specified flow direction and is manufactured into a machine learning learned model, and obtains an output result from the learned model as information related to an estimated result of the plate thickness of the glass plate.

[0014] One scheme is a glass plate manufacturing device, comprising: an information processing device, comprising a determination unit, which inputs information on manufacturing conditions of a glass plate that flows in a specified flow direction and is manufactured into a machine-learned learned model, and obtains an output result from the learned model as information related to an estimated result of the thickness of the glass plate; and an adjustment unit, which adjusts the manufacturing conditions based on the estimated result generated by the determination unit.

[0015] One scheme is an information processing method, in which a judgment unit inputs information on manufacturing conditions during the manufacturing of a glass plate that flows in a specified flow direction into a learned model of machine learning, and obtains an output result from the learned model as information related to an estimated result of the thickness of the glass plate.

[0016] One scheme is a glass plate manufacturing method, wherein a judgment unit inputs information on manufacturing conditions during the manufacturing of a glass plate that flows in a specified flow direction and is manufactured into a learned model of machine learning, obtains an output result from the learned model as information related to an estimated result of the thickness of the glass plate, and an adjustment unit adjusts the manufacturing conditions based on the estimated result generated by the judgment unit.

[0017] Effects of the Invention

[0018] According to the information processing device, information processing device, glass plate manufacturing device, information processing method and glass plate manufacturing method involved in the present disclosure, the quality maintenance of the thickness of the glass plate after manufacturing can be achieved by quickly adjusting the temperature when the thickness of the glass plate after manufacturing deviates from a constant range. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 It is a figure which shows the schematic structural example of the glass plate manufacturing apparatus which concerns on embodiment.

[0020] Figure 2 It is a diagram showing a configuration example of an information processing device according to an embodiment.

[0021] Figure 3 This is a diagram showing an example of input and output information of a learned model according to an embodiment.

[0022] Figure 4 It is a diagram showing an example of manufacturing conditions according to the embodiment.

[0023] Figure 5 This is a diagram showing an example of the state of the glass plate according to the embodiment.

[0024] Figure 6 This is a diagram showing an example of the characteristics of the thickness in the width direction of the glass plate according to the embodiment.

[0025] Fig. 7A This is a diagram showing an example of the characteristics of the time change of the plate thickness of the glass plate according to the embodiment.

[0026] Figure 7B This is a diagram showing an example of the characteristics of the time change of the plate thickness of the glass plate according to the embodiment.

[0027] Figure 7C This is a diagram showing an example of the characteristics of the time change of the plate thickness of the glass plate according to the embodiment.

[0028] Fig.7D This is a diagram showing an example of the characteristics of the time change of the plate thickness of the glass plate according to the embodiment.

[0029] Fig. 7E This is a diagram showing an example of the characteristics of the time change of the plate thickness of the glass plate according to the embodiment.

[0030] Figure 8 It is a figure which shows an example of the procedure of the process performed in the glass plate manufacturing apparatus which concerns on embodiment.

[0031] Fig. 9 It is a diagram showing another example of input and output information of a learned model according to the embodiment.

[0032] Fig.10 It is a cross-sectional view showing part of a glass sheet manufacturing apparatus using a float process.

[0033] Fig.11 It is a top view showing the molten metal tank.

[0034] Fig.12 It is a top view showing the molten metal tank.

[0035] Description of symbols

[0036] 1 ... glass plate manufacturing device, 11 ... melting furnace, 12, 810 ... molten metal tank, 13 ... slow cooling furnace, 14 ... cleaning inspection / cutting unit, 101 ... information processing device, 121 ... plate thickness detection unit, 122 ... manufacturing condition detection unit, 123 ... adjustment unit, 141 ... operator, 211 ... input unit, 212 ... output unit, 213 ... communication unit, 214 ... storage unit, 215 ... control unit, 231 ... operation unit, 251 ... display unit, 271 ... determination unit, 311, 311a ... learning completed model, 511 ... manufacturing condition, 531 ... temperature condition, 532 ... pressure condition, 533 ... flow channel control gate condition, 534 ... heater condition, 535 ... humidity condition, 536 ... plate thickness condition, 611 ... glass plate, 810a ... opening, 811a ... bottom, 811b ... top, 8 12… bottom body, 812a… storage tank, 813… bottom shell, 815… top shell, 818… flow channel control gate, 819… lip, 820-829… top roller, 824A… top roller body, 824B… rotating shaft, 830… heater, 840… control device, 2011, 2110, 2121-2122, 2131-2132… characteristics, AR… internal space, AR1… lower space, AR2… upper space, ARF… forming area, ARF1… upstream forming area, ARF2… downstream forming area, ARH… heater area, FL1-FL2… flow line, Gm… molten glass, GR… glass ribbon (glass sheet), M… molten metal, Ma… surface, PL1… dividing line, PL2… partition line, PL2a-PL2d… inner partition line DETAILED DESCRIPTION

[0037] Hereinafter, embodiments of the present disclosure will be described with reference to the drawings.

[0038] It should be noted that various information described in the following embodiments may also be referred to as data or the like.

[0039] [Glass sheet manufacturing device]

[0040] Figure 1 It is a figure which shows the schematic structural example of the glass plate manufacturing apparatus 1 which concerns on embodiment.

[0041] exist Figure 1 In the figure, for the convenience of explanation, a three-dimensional orthogonal coordinate system, namely an XYZ orthogonal coordinate system, is shown. In this embodiment, the positive direction of the Z axis represents the upward direction, and the negative direction of the Z axis represents the downward direction. In addition, the XY plane represents a plane horizontal to the up and down directions.

[0042] The glass sheet manufacturing apparatus 1 includes a melting furnace 11 , a molten metal tank 12 , a slow cooling furnace 13 , and a cleaning, inspection, and cutting unit 14 .

[0043] Moreover, the glass plate manufacturing apparatus 1 includes an information processing device 101 , a plate thickness detection unit 121 , a manufacturing condition detection unit 122 , and an adjustment unit 123 .

[0044] In addition, Figure 1 2 shows an operator 141 who operates the glass plate manufacturing apparatus 1. The operator 141 is, for example, a person. In addition, the operator may be referred to as a user or the like instead of the operator.

[0045] Here, in Figure 1 In the example, in order to make the figure easier to observe, the plate thickness detection unit 121, the manufacturing condition detection unit 122 and the adjustment unit 123 are illustrated in a manner separated from the glass plate manufacturing process part (melting furnace 11, molten metal tank 12, slow cooling furnace 13 and cleaning inspection / cutting unit 14), but for example, one or more of the plate thickness detection unit 121, the manufacturing condition detection unit 122 and the adjustment unit 123 may also be integrated with the glass plate manufacturing process part.

[0046] In the melting furnace 11, a melting process is performed.

[0047] In the melting step, glass raw materials prepared by mixing a plurality of raw materials are melted to obtain molten glass. After the glass raw materials are put into the melting furnace 11, they are melted by heating methods such as radiation heat of flames ejected from burners and electric melting to become molten glass.

[0048] In the molten metal tank 12, the forming process is performed.

[0049] In the forming process, the molten glass obtained in the melting process is continuously supplied to the molten tin in the bath, and the molten glass is flowed and formed on the molten tin to obtain a glass sheet (so-called glass ribbon) as a plate of glass. The glass sheet flows in a predetermined direction and is cooled, and is pulled up from the molten tin.

[0050] In the slow cooling furnace 13, a slow cooling process is performed.

[0051] In the slow cooling step, the glass sheet obtained in the forming step is slowly cooled in the slow cooling furnace 13. The glass sheet is slowly cooled while being horizontally conveyed on rollers from the entrance to the exit of the slow cooling furnace 13.

