Anomaly detection system and method for detecting abnormalities in a target material detector

Abstract

To provide an abnormality determination system and an abnormality determination method for a target material detector capable of accurately determining the occurrence of an abnormality in a target material detector on a production line.SOLUTION: An abnormality determination system for a target material detector 20 that detects the front and rear end positions of a target material by detecting the infrared radiation of the target material S conveyed on a production line, includes: a reference measuring device 30 that is provided together with the target material detector in the same conveying section where there is no cooling device for cooling the target material, and measures the infrared radiation of the target material as a reference radiation intensity; and an abnormality determination device 40 that determines an abnormality in the target material detector. The abnormality determination device includes: a data acquisition unit that, during operation, acquires reference data based on the infrared radiation intensity detected by the target material detector and the reference radiation intensity detected by the reference measuring device as operation-time data; a deviation degree calculation unit that calculates a deviation degree of the operation-time data on the basis of the previously acquired normal-time data; and an abnormality determination unit that determines an abnormality in the target material detector on the basis of the deviation degree.SELECTED DRAWING: Figure 3

Description

Technical Field

[0001] The present invention relates to an abnormality determination system and an abnormality determination method for a target material detector that determines an abnormality of a target material detector in a production line.

Background Art

[0002] In a production line such as a hot rolling line, a target material detector (tracking sensor) is installed at a predetermined position in the production line in order to detect (track) the positions of the leading end and the trailing end of a target material such as a rolled material to be conveyed. As the target material detector, an HMD (Hot Metal Detector) that detects the radiation intensity of infrared radiation of the target material is widely used. The HMD detects the presence or absence of the target material by detecting infrared radiation corresponding to a temperature of 400 ° C or higher emitted from the target material.

[0003] When the leading end of the target material passes through the detection position, the target material detector can detect that the leading end of the target material has passed because the detected state changes from the off state to the on state (starting the on operation). Further, when the trailing end of the target material passes through the detection position, the target material detector can detect that the trailing end of the target material has passed because the detected state changes from the on state to the off state (starting the off operation).

[0004] In addition, a target material detector that detects the passage of a target material by the above-described on / off operation may malfunction due to factors such as a decrease in sensitivity due to long-term use. When the target material detector malfunctions, tracking abnormalities occur, affecting the operation and setting calculation timing of equipment such as rolling mills and cooling devices, causing a sudden stop of the production line, and in some cases inducing damage to the equipment.

[0005] Therefore, if an abnormality of the target material detector can be grasped at an early stage, it becomes possible to take measures such as early replacement of the target material detector, and a sudden stop of the production line can be prevented. For this reason, an abnormality determination method (abnormality diagnosis method) for early grasping an abnormality of the target material detector has been conventionally performed.

[0006] For example, Patent Document 1 discloses a method for determining abnormalities using at least three tracking sensors (first HMD, second HMD, third HMD) installed on a rolling line. Specifically, it discloses a method for determining whether or not there is an abnormality in the tracking sensors by calculating a first travel distance (time from the start of ON operation of the first HMD to the start of ON operation of the second HMD × speed) and a second travel distance (time from the start of ON operation of the second HMD to the start of ON operation of the third HMD × speed).

[0007] Patent Document 2 discloses a method for determining abnormalities using tracking sensors installed at two locations at a predetermined interval. Specifically, the method calculates the time when the downstream tracking sensor starts on or off operation from the time when the upstream tracking sensor starts on or off operation and the transport speed of the rolled material. Then, the method discloses a method for determining that the tracking sensor is abnormal if the time difference between the calculated on or off operation start time and the actual time when the downstream tracking sensor started on or off operation exceeds an acceptable range. [Prior art documents] [Patent Documents]

[0008] [Patent Document 1] Japanese Patent Application Publication No. 9-218085 [Patent Document 2] Japanese Patent Publication No. 2011-50990 [Overview of the project] [Problems that the invention aims to solve]

[0009] However, the above-mentioned conventional technologies have the following problems.

[0010] The method disclosed in Patent Document 1 requires the use of at least three tracking sensors to determine abnormalities, which presents problems in terms of equipment costs. Furthermore, in the method disclosed in Patent Document 2, when the transport speed of the rolled material changes between two tracking sensors installed at a predetermined interval, the time difference between the on or off operation of the upstream and downstream tracking sensors changes. This necessitates setting a large tolerance range for the time difference, resulting in problems with the accuracy of abnormality determination.

[0011] The present invention has been made in view of the above circumstances, and its object is to provide a target material detector abnormality detection system and abnormality detection method that can accurately determine the occurrence of abnormalities in the target material detector in a manufacturing line. [Means for solving the problem]

