Casting blank cutting control method, device and equipment of continuous casting machine

Abstract

The invention discloses a casting blank cutting control method, device and equipment for a continuous casting machine, and the method comprises the steps: obtaining a first field image in the process that a cutting gun of the continuous casting machine cuts a casting blank; based on the first field image, cutting information is determined, and the cutting information comprises a first result used for representing whether the casting blank is successfully cut or not; and based on the cutting information, the cutting process of the casting blank is controlled. The technical problem that the continuous casting machine is low in production efficiency is solved.

Description

Technical Field

[0001] The present invention belongs to the technical field of steel smelting, and particularly relates to a method, device and equipment for controlling the cutting of continuous casting billets of a continuous caster. Background Art

[0002] In the process of steel production, the cutting of continuous casting billets of a continuous caster is a key link. Traditional cutting state monitoring mainly relies on manual visual inspection. Operators need to observe the cutting state in real time, with high labor intensity. Moreover, affected by human factors, there are risks of missed inspection and misjudgment. In addition, during the cutting process, due to unstable air pressure, insufficient gas combustion and other reasons, it is more likely for operators to misjudge that the billet has been cut when it has not been cut off.

[0003] The billet not being cut off means that the billet forms a continuous long strip shape on the roller table, which occupies a large physical space on the roller table. When these uncut billets move on the roller table, if their length or width exceeds the designed accommodation range of the roller table, it may cause friction, collision or even jamming between the billet and the roller table or other equipment, and there may be a need for shutdown inspection or maintenance, thus resulting in low production efficiency of the continuous caster. Summary of the Invention

[0004] Embodiments of the present invention provide a method, device and equipment for controlling the cutting of continuous casting billets of a continuous caster, which solve the technical problem of low production efficiency of the continuous caster.

[0005] In a first aspect, embodiments of the present invention provide a method for controlling the cutting of continuous casting billets of a continuous caster, including: during the process of the cutting gun of the continuous caster cutting the billet, acquiring a first on-site image; based on the first on-site image, determining cutting information, where the cutting information includes a first result for characterizing whether the billet is cut successfully; and controlling the cutting process of the billet based on the cutting information.

[0006] Combined with the first aspect of the present invention, in some embodiments, the determining the cutting information based on the first on-site image includes: preprocessing the first on-site image to obtain a second on-site image, where the preprocessing includes grayscale processing; and determining the cutting information based on the second on-site image.

[0007] Combined with the first aspect of the present invention, in some embodiments, the determining the cutting information based on the second on-site image includes: on the second on-site image, determining a first region of interest, where the first region of interest includes the region corresponding to the cutting section of the billet; if the cutting section exists in the first region of interest, determining that the first result is that the cutting is successful; if the cutting section does not exist in the first region of interest, determining that the first result is that the cutting is not successful.

[0008] In combination with the first aspect of the present invention, in some embodiments, it further includes: obtaining a first grayscale threshold; based on the first grayscale threshold, performing binarization processing on the first region of interest to obtain a second region of interest; obtaining the length and width of the white sub-region of the second region of interest; based on the length and width of the white sub-region, determining the aspect ratio of the white sub-region; if the deviation between the aspect ratio of the white sub-region and a preset ratio is less than a preset deviation threshold, it is determined that the cutting section exists in the first region of interest, otherwise, it is determined that the cutting section does not exist in the first region of interest.

[0009] In combination with the first aspect of the present invention, in some embodiments, the obtaining of the first grayscale threshold includes: obtaining the steel type of the continuous casting billet, the ambient temperature and ambient light intensity of the environment where the continuous casting billet is located; based on the steel type of the continuous casting billet, determining the specific heat capacity of the continuous casting billet; correcting a preset second grayscale threshold based on the specific heat capacity, the ambient temperature, and the ambient light intensity to obtain the first grayscale threshold; wherein, the smaller the specific heat capacity, the larger the grayscale threshold; the higher the ambient temperature, the larger the grayscale threshold; the stronger the ambient light intensity, the larger the grayscale threshold.

