Sheet end detection device and sheet end detection method

Abstract

This sheet end detection device comprises: one or more magnetic field generation units that are installed in a space generated between two adjacent conveyance rolls and a path for passing a steel sheet therethrough, and that generate a magnetic field; a plurality of detection units that are installed side by side in the width direction of the sheet end detection device corresponding to the sheet width direction of the steel sheet, and that detect a signal corresponding to the strength of a magnetic field obtained by reflecting the influence of the steel sheet on the magnetic field generated by the magnetic field generation units; and an arithmetic processing unit that detects the position of a sheet end of the steel sheet. The arithmetic processing unit acquires the relationship between a change amount f of a signal detected by each of the detection units and a position x of each of the detection units in the width direction, approximates the acquired relationships by a function f(x), and detects the position x0 calculated by formula (1) as the position of the sheet end of the steel sheet.　(1): f(x0)=Lmin+α×(Lmax-Lmin) In the formula, Lmax represents the maximum value of the function f(x), Lmin represents the minimum value of the function f(x), and α represents a predetermined coefficient.

Description

Plate edge detection device and plate edge detection method

[0001] The present disclosure relates to a plate edge detection device and a plate edge detection method.

[0002] For example, a cooling device for cooling a steel sheet has a plurality of conveying rolls installed therein, and the steel sheet is conveyed through the cooling device while being placed on the plurality of conveying rolls. When the steel sheet is conveyed, the steel sheet may meander due to thermal expansion of the conveying rolls or a defect in the flat shape of the steel sheet. In order to suppress the meandering of the steel sheet, for example, conveying roll position control is performed to change the positions of the conveying rolls. In order to perform conveying roll position control, it is necessary to accurately detect the positions of the ends of the steel sheet in the sheet width direction (hereinafter referred to as "sheet ends").

[0003] Here, the following techniques have been proposed as techniques for detecting the positions of the plate ends of a steel plate: For example, Japanese Patent Laid-Open No. 2009-250898 (Patent Document 1) discloses a technique in which the surface of a coil around which a steel plate is wound is irradiated with light, an image of the coil surface is captured, and the positions of the plate ends of the coil are detected based on the brightness of the captured image.

[0004] Furthermore, Japanese Patent Laid-Open Publication No. 55-147373 (Patent Document 2) discloses a technology that uses a sensor having a ferrite with three protrusions on the top and bottom, a reference coil wound around the protrusion in the center of the upper side of the ferrite, and a detection coil wound around the protrusion in the center of the lower side of the ferrite, and detects the position of the plate end of a steel plate based on the change in inductance of the detection coil when a magnetic field is generated in the reference coil.

[0005] However, since the inside of a cooling device that cools a steel plate is an environment in which visible light and infrared light are blocked by water vapor, a large amount of water, etc., it is difficult to obtain an image containing enough brightness to detect the position of the plate edge of the steel plate in such an environment. Therefore, it is difficult to detect the position of the plate edge of a steel plate being transported inside the cooling device using the technology described in Patent Document 1.

[0006] In contrast, magnetic fields are not affected by water vapor or large amounts of water, so when detecting the position of the plate end of a steel plate based on the change in inductance of a detection coil when a magnetic field is generated in a reference coil, as in the technology described in Patent Document 2, it is possible to avoid being affected by water vapor or large amounts of water.

[0007] However, when a steel sheet is transported while placed on the transport rolls, the position of the sheet edge may change in the width direction of the steel sheet, and the steel sheet may also move up and down (i.e., the pass line of the steel sheet may fluctuate up and down). Here, the technology described in Patent Document 2 does not take into consideration that the inductance of the detection coil changes not only when the position of the sheet edge of the steel sheet changes in the width direction of the steel sheet, but also when the steel sheet is lifted up from the transport rolls. Therefore, a change in inductance obtained when the steel sheet is lifted up from the transport rolls is mistaken for a change in inductance due to a change in the position of the sheet edge of the steel sheet in the width direction of the steel sheet, and therefore the position of the sheet edge of the steel sheet cannot be correctly detected when the steel sheet is lifted up from the transport rolls.

[0008] Therefore, the present disclosure aims to provide a plate end detection device and a plate end detection method that can accurately detect the position of the plate end of a steel plate in an environment where visible light and infrared light are blocked by water vapor or large amounts of water, such as inside a cooling device that cools steel plates, without being affected by the up and down movement of the steel plate during transportation that may normally occur during manufacturing.

[0009] A first aspect of the present disclosure is a sheet edge detection device that detects a position of an edge of a steel sheet in a cooling device that cools the steel sheet, the sheet edge detection device comprising: one or more magnetic field generation units that are installed in a space generated between two adjacent transport rolls out of a plurality of transport rolls that transport the steel sheet and a path through which the steel sheet passes, and that generate a magnetic field; a plurality of detection units that are installed side by side in a width direction of the sheet edge detection device corresponding to the sheet width direction of the steel sheet, and that detect a signal corresponding to the strength of a magnetic field obtained by reflecting the influence of the steel sheet on the magnetic field generated by the magnetic field generation units; and a calculation processing unit that detects the position of the sheet edge of the steel sheet based on a change f in the signal detected by each of the detection units and a position x of each of the detection units in the width direction, and the calculation processing unit obtains a relationship between the change f in the signal detected by each of the detection units and the position x of each of the detection units in the width direction, and approximates the relationship with a function f(x), and calculates a position x calculated from the following formula (1): ０ is the position of the edge of the steel plate. ０ )=Lmin+α×(Lmax−Lmin) (1) where Lmax is the maximum value of the function f(x), Lmin is the minimum value of the function f(x), and α is a predetermined coefficient.

[0010] A second aspect of the present disclosure is a plate edge detection method for detecting the position of the plate edge of a steel plate in a cooling device that cools the steel plate, the method comprising: one or more magnetic field generation units that are installed in a space generated between two adjacent transport rolls out of a plurality of transport rolls that transport the steel plate and a path through which the steel plate passes, and that generate a magnetic field; a plurality of detection units that are installed side by side in the width direction of the plate edge detection device corresponding to the plate width direction of the steel plate, and that detect a signal corresponding to the strength of the magnetic field obtained by reflecting the influence of the steel plate on the magnetic field generated by the magnetic field generation units; and a detection unit that detects the position of the plate edge of the steel plate based on an amount of change f of the signal detected by each of the detection units and a position x of each of the detection units in the width direction. and a calculation processing unit that detects the position of the edge of the steel sheet, using a sheet edge detection device having a plurality of the detection units, and a calculation processing unit that detects the position of the edge of the steel sheet, based on a change f in the signal detected by each of the detection units and a position x of each of the detection units in the width direction, and the calculation processing unit obtains a relationship between the change f in the signal detected by each of the detection units and the position x of each of the detection units in the width direction, approximates the relationship by a function f(x), and calculates the position x by the following formula (1): ０ is detected as the position of the edge of the steel plate. ０ )=Lmin+α×(Lmax−Lmin) (1) where Lmax is the maximum value of the function f(x), Lmin is the minimum value of the function f(x), and α is a predetermined coefficient.

[0011] According to the present disclosure, a plate end detection device and a plate end detection method are provided that can accurately detect the position of the plate end of a steel plate in an environment where visible light and infrared light are blocked by water vapor or large amounts of water, such as inside a cooling device that cools steel plates, without being affected by the up and down movement of the steel plate during transportation that may normally occur during manufacturing.

[0012] FIG. 1 is a plan view showing an example of the overall configuration of a strip edge detection device according to a first embodiment of the present disclosure and a steel sheet transported by a plurality of transport rolls. FIG. 2 is a side view showing an example of the overall configuration of a strip edge detection device according to the first embodiment and a steel sheet. FIG. 3 is a perspective view showing an example of a plurality of sensors according to the first embodiment. FIG. 4 is a front view showing an example of the configuration of a sensor according to the first embodiment. FIG. 5 is a diagram showing an example of the relationship between the amount of change in inductance and the position of a sensor. FIG. 6 is a diagram showing an example of the relationship between the detection error of the strip edge position and the amount of lift of the test steel sheet when the test steel sheet is horizontal. FIG. 7 is a block diagram showing an example of the configuration of a processing device according to the first embodiment. FIG. 8 is a flowchart showing an example of the flow of strip edge detection processing according to the first embodiment. FIG. 9 is a diagram showing an example of the relationship between the amount of change in inductance and the position of a sensor when the test steel sheet is inclined. FIG. 10 is a diagram showing an example of the relationship between the detected strip edge position and the number of times the starting point is shifted by one for each sensor position. FIG. 11 is a perspective view showing an example of a sensor according to a fourth embodiment.

[0013] First Embodiment First, a first embodiment of the present disclosure will be described.

[0014] 1 and 2 show an example of the overall configuration of a strip edge detection device 10 according to a first embodiment of the present disclosure and a steel sheet 14 transported by a plurality of transport rolls 12. For example, a cooling device 16 that cools the steel sheet 14 has a plurality of horizontally arranged transport rolls 12 installed therein. The transport rolls 12 are arranged parallel to one another. The steel sheet 14 is transported through the cooling device 16 while being placed on the transport rolls 12. The X-axis direction indicates the width direction of the steel sheet 14, and the Y-axis direction indicates the transport direction of the steel sheet 14. Hereinafter, the transport direction of the steel sheet 14 will be referred to as the "transport direction," and the width direction of the steel sheet 14 will be referred to as the "strip width direction." The width direction of the strip edge detection device 10 will be referred to as the "device width direction." The device width direction corresponds to the strip width direction. The device width direction refers to the direction perpendicular to the transport direction when viewed from the normal direction of the surface of the steel sheet 14.

[0015] 1 and 2, the steel plate 14 is shown to be transported in a horizontal direction, but the transport direction is not limited to the horizontal direction, and the steel plate 14 may be transported in a direction other than the horizontal direction, such as a vertical direction. In the following, for convenience, the explanation will be continued assuming that the steel plate 14 is transported in a horizontal direction.

[0016] The plate edge detection device 10 is a device that detects the plate edge (i.e., plate edge 14A) of a steel plate 14 being transported within a cooling device 16, and has a plurality of sensors 18 and a processing device 20.

[0017] The multiple sensors 18 are installed side by side in the width direction of the device in the space between two adjacent transport rolls 12 among the multiple transport rolls 12 and the path through which the steel sheet 14 passes. The configuration of each sensor 18 will be described in detail later, but each sensor 18 (see FIGS. 3 and 4 ) is a non-contact magnetic sensor having a yoke 24, an excitation coil 26, and a detection coil 28. The sensor 18 is installed on a stand 22 so that the magnetic flux emitted from the yoke 24 reaches the steel sheet 14 when the steel sheet 14 is directly above the sensor 18. However, the sensor 18 is positioned at a certain distance from the steel sheet 14 in order to be heat-resistant and impact-resistant. The multiple sensors 18 include a sensor 18 located outside the steel sheet 14, even when a steel sheet 14 of the maximum width that can be transported through the cooling device 16 is transported. Furthermore, depending on the minimum width of the steel plate 14 that can be transported through the cooling device 16 (i.e., the steel plate to be measured by the plate edge detection device 10), the sensor 18 does not have to be installed near the center in the device width direction of the plate edge detection device 10. In the example shown in Figures 1 and 2, as an example, the multiple sensors 18 include a sensor 18 located inside the minimum width of the steel plate 14. The excitation coil 26 is an example of a magnetic field generation unit of the present disclosure, and the detection coil 28 is an example of a detection unit of the present disclosure.

[0018] The processing device 20 is electrically connected to each of the multiple sensors 18, and applies a voltage to the excitation coil 26 and detects the position of the plate end of the steel plate 14 based on a signal detected by the detection coil 28 in accordance with the strength of the magnetic field generated in the excitation coil 26.

[0019] Fig. 3 shows an example of a plurality of sensors 18 arranged in the device width direction, and Fig. 4 shows an example of the configuration of each sensor 18. The plurality of sensors 18 have the same configuration. Each sensor 18 has a yoke 24, an excitation coil 26, a detection coil 28, and a pair of bobbins 30. The yoke 24 is made of a magnetic material such as a ferrite core, and has a pair of core portions 32 and a connecting portion 34. The pair of core portions 32 extend vertically, and the connecting portion 34 connects the lower ends of the pair of core portions 32 to each other.

