JP2020015938A - Tuyere monitoring device, tuyere monitoring program, and tuyere monitoring method - Google Patents

Abstract

To enable monitoring of operation state which is needed in point of view of understanding recognition of images by human body.SOLUTION: A tuyere monitoring device 1 has a learning device 15 leaning relationship between images from a tuyere imaged by an imaging device 106 and operation states of a blast furnace, an image data acquisition part 21 for acquiring image data exhibiting images imaged by the imaging device 106 per a prescribed imaging interval, an operation state acquisition part 22 for acquiring the operation states of the blast furnace out put from the learning device responding to input of the image data to the learning device, an operation state estimation part 23 for estimating presence or absence of abnormality of the operation states of the blast furnace on the basis of operation states acquired by the operation state acquisition part, and a warning signal output part 24 for outputting a warning signal when abnormality is estimated to exist in the operation states of the blast furnace, in which the operation state estimation part 23 determines that most frequent operation state in a plurality of the operations states acquired over first unit time longer than imaging time as an operation state of the blast furnace in the first unit time.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a tuyere monitoring device, a tuyere monitoring program, and a tuyere monitoring method.

  Conventionally, the stability of the gas flow and the condition of the blast furnace have been monitored based on the amount of change in the shaft pressure and the absolute value of the stave temperature. In addition, the reaction state and the fluctuation state in the raceway and the lower part of the cohesive zone are monitored by the operator visually observing the brightness inside the blast furnace, the turning state of coke, the drop of unreduced ore, etc. through the tuyere. It had been. However, with the visual monitoring of the operator, there is a possibility that the consistency, objectivity, and data preservation of monitoring of the reaction status such as the raceway may not be sufficiently ensured. For example, in the visual monitoring of the operator, since the criterion for determining whether or not an abnormal state has occurred depends on the subjectivity of the operator and the like, the criterion differs for each operator performing the monitoring work, and there is a possibility that consistency and objectivity may not be maintained. .

  There are known various techniques for observing a state in a tuyere using an image taken by an imaging device provided in the tuyere. For example, using a normalized image geometrically transformed so that the contour shape of the tuyere becomes a normalized circle in the thermal radiance image captured from the tuyere, the reaction state and the fluctuation state in the raceway and the lower part of the cohesive zone are used. Is known (for example, see Patent Documents 1 and 2). In the techniques described in Patent Literatures 1 and 2, in a binarized image generated by binarizing a normalized image, a state near a furnace tuyere can be easily determined based on a light-dark distribution in a radial direction of a normalized circle. Can be observed.

  Further, there is known a technique for estimating a temperature distribution of a combustion field in a raceway using thermal images obtained by imaging a combustion field in a raceway from tuyeres at two different wavelengths (for example, see Patent Documents 3 and 4). . By using the techniques described in Patent Documents 3 and 4, monitoring the reaction status of the raceway further improves consistency, objectivity, and data preservation.

JP-A-2006-60931 Japanese Patent No. 5810981 JP 2001-318002 A JP-A-2002-309307 JP-A-7-305104 JP-A-7-305105 JP 2007-284725 A

Alex Krizhevsky, Ilya Sutskever, and Geoffrey E Hinton, "ImageNetClassification with Deep Convolutional Neural Networks", Advances InNeural Information Processing Systems, Vol. 25, pp. 1106-1114, 2012.

  Items that the operator visually monitors from the tuyere are the coke swirling condition, the raw ore falling state, which is a phenomenon in which ore that has not been completely reduced is dropped, and whether the tuyere is damaged and the pulverized coal supply pipe is damaged. The presence or absence of breakage of the blast furnace equipment, such as the presence or absence, is further included. However, since the coke swirling state, the raw ore falling state, and the presence or absence of damage to the blast furnace equipment need to be performed from the viewpoint of human recognition of the image, the techniques described in Patent Documents 1 to 4 are used. It is not easy to monitor.

  In one embodiment, it is an object of the present invention to provide a technique capable of monitoring an operation state that needs to be performed from a viewpoint that a person can recognize an image, without relying on visual monitoring by an operator.

The gist of the present invention that solves such a problem is a tuyere monitoring device, a tuyere monitoring device program, and a tuyere monitoring device method described below.
(1) a plurality of images taken at different times by an imaging device having an imaging unit arranged to face any one of the tuyeres for supplying pulverized coal together with gas into the interior of the blast furnace; A learning device that learns a relationship between the blast furnace operation state including the first abnormal state and the second abnormal state,
An image data acquisition unit that acquires image data indicating an image captured by the imaging device at predetermined imaging intervals,
An operation state acquisition unit that acquires the operation state of the blast furnace output from the learning device in response to inputting the image data to the learning device,
An operation state estimation unit that estimates presence or absence of an abnormality in the operation state of the blast furnace based on the operation state acquired by the operation state acquisition unit,
A warning signal output unit that outputs a warning signal when the operation state of the blast furnace is estimated to be abnormal by the operation state estimation unit,
The operating state estimating unit determines the most frequent operating state among the plurality of operating states acquired during the first unit time longer than the imaging interval as the operating state of the blast furnace in the first unit time, and determines the operation state. A tuyere monitoring device, wherein the operating state of the blast furnace is stored as first state information.
(2) The tuyere monitoring device according to (1), wherein the first unit time is a time from 1 second to 60 seconds.
(3) The operating state estimating unit determines whether the operating state of the blast furnace is abnormal based on the operating state frequency of the blast furnace corresponding to the first state information stored during the second unit time longer than the first unit time. The tuyere monitoring device according to (1) or (2), which estimates presence or absence.
(4) The tuyere monitoring device according to (3), wherein the second unit time is a time of 5 minutes or more and 10 minutes or less.
(5) A second unit time change unit for estimating a collapse period of a raceway formed inside the blast furnace based on operating conditions of the blast furnace, and changing a second unit time according to the estimated collapse period of the raceway. The tuyere monitoring device according to (3) or (4), further comprising:
(6) The operation state estimating unit is configured to perform the operation based on a ratio of the number of the first state information indicating each of the first abnormal state and the second abnormal state to the number of the first state information stored during the second unit time. The tuyere monitoring device according to any one of (3) to (5), wherein the tuyere monitoring device estimates whether there is an abnormality in the operation state of the blast furnace.
(7) The first abnormal state is a state in which ore that has not been completely reduced drops.
The operating state estimating unit determines whether a ratio of the number of the first state information indicating the first abnormal state to the number of the first state information stored during the second unit time is larger than a predetermined first threshold ratio. Judge
The warning signal output unit drops the ore that has not been completely reduced when the operation state estimation unit determines that the ratio of the number of pieces of the first state information indicating the first abnormal state is larger than the first threshold ratio. (6) The tuyere monitoring device according to (6), which outputs a first caution signal indicating that there is a risk of occurrence of a state in which there is a risk of occurrence.
(8) The operation state estimating unit may be configured such that a ratio of the number of the first state information indicating the first abnormal state to the number of the first state information stored during the second unit time is larger than the first threshold ratio. Determining whether the ratio is greater than a second threshold ratio,
The warning signal output unit drops the ore that has not been completely reduced when the operating state estimating unit determines that the ratio of the number of pieces of the first state information indicating the first abnormal state is larger than the second threshold ratio. (7) The tuyere monitoring device according to (7), which outputs a first alarm signal indicating that the possibility of occurrence of a state of occurrence is high.
(9) The second abnormal state is a state in which the pulverized coal supply pipe for supplying pulverized coal to the tuyere is damaged,
The operating state estimating unit determines whether a ratio of the number of the first state information indicating the second abnormal state to the number of the first state information stored during the second unit time is larger than a predetermined third threshold ratio. Judge whether or not
The warning signal output unit is configured to supply the pulverized coal to the tuyere when the operation state estimating unit determines that the ratio of the number of pieces of the first state information indicating the second abnormal state is greater than the third threshold ratio. The tuyere monitoring device according to any one of (6) to (8), which outputs a second caution signal indicating that a state in which the coal supply pipe may be damaged may occur.
(10) The operating state estimating unit is configured such that a ratio of the number of the first state information indicating the second abnormal state to the number of the first state information stored during the second unit time is larger than the third threshold ratio. Determine whether it is greater than a fourth threshold ratio,
The warning signal output unit is configured to supply pulverized coal to the tuyere when the operation state estimating unit determines that the ratio of the number of pieces of the first state information indicating the second abnormal state is greater than the fourth threshold ratio. (9) The tuyere monitoring device according to (9), which outputs a second alarm signal indicating that a possibility that the coal supply pipe is damaged is high.
(11) The operation state of the blast furnace further includes a cautionary state indicating that an abnormality may occur in the operation state of the blast furnace,
The operating state estimating unit determines whether a ratio of the number of the first state information indicating the cautionary state to the number of the first state information stored during the second unit time is larger than a predetermined fifth threshold frequency. Judge,
The warning signal output unit may cause an abnormality in the operation state of the blast furnace when the operation state estimation unit determines that the ratio of the number of pieces of the first state information indicating the cautionary state is larger than the fifth threshold ratio. The tuyere monitoring device according to any one of (6) to (10), wherein the tuyere monitoring device outputs a third attention signal indicating that there is a warning.
(12) The operation state estimating unit determines the number of the first state information based on the moving average value of the frequency of the first state information stored during the plurality of second unit times, (6) to (11). The tuyere monitoring device according to any one of the above items.
(13) The operating state estimating unit stores the operating state of the blast furnace corresponding to the most frequent first state information among the first state information stored during the second unit time as second state information, When the second state information indicating any of the first abnormal state and the second abnormal state continues for a predetermined threshold number of times or more, it is estimated that there is an abnormality in the operation state of the blast furnace. (3) to (5) The tuyere monitoring device according to any one of the above.
(14) The operating state estimating unit estimates the presence or absence of an abnormality in the operating state of the blast furnace at every third unit time longer than the first unit time and defined independently of the second unit time, (3) The tuyere monitoring device according to any one of (1) to (13).
(15) The operation state estimating unit acquires lance purge time information indicating a lance purge time during which pulverized coal is not supplied from the tuyere to the inside of the blast furnace, and excludes image data indicating an image captured during the lance purge time to remove blast furnace. The tuyere monitoring device according to any one of (1) to (13), which estimates the presence or absence of an abnormality in the operating state of the tuyere.
(16) a plurality of images taken at different times by an imaging device having an imaging unit arranged to face any one of the tuyeres for supplying pulverized coal together with gas into the interior of the blast furnace; In a tuyere monitoring device having a learning device that has learned a relationship with an operation state of a blast furnace including a first abnormal state and a second abnormal state,
Acquiring image data indicating an image captured by the imaging device at predetermined imaging intervals,
In response to inputting the image data to the learning device, the operation state of the blast furnace output from the learning device is obtained,
Based on the obtained operating state, the presence or absence of an abnormality in the operating state of the blast furnace is estimated,
When the operation state of the blast furnace is estimated to be abnormal, a warning signal is output, and the computer is caused to execute processing,
The process of estimating the presence or absence of an abnormality in the operation state of the blast furnace includes, in the first unit time, the most frequent operation state among the plurality of operation states acquired during the first unit time longer than the imaging interval. A tuyere monitoring program comprising: determining a blast furnace operating state; and storing the determined blast furnace operating state as first state information.
(17) a plurality of images taken at different times by an imaging device having an imaging unit arranged to face any one of the tuyeres for supplying pulverized coal together with gas into the interior of the blast furnace; In a tuyere monitoring device having a learning device that has learned a relationship with an operation state of a blast furnace including a first abnormal state and a second abnormal state,
Acquiring image data indicating an image captured by the imaging device at predetermined imaging intervals,
In response to inputting the image data to the learning device, the operation state of the blast furnace output from the learning device is obtained,
Based on the obtained operating state, the presence or absence of an abnormality in the operating state of the blast furnace is estimated,
Outputting a warning signal when it is estimated that there is an abnormality in the operation state of the blast furnace, including a process,
The process of estimating the presence or absence of an abnormality in the operation state of the blast furnace includes, in the first unit time, the most frequent operation state among the plurality of operation states acquired during the first unit time longer than the imaging interval. A tuyere monitoring method comprising: determining a blast furnace operating state; and storing the determined blast furnace operating state as first state information.