[0010] In combination with the first aspect of the present invention, in some embodiments, the determining of the cutting information based on the second on-site image includes: inputting the second on-site image into a preset target detection model to obtain a second result and a third result, the second result being used to represent whether there is slag splashing during the cutting process, the third result being used to represent whether there is a fire risk during the cutting process, and the cutting information including the second result and the third result.

[0011] In combination with the first aspect of the present invention, in some embodiments, the target detection model is established through the following steps: obtaining training data, the training data including a plurality of slag splashing training images and a plurality of fire training images, each of the plurality of slag splashing training images having a slag splashing identifier, and each of the plurality of fire training images having a fire identifier; training the YOLO model based on the training data to obtain the target detection model.

[0012] In combination with the first aspect of the present invention, in some embodiments, the determining of the cutting information based on the second on-site image includes: based on the second on-site image, determining the current position of the continuous casting billet through contour detection; based on the current position of the continuous casting billet, determining the current offset of the continuous casting billet, and the cutting information including the current offset of the continuous casting billet.

[0013] Second aspect, an embodiment of the present invention provides a billet cutting control device for a continuous casting machine, including: an image acquisition unit configured to acquire a first on-site image during the process of a cutting torch of the continuous casting machine cutting a billet; an information determination unit configured to determine cutting information based on the first on-site image, where the cutting information includes a first result for characterizing whether the billet is successfully cut; and a cutting control unit configured to control the cutting process of the billet based on the cutting information.

[0014] Third aspect, an embodiment of the present invention provides an electronic device, including: a memory, a processor, and a computer program stored on the memory and executable on the processor, where when the processor executes the computer program, the method described in any item of the first aspect is implemented.

[0015] One or more technical solutions provided by the embodiments of the present invention at least achieve the following technical effects or advantages:

[0016] In the embodiment of the present invention, during the process of a cutting torch of a continuous casting machine cutting a billet, a first on-site image is acquired; based on the first on-site image, cutting information is determined, where the cutting information includes a first result for characterizing whether the billet is successfully cut; and based on the cutting information, the cutting process of the billet is controlled. By determining the cutting information based on the first on-site image, it can be accurately known whether the billet is successfully cut through the first result in the cutting information, avoiding possible misjudgments by operators, and thus avoiding moving the billet when it is not cut off, further avoiding friction, collision, or even jamming between the billet and the roller table or other equipment, and also avoiding shutdown inspection or maintenance. Therefore, the production efficiency of the continuous casting machine is improved. Description of the Drawings

[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for description in the embodiments. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0018] Figure 1 It is a flowchart of the billet cutting control method for a continuous casting machine in an embodiment of the present invention;

[0019] Figure 2 It is a schematic diagram of a second on-site image when the cutting is not successful in an embodiment of the present invention;

[0020] Figure 3 It is a schematic diagram of a second on-site image when the cutting is successful in an embodiment of the present invention;

[0021] Figure 4 It is a functional module diagram of the billet cutting control device for a continuous casting machine in an embodiment of the present invention;

[0022] Figure 5 This is a schematic structural diagram of an electronic device in an embodiment of the present invention. Detailed implementation manners

[0023] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without making creative efforts shall fall within the protection scope of the present invention.

[0024] In the present invention, descriptions such as "first" and "second" are only for descriptive purposes, and cannot be understood as indicating or implying their relative importance or implicitly indicating the quantity of the indicated technical features. Thus, features defined with "first" and "second" may explicitly or implicitly include at least one of such features. In addition, the technical solutions between various embodiments may be combined with each other, but it must be based on the fact that those of ordinary skill in the art can implement them. When the combination of technical solutions appears to be contradictory or unable to be implemented, it should be considered that such a combination of technical solutions does not exist and is not within the protection scope required by the present invention.