[0052] In the cleaning, inspection, and cutting section 14, a cleaning, inspection, and cutting process are performed.

[0053] In the cleaning and inspection step, the glass plate that has been slowly cooled in the slow cooling step is cleaned and inspected.

[0054] In the cutting process, the glass sheet that has been slowly cooled in the slow cooling process is cut into a predetermined size by a cutting machine. In the cutting process, both edges (so-called ears) in the width direction of the glass sheet are cut off. Both edges in the width direction of the glass sheet become thicker due to the influence of surface tension, etc., and are therefore cut off.

[0055] After the cutting process, the glass sheets are shipped.

[0056] The plate thickness detection unit 121 detects the plate thickness (plate thickness) of the glass plate. In this embodiment, the plate thickness is the plate thickness in the vertical direction (in Figure 1 In the example, the thickness is in the direction parallel to the Z axis).

[0057] Here, in this embodiment, the thickness detector 121 is disposed near the downstream exit of the glass sheet manufacturing apparatus 1 to detect the thickness of the glass sheet at this position. This position is downstream of the slow cooling furnace 13 and near the cleaning inspection / cutting unit 14.

[0058] The plate thickness detection unit 121 may also be configured to detect the direction of conveyance of the glass plate (in the direction of conveyance of the glass plate). Figure 1 In the example, the width direction (in the direction parallel to the X-axis, the direction from the negative side to the positive side of the X-axis) is perpendicular to the X-axis. Figure 1 In the example of the Y axis, the plate thickness is detected at a plurality of positions. These plurality of positions may be arranged at a constant interval in the width direction, for example, or may be arranged at other intervals.

[0059] In the present embodiment, the portion of the glass plate to be detected by the plate thickness detection unit 121 is a state where a predetermined time has passed after the glass plate has passed through the molten metal tank 12. The predetermined time is, for example, about 30 minutes, but may be other time periods.

[0060] The manufacturing condition detection unit 122 detects predetermined manufacturing conditions.

[0061] Here, in the present embodiment, the manufacturing condition detection unit 122 detects the manufacturing conditions in the molten metal tank 12 .

[0062] In addition, as another example, the manufacturing condition detection unit 122 may detect the manufacturing conditions at a location other than the molten metal tank 12 in the glass sheet manufacturing apparatus 1 .

[0063] The adjustment unit 123 has a function of adjusting manufacturing conditions.

[0064] Here, in the present embodiment, the adjustment unit 123 has a function of adjusting the manufacturing conditions in the molten metal tank 12 .

[0065] In addition, as another example, the adjustment unit 123 may have a function of adjusting the production conditions at a location other than the molten metal tank 12 in the glass sheet manufacturing apparatus 1 .

[0066] As an example, the adjustment unit 123 may include an operation unit operated by the operator 141 and adjust the manufacturing conditions according to the contents of the operation received by the operation unit.

[0067] As another example, the adjustment unit 123 may adjust the manufacturing conditions based on an instruction input from an external device (in this embodiment, the information processing device 101).

[0068] [Information processing device]

[0069] Figure 2 It is a diagram showing a configuration example of an information processing device 101 according to an embodiment.

[0070] The information processing device 101 is constituted by, for example, a computer.

[0071] The information processing device 101 includes an input unit 211 , an output unit 212 , a communication unit 213 , a storage unit 214 , and a control unit 215 .

[0072] The input unit 211 includes an operation unit 231 .

[0073] The output unit 212 includes a display unit 251 .

[0074] The control unit 215 includes a determination unit 271 .

[0075] The input unit 211 has a function of inputting information from the outside. For example, the operation unit 231 has a function of inputting information according to an operation performed by the operator 141.

[0076] The output unit 212 has a function of outputting information to the outside. For example, the display unit 251 has a function of outputting information of a display object on a screen.

[0077] The communication unit 213 has a function of communicating with an external device.

[0078] The storage unit 214 stores various information. In the present embodiment, the storage unit 214 stores the learned model 311 and the like.

[0079] The learned model 311 may be set in advance in the storage unit 214 , or may be set externally at an arbitrary timing.

[0080] Here, as the learning model of the learned model 311, any machine learning model may be used, for example, a neural network model may be used.

[0081] The control unit 215 performs various processes and controls.

[0082] The determination unit 271 inputs predetermined information to be determined into the learned model 311 , and acquires an output from the learned model 311 as information on the determination result.

[0083] The control unit 215 may output the information of the obtained determination result through the output unit 212, and as an example, may output the display content related to the information of the obtained determination result to the screen display of the display unit 251. In this case, for example, the operator 141 may view the display content and operate the adjustment unit 123 according to the display content, thereby controlling (manually controlled by the operator 141) the adjustment of the manufacturing conditions by the adjustment unit 123.

[0084] In addition, the control unit 215 may have a function (automatic control function) of controlling the adjustment of the manufacturing conditions by the adjustment unit 123 based on the information of the acquired determination result, for example.

[0085] As a specific example, the control unit 215 may also output the information of the obtained determination result or other information based on the information to the adjustment unit 123 as a control signal of the adjustment unit 123 through the communication unit 213, thereby controlling the adjustment of the manufacturing conditions by the adjustment unit 123. In this case, the adjustment unit 123 adjusts the manufacturing conditions according to the control signal received from the control unit 215.

[0086] Here, the determination unit 271 may, for example, adopt the output (estimation result) from the learned model 311 as it is as the determination result generated by the determination unit 271, or may perform further determination or analysis based on the output (estimation result) from the learned model 311 and use the result as the determination result generated by the determination unit 271.

[0087] In addition, the words "predict" or "forecast" may be used instead of "predict".

[0088] In the present embodiment, the information processing device 101 has a function of communicating with the plate thickness detection unit 121 by using the communication unit 213. Thus, the information processing device 101 can receive and acquire information on the detection result of the plate thickness detected by the plate thickness detection unit 121.

[0089] In the present embodiment, information processing device 101 has a function of communicating with manufacturing condition detection unit 122 using communication unit 213. Thus, information processing device 101 can receive and acquire information on the detection result of the manufacturing condition detected by manufacturing condition detection unit 122.

[0090] In the present embodiment, the information processing device 101 has a function of communicating with the adjustment unit 123 using the communication unit 213. Thus, the information processing device 101 can control the adjustment of the manufacturing conditions by the adjustment unit 123 by outputting (sending) instructions to the adjustment unit 123.

[0091] Here, each communication may be, for example, wired communication or wireless communication.

[0092] It should be noted that, in this embodiment, a structure is shown in which the information processing device 101 communicates with the plate thickness detection unit 121, the manufacturing condition detection unit 122 and the adjustment unit 123 respectively, but the exchange of information between two different devices can also be carried out by any method, for example, it can also be carried out using a portable external storage medium.

[0093] In addition, in the present embodiment, a configuration is shown in which the information processing device 101 can control the adjustment unit 123 , but as another example, a configuration in which the information processing device 101 does not control the adjustment unit 123 may be used.

[0094] <Example of input and output information of a learned model>

[0095] Figure 3 1 is a diagram showing an example of input and output information of the learned model 311 according to the embodiment. Information on the detection results of manufacturing conditions is input as input information to the learned model 311. Information on the detection results of manufacturing conditions detected by the manufacturing condition detection unit 122 can be used as this information.

[0096] exist Figure 3 The example of FIG. 1 shows a case where information on the detection result of the plate thickness of the glass plate is not used as input information to the learned model 311 .

[0097] Here, as the information of the detection result of the manufacturing conditions, for example, one piece of information may be used, or a plurality of pieces of information may be used.

[0098] It should be noted that, as another example, the input information may be information processed based on the information of the manufacturing condition detection result. In other words, the input information may not be the information of the manufacturing condition detection result itself, but may be converted into other information.