  In one embodiment, it is possible to monitor an operation state that needs to be performed from a viewpoint that a human recognizes an image without requiring visual monitoring by an operator.

It is a figure which shows typically the example of a structure of the blast furnace monitoring system containing the tuyere monitoring apparatus which concerns on embodiment. It is a figure showing the tuyere monitoring device concerning a 1st embodiment. 2 is an example of an image captured by the imaging device illustrated in FIG. 1. FIG. 2 is a diagram illustrating an example of an image associated with each of the operation states of the blast furnace illustrated in FIG. 1, (a) is an example of an image associated with a lance purge state, and (b) is an example of an image associated with a normal state. (C) is an example of an image associated with the caution required state, (d) is an example of an image associated with the raw ore fall state, and (e) is an example of an image associated with the lance breakage state. is there. It is a flowchart of the tuyere monitoring process performed by the tuyere monitoring apparatus shown in FIG. It is a figure showing the tuyere monitoring device concerning a 2nd embodiment. 7 is a flowchart of tuyere monitoring processing executed by the tuyere monitoring device shown in FIG. It is a figure showing the tuyere monitoring device concerning a 3rd embodiment. 9 is a flowchart of tuyere monitoring processing executed by the tuyere monitoring device shown in FIG. It is a figure showing the tuyere monitoring device concerning a 4th embodiment. It is a flowchart of the tuyere monitoring process performed by the tuyere monitoring apparatus shown in FIG. It is a figure which shows the time-dependent change of the operating state of a blast furnace when a raw ore fall state occurs. It is a figure which shows a time-dependent change of the operation state of a blast furnace when a lance breakage state occurs.

  Hereinafter, a tuyere monitoring device, a tuyere monitoring program, and a tuyere monitoring method will be described with reference to the drawings. However, the technical scope of the present invention is not limited to those embodiments.

(Overview of the tuyere monitoring device according to the embodiment)
The tuyere monitoring device according to the embodiment uses a learning device that has learned a relationship between a plurality of images captured at different times by an imaging device arranged to face the tuyere and the operating state of the blast furnace. Then, it is estimated whether there is any abnormality in the operation state of the blast furnace. The tuyere monitoring device according to the embodiment changes the most frequent operation state among the plurality of operation states acquired during a first unit time longer than the interval at which the imaging device captures an image, by using the blast furnace in the first unit time. The operation state is determined, and the presence or absence of an abnormality in the operation state is estimated using the determined operation state. The tuyere monitoring device according to the embodiment determines the mode value during the first unit time to be the operating state of the blast furnace at the first unit time, so that the internal state characteristics of the blast furnace can be determined by the same processing as the visual recognition of the operator. Can decide.

(Blast furnace monitoring system according to the embodiment)
FIG. 1 is a diagram schematically illustrating a configuration example of a blast furnace monitoring system including a tuyere monitoring device according to an embodiment.

  In the blast furnace 100, about 40 tuyeres 102 are arranged around a furnace body 101. A hot air supply pipe 103, a pulverized coal supply pipe 104, an observation window 105, and an imaging device 106 are arranged in each of the tuyeres 102, and supply pulverized coal together with a high-pressure gas into the blast furnace 100.

  One end of the hot air supply pipe 103 is a cylindrical member inserted into the furnace body 101 at the tuyere 102, and a high-temperature gas is blown into the inside of the blast furnace 100 at a high pressure from the other end. The raceway 107 is formed inside the blast furnace 100 by the wind pressure of the high-temperature gas blown into the inside of the blast furnace 100 via the hot-air supply pipe 103.

  One end of the pulverized coal supply pipe 104 is a cylindrical member inserted into the hot air supply pipe 103, and pulverized coal is blown into the blast furnace 100 from the other end. In one example, a pair of pulverized coal supply pipes 104 is arranged in each of the tuyeres 102. At the interface of the raceway 107, a coke charged into the blast furnace 100 and pulverized coal blown from the pulverized coal supply pipe 104 burn to generate a high-temperature combustion reaction in which carbon monoxide is generated. In order to prevent the inside of the pulverized coal supply pipe 104 from being clogged with pulverized coal, the blowing of the pulverized coal is stopped at predetermined time intervals, and a high-pressure gas such as nitrogen is blown instead of the pulverized coal. The operation state in which high-pressure gas is blown in place of pulverized coal is also called a lance purge state. Also, when the pulverized coal supply pipe 104 is clogged, high-pressure gas such as nitrogen is blown into the pulverized coal supply pipe 104 instead of pulverized coal, and the operation state of the blast furnace 100 is in a lance purge state.

  The observation window 105 is a light transmissive member fitted into a hole formed at the other end of the hot air supply pipe 103, and is arranged at a position where the raceway 107 can be visually recognized.

  The imaging device 106 is fixedly disposed near the tuyere 102. The imaging device 106 has an imaging unit 108 arranged opposite to the tuyere 102, and a plurality of images taken through the hot air supply pipe 103 and the pulverized coal supply pipe 104 arranged in the tuyere 102 are transmitted to the LAN 109. To the tuyere monitoring device 1 via the. The imaging device 106 can perform a high-speed imaging operation at predetermined imaging intervals, for example, every 10 ms, and sequentially transmits image data indicating captured images to the tuyere monitoring device 1.

(Tuyere monitoring device according to the first embodiment)
FIG. 2 is a diagram illustrating the tuyere monitoring device 1.

  The tuyere monitoring device 1 includes a communication unit 11, a storage unit 12, an input unit 13, an output unit 14, a processing unit 20, and a learning device 15. The communication unit 11, the storage unit 12, the input unit 13, the output unit 14, the processing unit 20, and the learning unit 15 are connected to each other via a bus 200. The tuyere monitoring device 1 estimates the presence or absence of an abnormality in the operating state of the blast furnace 100 using the image data transmitted from the imaging devices 106 arranged in each of the forty tuyeres 102. In one example, the tuyere monitoring device 1 may be integrated with a blast furnace monitoring device, which is a higher-level control device (not shown) that monitors the operation state of the blast furnace 100. In another example, the tuyere monitoring device 1 is a personal computer. There may be.

  The communication unit 11 has a wired communication interface circuit such as Ethernet (registered trademark). The communication unit 11 communicates with the imaging device 106 and a blast furnace monitoring device (not shown) via the LAN 109.

