Abnormality detection system

The unmanned anomaly detection system with drones and machine learning improves radiant tube inspection efficiency and safety by accurately identifying cracks and bends, reducing human error and information leakage.

JP2024163761AActive Publication Date: 2024-11-22JFE STEEL CORP
View PDF 5 Cites 0 Cited by

Patent Information

Application Number
JP2023079632
Authority / Receiving Office
JP · JP
Patent Type
Applications
Current Assignee / Owner
Filing Date
2023-05-12
Publication Date
2024-11-22
Estimated Expiration
2043-05-12

AI Technical Summary

Technical Problem

Conventional methods for inspecting radiant tubes in steel plants are inefficient and pose safety risks due to manual inspection, with potential for human error and lack of accurate anomaly detection systems.

Method used

An unmanned anomaly detection system using drones for imaging, combined with machine learning algorithms to identify cracks and bends in radiant tubes, and a dual-network communication system for secure and efficient data transmission.

Benefits of technology

The system reduces human risk, enhances inspection accuracy, and minimizes information leakage by utilizing drones and machine learning for precise anomaly detection with reduced manual intervention.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 2024163761000001_ABST
    Figure 2024163761000001_ABST
Patent Text Reader

Abstract

To improve abnormality inspection technology.SOLUTION: An abnormality detection system 100 includes an unmanned photographing device 1 which photographs an image of a facility to be inspected, a communication terminal 4 which acquires a photographed image from the unmanned photographing device 1, and an abnormality specifying device 6 which acquires a photographed image from the communication terminal 4 and specifies the abnormality of the inspection target from the photographed image. The unmanned photographing device 1 and the communication terminal 4 can communicate wirelessly through a first dedicated network. The communication terminal 4 and the abnormality specifying device 6 can perform radio communication through a second dedicated network. The communicable distance of the second dedicated network is longer than that of the first dedicated network.SELECTED DRAWING: Figure 1
Need to check novelty before this filing date? Find Prior Art

Description

[Technical field]

[0001] The present disclosure relates to an equipment anomaly detection system. [Background technology]

[0002] Radiant tubes (hereinafter, also referred to as RT) are commonly used as equipment in steelworks, for example, as heating devices in annealing furnaces. RTs deteriorate over time due to thermal expansion, and abnormalities such as cracks in the tubes and bending that causes the tubes themselves to sag occur, so if an abnormality is found in the RT, it is necessary to replace it appropriately. Conventional RT inspection involves workers climbing between rows of RTs using their hands and feet, visually inspecting the RT to be inspected, and writing down any abnormalities on a checklist, which causes the risk of falling and increases the number of man-hours. Summary of the Invention [Problem to be solved by the invention]

[0003] In the conventional inspection method described above, it was necessary to enter abnormalities in a checklist, and there was a possibility of human error, such as mistakes in entering information in the checklist or overlooking abnormalities in the RT. In recent years, it has become common to use unmanned photography means such as drones to inspect facilities at various locations, but no particular consideration has been given to a system configuration that ensures the safety of photography using unmanned photography devices, ensures confidentiality of information, and efficiently and highly accurately inspects facilities for abnormalities. In other words, there was room for improvement in conventional anomaly detection technology.

[0004] An object of the present disclosure, made in consideration of the above circumstances, is to improve equipment anomaly detection technology. [Means for solving the problem]

[0005] (1) An anomaly detection system according to an embodiment of the present disclosure includes an unmanned photography device that captures images of equipment to be inspected, a communication terminal that acquires the captured images from the unmanned photography device, and an anomaly identification device that acquires the captured images from the communication terminal and identifies an anomaly in the inspection target from the captured images, The unmanned photographing device and the communication terminal are capable of wireless communication through a first dedicated network, and the communication terminal and the anomaly identification device are capable of wireless communication through a second dedicated network, and a communication distance of the second dedicated network is longer than a communication distance of the first dedicated network.

[0006] (2) An anomaly detection system according to an embodiment of the present disclosure is the anomaly detection system according to (1), The anomaly identification device is capable of identifying an anomaly in a radiant tube using an anomaly identification program capable of identifying an anomaly including a crack in a radiant tube that heats a steel plate, The anomaly identification program is machine-learned based on first training data related to cracks in the radiant tube, and the first training data includes images that have been processed based on photographed images of the crack for the purpose of increasing information on the width, length, and clarity of the crack.

