Device for detecting leakage of molten metal, method for detecting leakage of molten metal, and method for continuously casting metal

Abstract

Provided is a molten metal leak detection device capable of suppressing erroneous detection of molten metal leaks. A leak detection device for molten metal detects a leak of molten metal from a molten metal processing facility, the leak detection device comprising: an image capturing unit that captures an image of the molten metal processing facility at predetermined time intervals and generates image data; a data acquisition unit that acquires time-series data indicating the number of pixels of a specific color included in a predetermined region of the image data; and a detection unit that uses the time-series data to detect the occurrence of a leak in the molten metal.

Description

Technical Field

[0001] This invention relates to a molten metal leakage detection device, a molten metal leakage detection method, and a continuous casting method for metal using the molten metal leakage detection method. Background Technology

[0002] In equipment handling molten metal, accidents involving molten metal leaks and significant equipment damage are inevitable. Therefore, when a molten metal leak occurs, it is crucial to detect it quickly to prevent further damage. A typical response includes emergency casting shutdown procedures such as stopping the supply of molten metal to the affected equipment.

[0003] Previously, the detection and prevention of escalating damage from abnormal conditions such as molten metal leaks were typically handled by operators. The procedure involved the operator visually or audibly sensing the abnormality, determining its occurrence, and pressing a button to immediately stop the supply of molten steel to the affected equipment. However, relying on operators for detection and prevention of escalation proved slow due to operator characteristics and workload at the time of the abnormality, sometimes resulting in significant equipment damage.

[0004] To address this problem, Patent Document 1 discloses a method and apparatus for automatically detecting and preventing the spread of damage caused by an abnormal condition occurring during a continuous casting process, namely, a steel leakage accident (steel leakage). Here, steel leakage refers to an abnormal condition where the outer skin of a casting sheet cracks for some reason, causing molten steel to leak out. According to Patent Document 1, the casting sheet in continuous casting is photographed, and if the number of pixels representing a predetermined specific color in the resulting image data exceeds a preset threshold, a steel leakage is determined to have occurred.

[0005] Existing technical documents

[0006] Patent documents

[0007] Patent Document 1: Japanese Patent Application Publication No. 2001-269770 Summary of the Invention

[0008] The problem that the invention aims to solve

[0009] In Patent Document 1, the instantaneous value of a pixel is used to detect the occurrence of molten metal leakage, i.e., the occurrence of an abnormal state, in a continuous casting equipment. However, even in a sound casting operation where molten metal leakage does not occur, sparks of molten metal will fly around the slab during casting due to splashing of molten metal.

[0010] The color of the scattered sparks is approximately the same as that of the molten metal leaking during a leak. Therefore, there is a problem that scattered sparks from molten metal in a healthy casting operation can be mistaken for a leak. This invention was made in view of this problem in the prior art, and its object is to provide a molten metal leak detection device and a molten metal leak detection method capable of suppressing false detections of molten metal leaks. Another object of this invention is to provide a continuous casting method for metal using a molten metal leak detection method.

[0011] Methods for solving problems

[0012] The methods used to solve the above problems are as follows.

[0013] [1] A molten metal leakage detection device for detecting leakage of molten metal from a molten metal processing equipment, wherein the molten metal leakage detection device comprises: an imaging unit for taking pictures of the molten metal processing equipment at predetermined time intervals to generate image data; a data acquisition unit for acquiring time-series data of the number of pixels representing a specific color contained in a predetermined area of ​​the image data; and a detection unit for using the time-series data to detect a molten metal leakage.

[0014] [2] According to the molten metal leakage detection device of [1], the detection unit uses the time series data to measure the duration for which the number of pixels representing a specific color continuously reaches a reference number or more, and detects that the molten metal leakage occurs when the duration reaches a predetermined threshold or more.

[0015] [3] According to the molten metal leakage detection device of [1], the detection unit uses the time series data to measure the cumulative value of the number of pixels representing a specific color, and detects that the molten metal leakage occurs when the cumulative value within a specified time becomes above a predetermined threshold.