  The storage unit 12 includes, for example, at least one of a semiconductor storage device, a magnetic tape device, a magnetic disk device, and an optical disk device. The storage unit 12 stores an operating system program, a driver program, an application program, data, and the like used for processing in the processing unit 20. For example, the storage unit 12 uses the image data transmitted from the imaging device 106 to cause the processing unit 20 to execute a tuyere monitoring program for estimating the presence or absence of an abnormality in the operation state of the blast furnace 100. And so on. The tuyere monitoring program may be installed in the storage unit 12 using a known setup program or the like from a computer-readable portable recording medium such as a CD-ROM or a DVD-ROM. The storage unit 12 stores various data used in the tuyere monitoring process. For example, the storage unit 12 stores a plurality of image data sequentially transmitted from the imaging device 106 in association with an imaging time.

  The input unit 13 may be any device as long as data can be input, and is, for example, a touch panel, a keyboard, or the like. The operator can use the input unit 13 to input characters, numbers, symbols, and the like. When operated by the operator, the input unit 13 generates a signal corresponding to the operation. Then, the generated signal is supplied to the processing unit 20 as an instruction of the operator.

  The output unit 14 may be any device that can display a video, an image, and the like, and is, for example, a liquid crystal display or an organic EL (Electro-Luminescence) display. The output unit 14 displays a video corresponding to the video data supplied from the processing unit 20, an image corresponding to the image data, and the like. The output unit 14 may be an output device that prints images, images, characters, and the like on a display medium such as paper.

The learning device 15 is a learning device that learns features of an image that has been abstracted by supervised learning or unsupervised learning. For example, a plurality of images captured by the imaging device 106 using a known machine learning technique such as deep learning. Learn the relationship between the image of the blast furnace and the operating state of the blast furnace. Deep learning is machine learning using a multilayer neural network composed of an input layer, a middle layer, and an output layer. Each feature vector of a plurality of images captured by the imaging device 106 is input to each node of the input layer. The feature vector may be, for example, a partial image obtained by cutting out a partial area of the image. Further, each of the partial images used as the feature vector may be cut out such that a part thereof is superimposed on another partial image. Each node in the hidden layer outputs the sum of values obtained by multiplying each feature vector output from each node in the input layer by a weight, and further, the output layer outputs each feature vector output from each node in the hidden layer. The sum of the values multiplied by the weight is output. The learning device 15 learns such that the difference between the output value from the output layer and each of the plurality of images captured by the imaging device 106 is reduced while adjusting each weight. The learning device 15 is realized by, for example, the technology described in Non-Patent Document 1. The learning by the learning unit 15 of the relationship between the plurality of images captured by the imaging device 106 and the operating state of the blast furnace is performed by, for example, a method disclosed in Japanese Patent Application No. No. (Your reference number: Enter the application number of FP171419). That is, the plurality of images captured by the imaging device 106 are clustered into an image group and associated with the operating state of the blast furnace, and the operating state of the blast furnace associated with the image group is determined to be a normal state, a first abnormal state, and a second abnormal state. When it is determined that the blast furnace includes the image and the operating state of the blast furnace, the learning device learns the relationship between the image and the operating state of the blast furnace, and the learning device 15 that has learned the relationship between the image and the operating state of the blast furnace is output. The operating state of the blast furnace including the normal state, the first abnormal state, and the second abnormal state will be described later.

  FIG. 3 is an example of an image captured by the imaging device 106. In FIG. 3, a pair of pulverized coal supply pipes 104 are imaged on both left and right sides, and pulverized coal supplied to the raceway 107 via the pulverized coal supply pipe 104 is imaged in the center. In the lower right part, an image of the coke circling in the raceway 107 is also taken.

  FIG. 4 is a diagram illustrating an example of an image associated with the operation state of the blast furnace 100 by the learning device 15. The operation states of the blast furnace 100 associated with a plurality of images are five states of a lance purge state, a normal state, a cautionary state, a first abnormal state, and a second abnormal state.

  FIG. 4A is an example of an image associated with the lance purge state. In the lance purge state, high-pressure gas such as nitrogen is blown instead of pulverized coal. Therefore, the pulverized coal supplied to the raceway 107 via the pulverized coal supply pipe 104 is captured in the image shown in FIG. Not.

  FIG. 4B is an example of an image in which the state inside the blast furnace 100 visually recognized from the tuyere 102 is associated with a favorable normal state. In the image shown in FIG. 4B, the portion other than the path of the pulverized coal blown from the pair of pulverized coal supply pipes 104 is a region where the luminance is high, and a favorable high-temperature combustion reaction occurs in the raceway 107. Is shown.

  FIG. 4C is an example of an image in which the state inside the blast furnace 100 visually recognized from the tuyere 102 is associated with a state requiring attention. The caution required state is, for example, a state immediately before the time at which the raw ore fall state and breakage of the blast furnace equipment occur.

  FIG. 4D is an example of an image associated with a raw ore falling state in which ore that has not been completely reduced is dropped. In this specification, the raw ore fall state is also referred to as a first abnormal state. In the image shown in FIG. 4D, a portion extending downward from the center of the image is a region having a low luminance, and indicates that ore that has not been completely reduced is dropped. That is, when the unreduced ore comes into contact with coke, the ore undergoes a direct reduction reaction, which is an endothermic reaction that absorbs a large amount of heat, so that the temperature around the ore decreases, and the ore is shown as a dark area on the screen.

  FIG. 4E is an example of an image associated with a lance breakage state in which the pulverized coal supply pipe 104 has been broken. In this specification, the lance breakage state is also referred to as a second abnormal state. The lance breakage state is a state in which the pulverized coal supply pipe 104 is bent or broken for some reason. In the image shown in FIG. 4E, one of the paths of the pulverized coal blown from the pair of pulverized coal supply pipes 104 is curved, indicating that one of the pair of pulverized coal supply pipes 104 is damaged.

  The processing unit 20 has one or a plurality of processors and their peripheral circuits. The processing unit 20 controls the overall operation of the tuyere monitoring device 1 and is, for example, a CPU. The processing unit 20 performs processing based on programs (a driver program, an operating system program, an application program, and the like) stored in the storage unit 12. Further, the processing unit 20 can execute a plurality of programs (such as application programs) in parallel.

  The processing unit 20 includes an image data acquisition unit 21, an operation state acquisition unit 22, an operation state estimation unit 23, and a warning signal output unit 24. Each of these units is a functional module realized by a program executed by a processor included in the processing unit 20. Alternatively, these units may be implemented in the tuyere monitoring device 1 as firmware.

  The image data acquisition unit 21 acquires image data indicating an image captured by the imaging device 106 at predetermined imaging intervals. The operation state acquisition unit 22 inputs the image data acquired by the image data acquisition unit 21 to the learning device 15 and acquires the operation state of the blast furnace 100 output from the learning device 15 in response to the input of the image data. I do. The operating state estimating unit 23 estimates whether the operating state of the blast furnace 100 is abnormal based on the operating state acquired by the operating state acquiring unit 22. The warning signal output unit 24 outputs a warning signal when the operation state estimation unit 23 estimates that the operation state of the blast furnace 100 is abnormal. The warning signal includes a caution signal indicating that an abnormality may occur in the operation state of the blast furnace 100 and an alarm signal indicating that there is a high possibility that an abnormality occurs in the operation state of the blast furnace 100.

(Tuyere monitoring process by tuyere monitoring device 1)
FIG. 5 is a flowchart of the tuyere monitoring process executed by the tuyere monitoring device 1. The tuyere monitoring process shown in FIG. 5 is executed mainly by the processing unit 20 in cooperation with each element of the tuyere monitoring device 1 based on a program stored in the storage unit 12 in advance. The tuyere monitoring process uses the image data transmitted from the imaging device 106 disposed in each of the tuyeres 102 because the states of the raceways 107 visually recognized from the forty tuyeres 102 are different from each other. Then, it is executed for each of the tuyere 102.

  First, the image data acquisition unit 21 acquires image data indicating an image captured by the imaging device 106 (S101), and stores the acquired image data in the storage unit 12. The image corresponding to the image data acquired by the image data acquisition unit 21 is an image captured at a predetermined imaging interval such as every 10 ms.

  Next, the operation state acquisition unit 22 inputs the image data acquired in the process of S101 to the learning device 15, and operates the blast furnace 100 output from the learning device 15 in response to inputting the image data to the learning device 15. The status is acquired (S102). The operating state acquisition unit 22 stores the operating state of the blast furnace 100 output from the learning unit 15 in the storage unit 12 in association with the image data input to the learning unit 15.

  Next, the image data acquisition unit 21 determines whether or not a predetermined first unit time longer than the imaging interval has elapsed since the first acquisition of the image data in the process of S101 (S103). The image data acquisition unit 21 has a timer, measures the time since the first acquisition of the image data in the process of S101, resets the timer when the first unit time has elapsed, and restarts the time measurement. If the image data acquisition unit 21 determines that the first unit time has not elapsed since the first acquisition of the image data in the process of S101 (S103-NO), the image data indicating the image captured by the imaging device 106 Is acquired (S101). The first unit time is a time from 1 second to 60 seconds. When the imaging interval at which the imaging device 106 captures an image is 10 ms, and the first unit time is 1 second, 100 image data and the image data are associated with the image data in the processing of S102 during the first unit time. The operation state is stored in the storage unit 12. Also, when the imaging interval at which the imaging device 106 captures an image is 10 ms and the first unit time is 60 seconds, 6000 pieces of image data are associated with the image data in the processing of S102 during the first unit time. The operation state thus set is stored in the storage unit 12.