[0007] (3) An anomaly detection system according to an embodiment of the present disclosure is the anomaly detection system according to (1) or (2), The anomaly identification device is capable of identifying an anomaly in the radiant tube by using an anomaly identification program capable of identifying an anomaly including a bend in the radiant tube that heats the steel plate, The anomaly identification program is machine-learned based on second training data related to the bends in the radiant tube, and the second training data includes images of the same bend taken from multiple angles.

[0008] (4) An anomaly detection system according to an embodiment of the present disclosure is the anomaly detection system according to (2) or (3), when the abnormality identifying device identifies an abnormality in the radiant tube, the abnormality identifying device transmits result information indicating the existence of the identified abnormality to the communication terminal; The result information includes a risk indicator that indicates the risk of an abnormality.

[0009] (5) An anomaly detection system according to an embodiment of the present disclosure is the anomaly detection system according to any one of (2) to (4), The anomaly identification device acquires operating conditions related to the radiant tubes, predicts whether or not an anomaly will occur in each radiant tube and the level of risk, sets a priority for each radiant tube based on the predicted level of risk, and the result information includes a priority indication indicating the priority. Effect of the Invention

[0010] According to the facility anomaly detection system of an embodiment of the present disclosure, the facility anomaly detection system photographs the facility using an unmanned camera, eliminating the need for inspectors to climb the facility and reducing the risk of disasters. Furthermore, the communication in the facility anomaly detection system is performed using the first dedicated network or the second dedicated network, reducing the risk of information leakage. In other words, according to the facility anomaly detection system of an embodiment of the present disclosure, the facility anomaly detection technology can be improved. [Brief description of the drawings]

[0011] [Figure 1] FIG. 1 is a diagram illustrating a configuration example of an anomaly detection system according to an embodiment of the present disclosure. [Diagram 2] FIG. 2 is a diagram showing an example of a flight route of the unmanned photography device. [Diagram 3] FIG. 1 is a diagram showing an overall view of a radiant tube. [Figure 4] FIG. 2 is a schematic diagram showing the configuration of a radiant tube. [Diagram 5] 11 is an example of result information displayed by the communication terminal. [Figure 6] 1 is a diagram showing a hardware configuration of an unmanned photography device 1 according to an embodiment of the present disclosure. [Figure 7] 2 is a diagram illustrating a hardware configuration of a communication terminal 2 according to an embodiment of the present disclosure. [Figure 8] 2 is a diagram illustrating a hardware configuration of a communication terminal 4 according to an embodiment of the present disclosure. [Figure 9] 2 is a diagram illustrating a hardware configuration of an abnormality identifying device 6 according to an embodiment of the present disclosure. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0012] An anomaly detection system according to an embodiment of the present disclosure will be described below with reference to the drawings. Note that, although an RT used as a heating device for an annealing furnace will be described as an example of equipment to which the anomaly detection system of the present invention is applied, the equipment to be inspected is not limited to the RT. Examples of equipment other than the RT include a crane girder and a crack inside a chimney.

[0013] In each drawing, the same or corresponding parts are denoted by the same reference numerals. In the description of this embodiment, the description of the same or corresponding parts will be omitted or simplified as appropriate.

[0014] 1 shows an example of the configuration of an anomaly detection system 100 according to this embodiment. The anomaly detection system 100 includes an unmanned photography device 1, a communication terminal 2, a communication terminal 4, and an anomaly identification device 6.

[0015] The unmanned photography device 1 has an imaging unit such as a camera. For example, the unmanned photography device 1 may be any device that can capture images using the imaging unit without being attached to a human body (e.g., held in a hand or attached to a helmet), and may be, for example, an unmanned aerial vehicle. Unmanned aerial vehicles include drones. A captured image 3 captured by the imaging unit is transmitted from the unmanned photography device 1 to a communication terminal 2. The unmanned photography device 1 and the communication terminal 2 communicate with each other using a short-range wireless communication standard (a first dedicated network described later). In the following, in this embodiment, a case where the unmanned photography device 1 is a drone will be described, but this is not limited to this.