[0012] [1] An abnormality determination system for a target material detector that detects the position of the leading and trailing ends of a target material by detecting infrared radiation emitted by the target material being transported in a manufacturing line, comprising: a reference measuring instrument provided together with the target material detector in the same transport section where no cooling device for cooling the target material is provided, and which measures the infrared radiation of the target material as a reference radiation intensity; and an abnormality determination device for determining an abnormality of the target material detector, wherein the abnormality determination device includes: a data acquisition unit that acquires reference data as operational data during operation based on the infrared radiation intensity of the infrared radiation of the target material detected by the target material detector and the reference radiation intensity measured by the reference measuring instrument; a deviation degree calculation unit that calculates the degree of deviation of the operational data from the normal data based on normal data of the target material detector and the reference measuring instrument acquired in advance; and an abnormality determination unit that determines an abnormality of the target material detector based on the deviation degree. [2] The reference measuring instrument is a radiation thermometer, an abnormality determination system for the target material detector as described in [1]. [3] A method for determining an abnormality in a target material detector, which detects the positions of the leading and trailing ends of a target material by detecting infrared radiation emitted by the target material being transported on a manufacturing line, comprising: a data acquisition step of acquiring reference data as operational data based on the infrared radiation intensity of the infrared radiation of the target material detected by the target material detector and the reference radiation intensity measured by a reference measuring instrument installed together with the target material detector in the same transport section where no cooling device for cooling the target material is installed; a deviation degree calculation step of calculating the deviation degree of the operational data from the normal data based on normal data of the target material detector and the reference measuring instrument acquired in advance; and an abnormality determination step of determining an abnormality in the target material detector based on the deviation degree. [4] The method for determining an abnormality of a target material detector as described in [3], wherein in the step of calculating the degree of deviation, the degree of deviation of the operating data from the normal data is calculated based on the principal components identified by the principal component analysis of the normal data. [5] The method for determining an abnormality of a target material detector according to [3] or [4], wherein in the abnormality determination step, it is determined that an abnormality has occurred in the target material detector if the degree of deviation exceeds a preset threshold. [6] The method for determining an abnormality of a target material detector according to [3] or [4], wherein in the abnormality determination step, it is determined that an abnormality has occurred in the target material detector if the number of times the deviation exceeds a preset threshold exceeds a predetermined number of times. [Effects of the Invention]

[0013] According to the present invention, it is possible to accurately determine the occurrence of abnormalities in the target material detector on the manufacturing line. [Brief explanation of the drawing]

[0014] [Figure 1] This figure shows a schematic side view of an example of a manufacturing facility. [Figure 2] This diagram schematically shows the configuration of the target material detector. [Figure 3]It is a diagram schematically showing the configuration etc. of a reference measuring device. [Figure 4] It is a diagram schematically showing the configuration etc. of an abnormality determination device. [Figure 5] It is a diagram showing the correlation between the infrared radiation intensity acquired by a target material detector and the reference radiation intensity acquired by a reference measuring device. [Figure 6] It is a diagram showing the principal component vector in the correlation between the infrared radiation intensity acquired by a target material detector and the reference radiation intensity acquired by a reference measuring device. [Figure 7] It is a diagram showing the result of calculating the deviation degree (Q statistic) for a plurality of target materials during operation.

Mode for Carrying Out the Invention

[0015] <First Embodiment> Hereinafter, the first embodiment of the present invention will be specifically described. FIG. 1 shows a schematic side view as an example of manufacturing equipment 1. The manufacturing equipment 1 includes a manufacturing line 10 such as a hot rolling line, an abnormality determination system 50, a control device 60, a control computer 70, and a host computer 80. The manufacturing line 10 includes a heating furnace 11, an oxide removal device 12 which is a descaling device, a width reduction device 13, a rough rolling mill 14, a finishing rolling mill 15, a cooling device 16 which is a water cooling device, and a coiling device 17. The target material S may be a rolled material such as a slab or a steel plate after casting.

[0016] In manufacturing line 10, the target material S (not shown) is charged into the heating furnace 11, heated to a predetermined set temperature, and then extracted from the heating furnace 11 as a hot slab. After the primary scale formed on the surface of the target material S extracted from the heating furnace 11 is removed by the oxide removal device 12, the width of the target material S is reduced to a predetermined set width by the width reduction device 13. The target material S with reduced width is rolled to a predetermined thickness in the rough rolling mill 14 and then conveyed to the finishing rolling mill 15 as a rough bar. The rough rolling mill 14 may have a reversible rolling mill 14a and a non-reversible rolling mill 14b. The target material S is rolled to the product thickness by a continuous rolling mill with 5 to 7 rolling stands in the finishing rolling mill 15. On the downstream side of the finishing rolling mill 15, a cooling device 16 is provided in a facility called a run-out table. The target material S is cooled to a predetermined temperature by the cooling device 16 and then wound into a coil by the coiling device 17.

[0017] The manufacturing line 10 has a conveying mechanism 18 for conveying the target material S from the heating furnace 11 to the coiling device 17 in the conveying direction A of the target material S. As shown in FIGS. 2 and 3, the conveying mechanism 18 may be configured to arrange a plurality of conveying rolls.

[0018] The heating furnace 11 heats the target material S to 1100 - 1300°C. The temperature of the target material S becomes 950 - 1150°C at the stage when the rough rolling by the rough rolling mill 14 is completed, and becomes 850 - 1200°C at the stage when it is charged into the finishing rolling mill 15. The temperature of the target material S becomes 700 - 1000°C at the stage when it is unloaded from the finishing rolling mill 15, and becomes 400 - 800°C before being wound by the coiling device 17. Note that an induction heating device may be arranged on the upstream side of the finishing rolling mill 15. In this case, the temperature of the target material S at the stage when it is charged into the finishing rolling mill 15 is higher than the temperature of the target material S at the stage when the rough rolling by the rough rolling mill 14 is completed.

[0019] The control device 60 controls each piece of equipment that makes up the manufacturing line 10 as a PLC. The control computer 70 transmits control signals to the control device 60 as a process computer. The higher-level computer 80 transmits control signals to the control computer 70 and the control device 60 for manufacturing the target material S on the manufacturing line 10.

[0020] The control computer 70 sets control target values ​​for each piece of equipment constituting the manufacturing line 10 based on the control signals transmitted from the higher-level computer 80, and transmits these control target values ​​to the control device 60. The control device 60 sets the operating conditions for each piece of equipment based on the control target values ​​transmitted from the control computer 70. Based on the set operating conditions for each piece of equipment, the control device 60 performs control of the plate thickness and plate width of the target material S in the manufacturing line 10.

[0021] The control device 60 collects information from various sensors (object detector, reference measuring instrument, thermometer, plate thickness gauge, etc.) installed on the manufacturing line 10 at a predetermined sampling period. The control device 60 transmits the collected information from the various sensors to the control computer 70.

