Converter iron mixing safety identification and positioning detection device and method

Abstract

The invention relates to a converter iron mixing safety identification and positioning detection device and method. The device comprises an RFID tag, a reader-writer, an industrial camera, a ladle, a controller, a crown block, a light supplementing lamp and an alarm. The method comprises the steps that a high-definition industrial camera collects a converter mouth area image in real time, and the conditions of molten iron ladle and converter area personnel are recognized through a deep learning model; the crown block is positioned through a reader-writer and an RFID tag which are arranged on the crown block and a crown block walking board; the central processing unit fuses vision and RFID data, judges an identification area, converter tapping angles, crown block positions and ladle identification, when a person is in the identification area, alarm information is sent out, the active early warning problem that the person enters a dangerous area by mistake during iron exchange is solved, and whole-process safety monitoring of iron exchange operation is achieved.

Description

TECHNICAL FIELD

[0001] The patent application belongs to the technical field of metallurgical safety production, and more particularly relates to a converter iron mixing safety identification and positioning detection device and method. BACKGROUND

[0002] During the converter iron mixing process, there are high risks of high-temperature molten iron splashing and equipment collision. The traditional manual monitoring has blind spots, the existing infrared detection technology is easily disturbed by high temperature, and the laser positioning has insufficient precision in a smoky environment. Therefore, there is an urgent need for an active protection system and method which can identify personnel in a dangerous area in real time and accurately position. SUMMARY

[0003] The technical problem to be solved by the present application is to provide a converter iron mixing safety identification and positioning detection device and method, which solves the active early warning problem of personnel mistakenly entering a dangerous area during iron mixing and realizes full-process safety monitoring of iron mixing operation.

[0004] To solve the above problems, the technical scheme adopted by the present application is as follows: A converter iron mixing safety identification and positioning detection method, including a converter iron mixing safety identification and positioning detection device corresponding to a crown block, the crown block is provided with a ladle in the middle part, and the device comprises: An image acquisition unit is installed on a converter operation platform of the crown block, is used for aiming at an iron mixing operation area, and identifies the personnel situation in the ladle and converter area through a deep learning model (the deep learning model is a trained model, the deep model is understood as a use unit, the input is an image, and the output is whether there is a person in the image area); A crown block positioning unit is arranged on the crown block and a crown block walkway to realize positioning of the crown block; A safety alarm unit receives alarm information and executes an alarm action (issues an alarm sound); A central processing unit is connected with the image acquisition unit, the crown block positioning unit and the safety alarm unit, judges the identification area, the converter tapping angle and the position of the crown block, and identifies the ladle, and when there is personnel in the identification area, issues alarm information.

[0005] Further, the image acquisition unit comprises an industrial camera and a fill-in light, the industrial camera and the fill-in light are both arranged towards the ladle and the converter area, and the industrial camera and the fill-in light are both connected with the central processing unit.

[0006] Further, the crown block positioning unit comprises an RFID acquisition unit and an RFID read-write unit, the RFID acquisition unit comprises an RFID tag arranged at the end of the crown block and a reader-writer arranged on the crown block walkway, and the RFID tag cooperates with the reader-writer. The RFID read-write unit is used to read the tag book in the read-write unit and send the tag data to the central processing unit. That is, the RFID read-write unit is part of the crane positioning unit, and the RFID read-write unit is used to read the RFID tag book and send the RFID tag data.

[0007] Further, the safety alarm unit is an alarm, and the alarm is connected with the central processing unit.

[0008] Further, the central processing unit includes a control processing unit, and a target recognition unit connected with the control processing unit, and the control processing unit is a central processing unit; The target recognition unit includes an input module and an output module, the input module is connected with the image acquisition unit, and the output module is connected with the control processing unit; The crane positioning unit sends the position information of the crane to the target recognition unit, and the target recognition unit judges according to the position information of the crane, and if there is no crane in the position of the converter, it is not recognized; Further, the control processing unit includes a ladle recognition model, a crane RFID positioning algorithm unit, and a crane area safety algorithm unit, The ladle recognition model is used to complete the recognition of whether there is a ladle in front of the converter, which can identify whether there is a ladle, and it uses an improved YOLOv5 architecture, and the training data set contains 10000+ labeled converter iron mixing scene images; The crane RFID positioning algorithm unit realizes three-dimensional positioning of the crane through the TDOA (time difference of arrival) algorithm, and the positioning accuracy is ≤10 cm; The crane area safety algorithm unit defines the identification area in the window, and issues an alarm if there is a person in the identification area.