[0016] The communication terminal 2 and the communication terminal 4 are communication terminals equipped with a function of controlling and monitoring the unmanned aerial vehicle owned by an operator (hereinafter, simply referred to as an operator) such as a supervisor of the unmanned photographing device 1 or other related parties. In this embodiment, the communication terminal 2 is equipped with a function of controlling the unmanned aerial vehicle. The communication terminal 2 also transmits the photographed image 3 received from the unmanned photographing device 1 to the communication terminal 4. In other words, the communication terminal 4 acquires the photographed image from the unmanned photographing device 1. For example, the communication terminal 2 is a control controller of the unmanned photographing device 1. The communication terminal 4 is equipped with a function of monitoring the unmanned aerial vehicle. The communication terminal 4 also has a display unit that displays the photographed image 3 transmitted from the communication terminal 2. The display unit also displays the abnormality diagnosis result of the RT described later in a viewable manner. The communication terminal 4 is, for example, a tablet terminal. The communication terminal 2 and the communication terminal 4 are capable of wireless communication via a first dedicated network. The unmanned photographing device 1 and the communication terminal 4 may communicate directly. For example, the communication terminal 4 may directly receive the photographed image from the unmanned photographing device 1. In this case, the unmanned photography device 1 and the communication terminal 4 communicate via a first dedicated network. The functions of the communication terminals 2 and 4 may be configured by a single communication terminal. For example, the communication terminal 4 may have a function for controlling and monitoring the unmanned aerial vehicle. In this case, the unmanned photography device 1 and the communication terminal 4 communicate via the first dedicated network.

[0017] Anomaly identification device 6 is a server device or the like having an anomaly identification program that detects the presence or absence of anomaly based on captured image 5 of the inspection target. Captured image 5 is transmitted to anomaly identification device 6 from communication terminal 4. In other words, anomaly identification device 6 acquires the captured image from communication terminal 4 and identifies an anomaly of the inspection target from the captured image. Although an example in which anomaly detection system 100 is equipped with one anomaly identification device 6 is shown in Fig. 1, this is not limiting. Anomaly detection system 100 may be equipped with two or more anomaly identification devices 6.

[0018] The anomaly identification program is constructed so that anomalies occurring in the equipment to be inspected can be identified by machine learning. In this embodiment, the anomaly identification program is constructed by learning a large amount of teacher data (hereinafter also referred to as first teacher data) including images of abnormalities related to cracks as RT abnormalities, and teacher data (hereinafter also referred to as second teacher data) including images of abnormalities related to bending.

[0019] The first teacher data includes an image in which pixels have been adjusted based on the captured image of the crack in order to increase information on the width, length, and clarity of the crack. In other words, the first teacher data includes an image that has been processed based on the captured image of the crack in order to increase information on the width, length, and clarity of the crack. The second teacher data includes images of the same bend as bend data, for example, captured from multiple angles from the front, back, left, and right. In other words, the second teacher data includes images of the same bend captured at multiple angles. In this way, machine learning can be performed on the bend using images captured at multiple angles to improve the accuracy of anomaly detection related to anomalies in the bend.

[0020] The anomaly identification device 6 can determine whether an abnormality has occurred in the RT by inputting images captured by an unmanned photography device 1 such as a drone into an anomaly identification program. Furthermore, when an abnormality has occurred, the anomaly identification device 6 can identify whether a crack, a bend, or both have occurred. The abnormalities to be learned may be set appropriately according to the equipment.

[0021] In this embodiment, the unmanned photography device 1, the communication terminal 2, and the communication terminal 4, and the anomaly identification device 6 are capable of wireless communication using different communication standards. Hereinafter, the communication standard between the unmanned photography device 1, the communication terminals 2, and 4 is also referred to as a first dedicated network, and the communication standard between the communication terminal 4 and the anomaly identification device 6 is also referred to as a second dedicated network. Here, the "dedicated network" refers to a network that can be accessed by devices that have been granted access rights within a limited range.