[0016] [4] A molten metal leakage detection device according to any one of [1] to [3], wherein the molten metal processing equipment is a continuous casting machine for metal, and the imaging unit is disposed at a position capable of taking pictures below the casting mold.

[0017] [5] A method for detecting leakage of molten metal, wherein the method includes: an image capture step, wherein the molten metal processing equipment is photographed at predetermined time intervals to generate image data; a data acquisition step, wherein time series data is acquired of the number of pixels representing a specific color contained in a specified area of ​​the image data; and a detection step, wherein the time series data is used to detect leakage of molten metal.

[0018] [6] According to the molten metal leakage detection method of [5], in the detection step, the time series data is used to measure the duration for which the number of pixels representing a specific color continuously reaches a reference number or more, and the leakage of the molten metal is detected when the duration reaches a predetermined threshold or more.

[0019] [7] According to the molten metal leakage detection method of [5], in the detection step, the time series data is used to measure the cumulative value of the number of pixels representing a specific color, and the leakage of the molten metal is detected when the cumulative value within a specified time exceeds a predetermined threshold.

[0020] [8] The method for detecting leakage of molten metal according to any one of [5] to [7], wherein, The molten metal processing equipment is a continuous casting machine for metals. In the shooting step, image data is generated by shooting the area below the mold.

[0021] [9] A continuous casting method for metal, wherein, if a leak of molten metal is detected by the molten metal leakage detection method described in [8], the supply of molten metal to the mold is stopped.

[0022] Invention Effects

[0023] In the molten metal leakage detection device and molten metal leakage detection method of the present invention, time series data representing the number of pixels of a specific color is used to detect molten metal leakage, thereby suppressing false detection of molten metal leakage. Attached Figure Description

[0024] Figure 1 This is a cross-sectional schematic diagram illustrating an example of applying the molten metal leakage detection device of this embodiment to a continuous casting machine for metal with a mold.

[0025] Figure 2 This is a schematic diagram illustrating an example of the configuration of an image analysis device.

[0026] Figure 3It is a graph representing the time-varying number of pixels of a specific color in a continuous casting machine for steel. Detailed Implementation

[0027] The following description uses an example of applying the molten metal leakage detection device of this embodiment to a continuous casting machine for metal. A continuous casting machine for metal is an example of molten metal processing equipment. Other examples of molten metal processing equipment include conveying containers and conveying pots for transporting molten metal; the molten metal leakage detection device of this embodiment can also be applied to these devices.

[0028] First, the continuous casting machine 10 for metal will be described. Figure 1 This is a cross-sectional schematic diagram illustrating an example of applying the molten metal leakage detection device 40 of this embodiment to a continuous casting machine 10 having a mold 12. The continuous casting equipment 10 has a mold 12, an tundish 14 disposed above the mold 12, and a plurality of casting support rollers 16 arranged below the mold 12. Although not shown in the figure, a ladle containing molten metal 18 is disposed above the tundish 14, and molten metal 18 is injected into the tundish 14 from the bottom of the ladle. A sliding gate 19 and an immersion nozzle 20 are disposed at the bottom of the tundish 14. The sliding gate 19 slides, and when the closure of the sliding gate 19 is released, the molten metal 18 is injected into the mold 12 via the immersion nozzle 20. The molten metal 18 dissipates heat from the inner surface of the mold 12 and solidifies, forming a solidified shell 22. Thus, a casting sheet 26 is formed with the solidified shell 22 as the outer shell and having an unsolidified layer 24 composed of molten metal 18 inside.

[0029] Between adjacent casting support rollers 16 in the casting direction, a plurality of secondary cooling zones 28, each equipped with spray nozzles (not shown), are arranged along the casting direction directly below the mold 12. The casting sheet 26 is cooled as it is pulled out by cooling water sprayed from the spray nozzles of the secondary cooling zones 28. During the passage of the casting sheet 26 through the plurality of secondary cooling zones 28 by the casting support rollers 16, the solidified shell 22 is appropriately cooled, and the unsolidified layer 24 solidifies, thus completing the solidification of the casting sheet 26.