[0099] Information on the estimated result of the plate thickness is output as output information from the learned model 311. This information is, for example, information on the plate thickness estimated to be detected by the plate thickness detection unit 121.

[0100] It should be noted that, as another example, the output information may be information processed based on the information of the estimated result of the plate thickness. In other words, the output information may not be the information of the estimated result of the plate thickness itself, but may be converted into other information.

[0101] As a specific example, the output information may be information indicating whether the estimated result of the plate thickness satisfies a predetermined condition. For example, the condition may be that the plate thickness (estimated result) is abnormal if it is not within a predetermined range (between a lower limit and an upper limit).

[0102] In addition, for example, estimated results of the plate thickness at multiple positions in the width direction of the glass plate may be obtained, and machine learning may be performed in such a manner that there is no position where the estimated result of the plate thickness in the width direction deviates significantly (for example, deviates by more than a specified threshold).

[0103] Here, in the present embodiment, the learned model 311 may be generated by any learning method.

[0104] As an example, information on the detection results of manufacturing conditions can be used as input information of the model before learning, and information on the plate thickness detected by the plate thickness detection unit 121 can be used as training data (true value data) relative to the output information of the model before learning, thereby generating a learned model 311 through machine learning.

[0105] In addition, in this embodiment, the case where the learned model 311 that has been learned before the determination by the determination unit 271 is used is shown, but as another example, a learning model in which the learned model content is successively updated (corrected) in a time series may be used (in this embodiment, this is also regarded as an example of the learned model 311). As such a learning model, for example, a state space model may be used.

[0106] As described above, as the learned model 311 , a learned model whose learning has been completed and whose model contents are fixed may be used, or a learned model whose model contents are further updated sequentially in a time series may be used.

[0107] [Example of manufacturing conditions]

[0108] Figure 4 It is a diagram showing an example of manufacturing conditions 511 according to the embodiment.

[0109] The manufacturing conditions 511 include, for example, a temperature condition 531 , a pressure condition 532 , a flow channel control gate condition 533 , a heater condition 534 , a humidity condition 535 , and a plate thickness condition 536 .

[0110] It should be noted that the manufacturing condition 511 may not be used. Figure 4 All the conditions shown may be used as any one or more conditions.

[0111] In addition, as the manufacturing conditions 511, conditions different from those exemplified in this embodiment may be used.

[0112] Here, Figure 4 The specific example of the manufacturing condition 511 shown is an example and is not necessarily limited to the specific example. For example, it is also possible to configure it without using Figure 4 The manufacturing conditions shown (this specific example) are different from those used.

[0113] Each manufacturing condition may be detected in real time by the manufacturing condition detection unit 122 during the manufacturing process of the glass sheet, for example.

[0114] <Temperature conditions>

[0115] As manufacturing conditions, conditions related to temperature (temperature conditions 531 ) may also be used.

[0116] As the temperature condition 531 , for example, the temperature inside the molten metal tank 12 may be used, or another temperature may be used.

[0117] For example, the temperature of one location may be used as the temperature condition 531, or the temperatures of a plurality of different locations may be used. The plurality of different locations may be distributed in the conveying direction of the glass plate (in this embodiment, the direction parallel to the X-axis), or in the width direction of the glass plate (in this embodiment, the direction parallel to the Y-axis), or in both directions.

[0118] In addition, as the temperature condition 531, for example, a temperature at one time may be used for each part, or temperatures at a plurality of different times may be used.

[0119] It should be noted that the temperature of the predetermined location may be detected by, for example, a temperature sensor provided at the predetermined location (or in the vicinity thereof).

[0120] Alternatively, a value obtained by averaging the temperatures at a plurality of locations (spatial average value) may be used, or a value obtained by averaging the temperatures at a plurality of times (temporal average value) may be used.

[0121] <Pressure conditions>

[0122] As manufacturing conditions, conditions related to pressure (pressure conditions 532) may also be used.

[0123] As the pressure condition 532 , for example, the pressure inside the molten metal tank 12 may be used, or other pressures may be used.

[0124] As the pressure condition 532, for example, pressure at one location may be used, or pressure at multiple different locations may be used. The multiple different locations may be distributed in the conveying direction of the glass plate (in this embodiment, the direction parallel to the X-axis), or in the width direction of the glass plate (in this embodiment, the direction parallel to the Y-axis), or in both directions.

[0125] In addition, as the pressure condition 532, for example, the pressure at one time may be used for each part, or the pressure at a plurality of different times may be used.

[0126] It should be noted that the pressure at the predetermined location may be detected by, for example, a pressure sensor provided at the predetermined location (or in the vicinity thereof).

[0127] Alternatively, a value obtained by averaging the pressures at a plurality of locations (spatial average value) may be used, or a value obtained by averaging the pressures at a plurality of times (temporal average value) may be used.

[0128] <Flow channel control gate conditions>

[0129] As the manufacturing conditions, conditions related to the flow path control gate (flow path control gate conditions 533) may be used.

[0130] In the present embodiment, the flow channel control gate is a portion that controls the flow rate of the molten glass supplied from the upstream side of the molten metal tank 12 .

[0131] As the flow path control gate condition 533 , for example, a value indicating the degree of opening and closing of the flow path control gate (for example, an angle) may be used, or other values ​​may be used.

[0132] In addition, as the flow path control gate condition 533, for example, a value at one time may be used, or values ​​at a plurality of different time points may be used.

[0133] It should be noted that the value indicating the degree of opening and closing of the flow path control gate (for example, the angle) may be determined by, for example, a function of controlling the flow path control gate (for example, the function of the adjustment unit 123 ).

[0134] Alternatively, a value obtained by averaging values ​​at a plurality of times (time average value) may be used.

[0135] <Heater Conditions>

[0136] As the manufacturing conditions, conditions related to the heater (heater conditions 534 ) may also be used.

[0137] In the present embodiment, the heater is a part that adjusts the temperature inside the molten metal tank 12. A plurality of heaters may be provided as the heater.

[0138] As the heater condition 534 , for example, a value indicating a heating condition by the heater (for example, the temperature of the heater) may be used, or other values ​​may be used.

[0139] For example, a value related to a heater at one location may be used as the heater condition 534, or values ​​related to heaters at multiple different locations may be used. The multiple different locations may be distributed in the conveying direction of the glass plate (in this embodiment, a direction parallel to the X-axis), or in the width direction of the glass plate (in this embodiment, a direction parallel to the Y-axis), or in both directions.

[0140] In addition, as the heater condition 534, for example, a value at one time may be used for each portion, or values ​​at a plurality of different time may be used.

[0141] It should be noted that the value indicating the heating status by the heater may be determined by, for example, a function for controlling the heater (for example, the function of the adjustment unit 123 ).

[0142] Alternatively, a value obtained by averaging values ​​related to heaters at a plurality of locations (spatial average value) may be used, or a value obtained by averaging values ​​at a plurality of times (temporal average value) may be used.

[0143] <Humidity conditions>

[0144] As manufacturing conditions, conditions related to humidity (humidity conditions 535) may also be used.

[0145] As the humidity condition 535 , for example, the humidity inside the molten metal tank 12 may be used, or other humidity may be used.

[0146] For example, the humidity at one location may be used as the humidity condition 535, or the humidity at multiple different locations may be used. The multiple different locations may be distributed in the conveying direction of the glass plate (in this embodiment, the direction parallel to the X-axis), or in the width direction of the glass plate (in this embodiment, the direction parallel to the Y-axis), or in both directions.

[0147] In addition, as the humidity condition 535, for example, the humidity at one time may be used for each part, or the humidity at a plurality of different times may be used.

[0148] It should be noted that the humidity at the predetermined location may be detected by, for example, a humidity sensor provided at the predetermined location (or in the vicinity thereof).