[0022] As shown in Figure 1, multiple target material detectors 20 are arranged in the manufacturing line 10. The target material detectors 20 are placed at multiple locations in the manufacturing line 10, between the heating furnace 11 and the finishing rolling mill 15. The target material detectors 20 detect the positions of the leading and trailing ends of the target material S by detecting the infrared radiation emitted by the target material S being transported in the manufacturing line 10.

[0023] Specifically, as shown in Figure 1, the multiple target material detectors 20 are arranged as follows: target material detector 20a is located downstream of the heating furnace 11, target material detector 20b is located downstream of the oxide removal device 12, target material detector 20c is located downstream of the width reduction device 13, target material detector 20d is located downstream of the reversible rolling mill 14a, and target material detector 20e is located downstream of the irreversible rolling mill 14b. In addition, target material detector 20f is located downstream of target material detector 20e and between the irreversible rolling mill 14b and the finishing rolling mill 15. In the embodiment shown in Figure 1, the multiple target material detectors 20 are located between the heating furnace 11 and the finishing rolling mill 15, but it is also possible to configure the target material detectors 20 to be located downstream of the finishing rolling mill 15.

[0024] In the manufacturing line 10, reference measuring instruments 30 for measuring the temperature of the target material S are placed at multiple locations. Specifically, as shown in Figure 1, reference measuring instrument 30a is placed upstream of the width reduction device 13, reference measuring instrument 30b is placed between the reversible rolling mill 14a and the irreversible rolling mill 14b, reference measuring instrument 30c is placed downstream of the irreversible rolling mill 14b, and reference measuring instrument 30d is placed downstream of reference measuring instrument 30c and upstream of the finishing rolling mill 15. The reference measuring instrument 30 may be an infrared thermometer for measuring the temperature of the target material S.

[0025] Here, the reference measuring instrument 30a may be used to specify the operating conditions of the downstream width reduction device 13. In addition, multiple reference measuring instruments 30 may be used to specify the operating conditions of the roughing mill 14 and the finishing mill 15 in the production line 10. For example, the reference measuring instrument 30d measures the surface temperature of the target material S before finishing rolling. The surface temperature of the target material S measured by the reference measuring instrument 30d may be used in the control device 60 to calculate various setting values ​​such as the roll gap when the target material S is fed into the finishing mill 15. Furthermore, the surface temperature of the target material S measured by the reference measuring instrument 30d also serves to initiate the setting calculation in the finishing mill 15 and to transmit the temperature data of the target material S to the control device 60 and the control computer 70. This role may also be performed by using the surface temperature of the target material S measured by the reference measuring instrument 30c located downstream of the roughing mill 14.

[0026] Next, the configuration of the target material detector 20 will be explained using Figure 2. Figure 2 is a schematic diagram showing the configuration of the target material detector 20. As shown in Figure 2, the target material detector 20 has an infrared radiation detection unit 21, a detection value amplification unit 22, and a detection value calculation unit 23.

[0027] The infrared radiation detection unit 21 detects infrared radiation (wavelength 0.8 to 1000 μm) and converts the detected infrared radiation intensity into an electrical signal. The infrared radiation detection unit 21 is, for example, a thermoelectric element such as a thermopile or a pyroelectric infrared sensor (pyrosensor), or a photoelectric element using a photodiode or phototransistor. As the photoelectric element, any of the following elements may be used: lead selenide (PbSe), lead sulfide (PbS), indium gallium arsenide (InGaAs), or silicon (Si). The peak wavelengths for measurement of each photoelectric element are 4 μm for lead selenide, 2 μm for lead sulfide, 1.55 μm for indium gallium arsenide, and 0.9 μm for silicon. The infrared radiation detection unit 21 detects the received intensity of infrared radiation at a specific wavelength according to the corresponding peak wavelength and converts it into an electrical signal. Furthermore, a germanium (Ge) photodiode may be used as the element of the infrared radiation detection unit 21. In this case, the measurement wavelength will be 0.8 μm to 1.8 μm.

[0028] The detected value amplification unit 22 amplifies the electrical signal of the infrared radiation intensity detected by the infrared radiation detection unit 21 and outputs it to the detected value calculation unit 23.

[0029] The detection value calculation unit 23 calculates whether the target material S has reached the detection position M on the manufacturing line 10 where the target material detector 20 performs detection, based on the infrared radiation intensity signal detected by the infrared radiation detection unit 21. Specifically, as shown in Figure 2, a threshold is set in advance for the time-series signal of infrared radiation intensity input to the detection value calculation unit 23. Then, the detection value calculation unit 23 determines that the target material S is at the detection position M (ON state) if the infrared radiation intensity is greater than or equal to the threshold, and determines that the target material S is not at the detection position M (OFF state) if the value is less than the threshold.

[0030] Specifically, the detection value calculation unit 23 detects that the tip of the target material S has reached the detection position M when the state of the target material S at the detection position M changes from the OFF state to the ON state. On the other hand, it detects that the tail end of the target material S has passed the detection position M when the state of the target material S at the detection position M changes from the ON state to the OFF state. The state of the target material S at the detection position M detected by the detection value calculation unit 23 is transmitted to the control device 60. Based on the reception of the state of the target material S transmitted by the detection value calculation unit 23 (target material detector 20), the control device 60 controls the operation of each piece of equipment in the manufacturing line 10.

[0031] Next, the configuration of the reference measuring instrument 30 will be explained using Figure 3. Figure 3 is a schematic diagram showing the configuration of the reference measuring instrument 30. As shown in Figure 3, the reference measuring instrument 30 has a measuring unit 31, a measurement value amplification unit 32, and a measurement value calculation unit 33.