  When it is determined that the first unit time has elapsed since the first acquisition of the image data in the processing of S101 (S103-YES), the operating state estimating unit 23 determines the operating state of the blast furnace 100 in the first unit time. (S104). Specifically, the operating state estimating unit 23 determines the most frequent operating state among the plurality of operating states acquired during the first unit time and stored in the storage unit 12 in the blast furnace in the first unit time. 100 operating states are determined. For example, when the normal state is the most frequent operation state among the plurality of operation states acquired during the first unit time, the operation state estimation unit 23 changes the normal state of the blast furnace 100 in the first unit time. Determine the operating state.

  The operating state estimating unit 23 stores the operating state determined as the operating state of the blast furnace 100 in the first unit time in the storage unit 12 as first state information, and image data and image data acquired over the first unit time. Is deleted from the storage unit 12.

  Next, the image data acquisition unit 21 determines whether or not a predetermined second unit time longer than the first unit time has elapsed since the first acquisition of the image data in the process of S101 (S105). The image data acquisition unit 21 has a timer different from the timer that measures the first unit time, measures the time since the first acquisition of the image data in the process of S101, and sets the timer when the second unit time has elapsed. Reset and restart timing. If the image data acquisition unit 21 determines that the second unit time has not elapsed since the first acquisition of the image data in the processing of S101 (S105-NO), the image data indicating the image captured by the imaging device 106 Is acquired (S101). The second unit time is a time of 5 minutes or more and 10 minutes or less.

  If it is determined that the second unit time has elapsed since the first acquisition of the image data in the processing of S101 (S105-YES), the operating state estimating unit 23 estimates whether or not the operating state of the blast furnace 100 is abnormal. . Specifically, the operation state estimating unit 23 determines whether there is an abnormality in the operation state of the blast furnace 100 based on the frequency of the operation state of the blast furnace 100 corresponding to the first state information stored during the second unit time. presume.

  The operating state estimating unit 23 calculates the ratio of the number of pieces of the first state information indicating each of the operating states of the blast furnace 100 to the number of all pieces of the first state information stored during the second unit time (S106). . The operating state estimating unit 23 extracts the number of pieces of first state information indicating the lance purge state, the normal state, the cautionary state, the first abnormal state, and the second abnormal state during the second unit time. Next, the operating state estimating unit 23 divides the number of the extracted first state information indicating the respective operating states by the number of all the first state information stored during the second unit time to obtain the blast furnace 100. Calculate the ratio of the number of pieces of first state information indicating each of the operation states.

  Next, the operating state estimating unit 23 determines that the ratio of the number of the first state information indicating the first abnormal state to the number of all the first state information stored during the second unit time is a predetermined first threshold value. It is determined whether the ratio is larger than the ratio (S107). The first abnormal state is a state in which ore that has not been completely reduced drops, and the first threshold ratio is, for example, 30%. When the operating state estimating unit 23 determines that the ratio of the number of pieces of the first state information indicating the first abnormal state is equal to or less than the first threshold value ratio (S107-NO), the process proceeds to S111.

  If the operating state estimating unit 23 determines that the ratio of the number of pieces of the first state information indicating the first abnormal state is larger than the first threshold ratio (S107-YES), the process proceeds to S108. The operation state estimating unit 23 determines that the ratio of the number of the first state information indicating the first abnormal state to the number of all the first state information stored during the second unit time is larger than the first threshold ratio. It is determined whether the ratio is larger than a predetermined second threshold ratio (S108). The second threshold ratio is, for example, 50%.

  When the operating state estimating unit 23 determines that the ratio of the number of pieces of the first state information indicating the first abnormal state is equal to or less than the second threshold ratio (S108-NO), the warning signal output unit 24 outputs the first signal. An attention signal is output (S109). The first caution signal is a signal indicating that a state in which ore that has not been completely reduced may drop may occur. Next, the process proceeds to S111.

  When the operating state estimating unit 23 determines that the ratio of the number of pieces of the first state information indicating the first abnormal state is larger than the second threshold ratio (S108-YES), the warning signal output unit 24 outputs the first warning. A signal is output (S110). The first alarm signal is a signal indicating that there is a high possibility that a state in which ore that has not been completely reduced drops will occur. Next, the process proceeds to S111.

  The operation state estimating unit 23 determines that the ratio of the number of the first state information indicating the second abnormal state to the number of all the first state information stored during the second unit time is higher than a predetermined third threshold ratio. It is determined whether it is larger (S111). The second abnormal state is a state in which the pulverized coal supply pipe 104 that supplies the pulverized coal to the tuyere 102 is damaged, and the second threshold ratio is, for example, 30%. When the operating state estimating unit 23 determines that the ratio of the number of pieces of the first state information indicating the second abnormal state is equal to or less than the third threshold ratio (S111-NO), the process proceeds to S115.

  When the operating state estimating unit 23 determines that the ratio of the number of pieces of the first state information indicating the second abnormal state is larger than the third threshold ratio (S111-YES), the process proceeds to S112. The operation state estimating unit 23 determines that the ratio of the number of the first state information indicating the second abnormal state to the number of all the first state information stored during the second unit time is larger than the third threshold ratio. It is determined whether the ratio is larger than a predetermined fourth threshold ratio (S112). The fourth threshold ratio is, for example, 50%.

  If the operating state estimating unit 23 determines that the ratio of the number of pieces of the first state information indicating the second abnormal state is equal to or less than the fourth threshold value ratio (S112-NO), the warning signal output unit 24 sets the second signal. An attention signal is output (S113). The second caution signal is a signal indicating that the pulverized coal supply pipe 104 that supplies the pulverized coal to the tuyere 102 may be damaged. Next, the process proceeds to S115.

  When the operating state estimating unit 23 determines that the ratio of the number of pieces of the first state information indicating the second abnormal state is larger than the fourth threshold ratio (S112-YES), the warning signal output unit 24 outputs the second warning. A signal is output (S114). The second alarm signal is a signal indicating that there is a high possibility that the pulverized coal supply pipe 104 that supplies the pulverized coal to the tuyere 102 will be damaged. Next, the process proceeds to S115.

  The operating state estimating unit 23 determines that the ratio of the number of the first state information indicating the cautionary state to the number of all the first state information stored during the second unit time is larger than a predetermined fifth threshold ratio. It is determined whether or not (S115). The caution required state is a state indicating that an abnormality may occur in the operation state of the blast furnace 100, and the fifth threshold ratio is, for example, 30%. When the operating state estimating unit 23 determines that the ratio of the number of pieces of the first state information indicating the cautionary state is equal to or smaller than the predetermined fifth threshold ratio (S115-NO), the process proceeds to S117.

  When the operating state estimating unit 23 determines that the ratio of the number of pieces of the first state information indicating the cautionary state is larger than the predetermined fifth threshold ratio (S115-YES), the warning signal output unit 24 outputs the third signal. An attention signal is output (S116). The third caution signal is a signal indicating that an abnormality may occur in the operation state of the blast furnace 100. Next, the process proceeds to S117.

  The image data acquisition unit 21 determines whether or not a monitoring end instruction to end the monitoring of the tuyere 102 has been received from a blast furnace monitoring device (not shown) (S117). If the image data acquisition unit 21 determines that the monitoring end instruction for instructing the end of the monitoring of the tuyere 102 has not been received from the blast furnace monitoring device (not shown) (S117-NO), the process returns to S101. If it is determined by the image data acquisition unit 21 that a monitoring end instruction to end the monitoring of the tuyere 102 has been received from a blast furnace monitoring device (not shown) (S117-YES), the process ends.

(Operation and effect of the tuyere monitoring device according to the first embodiment)
The tuyere monitoring device 1 determines the most frequent operation state among the plurality of operation states acquired during the first unit time longer than the interval at which the imaging device 106 takes an image, and sets the blast furnace 100 at the first unit time. The operation state is determined, and the presence or absence of an abnormality in the operation state is estimated using the determined operation state. Since the tuyere monitoring device 1 determines the most frequent operation state among the plurality of operation states acquired during the first unit time as the operation state of the blast furnace 100 in the first unit time, the tuyere monitoring device 1 sets the most frequent value as the most frequent value. Time-average features can be extracted. When the operator observes the internal state of the blast furnace 100 such as the raceway 107 from the tuyere 102, the most frequent state of the internal state of the blast furnace 100 that changes every moment is visually recognized as the internal state of the blast furnace 100. The tuyere monitoring device 1 determines not the instantaneous value and the average value of the plurality of operating states during the first unit time but the mode value during the first unit time as the operating state of the blast furnace 100 in the first unit time. Therefore, the internal state characteristics of the blast furnace 100 can be determined by the same processing as the visual recognition of the operator. That is, the tuyere monitoring device 1 enables monitoring of an operation state that needs to be performed from a viewpoint that a human recognizes an image, without relying on visual monitoring by an operator.

  Further, the tuyere monitoring device 1 uses image data indicating an image captured at an imaging interval of, for example, the order of μs to convert the first state information indicating the operation state of the blast furnace 100 to a first state, such as 1 second, for example. Determined every unit time. The tuyere monitoring device 1 determines and stores the first state information for each first unit time based on the image data indicating the image captured at the imaging interval, so as to estimate the presence or absence of an abnormality in the operation state. In addition, the amount of data to be processed can be compressed. The tuyere monitoring device 1 can compress the amount of data used for estimating the presence or absence of an abnormality in the operating state, so that the estimation of the presence or absence of an abnormality in the operating state of all of the about 40 tuyeres 102 is instantaneous. And it can be performed continuously.