[0030] Multiple conveying rollers 17 are arranged downstream in the casting direction to continue conveying the casting sheets 26. Above the conveying rollers 17 is a casting sheet cutter 30 for cutting the casting sheets 26. After solidification, the casting sheets 26 are cut by the casting sheet cutter 30 to continuously cast casting sheets 26a of a specified length.

[0031] In such a continuous casting machine 10, molten metal 18 sometimes leaks from the solidified shell 22 formed on the casting sheet 26 to the outside. The molten metal leakage detection device 40 of this embodiment is used to detect such leakage of molten metal 18 to the outside.

[0032] The molten metal leakage detection device 40 includes a camera 42 and an image analysis device 44. The camera 42 performs an image capture step, for example, by taking images of the continuous casting machine 10, including the casting sheet 26 below the mold 12, every 100 ms to generate image data, and then sends the generated image data to the image analysis device 44. Here, 100 ms is an example of a predetermined time interval.

[0033] The solidified shell 22 of the cast sheet 26 immediately after being pulled from the mold 12 is thin, making it prone to leakage. Therefore, the area captured by the camera 42 is preferably the area directly below the mold 12. However, depending on the structure of the continuous casting machine 10, it is sometimes difficult to capture the area directly below the mold 12. In this case, it is sufficient to capture the area below the mold 12, which is close to the area directly below the mold 12.

[0034] The image analysis device 44 obtains time-series data, i.e., time-varying data, of the number of pixels representing a specific color contained within a defined area of ​​image data acquired from the camera 42. Here, the specific color represents the color of the molten metal 18 contained within the defined area of ​​the image data. The image analysis device 44 uses this time-series data to detect molten metal leakage. The camera 42 is an example of an imaging unit; for example, it is a digital camera or camcorder with a CCD image sensor or a CMOS image sensor capable of generating color image data.

[0035] To capture a full circumference below the mold 12, the molten metal leakage detection device 40 may also have two or more cameras 42. In this case, the image analysis device 44 obtains time-series data, representing the time variation of the number of pixels for a specific color, from the image data acquired by each camera, and uses this time-series data to detect leakage of the molten metal 18.

[0036] Next, the image analysis device 44 will be described. Figure 2 This is a schematic diagram illustrating an example configuration of the image analysis device 44. The image analysis device 44 is, for example, a workstation, a personal computer, or other general-purpose computer. The image analysis device 44 includes a control unit 46, an input unit 48, an output unit 50, and a storage unit 52.

[0037] The control unit 46, such as a CPU, executes various programs stored in the storage unit 52, enabling it to function as both a data acquisition unit 54 and a detection unit 56. The input unit 48 is, for example, a keyboard or a touch panel integrated with the display. The output unit 50 is, for example, an LCD, a CRT display, or a warning light. The storage unit 52 is, for example, an updatable flash memory, a built-in hard drive or memory card connected via a data communication terminal, or other information recording media and their read / write devices. The storage unit 52 stores programs and data for acquiring time-series data from image data received from the camera 42 and using this time-series data to detect leaks of molten metal 18.

[0038] Next, the processing performed by the data acquisition unit 54 and the detection unit 56 will be described. When the data acquisition unit 54 receives image data generated by the camera 42, it performs a data acquisition step, measuring the number of pixels representing a specific color contained within a defined area of ​​the image data. By repeatedly performing the image data reception and this measurement, the data acquisition unit 54 acquires time-series data, i.e., time-varying data, of the time variation in the number of pixels representing a specific color.

[0039] Specifically, when the data acquisition unit 54 receives image data, it reads information representing a predetermined area from the storage unit 52 and determines the predetermined area. The predetermined area is a region set to eliminate unavoidable interference entering the shooting area based on the camera's setting position. The predetermined area is preset during the tuning performed when the molten metal leakage detection device 40 is installed, and information representing the predetermined area is input from the input unit 48 and stored in the storage unit 52.

[0040] Furthermore, the data acquisition unit 54 converts the RGB values ​​of each pixel into HSV values. This is because the range of a specific color is set to the range of HSV values. The range of a specific color can be set to the range of coordinates in a color space such as RGB values ​​or HSV values. In this case, the data acquisition unit 54 converts pixels in accordance with the color space in which the range of the specific color is set.