[0020] Each bobbin 30 is formed in a cylindrical shape. A core portion 32 is inserted into the inside of each bobbin 30, thereby attaching each bobbin 30 to the core portion 32. The excitation coil 26 is wound around one of the pair of core portions 32 via the bobbin 30, and the detection coil 28 is wound around the other of the pair of core portions 32 via the bobbin 30.

[0021] The yoke 24 is a member that functions as a ferromagnetic core material for the excitation coil 26 and the detection coil 28. Each yoke 24 is installed with a pair of core portions 32 aligned in the conveyance direction. The multiple yokes 24 are also arranged side by side in the device width direction. The detection coil 28 provided in each yoke 24 corresponds to the excitation coil 26 provided in the same sensor 18.

[0022] The sensor 18 may be configured such that the excitation coil 26 and the detection coil 28 are separate coils that are wound independently around each core portion 32, or a single coil may be wound continuously around each core portion 32 as the excitation coil 26 and the detection coil 28. The following description will continue using, as an example, a configuration in which the sensor 18 is configured such that a single coil is wound continuously around each core portion 32 as the excitation coil 26 and the detection coil 28.

[0023] When the steel sheet 14 is not present above the sensor 18, the magnetic flux emitted from the yoke 24 on which the excitation coil 26 is mounted is absorbed by the south pole of the yoke 24 on which the detection coil 28 of the same sensor 18 is mounted. On the other hand, when the steel sheet 14 is present above the sensor 18, eddy currents are generated due to the time-dependent change in the magnetic flux. The eddy currents generate a magnetic field that cancels out the magnetic flux emitted from the yoke 24 on which the excitation coil 26 is mounted. The strength of the magnetic field detected by the detection coil 28 changes when the spatial distribution of the magnetic flux is canceled out by the eddy currents generated by the presence of the conductive steel sheet 14. By utilizing this phenomenon, a signal detected by the detection coil 28 corresponding to the magnetic field strength obtained by reflecting the influence of the steel sheet 14 on the magnetic field generated by the excitation coil 26 can be acquired for each position of the sensor 18. The position of the edge of the steel sheet 14 can be detected based on the relationship between the change in the acquired signal and the position of the sensor 18. The signal corresponding to the magnetic field strength can be expressed by electromagnetic characteristics such as impedance (inductance), voltage, and current. In the following, inductance is used as an example of a signal corresponding to the magnetic field strength.

[0024] FIG. 5 shows an example of the relationship between the amount of change in inductance and the position of the sensor 18 obtained through experiment and analysis. In the experiment, a test steel plate 36 serving as a substitute for the steel plate 14 and multiple sensors 18 installed in the width direction of the device were used. In FIG. 5 , the test steel plate 36 used in the experiment and the positions at which the multiple sensors 18 were installed are shown on the horizontal axis of the graph. The vertical axis indicates a value corresponding to the amplitude of the signal detected by the detection coil 28. Specifically, it indicates the amount of change in inductance that changes depending on the relative position between the yoke 24 and the test steel plate 36. The first vertical axis, which is the vertical axis on the left side of FIG. 5 , indicates the experimental value, and the second vertical axis, which is the vertical axis on the right side of FIG. 5 , indicates the analytical value. The horizontal axis indicates the relative position of the sensor 18 installed in the width direction of the device. The first vertical axis and the second vertical axis are examples of the first axis of the present disclosure, and the horizontal axis is an example of the second axis of the present disclosure. The position of the sensor 18 is also an example of the position of the detection coil of the present disclosure. In this disclosure, in addition to the first embodiment, the second and third embodiments, etc. will, in principle, be described only on one side of the plate end detection device 10 in the device width direction for convenience, and a description of the opposite side of the plate end detection device 10 in the device width direction will be omitted.

[0025] In the experiment, multiple sensors 18 were installed in the width direction of the device, from the outside of the edge of the test steel plate 36 toward the inside of the test steel plate 36. At each plot position in the graph, an AC voltage was applied to the excitation coil 26 of each sensor 18, and a signal (inductance) was obtained from the detection coil 28 in accordance with the strength of the magnetic field obtained by reflecting the influence of the test steel plate 36 on the magnetic field generated in the excitation coil 26. The relationship between the change in inductance and the position of the sensor 18 was then graphed. The plots indicated by white circles represent actual measured values ​​obtained in the experiment, and graph line G1 is a line smoothly connecting the experimental values ​​indicated by white circles to facilitate understanding of the experimental results.

[0026] Meanwhile, in the analysis, the test steel plate 36 used in the experiment and the multiple sensors 18 were represented on a computer using software capable of simulating the strength and distribution of a magnetic field, and signals (inductance) detected by the detection coils 28 were obtained using the same method as in the experiment. Then, the relationship between the relative amount of change in inductance and the position of the sensor 18 was graphed. The plots indicated by black circles show the analytical values ​​obtained in the analysis, and graph line G2 is a line smoothly connecting the analytical values ​​indicated by black circles to make the analysis results easier to understand.

[0027] The sensor 18, located far enough away from the edge of the test steel plate 36 toward the outside of the test steel plate 36, can detect signals without the magnetic flux emitted from the yoke 24 being blocked by the test steel plate 36, so the values ​​on the vertical axis of the graph lines G1 and G2 remain unchanged and remain at the upper limit. This value on the vertical axis that remains unchanged can be considered an upper limit. In other words, the value that can be considered an upper limit here refers to the maximum value on the vertical axis that is saturated in the relationship between the amount of inductance change and the position of the sensor 18 shown in FIG. 5 . As the position of the sensor 18 approaches the edge of the test steel plate 36 from the outside, the amount of magnetic flux blocked by the test steel plate 36 gradually increases, so the values ​​on the vertical axis of the graph lines G1 and G2 decrease. As the position of the sensor 18 passes the edge of the test steel plate 36 and moves toward the inside of the test steel plate 36, the amount of magnetic flux blocked by the test steel plate 36 further increases, so the values ​​on the vertical axis of the graph lines G1 and G2 further decrease. For the sensor 18 positioned inside the test steel plate 36, the magnetic flux emitted from the yoke 24 is blocked by the test steel plate 36, so the values ​​on the vertical axis of the graph lines G1 and G2 remain at the lower limit. These values ​​on the vertical axis that remain at the lower limit can be considered to be the lower limit. In other words, the value that can be considered to be the lower limit here refers to the minimum value on the vertical axis that is saturated in the relationship between the amount of change in inductance and the position of the sensor 18 shown in Figure 5.

[0028] Here, the inventors considered that the results of the experiment and analysis (graph lines G1 and G2) could be approximated to a function. By approximating to a function, it is possible to detect values ​​that can be considered as upper and lower limits without being affected by noise. Examples of functions include a sigmoid function and a ramp function. Note that other functions that can represent a distribution involving saturation of maximum and minimum values ​​may also be used. The following description will continue using a sigmoid function as an example of a function. The sigmoid function is an example of a function in the present disclosure.

[0029] Furthermore, the inventors conducted multiple experiments and analyses by changing the position of the plate end of the test steel plate 36 in the device width direction, and as a result of detecting the plate end, they found that the position on the horizontal axis at which the range between a value considered to be the upper limit and a value considered to be the lower limit of the vertical axis value of the approximated sigmoid function is divided at a predetermined ratio (hereinafter referred to as "division ratio a:b") corresponds to the position of the plate end of the test steel plate 36. For example, the division ratio a:b is 3:7, but is not limited to this.

[0030] Furthermore, the inventors conducted evaluation tests by varying the vertical position of the test steel plate 36 while keeping the test steel plate 36 horizontal, assuming possible vertical movement of the steel plate 14 during transport during normal manufacturing. In this evaluation test, the position of the end of the test steel plate 36 in the device width direction was fixed, while the floating amount of the test steel plate 36 was varied. Tests were conducted to detect the end position of the test steel plate 36 five times for each floating amount. In each test, the position on the horizontal axis that divides the range between the upper and lower limits of the vertical axis value of the sigmoid function obtained in the test at a ratio of a:b was calculated as the end position of the test steel plate 36. The difference between the calculated end position and the reference position was defined as the detection error for the end position. The average, maximum, and minimum values ​​of the detection error for the end position obtained in the five tests were then calculated. Here, the floating amount is an amount that indicates the vertical distance between the pass line and the (steel plate 14 or) test steel plate 36, when the pass line is the path along which the (steel plate 14 or) test steel plate 36 is transported while in contact with the transport rolls (without bouncing).

[0031] 6 shows an example of the relationship between the detection error of the position of the plate edge of the test steel plate 36 and the floating amount of the test steel plate 36. As shown in FIG. 6, it was found that the detection error can be kept within an allowable range (for example, ±5 mm or less) as long as the range of vertical movement of the steel plate 14 during transportation that can be expected during normal manufacturing (for example, a floating amount of 40 mm or less). Furthermore, from this evaluation test, the inventors found that the above-mentioned division ratio a:b does not change substantially even if the steel plate 14 moves up and down while remaining horizontal within the range of vertical movement of the steel plate 14 during transportation that can be expected during normal manufacturing.

[0032] From the above, it has been found that for a steel plate 14 that moves up and down while remaining horizontal within the range of up and down movement that can be expected during transportation of the steel plate 14 during normal manufacturing, the plate end detection device 10 can be used to obtain the relationship between the change in signal obtained for each position of the sensor 18 and the position of the sensor 18, and the obtained relationship can be approximated using a sigmoid function.The position of the horizontal axis that divides the range between the value that can be considered the upper limit and the value that can be considered the lower limit at a division ratio of a:b can be derived for the vertical axis value of the approximated sigmoid function, and the derived position on the horizontal axis can be detected as the position of the plate end of the steel plate 14.

[0033] When the division ratio a:b is determined by experiment, it is expected that the division ratio a:b will vary slightly if the position of the plate end is changed in the width direction of the device for each experiment. Therefore, multiple experiments may be performed by changing the position of the plate end in the width direction of the device, and the average of the multiple division ratios a:b obtained for each experiment may be calculated. The calculated division ratio a:b may be used when actually detecting the position of the plate end of the steel plate 14. Similarly, when the division ratio a:b is determined by analysis, it is expected that the division ratio a:b will vary slightly if the position of the plate end is changed in the width direction of the device for each analysis. Therefore, it is also possible to perform multiple analyses by changing the position of the plate end in the width direction of the device, and the average of the multiple division ratios a:b obtained for each analysis may be calculated. The calculated division ratio a:b may be used when actually detecting the position of the plate end of the steel plate 14. Hereinafter, the division ratio a:b determined in advance by calculating the average of multiple division ratios a:b will be referred to as a "specific ratio."

[0034] Next, a specific configuration of the processing device 20 according to the first embodiment will be described. Fig. 7 shows an example of the configuration of the processing device 20. The processing device 20 is a device that performs various controls and various calculations related to the strip edge detection device 10, and is configured by a computer. The processing device 20 has a processor 42, a volatile memory 44, a non-volatile memory 46, an output circuit 48, and an input circuit 50. The processor 42, the volatile memory 44, the non-volatile memory 46, the input circuit 50, and the output circuit 48 are connected to each other so as to be able to communicate with each other via a bus 52 or the like.

[0035] The processor 42 includes, for example, a central processing unit (CPU) or a microprocessor unit (MPU). The volatile memory 44 includes, for example, random access memory (RAM) and temporarily stores programs and data as a working area. The non-volatile memory 46 includes, for example, read only memory (ROM), a hard disk drive (HDD), or a solid state drive (SSD), and stores various programs including an operating system and various data.

[0036] The non-volatile memory 46 stores a program for executing a process for detecting the position of the plate edge of the steel plate 14 (hereinafter referred to as "plate edge detection process"). The processor 42 reads the program from the non-volatile memory 46 and executes the program using the volatile memory 44 as a work area. The processor 42 controls the output circuit 48 and performs arithmetic processing based on signals input from the input circuit 50 in accordance with the program stored in the non-volatile memory 46.