[0025] An embodiment of the present invention provides a method for controlling the cutting of a continuous casting billet. Refer to Figure 1 As shown, the method includes the following steps S101 to S103:

[0026] S101: During the process of the cutting gun of the continuous casting machine cutting the billet, obtain a first on-site image.

[0027] It should be noted that the first on-site image may include the area where the cutting gun is located, the area where the billet is located, and the area corresponding to the cutting section of the billet.

[0028] In some implementation manners, before obtaining the first on-site image, it may further include: in response to the cutting completion signal of the cutting gun, wait for a preset duration.

[0029] It should be noted that the cutting completion signal of the cutting gun does not represent the successful cutting of the billet, but means that the current cutting duration reaches the preset cutting duration, and at this time the cutting gun returns to the initial position. Therefore, at the moment corresponding to the cutting completion signal, the billet may or may not be successfully cut. If the cutting is successful, the roller table will separate the billet at this time. Therefore, obtaining the first on-site image after waiting for the preset duration ensures that the cutting section can be photographed in the case of successful cutting, thereby improving the accuracy of judging whether the billet is successfully cut.

[0030] S102: Based on the first on-site image, determine the cutting information, where the cutting information includes a first result for characterizing whether the slab cutting is successful.

[0031] In some embodiments, step S102 may include: preprocessing the first on-site image to obtain a second on-site image, where the preprocessing includes grayscale processing; and determining the cutting information based on the second on-site image.

[0032] It should be noted that the preprocessing may further include denoising and enhancement. Through denoising, enhancement, and grayscale processing, it is possible to adapt to different lighting and temperature conditions and improve the image clarity.

[0033] In some embodiments, determining the cutting information based on the second on-site image may include: on the second on-site image, determine a first region of interest, where the first region of interest includes the region corresponding to the cutting section of the slab; if there is a cutting section in the first region of interest, determine that the first result is successful cutting; if there is no cutting section in the first region of interest, determine that the first result is unsuccessful cutting.

[0034] Reference Figure 2 and Figure 3 as shown, Figure 2 is a schematic diagram of the second on-site image with unsuccessful cutting in the embodiment of the present invention, Figure 3 is a schematic diagram of the second on-site image with successful cutting in the embodiment of the present invention. The first region of interest includes the region corresponding to the cutting section of the slab, which means that in the case of successful cutting and the slab being separated, a cutting section can be detected in the first region of interest, and in the case of unsuccessful cutting, a cutting section cannot be detected in the first region of interest.

[0035] It should be noted that in machine vision and image processing, outlining the area to be processed in the form of a square, circle, ellipse, irregular polygon, etc. from the processed image is called the region of interest (ROI, region of interest). The first region of interest can be adjusted in real time according to the position of the cutting torch and set along the lower edge of the flame of the cutting torch to avoid false detection caused by the flame invading the detection area. In addition, by determining the first region of interest on the second on-site image and then judging whether there is a cutting section within the first region of interest, the influence of the rest of the second on-site image on the judgment of the cutting section is avoided, thereby improving the accuracy of judging whether the slab cutting is successful.

[0036] In some embodiments, the method for controlling the cutting of a continuous casting slab may further include: obtaining a first gray-scale threshold; based on the first gray-scale threshold, performing binarization processing on a first region of interest to obtain a second region of interest; obtaining the length and width of the white sub-region of the second region of interest; based on the length and width of the white sub-region, determining the aspect ratio of the white sub-region; if the deviation between the aspect ratio of the white sub-region and a preset ratio is less than a preset deviation threshold, it is determined that there is a cutting section in the first region of interest; otherwise, it is determined that there is no cutting section in the first region of interest.