[0022] The first dedicated network is a short-distance wireless communication standard, such as NFC communication, BlueTooth (registered trademark), WiFi (registered trademark), etc. The first dedicated network is a private dedicated network, and only communication devices authenticated at the time of initial connection or the like can communicate. In this embodiment, the unmanned photography device 1, the communication terminal 2, and the communication terminal 4 are assumed to be authenticated in advance. In this embodiment, a case where WiFi (registered trademark) is used will be described as an example.

[0023] The second dedicated network is a communication standard with a longer communication distance than the first dedicated network, and performs wireless communication via a communication network. For example, the second dedicated network is composed of a wireless communication network such as local 5G or private LTE, and a base station. The second dedicated network is a private dedicated network, and only communication devices authenticated at the time of initial connection or the like can communicate. In this embodiment, the communication terminal 4 and the anomaly identification device 6 are assumed to be authenticated in advance. In this embodiment, a case where private LTE is used will be described as an example.

[0024] FIG. 2 shows an example of the flight route of the unmanned camera 1 when inspecting RT8. Arrow 7 indicates the flight route of the unmanned camera 1. As shown in FIG. 2, a starting point for the unmanned camera 1 is provided at the bottom of the furnace directly below the RT8 row. The unmanned camera 1 is set at the starting point. With the unmanned camera 1 set at the starting point, a control signal for starting inspection is sent from the communication terminal 4 via the communication terminal 2, and the inspection starts when the unmanned camera 1 receives the control signal. An overall image and a schematic diagram of the RT8 are shown in FIG. 3 and FIG. 4. As shown in FIG. 3, the RT8 are arranged in multiple rows in the annealing furnace, and the steel plate is heated by the RT8. Also, as shown in FIG. 4, inside the tube of the RT8, the air is heated by a burner 81.

[0025] The flight route of the unmanned photography device 1 is determined in advance as described above. In this embodiment, the RT1 row is vertically divided into two, one of which flies while ascending and the other flies while descending. For example, in the flight route of arrow 7a, the unmanned photography device 1 flies while ascending. On the other hand, in the flight route of arrow 7b, the unmanned photography device 1 flies while descending.

[0026] Waypoints are set in advance for the flight route, and the route returns to the starting and ending points while taking RT8 photos at each waypoint. As described above, the unmanned photography device 1 takes photos at the set waypoints and transmits the captured images to the communication terminal 4 via the communication terminal 2 and the first dedicated network at any time. Such captured images include shooting position information indicating the time when the drone took the image and at which waypoint set on the flight route the image was taken. The communication terminal 4 transmits the received captured image 5 to the anomaly identification device 6 via the second dedicated network, and the anomaly identification device 6 that receives the captured image 5 executes an anomaly identification program to diagnose whether or not there is an anomaly in the RT8. The diagnosis result is transmitted to the communication terminal 4 via the second dedicated network, and the communication terminal 4 displays the diagnosis result. The operator can check the diagnosis result using the communication terminal 4.

[0027] In this embodiment, the reason why the communication between the unmanned photography device 1 and the communication terminal 2 and communication terminal 4 is performed via the first dedicated network, which is short-range wireless communication, is that the flight state / inspection progress of the unmanned photography device 1 should be monitored at a distance visible to the operator, and to ensure the image transmission speed from the unmanned photography device 1. On the other hand, the reason why the communication standard between the communication terminal 4 and the anomaly identification device 6 is the second dedicated network is that the execution of the anomaly identification program places a load on the user, and therefore it is necessary to transmit captured images to the anomaly identification device 6 installed in a remote location different from the inspection location (measurement site).

[0028] In this embodiment, the unmanned photography device 1 and the anomaly identification device 6 do not communicate directly. That is, the image captured by the unmanned photography device 1 is always transmitted to the anomaly identification device 6 via the communication terminals 2 and 4. With this configuration, for example, when a failure occurs in the second dedicated network, the operator can check the captured image itself using the communication terminal 4 to confirm whether or not there is an anomaly. When transmitting the captured image to the anomaly identification device 6, by adding facility information (information indicating which of the multiple RTs the image is captured for), it is possible to transmit diagnostic result information (hereinafter also referred to as result information) including past diagnostic results for a specific RT stored in the anomaly identification device 6 to the communication terminal 4 in a lump, and the result information can be viewed on the communication terminal 4.