[0037] The processing device 20 may have an electronic circuit such as a programmable logic device (PLD) or an application specific integrated circuit (ASIC) instead of or in addition to the processor 42. Some or all of the functions of the processor 42 may be realized by the electronic circuit such as the PLD or ASIC.

[0038] The excitation coil 26 of each sensor 18 is electrically connected to the output circuit 48. The output circuit 48 is controlled by the processor 42 to apply a voltage to each excitation coil 26. The detection coil 28 of each sensor 18 is electrically connected to the input circuit 50. When a signal output from the detection coil 28 is input to the input circuit 50, the input circuit 50 A / D converts the analog signal output from the detection coil 28 into a digital signal and outputs the digitized signal to the processor 42.

[0039] The processor 42 reads out a program for executing the strip edge detection process from the non-volatile memory 46, and deploys and executes the read program in the volatile memory 44, thereby functioning as each functional unit of the processing device 20. Specifically, the processor 42 functions as a voltage control unit 62 and an arithmetic processing unit 66.

[0040] The voltage control unit 62 is a functional unit that controls the output circuit 48 to apply voltage to the excitation coils 26 of the multiple yokes 24. When the voltage control unit 62 applies an AC voltage to the excitation coils 26, a signal corresponding to the strength of the magnetic field obtained by reflecting the influence of the steel plate 14 on the magnetic field generated in the excitation coil 26 is detected by the detection coil 28 provided in the same sensor 18.

[0041] The arithmetic processing unit 66 detects the position of the plate end of the steel sheet 14 from the relationship between the amount of change in the signals detected by all of the detection coils 28 and the positions of the sensors 18. Specifically, the arithmetic processing unit 66 acquires the signals detected by the detection coils 28 for each position of the sensors 18 in accordance with the strength of the magnetic field obtained by reflecting the influence of the steel sheet 14 on the magnetic field generated by the excitation coil 26. For example, the input circuit 50 is provided with multiple channels to which signals from the detection coils 28 are input. Each channel is assigned a position of each sensor 18. When acquiring a signal from each channel, the arithmetic processing unit 66 identifies the position of the sensor 18 corresponding to each signal based on the identification information of each channel. Then, the arithmetic processing unit 66 approximates the relationship between the amount of change in the signals acquired for each position of the sensors 18 and the positions of the sensors 18 using a sigmoid function, and detects, as the position of the plate end of the steel sheet 14, the position on the horizontal axis that divides the range between the value considered to be the upper limit and the value considered to be the lower limit of the vertical axis of the approximated sigmoid function at the predetermined specific ratio.

[0042] For example, the arithmetic processing unit 66 may detect the position of the plate end of the steel plate 14 by performing the following arithmetic processing. That is, the arithmetic processing unit 66 obtains the relationship between the amount of change f of the signal detected by each detection coil 28 and the position x of each detection coil 28 in the device width direction, approximates the obtained relationship with a sigmoid function f(x), and obtains the position x calculated from the following formula (1): ０ may be detected as the position of the plate end of the steel plate 14. ０) = Lmin + α × (Lmax - Lmin) (1) Here, Lmax is the maximum value of the sigmoid function f(x). The maximum value Lmax may be a value that can be considered the upper limit of the vertical axis values ​​of the above-mentioned sigmoid function, that is, for example, the maximum value of the vertical axis that is saturated in the relationship between the amount of change in inductance and the position of the sensor 18 shown in FIG. 5. Furthermore, Lmin is the minimum value of the sigmoid function f(x). The minimum value Lmin may be a value that can be considered the lower limit of the vertical axis values ​​of the above-mentioned sigmoid function, that is, for example, the minimum value of the vertical axis that is saturated in the relationship between the amount of change in inductance and the position of the sensor 18 shown in FIG. 5. Furthermore, α is a predetermined coefficient. The coefficient α is a ratio corresponding to the above-mentioned specific proportion, and is given by α = b / (a + b).

[0043] Furthermore, the coefficient α may be stored in advance in the volatile memory 44 or the non-volatile memory 46. The coefficient α may be stored in advance in the volatile memory 44 or the non-volatile memory 46 by a user of the plate edge detection device 10 inputting the value of α to the processing device 20. Then, the calculation processing unit 66 may acquire the coefficient α stored in advance in the volatile memory 44 or the non-volatile memory 46. The volatile memory 44 or the non-volatile memory 46 is an example of a "storage unit" in the present disclosure.

[0044] The coefficient α is a coefficient set based on a specific coefficient setting process using a test steel plate that serves as a substitute for the steel plate 14. The coefficient setting process uses the test steel plate to obtain the relationship between the amount of change f in the signal detected by each detection coil 28 and the position x of each detection coil 28 in the device width direction, approximates this relationship with a sigmoid function f(x), and then applies the actual position x of the plate end of the test steel plate to the sigmoid function f(x). ０ ' to calculate the value of the function f(x0'). Then, add the function f(x ０ α is calculated by substituting the maximum value Lmax and minimum value Lmin of the function f(x) as the coefficient. ０(Lmax-Lmin) / (Lmax-Lmin) (1') That is, using a test steel plate, the relationship between the amount of change f in the signal detected by each detection coil 28 and the position x of each detection coil 28 in the device width direction is obtained, and this relationship is approximated by a sigmoid function f(x), and the actual position x of the plate end of the test steel plate is calculated. ０ ' to enter the function f(x ０ Then, the range between the maximum and minimum values ​​of the sigmoid function f(x) is calculated by the function f(x ０ The method also includes calculating a division ratio a:b, which is the ratio when the test steel plate is divided by the value of the detection coils 28 (a:b'), obtaining multiple division ratios a:b obtained by changing the position of the plate edge of the test steel plate in the device width direction relative to the positions of the multiple detection coils 28, and setting the coefficient α based on the multiple division ratios a:b. For example, the coefficient α may be set by calculating the average value of the multiple division ratios a:b and then calculating the ratio b / (a+b) from the calculated division ratio a:b. For example, as described above, if the average value of the division ratio a:b is 3:7, α is set to 0.70. The value of α is a characteristic unique to each plate edge detection device 10. The value of α depends on the characteristics unique to the sensor 18 (such as magnetic characteristics), the distance between the sensor 18 and the steel plate 14 (including the test steel plate that serves as a substitute for the steel plate 14), and the type of signal used (e.g., impedance (inductance), voltage, current, etc.). Therefore, the value of α must be set during the coefficient setting process for each plate edge detection device 10. If necessary, the value of α may be set by periodically performing a coefficient setting process for each strip edge detection device 10. The value of α is often about 0.70, that is, in the range of 0.60 to 0.80.

[0045] 8 shows an example of the flow of the edge detection process executed by the processor 42 of the processing device 20. The edge detection method by the edge detection device 10 is executed by the processor 42 executing the edge detection process.

[0046] First, in step S10, the voltage control unit 62 controls the voltage applied to the excitation coils 26 of the multiple yokes 24. Specifically, the voltage control unit 62 selects a yoke 24 from the multiple yokes 24, and performs control to apply an AC voltage to the excitation coil 26 of the selected yoke 24.

[0047] Next, in step S12, when an AC voltage is applied to the excitation coil 26 of the corresponding yoke 24, the detection coil 28 detects a signal corresponding to the strength of the magnetic field obtained by reflecting the influence of the steel plate 14 on the magnetic field generated in the excitation coil 26. Step S12 is an example of a detection step in the present disclosure.

[0048] Next, in step S14, the arithmetic processing unit 66 detects the position of the plate end of the steel plate 14 based on the relationship between the amount of change in the signal detected by the detection coil 28 and the position of the sensor 18. Specifically, the arithmetic processing unit 66 acquires the signal detected by the detection coil 28 for each position of the sensor 18 in accordance with the strength of the magnetic field obtained by reflecting the influence of the steel plate 14 on the magnetic field generated by the excitation coil 26, and approximates the relationship between the amount of change in the acquired signal and the position of the sensor 18 using a sigmoid function. Then, the arithmetic processing unit 66 detects, as the plate end position of the steel plate 14, the position on the horizontal axis that divides the range between the value considered to be the upper limit and the value considered to be the lower limit of the vertical axis value of the approximated sigmoid function at the predetermined specific ratio described above. The arithmetic processing unit 66 may detect the plate end position of the steel plate 14 by performing the above-described arithmetic processing. After step S14, the plate end detection processing ends. Step S14 is an example of a arithmetic processing step of the present disclosure.

[0049] As described above, in the strip edge detection device 10 according to the first embodiment, the arithmetic processing unit 66 acquires, for each position of the sensor 18, a signal detected by the detection coil 28 in accordance with the strength of the magnetic field obtained by reflecting the influence of the steel sheet 14 on the magnetic field generated in the excitation coil 26, and approximates the relationship between the amount of change in the acquired signal and the position of the sensor 18 using a sigmoid function. Then, the arithmetic processing unit 66 detects, as the position of the strip edge of the steel sheet 14, the position on the horizontal axis that divides the range between the value considered to be the upper limit and the value considered to be the lower limit of the vertical axis value of the approximated sigmoid function at the above-mentioned predetermined specific ratio.

[0050] Here, as a result of investigations by the inventors, it was found that the above-mentioned specific ratio remains substantially unchanged even if the steel plate 14 moves up and down while remaining horizontal within the range of vertical movement that can be expected during transportation of the steel plate 14 in normal manufacturing (see FIG. 6). Therefore, even if the steel plate 14 moves up and down within the range that can be expected during transportation of the steel plate 14 in normal manufacturing, the position of the plate end of the steel plate 14 can be correctly detected without being affected by the vertical movement of the steel plate 14.

[0051] Furthermore, the sheet edge detection device 10 according to the first embodiment detects the position of the sheet edge of the steel sheet 14 using a magnetic field that is not affected by water vapor, large amounts of water, etc. Therefore, when detecting the position of the sheet edge of the steel sheet 14, it is possible to avoid being affected by water vapor, large amounts of water, etc.

[0052] In this way, with the plate end detection device 10 according to the first embodiment, the position of the plate end of the steel plate 14 can be correctly detected in an environment where visible light and infrared light are blocked by water vapor or a large amount of water, such as inside the cooling device 16 that cools the steel plate 14, without being affected by the up and down movement of the steel plate 14 during transportation that may normally occur during manufacturing.

[0053] 5 uses inductance as the detected signal, the relationship between the vertical and horizontal axes is such that the value on the vertical axis decreases as the value on the horizontal axis increases, but depending on the method of calculating the value on the vertical axis (including changing the detection target to voltage or current), it is possible that the value on the vertical axis increases as the value on the horizontal axis increases (i.e., the graph in FIG. 5 is displayed upside down). Even in this case, the calculation processing unit 66 may take the maximum value Lmax to be a value that can be regarded as the upper limit of the vertical axis values ​​of the sigmoid function, i.e., for example, the saturated maximum value on the vertical axis, and may take the minimum value Lmin to be a value that can be regarded as the lower limit of the vertical axis values ​​of the sigmoid function, i.e., the saturated minimum value on the vertical axis.

[0054] Second Embodiment Next, a second embodiment of the present disclosure will be described. Contents that overlap with those of the first embodiment will be omitted by using similar terms.

[0055] In the first embodiment described above, it is assumed that the steel plate 14 moves up and down while remaining horizontal, but it is assumed that during normal manufacturing, the steel plate 14 will be inclined with respect to the device width direction when viewed from the conveying direction. According to the first embodiment, even when the steel plate 14 is inclined, the position of the plate edge of the steel plate 14 can be detected with a certain degree of accuracy, but it is also assumed that higher detection accuracy is required. The second embodiment aims to improve the detection accuracy when detecting the position of the plate edge of the steel plate 14 compared to the first embodiment.