[0037] It should be noted that the method for determining whether there is a cutting section in the first region of interest may also be based only on the length or width of the white sub-region, but there is a problem of low judgment accuracy. Taking the length as an example, different placement positions of the camera may cause significant differences between the first on-site image and the second on-site image, which in turn may cause changes in the length of the white sub-region of the second region of interest, making it impossible to determine whether there is a cutting section in the first region of interest based on the comparison between the length of the white sub-region and a preset length. In addition, in the case of successful cutting, the white sub-region of the second region of interest is the cutting section, but in the case of unsuccessful cutting, the white sub-region of the second region of interest may be the flame of the cutting torch. If the length corresponding to the flame region of the cutting torch happens to be the same as the length of the cutting section, it will also lead to misjudgment. To solve the above problems, considering that the aspect ratio of the cutting section does not change, the embodiments of the present invention define that whether there is a cutting section in the first region of interest is determined based on the aspect ratio of the white sub-region, achieving accurate determination of whether there is a cutting section in the first region of interest, and further improving the accuracy of determining whether the continuous casting slab is successfully cut.

[0038] In some embodiments, obtaining the first gray-scale threshold may include: obtaining the steel type of the continuous casting slab, the ambient temperature and the ambient light intensity of the environment where the continuous casting slab is located; based on the steel type of the continuous casting slab, determining the specific heat capacity of the continuous casting slab; correcting a preset second gray-scale threshold based on the specific heat capacity, the ambient temperature and the ambient light intensity to obtain the first gray-scale threshold; wherein, the smaller the specific heat capacity, the larger the gray-scale threshold; the larger the ambient temperature, the larger the gray-scale threshold; the larger the ambient light intensity, the larger the gray-scale threshold.

[0039] It should be noted that the first region of interest can also be directly binarized by a preset second grayscale threshold to obtain a second region of interest. However, there is a problem of low accuracy in judging whether the slab is cut successfully. The following is an explanation: The smaller the specific heat capacity of the slab, the greater the temperature rise of the slab after absorbing the same amount of heat. Therefore, the brighter the outer wall of the slab in the second on-site image, that is, the greater the grayscale value of the outer wall of the slab in the second on-site image; the higher the ambient temperature, the higher the temperature of the outer wall of the slab. Therefore, the brighter the outer wall of the slab in the second on-site image, that is, the greater the grayscale value of the outer wall of the slab in the second on-site image; the greater the ambient light intensity, the brighter the outer wall of the slab in the second on-site image, that is, the greater the grayscale value of the outer wall of the slab in the second on-site image. From the above, it can be seen that if the first region of interest is directly binarized by a fixed grayscale threshold, then different specific heat capacities, ambient temperatures, and ambient light intensities will all result in different grayscale values for each pixel point, and thus different second regions of interest will be obtained. That is, the white sub-region of the second region of interest may not only include the cutting section, but may also include part of the outer wall of the slab. This will lead to an incorrect judgment of whether there is a cutting section in the first region of interest, and thus a low accuracy in judging whether the slab is cut successfully. To solve the above problems, the embodiments of the present invention correct the fixed grayscale threshold in combination with the specific heat capacity, ambient temperature, and ambient light intensity to adapt to the actual situation, achieve accurate distinction between the cutting section and the outer wall of the slab, avoid the situation where the white sub-region of the second region of interest includes the cutting section and the outer wall of the slab, and thus improve the accuracy of judging whether there is a cutting section in the first region of interest, and further improve the accuracy of judging whether the slab is cut successfully.

[0040] In some embodiments, based on the second on-site image, determining the cutting information may include: inputting the second on-site image into a preset target detection model to obtain a second result and a third result. The second result is used to characterize whether there is slag splashing during the cutting process, and the third result is used to characterize whether there is a fire risk during the cutting process. The cutting information includes the second result and the third result.

[0041] In some embodiments, the target detection model is established through the following steps: obtaining training data, where the training data includes a plurality of slag splashing training images and a plurality of fire training images. Each image in the plurality of slag splashing training images has a slag splashing identifier, and each image in the plurality of fire training images has a fire identifier; training the YOLO (You Only Look Once) model based on the training data to obtain the target detection model.