[0032] The measurement unit 31 measures infrared radiation (wavelength 0.8 to 1000 μm) and converts the measured infrared radiation intensity into an electrical signal. Like the infrared radiation detection unit 21, the measurement unit 31 is a thermoelectric element such as a thermopile or pyroelectric infrared sensor (pyrosensor), or a photoelectric element such as a photodiode or phototransistor. The measurement unit 31 detects the received intensity of infrared radiation at a specific wavelength according to the corresponding peak wavelength and converts it into an electrical signal.

[0033] The measurement value amplification unit 32 amplifies the electrical signal of the infrared radiation intensity measured by the measurement unit 31 and outputs it to the measurement value calculation unit 33.

[0034] The measurement value calculation unit 33 converts the signal measured by the measurement unit 31 and amplified by the measurement value amplification unit 32 into temperature information, based on the predetermined relationship between infrared radiation intensity and the temperature of the target material S, and the emissivity of the target material S.

[0035] The temperature information of the target material S at the measurement position N, calculated by the measurement value calculation unit 33, is transmitted to the control device 60. When the control device 60 receives the temperature information of the target material S transmitted by the measurement value calculation unit 33 (reference measuring instrument 30), it sets the operation of each piece of equipment in the manufacturing line 10 based on that temperature information.

[0036] The abnormality detection system for a target material detector according to the present invention uses a reference measuring instrument 30, which is also placed on the manufacturing line 10 and measures the infrared radiation intensity of the target material S, in order to determine if the target material detector 20 is abnormal. The reference measuring instrument 30 may be a radiation thermometer capable of measuring the infrared radiation intensity of the target material S.

[0037] The reference measuring instrument 30 transmits temperature information used for various controls such as thickness control and temperature control of the target material S to the control device 60 and the control computer 70 in the manufacturing line 10, and is therefore strictly managed and maintained. In particular, the reference measuring instrument 30 is installed in an environment where the risk of malfunction due to factors such as the generation of water vapor in the surroundings, a decrease in light intensity, and the inclusion of foreign matter is reduced, and is adjusted frequently in accordance with quality control requirements. Therefore, by using the reference measuring instrument 30 for the target material detector 20 that is subject to abnormality judgment, abnormalities in the target material detector 20 can be determined with high accuracy.

[0038] Furthermore, the abnormality determination system for the target material detector according to the present invention includes an abnormality determination device 40. During operation, as shown in Figure 3, the abnormality determination device 40 acquires the infrared radiation intensity detected by the infrared radiation detection unit 21 of the target material detector 20 and amplified by the detection value amplification unit 22. In addition, during operation, the abnormality determination device 40 acquires reference data based on the reference radiation intensity measured by the measurement unit 31 of the reference measuring instrument 30. The reference data refers to the reference radiation intensity measured by the measurement unit 31 of the reference measuring instrument 30 and amplified by the measurement value amplification unit 32, or the temperature of the target material S calculated based on the reference radiation intensity amplified by the measurement value amplification unit 32. That is, the reference measuring instrument 30 may be a radiation thermometer to which the reference radiation intensity or the temperature of the target material S calculated based on the reference radiation intensity can be applied as reference data to be transmitted to the abnormality determination device 40.

[0039] Here, it is preferable that the target material detector 20 and the reference measuring instrument 30 are both provided in the same transport section where a cooling device 16 for cooling the target material S is not provided. The transport section in the manufacturing line 10 where a cooling device 16 is not provided is a section in which the temperature of the target material S does not drop by 50°C or more during the transport process.

[0040] Specifically, in the embodiment shown in Figure 1, the transport section between the oxide removal device 12 and the reversible rolling mill 14a, the transport section between the reversible rolling mill 14a and the irreversible rolling mill 14b, and the transport section between the irreversible rolling mill 14b and the finishing rolling mill 15 correspond to transport sections where a cooling device is not installed. If a cooling device is installed upstream of the finishing rolling mill 15, the section between the irreversible rolling mill 14b and the cooling device corresponds to a transport section where a cooling device is not installed. In other words, the configuration in which the target material detector 20 and the reference measuring instrument 30 are installed in the same transport section in the manufacturing line 10 where the cooling device 16 is not installed means, for example, a configuration in which both the target material detector 20 and the reference measuring instrument 30 are installed in the transport section between the oxide removal device 12 and the reversible rolling mill 14a.

[0041] In the manufacturing line 10, the infrared radiation intensity in the wavelength range of infrared radiation emitted from the target material S changes with changes in the temperature of the target material S. Furthermore, when the target material detector 20 and the reference measuring instrument 30 are both placed in the same transport section where no cooling device is installed, a correlation exists between the output of the infrared radiation detection unit 21 and the output of the measuring unit 31. For this reason, the infrared radiation detection unit 21 in the target material detector 20 and the measuring unit 31 in the reference measuring instrument 30 are not limited to using the same elements. In other words, by installing both the target material detector 20 and the reference measuring instrument 30 in the same transport section as described above, abnormalities in the target material detector 20 can be accurately determined based on the correlation between their outputs.

[0042] Next, the configuration of the abnormality detection device 40 will be explained using Figure 4. Figure 4 is a schematic diagram showing the configuration of the abnormality detection device 40. As shown in Figure 4, the abnormality detection device 40 has a data acquisition unit 41, a deviation degree calculation unit 42, an abnormality detection unit 43, and a normal data processing unit 44. The abnormality detection system for the target material detector according to the present invention has the target material detector 20 and the reference measuring instrument 30 installed in the same transport section as described above, and has an abnormality detection device 40 that determines abnormalities in the target material detector 20 using values ​​acquired by the infrared radiation detection unit 21 (target material detector 20) and the measurement unit 31 (reference measuring instrument 30).