[0056] FIG. 9 shows an example of the relationship between the change in inductance and the position of the sensor 18 obtained by an experiment. In the experiment, inclined test steel plates 36A and 36B were used as test steel plates substituting for the steel plate 14. In FIG. 9, the inclined test steel plates 36A and 36B and multiple sensors 18 are shown corresponding to the horizontal axis of the graph. The test steel plate 36A is inclined so that the floating amount at the plate end position relative to the reference height of the pass line is 0 mm and the floating amount increases at a gradient of 8 / 120 toward the inside of the test steel plate 36A. The test steel plate 36B is inclined so that the floating amount at the plate end position relative to the reference height of the pass line is 30 mm and the floating amount decreases at a gradient of 8 / 120 toward the inside of the test steel plate 36B.

[0057] The height L1 from the reference height of the sensor 18 to the reference height of the pass line is set to 30 mm. The lift-off amount L indicates the distance from the reference height of the sensor 18 to the steel plate 14 or test steel plate 36 at a specific position in the width direction of the device. In other words, the lift-off amount L for each sensor 18 is defined as the value obtained by adding the value of the lift-off amount to the value of the height L1 from the reference height of the sensor 18 to the reference height of the pass line. The reference height of the sensor 18 may be the average value of the heights of multiple sensors 18. The reference height of the pass line may be the average value of the entire length of the pass line along which the steel plate 14 is transported.

[0058] The vertical axis indicates values ​​corresponding to the amplitude of the signal obtained by the experiment, specifically, the relative change in inductance detected by the multiple sensors 18 installed in the width direction of the device. The horizontal axis indicates the relative position of the sensor 18 when a predetermined reference position in the width direction of the device is set to 0. The reference position is the position of the plate end set in the experiment.

[0059] In the experiment, the above-mentioned plate edge detection device 10 was used, and an AC voltage was applied to the excitation coil 26 of each sensor 18, and a signal (inductance) was obtained from the detection coil 28 in accordance with the strength of the magnetic field obtained by reflecting the influence of the steel plate on the magnetic field generated in the excitation coil 26. Then, the relationship between the amount of change in inductance and the position of the sensor 18 was approximated using a sigmoid function.

[0060] Graph line A1 shows the inductance change and the position of the sensor 18 obtained for the test steel sheet 36A. Graph line A2 shows the inductance change and the position of the sensor 18 obtained for the test steel sheet that was horizontally positioned and had a lift-off amount L of 30 mm. Graph line A3 shows the inductance change and the position of the sensor 18 obtained for the test steel sheet that was horizontally positioned and had a lift-off amount L of 35 mm. Graph line A4 shows the inductance change and the position of the sensor 18 obtained for the test steel sheet that was horizontally positioned and had a lift-off amount L of 40 mm. Graph line B1 shows the inductance change and the position of the sensor 18 obtained for the test steel sheet 36B. Graph line B2 shows the inductance change and the position of the sensor 18 obtained for the test steel sheet that was horizontally positioned and had a lift-off amount L of 50 mm. Graph line B3 shows the inductance change and the position of the sensor 18 obtained for the test steel sheet that was horizontally positioned and had a lift-off amount L of 60 mm. The graph line B4 shows the change in inductance versus the position of the sensor 18 obtained for a test steel plate that is placed horizontally and has a lift-off amount L of 70 mm.

[0061] Here, the inventors compared graph line A1 with graph lines A2 to A4 and found that, in the case of the inclined test steel plate 36A, when the position of the sensor 18 is in the range of -70 mm to 40 mm, the amount of change in inductance changes in a similar manner to the case of a test steel plate that floats while remaining horizontal, but when the position of the sensor 18 is in the range of more than 40 mm, the amount of change in inductance deviates from the case of a test steel plate that floats while remaining horizontal.Furthermore, the inventors compared graph line B1 with graph lines B2 to B4 and found that, in the case of the inclined test steel plate 36B, when the position of the sensor 18 is in the range of -70 mm to 40 mm, the amount of change in inductance changes in a similar manner to the case of a test steel plate that floats while remaining horizontal, but when the position of the sensor 18 is in the range of more than 40 mm, the amount of change in inductance deviates from the case of a test steel plate that floats while remaining horizontal.

[0062] Based on the above findings, the inventors thought that in the case of an inclined steel plate 14, if data in the range R in which the amount of change in inductance changes in the same manner as in the case of a steel plate 14 floating while remaining horizontal is used (in other words, if data in the range in which the amount of change in inductance deviates from the case of a steel plate 14 floating while remaining horizontal is not used), the position of the end of the steel plate 14 can be detected while mitigating the effects of the inclination of the steel plate 14.

[0063] Specifically, the inventors considered that it would be sufficient to determine the position of the sensor 18 located at the outermost position in the width direction of the device among the points where the vertical axis value is considered to be the upper limit, in order from the outer position to the inner position, as the starting point, and the position of the sensor 18 corresponding to the point where the vertical axis value is the smallest among the points where the vertical axis value is considered to be the lower limit, where the vertical axis value is not affected even if the steel plate 14 is tilted (for example, the point corresponding to 40 mm in Figure 9), among the points where the vertical axis value is considered to be the lower limit, as the end point, and to detect the position of the plate end of the steel plate 14 based on the amount of change in inductance obtained from the sensors 18 present in the section from the start point to the end point out of the multiple sensors 18.

[0064] Based on the above concept, in the second embodiment, the arithmetic processing unit 66 operates as follows: That is, when the relationship between the amount of change in the signal corresponding to the strength of the magnetic field and the position of the sensor 18, which is shown in a graph (see, for example, FIG. 9 ) having a vertical axis corresponding to the amount of change in the signal acquired for each position of the sensor 18 and a horizontal axis corresponding to the position of the sensor 18, is approximated by a sigmoid function, the arithmetic processing unit 66 sets sections (for example, sections corresponding to range R in FIG. 9 ) starting from the position of the sensor 18 located outermost in the device width direction among points where the value on the vertical axis is a value that can be considered to be the upper limit, in order from the outside to the inside in the device width direction, and ending at the position of the sensor 18 corresponding to the smallest value on the vertical axis among points where the value on the vertical axis is a value that can be considered to be the lower limit, where the value on the vertical axis is not affected even if the steel sheet 14 is tilted, among points where the value on the vertical axis is a value that can be considered to be the lower limit. The arithmetic processing unit 66 then detects the position of the plate end of the steel plate 14 based on the relationship between the amount of change in the signal corresponding to the strength of the magnetic field obtained from the sensor 18 present in the set section and the position of the sensor 18. Specifically, the arithmetic processing unit 66 detects, as the plate end of the steel plate 14, the position on the horizontal axis that divides the range corresponding to the set section (for example, range R in FIG. 9 ) between a value that can be considered as the upper limit (for example, a value corresponding to −70 mm in FIG. 9 ) and a value that can be considered as the lower limit (for example, a value corresponding to 40 mm in FIG. 9 ) at the above-mentioned predetermined specific ratio on the vertical axis.

[0065] More specifically, the calculation processing unit 66 may detect the position of the plate end of the steel plate 14 by performing the following calculation process. That is, the calculation processing unit 66 may first predict the position of the plate end of the steel plate 14, and set the above-mentioned section (for example, the section corresponding to range R in FIG. 9 ) so as to include the position of the plate end of the steel plate 14. Then, the calculation processing unit 66 may detect the position of the plate end of the steel plate 14 based on the relationship between the amount of change f of the signal detected by each detection coil 28 in the set section and the position x of each detection coil.

[0066] When setting the above-mentioned interval (for example, the interval corresponding to range R in Figure 9), the calculation processing unit 66 may predict the position of the plate end of the steel plate 14 (for example, the position where the value on the horizontal axis in Figure 9 is 0) by performing the following prediction processing.

[0067] For example, the calculation processing unit 66 may approximate the relationship between the amount of change in the signal acquired for each position of the sensor 18 and the position of the sensor 18 using a sigmoid function, and detect the horizontal axis position that divides the range between a value that can be considered as the upper limit and a value that can be considered as the lower limit of the vertical axis value of the approximated sigmoid function at the above-mentioned predetermined specific ratio as the position of the plate end of the steel plate 14, thereby predicting the position of the plate end of the steel plate 14. In this way, when approximation using a sigmoid function (approximation in the second embodiment) is performed to predict the position of the plate end of the inclined steel plate 14, an approximation to predict the position of the plate end of the steel plate 14 (for example, the approximation in the first embodiment, which will be referred to as the first approximation for convenience) is performed in advance, and then the above-mentioned interval set in the second embodiment (first position x described later) is detected based on the position of the plate end of the steel plate 14 predicted based on the sigmoid function obtained by the first approximation. １ to the second position x ２ It is preferable to perform an approximation (the approximation in the second embodiment, which will be referred to as the second approximation for convenience) to accurately detect the position of the plate end of the steel plate 14 (even if the steel plate 14 is inclined) for the section from the end of the steel plate 14 to the end of the steel plate 14.

[0068] Furthermore, the calculation processing unit 66 may acquire the signal change amounts from the multiple sensors 18, determine whether or not the differential value of the signal change amount calculated for each adjacent sensor 18 in order from the outside to the inside in the device width direction exceeds a predetermined threshold, and predict the position of the sensor 18 located on the inside in the device width direction among the adjacent sensors 18 whose differential value is determined to exceed the threshold as the plate end position of the steel plate 14. With this prediction method, the first approximation described above is not necessary, and only the second approximation described above is required. Because there is no first approximation, the calculation time (operation time) in the calculation processing unit 66 can be shortened.

[0069] Furthermore, the calculation processing unit 66 may acquire the amount of change in signals from the multiple sensors 18, and predict the position on the horizontal axis that divides the range between a value that can be considered as the upper limit and a value that can be considered as the lower limit on the vertical axis into the above-mentioned predetermined specific ratio as the position of the plate end of the steel plate 14. This prediction method also does not require a first approximation, and only a second approximation is required, which can shorten the calculation time (operation time).

[0070] The length of the section is, for example, the length of the first position x defined below in a preliminary test using a test steel plate that serves as a substitute for the steel plate. １ and the second position x ２ and the first position x １ to the second position x ２ The distance may be set to a length corresponding to the distance to the

[0071] 1st position x １ is the position x of the plate end ０ The maximum value of the function f(x) may be a value that corresponds to the position of the sensor 18 located at the outermost position in the device width direction among values ​​that can be considered as upper limits, or may be the maximum value among values ​​that can be considered as upper limits.

[0072] Although it depends on the function used, the position corresponding to the maximum value (Lmax) or minimum value (Lmin) of the function f(x) is often located at a position infinitely far outward in the device width direction. １When determining the function f(x), the position of the outermost sensor 18 in the width direction of the device may be regarded as the position corresponding to the maximum or minimum value of the function f(x). Furthermore, in order to shorten the calculation time (operation time) by reducing the number of sensors to be approximated, a value smaller (or larger) than the value of the function f(x) at the position of the outermost sensor 18 in the width direction of the device may be regarded as the maximum (or minimum) value of the function f(x). However, in this case, the allowable range of values ​​smaller (or larger) than the value of the function f(x) at the position of the outermost sensor 18 in the width direction of the device that can be regarded as the maximum (or minimum) value of the function f(x) is determined in advance by preliminary testing, magnetic analysis, or the like, depending on the required prediction accuracy of the strip edge position, the target calculation time, and the like. Those skilled in the art can easily determine the allowable range that can be regarded as the maximum (or minimum) value of the function f(x) by preliminary testing, magnetic analysis, or the like. Note that, like the setting of α described above, this tolerance depends on the characteristics of the strip edge detection device 10, and is preferably determined for each strip edge detection device 10 depending on the prediction accuracy of the target strip edge position, the calculation time, etc. This tolerance may be periodically reviewed as necessary. Generally, this tolerance is no more than about several percent of the difference (Lmax-Lmin) between the maximum and minimum values ​​of the function f(x), and may be, for example, no more than 5%, no more than 3%, no more than 2%, no more than 1%, or no more than 0.5% of that difference.