[0042] It should be noted that the YOLO model is a set of deep learning models for object detection. This model is known for its speed and efficiency and has received extensive attention and applications in the field of computer vision. The YOLO model uses a convolutional neural network for object detection. It applies a single neural network to the entire image, then divides the image into regions, and predicts bounding boxes and probabilities for each region. These bounding boxes are weighted by the predicted probabilities. Compared with traditional object detection methods, the YOLO model does not require complex processing procedures such as region proposals and feature extraction, so it can achieve real-time object detection.

[0043] It should be noted that the slag splashing and fire can be detected through the object detection model. When slag splashing or fire is detected, an alarm message is output in a timely manner to prompt relevant personnel to take safety measures, avoiding the occurrence of safety accidents, improving the safety of billet cutting, and thus enhancing the production safety of the continuous casting machine.

[0044] In some embodiments, determining the cutting information based on the second on-site image may include: determining the current position of the billet through contour detection based on the second on-site image; determining the current offset of the billet based on the current position of the billet, and the cutting information includes the current offset of the billet.

[0045] It should be noted that if the current offset of the billet is too large, it may cause the billet to collide with surrounding equipment, resulting in downtime for maintenance and reducing the production efficiency of the continuous casting machine. Therefore, in the embodiments of the present invention, the current offset of the billet is monitored in real time. If the current offset of the billet is greater than the preset offset threshold, a third alarm message is output to prompt relevant personnel to take measures to avoid the occurrence of safety accidents, improving the production safety of the continuous casting machine, avoiding equipment collisions, and thus avoiding equipment downtime for maintenance and improving the production efficiency of the continuous casting machine.

[0046] S103: Control the cutting process of the billet based on the cutting information.

[0047] In some embodiments, step S103 may include: if the first result is successful cutting, control the cutting gun to stop cutting the billet; if the first result is unsuccessful cutting, control the cutting gun to cut the billet; if the second result is the existence of slag splashing, output the first alarm message; if the third result is the existence of a fire risk, output the second alarm message; if the current offset of the billet is greater than the preset offset threshold, output the third alarm message.

[0048] It should be noted that the existing camera monitoring system can only provide on-site images, but lacks intelligent analysis capabilities, cannot automatically identify the cutting state, and is even less able to participate in automatic control. In the monitoring of the continuous casting billet cutting state of the existing technical solutions, manual monitoring is mainly adopted. The cutting site images are transmitted to the monitoring room through cameras, and the operators observe in real time and manually judge the cutting state. The problems existing in the existing technology are as follows: relying on manual labor, with a large labor intensity. Manual monitoring requires operators to concentrate their attention for a long time, which is easy to cause fatigue and is affected by subjective factors. The detection accuracy is low. Simple image processing algorithms cannot adapt to the changes in different steel types, temperatures, and lighting conditions, and the false detection rate is high. It cannot participate in automatic control. The existing system lacks the linkage ability with the production control system and cannot achieve automatic control. There is a lack of multi-function integration. The existing technology usually only targets a single detection object and fails to make full use of image resources to achieve multi-object detection. To solve the above problems, the embodiments of the present invention use advanced image processing and recognition algorithms to detect the cutting state of the billet in real time and accurately, and are linked with control systems such as PLC to achieve automatic control. At the same time, the system also has multiple functions such as detecting cutting slag splashing, fire warning, and billet position deviation, making full use of image resources to provide comprehensive protection for production and safety. Improve the accuracy and real-time performance of billet cutting state detection; achieve the linkage between the system and production control equipment and automatically control the cutting process; multi-function integration, detect cutting slag splashing, fire warning, and billet position deviation; reduce the labor intensity of manual monitoring and reduce the influence of human factors.