[0029] FIG. 5 shows an example of result information displayed by the communication terminal 4. The result information displayed by the communication terminal 4 includes a rectangular frame object 9 indicating whether or not an abnormality has occurred in the RT to be inspected, a display object 10 (displaying a "crack") indicating the type of abnormality, and an abnormality judgment display 11. The rectangular frame object 9 allows the operator to easily grasp the location of the abnormality. Furthermore, the display object 10 allows the operator to easily grasp the type of abnormality. The abnormality judgment display 11 is an object indicating that the image contains an abnormality. This allows the operator to easily grasp result information containing an abnormality. Note that the display mode of the diagnosis result is not limited to this, and the presence or absence and type of abnormality may be indicated by a single display method, such as changing the frame color of the rectangular frame object 9 depending on the type of abnormality.

[0030] Also, the degree of danger may be displayed according to the degree of abnormality, such as the size of the crack, the size of the bend, etc. In other words, the result information may include a danger level display that indicates the degree of danger of the abnormality. The text object 12 in Fig. 5 is an example of a danger level display. For example, the danger level display may be in three stages: high, medium, and low.

[0031] The abnormality identification device 6 may obtain the operating conditions for the radiant tubes, predict the occurrence of an abnormality and the degree of risk in each radiant tube, and set a priority for each radiant tube based on the predicted degree of risk. In this case, the result information may include a priority indication indicating such priority. By displaying the priority, for example, when an abnormality occurs in multiple RTs (RTs located at different positions), a maintenance worker can easily determine which RT to start maintenance work on. Alternatively, it is possible to encourage maintenance work in the order of priority by ranking the RTs from the highest priority and transmitting the result information to the communication terminal 4. The text object 13 in FIG. 5 is an example of a priority indication. For example, the priority indication may be in three stages, high, medium, and low. Note that FIG. 5 shows an example in which only the priority of one location is displayed, but this is not limited to this. The communication terminal 4 may display the priorities of the result information of multiple locations side by side. In this case, the priority may be represented by the color of each rectangular frame 14 related to each image. For example, when the priority is represented in three stages, high, medium, and low, the color of the rectangular frame 14 is red when the priority is high. Moreover, when the priority is medium, the color of the rectangular frame 14 is set to yellow. On the other hand, when the priority is low, the color of the rectangular frame 14 is set to green. This allows the operator to easily recognize images with high priority.

[0032] In the above description, the unmanned camera 1 flies autonomously along a predetermined flight route, but the present invention can also be applied to manual drone operation by a pilot. The present invention can also be applied to videos captured by the unmanned camera 1, but from the viewpoint of reducing the data size of data transmission and reception, application to images is preferable. By converting the data into images, it is possible to transfer data at any time during the flight of the unmanned camera 1 and perform AI analysis (anomaly detection) in real time. The unmanned camera means is not limited to the unmanned camera 1 such as a drone, and any method can be adopted as long as it is a method that allows unmanned shooting without requiring workers to climb up the equipment while carrying a camera. For example, an omnidirectional camera inspection unit equipped with an imaging unit such as a camera may be used.

[0033] (Example) Table 1 shows a comparison of the anomaly identification rate and identification time when the anomaly detection system 100 according to this embodiment is introduced with the conventional method. The anomaly identification rate for cracks and bends has improved significantly, and it has also become possible to significantly shorten the identification time.

[0034] [Table 1]

[0035] (Hardware configuration) 6 is an example of a hardware configuration of the unmanned photography device 1 according to an embodiment of the present disclosure. As shown in FIG. 6, the unmanned photography device 1 includes a control unit 101, a storage unit 102, a communication unit 103, a positioning unit 104, and an imaging unit 105.

[0036] The control unit 101 includes at least one processor, at least one dedicated circuit, or a combination of these. The processor is a general-purpose processor such as a central processing unit (CPU) or a graphics processing unit (GPU), or a dedicated processor specialized for a specific process. The dedicated circuit is, for example, a field-programmable gate array (FPGA) or an application specific integrated circuit (ASIC). The control unit 101 executes processes related to the operation of the unmanned photography device 1 while controlling each part of the unmanned photography device 1.