[0073] 2nd position x ２ is the position x of the plate end ０The allowable error range is a position inside the position in the width direction of the device, and is a position where the error between the value of the sigmoid function f(x) obtained in a preliminary test for a test steel plate inclined relative to the multiple detection coils 28 (e.g., test steel plates 36A and 36B in FIG. 9 ) and the value of the sigmoid function f(x) obtained for a test steel plate levitating while remaining horizontal relative to the multiple detection coils 28 (e.g., the test steel plate shown by graph lines A2 to A4 and B2 to B4 in FIG. 9 ) is the maximum value within a predetermined allowable error range. The allowable error range may be determined experimentally. For example, the allowable error range may be determined in advance through a preliminary test depending on the required prediction accuracy of the plate edge position, calculation time, etc. Those skilled in the art can easily determine the allowable error range through preliminary tests or magnetic analysis, etc. This allowable error range depends on the characteristics of the device and is preferably determined for each device depending on the prediction accuracy of the target plate edge position, etc. This allowable error range may be periodically reviewed as necessary. Generally, this tolerance is a few percent or less of the difference (Lmax-Lmin) between the maximum and minimum values ​​of the function f(x), and may be, for example, 5%, 3%, 2%, 1%, or 0.5% of that difference.

[0074] The information about the length of the section may be stored in advance in the volatile memory 44 or the nonvolatile memory 46, and the calculation processing unit 66 may acquire the information about the length of the section stored in advance in the volatile memory 44 or the nonvolatile memory 46. Note that the information about the length of the section may be, for example, the information about the distance between the position of the plate end of the steel plate 14 (for example, the position where the value on the horizontal axis in FIG. 9 is 0) and the first position x １ the distance (length) between the end of the steel plate 14 and the second position x ２ The first specific section is the interval (length) between the first specific section and the second specific section. If information about the length of this section is stored in advance, the specific position of the first specific section in the device width direction can be determined based on the approximate position of the end of the steel sheet 14 predicted by a first approximation or the like (including the method without approximation described above) and the information about the length of the section stored in advance. Next, the position of the end of the steel sheet 14 can be accurately detected based on the relationship between the position x of each detection unit within this determined first specific section and the amount of change f in the signal of each detection unit.

[0075] As described above, in the second embodiment, in a preliminary test using a test steel plate as a substitute for a steel plate, １ and the second position x ２ and the first position x １ to the second position x ２ When a section having a length equivalent to the distance to the target position is defined as the first specific section, the arithmetic processing unit 66 may predict the position of the end of the steel plate 14 and set the first specific section to a range including the position of the end of the steel plate 14. Then, the arithmetic processing unit 66 may detect the position of the end of the steel plate 14 based on the relationship between the amount of change f of the signal detected by each detection coil 28 in the first specific section and the position x of each detection coil 28 in the device width direction in the first specific section.

[0076] According to the second embodiment, the position of the plate end of the steel plate 14 is detected using data corresponding to a section (for example, the section from -70 mm to 40 mm shown in Figure 9) excluding the section affected by the tilt of the steel plate 14 (for example, the section over 40 mm shown in Figure 9) from the entire section of the multiple sensors 18.Therefore, the position of the plate end of the steel plate 14 can be detected while mitigating the effect of the tilt of the steel plate 14.

[0077] In the second embodiment, as described above, the calculation processing unit 66 may perform a first approximation using a sigmoid function to predict the position of the end of the steel sheet 14, and then perform a second approximation using the sigmoid function to determine the relationship between the position of the sensor 18 and the amount of change in the signal corresponding to the magnetic field strength corresponding to a range set based on the sigmoid function obtained by the first approximation. The data used in the second approximation may be data corresponding to the range among the data used in the first approximation. The calculation processing unit 66 may then detect the position of the end of the steel sheet 14 by deriving a position on the horizontal axis that divides the range between the value considered to be the upper limit and the value considered to be the lower limit at the predetermined specific ratio, based on the vertical axis values ​​of the sigmoid function obtained by the second approximation. In this way, a new sigmoid function is obtained by performing the second approximation without using data in a range in which the amount of change in the signal corresponding to the magnetic field strength deviates from the case of a horizontal steel sheet 14. Therefore, the position of the end of the steel sheet 14 can be detected while further mitigating the influence of tilting the steel sheet 14. That is, by performing the second approximation, it is possible to improve the detection accuracy when detecting the position of the plate end of the steel plate 14 compared to when performing the approximation once.

[0078] Furthermore, when performing two approximations, the calculation processing unit 66 may detect the approximate position of the strip end by deriving a horizontal axis position that divides the range between the values ​​considered to be the upper limit and the values ​​considered to be the lower limit of the vertical axis of the sigmoid function obtained by the first approximation into the predetermined specific ratio. Then, among multiple points where the vertical axis value is considered to be the upper limit during the first approximation (e.g., points corresponding to -70 to -40 mm in FIG. 9 ), the calculation processing unit 66 may extract a point closest to the point corresponding to the approximate position of the strip end (e.g., points corresponding to -40 mm in FIG. 9 ) and use the position of the sensor 18 corresponding to the extracted point as the starting point of the interval used during the second approximation. In this way, the second approximation is performed without using data from a range of consecutive values ​​considered to be the upper limit that does not affect the approximation (e.g., points corresponding to -70 to -30 mm in FIG. 9 ). This reduces the burden on the processor 42 during the second approximation.

[0079] 9 uses inductance as the detected signal, the relationship between the vertical and horizontal axes is such that the value on the vertical axis decreases as the value on the horizontal axis increases, but depending on the method of calculating the value on the vertical axis (including changing the detection target to voltage or current), it is also possible that the value on the vertical axis increases as the value on the horizontal axis increases (i.e., the graph in FIG. 9 is displayed upside down). In this case, the calculation processing unit 66 may set as the start point the position of the sensor 18 located at the outermost position in the device width direction among the points where the value on the vertical axis is considered to be the lower limit, and as the end point the position of the sensor 18 corresponding to the point where the value on the vertical axis is maximum among the points where the value on the vertical axis is considered to be the upper limit, where the value on the vertical axis is not affected even if the steel sheet 14 is tilted, among the points where the value on the vertical axis is considered to be the upper limit. The calculation processing unit 66 may then detect the position of the end of the steel plate 14 based on the amount of change in the signal corresponding to the strength of the magnetic field obtained from one of the multiple sensors 18 located in the section from the starting point to the end point.

[0080] Furthermore, when the value of the sigmoid function f(x) obtained for the test steel plate increases as the position x of the detection coil 28 changes outward in the device width direction (as shown in the graph of FIG. 9 ), the first position x １ may be set at a position corresponding to the maximum value of the sigmoid function f(x). However, if the value of the sigmoid function f(x) obtained for the test steel plate decreases as the position x of the detection coil 28 changes outward in the device width direction (when the graph in FIG. 9 is shown upside down), the first position x １ may be at a position corresponding to the minimum of the sigmoid function f(x).

[0081] Third Embodiment Next, a third embodiment of the present disclosure will be described. Contents that overlap with those of the first embodiment will be omitted by using similar terms.

[0082] In the second embodiment described above, when the relationship between the change in the signals detected by all the sensors 18 and the positions of the sensors 18 is approximated by a sigmoid function, a section (for example, the section from -70 mm to 40 mm shown in Figure 9) is set, excluding the section affected by the tilt of the steel plate 14 (for example, the section exceeding 40 mm shown in Figure 9) from the entire section of the multiple sensors 18, and the position of the plate end of the steel plate 14 is detected using data corresponding to the set section.

[0083] In response to this, the inventors conducted an experiment in which they preset a section narrower than the entire section of the multiple sensors 18 (for example, a section equivalent to 80 mm from -40 mm to 40 mm as shown in Figure 9), and while shifting the start point of the section from the outer position in the width direction of the device (for example, the -70 mm position as shown in Figure 9) to the inside, approximated the relationship between the change in signal corresponding to the strength of the magnetic field obtained for each section and the position of the sensor 18 with a sigmoid function, and detected the position on the horizontal axis that divides the range between the value considered to be the upper limit and the value considered to be the lower limit of the vertical axis of the approximated sigmoid function into the above-mentioned predetermined specific ratio as the end of the plate.

[0084] 10 shows an example of the relationship between the position of the plate end detected by an experiment and the number of times the starting point was shifted by one for each position of the sensor 18. The vertical axis shows the detected position of the plate end relative to the actual position of the plate end, which is set to 0, and the horizontal axis shows the number of times the starting point was shifted by one for each position of the sensor 18. The inventors have found that when the starting point is shifted by one for each position of the sensor 18 from one to three times and nine times or more, the detected position of the plate end fluctuates, but when the starting point is shifted by one for each position of the sensor 18 from four to eight times, the detected position of the plate end remains stable near 0, even if the starting point is shifted. In this way, even if the starting point is shifted, the detected position of the plate end stabilizes near 0. This is thought to be because, when the starting point is shifted by one point for each position of sensor 18 from 4 to 8 times, the actual plate end is located near the position on the horizontal axis that divides the range between the value that can be considered the upper limit and the value that can be considered the lower limit of the vertical axis value of the approximated sigmoid function into the predetermined specific ratio described above.

[0085] Based on the above findings, the inventors have considered that it would be sufficient to select a predetermined number of sensors 18 from the sensor 18 that serves as a start point (for example, sensors 18 in a section corresponding to 80 mm from -40 mm to 40 mm as shown in Figure 9) from the plurality of sensors 18 in order from an outer position in the device width direction (for example, a position of -70 mm as shown in Figure 9) to the inner side, and to identify a predetermined number of sensors 18 from the sensor 18 that serves as the start point, which are within a range where the detected position of the plate end of the steel plate 14 can be regarded as constant even if the start point is shifted, from the plurality of sensors 18, and to derive the correct position of the plate end of the steel plate 14 from the position of the plate end of the steel plate 14 detected based on the signals detected by the identified sensors 18 when the position of the plate end of the steel plate 14 is detected based on the signals detected by the selected sensors 18 while shifting the start point.

[0086] Specifically, the inventors thought that if the state in which the detected position of the plate end of the steel plate 14 falls within a predetermined error range X continues for a predetermined number of times (e.g., N = 5) or more, even when the starting point is shifted, then a representative value (e.g., average or median) of the positions of the plate end of the steel plate 14 detected in each of the N times should be calculated, and the calculated representative value should be used as the correct position of the plate end of the steel plate 14.

[0087] The number of sensors 18 selected as the predetermined number of sensors 18 from the above-mentioned starting sensor 18 may be set to a number corresponding to a range from a point closest to the point corresponding to the position of the sheet end (e.g., a point corresponding to -40 mm in FIG. 9 ) to a point corresponding to a value considered to be the lower limit (e.g., a point corresponding to 40 mm in FIG. 9 ) among multiple points (e.g., points corresponding to -70 to -40 mm in FIG. 9 ) at which the value on the vertical axis of a sigmoid function obtained based on signals detected by all sensors 18 (see FIG. 9 ) is considered to be an upper limit. The values ​​of X and N may be set based on experimental or analytical results of the distribution of the amplitude of signals detected by the multiple sensors 18, assuming, for example, the height L1 from the reference height of the sensor 18 to the reference height of the pass line, the floating amount of the steel sheet 14, the inclination of the steel sheet 14, etc.

[0088] Based on the above idea, in the third embodiment, the arithmetic processing unit 66 operates as follows. That is, the arithmetic processing unit 66 selects a predetermined number of sensors 18 from the sensor 18 that is the starting point from among the plurality of sensors 18, in order from the outer position to the inner position in the device width direction, and sets a section (for example, a section from -40 mm to 40 mm shown in FIG. 9 ). While maintaining the interval (length) of the section (for example, a first position x １ to the second position x ２ The start point of the set section is shifted by one for each position of the sensors 18 (while maintaining the distance to the sensor 18), and the position of the plate end of the steel plate 14 is detected based on the signals detected by the selected sensors 18 corresponding to the section. Furthermore, the calculation processing unit 66 identifies a predetermined number of sensors 18, starting from the sensor 18 serving as the start point, that are within a range in which the detected position of the plate end of the steel plate 14 can be considered constant even when the start point is shifted. Then, the calculation processing unit 66 derives the correct position of the plate end of the steel plate 14 from the position of the plate end of the steel plate 14 detected based on the signals detected by the identified sensors 18. Specifically, if the detected position of the plate end of the steel plate 14 remains within a predetermined error range X for a predetermined number of times (e.g., N = 5) or more, even when the starting point is shifted, the calculation processing unit 66 calculates a representative value (e.g., average or median) of the positions of the plate end of the steel plate 14 detected in each of the N times, and adopts the calculated representative value as the correct position of the plate end of the steel plate 14.