[0043] Here, the anomaly detection device 40 may be composed of a general-purpose device such as a personal computer or workstation having a CPU, ROM, and RAM. The ROM stores data acquired by the infrared radiation detection unit 21 (target material detector 20) and the measurement unit 31 (reference measuring instrument 30), and stores a control program for executing the anomaly detection method for the target material detector according to the present invention, as well as processing programs and processing data that control the processing of the entire anomaly detection system. The CPU controls the processing of the anomaly detection system according to the control program and processing programs stored in the ROM. The RAM temporarily stores processing programs and processing data related to the processing executed by the CPU and functions as the CPU's working area. In addition to the normal data from the infrared radiation detection unit 21 (target material detector 20) and the measurement unit 31 (reference measuring instrument 30), the ROM also stores statistical features of the normal data identified by the normal data identification unit 46 (described later) and thresholds used for anomaly detection.

[0044] The data acquisition unit 41 acquires reference data as operational data during operation of the manufacturing line 10, based on the infrared radiation intensity detected by the infrared radiation detection unit 21 (target material detector 20) (hereinafter also referred to as "operation detection data") and the reference radiation intensity measured by the measurement unit 31 (reference measuring instrument 30) (hereinafter also referred to as "operation measurement data").

[0045] The deviation calculation unit 42 calculates the degree of deviation of the operational data from the normal data based on the normal data of the target material detector 20 and the reference measuring instrument 30 that have been acquired in advance. Here, the normal data refers to the infrared radiation intensity detected by the infrared radiation detection unit 21 (target material detector 20) (hereinafter also referred to as "normal detection data") and the reference radiation intensity measured by the measuring unit 31 (reference measuring instrument 30) (hereinafter also referred to as "normal measurement data") when the target material detector 20 is operating normally (when no abnormality occurs).

[0046] In other words, the normal operation data consists of multiple datasets, each dataset containing normal detection data and normal measurement data when the target material detector 20 is functioning normally. Furthermore, the state in which the target material detector 20 is functioning normally refers to a state in which, for a certain period of time during which the manufacturing line 10 is in operation, the control device 60 that controls the manufacturing line 10 does not issue any errors or alarms.

[0047] The abnormality determination unit 43 determines an abnormality in the target material detector 20 based on the degree of deviation calculated by the degree of deviation calculation unit 42.

[0048] The normal data processing unit 44 includes a normal data storage unit 45 and a normal data identification unit 46. The normal data storage unit 45 pre-stores multiple datasets, with normal detection data and normal measurement data forming a single dataset. The normal data identification unit 46 identifies the statistical features of the normal data. The statistical features of the normal data refer to indicators that represent the correlation between the normal detection data and the normal measurement data that constitute the normal data. The normal data processing unit 44 transmits the identified statistical features of the normal data to the deviation calculation unit 42.

[0049] Here, the method for determining abnormalities in the target material detector 20 of the abnormality detection device 40 will be specifically explained using Figure 5. Figure 5 is a diagram showing the correlation between the infrared radiation intensity acquired by the target material detector 20 and the reference radiation intensity acquired by the reference measuring instrument.

[0050] The normal operation data processing unit 44 acquires normal detection data and normal measurement data via the data acquisition unit 41 when the target material detector 20 is operating normally, and stores the acquired normal detection data and normal measurement data in the normal operation data storage unit 45. The number of normal operation data sets stored in the normal operation data storage unit 45 should be between 100 and 10,000. If the number of normal operation data sets is less than 100, the accuracy of the abnormality detection of the target material detector 20 may decrease. Also, if the number of normal operation data sets exceeds 10,000, it becomes difficult to improve the accuracy of the abnormality detection of the target material detector 20.

[0051] The normal data identification unit 46 identifies the statistical features of the normal data when the normal data is stored in the normal data storage unit 45. Specifically, as shown in Figure 5, for example, the normal data identification unit 46 plots the relationship between the infrared radiation intensity acquired by the target material detector 20 and the reference radiation intensity acquired by the reference measuring instrument 30 as a correlation graph, and identifies the statistical features of the normal data.

[0052] In Figure 5, which shows the correlation between the infrared radiation intensity acquired by the target material detector 20 and the reference radiation intensity acquired by the reference measuring instrument 30, the values ​​on the vertical and horizontal axes represent the count. The count is a value obtained by converting the voltage values ​​corresponding to the infrared radiation intensity detected by the target material detector 20 and the reference radiation intensity measured by the reference measuring instrument 30 into integer values ​​for internal computer processing. The count is a numerical value (integer) that has a one-to-one correspondence with the voltage values ​​corresponding to the infrared radiation intensity detected by the target material detector 20 and the reference radiation intensity measured by the reference measuring instrument 30.

[0053] As shown in Figure 5, the normal detection data and normal measurement data that constitute the normal data have a high correlation. Therefore, the normal data identification unit 46 may identify equation (1) below by performing a multiple regression analysis on the relationship between the normal detection data and normal measurement data that constitute the normal data. Here, Y is the normal detection data for the target material detector 20, X is the normal measurement data for the reference measuring instrument 30, and α and β are the regression coefficients in equation (1). The normal detection data Y and normal measurement data X may be the voltage values ​​corresponding to the infrared radiation intensity detected by the target material detector 20 and the reference radiation intensity measured by the reference measuring instrument 30, or they may be the counts converted to integer values ​​for internal processing by the computer. Y = αX + β ... (1)

[0054] Furthermore, the standard deviation σ may be calculated as the variability in the normal data based on the multiple regression analysis of the normal detection data and the normal measurement data that constitute the normal data. That is, the normal data identification unit 46 may identify the regression coefficients α and β of equation (1) and the standard deviation σ as statistical features of the normal data.

[0055] Next, during operation, the data acquisition unit 41 acquires operation detection data and operation measurement data that constitute the operation data. Then, the deviation calculation unit 42 calculates the degree of deviation of the operation data from the normal data. Specifically, with respect to the target material detector 20, the deviation of the operation detection data of the operation data from the normal detection data.