[0149] Alternatively, a value obtained by averaging the humidity at a plurality of locations (spatial average value) may be used, or a value obtained by averaging the humidity at a plurality of times (temporal average value) may be used.

[0150] <Thickness Conditions>

[0151] As the manufacturing conditions, conditions related to the thickness of the glass plate (thickness conditions 536 ) may be used.

[0152] As the plate thickness condition 536 , for example, the plate thickness of a predetermined portion of the glass plate in a state of being present in the molten metal tank 12 may be used, or the plate thickness of the glass plate in another state may be used.

[0153] For example, the plate thickness of one location may be used as the plate thickness condition 536, or the plate thickness of a plurality of different locations may be used. The plurality of different locations may be distributed in the conveying direction of the glass plate (in this embodiment, the direction parallel to the X-axis), or in the width direction of the glass plate (in this embodiment, the direction parallel to the Y-axis), or in both directions.

[0154] In addition, as the plate thickness condition 536, for example, the plate thickness at one time may be used for each portion, or the plate thickness at a plurality of different times may be used.

[0155] It should be noted that the plate thickness at the predetermined location may be detected by, for example, a plate thickness sensor disposed at the predetermined location (or in the vicinity thereof). In the case where the predetermined location is the same as the location where the plate thickness is detected by the plate thickness detection unit 121, the function of the plate thickness detection unit 121 may also be used as the function of detecting the plate thickness at the predetermined location.

[0156] Alternatively, a value obtained by averaging the plate thickness at a plurality of locations (spatial average value) may be used, or a value obtained by averaging the plate thickness at a plurality of times (temporal average value) may be used.

[0157] Here, the temperature, pressure, and humidity may change due to disturbances corresponding to, for example, weather conditions or time periods of a day (for example, the rising and setting of the sun), and may be difficult to control to be constant.

[0158] Furthermore, the expansion state in the width direction during the manufacture of the glass sheet may change due to disturbance, and it may be difficult to control it to be constant.

[0159] It should be noted that in this embodiment, the flow path control gate and the heater are controllable.

[0160] <Example of what a glass plate looks like>

[0161] Figure 5 This is a diagram showing an example of the appearance of the glass plate 611 according to the embodiment.

[0162] exist Figure 5 For the convenience of explanation, the Figure 1 The XYZ orthogonal coordinate system shown is the same XYZ orthogonal coordinate system.

[0163] in addition, Figure 5 The shape of the glass plate 611 shown is a schematic shape for explanation.

[0164] The glass plate 611 is in the conveying direction (in Figure 5 In the example, the distribution is parallel to the X-axis) and the distribution is more distributed in the width direction (in the Figure 5 In the example above, the image expands in a direction parallel to the Y axis).

[0165] exist Figure 5 In the example of FIG. 1 , a case is shown in which the width of the glass plate 611 extends symmetrically with respect to the center in the width direction of the glass plate 611 .

[0166] exist Figure 5 In the example of FIG. 6 , six positions (position x0 to position x5) are shown on the glass plate 611 in the conveying direction of the glass plate 611.

[0167] For example, the glass component present at position x0 at time t0 moves to position x1 at time t1, and similarly moves to positions x0, x2, x3, x4, and x5 at times t2, t3, t4, and t5, respectively.

[0168] Here, among positions x0 to x5, the larger the number, the more downstream the position. Also, among times t0 to t5, the larger the number, the more advanced the time (that is, the later the time).

[0169] exist Figure 5 In the example of FIG. 6 , five positions (position y0, position yR1, position yR2, position yL1, and position yL2) are shown on the glass plate 611 in the width direction.

[0170] The position y0 indicates the center position of the glass plate 611 in the width direction.

[0171] The position yR1 and the position yL1 represent the positions symmetrically in the width direction with respect to the position y0 (in Figure 5 In the example of , the upward direction and the downward direction) are positions that are the same distance away from position y0.

[0172] Similarly, the position yR2 and the position yL2 represent the positions symmetrically in the width direction with respect to the position y0 (in Figure 5 In the example of , the upward direction and the downward direction) are positions that are the same distance away from position y0.

[0173] Here, the position yR2 and the position yL2 are separated from the position y0 by a larger distance than the position yR1 and the position yL1.

[0174] Figure 6 This is a diagram showing an example of the characteristics of the thickness in the width direction of the glass plate according to the embodiment.

[0175] It should be noted that Figure 6 The characteristics shown in the examples are illustrative and are not necessarily strict characteristics.

[0176] exist Figure 6 In the graph shown, the horizontal axis represents the width direction of the glass plate (in the present embodiment, the direction parallel to the Y axis), and the vertical axis represents the plate thickness.

[0177] exist Figure 6 The characteristic 2011 of the plate thickness in the width direction of the glass plate is shown.

[0178] The horizontal axis shows the center position h0 in the width direction of the glass plate, positions hR1 and hL1 symmetrical to the center position h0 , and positions hR2 and hL2 symmetrical to the center position h0 .

[0179] Here, the position hR1 is relative to the center position h0 in a predetermined direction (in Figure 6 In the example, the positive direction of the Y axis is a position away from the specified distance, and the position hR2 is a position further away. On the contrary, the position hL1 is a position in the direction opposite to the specified direction (in the positive direction of the Y axis) relative to the center position h0. Figure 6 In the example, the negative direction of the Y axis) is away from the position of a specified distance, and the position hL2 is further away.

[0180] For example, the thickness of the glass plate to be detected by the plate thickness detection unit 121 may have the following characteristics: Figure 6 Features shown.

[0181] In addition, for example, as in the metal bath 12, at a position upstream of the thickness detection unit 121, the characteristics of the thickness of the glass plate may be as follows: Figure 6 Features shown.

[0182] but, Figure 6The characteristic 2011 shown is an example for explanation, and actually, the characteristic of the plate thickness of the glass plate may change as the glass plate proceeds from upstream to downstream.

[0183] Figure 7A to Figure 7E Each of them is a diagram showing an example of the characteristics of the time change of the plate thickness of the glass plate according to the embodiment.

[0184] exist Figure 7A to Figure 7E In each of the graphs shown, the horizontal axis represents time (moment) and the vertical axis represents plate thickness.

[0185] exist Fig. 7A It is shown in Figure 6 Characteristic 2110 of the time variation of the thickness of the glass sheet at position h0 is shown.

[0186] exist Figure 7B It is shown in Figure 6 The characteristic 2121 of the time variation of the plate thickness of the glass plate at the position hR1 is shown.

[0187] exist Figure 7C It is shown in Figure 6 A characteristic 2122 of the time variation of the thickness of the glass sheet at the position hR2 is shown.

[0188] exist Fig.7D It is shown in Figure 6 A characteristic 2131 of the time variation of the thickness of the glass plate at the position hL1 is shown.

[0189] exist Fig. 7E It is shown in Figure 6 A characteristic 2132 of the time variation of the thickness of the glass sheet at the position hL2 is shown.

[0190] For example, Figure 7A to Figure 7E As shown, the plate thickness may be estimated at a plurality of positions in the width direction of the glass plate (determination based on the learned model 311 ).

[0191] In addition, although five positions in the width direction of the glass plate are shown in this example, any number, such as 10, may be used as the number of the plurality of positions in the width direction of the glass plate.

[0192] Figure 8 It is a figure which shows an example of the procedure of the process performed in the glass plate manufacturing apparatus 1 which concerns on embodiment.

[0193] (Step S1)

[0194] In the glass plate manufacturing apparatus 1, the determination unit 271 acquires input information (in the present embodiment, information on the detection result of the manufacturing conditions) to the learned model 311. Then, the process proceeds to step S2.