[0089] The length of the section is, for example, the length of the first position x defined below in a preliminary test using a test steel plate that serves as a substitute for the steel plate. １ ' and the second position x ２ and the first position x １ to the second position x ２ The distance may be set to a length corresponding to the distance to the

[0090] 1st position x １ ' is the position x of the plate end ０In the preliminary test, if the value of the function f(x) obtained for the test steel plate increases as the position x of the detector coil 28 moves outward in the width direction of the device, this is the innermost position in the width direction of the device among the sections in which the maximum value of the function f(x) is obtained for the test steel plate, and if the value of the function f(x) obtained for the test steel plate decreases as the position x of the detector coil 28 moves outward in the width direction of the device, this is the innermost position in the width direction of the device among the sections in which the minimum value of the function f(x) is obtained for the test steel plate. The section in which the maximum value of the function f(x) is obtained may be the section in which a value that can be considered as the upper limit described above is obtained (for example, the section from -70 to -40 mm in Figure 9).

[0091] Although it depends on the function used, the "position corresponding to the maximum value (or minimum value) of the function f(x) for the test steel sheet" is often located at a position infinitely far outward in the width direction of the device. In this case, the "section where the minimum value (or minimum value) of the function f(x) for the test steel sheet is obtained" is actually located at a position infinitely far outward in the width direction of the device. However, １ When determining "f(x)", the "minimum value (or minimum value) of the function f(x)" is a value that can be considered to be the "minimum value (or minimum value) of the function f(x)". The allowable range, which can be considered to be the "minimum value (or minimum value) of the function f(x)" regardless of the extent to which it varies from the original "minimum value (or minimum value) of the function f(x)", can be determined in advance through preliminary testing, magnetic analysis, etc., depending on the required prediction accuracy of the strip edge position and the target calculation time. In other words, a person skilled in the art can easily determine the allowable range that can be considered to be the maximum (or minimum value) of the function f(x) through preliminary testing, magnetic analysis, etc. Note that this allowable range depends on the characteristics of the strip edge detection device 10 and is preferably determined for each strip edge detection device 10 depending on the prediction accuracy of the target strip edge position, etc. This allowable range may be periodically reviewed as necessary. Generally, this tolerance is no more than a few percent of the difference (Lmax-Lmin) between the maximum and minimum values ​​of the function f(x), and may be, for example, no more than 5%, no more than 3%, no more than 2%, no more than 1%, no more than 0.5%, or no more than 0.3% of that difference.

[0092] 2nd position x ２ is the position x of the plate end ０ The second position x is a position inside the position in the width direction of the device, and is a position where the error between the value of the sigmoid function f(x) obtained in a preliminary test for a test steel plate inclined with respect to the multiple detection coils 28 (e.g., test steel plates 36A and 36B in FIG. 9 ) and the value of the sigmoid function f(x) obtained for a test steel plate levitating while remaining horizontal with respect to the multiple detection coils 28 (e.g., the test steel plate shown by graph lines A2 to A4 and B2 to B4 in FIG. 9 ) is the maximum value of a predetermined allowable error range. The allowable error range may be determined experimentally. For example, the required allowable error range may be determined in advance by a preliminary test depending on the required prediction accuracy of the plate end position, the target calculation time, etc. ２ is the second position x in the second embodiment described above. ２ Therefore, the second position x in this paragraph is the same as ２ The method of determining the value and the range thereof will not be described (see the second embodiment for details).

[0093] The information about the length of the section may be stored in advance in the volatile memory 44 or the nonvolatile memory 46, and the calculation processing unit 66 may acquire the information about the length of the section stored in advance in the volatile memory 44 or the nonvolatile memory 46. Note that the information about the length of the section may be, for example, the information about the distance between the position of the plate end of the steel plate 14 (for example, the position where the value on the horizontal axis in FIG. 9 is 0) and the first position x １ ', the position of the end of the steel plate 14, and the second position x ２ The second specific section is the distance (length) between the first and second specific sections. Alternatively, the second specific section may be the sum of these two distances (lengths). If information about the length of this section is stored in advance, the start point (and end point) of the second specific section can be shifted in the device width direction for each position of the detection unit while maintaining the distance (length) of the second specific section. In other words, the second specific section can be moved in the device width direction.

[0094] The target movement section for moving the second specific section in the device width direction does not have to be the entire width of the device, but may be, for example, near the approximate position of the plate end of the steel plate 14 predicted by the first approximation in the second embodiment (including the method without approximation).

[0095] Then, in a preliminary test using a test steel plate as a substitute for the steel plate, the first position x defined above １ ' and the second position x ２ and the first position x １ ' to the second position x ２ In the case where the second specific section is defined as a section having a length equivalent to the distance to the target position, the arithmetic processing unit 66 may detect the position of the plate end of the steel plate 14 by performing the following arithmetic processing. That is, the arithmetic processing unit 66 may repeat the process of deriving the position of the plate end of the steel plate 14 based on the relationship between the amount of change f in the signal detected by each detection coil 28 in the second specific section and the position x of each detection coil 28 in the second specific section in the device width direction while moving the second specific section in the device width direction by shifting the start point of the second specific section in the device width direction, identify the minimum range (for example, the range when the second specific section is shifted from 4 to 8 times in FIG. 10 ) in which the amount of change in the position of the plate end of the steel plate 14 due to the movement of the second specific section is the smallest, and detect the position of the plate end of the steel plate 14 obtained in the minimum range as the position of the plate end of the steel plate 14. In addition, the calculation processing unit 66 may detect a representative value (e.g., average value, median value, maximum value, or minimum value) of the position of the plate end of the steel plate 14 obtained in the minimum range as the position of the plate end of the steel plate 14.

[0096] According to the third embodiment, among the multiple sensors 18, a predetermined number of sensors 18 are identified, starting from the sensor 18 that serves as the starting point, within a range where the detected position of the plate edge of the steel plate 14 can be considered constant even if the starting point is shifted, and the correct position of the plate edge of the steel plate 14 is derived from the position of the plate edge of the steel plate 14 detected based on the signal detected by the identified sensor 18. Therefore, the detection accuracy of the position of the plate edge of the steel plate 14 can be improved.

[0097] Furthermore, the approximation is performed without using data from the range of consecutive values ​​that can be considered as upper limits and that do not affect the approximation (for example, the point corresponding to -70 to -30 mm in Figure 9), thereby reducing the burden on the processor 42 when performing the approximation.

[0098] Furthermore, the position of the plate end of the steel plate 14 is detected using data corresponding to a section (for example, the section from -40 mm to 40 mm shown in Figure 9) excluding the section affected by the tilt of the steel plate 14 (for example, the section over 40 mm shown in Figure 9) from the entire section of the multiple sensors 18, so the position of the plate end of the steel plate 14 can be detected while mitigating the effect of the tilt of the steel plate 14.

[0099] In addition, when the value of the sigmoid function f(x) obtained for the test steel plate increases as the position x of the detection coil 28 changes outward in the device width direction (in the case of the graph shown in FIG. 9 , which slopes upward to the left), １ The first position x ′ may be set to the innermost position in the width direction of the device among the sections where the maximum value of the function f(x) for the test steel plate is obtained. However, if the value of the sigmoid function f(x) obtained for the test steel plate decreases as the position x of the detection coil 28 moves outward in the width direction of the device (if the graph in FIG. 9 is shown upside down), １ The position f(x)′ may be the innermost position in the width direction of the device among the sections in which the minimum value of the function f(x) is obtained for the test steel sheet in the preliminary test. The section in which the minimum value of the function f(x) is obtained may be a section corresponding to the section in which a value that can be regarded as the upper limit described above is obtained (for example, the section from −70 to −40 mm in FIG. 9 ).

[0100] Furthermore, in each of the above embodiments, the case where a plurality of sensors 18 is used has been described as an example of the strip edge detection device 10, but the strip edge detection device 10 may be configured with a single sensor 18. While a single sensor 18 having a yoke 24, an excitation coil 26, and a detection coil 28 is moved in the width direction of the device, the position of the strip edge of the steel sheet 14 can be detected based on the amount of change in the signal corresponding to the strength of the magnetic field at the position of the sensor 18.

[0101] Furthermore, by using a yoke extending in the width direction of the device, an excitation coil corresponding to the yoke, and multiple detection coils arranged in the width direction of the device, the position of the plate end of the steel plate 14 can be detected based on the amount of change in signal corresponding to the strength of the magnetic field at the positions of the multiple detection coils 28. This will be described in detail in the fourth embodiment.

[0102] Here, an example is given in which the excitation coil 26 and detection coil 28 used in the sensor are attached to the yoke as separate coils, but the excitation coil 26 and detection coil 28 may also be configured as the same coil, and the sensor may detect the effect of the magnetic field generated by the coil itself being affected by the steel plate.

[0103] Furthermore, although the above examples show examples in which a yoke is used as the ferromagnetic core material, it is not limited to a yoke. Furthermore, the magnetic field generating unit may be any unit capable of generating a magnetic field, and may be an excitation coil without a ferromagnetic core material. Furthermore, the detecting unit may be any unit capable of detecting a signal corresponding to the strength of the magnetic field, and may be a detection coil without a ferromagnetic core material. [Fourth Embodiment] Next, a fourth embodiment of the present disclosure will be described.

[0104] 11 and 12 show an example of a sensor 118 according to the fourth embodiment. In the fourth embodiment, a single sensor 118 is used instead of the multiple sensors 18 of the first embodiment. The sensor 118 includes a yoke 124, an excitation coil 126, and multiple detection coils 128. The yoke 124 is made of a magnetic material such as a ferrite core, and includes a pair of core portions 132 and a connecting portion 134. The pair of core portions 132 extend vertically, and the connecting portion 134 connects the lower ends of the pair of core portions 132 together. The yoke 124 is provided to extend in the conveyance direction.

[0105] The excitation coil 126 is wound around one of a pair of core portions 132 via a bobbin (not shown). The excitation coil 126 is wound around one of the core portions 132 and extends in the conveyance direction. The multiple detection coils 128 are arranged side by side in the device width direction. Each detection coil 128 is arranged above the other core portion 132 of the pair of core portions 132. The sensor 118 has one excitation coil 126 for the multiple detection coils 128, and the excitation coil 126 is used in common for the multiple detection coils 128.

[0106] The above describes one embodiment of the present disclosure, but the present disclosure is not limited to the above, and it goes without saying that various modifications can be made to the present disclosure without departing from the spirit of the present disclosure.

[0107] The disclosure of Japanese Application No. 2024-035343, filed on March 7, 2024, is incorporated herein by reference in its entirety.