  The tuyere monitoring device 1 uses the first state information instead of the image data indicating the image captured by the imaging device 106 to estimate whether or not the operating state of the blast furnace 100 is abnormal. Is determined, the image data indicating the captured image is not used. After the first state information is determined, the tuyere monitoring device 1 deletes the image data indicating the captured image without using the image data, thereby reducing the amount of data stored in the storage unit 12 and various processes. The amount of calculation at the time of execution can be greatly reduced.

  In addition, the tuyere monitoring device 1 sets the first unit time to any time of 1 second or more and 60 seconds or less so that the first unit time coincides with the acquisition interval at which other operation data of the blast furnace 100 is acquired. The operation state of the mouth monitoring device 1 and other operation data of the blast furnace 100 can be easily compared. For example, when pressure data indicating various pressures inside the blast furnace 100 is acquired at one-second intervals, by setting the first unit time to 1 second, the time-dependent changes in various pressures inside the blast furnace 100 and the tuyere monitoring This makes it easy to compare the operating state of the apparatus 1 with a change over time. Further, when the stave temperature data indicating the stave temperature of the blast furnace 100 is acquired at intervals of 60 seconds, the first unit time is set to 60 seconds, so that the stave temperature inside the blast furnace 100 changes over time and the tuyere monitoring device 1 It becomes easy to compare with the change over time of the operating state of the apparatus.

  In addition, the tuyere monitoring device 1 determines the operating state of the blast furnace 100 based on the frequency of the operating state of the blast furnace 100 corresponding to the first state information stored during the second unit time longer than the first unit time. Estimate the presence or absence of abnormality. The image captured by the imaging device 106 through the tuyere 102 has a greater variation than the change inside the blast furnace 100, and the estimation based on the first unit time of 1 second or more and 60 seconds or less constantly changes the operating state of the blast furnace 100. It may not be easy to estimate. The tuyere monitoring device 1 estimates whether or not there is an abnormality in the operation state of the blast furnace 100 at intervals of a longer second unit time, and based on whether or not an abnormality has constantly occurred over the second unit time. Since the presence or absence of an abnormality is estimated, it is less likely that an unsteady phenomenon is estimated as an abnormal state.

  In addition, the tuyere monitoring device 1 executes the estimation process for estimating the presence or absence of an abnormality in the operation state of the blast furnace every second unit time of 5 minutes or more and 10 minutes or less. It is possible to prevent the estimating process from being executed as frequently as necessary. In addition, the tuyere monitoring device 1 sets the second unit time so as to include one cycle of the collapse of the raceway 107, which is generally about 5 to 10 minutes, so that the initial luminance of the collapse of the raceway 107 can be improved. Erroneous estimation caused by the difference between the luminance in the second period and the luminance in the latter period can be prevented.

  In addition, the tuyere monitoring device 1 sends a warning signal indicating that an abnormality may occur and an alarm signal indicating that the abnormality is likely to occur to all first state information during the second unit time. Is output in accordance with the ratio of the number of pieces of the first state information indicating the abnormal state to the number of pieces of the first state information. The tuyere monitoring device 1 can prompt the operator to take measures according to the signal output by the tuyere monitoring device 1 by sequentially outputting the caution signal and the alarm signal according to the ratio indicating the abnormal state.

  In addition, the tuyere monitoring device 1 outputs the caution signal according to the ratio during the second unit time of the cautionary state indicating that there is a possibility that an abnormality may occur in the operation state of the blast furnace 100. The operator is expected that some abnormal condition may occur in the system.

(Tuyere monitoring device according to the second embodiment)
FIG. 6 is a diagram illustrating a tuyere monitoring device according to the second embodiment. The tuyere monitoring device 2 according to the second embodiment is arranged in place of the tuyere monitoring device 1 according to the first embodiment in the blast furnace monitoring system shown in FIG. 1 and uses image data transmitted from the imaging device 106. Then, the presence or absence of an abnormality in the operation state of the blast furnace 100 is estimated.

  The tuyere monitoring device 2 differs from the tuyere monitoring device 1 in having a processing unit 30 instead of the processing unit 20. The processing unit 30 is different from the processing unit 20 in having a second unit time changing unit 35. The configuration and function of the components of the tuyere monitoring device 2 other than the second unit time changing unit 35 are the same as the configuration and functions of the components of the tuyere monitoring device 1 denoted by the same reference numerals, and will be described in detail here. Is omitted.

(Tuyere monitoring process by tuyere monitoring device 2)
FIG. 7 is a flowchart of the tuyere monitoring process executed by the tuyere monitoring device 2. The tuyere monitoring process shown in FIG. 7 is executed mainly by the processing unit 30 in cooperation with each element of the tuyere monitoring device 2 based on a program stored in the storage unit 12 in advance. The tuyere monitoring process uses the image data transmitted from the imaging device 106 disposed at each of the tuyeres 102 because the states of the raceways 107 visually recognized from the forty tuyeres 102 are different from each other. Then, it is executed for each of the tuyere 102.

  Since the processing of S201 to S216 is the same as the processing of S101 to S116, a detailed description is omitted here. The second unit time changing unit 35 estimates the collapse period of the raceway 107 formed inside the blast furnace 100 based on the operating conditions of the blast furnace 100 (S217). The second unit time changing unit 35 is formed inside the blast furnace 100 based on operating conditions of the blast furnace 100 such as a tuyere diameter, a tuyere wind speed, a pulverized coal ratio, and the like, by techniques described in Patent Documents 5 to 7, for example. The collapse period of the raceway 107 is estimated (S217). Techniques for estimating the collapse period of the raceway 107 formed inside the blast furnace 100 are well known, and therefore detailed description is omitted here.

  Next, the second unit time changing unit 35 changes the second unit time according to the collapse period of the raceway 107 estimated in the process of S217 (S218). The second unit time changing unit 35 may change the second unit time so as to be equal to the collapse period of the raceway 107 estimated in the process of S217, and may change the second unit time based on the collapse period of the raceway 107 estimated in the process of S217. The second unit time may be changed so that is also longer. For example, the second unit time change unit 35 may change the second unit time so as to match the collapse period of the raceway 107 estimated in the process of S217, and may change the collapse of the raceway 107 estimated in the process of S217. The second unit time may be changed so as to be 1.2 times the cycle. In addition, the second unit time change unit 35 may change the second unit time so that the second unit time becomes 1.5 times the collapse period of the raceway 107 estimated in the process of S217. Since the process of S219 is the same as the process of S117, a detailed description is omitted here.

(Operation and effect of the tuyere monitoring device according to the second embodiment)
Since the tuyere monitoring device 2 changes the second unit time in accordance with the estimated collapse period of the raceway 107, even when the state of the raceway 107 is fluctuating, the tuyere monitoring device 2 has the initial state of the collapse period of the raceway 107. The occurrence of erroneous estimation due to the difference between the luminance and the luminance in the latter period can be prevented with higher accuracy.

(Modification of the tuyere monitoring device according to the first embodiment and the second embodiment)
In the tuyere monitoring devices 1 and 2, the operating state estimating unit 23 estimates whether there is an abnormality in the operating state of the blast furnace 100 when it is determined that the second unit time has elapsed since the first acquisition of the image data. In the tuyere monitoring devices 1 and 2, the second unit time is set without overlapping each other, but the setting method of the second unit time is not limited to this. For example, each time the first unit time elapses, that is, each time the image data is acquired and the operation state is acquired, the time from that time to the time that is traced back by the second unit time is defined as the second unit time, and the second unit time is calculated. The presence or absence of an abnormality in the operating state of the blast furnace 100 may be estimated based on the frequency of the operating state of the blast furnace 100 corresponding to the first state information stored in between.

  In this case, the second unit time is set so as to be shifted and overlapped by the first unit time, and the estimation process for estimating the presence or absence of an abnormality in the operation state of the blast furnace is executed every time the first unit time elapses. Will be done.

(Tuyere monitoring device according to the third embodiment)
FIG. 8 shows a tuyere monitoring device according to the third embodiment that is effective when the second unit time is set to be duplicated. The tuyere monitoring device 3 according to the third embodiment is arranged in place of the tuyere monitoring device 1 according to the first embodiment in the blast furnace monitoring system shown in FIG. 1 and uses image data transmitted from the imaging device 106. Then, the presence or absence of an abnormality in the operation state of the blast furnace 100 is estimated.

  The tuyere monitoring device 3 differs from the tuyere monitoring device 1 in having a processing unit 40 instead of the processing unit 30. The processing unit 40 differs from the processing unit 20 in that the processing unit 40 includes an image data acquisition unit 41 and an operation state estimation unit 43 instead of the image data acquisition unit 21 and the operation state estimation unit 23. The configuration and function of the components of the tuyere monitoring device 3 other than the image data acquisition unit 41 and the operation state estimation unit 43 are the same as those of the components of the tuyere monitoring device 1 denoted by the same reference numerals. A detailed description will be omitted.

(Tuyere monitoring process by tuyere monitoring device 3)
FIG. 9 is a flowchart of the tuyere monitoring process executed by the tuyere monitoring device 3. The tuyere monitoring process shown in FIG. 9 is mainly executed by the processing unit 40 in cooperation with each element of the tuyere monitoring device 3 based on a program stored in the storage unit 12 in advance. The tuyere monitoring process uses the image data transmitted from the imaging device 106 disposed at each of the tuyeres 102 because the states of the raceways 107 visually recognized from the forty tuyeres 102 are different from each other. Then, it is executed for each of the tuyere 102.