[0009] Further, the ladle recognition model is an AI recognition model.

[0010] A converter iron mixing safety recognition and positioning detection method uses the above device, including the following steps: Step 1: Install the image acquisition unit composed of an industrial camera and a fill light on the converter operation platform, align the iron mixing operation area, and identify the ladle and the personnel in the converter area through a deep learning model; Step 2: Arrange the crane positioning unit read-write unit and RFID tag to realize crane positioning; Step 3: The central processor fuses visual and RFID data to judge the identification area, the converter tapping angle, the crane position, and the ladle recognition, and issues an alarm when there is a person in the identification area. Step 4: The safety alarm unit alarm is based on the interlocking relationship between the converter tapping angle, the crane position, and the ladle identification model. Since the prerequisite for the converter to charge iron is that the tapping angle and the position of the converter are both above the crane, the ladle identification model only identifies and sends signals when the tapping angle (the angle of charging iron) and the crane position are above the converter position. When the conditions are met, an alarm sound is emitted.

[0011] Furthermore, the identification area is implemented by the crane area safety algorithm unit. The crane area safety algorithm unit delineates the identification sub-area within the window. If there are personnel in the identification sub-area, an alarm message is issued.

[0012] Due to the adoption of the above technical solution, the beneficial effects achieved by this invention are: 1. Multimodal perception fusion: Improves overall positioning accuracy in high-temperature and dusty environments; 2. Proactive safety protection: The entire process of "monitoring-alarm" is automated through a tiered early warning mechanism; 3. Strong compatibility: The modular design can be adapted to different converter furnace types and overhead crane systems. Attached Figure Description

[0013] Figure 1 This is a schematic diagram of the system of the present invention.

[0014] Figure 2 This is a connection block diagram of the present invention.

[0015] Figure 3 This is a flowchart of the present invention.

[0016] The components include: 1. RFID tag; 2. Reader; 3. Industrial camera; 4. Ladle; 5. Controller; 6. Overhead crane; 7. Supplemental lighting; 8. Alarm. Detailed Implementation

[0017] The present invention will be further described in detail below with reference to the embodiments.

[0018] A converter molten iron charging safety identification and positioning detection device, corresponding to the overhead crane 6, wherein a molten iron ladle 4 is provided in the middle of the overhead crane 6, comprising: The image acquisition unit, installed on the converter operation platform of the overhead crane 6, is used to align with the molten iron charging operation area and identify the personnel situation in the molten iron ladle 4 and the converter area through a deep learning model (the deep learning model is pre-trained; the deep model can be understood as a user unit, the input is an image, and the output is whether there are people in the image area). The crane positioning unit is arranged on crane 6 and crane platform to realize the positioning of crane 6; The safety alarm unit receives alarm information and executes alarm actions (issues an alarm sound). The central processing unit is connected with the image acquisition unit, the overhead traveling crane positioning unit and the safety alarm unit, judges and identifies the recognition area, the converter tapping angle and the position of the overhead traveling crane 6, and performs molten iron ladle 4 recognition, and sends an alarm information when there is a person in the recognition area.

[0019] The image acquisition unit comprises an industrial camera 3 and a fill-in light 7, both of which are arranged towards the molten iron ladle 4 and the converter area, and are connected with the central processing unit.

[0020] The overhead traveling crane positioning unit comprises an RFID acquisition unit and an RFID read-write unit, the RFID acquisition unit comprises an RFID tag 1 arranged at the end of the overhead traveling crane 6 and a read-write device 2 arranged on the overhead traveling crane runway, and the RFID tag 1 cooperates with the read-write device 2. The RFID read-write unit is used for reading the tag book in the read-write device 2 and sending the tag data to the central processing unit, that is, the RFID read-write unit is part of the overhead traveling crane positioning unit, and is used for reading the RFID tag data.

[0021] The safety alarm unit is an alarm 8 connected with the central processing unit.