[0108] The following are supplementary notes regarding the present disclosure: (Supplementary Note 1) A sheet edge detection device for detecting the position of the sheet edge of a steel sheet in a cooling device for cooling the steel sheet, comprising: a magnetic field generation unit that is installed in a space between two adjacent transport rolls out of a plurality of transport rolls that transport the steel sheet and a path through which the steel sheet passes, and that generates a magnetic field; a plurality of detection units that are installed side by side in the width direction of the sheet edge detection device corresponding to the sheet width direction of the steel sheet, and that detect, at their respective positions, signals corresponding to the strength of a magnetic field obtained by reflecting the influence of the steel sheet on the magnetic field generated by the magnetic field generation units; and a calculation processing unit that detects the position of the sheet edge of the steel sheet from the relationship between the amount of change in the signal detected by the detection units and the positions of the detection units, wherein the calculation processing unit approximates, with a function, the relationship between the amount of change in the signal and the positions of the detection units, which is shown in a graph having a first axis corresponding to the amount of change in the signal and a second axis corresponding to the positions of the detection units, a plate edge detection device that detects, as the position of the plate edge of the steel plate, the position of the second axis that divides a range between a value that can be regarded as an upper limit and a value that can be regarded as a lower limit in the first axis values ​​of the function obtained by approximation, at a predetermined specific ratio. (Supplementary Note 2) The plate edge detection device according to Supplementary Note 1, wherein the specific ratio is determined by: obtaining a relationship between the amount of change in the signal and the position of the detection unit using a test piece that serves as a substitute for the steel plate; approximating, with the function, the relationship between the amount of change in the signal and the position of the detection unit shown in the graph obtained using the test piece; determining a division ratio that is a ratio when the range between a value that can be regarded as an upper limit and a value that can be regarded as a lower limit in the first axis values ​​of the function obtained by approximation is divided into positions on the second axis that correspond to the position of the plate edge of the test piece; and calculating an average value of a plurality of the division ratios obtained by changing the position of the plate edge of the test piece in the width direction of the device with respect to the position of the detection unit.(Supplementary Note 3) When the relationship between the amount of change in the signal and the position of the detection unit shown in the graph is approximated by the function, the calculation processing unit, in order from the outer position of the sheet end along the device width direction of the steel sheet, sets the position of the detection unit corresponding to a point where the value of the first axis is a value that can be regarded as an upper limit as a start point, and sets the position of the detection unit corresponding to a point where the value of the first axis is a value that can be regarded as a lower limit, among points on the way where the value of the first axis decreases from a value that can be regarded as an upper limit to a lower limit, where the value of the first axis is not affected even if the steel sheet is tilted, and where the value of the first axis is a value that can be regarded as a lower limit, as an end point; or sets the position of the detection unit corresponding to a point where the value of the first axis is a value that can be regarded as a lower limit, among points on the way where the value of the first axis increases from a value that can be regarded as a lower limit to an upper limit, where the value of the first axis is not affected even if the steel sheet is tilted, as an end point, The plate edge detection device according to Supplementary Note 1 or Supplementary Note 2, which detects the position of the plate edge of the steel plate based on an amount of change in the signal obtained from detection units among the plurality of detection units that are present in a section from the start point to the end point. (Supplementary Note 4) The plate edge detection device according to Supplementary Note 1 or Supplementary Note 2, wherein the arithmetic processing unit selects a predetermined number of detection units from the detection unit that serves as a start point, in order from an outer side position of the plate edge along the device width direction of the steel plate, and when detecting the position of the plate edge of the steel plate based on the signals detected by the selected detection units while shifting the start point, identifies a predetermined number of detection units from the detection unit that serves as the start point that are within a range where the detected position of the plate edge of the steel plate can be considered constant even if the start point is shifted, and derives a correct position of the plate edge of the steel plate from the position of the plate edge of the steel plate detected based on the signals detected by the identified detection units.(Supplementary Note 5) A sheet edge detection device for detecting the position of an edge of a steel sheet in a cooling device for cooling the steel sheet, comprising: a magnetic field generating unit that is installed in a space between two adjacent transport rolls out of a plurality of transport rolls that transport the steel sheet and a path through which the steel sheet passes, and that generates a magnetic field; a plurality of detection units that are installed side by side in the width direction of the sheet edge detection device corresponding to the sheet width direction of the steel sheet, and that detect signals corresponding to the strength of the magnetic field obtained by reflecting the influence of the steel sheet on the magnetic field generated by the magnetic field generating units; and an arithmetic processing unit that detects the position of the sheet edge of the steel sheet based on an amount of change f in the signal detected by each of the detection units and a position x of each of the detection units in the width direction, wherein the arithmetic processing unit obtains the relationship between the amount of change f in the signal detected by each of the detection units and the position x of each of the detection units in the width direction, and approximates the relationship with a function f(x), and calculates the position x using the following formula (1): ０ as the position of the edge of the steel plate. ０ ) = Lmin + α × (Lmax - Lmin) ... (1) where Lmax is the maximum value of the function f(x), Lmin is the minimum value of the function f(x), and α is a predetermined coefficient. (Note 6) A storage unit for storing the coefficients is provided, and the coefficients are coefficients set based on a specific coefficient setting process using a test steel plate that serves as a substitute for the steel plate, and the coefficient setting process includes: using the test steel plate, obtaining a relationship between an amount of change f in the signal detected by each of the detection units and a position x of each of the detection units in the width direction, approximating the relationship with the function f(x), and incorporating an actual position x of the plate end of the test steel plate into the function f(x). ０ ' to calculate the value of the function f(x0'), and the range between the maximum and minimum values ​​of the function f(x) is calculated by inputting ０ (Supplementary Note 7) In a preliminary test using a test steel plate that serves as a substitute for the steel plate, a division ratio is calculated, which is a ratio when the end of the test steel plate is divided by a value of the first position x defined below, and a plurality of division ratios are obtained by changing the position of the end of the test steel plate in the width direction with respect to the positions of the plurality of detection units, and the coefficient is set based on the plurality of division ratios.１ and the second position x ２ and determining the first position x １ to the second position x ２ The sheet edge detection device according to Supplementary Note 5 or Supplementary Note 6, wherein, when a section having a length equivalent to a distance to a point is defined as a first specific section, the arithmetic processing unit predicts the position of the sheet edge of the steel sheet, sets the first specific section to a range including the position of the sheet edge of the steel sheet, and detects the position of the sheet edge of the steel sheet based on the relationship between an amount of change f of the signal detected by each of the detection units in the first specific section and a position x of each of the detection units in the first specific section in the width direction. １ is the position x ０ and when the value of the function f(x) obtained for the test steel plate in the preliminary test increases as the position x of the detection unit changes outward in the width direction, the second position x is a position corresponding to the maximum value of the function f(x), and when the value of the function f(x) obtained for the test steel plate in the preliminary test increases as the position x of the detection unit changes outward in the width direction, the second position x is a position corresponding to the minimum value of the function f(x); ２ is the position x ０ and a position at which the error between the value of the function f(x) obtained for the test steel plate tilted relative to the plurality of detection units in the preliminary test and the value of the function f(x) obtained for the test steel plate floating while remaining horizontal relative to the plurality of detection units is the maximum value of a predetermined allowable error range. １ ' and the second position x ２ and determining the first position x １ ' to the second position x ２the calculation processing unit repeats a process of deriving the position of the edge of the steel plate based on the relationship between the amount of change f in the signal detected by each of the detection units in the second specific section and the position x of each of the detection units in the second specific section in the width direction while moving the second specific section in the width direction by shifting the start point of the second specific section in the width direction for each position of the detection unit, identifies a minimum range within the range to which the second specific section is moved, in which the amount of change in the position of the edge of the steel plate due to the movement of the second specific section is the smallest, and detects the position of the edge of the steel plate obtained in the minimum range as the position of the edge of the steel plate. １ ' is the position x ０ and, in the preliminary test, if the value of the function f(x) obtained for the test steel plate increases as the position x of the detection unit changes outward in the width direction, the second position x is the innermost position in the width direction within the section in which the maximum value of the function f(x) is obtained for the test steel plate, and, if the value of the function f(x) obtained for the test steel plate decreases as the position x of the detection unit changes outward in the width direction, the second position x is the innermost position in the width direction within the section in which the minimum value of the function f(x) is obtained for the test steel plate, ２ is the position x ０and the position where the error between the value of the function f(x) obtained in the preliminary test for the test steel plate inclined with respect to the plurality of detection parts and the value of the function f(x) obtained for the test steel plate floating while remaining horizontal with respect to the plurality of detection parts is the maximum value of a predetermined allowable error range. (Supplementary Note 9) A plate edge detection method for detecting the position of the plate edge of a steel plate in a cooling device that cools the steel plate, comprising: a plate edge detection device having one or more magnetic field generating units that are installed in a space between two adjacent transport rolls out of a plurality of transport rolls that transport the steel plate and a path through which the steel plate passes, and that generate a magnetic field; a plurality of detection units that are installed side by side in the device width direction of the steel plate, and that detect, at their respective installed positions, signals corresponding to the strength of a magnetic field obtained by reflecting the influence of the steel plate on the magnetic field generated by the magnetic field generating units; and an arithmetic processing unit that detects the position of the plate edge of the steel plate from the relationship between the amount of change in the signal detected by the detection units and the positions of the detection units; a detection step of detecting, at their respective installed positions, signals corresponding to the strength of a magnetic field obtained by reflecting the influence of the steel plate on the magnetic field generated by the magnetic field generating units, using the plurality of detection units; and an arithmetic processing step of detecting the position of the plate edge of the steel plate using the arithmetic processing unit, from the relationship between the amount of change in the signal detected by the detection units and the positions of the detection units, A plate end detection method comprising: approximating, with a function, the relationship between the amount of change in the signal and the position of the detection unit, which is shown in a graph having a first axis corresponding to the amount of change in the signal and a second axis corresponding to the position of the detection unit; and detecting, as the position of the plate end of the steel plate, the position of the second axis that divides a range between a value that can be considered as an upper limit and a value that can be considered as a lower limit in the first axis values ​​of the function obtained by the approximation, at a predetermined specific ratio.(Supplementary Note 10) A plate edge detection method for detecting the position of the plate edge of a steel plate in a cooling device that cools the steel plate, comprising: one or more magnetic field generating units that are installed in a space generated between two adjacent transport rolls out of a plurality of transport rolls that transport the steel plate and a path through which the steel plate passes, and that generate a magnetic field; a plurality of detection units that are installed side by side in the width direction of the plate edge detection device corresponding to the plate width direction of the steel plate, and that detect a signal corresponding to the strength of the magnetic field obtained by reflecting the influence of the steel plate in the magnetic field generated by the magnetic field generating units; and an arithmetic processing unit that detects the position of the plate edge of the steel plate based on a change f in the signal detected by each of the detection units and a position x of each of the detection units in the width direction; a detection step, using the plurality of detection units, that detects a signal corresponding to the strength of the magnetic field obtained by reflecting the influence of the steel plate in the magnetic field generated by the magnetic field generating units; and an arithmetic processing step, using the arithmetic processing unit, that detects the position of the plate edge of the steel plate based on the change f in the signal detected by each of the detection units and the position x of each of the detection units in the width direction. The calculation processing step obtains a relationship between an amount of change f in the signal detected by each of the detection units and a position x of each of the detection units in the width direction, approximates the relationship with a function f(x), and calculates the position x using the following formula (1): ０ as the position of the plate end of the steel plate. ０ )=Lmin+α×(Lmax−Lmin) (1) where Lmax is the maximum value of the function f(x), Lmin is the minimum value of the function f(x), and α is a predetermined coefficient. (Appendix 11) The plate edge detection device has a storage unit that stores the coefficients, and the coefficients are set based on a specific coefficient setting process using a test steel plate that serves as a substitute for the steel plate, and the coefficient setting process includes: using the test steel plate, obtaining a relationship between an amount of change f in the signal detected by each of the detection units and a position x of each of the detection units in the width direction; approximating the relationship with the function f(x), and calculating the actual position x of the plate edge of the test steel plate in the function f(x). ０' to calculate the value of the function f(x0'), and the range between the maximum and minimum values ​​of the function f(x) is calculated by inputting ０ (Supplementary Note 12) In a preliminary test using a test steel plate that serves as a substitute for the steel plate, a division ratio is calculated, which is a ratio when the end of the test steel plate is divided by a value of the first position x defined below, and a plurality of division ratios are obtained by changing the position of the end of the test steel plate in the width direction with respect to the positions of the plurality of detection units. １ and the second position x ２ and determining the first position x １ to the second position x ２ The method for detecting an edge of the steel plate according to Supplementary Note 10 or Supplementary Note 11, wherein the first specific section is defined as a section having a length corresponding to a distance from the steel plate to the end of the steel plate, and the calculation step includes predicting the position of the edge of the steel plate, setting the first specific section to a range including the position of the edge of the steel plate, and detecting the position of the edge of the steel plate based on the relationship between the amount of change f of the signal detected by each of the detection units in the first specific section and the position x of each of the detection units in the first specific section in the width direction. １ is the position x ０ and when the value of the function f(x) obtained for the test steel plate in the preliminary test increases as the position x of the detection unit changes outward in the width direction, the second position x is a position corresponding to the maximum value of the function f(x), and when the value of the function f(x) obtained for the test steel plate in the preliminary test increases as the position x of the detection unit changes outward in the width direction, the second position x is a position corresponding to the minimum value of the function f(x); ２ is the position x ０ and a position at which the error between the value of the function f(x) obtained for the test steel plate tilted relative to the plurality of detection units in the preliminary test and the value of the function f(x) obtained for the test steel plate floating while remaining horizontal relative to the plurality of detection units is the maximum value of a predetermined allowable error range.１ ' and the second position x ２ and determining the first position x １ ' to the second position x ２ the second specific section is defined as a section having a length equivalent to a distance to the first position, the calculation step repeats a process of deriving the position of the end of the steel plate based on a relationship between an amount of change f in the signal detected by each of the detection units in the second specific section and a position x of each of the detection units in the second specific section in the width direction while moving the second specific section in the width direction by shifting a start point of the second specific section in the width direction for each position of the detection unit, and identifies a minimum range within the range to which the second specific section is moved, in which the amount of change in the position of the end of the steel plate due to the movement of the second specific section is smallest, and detects the position of the end of the steel plate obtained in the minimum range as the position of the end of the steel plate. １ ' is the position x ０ and, in the preliminary test, if the value of the function f(x) obtained for the test steel plate increases as the position x of the detection unit changes outward in the width direction, the second position x is the innermost position in the width direction within the section in which the maximum value of the function f(x) is obtained for the test steel plate, and, if the value of the function f(x) obtained for the test steel plate decreases as the position x of the detection unit changes outward in the width direction, the second position x is the innermost position in the width direction within the section in which the minimum value of the function f(x) is obtained for the test steel plate, ２ is the position x ０ and the position where the error between the value of the function f(x) obtained in the preliminary test for the test steel plate inclined with respect to the plurality of detection parts and the value of the function f(x) obtained for the test steel plate floating while remaining horizontal with respect to the plurality of detection parts is the maximum value of a predetermined allowable error range.