[0056] The degree of deviation may be defined as the amount of deviation from the predicted value obtained using equation (1) for the operational data (operational detection data) acquired during operation. Specifically, the operational measurement data of the reference measuring instrument 30, which is acquired as operational data, is input into equation (1) to obtain the predicted value as the output Y of the target material detector 20. Then, as shown in Figure 5, the absolute value of the difference between the predicted value obtained from equation (1) and the output of the target material detector 20 (operational detection data) acquired as operational data is calculated and output to the abnormality determination unit 43 as the degree of deviation.

[0057] Here, the operational measurement data X input to equation (1) may be a voltage value corresponding to the reference radiation intensity measured by the reference measuring instrument 30, or a count number converted to an integer value for internal computer processing. Then, the operational detection data Y calculated in equation (1) may be applied as a voltage value corresponding to the infrared radiation intensity that can be detected by the target material detector 20, or as a count number that can be converted to an integer value for internal computer processing.

[0058] More specifically, if the voltage value corresponding to the reference radiation intensity measured by the reference measuring instrument 30 is used as the operational measurement data X input to equation (1), then in equation (1), the voltage value corresponding to the infrared radiation intensity that can be detected by the target material detector 20 may be calculated as the operational detection data Y. Then, the absolute value (deviation) of the difference between the calculated voltage value and the voltage value corresponding to the infrared radiation intensity at the target material detector 20 acquired as operational data may be determined and output to the abnormality determination unit 43 as the degree of deviation.

[0059] Furthermore, if the count number corresponding to the reference radiation intensity measured by the reference measuring instrument 30 is used as the operational measurement data X input to equation (1), then the count number corresponding to the infrared radiation intensity that can be detected by the target material detector 20 may be calculated as the operational detection data Y in equation (1). Then, the absolute value (deviation amount) of the difference between the calculated count number and the count number corresponding to the infrared radiation intensity at the target material detector 20 acquired as operational data may be determined and output to the abnormality determination unit 43 as the degree of deviation.

[0060] The abnormality determination unit 43 determines an abnormality in the target material detector 20 based on the deviation degree calculated by the deviation degree calculation unit 42. The abnormality determination unit 43 may determine that an abnormality has occurred in the target material detector 20 if the deviation degree calculated by the deviation degree calculation unit 42 exceeds a preset threshold. On the other hand, if the calculated deviation degree is less than or equal to the preset threshold, the abnormality determination unit 43 may determine that the target material detector 20 is normal.

[0061] The threshold may be set within the range of 3σ to 20σ based on the standard deviation σ identified by the normal data identification unit 46. Preferably, the threshold is set within the range of 3σ to 10σ based on the standard deviation σ identified by the normal data identification unit 46. By setting the threshold to a small value, it is possible to reliably determine the occurrence of an abnormality in the target material detector 20 even if the deviation of the operation detection data from the normal detection data is small. On the other hand, by setting the threshold to a large value, it is possible to determine sudden abnormalities occurring in the target material detector 20, such as loss of output due to a break in the signal system of the target material detector 20.

[0062] The abnormality determination unit 43 may determine that an abnormality has occurred in the target material detector 20 if the number of times the deviation calculated by the deviation calculation unit 42 exceeds a preset threshold exceeds a predetermined number of times. On the other hand, if the number of times the calculated deviation exceeds a preset threshold is less than or equal to a predetermined number of times, the target material detector 20 may be determined to be normal.

[0063] The area surrounding the target material S being transported on the manufacturing line 10 may contain water vapor or other disturbances that affect the infrared radiation intensity detected by the target material detector 20. Furthermore, operational problems on the manufacturing line 10 may interrupt the transport of the target material S, causing its temperature to drop unintentionally. In such cases, the abnormality detection unit 43 may incorrectly determine an abnormality even though no abnormality has occurred in the target material detector 20. In other words, the abnormality detection unit 43 may make a false determination.

[0064] To prevent this, it is preferable to determine that an abnormality has occurred in the target material detector 20 when the number of times the deviation exceeds a preset threshold exceeds a predetermined number. This prevents misjudgments that may occur due to sudden phenomena caused by disturbances such as water vapor or operational troubles, and ensures reliable detection of abnormalities occurring in the target material detector 20. Furthermore, since it is possible to detect abnormalities in which the deviation gradually increases due to deterioration of the target material detector 20, it is possible to detect not only the occurrence of an abnormality but also signs of an impending abnormality.

[0065] From the above perspective, it is preferable that the threshold is set from the range of 3σ to 20σ based on the standard deviation σ identified by the normal data identification unit 46, and that the predetermined number of times the threshold is exceeded is set from the range of 3 to 20 times. If the predetermined number is set to less than 3 times, it may be difficult to exclude false judgments. Also, if the predetermined number is set to 20 times or more, it may not be possible to determine an abnormality in the target material detector 20.

[0066] If the abnormality determination unit 43 determines that an abnormality has occurred in the target material detector 20, the abnormality determination device 40 (abnormality determination unit 43) may transmit information regarding the occurrence of an abnormality in the target material detector 20 to the control computer 70, or to the control computer 70 via the control device 60. When the control computer 70 receives information regarding the occurrence of an abnormality in the target material detector 20, it may transmit a signal to the control device 60 to stop operations on the manufacturing line 10. Alternatively, when the control computer 70 receives information regarding the occurrence of an abnormality in the target material detector 20, it may transmit a signal to the control device 60 to replace and repair the target material detector 20 in which an abnormality has been determined, in order to execute the pre-set maintenance plan for the manufacturing line 10 ahead of schedule.

[0067] As described above, the abnormality detection system for the target material detector according to the present invention uses a reference measuring instrument 30 that can measure a reference radiation intensity that has a high correlation with the infrared radiation intensity detected by the target material detector 20, thereby enabling accurate detection of the occurrence of an abnormality in the target material detector 20.