[0022] The central processing unit comprises a control processing unit and a target identification unit connected with the control processing unit, and the control processing unit is a central processing unit 5. The target identification unit comprises an input module and an output module, the input module is connected with the image acquisition unit, and the output module is connected with the control processing unit. The overhead traveling crane positioning unit sends the position information of the overhead traveling crane to the target identification unit, and the target identification unit judges according to the position information of the overhead traveling crane, and if there is no overhead traveling crane at the position of the converter, it is not identified. The control processing unit comprises a molten iron ladle recognition model, an overhead traveling crane RFID positioning algorithm unit and an overhead traveling crane area safety algorithm unit. The molten iron ladle recognition model is used for completing the recognition of whether there is a molten iron ladle in front of the converter, can identify whether there is a molten iron ladle, adopts an improved YOLOv5 architecture, and the training data set contains 10000+ labeled converter iron mixing scene image; The overhead traveling crane RFID positioning algorithm unit realizes three-dimensional positioning of the overhead traveling crane through a TDOA (time difference of arrival) algorithm, and the positioning accuracy is ≤10 cm. The overhead traveling crane area safety algorithm unit draws a recognition area in a window, and sends an alarm if there is a person in the recognition area.

[0023] The molten iron ladle recognition model is an AI recognition model.

[0024] Here, combined with Figure 2 , Figure 3 : Iron ladle identification model: Completes the identification of whether there is an iron ladle in front of the converter furnace.

[0025] The overhead crane RFID positioning algorithm unit: A positioning card / reader 2 (the positioning card is a symbol, which can be a number such as "1" or "100") is installed on the overhead crane platform of the converter. An RFID identification module / RFID tag 1 is installed on the overhead crane. For example, when the overhead crane travels to the iron-collecting position of converter No. 1 (for example, the symbol "1" representing converter No. 1), the ground server receives the symbol "1", which can prove that the overhead crane is collecting iron in converter No. 1.

[0026] The overhead crane area safety algorithm unit identifies whether personnel are present before the converter iron is added, under the premise of overhead crane iron addition.

[0027] Control and processing unit: mainly the server, which is responsible for identifying steel ladles, identifying whether personnel are present in the safe area, and determining the position of the overhead crane.

[0028] Target recognition unit: mainly includes two aspects: molten iron ladle recognition model and crane area safety algorithm.

[0029] RFID reader / writer unit: mainly includes tags, reader / writer, and a control processing unit for sending tag data.

[0030] At the same time, this invention also discloses a method for safety identification and positioning detection of converter iron charging, such as... Figure 1 Follow these steps: Step 1: The high-definition industrial camera 3 and the supplementary light 7 form an image acquisition unit, which is installed on the converter operation platform and aimed at the molten iron charging area. The deep learning model is used to identify the molten iron ladle 4 and the personnel in the converter area. Step 2: The crane is positioned by placing RFID tags 1 and readers 2 on the crane and the crane walkway respectively. Step 3: The central processing unit 5 integrates visual and RFID data to determine the identification area, converter tapping angle, crane position, and molten iron ladle identification. When personnel are in the identification area, the alarm 8 sends an alarm message.

[0031] Step Four: The safety alarm unit (i.e., alarm 8) establishes an interlocking relationship between the converter tapping angle, the overhead crane position, and the ladle identification model. Since the prerequisite for molten iron charging in the converter is the presence of an overhead crane at both the tapping angle and the converter position, the ladle identification model only identifies and sends signals when the tapping angle (iron charging angle) and the overhead crane position are within the converter's location. An alarm sound is emitted when these conditions are met. The identification area is implemented by the overhead crane area safety algorithm unit. This unit delineates subdivided identification areas within a window. If personnel are present within the identified subdivided area, an alarm is issued.

[0032] In the present embodiment, reference is made to the accompanying drawings Figure 1 , the device includes a high-definition industrial camera module, an anti-metal RFID positioning module, a central processing module, and a pre-warning execution module. The high-definition camera collects real-time images of the converter mouth area, and identifies the condition of the ladle and the personnel in the converter area through a deep learning model. The RFID positioning device realizes the positioning of the crane through the reader and tag arranged on the crane and the crane track. The central processor fuses visual and RFID data to judge and identify the recognition area, the converter tapping angle, the crane position, and the ladle recognition, and sends an alarm information when there is personnel in the recognition area. The present application solves the problem of active pre-warning of personnel mistakenly entering the dangerous area during iron making, and realizes the whole-process safety monitoring of iron making operation.