  Since the processing of S301 to S304 is the same as the processing of S101 to S104, detailed description is omitted here. Next, the image data obtaining unit 41 first obtains the image data in the process of S301, and a predetermined third unit time longer than the first unit time and defined independently of the second unit time has elapsed. It is determined whether or not it is (S305). The image data acquisition unit 41 has a timer different from the timer that measures the first unit time and the second unit time, measures the time since the first acquisition of the image data in the processing of S301, and When the time elapses, the timer is reset and time measurement is restarted. The third unit time may be longer than the first unit time and the second unit time, may be shorter than the second unit time, and may be the same time as the second unit time. For example, when the first unit time is 1 second and the second unit time is 10 minutes, the third unit time may be 1 minute, 5 minutes, or 10 minutes. Or 15 minutes.

  If it is determined that the third unit time has elapsed since the first acquisition of the image data in the processing of S301 (S305-YES), the operating state estimating unit 43 estimates whether there is an abnormality in the operating state of the blast furnace 100. . Specifically, the operating state estimation unit 43 determines the operating state of the blast furnace 100 based on the frequency of the operating state of the blast furnace 100 corresponding to the first state information stored from the second unit time before to the present. Estimate the presence or absence of abnormality.

  The operating state estimating unit 43 calculates the ratio of the number of the first state information indicating each operating state of the blast furnace 100 to the number of the first state information stored from the second unit time before to the present (S306). ). The operation state estimation unit 43 extracts the number of pieces of first state information indicating each of the lance purge state, the normal state, the cautionary state, the first abnormal state, and the second abnormal state from the second unit time before to the present. . Next, the operation state estimation unit 43 divides the number of the extracted first state information indicating each operation state by the number of the first state information stored from the second unit time before to the present, The ratio of the number of pieces of first state information indicating each of the operation states of the blast furnace 100 is calculated.

  The processing in S307 to S317 is the same as the processing in S107 to S117, and a detailed description is omitted here.

(Operation and effect of the tuyere monitoring device according to the third embodiment)
The tuyere monitoring device 3 sets the third unit time, which is the interval for performing the estimation process for estimating the presence or absence of an abnormality in the operation state of the blast furnace 100, as first state information in order to estimate the presence or absence of the abnormality in the operation state of the blast furnace 100 Can be defined independently of the second unit time, which is the period for acquiring This allows
For example, even when the second unit time is set so as to be shifted while being overlapped by the first unit time, it is possible to prevent the estimation process from being executed at an unnecessarily high frequency, such as every time the first unit time elapses. . In addition, the tuyere monitoring device 3 is configured to set the third unit time to 1 minute when the operation state of the blast furnace 100 is unstable and the second unit time is 10 minutes. Monitoring of 100 operating conditions can be enhanced.

(Tuyere monitoring device according to the fourth embodiment)
FIG. 10 is a diagram showing a tuyere monitoring device according to the fourth embodiment. The tuyere monitoring device 4 according to the fourth embodiment is arranged in place of the tuyere monitoring device 1 according to the first embodiment in the blast furnace monitoring system shown in FIG. 1 and uses image data transmitted from the imaging device 106. Then, the presence or absence of an abnormality in the operation state of the blast furnace 100 is estimated.

  The tuyere monitoring device 4 differs from the tuyere monitoring device 1 in having a processing unit 50 instead of the processing unit 20. The processing unit 50 differs from the processing unit 20 in having the operation state estimation unit 53 instead of the operation state estimation unit 23. The configuration and function of the components of the tuyere monitoring device 4 other than the operation state estimating unit 53 are the same as the configuration and functions of the components of the tuyere monitoring device 1 denoted by the same reference numerals, and thus detailed description is omitted here. I do.

(Tuyere monitoring process by tuyere monitoring device 4)
FIG. 11 is a flowchart of the tuyere monitoring process executed by the tuyere monitoring device 4. The tuyere monitoring process shown in FIG. 11 is mainly executed by the processing unit 50 in cooperation with each element of the tuyere monitoring device 4 based on a program stored in the storage unit 12 in advance. The tuyere monitoring process uses the image data transmitted from the imaging device 106 disposed at each of the tuyeres 102 because the states of the raceways 107 visually recognized from the forty tuyeres 102 are different from each other. Then, it is executed for each of the tuyere 102.

  Since the processing of S401 to S405 is the same as the processing of S101 to S105, detailed description is omitted here. Next, the operating state estimating unit 53 stores the operating state of the blast furnace 100 corresponding to the most frequent first state information among the first state information stored during the second unit time as second state information. . The operating state estimating unit 53 estimates that the operating state of the blast furnace 100 is abnormal when the second state information indicating one of the first abnormal state and the second abnormal state continues for a predetermined threshold number of times or more. .

  The operation state estimating unit 53 determines the most frequent operation state in the first state information stored during the second unit time (S406). The operation state estimation unit 53 extracts the number of pieces of first state information indicating the lance purge state, the normal state, the cautionary state, the first abnormal state, and the second abnormal state during the second unit time. Next, the operating state estimating unit 53 determines the operating state with the largest number of extracted pieces as the most frequent operating state. The operating state estimating unit 53 stores the determined most frequent operating state in the storage unit 12 as second state information in association with the current time.

  Next, the operating state estimating unit 53 determines whether or not the most frequent operating state corresponding to the second state information stored in the processing of S406 is the first abnormal state (S407). If the operation state estimating unit 53 determines that the most frequent operation state is not the first abnormal state (S407-NO), the process proceeds to S412.

  When the operating state estimating unit 53 determines that the most frequent operating state is the first abnormal state (S407-YES), the operating state estimating unit 53 checks the second state information over the second unit time of the past several times, and It is determined whether the second state information indicating the abnormal state has continued more than a predetermined first threshold number of times (S408). The first threshold number of times is defined according to the length of the first unit time and the second unit time. For example, when the first unit time is 30 seconds and the second unit time is 5 minutes, the first threshold number may be three times, and the first unit time is 60 seconds and the second unit time is When the time is 10 minutes, the first threshold number may be three. If the operation state estimating unit 53 determines that the number of times the second state information indicating the first abnormal state is continuous is equal to or less than the first threshold number (S408-NO), the process proceeds to S412.

  When the operating state estimating unit 53 determines that the second state information indicating the first abnormal state has continued more than the first threshold number of times (S408-YES), the process proceeds to S409. The operation state estimating unit 53 checks the second state information over the second unit time for the past several times, and determines that the second state information indicating the first abnormal state is larger than the first threshold number of times. It is determined whether or not the number of consecutive times exceeds the threshold value (S409). The second threshold number of times is defined according to the length of the first unit time and the second unit time. For example, when the first unit time is 30 seconds and the second unit time is 5 minutes, the second threshold number may be set to 5, and the first unit time is 60 seconds and the second unit time is When the time is 10 minutes, the second threshold number may be set to five.

  When the operation state estimating unit 53 determines that the number of times the second state information indicating the first abnormal state is continuous is equal to or less than the second threshold number (S409-NO), the warning signal output unit 24 outputs the first signal. An attention signal is output (S410). The first caution signal is a signal indicating that a state in which ore that has not been completely reduced may drop may occur. Next, the process proceeds to S412.

  When the operating state estimating unit 53 determines that the second state information indicating the first abnormal state has continued more than the second threshold number of times (S409-YES), the warning signal output unit 24 outputs the first warning signal. It outputs (S411). The first alarm signal is a signal indicating that there is a high possibility that a state in which ore that has not been completely reduced drops will occur. Next, the process proceeds to S412.

  Next, the operating state estimating unit 53 determines whether or not the most frequent operating state corresponding to the second state information stored in the processing of S406 is the second abnormal state (S412). If the operation state estimating unit 53 determines that the most frequent operation state is not the second abnormal state (S412-NO), the process proceeds to S417.

  When the operating state estimating unit 53 determines that the most frequent operating state is the second abnormal state (S412-YES), the operating state estimating unit 53 checks the second state information over the second unit time for the past several times, and It is determined whether or not the second state information indicating the abnormal state has continued more than a predetermined third threshold number of times (S413). The third threshold number of times is defined according to the length of the first unit time and the second unit time. For example, when the first unit time is 30 seconds and the second unit time is 5 minutes, the number of times of the third threshold may be 3, and the first unit time is 60 seconds and the second unit time is When the time is 10 minutes, the third threshold number may be three. When the operating state estimating unit 53 determines that the number of times the second state information indicating the second abnormal state is continuous is equal to or less than the third threshold number (S413-NO), the process proceeds to S417.

  If the operating state estimating unit 53 determines that the second state information indicating the second abnormal state has continued more than the third threshold number of times (S413-YES), the process proceeds to S414. The operating state estimating unit 53 checks the second state information over the second unit time for the past several times, and determines that the second state information indicating the second abnormal state is larger than the third threshold number of times. It is determined whether or not the number of consecutive times exceeds the threshold value (S414). The fourth threshold number of times is defined according to the length of the first unit time and the second unit time. For example, when the first unit time is 30 seconds and the second unit time is 5 minutes, the fourth threshold number may be set to 5, the first unit time is 60 seconds, and the second unit time is When it is 10 minutes, the fourth threshold number may be set to five times.