[0037] The storage unit 102 includes at least one semiconductor memory, at least one magnetic memory, at least one optical memory, or a combination of at least two of these. The semiconductor memory is, for example, a random access memory (RAM) or a read only memory (ROM). The RAM is, for example, a static random access memory (SRAM) or a dynamic random access memory (DRAM). The ROM is, for example, an electrically erasable programmable read only memory (EEPROM). The storage unit 102 functions as, for example, a main storage device, an auxiliary storage device, or a cache memory. The storage unit 102 stores data used in the operation of the unmanned photography device 1 and data obtained by the operation of the unmanned photography device 1.

[0038] The communication unit 103 includes at least one wireless communication interface. The wireless communication interface is an interface that supports communication via the first dedicated network described above. The communication unit 103 receives data used in the operation of the unmanned photography device 1, and transmits data obtained by the operation of the unmanned photography device 1.

[0039] The positioning unit 104 includes a receiver compatible with a satellite positioning system. The receiver is compatible with, for example, a GPS (Global Positioning System), but is not limited thereto, and may be compatible with any satellite positioning system. The positioning unit 104 also includes, for example, a gyro sensor, a geomagnetic sensor, and an air pressure sensor. In this embodiment, the unmanned photography device 1 can obtain its own photography position information, the direction in which the device is facing, and the inclination of the device using the positioning unit 104. The position information may include two-dimensional coordinate data including latitude and longitude, or may include three-dimensional coordinate data including altitude in addition to latitude and longitude.

[0040] The imaging unit 105 includes an imaging device such as a camera that captures an image of an object. The imaging device may be configured with, for example, a complementary metal oxide semiconductor (CMOS) image sensor or a charge coupled device (CCD).

[0041] 7 is an example of a hardware configuration of a communication terminal 2 according to an embodiment of the present disclosure. As shown in FIG. 7, the communication terminal 2 includes a control unit 201, a storage unit 202, a communication unit 203, and an input unit 204.

[0042] The control unit 201 includes at least one processor, at least one dedicated circuit, or a combination of these. The processor is a general-purpose processor such as a central processing unit (CPU) or a graphics processing unit (GPU), or a dedicated processor specialized for a specific process. The dedicated circuit is, for example, a field-programmable gate array (FPGA) or an application specific integrated circuit (ASIC). The control unit 201 executes processes related to the operation of the communication terminal 2 while controlling each unit of the communication terminal 2.

[0043] The storage unit 202 includes at least one semiconductor memory, at least one magnetic memory, at least one optical memory, or a combination of at least two of these. The semiconductor memory is, for example, a random access memory (RAM) or a read only memory (ROM). The RAM is, for example, a static random access memory (SRAM) or a dynamic random access memory (DRAM). The ROM is, for example, an electrically erasable programmable read only memory (EEPROM). The storage unit 202 functions as, for example, a main storage device, an auxiliary storage device, or a cache memory. The storage unit 202 stores data used in the operation of the communication terminal 2 and data obtained by the operation of the communication terminal 2.

[0044] The communication unit 203 includes at least one wireless communication interface. The wireless communication interface is an interface corresponding to communication via the first dedicated network described above. The communication unit 203 receives data used in the operation of the communication terminal 2, and transmits data obtained by the operation of the communication terminal 2.

[0045] The input unit 204 includes at least one input interface. The input interface is, for example, a physical key, a capacitive key, a pointing device, or a touch screen that is integrated with a display. The input unit 204 accepts an operation for inputting data used for the operation of the communication terminal 2.

[0046] 8 is an example of a hardware configuration of a communication terminal 4 according to an embodiment of the present disclosure. As shown in FIG. 8, the communication terminal 4 includes a control unit 401, a storage unit 402, a communication unit 403, an input unit 404, and a display unit 405.

[0047] The control unit 401 includes at least one processor, at least one dedicated circuit, or a combination of these. The processor is a general-purpose processor such as a central processing unit (CPU) or a graphics processing unit (GPU), or a dedicated processor specialized for a specific process. The dedicated circuit is, for example, a field-programmable gate array (FPGA) or an application specific integrated circuit (ASIC). The control unit 401 executes processes related to the operation of the communication terminal 4 while controlling each unit of the communication terminal 4.