[0068] Therefore, the following method for determining abnormalities in the target material detector may be used. Specifically, the method for determining abnormalities in the target material detector is a method for determining abnormalities in the target material detector 20, which detects the position of the leading and trailing ends of the target material S by detecting the infrared radiation emitted by the target material S being transported in the manufacturing line 10, and may include a data acquisition step of acquiring reference data as operational data based on the infrared radiation intensity of the infrared radiation of the target material S detected by the target material detector 20 during operation, and the reference radiation intensity measured by a reference measuring instrument 30 installed together with the target material detector 20 in the same transport section where a cooling device 16 for cooling the target material S is not installed; a deviation degree calculation step of calculating the deviation degree of the operational data from the normal data based on the normal data of the target material detector 20 and the reference measuring instrument 30 acquired in advance; and an abnormality determination step of determining an abnormality in the target material detector 20 based on the deviation degree. Furthermore, in the abnormality determination step, if the deviation degree exceeds a preset threshold, it may be determined that an abnormality has occurred in the target material detector 20.

[0069] <Second Embodiment> Next, a second embodiment of the present invention will be described in detail. In the abnormality determination system for the target material detector according to this embodiment, the deviation degree calculation unit 42 of the abnormality determination device 40 may calculate the degree of deviation of the operating data from the normal data based on the principal components identified by principal component analysis of the normal data. In this case, the abnormality determination unit 43 may determine that an abnormality has occurred in the target material detector 20 if the calculated deviation degree exceeds a preset threshold. On the other hand, if the calculated deviation degree is less than or equal to a preset threshold, the target material detector 20 may be determined to be normal.

[0070] Here, principal component analysis refers to an analytical method that synthesizes variables called principal components, which represent the overall variability of the data across a number of correlated variables, by reducing the dimensionality of the data set. It then calculates principal component vectors based on the principal components and includes the operation of projecting the data across the number of variables into the space constructed by the principal component vectors.

[0071] In this embodiment, the degree of deviation of operational data from normal data is calculated by identifying the correlation between variables under normal conditions through principal component analysis of the normal data, and then using the Q statistic to calculate the degree of deviation from normal data.

[0072] Specifically, N (N≧5) normal data points may be acquired, and the degree of deviation in the principal component space for the operational data may be calculated using the k principal components of the normal data obtained by principal component analysis. Then, in the deviation calculation unit 42, the Q statistic shown in equation (2) below may be calculated as the deviation δ using the principal component vector [a, b] of the normal data, the output x of the target material detector 20 of the operational data, and the output y of the reference measuring instrument 30. TIFF0007861740000001.tif10161

[0073] In this case, the k principal components may be determined as the main principal components whose cumulative contribution rate is greater than or equal to a predetermined value (e.g., 0.8). This allows for the calculation of the k principal components and outlier components, which are principal components lower than these k, for the operational data.

[0074] Based on the principal component analysis described above, the Q statistic can be obtained from the calculated residuals. The Q statistic is then defined as the degree of deviation, and if the degree of deviation exceeds a predetermined threshold, it can be determined that an abnormality has occurred in the target material detector 20. The threshold for the degree of deviation may be set, for example, by calculating the Q statistic of normal data and setting it to a value of approximately 2 to 10 times the Q statistic of normal data.

[0075] <Third Embodiment> Next, a third embodiment of the present invention will be described in detail. The abnormality determination system for the target material detector according to this embodiment may calculate the degree of deviation of the operating data from the normal data based on the k-Nearest Neighbor Algorithm.

[0076] The k-nearest neighbors method is a classification technique based on the closest training examples in the feature space. Specifically, it involves acquiring multiple normal-state data points and plotting them in a model space (vector space). The distance (Euclidean distance) in the model space between the operational data points acquired during operation and the normal-state data points is then calculated. Finally, the k (k≧1) data points that are closest in distance are acquired, their distances (Euclidean distances) to the operational data points are calculated, and their average distance is calculated.

[0077] Subsequently, the abnormality determination unit 43 may determine that an abnormality has occurred in the target material detector 20 if the calculated average distance is used as the deviation degree and the deviation degree exceeds a preset threshold. On the other hand, if the calculated deviation degree is less than or equal to the preset threshold, the target material detector 20 may be determined to be normal.

[0078] In this case, the abnormality determination unit 43 may be configured to determine whether inspection or repair of the target material detector 20 is necessary by thresholding the number of times an abnormality is determined in the target material detector 20 over a predetermined period using a predetermined threshold. This is because if the output of the target material detector 20 gradually decreases due to dirt or other reasons, the degree of deviation will gradually increase as operation progresses, and the number of times it exceeds the threshold will increase. Therefore, by individually setting the threshold for the number of times an abnormality is determined in the target material detector 20, it is possible to determine not only when the target material detector 20 actually fails, but also the signs before failure. [Examples]

[0079] Next, we will explain the results of performing abnormality detection on a target material detector 20 installed on a manufacturing line 10 using the abnormality detection system and abnormality detection method for target material detectors according to the present invention.

[0080] The target material detector 20, which is the subject of abnormality detection, was a target material detector 20e located downstream of the roughing mill 14 and upstream of the finishing mill 15 in the manufacturing line 10 shown in Figure 1. The reference measuring instrument 30 was a reference measuring instrument 30c located 3m downstream from the irreversible rolling mill 14b. Therefore, both the target material detector 20 and the reference measuring instrument 30 are located in the same transport section of the manufacturing line 10 where the cooling device 16 for cooling the target material S is not installed. The detection position M of the target material S in the target material detector 20e and the measurement position N of the target material S in the reference measuring instrument 30c were 5m apart in the transport direction A.