[0041] The data acquisition unit 54 reads information representing the range of a specific color from the storage unit 52, compares the HSV values ​​of all pixels contained in the specified area with the information representing the range, and measures the number of pixels within the range of the specific color.

[0042] The range of a specific color is predetermined during tuning when the molten metal leakage detection device 40 is installed. Information representing the range of that specific color is input from the input unit 48 and stored in the storage unit 52. The data acquisition unit 54 acquires time-series data representing the time variation of the number of pixels representing a specific color by repeatedly receiving image data and measuring the number of pixels representing a specific color.

[0043] The detection unit 56 performs a detection step, using time-series data—the time variation of the number of pixels representing a specific color—acquired by the data acquisition unit 54 to detect leakage of molten metal 18. Using this time-series data, the detection unit 56 measures the duration for which the number of pixels representing a specific color continuously exceeds a reference number. The detection unit 56 reads information representing the reference number and a threshold from the storage unit 52. If the duration for which the number continuously exceeds the reference number is greater than or equal to the threshold, leakage of molten metal 18 is detected. On the other hand, even if the number of pixels representing a specific color is greater than or equal to the reference number, if the duration for which the number continuously exceeds the reference number is not greater than or equal to a predetermined threshold, the detection unit 56 does not detect leakage of molten metal 18.

[0044] The number of references is, for example, 1000, and the threshold for the duration of the period exceeding the number of references is, for example, 2 seconds. The number of references and the threshold are preset during the tuning performed when the molten metal leakage detection device 40 is installed. Information indicating the number of references and information indicating the threshold are input from the input unit 48 and saved in the storage unit 52.

[0045] Figure 3 It is a graph representing the time-varying number of pixels of a specific color in a continuous casting machine for steel. Figure 3 The horizontal axis represents time (seconds), and the vertical axis represents the number of pixels. Figure 3 The solid line represents the time change in the number of pixels of a specific color when steel leakage or molten steel spillage occurs in a continuous casting machine. Figure 3 The dashed line represents the time variation in the number of pixels of a specific color when there is no leakage of molten steel from the casting and no sparks from the molten steel. The dashed line represents the baseline number (1000) of pixels of a specific color.

[0046] The splashing sparks generated by the splashing of molten steel during normal operation have the same color as the leaked molten steel. For example... Figure 3 As shown, even when sparks of molten steel are generated, the number of pixels of a specific color can sometimes instantly increase to the same number of pixels as when molten steel leaks. Therefore, it can be concluded that if a baseline number of pixels of a specific color is used as a threshold to detect molten steel leakage, then the mere generation of sparks of molten steel will be mistakenly detected as a molten steel leak.

[0047] In contrast, in the molten metal leakage detection device 40 of this embodiment, time-series data representing the time variation of the number of pixels of a specific color is used, and leakage of molten metal 18 is detected by utilizing the duration for which the number of pixels representing a specific color continuously exceeds a reference number. Figure 3As shown, compared to the sudden generation of scattered sparks, if a leak occurs and molten steel leaks, the molten steel continues to flow out. Therefore, by using time-series data representing the time variation of the number of pixels of a specific color, and utilizing the duration for which the number of pixels representing a specific color consistently exceeds a baseline number, the leakage of molten metal 18 can be detected. This allows for the detection of leakage of molten metal 18 while suppressing false detections of molten metal leakage.

[0048] Refer again Figure 2 The detection unit 56 detects leakage of molten metal 18 if the number of pixels representing a specific color remains at or above a reference number for a duration exceeding a threshold. When the detection unit 56 detects leakage of molten metal 18, it causes the LCD or CRT display, which serves as the output unit 50, to display a message indicating that leakage of molten metal 18 has occurred, or it illuminates a warning light. This allows the surrounding environment to be notified that leakage of molten metal 18 has occurred.