Claims

1. A converter molten iron charging safety identification and positioning detection device, corresponding to a crane (6), wherein a molten iron ladle (4) is provided in the middle of the crane (6), characterized in that include: The image acquisition unit is installed on the converter operation platform of the overhead crane (6) and is used to align with the iron-pouring operation area. It identifies the molten iron ladle (4) and the personnel situation in the converter area through a deep learning model. The crane positioning unit is arranged on the crane (6) and the crane platform to realize the positioning of the crane (6); The safety alarm unit receives alarm information and executes alarm actions (issues an alarm sound). The central processing unit is connected to the image acquisition unit, the overhead crane positioning unit, and the safety alarm unit. It determines the identification area, the converter tapping angle, and the position of the overhead crane (6), and identifies the molten iron ladle (4). When personnel are in the identification area, it sends out alarm information.

2. The converter iron charging safety identification and positioning detection device according to claim 1, characterized in that: The image acquisition unit includes an industrial camera (3) and a fill light (7). The industrial camera (3) and the fill light (7) are both positioned facing the molten iron ladle (4) and the converter area. The industrial camera (3) and the fill light (7) are both connected to the central processing unit.

3. The converter iron charging safety identification and positioning detection device according to claim 2, characterized in that: The crane positioning unit includes an RFID acquisition unit and an RFID reading and writing unit. The RFID acquisition unit includes an RFID tag (1) arranged at the end of the crane (6) and a reader (2) arranged on the crane platform. The RFID tag (1) and the reader (2) work together. The RFID reading and writing unit is used to read the tag book in the reader (2) and send the tag data to the central processing unit.

4. The converter iron charging safety identification and positioning detection device according to claim 3, characterized in that: The safety alarm unit is an alarm (8), which is connected to the central processing unit.

5. The converter iron charging safety identification and positioning detection device according to claim 4, characterized in that: The central processing unit includes a control processing unit and a target recognition unit connected to the control processing unit. The control processing unit is a central processing unit (5). The target recognition unit includes an input module and an output module. The input module is connected to the image acquisition unit, and the output module is connected to the control processing unit. The crane positioning unit sends the crane's position information to the target identification unit. The target identification unit makes a judgment based on the crane's position information. If there is no crane at the converter location, it will not be identified.

6. The converter iron charging safety identification and positioning detection device according to claim 5, characterized in that: The control and processing unit includes a molten iron ladle identification model, an overhead crane RFID positioning algorithm unit, and an overhead crane area safety algorithm unit. The molten iron ladle recognition model is used to identify whether there is a molten iron ladle in front of the converter furnace. It can identify whether there is a molten iron ladle. It adopts an improved YOLOv5 architecture, and the training dataset contains 10,000+ labeled converter iron-pouring scene images. The overhead crane RFID positioning algorithm unit achieves three-dimensional positioning of the overhead crane through the TDOA (Time Difference of Arrival) algorithm, with a positioning accuracy of ≤10cm; The overhead crane area safety algorithm unit defines an identification area within a window and issues an alarm if personnel are present in that area.

7. The converter iron charging safety identification and positioning detection device according to claim 6, characterized in that: The molten iron ladle recognition model is an AI recognition model.

8. A method for safety identification and positioning detection of converter iron charging, utilizing the device described in any one of claims 1-7, characterized in that... Includes the following steps: Step 1: Install the image acquisition unit on the converter operation platform, align it with the molten iron charging area, and use a deep learning model to identify the molten iron ladle (4) and the personnel situation in the converter area; Step 2: Deploy the crane positioning unit to achieve crane positioning; Step 3: The central processing unit (5) integrates visual and RFID data to determine the identification area, converter tapping angle, crane position, and molten iron ladle identification. When personnel are in the identification area, an alarm message is sent. Step 4: The safety alarm unit will sound an alarm when the conditions are met based on the interlocking relationship between the converter tapping angle, the crane position, and the molten iron ladle identification model.

9. A method for safety identification and positioning detection of converter iron charging according to claim 8, characterized in that: The identification area is implemented by the crane area safety algorithm unit. The crane area safety algorithm unit delineates the identification sub-area within the window. If there are personnel in the identification sub-area, an alarm message is issued.

Images