[0195] (Step S2)

[0196] In the glass plate manufacturing apparatus 1, the determination unit 271 inputs the acquired input information to the learned model 311. Then, the process proceeds to step S3.

[0197] (Step S3)

[0198] In the glass plate manufacturing apparatus 1, the determination unit 271 acquires output information (in the present embodiment, information on the estimation result of the plate thickness) from the learned model 311. Then, the process proceeds to step S4.

[0199] (Step S4)

[0200] In the glass plate manufacturing apparatus 1, the control part 215 performs control of adjustment of a manufacturing condition based on the output information acquired by the determination part 271. Then, it transfers to the process of step S5.

[0201] (Step S5)

[0202] In the glass plate manufacturing apparatus 1, the control unit 215 determines whether to end the processing of this flow.

[0203] As a result of this determination, when the control unit 215 determines that the processing of this flow is to be terminated (step S5 : Yes), the processing of this flow is terminated.

[0204] On the other hand, as a result of this determination, when the control unit 215 determines not to terminate the processing of this flow (step S5 : No), the process proceeds to the process of step S1 .

[0205] Here, in Figure 8 In the example, for convenience of explanation, the case where the control unit 215 determines whether to end the processing of this flow in the process of step S5 is shown, but any structure can also be used as a structure for determining whether to end the processing of this flow.

[0206] In addition, Figure 8 In the example of , a processing flow having a processing step (step S4) of adjusting the manufacturing conditions by the control adjustment unit 123 is shown, but as another example, a processing flow not including such a processing step (step S4) may also be executed.

[0207] <Other examples of input and output information of the learned model>

[0208] Fig. 9It is a diagram showing another example of input and output information of the learned model 311 a according to the embodiment.

[0209] Here, the learned model 311 a is a modified example of the learned model 311 .

[0210] Roughly speaking, the learned model 311 a is the same as the learned model 311 except that the input information is different. Here, the different points will be described in detail, and the detailed description of the same points will be omitted.

[0211] exist Fig. 9 In the example of , information on the detection result of the plate thickness of the glass plate is included in the input information to the learned model 311a.

[0212] Information on the detection results of the manufacturing conditions is input as input information to the learned model 311 a .

[0213] The input information may be various information, and in this example, includes information on the detection result of the plate thickness other than the estimation purpose.

[0214] Here, the plate thickness other than the estimation target plate thickness is a plate thickness other than the plate thickness that is the target plate thickness estimated by the learned model 311 a (that is, the plate thickness itself that is to be obtained as the estimation result).

[0215] For example, if the estimated target plate thickness is set as the plate thickness at a predetermined time in a predetermined position of the glass plate (at Figure 1 In the example of FIG. 1 , the thickness of the glass plate at the time when the specified portion of the glass plate is actually manufactured and detected by the thickness detection unit 121 is used as the thickness for purposes other than estimation. For example, the thickness of a different portion of the glass plate (a portion other than the specified portion) or the thickness of the same portion of the glass plate (the specified portion) but at different times (times other than the specified time) can be used.

[0216] The information of the estimated result of the plate thickness is outputted from the learned model 311a as output information. This information is, for example, information of the plate thickness estimated to be detected by the plate thickness detection unit 121.

[0217] It should be noted that other information (for example, information that has been converted) may be used as the input information and the output information.

[0218] Here, the learned model 311 a may be generated by any learning method.

[0219] In addition, in this example, the case where the learned model 311a that has been learned before the determination by the determination unit 271 is used is shown, but as another example, a learning model in which the learned model content is successively updated (corrected) in a time series may be used (in this embodiment, this is also regarded as an example of the learned model 311a). As such a learning model, for example, a state space model may be used.

[0220] As described above, as the learned model 311 a , a learned model whose learning has been completed and whose model contents are fixed may be used, or a learned model whose model contents are further updated sequentially in a time series may be used.

[0221] <Time Series Information of Plate Thickness During Production of Glass Plate>

[0222] The input information of the learned model 311 a may include time series information of the thickness of the glass sheet during manufacture.

[0223] Here, as the time series information of the plate thickness during the manufacture of the glass plate, for example, information on the plate thickness at a plurality of times of a portion (glass component) in the glass plate that is the target of estimation of the plate thickness may be used.

[0224] These multiple time points may be, for example, time points at constant intervals, or time points at other intervals.

[0225] The start time and end time of these multiple times may also be various.

[0226] As an example, the time when the manufacture of the glass component starts (or a time near it) may be used as the start time, and the time when the manufacture of the glass component ends (or a time near it) may be used as the end time.

[0227] As another example, we can also focus on a specific process in a part of the glass manufacturing process (for example, the manufacturing process in the molten metal tank 12), in which the time when the manufacturing of the glass component starts (or a time near it) is used as the start time, and the time when the manufacturing of the glass component ends (or a time near it) is used as the end time.

[0228] As a specific example, in a case where the glass component in a glass plate takes one hour to manufacture from the start of manufacturing to the end, the information of the thickness of the glass component every minute from the start of manufacturing to the end of manufacturing (that is, the information of the thickness at 60 moments) can also be used as the time series information of the thickness of the glass plate during manufacturing.

[0229] It should be noted that in the present embodiment, a situation is shown where machine learning is performed assuming that the various glass components in a glass plate flow in a straight line along the conveying direction of the glass plate (in the present embodiment, a direction parallel to the X-axis), but in fact, the various glass components may flow along a path having a portion curved in the width direction (a path that turns in the width direction), and therefore a structure for performing machine learning that takes such a path into consideration may also be used.

[0230] However, even if it is assumed that the various glass components in the glass plate flow in a straight line along the conveying direction of the glass plate (in this embodiment, the direction parallel to the X-axis), as in the present embodiment, by performing machine learning under this assumption, an estimated result with sufficient accuracy for practical use can be obtained as an output result from the learned model.

[0231] [Example of a heater in a float metal bath]

[0232] Reference Figure 10-11 An example of a heater in a float metal bath 810 will be described. In this example, a structure example in which the heater is divided is shown.

[0233] Figure 10-11 The structure of the heater in the molten metal tank 810 shown can also be applied to the molten metal tank 12 involved in the embodiment. In this case, Figure 1 The adjustment unit 123 shown can also be used to adjust the manufacturing conditions. Fig.10 Adjustment of the heating conditions of more than one of the plurality of heaters 830 shown.

[0234] In addition, about Figure 10-11 The structures other than the heater in the molten metal tank 810 shown may be partially or entirely applied to the molten metal tank 12 according to the embodiment.

[0235] Fig.10 It is a cross-sectional view showing part of the glass plate manufacturing apparatus 1 of the float process.

[0236] Fig.11 and Fig.12 Each of them is a top view showing a molten metal tank.

[0237] exist Fig.10 For the convenience of explanation, the Figure 1 The XYZ orthogonal coordinate system shown is the same XYZ orthogonal coordinate system.

[0238] Fig.10 The cross-sectional view shown is an XZ cross-sectional view.

[0239] It should be noted that Figure 1Likewise, the Z-axis direction is the up-down direction, the X-axis direction is the length direction of the molten metal tank 810 , and the Y-axis direction is the width direction of the molten metal tank 810 .

[0240] The length direction of the metal liquid tank 810 is Figure 10 to Figure 12 In this embodiment, the left-right direction is the conveying direction of the glass ribbon GR (glass sheet). Figure 10 to Figure 12 The up-down direction in φ is a width direction perpendicular to the conveyance direction of the glass ribbon GR.

[0241] The conveyance direction of the glass ribbon GR is a direction in which the glass ribbon GR is conveyed in a plan view.

[0242] The upstream and downstream sides refer to the conveying direction (X-axis direction) of the glass ribbon GR in the glass sheet manufacturing apparatus 1 , where the +X side (positive side of the X-axis) is the downstream side and the -X side (negative side of the X-axis) is the upstream side.