  If the operating state estimating unit 53 determines that the number of times the second state information indicating the second abnormal state is continuous is equal to or less than the fourth threshold number (S414-NO), the warning signal output unit 24 outputs the second signal. An attention signal is output (S415). The second caution signal is a signal indicating that the pulverized coal supply pipe 104 that supplies the pulverized coal to the tuyere 102 may be damaged. Next, the process proceeds to S417.

  When the operating state estimating unit 53 determines that the second state information indicating the second abnormal state has continued more than the second threshold number of times (S414-YES), the warning signal output unit 24 outputs the second warning signal. It outputs (S416). The second alarm signal is a signal indicating that there is a high possibility that the pulverized coal supply pipe 104 that supplies the pulverized coal to the tuyere 102 will be damaged. Next, the process proceeds to S417.

  Next, the operating state estimating unit 53 determines whether or not the most frequent operating state corresponding to the second state information stored in the processing of S406 is the cautionary state (S417). If the operation state estimating unit 53 determines that the most frequent operation state is not the cautionary state (S417-NO), the process proceeds to S420.

  When the operating state estimating unit 53 determines that the most frequent operating state is the cautionary state (S417-YES), the operating state estimating unit 53 checks the second state information over the second unit time for the past several times, and It is determined whether or not the second state information indicating the number of consecutive times exceeds a predetermined fifth threshold value (S418). The fifth threshold number of times is defined according to the length of the first unit time and the second unit time. For example, when the first unit time is 30 seconds and the second unit time is 5 minutes, the fifth threshold number may be set to 3, the first unit time is 60 seconds and the second unit time is When the time is 10 minutes, the fifth threshold number of times may be three. If the operating state estimating unit 53 determines that the number of times the second state information indicating the cautionary state is continuous is equal to or less than the fifth threshold number (S418-NO), the process proceeds to S420.

  When the operating state estimating unit 53 determines that the second state information indicating the cautionary state has continued more than the fifth threshold number of times (S418-YES), the warning signal output unit 24 outputs a third caution signal. (S419). The third caution signal is a signal indicating that an abnormality may occur in the operation state of the blast furnace 100. Next, the process proceeds to S420.

  Since the process of S420 is the same as the process of S117, a detailed description is omitted here.

(Operation and Effect of the Tuyere Monitoring Device According to the Fourth Embodiment)
The tuyere monitoring device 4 estimates that the operation state of the blast furnace 100 is abnormal when the second unit time indicating the abnormal state continues for a predetermined threshold number of times or more. And the likelihood of being estimated can be reduced.

(Modification of tuyere monitoring device according to embodiment)
The tuyere monitoring devices 1 to 4 employ the first abnormal state, the second abnormal state, the cautionary state, and the normal state which does not correspond to any of them as the operating state of the blast furnace 100. Is not limited to this, and may include a lance purge state and the like, and may include at least three states of a first abnormal state, a second abnormal state, and a normal state. In the tuyere monitoring devices 1 to 4, the first abnormal state is used as the raw ore falling state, and the second abnormal state is used as the lance breakage state. May be used as an abnormal state.

  In addition, the tuyere monitoring devices 1 to 4 calculate the ratio of the number of pieces of the first state information indicating the operation states of the blast furnace 100 based only on the first state information stored during the latest second unit time. Calculate. However, the tuyere monitoring device according to the embodiment, based on the moving average of the number of pieces of the first state information indicating each of the operating states of the blast furnace 100 stored during the plurality of second unit times, The ratio of the number of pieces of the first state information indicating each of the operation states may be calculated. The tuyere monitoring device according to the embodiment is configured to calculate the number of pieces of first state information based on a moving average of the numbers of pieces of first state information indicating the operating states of the blast furnace 100 stored during a plurality of second unit times. By calculating the ratio, the risk of erroneous estimation due to variation in the second unit time can be reduced. It is preferable that the tuyere monitoring device according to the embodiment calculates the ratio of the number of the first state information based on the moving average value of the number of the first state information for 15 minutes or more and 30 minutes or less. For example, when the second unit time is 5 minutes, the tuyere monitoring device according to the embodiment uses the first state information based on the moving average value of the number of the first state information in the second unit time of 3 or more and 6 or less. Is preferably calculated.

  In addition, the tuyere monitoring device according to the embodiment acquires lance purge time information indicating the lance purge time during which pulverized coal is not supplied into the inside of the blast furnace 100 from the tuyere 102, and excludes the lance purge time to operate the blast furnace 100. May be estimated. The tuyere monitoring device according to the embodiment acquires the lance purge time information from the storage unit 12 when the lance purge time information is stored in the storage unit 12 in advance. When the lance purge time information is not stored in the storage unit 12 in advance, the tuyere monitoring device according to the embodiment acquires the lance purge time information from the blast furnace control device (not shown) via the LAN 109. The tuyere monitoring device according to the embodiment performs tuyere monitoring processing using less data and with a smaller calculation amount by estimating the presence or absence of an abnormality in the operation state of the blast furnace 100 excluding the lance purge time. be able to.

  Further, the tuyere monitoring device according to the embodiment continues to output the warning signal every time the process is repeated even after outputting the warning signal, but does not output the warning signal after outputting the warning signal once or a predetermined number of times. You may do so. For example, the tuyere monitoring device according to the embodiment outputs the first alarm signal a predetermined number of times, and then outputs the first alarm signal until the frequency of the first abnormal state in the second unit time falls below the frequency of the normal state in the second unit time. The warning signal may not be output. Similarly, the tuyere monitoring device according to the embodiment outputs the second alarm signal a predetermined number of times, and continues until the frequency of the second abnormal state in the second unit time falls below the frequency of the normal state in the second unit time. (2) The warning signal may not be output.

  FIG. 12 is a diagram showing a change over time in the operating state of the blast furnace when a raw ore falling state, also referred to as a first abnormal state, has occurred. In FIG. 12, the horizontal axis indicates elapsed time, and the left vertical axis indicates the respective ratios of the lance purge state, the normal state, the cautionary state, the first abnormal state, and the second abnormal state in the first unit time. The right vertical axis shows the number of times of the lance purge state, the normal state, the cautionary state, the first abnormal state, and the second abnormal state in the first unit time. Also, a US mark indicates a lance purge state, a circle indicates a normal state, a triangle indicates a cautionary state, a cross indicates a first abnormal state, and a square indicates a second abnormal state. The lance purge state is determined based on lance purge time information indicating the lance purge time acquired from the blast furnace control device. That is, here, the presence or absence of an abnormality in the operation state of the blast furnace is estimated excluding the lance purge time.

  In the example shown in FIG. 12, the first unit time is 1 second, the second unit time is 5 minutes, and the first threshold ratio for determining whether or not to output the first attention signal is 30%. The second threshold ratio for judging the necessity of outputting the first alarm signal is 50%.

  The normal state is dominant until the elapsed time reaches 2 hours 30 minutes, but the cautionary state starts to increase after the elapsed time exceeds 2 hours 30 minutes and needs attention when the elapsed time exceeds 3 hours The frequency of the condition exceeds the normal condition requiring caution. At the same time, after the elapsed time exceeded 3 hours, the ratio of the first abnormal state began to increase.

  When the elapsed time is 3 hours and 25 minutes, the first warning signal is output because the ratio of the first abnormal state exceeds 30%, and when the elapsed time is 3 hours and 35 minutes, the first abnormal state is output. The first alarm signal was output when the ratio of .gtoreq. Exceeded 50%. In response to the first alarm signal being output when the elapsed time was 3 hours and 35 minutes, the operator increased the ratio of the reducing material in order to eliminate the raw ore falling state.

  When the elapsed time is 3 hours and 55 minutes, the ratio of the first abnormal state reaches 90% and shows a peak, but the ratio of the first abnormal state decreases in response to the operator increasing the ratio of the reducing material. I do. When the elapsed time is 4 hours and 25 minutes, the ratio of the normal state exceeds the ratio of the first abnormal state, and when the elapsed time exceeds 6 hours, the ratio of the normal state exceeds 80%, and the blast furnace is operated. The state has returned to normal.

  FIG. 13 is a diagram illustrating a temporal change in the operating state of the blast furnace when a lance breakage state, also referred to as a second abnormal state, has occurred. In FIG. 13, the horizontal axis indicates elapsed time, and the left vertical axis indicates the respective ratios of the lance purge state, the normal state, the cautionary state, the first abnormal state, and the second abnormal state in the first unit time. The right vertical axis shows the number of times of the lance purge state, the normal state, the cautionary state, the first abnormal state, and the second abnormal state in the first unit time. Also, a US mark indicates a lance purge state, a circle indicates a normal state, a triangle indicates a cautionary state, a cross indicates a first abnormal state, and a square indicates a second abnormal state. The lance purge state is determined based on lance purge time information indicating the lance purge time acquired from the blast furnace control device. That is, here, the presence or absence of an abnormality in the operation state of the blast furnace is estimated excluding the lance purge time.

  In the example illustrated in FIG. 13, the first unit time is 1 second, the second unit time is 5 minutes, and the third threshold ratio for determining whether to output the second attention signal is 30%. The fourth threshold ratio for judging the necessity of outputting the second alarm signal is 50%.

  The normal state is dominant until the elapsed time reaches 4 hours 30 minutes, but the cautionary state starts to increase after the elapsed time exceeds 4 hours 30 minutes, and after the elapsed time exceeds 5 hours, the second state is changed. The rate of abnormal conditions has begun to rise.