[0049] If leakage of molten metal 18 occurs, and molten metal 18 is continuously supplied to the mold 12, the leakage will continue, thus amplifying the damage. Therefore, when the detection unit 56 detects leakage of molten metal 18, it is preferable to reverse the sliding gate 19 to stop the supply of molten metal 18 to the mold 12. In this way, by stopping the supply of molten metal 18 when leakage is detected, the amplification of damage caused by leakage of molten metal 18 can be suppressed in continuous casting methods using a continuous casting machine employing metal.

[0050] As explained above, in the molten metal leakage detection device 40 of this embodiment, time-series data representing the time change of the number of pixels of a specific color is used, and leakage of molten metal 18 is detected by utilizing the duration for which the number of pixels representing a specific color continuously reaches a reference number or more. Therefore, in the molten metal leakage detection device 40 of this embodiment, false detection of molten metal leakage can be suppressed.

[0051] In the above embodiment, an example is shown where the detection unit 56 detects leakage of molten metal 18 based on a duration in which the number of pixels representing a specific color continuously exceeds a reference number, but this is not a limitation. The detection unit 56 may also use time-series data, i.e., time-varying changes in the number of pixels representing a specific color, to measure the cumulative value of the number of pixels representing a specific color, and detect leakage of molten metal 18 based on this cumulative value over a predetermined time period (e.g., 0.5 seconds). In this case, the detection unit 56 accumulates the time-series data, and detects leakage of molten metal 18 when the cumulative value exceeds a predetermined threshold. This threshold over the predetermined time period is also predetermined during tuning performed when the molten metal leakage detection device 40 is set up, and information representing this threshold is input from the input unit 48 and stored in the storage unit 52.

[0052] like Figure 3 As shown, since the increase in the number of pixels of a specific color in the case of sparks from molten steel is instantaneous, the number of pixels of that specific color decreases sharply over time. Therefore, the leakage of molten metal 18 is detected by using the cumulative value of the number of pixels of a specific color in multiple image data sets. Thus, compared to detecting the leakage of molten metal 18 by using the cumulative value of the number of pixels of a specific color in a single image data set, the leakage of molten metal 18 can be detected while suppressing false detections caused by sparks from molten metal 18.

[0053] Furthermore, by using the cumulative value of the number of pixels of a specific color to detect the leakage of molten metal 18, a leakage of molten metal 18 can be detected in a shorter time compared to detecting the leakage of molten metal 18 by using a duration that is longer than a baseline number. The cumulative value of the number of pixels of a specific color in five image data points when molten metal 18 is leaking is significantly different from the cumulative value of the number of pixels of a specific color in five image data points when sparks of molten metal 18 are generated. Therefore, by setting a threshold that can distinguish the cumulative values ​​of both, the leakage of molten metal 18 can be detected, for example, when image data is generated every 100 ms, within 0.5 seconds after the leakage of molten metal 18. Thus, it can be seen that by using the cumulative value of the number of pixels of a specific color in multiple image data points to detect the leakage of molten metal 18, a leakage of molten metal 18 can be detected in a shorter time.

[0054] In the above embodiment, an example of sending image data generated by camera 42 to data acquisition unit 54 was described, but it is not limited to this. Alternatively, the image data generated by camera 42 may be stored in storage unit 52, and data acquisition unit 54 may read the image data stored in storage unit 52 and measure the number of pixels representing a specific color contained in a defined area of ​​the image data.

[0055] Embodiment

[0056] Next, an embodiment will be described in which when continuously casting steel using the continuous casting machine 10 for metals shown in Figure 1 , a molten metal leakage detection device 40 is used to detect leakage of molten steel from the continuous casting machine. In this embodiment, transition data of the number of pixels of a specific color when shooting the scattered sparks of molten steel generated during normal operation and transition data of the number of pixels of a specific color when shooting a breakout (leakage of molten steel) are each prepared for ten samples. For the ten prepared samples, the breakout is detected by the following method.

[0057] Pixels having HSV values that satisfy a specific color: H < 45 or H > 190, 50 < S < 150, and 50 < V < 150 are defined as the specific color (0 ≤ H, S, V ≤ 256).