[0048] The storage unit 402 includes at least one semiconductor memory, at least one magnetic memory, at least one optical memory, or a combination of at least two of these. The semiconductor memory is, for example, a random access memory (RAM) or a read only memory (ROM). The RAM is, for example, a static random access memory (SRAM) or a dynamic random access memory (DRAM). The ROM is, for example, an electrically erasable programmable read only memory (EEPROM). The storage unit 402 functions as, for example, a main storage device, an auxiliary storage device, or a cache memory. The storage unit 402 stores data used in the operation of the communication terminal 4 and data obtained by the operation of the communication terminal 4.

[0049] The communication unit 403 includes at least two wireless communication interfaces. Specifically, the wireless communication interfaces include an interface corresponding to communication via the first dedicated network and an interface corresponding to communication via the second dedicated network. The communication unit 403 receives data used for the operation of the communication terminal 4, and transmits data obtained by the operation of the communication terminal 4.

[0050] The input unit 404 includes at least one input interface. The input interface is, for example, a physical key, a capacitive key, a pointing device, or a touch screen integrally provided with a display. The input interface may also be, for example, a microphone that accepts voice input, or a camera that accepts gesture input. The input unit 404 accepts an operation to input data used for the operation of the communication terminal 4. The input unit 404 may be connected to the communication terminal 4 as an external input device instead of being provided in the communication terminal 4. As a connection method, any method such as USB (Universal Serial Bus), HDMI (registered trademark) (High-Definition Multimedia Interface), or Bluetooth (registered trademark) may be used.

[0051] The display unit 405 includes at least one display output interface. The display output interface is, for example, a display that outputs information as a video. The display is, for example, a liquid crystal display (LCD) or an organic EL (electro luminescence) display. The display unit 405 displays and outputs data obtained by the operation of the communication terminal 4. The display unit 405 may be connected to the communication terminal 4 as an external output device instead of being provided in the communication terminal 4. As a connection method, any method such as USB, HDMI (registered trademark), or Bluetooth (registered trademark) can be used.

[0052] The functions of the communication terminal 4 are realized by executing a program according to this embodiment on a processor corresponding to the communication terminal 4. That is, the functions of the communication terminal 4 are realized by software. The program causes a computer to execute the operations of the communication terminal 4, thereby causing the computer to function as the communication terminal 4. That is, the computer functions as the communication terminal 4 by executing the operations of the communication terminal 4 in accordance with the program.

[0053] 9 illustrates an example of a hardware configuration of an abnormality identification device 6 according to an embodiment of the present disclosure. As illustrated in FIG. 9, the abnormality identification device 6 includes a control unit 601, a storage unit 602, and a communication unit 603.

[0054] The control unit 601 includes at least one processor, at least one dedicated circuit, or a combination of these. The processor is a general-purpose processor such as a CPU or a GPU, or a dedicated processor specialized for a specific process. The dedicated circuit is, for example, an FPGA or an ASIC. The control unit 601 executes processes related to the operation of the abnormality identification device 6 while controlling each part of the abnormality identification device 6.

[0055] The storage unit 602 includes at least one semiconductor memory, at least one magnetic memory, at least one optical memory, or a combination of at least two of these. The semiconductor memory is, for example, a RAM or a ROM. The RAM is, for example, an SRAM or a DRAM. The ROM is, for example, an EEPROM. The storage unit 602 functions as, for example, a main storage device, an auxiliary storage device, or a cache memory. The storage unit 602 stores data used in the operation of the anomaly identification device 6 and data obtained by the operation of the anomaly identification device 6.

[0056] The communication unit 603 includes at least one wireless communication interface. Specifically, the wireless communication interface includes an interface corresponding to communication via the above-mentioned second dedicated network. The communication unit 603 receives data used in the operation of the anomaly identification device 6, and transmits data obtained by the operation of the anomaly identification device 6.