[0109] REFERENCE SIGNS LIST 10 Sheet edge detection device 12 Conveyor roll 14 Steel sheet 16 Cooling device 18 Sensor 20 Processing device 24 Yoke 26 Excitation coil 28 Detection coil 36 Test steel sheet

Claims

1. A sheet edge detection device for detecting the position of the sheet edge of a steel sheet in a cooling device for cooling the steel sheet, comprising: one or more magnetic field generating units that are installed in a space between two adjacent transport rolls out of a plurality of transport rolls that transport the steel sheet and a path through which the steel sheet passes, and that generate a magnetic field; a plurality of detection units that are installed side by side in the width direction of the sheet edge detection device corresponding to the sheet width direction of the steel sheet, and that detect signals corresponding to the strength of the magnetic field obtained by reflecting the influence of the steel sheet on the magnetic field generated by the magnetic field generating units; and a calculation processing unit that detects the position of the sheet edge of the steel sheet based on a change f in the signal detected by each of the detection units and a position x of each of the detection units in the width direction, wherein the calculation processing unit obtains the relationship between the change f in the signal detected by each of the detection units and the position x of each of the detection units in the width direction, and approximates the relationship with a function f(x), and calculates the position x using the following formula (1): ０ as the position of the edge of the steel plate. ０ )=Lmin+α×(Lmax−Lmin) (1) where Lmax is the maximum value of the function f(x), Lmin is the minimum value of the function f(x), and α is a predetermined coefficient.

2. A storage unit for storing the coefficients is provided, and the coefficients are set based on a specific coefficient setting process using a test steel plate that serves as a substitute for the steel plate, and the coefficient setting process involves: obtaining, using the test steel plate, a relationship between the amount of change f in the signal detected by each of the detection units and the position x of each of the detection units in the width direction; approximating the relationship with the function f(x); and incorporating the actual position x of the plate end of the test steel plate into the function f(x). ０ ' to calculate the value of the function f(x0'), and then add the function f(x ０ 2. The plate edge detection device according to claim 1, wherein α is calculated by substituting the maximum value Lmax and minimum value Lmin of the function f(x) into α=(f(x′), and the coefficient is set as α. ０ ')-Lmin) / (Lmax-Lmin)...(1') 3. In a preliminary test using a test steel plate that substitutes for the steel plate, the first position x defined below １ and the second position x ２ and determining the first position x １ to the second position x ２ 3. The sheet edge detection device according to claim 1 or 2, wherein, in a case where a section having a length equivalent to a distance to a point is defined as a first specific section, the arithmetic processing unit predicts the position of the sheet edge of the steel sheet, sets the first specific section to a range including the position of the sheet edge of the steel sheet, and detects the position of the sheet edge of the steel sheet based on a relationship between an amount of change f of the signal detected by each of the detection units in the first specific section and a position x of each of the detection units in the first specific section in the width direction. １ is the position x ０ and when the value of the function f(x) obtained for the test steel plate in the preliminary test increases as the position x of the detection unit changes outward in the width direction, the second position x is a position corresponding to the maximum value of the function f(x), and when the value of the function f(x) obtained for the test steel plate in the preliminary test increases as the position x of the detection unit changes outward in the width direction, the second position x is a position corresponding to the minimum value of the function f(x); ２ is the position x ０ and the position where the error between the value of the function f(x) obtained in the preliminary test for the test steel plate inclined with respect to the plurality of detection parts and the value of the function f(x) obtained for the test steel plate floating while remaining horizontal with respect to the plurality of detection parts is the maximum value of a predetermined allowable error range.

4. In a preliminary test using a test steel plate that substitutes for the steel plate, the first position x defined below １ ' and the second position x ２ and determining the first position x １ ' to the second position x ２ 3. The plate edge detection device according to claim 1, wherein the calculation processing unit repeats a process of deriving the position of the plate edge of the steel plate based on a relationship between an amount of change f in the signal detected by each of the detection units in the second specified section and a position x of each of the detection units in the second specified section in the width direction while moving the second specified section in the width direction by shifting a start point of the second specified section in the width direction for each of the detection units, identifies a minimum range within a range in which the amount of change in the position of the plate edge of the steel plate due to the movement of the second specified section is smallest, and detects the position of the plate edge of the steel plate obtained in the minimum range as the position of the plate edge of the steel plate. １ ' is the position x ０ and, in the preliminary test, if the value of the function f(x) obtained for the test steel plate increases as the position x of the detection unit changes outward in the width direction, the second position x is the innermost position in the width direction within the section in which the maximum value of the function f(x) is obtained for the test steel plate, and, if the value of the function f(x) obtained for the test steel plate decreases as the position x of the detection unit changes outward in the width direction, the second position x is the innermost position in the width direction within the section in which the minimum value of the function f(x) is obtained for the test steel plate, ２ is the position x ０ and the position where the error between the value of the function f(x) obtained in the preliminary test for the test steel plate inclined with respect to the plurality of detection parts and the value of the function f(x) obtained for the test steel plate floating while remaining horizontal with respect to the plurality of detection parts is the maximum value of a predetermined allowable error range.

5. A plate edge detection method for detecting the position of the plate edge of a steel plate in a cooling device that cools the steel plate, comprising: one or more magnetic field generating units that are installed in a space between two adjacent transport rolls out of a plurality of transport rolls that transport the steel plate and a path through which the steel plate passes, and that generate a magnetic field; a plurality of detection units that are installed side by side in the width direction of the plate edge detection device corresponding to the plate width direction of the steel plate, and that detect a signal corresponding to the strength of the magnetic field obtained by reflecting the influence of the steel plate in the magnetic field generated by the magnetic field generating units; and an arithmetic processing unit that detects the position of the plate edge of the steel plate based on a change f in the signal detected by each of the detection units and a position x of each of the detection units in the width direction; and a detection step that uses the plurality of detection units to detect a signal corresponding to the strength of the magnetic field obtained by reflecting the influence of the steel plate in the magnetic field generated by the magnetic field generating units; and an arithmetic processing step that uses the arithmetic processing unit to detect the position of the plate edge of the steel plate based on the change f in the signal detected by each of the detection units and the position x of each of the detection units in the width direction, The relationship between the amount of change f of the signal detected by each of the detecting units and the position x of each of the detecting units in the width direction is obtained, the relationship is approximated by a function f(x), and the position x is calculated from the following formula (1): ０ as the position of the plate end of the steel plate. ０ )=Lmin+α×(Lmax−Lmin) (1) where Lmax is the maximum value of the function f(x), Lmin is the minimum value of the function f(x), and α is a predetermined coefficient.

6. The plate edge detection device has a memory unit that stores the coefficients, and the coefficients are set based on a specific coefficient setting process using a test steel plate that serves as a substitute for the steel plate, and the coefficient setting process involves: using the test steel plate, obtaining a relationship between the amount of change f in the signal detected by each of the detection units and the position x of each of the detection units in the width direction; approximating the relationship with the function f(x); and incorporating the actual position x of the plate edge of the test steel plate into the function f(x). ０ ' to calculate the value of the function f(x0'), and then add the function f(x ０ 6. The method for detecting a plate edge according to claim 5, wherein α calculated by substituting the maximum value Lmax and the minimum value Lmin of the function f(x) is set as the coefficient. ０ ')-Lmin) / (Lmax-Lmin)...(1') 7. In a preliminary test using a test steel plate that substitutes for the steel plate, the first position x defined below １ and the second position x ２ and determining the first position x １ to the second position x ２ 7. The method for detecting an edge of the steel plate according to claim 5 or 6, wherein the calculation step includes the steps of: predicting a position of an edge of the steel plate; setting the first specific section to a range including the position of the edge of the steel plate; and detecting the position of the edge of the steel plate based on a relationship between an amount of change f of the signal detected by each of the detection units in the first specific section and a position x of each of the detection units in the first specific section in the width direction. １ is the position x ０ and when the value of the function f(x) obtained for the test steel plate in the preliminary test increases as the position x of the detection unit changes outward in the width direction, the second position x is a position corresponding to the maximum value of the function f(x), and when the value of the function f(x) obtained for the test steel plate in the preliminary test increases as the position x of the detection unit changes outward in the width direction, the second position x is a position corresponding to the minimum value of the function f(x); ２ is the position x ０ and the position where the error between the value of the function f(x) obtained in the preliminary test for the test steel plate inclined with respect to the plurality of detection parts and the value of the function f(x) obtained for the test steel plate floating while remaining horizontal with respect to the plurality of detection parts is the maximum value of a predetermined allowable error range.

8. In a preliminary test using a test steel plate that substitutes for the steel plate, the first position x defined below １ ' and the second position x ２ and determining the first position x １ ' to the second position x ２ 7. The plate edge detection method according to claim 5 or 6, wherein the calculation step repeats a process of deriving the position of the plate edge of the steel plate based on a relationship between an amount of change f in the signal detected by each of the detectors in the second specified section and a position x of each of the detectors in the second specified section in the width direction while moving the second specified section in the width direction by shifting a start point of the second specified section in the width direction for each of the detectors, and identifies a minimum range within a range in which the amount of change in the position of the plate edge of the steel plate due to the movement of the second specified section is smallest, and detects the position of the plate edge of the steel plate obtained in the minimum range as the position of the plate edge of the steel plate. １ ' is the position x ０ and, in the preliminary test, if the value of the function f(x) obtained for the test steel plate increases as the position x of the detection unit changes outward in the width direction, the second position x is the innermost position in the width direction within the section in which the maximum value of the function f(x) is obtained for the test steel plate, and, if the value of the function f(x) obtained for the test steel plate decreases as the position x of the detection unit changes outward in the width direction, the second position x is the innermost position in the width direction within the section in which the minimum value of the function f(x) is obtained for the test steel plate, ２ is the position x ０ and the position where the error between the value of the function f(x) obtained in the preliminary test for the test steel plate inclined with respect to the plurality of detection parts and the value of the function f(x) obtained for the test steel plate floating while remaining horizontal with respect to the plurality of detection parts is the maximum value of a predetermined allowable error range.

Images