[0243] The inner side in the width direction (Y-axis direction) is the side where the center of the molten metal tank 810 in the width direction is located. The outer side in the width direction is the side opposite to the side where the center of the molten metal tank 810 in the width direction is located.

[0244] In the following description, unless otherwise specified, the width direction means the width direction of the molten metal tank 810 and the width direction of the glass ribbon GR, and the conveying direction means the conveying direction of the glass ribbon GR.

[0245] exist Fig.11 and Fig.12 In FIG. 8 , the top 811b is omitted. Fig.12 In the figure, the top rollers 820 to 829 are omitted.

[0246] The glass plate manufacturing apparatus 1 includes a molten metal tank 810 having an internal space AR. Fig.10 The opening 810a of the molten metal tank 810 is shown in FIG.

[0247] The molten metal tank 810 is a device for forming the glass ribbon GR in the interior of the internal space AR. A melting furnace ( Figure 1 The melting furnace 11 shown in FIG. 1 supplies molten glass Gm from the upstream side to the surface Ma of the molten metal M accumulated in the metal tank 810 via the lip 819. The molten glass Gm is continuously supplied to the surface Ma of the molten metal M with the flow rate controlled by the flow channel control gate 818.

[0248] A slow cooling furnace ( Figure 1The slow cooling furnace cools the glass ribbon GR formed in the molten metal tank 810. The slow cooling furnace is provided with a plurality of conveying rollers, for example. The glass ribbon GR is conveyed from the upstream side (-X side) to the downstream side by being pulled toward the downstream side by the conveying rollers of the slow cooling furnace.

[0249] That is, the glass sheet manufacturing apparatus 1 continuously supplies molten glass Gm onto the surface Ma of the molten metal M stored in the molten metal tank 810 to form a glass ribbon GR, and conveys the glass ribbon GR in the conveyance direction (X-axis direction).

[0250] The metal liquid tank 810 has a bottom 811a, a top 811b, a plurality of heaters 830 and a plurality of control devices 840. Fig.11 As shown, the molten metal tank 810 has a plurality of top rollers 820 - 829 .

[0251] like Fig.10 As shown, the molten metal M is accumulated at the bottom 811a. The top 811b covers the upper side (+Z side) of the bottom 811a. A plurality of heaters 830 are provided at the top 811b. Each heater 830 is controlled by a control device 840 and can heat the glass ribbon GR formed on the molten metal M.

[0252] Hereinafter, each part of the molten metal tank 810 will be described in detail.

[0253] <Bottom>

[0254] The bottom portion 811 a has a bottom body 812 and a bottom housing 813 .

[0255] A storage groove 812a that is recessed toward the lower side (-Z side) is formed on the upper side (+Z side) of the bottom body 812. Molten metal M is stored in the storage groove 812a. The molten metal M is, for example, molten tin or molten tin alloy. The material of the bottom body 812 is, for example, clay bricks.

[0256] The bottom shell 813 covers the outer side of the bottom body 812. The bottom shell 813 is made of steel, for example.

[0257] <Top>

[0258] The top portion 811 b is disposed on the upper side (+Z side) of the bottom portion 811 a . The top portion 811 b includes a top outer shell 815 and a top brick layer 816 .

[0259] The top shell 815 is suspended from an upper structure (not shown) such as a beam of a building in which the molten metal tank 810 is installed. The top shell 815 is a box-shaped structure opened to the lower side (-Z side). A gas inlet (not shown) is provided in the top shell 815. The top shell 815 is made of steel, for example.

[0260] A side wall (not shown) is fixed to the inner side of the lower side (-Z side) of the top casing 815. The material of the side wall is, for example, thermal insulation brick or sillimanite.

[0261] The top brick layer 816 is provided on the inner side of the side wall. The top brick layer 816 is formed by placing substantially rectangular parallelepiped bricks called PBA on a frame assembled in a lattice shape (not shown). The frame assembled in a lattice shape is, for example, suspended from the top surface of the inner side of the top housing 815. Thus, the top brick layer 816 is arranged at a desired height.

[0262] The top brick layer 816 divides the inner space AR of the molten metal tank 810 into a lower space AR1 and an upper space AR2.

[0263] The lower space AR1 is a space located between the bottom 811a and the top 811b in the internal space AR of the molten metal tank 810. The molten metal M and the glass ribbon GR formed on the molten metal M are in contact with the lower space AR1.

[0264] The upper space AR2 is a space located inside the top 811b of the internal space AR of the molten metal tank 810. The upper space AR2 is located above the top brick layer 816 and the side wall (+Z side). The upper space AR2 communicates with the lower space AR1 via a through hole provided in the top brick layer 816.

[0265] A side seal (not shown) is provided between the bottom 811a and the top 811b in the vertical direction. The side seal seals the gap between the bottom 811a and the top 811b in the vertical direction. Thus, a substantially closed internal space AR surrounded by the bottom 811a, the top 811b and the side seal is formed.

[0266] <Heater>

[0267] The plurality of heaters 830 are disposed above the bottom 811a. In the molten metal tank 810, the plurality of heaters 830 are disposed, for example, in the number of several thousand. The plurality of heaters 830 extend in the up-down direction (Z-axis direction). The plurality of heaters 830 are disposed in a manner penetrating the top brick layer 816. The plurality of heaters 830 are, for example, in the shape of a solid cylinder. The material of the plurality of heaters 830 is, for example, silicon carbide (SiC).

[0268] like Fig.12 As shown, when observing the molten metal tank 810 from above, the heater area ARH where the plurality of heaters 830 are arranged has a plurality of partitions divided along the conveying direction (X-axis direction) of the glass ribbon GR formed on the molten metal M and the width direction (Y-axis direction) perpendicular to the conveying direction. Fig.12In the example of FIG. 1 , the heater area ARH is divided into columns A to H along the conveying direction, and columns B to G are divided into five sections along the width direction.

[0269] A plurality of heaters 830 are provided in each sub-area. The plurality of heaters 830 provided in one sub-area are connected to the same control device 840. That is, the plurality of heaters 830 provided in one sub-area are collectively controlled by one control device 840. The plurality of heaters 830 in a sub-area are provided, for example, so that the amount of heat radiated from the plurality of heaters 830 toward the glass ribbon GR is substantially uniform in one sub-area.

[0270] It should be noted that the heater region ARH is a region located inside the storage groove 812a when viewed from above.

[0271] <Control device>

[0272] The control device 840 controls the heater 830. The control device 840 is provided for each partition of the heater region ARH. The control device 840 collectively controls the plurality of heaters 830 provided in the corresponding partition. The configuration of the control device 840 is not particularly limited as long as the outputs of the plurality of heaters 830 in one partition can be collectively adjusted.

[0273] <Top Roller>

[0274] like Fig.11 As shown, a plurality of top rollers 820 to 829 are arranged opposite to each other on both sides of the width direction (Y-axis direction) of the glass ribbon GR. The top rollers 820 to 829 prevent the width of the glass ribbon GR from being narrowed due to surface tension. The number of top rollers is appropriately set according to the forming conditions such as the type of glass or the target thickness. The number of top rollers is, for example, more than 4 pairs and less than 30 pairs, preferably more than 10 pairs and less than 30 pairs. Fig.11 In the example, five pairs of top rollers are provided, namely, top rollers 820, 825, top rollers 821, 826, top rollers 822, 827, top rollers 823, 828, and top rollers 824, 829. The number of top rollers provided tends to increase as the thickness of the float glass to be manufactured decreases.

[0275] It should be noted that the top rollers 820 to 829 have the same structure except that they are provided at different positions, and therefore, in the following description, only the top roller 824 may be described as a representative.