[0058] Existing example: When the number of pixels representing the specific color is greater than or equal to a reference number (1000), a breakout is detected.

[0059] Inventive example: When the duration for which the number of pixels representing the specific color continuously remains greater than or equal to the reference number (1000) is 2 seconds or more, a breakout is detected.

[0060] The detection results of the above existing example and inventive example are shown in Table 1 below.

[0061] [Table 1]

[0062] As shown in Table 1, in the existing example, scattered sparks caused by molten steel splashes generated during 4 / 10 of normal operations were erroneously detected as breakouts. In contrast, in the inventive example, scattered sparks caused by molten steel splashes generated during normal operations were not erroneously detected as breakouts. Based on this result, it was confirmed that by using the molten metal leakage detection device of the present embodiment, which uses time series data of the temporal change in the number of pixels representing a specific color to detect leakage of molten metal 18, false detection of molten metal leakage can be suppressed.

[0063] Reference Numeral Explanation

[0064] 10 Continuous casting machine for metals

[0065] 12 Mold

[0066] 14 Tundish

[0067] 16 Ingot support roll

[0068] 17 Conveyor roll

[0069] 18 Molten metal

[0070] 19 Sliding gate

[0071] 20 Dipping nozzles

[0072] 22. Solidified shell

[0073] 24 Uncured layer

[0074] 26 castings

[0075] 26a casting

[0076] 28 Secondary cooling zone

[0077] 30 Casting Sheet Cutting Machine

[0078] 40. Molten metal leakage detection device

[0079] 42 cameras

[0080] 44 Image Analysis Device

[0081] 46 Control Department

[0082] 48 Input Section

[0083] 50 Output Section

[0084] 52 Preservation Department

[0085] 54 Data Acquisition Department

[0086] 56. Testing Department.

Claims

1. A molten metal leakage detection device for detecting leakage of molten metal from molten metal processing equipment, wherein, The molten metal leakage detection device includes: The imaging unit takes pictures of the molten metal processing equipment at predetermined time intervals to generate image data; The data acquisition unit acquires time-series data of the number of pixels representing a specific color contained within a defined area of ​​the image data; and The detection unit uses the time-series data to detect leaks of molten metal.

2. The molten metal leakage detection device according to claim 1, wherein, The detection unit uses the time series data to measure the duration for which the number of pixels representing a specific color continuously exceeds a reference number, and detects leakage of the molten metal when the duration exceeds a predetermined threshold.

3. The molten metal leakage detection device according to claim 1, wherein, The detection unit uses the time series data to measure the cumulative number of pixels representing a specific color, and detects a leak of molten metal if the cumulative value exceeds a predetermined threshold within a specified time.

4. The molten metal leakage detection device according to any one of claims 1 to 3, wherein, The molten metal processing equipment is a continuous casting machine for metal. The camera unit is positioned to take pictures of the area below the mold.

5. A method for detecting leakage of molten metal, wherein, The method for detecting leakage of molten metal includes: The imaging step involves taking pictures of the molten metal processing equipment at predetermined time intervals to generate image data; The data acquisition step involves acquiring time-series data of the number of pixels representing a specific color contained within a defined area of ​​the image data; and The detection step uses the time series data to detect leaks of molten metal.

6. The method for detecting leakage of molten metal according to claim 5, wherein, In the detection step, the time series data is used to measure the duration for which the number of pixels representing a specific color continuously exceeds a baseline number, and leakage of the molten metal is detected when the duration exceeds a predetermined threshold.

7. The method for detecting leakage of molten metal according to claim 5, wherein, In the detection step, the time series data is used to measure the cumulative number of pixels representing a specific color, and leakage of the molten metal is detected when the cumulative value exceeds a predetermined threshold within a specified time.

8. The method for detecting leakage of molten metal according to any one of claims 5 to 7, wherein, The molten metal processing equipment is a continuous casting machine for metals. In the shooting step, image data is generated by shooting the area below the mold.

9. A continuous casting method for metal, wherein, If a leak of molten metal is detected by the molten metal leakage detection method according to claim 8, the supply of molten metal to the mold shall be stopped.

Images