[0057] The functions of anomaly identification device 6 are realized by executing a program according to this embodiment on a processor equivalent to anomaly identification device 6. That is, the functions of anomaly identification device 6 are realized by software. The program causes a computer to execute the operations of anomaly identification device 6, thereby causing the computer to function as anomaly identification device 6. That is, the computer functions as anomaly identification device 6 by executing the operations of anomaly identification device 6 in accordance with the program.

[0058] In this embodiment, the program can be recorded on a computer-readable recording medium. The computer-readable recording medium includes a non-transitory computer-readable medium, such as a magnetic recording device, an optical disc, a magneto-optical recording medium, or a semiconductor memory. The program is distributed, for example, by selling, transferring, or lending a portable recording medium such as a DVD (digital versatile disc) or a CD-ROM (compact disc read only memory) on which the program is recorded. The program may also be distributed by storing the program in the storage of an external server and transmitting the program from the external server to another computer. The program may also be provided as a program product.

[0059] Although the present disclosure has been described based on the drawings and embodiments, it should be noted that those skilled in the art can easily make various modifications and corrections based on the present disclosure. Therefore, it should be noted that these modifications and corrections are included in the scope of the present disclosure. For example, the functions included in each means or step can be rearranged so as not to be logically inconsistent, and multiple means or steps can be combined or divided into one. [Explanation of symbols]

[0060] 100 Anomaly Detection System 1. Unmanned Camera 2. Communication terminals 3. Photographed images 4. Communication terminals 5. Photographed images 6 Abnormality identification device 7, 7a, 7b Arrows 8 RT 81 Burner 9 Rectangular Frame Object 10 Display Objects 11 Abnormality judgment display 12, 13 Text Objects 14 Rectangular Frame 101, 201, 401, 601 Control unit 102, 202, 402, 602 Storage section 103, 203, 403, 603 Communications Department 104 Positioning unit 105 Imaging unit 204, 404 input section 405 Display section

Claims

1. An anomaly detection system including an unmanned photography device that captures an image of an inspection target facility, a communication terminal that acquires the captured image from the unmanned photography device, and an anomaly identification device that acquires the captured image from the communication terminal and identifies an anomaly of the inspection target from the captured image, An anomaly detection system characterized in that the unmanned photography device and the communication terminal are capable of wireless communication via a first dedicated network, the communication terminal and the anomaly identification device are capable of wireless communication via a second dedicated network, and the communication distance of the second dedicated network is longer than the communication distance of the first dedicated network.

2. The anomaly identification device is capable of identifying an anomaly in a radiant tube using an anomaly identification program capable of identifying an anomaly including a crack in a radiant tube that heats a steel plate, The anomaly detection system described in claim 1, characterized in that the anomaly identification program is machine-learned based on first teacher data related to cracks in the radiant tube, and the first teacher data includes images that have been processed for the purpose of increasing information on the width, length, and clarity of the crack based on photographed images of the crack.

3. The anomaly identification device is capable of identifying an anomaly in the radiant tube by using an anomaly identification program capable of identifying an anomaly including a bend in the radiant tube that heats the steel plate, The anomaly detection system described in claim 1 or claim 2, characterized in that the anomaly identification program is machine-learned based on second teacher data related to the bend of the radiant tube, and the second teacher data includes images of the same bend taken at multiple angles.

4. when the abnormality identifying device identifies an abnormality in the radiant tube, the abnormality identifying device transmits result information indicating the existence of the identified abnormality to the communication terminal; The anomaly detection system according to claim 3 , wherein the result information includes a risk indication indicating a risk of the anomaly.

5. The anomaly detection system described in claim 4, characterized in that the anomaly identification device acquires operating conditions related to the radiant tubes, predicts the occurrence and degree of risk of an anomaly in each radiant tube, sets a priority for each radiant tube based on the predicted degree of risk, and the result information includes a priority indication indicating the priority.

Citation Information

Patent Citations

  • Roll surface detection, identification and analysis method based on neural network

    CN115375605A

  • Portable electronic apparatus, flying body, and telephone system

    JP2019029776A

  • Inspection system, inspection method, server device, and program

    JP2019052954A

  • Camber amount measurement method of steel plate, camber amount measurement device of steel plate and calibration method of camber amount measurement device of steel plate

    JP2020052060A

  • Unmanned aerial vehicle control device, and storage medium

    WO2022107761A1