[0276] The top roller 824 is composed of a top roller body 824A that supports the end portion of the glass ribbon GR in the width direction (Y-axis direction) and a rotation shaft 824B connected to the top roller body 824A.

[0277] These top rollers 820 to 829, more precisely, the top roller body 824A is provided in the forming region ARF of the glass ribbon GR. The forming region ARF of the glass ribbon GR is a region where the viscosity of the glass ribbon GR becomes 10 4.5 dPa·s or more and 10 7.5 The forming region ARF corresponds to the region where the rows B to G are provided in the heater region ARH.

[0278] The forming region ARF includes an upstream forming region ARF1 and a downstream forming region ARF2.

[0279] The upstream forming region ARF1 is a region located on the upstream side (-X side) of the forming region ARF. The upstream forming region ARF1 is a region where the viscosity of the glass ribbon GR is 10 4.5 dPa·s or more and greater than 10 5.7 In this example, the upstream forming region ARF1 corresponds to the region where rows B to E are provided in the heater region ARH. In the upstream forming region ARF1, the thickness of the glass ribbon GR is adjusted by adjusting the output of the heater 830 in each section of row B to row E.

[0280] The downstream forming region ARF2 is a region located on the downstream side (+X side) of the forming region ARF. The downstream forming region ARF2 is a region where the viscosity of the glass ribbon GR becomes 10 5.7 dPa·s or more and 10 7.5 That is, the viscosity of the glass ribbon GR in the downstream forming region ARF2 is higher than the viscosity of the glass ribbon GR in the upstream forming region ARF1. In this example, the downstream forming region ARF2 corresponds to the region where the rows F and G are provided in the heater region ARH.

[0281] In the downstream forming area ARF2, the output of the heater 830 in each partition of the column F, G is adjusted so that the temperature of the glass ribbon GR after the thickness is adjusted in the upstream forming area ARF1 gradually decreases from the upstream side (-X side) to the downstream side (+X side).

[0282] It should be noted that in Fig.11 and Fig.12 , for convenience of explanation, a dividing line PL1 is shown that divides two adjacent columns in the conveying direction (X-axis direction). The dividing line PL1 is located substantially in the center between the heaters 830 adjacent to each other in the conveying direction. The dividing line PL1 is, for example, a straight line extending in the width direction.

[0283] In addition, Fig.11 and Fig.12In FIG. 1 , for convenience of explanation, a dividing line PL2 is shown that divides two adjacent partitions in the width direction (Y-axis direction). The dividing line PL2 is located approximately in the center between the heaters 830 adjacent to each other in the width direction of the glass ribbon GR. The dividing line PL2 is, for example, a straight line extending in the conveying direction.

[0284] The dividing lines PL2 include internal dividing lines PL2a to PL2d.

[0285] It should be noted that in Fig.12 In FIG. 1 , for convenience of description, streamlines FL1 and FL2 of the glass ribbon GR are shown.

[0286] exist Fig.12 In the example of , the streamline FL1 has a portion that is curved so as to bulge toward one side (+Y side) in the width direction with respect to the conveying direction (X-axis direction).

[0287] In addition, Fig.12 In the example of FIG. 1 , the streamline FL2 has a portion that is curved so as to bulge toward the other side (−Y side) in the width direction with respect to the conveying direction (X-axis direction).

[0288] The plurality of heaters 830 may be arranged, for example, along the flow lines (for example, flow lines FL1 and FL2) of the glass ribbon GR, and may reduce (for example, minimize) the temperature difference in the direction of the flow lines.

[0289] Here, in Figure 10-11 When the structure of the heater in the metal liquid tank 810 shown in the figure is applied to the metal liquid tank 12 involved in the embodiment, for example, the function of the control device 840 can also be set in Figure 1 The adjustment section 123 is shown.

[0290] As described above, in the glass plate manufacturing apparatus 1 according to the present embodiment, the thickness of the glass plate (or information related thereto) finally manufactured can be estimated by a machine learning model using the manufacturing conditions of the glass plate during manufacturing as input. Thus, in the glass plate manufacturing apparatus 1, when the thickness of the glass plate after manufacturing deviates from a constant range (a range having a width, but also including the case where the width is a constant value that can be regarded as 0), the temperature is quickly adjusted, etc., so that the quality of the thickness of the glass plate after manufacturing can be maintained.

[0291] In addition, in the glass plate manufacturing device 1 involved in this embodiment, the manufacturing conditions of the glass plate can be adjusted based on the estimated results. For example, when it is estimated that a defect in plate thickness will occur, the temperature can be adjusted to correct the defect early, thereby minimizing the period of yield reduction.

[0292] Moreover, in the glass plate manufacturing apparatus 1 which concerns on this embodiment, the time change of the board thickness of a glass plate (or the information related thereto) can be estimated, and thereby, for example, the occurrence of a defect in board thickness can be continuously monitored in real time.

[0293] Moreover, in the glass plate manufacturing apparatus 1 which concerns on this embodiment, it is possible to continuously adjust the manufacturing conditions of a glass plate based on the result of this estimation, and it is possible to suppress the occurrence of a defect of a plate thickness for a long period of time.

[0294] As a structural example, in a glass plate manufacturing device 1, in an information processing device 101, a determination unit 271 inputs information on manufacturing conditions during the manufacturing of a glass plate that flows in a prescribed flow direction (conveying direction) and is manufactured into a learned model 311 of machine learning, and obtains an output result from the learned model 311 as information related to an estimated result of the thickness of the glass plate.

[0295] Therefore, in the glass plate manufacturing apparatus 1, the plate thickness (or information related thereto) of the glass plate finally manufactured can be estimated by a machine-learned model (learned model 311) using the manufacturing conditions of the glass plate during manufacturing as input.

[0296] As one configuration example, in the glass plate manufacturing apparatus 1, in the information processing device 101, the information of the manufacturing conditions input to the learned model 311 includes information on the plate thickness of the glass plate when the glass plate is manufactured.

[0297] Therefore, in the glass plate manufacturing apparatus 1, by including information on the thickness of the glass plate during manufacturing as information input to the learned model 311, it is possible to improve the accuracy of the output (estimation result) from the learned model 311.

[0298] As a configuration example, in the glass plate manufacturing apparatus 1 , in the information processing apparatus 101 , information on the plate thickness of the glass plate during the manufacturing of the glass plate includes time series information on the plate thickness during the manufacturing of the glass plate.

[0299] Therefore, in the glass plate manufacturing apparatus 1, by including the time series information of the thickness of the glass plate during manufacturing as the information input to the learned model 311, the accuracy of the output (estimation result) from the learned model 311 can be further improved.

[0300] As one configuration example, in the glass plate manufacturing apparatus 1 , the adjustment unit 123 adjusts the manufacturing conditions based on the estimation result generated by the determination unit 271 .

[0301] Therefore, in the glass plate manufacturing apparatus 1, the plate thickness of the manufactured glass plate can be maintained within a constant range.

[0302] The plate thickness of the glass plate is an important quality in the production of the glass plate.

[0303] In the conventional manufacturing process, the thickness of the glass sheet is always monitored, and the thickness of the glass sheet is maintained constant by adjusting the temperature in the glass sheet forming device, etc. Here, the thickness monitored in the prior art is the thickness measured by the inspection machine after the temperature of the glass sheet drops downstream of the forming device. That is, conventionally, the thickness is detected at a position downstream of the forming device (for example, a position near the exit) to determine good products, and this information is fed back to the temperature adjustment in the molten metal tank, etc., thereby achieving constant thickness maintenance.

[0304] However, in such conventional monitoring methods, there is a large time difference (e.g., a time difference of 30 minutes) between the position where the temperature adjustment is performed and the position where the good product is judged. Therefore, if the temperature adjustment is incomplete for some reason, the correction may be greatly delayed. In other words, during the time from when the thickness defect is determined by the inspection machine to when the temperature adjustment is corrected, glass sheets with thickness defect continue to be manufactured, resulting in a situation where the yield rate decreases for a long time.