[0049] It should be noted that the composition of the system for implementing the embodiments of the present invention is as follows. The first part is the image acquisition unit. For camera installation, high-definition cameras can be installed above the cutting area of each strand at the caster walking bridge, covering the cutting area. The cameras have high frame rates and high resolutions to meet the real-time detection requirements. For camera parameters, the cameras have high dynamic range (HDR) and low-light performance, and can obtain clear images in high-temperature, high-brightness, and smoky environments. The second part is the image processing unit. For image preprocessing, the acquired images can be denoised, enhanced, and grayscaled to adapt to different lighting and temperature conditions. For dynamic ROI setting, the region of interest (ROI) can be adjusted in real time according to the position of the cutting torch. The ROI is set along the lower edge of the cutting flame to avoid false detection caused by the flame invading the detection area. The third part is the target detection module. For the contour detection algorithm, an adaptive threshold contour detection algorithm is adopted for the bright features of the billet cutting section, and the threshold is dynamically adjusted to adapt to the changes of different steel grades and temperatures, accurately identifying the cutting section. For the application of the YOLO model, a trained YOLO model can be introduced to detect the slag splashing and early fire characteristics during the cutting process, realizing multi-target recognition. The fourth part is the state recognition module. For cutting state determination, the start and end state information of cutting can be obtained by reading the PLC signals of the cutting carriage. For automatic and manual mode recognition, the running mode of the cutting carriage can be detected, and the system detection is only enabled in the automatic mode to avoid false judgment during manual operation. The fifth part is the control linkage module. For automatically controlling the cutting process, when it is detected that the cutting is not successful, the system automatically sends an instruction to the PLC to control the cutting carriage to continue cutting until the cutting is successful, ensuring production continuity. For the delay judgment mechanism, after the cutting end signal of the cutting carriage, a set time (such as 4 seconds) is delayed, and whether the bright surface of the section appears is detected to increase the reliability of the judgment. The sixth part is the alarm and early warning module. For slag splashing detection and alarm, the YOLO model is used to detect the slag splashing phenomenon during the cutting process, and an alarm is issued in a timely manner to prompt the operator to adjust the cutting parameters to prevent equipment damage. For fire early warning, the system has a fire detection model, which can detect abnormalities in the early stage of a fire and issue an early warning to avoid fire accidents. For billet position offset detection, the position and width of the billet are monitored in real time through contour detection. If an offset is found, the system immediately issues an alarm to remind for maintenance. The seventh part is the data storage and analysis module. For data storage, the detection results, alarm information, and relevant image data are stored in the database, classified by time and event type, for subsequent query and analysis. For data analysis and reporting, data statistics and analysis functions are provided to generate reports, providing a basis for production optimization and safety management. The eighth part is the system self-check and maintenance. For the self-check function, the system regularly self-checks each module and notifies the maintenance personnel in time when abnormalities are found. For remote update and upgrade, remote software update is supported to ensure the continuous improvement of system functions and the optimization of algorithms.

[0050] It should be noted that regarding image acquisition and preprocessing, the camera captures the images of the cutting area in real time and performs preprocessing to improve the image quality. Dynamic ROI setting: According to the position of the cutting torch, the ROI is adjusted in real time to ensure the accuracy of the detection area and avoid flame interference. Contour detection and adaptive threshold: The contour detection algorithm with adaptive threshold can be used to identify the bright surface features of the billet cutting section. Status determination and interlock control: 1. Read the PLC signal to obtain the cutting status and operation mode of the cutting vehicle. 2. When the cutting end signal of the cutting vehicle appears, the system delays to detect the bright surface of the section. If the bright surface is detected: It is judged that the cutting is successful, and the system records and enters the next process. If the bright surface is not detected: It is judged that the cutting is unsuccessful, and the system sends an instruction to control the cutting vehicle to continue cutting until the bright surface is detected. Regarding slag splashing and fire detection, 1. Slag splashing detection: Use the YOLO model to identify the slag splashing features and issue a prompt or alarm. 2. Fire early warning: Detect the visual features in the initial stage of the fire and give an early warning in time. Billet position and width detection: Through contour analysis, the position and size of the billet are monitored in real time, and an alarm is given immediately when an abnormality is found. Data storage and analysis: Store all detection results and related data, and provide reporting and analysis functions. System self-check and maintenance: The system performs self-check regularly and supports remote maintenance and upgrade. The embodiment of the present invention has strong adaptability and is easy to promote: The system has the functions of automatic threshold adjustment and automatic ROI adjustment, can adapt to different light conditions and steel grade changes, and the detection algorithm has strong robustness. This enables the system to operate stably in different production environments and is easy to be promoted and applied in other production lines or factories. Reduce equipment costs: Since the system has relatively low requirements for the camera, ordinary industrial cameras can meet the needs, reducing the investment in high-end equipment and lowering the construction and maintenance costs of the overall system. Improve detection accuracy: Adopt the contour detection algorithm with adaptive threshold and the YOLO object detection model, which can adapt to different steel grades, temperatures and environmental conditions, and significantly improve the detection accuracy and stability. Achieve automatic control: The system is interlocked with control devices such as PLC, can automatically judge the cutting status, and control the operation of the cutting vehicle, realizing the automatic control of the cutting process and reducing human intervention. Multifunctional integration: The system can not only detect the billet cutting status, but also has multiple functions such as slag splashing detection, fire early warning and billet position offset detection, with multiple functions in one machine, making full use of the image resources. Reduce labor intensity: Reduce the need for manual real-time monitoring, and reduce the labor intensity of operators and the risks brought by human factors. Improve safety: By detecting abnormalities such as slag splashing and fire in the early stage and giving an early warning in time, the safety of production equipment and personnel is guaranteed.