  When the elapsed time is 5 hours and 35 minutes, the second warning signal is output because the ratio of the second abnormal state exceeds 30%. When the elapsed time is 5 hours and 45 minutes, the second abnormal state is output. The second alarm signal was output when the ratio of the data exceeded 50%. In response to the second alarm signal being output when the elapsed time is 5 hours and 45 minutes, the operator confirms that the lance has been damaged from the tuyere, and then when the elapsed time is 6 hours and 10 minutes. Switched to lance purge state. If the switching to the lance purge state is delayed, there is a possibility that the tuyere is melted and the blast furnace must be stopped. The tuyere monitoring device outputs the second caution signal and the second alarm signal to urge the operator to take measures, thereby enabling the blast furnace to operate stably.

1-4 Tuyere monitoring device 15 Learning unit 21, 41 Image data acquisition unit 22 Operation state acquisition unit 23, 43, 53 Operation state estimation unit 24 Warning signal output unit 35 Second unit time change unit 100 Blast furnace 101 Furnace body 102 blades Mouth 103 Hot air supply pipe 104 Pulverized coal supply pipe 105 Observation window 106 Imaging device 107 Raceway 108 Imaging unit

Claims (17)

A plurality of images taken at different times by an imaging device having an imaging unit arranged to face any one of the tuyeres for supplying pulverized coal together with gas into the inside of the blast furnace, and at least a normal state and a first abnormality A learning device that has learned a relationship with the operating state of the blast furnace including a state and a second abnormal state;
An image data acquisition unit that acquires image data indicating an image captured by the imaging device at predetermined imaging intervals,
An operation state acquisition unit that acquires the operation state of the blast furnace output from the learning device in response to inputting the image data to the learning device,
An operation state estimation unit that estimates the presence or absence of an abnormality in the operation state of the blast furnace, based on the operation state acquired by the operation state acquisition unit,
A warning signal output unit that outputs a warning signal when the operation state of the blast furnace is estimated to be abnormal by the operation state estimation unit,
The operating state estimating unit calculates a most frequent operating state among the plurality of operating states acquired during a first unit time longer than the imaging interval, and sets an operating state of the blast furnace in the first unit time. And the determined operation state of the blast furnace is stored as first state information.   The tuyere monitoring device according to claim 1, wherein the first unit time is a time of 1 second or more and 60 seconds or less.   The operating state estimating unit is configured to determine an operating state of the blast furnace based on a frequency of the operating state of the blast furnace corresponding to the first state information stored during a second unit time longer than the first unit time. The tuyere monitoring device according to claim 1, wherein the presence or absence of an abnormality is estimated.   The tuyere monitoring device according to claim 3, wherein the second unit time is a time of 5 minutes or more and 10 minutes or less.   A second unit time change for estimating a collapse period of a raceway formed inside the blast furnace based on operating conditions of the blast furnace and changing the second unit time according to the estimated collapse period of the raceway; The tuyere monitoring device according to claim 3, further comprising a unit.   The operation state estimating unit is a ratio of the number of the first state information indicating each of the first abnormal state and the second abnormal state to the number of the first state information stored during the second unit time. The tuyere monitoring device according to any one of claims 3 to 5, wherein the presence or absence of an abnormality in an operation state of the blast furnace is estimated based on the condition. The first abnormal state is a state in which ore that has not been completely reduced drops.
The operating state estimating unit may determine that a ratio of the number of the first state information indicating the first abnormal state to the number of the first state information stored during the second unit time is a predetermined first threshold. Judge whether it is larger than the ratio,
The warning signal output unit is completely reduced when the operating state estimating unit determines that the ratio of the number of pieces of the first state information indicating the first abnormal state is larger than the first threshold ratio. The tuyere monitoring device according to claim 6, wherein the tuyere monitoring device outputs a first caution signal indicating that there is a possibility that a state in which unreacted ore is dropped may occur. The operating state estimating unit may determine that a ratio of the number of the first state information indicating the first abnormal state to the number of the first state information stored during the second unit time is the first threshold ratio. It is determined whether it is greater than a second threshold ratio that is greater than
The warning signal output unit completely returns when the operating state estimating unit determines that the ratio of the number of pieces of the first state information indicating the first abnormal state is larger than the second threshold ratio. The tuyere monitoring device according to claim 7, wherein the tuyere monitoring device outputs a first alarm signal indicating that there is a high possibility that a state in which unremoved ore is dropped is high. The second abnormal state is a state in which a pulverized coal supply pipe that supplies pulverized coal to the tuyere is damaged,
The operating state estimating unit may determine that a ratio of the number of the first state information indicating the second abnormal state to the number of the first state information stored during the second unit time is a predetermined third threshold value. Determine whether or not it is greater than the ratio,
The warning signal output unit outputs the tuyere to the tuyere when the operating state estimating unit determines that the ratio of the number of the first state information indicating the second abnormal state is larger than the third threshold ratio. The tuyere monitoring device according to any one of claims 6 to 8, which outputs a second caution signal indicating that a state in which the pulverized coal supply pipe that supplies the pulverized coal may be damaged may occur. The operating state estimating unit may determine that a ratio of the number of the first state information indicating the second abnormal state to the number of the first state information stored during the second unit time is the third threshold ratio. It is determined whether or not it is greater than a fourth threshold ratio that is greater than
The warning signal output unit outputs the tuyere to the tuyere when the operating state estimating unit determines that the ratio of the number of the first state information indicating the second abnormal state is greater than the fourth threshold ratio. The tuyere monitoring device according to claim 9, wherein the tuyere monitoring device outputs a second alarm signal indicating that a state in which the pulverized coal supply pipe that supplies the pulverized coal is likely to be damaged is high. The operating state of the blast furnace further includes a cautionary state indicating that an abnormality may occur in the operating state of the blast furnace,
The operating state estimating unit may determine that a ratio of the number of the first state information indicating the cautionary state to the number of the first state information stored during the second unit time is a predetermined fifth threshold frequency. Judge whether it is greater than
The operating state of the blast furnace when the operating state estimating unit determines that the ratio of the number of pieces of the first state information indicating the cautionary state is larger than the fifth threshold ratio. The tuyere monitoring device according to any one of claims 6 to 10, wherein the tuyere monitoring device outputs a third caution signal indicating that an abnormality may occur in the tuyere.   The said operation state estimation part determines the number of the said 1st state information based on the moving average value of the frequency of the said 1st state information memorize | stored during several said 2nd unit time. The tuyere monitoring device according to claim 1.   The operating state estimating unit stores, as second state information, an operating state of the blast furnace corresponding to the most frequent first state information among the first state information stored during the second unit time. When the second state information indicating one of the first abnormal state and the second abnormal state continues for a predetermined threshold number of times or more, it is estimated that there is an abnormality in the operation state of the blast furnace. The tuyere monitoring device according to any one of Items 3 to 5.   The operation state estimating unit estimates the presence or absence of an abnormality in the operation state of the blast furnace at every third unit time longer than the first unit time and defined independently of the second unit time. The tuyere monitoring device according to any one of claims 13 to 13.   The operation state estimation unit acquires lance purge time information indicating a lance purge time during which pulverized coal is not supplied from the tuyere to the inside of the blast furnace, and excludes image data indicating an image captured during the lance purge time, The tuyere monitoring device according to any one of claims 1 to 13, which estimates presence or absence of an abnormality in an operation state of the blast furnace. A plurality of images taken at different times by an imaging device having an imaging unit arranged to face any one of the tuyeres for supplying pulverized coal together with gas into the inside of the blast furnace, and at least a normal state and a first abnormality In a tuyere monitoring device having a learning device that has learned a relationship with the operating state of the blast furnace including a state and a second abnormal state,
Acquiring image data indicating an image captured by the imaging device at predetermined imaging intervals,
In response to inputting the image data to the learning device, to obtain the operating state of the blast furnace output from the learning device,
Based on the obtained operating state, to estimate the presence or absence of an abnormality in the operating state of the blast furnace,
When it is estimated that there is an abnormality in the operation state of the blast furnace, to output a warning signal, causing the computer to perform processing,
The process of estimating the presence or absence of an abnormality in the operating state of the blast furnace includes the step of determining the most frequent operating state among the plurality of operating states acquired over a first unit time longer than the imaging interval. A tuyere monitoring program, comprising: determining an operating state of the blast furnace per unit time, and storing the determined operating state of the blast furnace as first state information. A plurality of images taken at different times by an imaging device having an imaging unit arranged to face any one of the tuyeres for supplying pulverized coal together with gas into the inside of the blast furnace, and at least a normal state and a first abnormality In a tuyere monitoring device having a learning device that has learned a relationship with the operating state of the blast furnace including a state and a second abnormal state,
Acquiring image data indicating an image captured by the imaging device at predetermined imaging intervals,
In response to inputting the image data to the learning device, to obtain the operating state of the blast furnace output from the learning device,
Based on the obtained operating state, to estimate the presence or absence of an abnormality in the operating state of the blast furnace,
When it is estimated that there is an abnormality in the operation state of the blast furnace, output a warning signal, including processing,
The process of estimating the presence or absence of an abnormality in the operating state of the blast furnace includes the step of determining the most frequent operating state among the plurality of operating states acquired over a first unit time longer than the imaging interval. A tuyere monitoring method comprising: determining the operating state of the blast furnace in a unit time, and storing the determined operating state of the blast furnace as first state information.