[0305] Therefore, in this embodiment, the thickness distribution of the glass sheet in the molten metal tank is estimated by using a machine learning model using parameters related to the manufacture of the glass sheet. Thus, in this embodiment, the defect in thickness distribution can be corrected at an early stage, and the period of yield reduction can be suppressed to a minimum.

[0306] It should be noted that, in the above embodiments, the case of manufacturing a glass sheet by a float process is described, but the technology involved in the above embodiments can also be applied to the case of manufacturing a glass sheet by other methods, for example, it can also be applied to the case of manufacturing a glass sheet by a melting process.

[0307] Furthermore, the technology according to the above-described embodiment can be applied to, for example, the manufacture of various types of glass plates, and as an example, can be applied to the manufacture of glass plates used for flat panel displays (FPDs).

[0308] It should be noted that the program for realizing the function of any structural part in any device described above can also be recorded in a computer-readable recording medium, so that the computer system reads the program and executes it. It should be noted that the "computer system" mentioned here includes hardware such as an operating system or peripheral devices. In addition, "computer-readable recording medium" refers to removable media such as floppy disks, optical magnetic disks, ROMs, CD (Compact Disc)-ROM (Read Only Memory), and storage devices such as hard disks built into the computer system. Moreover, the so-called "computer-readable recording medium" also includes a medium that keeps the program for a certain period of time, such as a volatile memory inside the computer system of a server or client when the program is sent via a network such as the Internet or a communication line such as a telephone line. The volatile memory may also be a RAM (Random Access Memory), for example. The recording medium may also be a non-temporary recording medium, for example.

[0309] In addition, the above-mentioned program can also be transmitted from a computer system storing the program in a storage device, etc. to another computer system via a transmission medium or a transmission wave in the transmission medium. Here, the "transmission medium" for transmitting the program refers to a medium having the function of transmitting information, such as a network such as the Internet or a communication line such as a telephone line.

[0310] In addition, the above-mentioned program may also be a program for realizing a part of the above-mentioned functions. Moreover, the above-mentioned program may also be a program that can realize the above-mentioned functions by combining the above-mentioned functions with a program already recorded in the computer system, i.e., a so-called differential file. The differential file may also be called a differential program.

[0311] In addition, the functions of any structural parts in any of the devices described above can also be implemented by a processor. For example, each process in the embodiment can also be implemented by a processor that acts based on information such as a program and a computer-readable recording medium that stores information such as a program. Here, with respect to the processor, for example, the functions of each part can be implemented by separate hardware, or the functions of each part can be implemented by integrated hardware. For example, it can also be that the processor includes hardware, and the hardware includes at least one of a circuit for processing digital signals and a circuit for processing analog signals. For example, the processor can also be composed of one or more circuit devices or one or more circuit elements mounted on a circuit substrate or both. As a circuit device, an IC (Integrated Circuit) or the like can also be used, and as a circuit element, a resistor or a capacitor or the like can also be used.

[0312] Here, the processor may be, for example, a CPU. However, the processor is not limited to the CPU, and various processors such as a GPU (Graphics Processing Unit) or a DSP (Digital Signal Processor) may be used. In addition, the processor may be, for example, a hardware circuit formed by an ASIC (Application Specific Integrated Circuit). In addition, the processor may be, for example, composed of a plurality of CPUs, or may be composed of a plurality of hardware circuits formed by ASICs. In addition, the processor may be, for example, composed of a combination of a plurality of CPUs and a plurality of hardware circuits formed by ASICs. In addition, the processor may include, for example, one or more of an amplifier circuit and a filter circuit for processing analog signals.

[0313] As mentioned above, although the embodiment of this disclosure is described in detail with reference to the drawings, the specific structure is not limited to this embodiment, and also includes the design etc. within the scope which does not deviate from the summary of this disclosure.

[0314] [Note]

[0315] [Structural Example 1] to [Structural Example 6] are shown.

[0316] [Structural Example 1]

[0317] An information processing device comprises a determination unit, wherein the determination unit inputs information on manufacturing conditions of a glass plate manufactured by flowing in a specified flow direction into a machine-learned learned model, and obtains an output result from the learned model as information related to an estimated result of the thickness of the glass plate.

[0318] [Structural Example 2]

[0319] According to the information processing device described in [Structural Example 1],

[0320] The information on the manufacturing conditions input to the learned model includes information on the thickness of the glass sheet when the glass sheet is manufactured.

[0321] [Structural Example 3]

[0322] According to the information processing device described in [Structural Example 2],

[0323] The information on the plate thickness of the glass plate when the glass plate is manufactured includes time series information on the plate thickness when the glass plate is manufactured.

[0324] [Structural Example 4]

[0325] A glass plate manufacturing device, comprising:

[0326] An information processing device including a determination unit that inputs information on manufacturing conditions of a glass sheet manufactured by flowing in a predetermined flow direction into a machine-learned learned model, and obtains an output result from the learned model as information related to an estimated result of a sheet thickness of the glass sheet; and

[0327] An adjustment unit adjusts the manufacturing condition based on the estimation result generated by the determination unit.

[0328] For example, it is possible to implement a method of processing performed in an information processing device.

[0329] [Structural Example 5]

[0330] An information processing method,

[0331] The determination unit inputs information on manufacturing conditions of a glass sheet manufactured by flowing in a predetermined flow direction into a machine-learned learned model, and obtains an output result from the learned model as information related to an estimated result of the thickness of the glass sheet.

[0332] For example, a method of processing performed in a glass plate manufacturing apparatus can be implemented.

[0333] [Structural Example 6]

[0334] A method for manufacturing a glass plate,

[0335] The determination unit inputs information on manufacturing conditions of a glass sheet manufactured by flowing in a predetermined flow direction into a machine-learned learned model, and obtains an output result from the learned model as information related to an estimated result of a sheet thickness of the glass sheet.

[0336] An adjustment unit adjusts the manufacturing condition based on the estimation result generated by the determination unit.

Claims

1. An information processing device, The information processing device includes a determination unit, which inputs information on manufacturing conditions during the manufacturing of a glass plate that flows in a specified flow direction and is manufactured into a learned model of machine learning, and obtains an output result from the learned model as information related to an estimated result of the thickness of the glass plate.

2. The information processing device according to claim 1, in, The information on the manufacturing conditions input to the learned model includes information on the thickness of the glass sheet when the glass sheet is manufactured.

3. The information processing device according to claim 2, in, The information on the plate thickness of the glass plate when the glass plate is manufactured includes time series information on the plate thickness when the glass plate is manufactured.

4. A glass sheet manufacturing device, comprising: An information processing device including a determination unit that inputs information on manufacturing conditions of a glass sheet manufactured by flowing in a predetermined flow direction into a machine-learned learned model, and obtains an output result from the learned model as information related to an estimated result of a sheet thickness of the glass sheet; and An adjustment unit adjusts the manufacturing condition based on the estimation result generated by the determination unit.

5. An information processing method, The determination unit inputs information on manufacturing conditions of a glass sheet manufactured by flowing in a predetermined flow direction into a machine-learned learned model, and obtains an output result from the learned model as information related to an estimated result of the thickness of the glass sheet.

6. A method for manufacturing a glass sheet, The determination unit inputs information on manufacturing conditions of a glass sheet manufactured by flowing in a predetermined flow direction into a machine-learned learned model, and obtains an output result from the learned model as information related to an estimated result of a sheet thickness of the glass sheet. An adjustment unit adjusts the manufacturing condition based on the estimation result generated by the determination unit.

Images