[0081] In this embodiment, the normal data processing unit 44 acquired 400 normal data points in advance. Specifically, the infrared radiation intensity (normal detection data) and reference radiation intensity (normal measurement data) data acquired by the target material detector 20e and the reference measuring instrument 30c were used to calculate the average value of the data at each position in the longitudinal direction of the target material S, and stored in the normal data storage unit 45. That is, one data set was stored as normal data for each target material S. Then, the normal data identification unit 46 determined one principal component of the normal data acquired by principal component analysis. In this case, the unit vector of the principal component was represented as [a, b].

[0082] Here, Figure 6 shows the state of the principal component vectors in the normal data. Specifically, Figure 6 is a diagram showing the principal component vectors in the correlation between the infrared radiation intensity acquired by the target material detector 20e and the reference radiation intensity acquired by the reference measuring instrument 30c. As shown in Figure 6, although the target material S transported on the manufacturing line 10 undergoes temperature changes depending on the individual target material S, a high correlation was confirmed between the outputs of the two because the target material detector 20e and the reference measuring instrument 30c are located in the same transport section.

[0083] Next, during the operation of the manufacturing line 10, the abnormality detection device 40 acquired the outputs of the target material detector 20e and the reference measuring instrument 30c, and performed an abnormality determination of the target material detector 20e. Here, the outputs of the target material detector 20e and the reference measuring instrument 30c were acquired as time-series data in real time by the data acquisition unit 41. During operation, the outputs of the target material detector 20e and the reference measuring instrument 30c were continuously acquired, and the abnormality determination of the target material detector 20e was continuously performed. The deviation degree calculation unit 42 calculated the Q statistic, and the calculated Q statistic was used as the deviation degree.

[0084] Figure 7 shows the results of calculating the degree of deviation (Q statistic) for multiple target materials S (steel materials S1 to S3) during operation. The threshold used for abnormality detection by the abnormality detection unit 43 was set to 70 in advance. When the degree of deviation exceeds the threshold, the abnormality detection unit 43 determines that there is an abnormality in the target material detector 20e and sends a signal to the control computer 70 to output an alarm. When the control computer 70 receives the signal to output an alarm, it outputs an alarm via the control device 60.

[0085] As shown in Figure 7, in the manufacturing line 10, when the target material detector 20e detected the transport of the target material S (steel material S3), the deviation calculated by the deviation calculation unit 42 exceeded the threshold, and an alarm was output. In other words, the abnormality detection device 40 was able to accurately determine the occurrence of an abnormality in the target material detector 20e. [Explanation of symbols]

[0086] 1 Manufacturing equipment 10 Production lines 11 Heating furnace 12. Oxide removal device 13 Width reduction device 14 Roughing mill 15. Finishing Rolling Mill 16 Cooling device 17 Winding device 18 Conveying mechanism 20 Target material detector 21 Infrared radiation detection unit 22 Detected Value Amplification Section 23 Detected Value Calculation Unit 30 Reference measuring instrument 31 Measuring part 32 Measurement Amplification Section 33 Measurement Calculation Unit 40 Abnormality determination device 41 Data Acquisition Unit 42. Deviation Calculation Unit 43 Abnormality determination section 44 Normal Data Processing Unit 45 Normal Data Storage Unit 46 Normal Data Identification Unit 50 Anomaly Detection System 60 Control device 70 Control computer 80 high-level calculator A Conveying direction S Target material M detection position N measurement position

Claims

1. An abnormality detection system for a target material detector that detects the position of the leading and trailing ends of a target material by detecting infrared radiation emitted by the target material being transported on a manufacturing line, A reference measuring instrument is provided in the same transport section where no cooling device for cooling the target material is located, together with the target material detector, and measures the infrared radiation of the target material as the reference radiation intensity. The system includes an abnormality determination device for determining abnormalities in the target material detector, The abnormality detection device is, During operation, a data acquisition unit acquires reference data as operational data based on the infrared radiation intensity of the target material detected by the target material detector and the reference radiation intensity measured by the reference measuring instrument. A deviation calculation unit calculates the degree of deviation of the operating data from the normal data based on the normal data of the target material detector and the reference measuring instrument acquired in advance, An abnormality determination unit that determines an abnormality of the target material detector based on the degree of deviation, An abnormality detection system for a target material detector, comprising the above.

2. The abnormality determination system for a target material detector according to claim 1, wherein the reference measuring instrument is a radiation thermometer.

3. A method for determining abnormalities in a target material detector, which detects the position of the leading and trailing ends of a target material by detecting infrared radiation emitted by the target material being transported on a manufacturing line, During operation, a data acquisition step is performed to acquire reference data as operational data based on the infrared radiation intensity of the target material detected by the target material detector, and the reference radiation intensity measured by a reference measuring instrument installed together with the target material detector in the same transport section where no cooling device for cooling the target material is installed. A deviation calculation step that calculates the degree of deviation of the operating data from the normal data based on the normal data of the target material detector and the reference measuring instrument acquired in advance, An abnormality determination step in which an abnormality is determined based on the degree of deviation of the target material detector, A method for determining abnormalities in a target material detector, comprising the above.

4. The method for determining an abnormality of a target material detector according to claim 3, wherein in the deviation degree calculation step, the deviation degree of the operating data from the normal data is calculated based on the principal components identified by the principal component analysis of the normal data.

5. The method for determining an abnormality in a target material detector according to claim 3 or 4, wherein in the abnormality determination step, it is determined that an abnormality has occurred in the target material detector if the degree of deviation exceeds a preset threshold.

6. The method for determining an abnormality in a target material detector according to claim 3 or 4, wherein in the abnormality determination step, it is determined that an abnormality has occurred in the target material detector if the number of times the deviation exceeds a preset threshold exceeds a predetermined number of times.

Images