[0051] In an embodiment of the present invention, during the process of cutting a billet by a cutting torch of a continuous casting machine, a first on-site image is acquired; based on the first on-site image, cutting information is determined, and the cutting information includes a first result for characterizing whether the billet is successfully cut; based on the cutting information, the cutting process of the billet is controlled. By determining the cutting information based on the first on-site image, it can be accurately known whether the billet is successfully cut through the first result in the cutting information, avoiding possible misjudgments by operators, thus avoiding moving the billet when it is not cut off, and further avoiding friction, collision or jamming between the billet and the roller table or other equipment, and also avoiding shutdown inspection or maintenance. Therefore, the production efficiency of the continuous casting machine is improved.

[0052] Based on the same inventive concept, referring to Figure 4 As shown, an embodiment of the present invention provides a billet cutting control device 10 for a continuous casting machine, including: an image acquisition unit 110, configured to acquire a first on-site image during the process of cutting a billet by a cutting torch of the continuous casting machine; an information determination unit 120, configured to determine cutting information based on the first on-site image, and the cutting information includes a first result for characterizing whether the billet is successfully cut; a cutting control unit 130, configured to control the cutting process of the billet based on the cutting information.

[0053] It can be understood that the information determination unit 120 includes: a preprocessing subunit, configured to preprocess the first on-site image to obtain a second on-site image, and the preprocessing includes grayscale processing; a determination subunit, configured to determine the cutting information based on the second on-site image.

[0054] It can be understood that the determination subunit is specifically configured to: on the second on-site image, determine a first region of interest, and the first region of interest includes the region corresponding to the cutting section of the billet; if a cutting section exists in the first region of interest, determine that the first result is a successful cut; if no cutting section exists in the first region of interest, determine that the first result is an unsuccessful cut.

[0055] It can be understood that the billet cutting control device 10 of the continuous casting machine further includes: a determination unit, specifically used for: obtaining a first gray threshold; based on the first gray threshold, performing binarization processing on a first region of interest to obtain a second region of interest; obtaining the length and width of the white sub-region of the second region of interest; based on the length and width of the white sub-region, determining the aspect ratio of the white sub-region; if the deviation between the aspect ratio of the white sub-region and a preset ratio is less than a preset deviation threshold, it is determined that there is a cutting section in the first region of interest, otherwise, it is determined that there is no cutting section in the first region of interest. Among them, obtaining the first gray threshold includes: obtaining the steel type of the billet, the ambient temperature and ambient light intensity of the environment where the billet is located; based on the steel type of the billet, determining the specific heat capacity of the billet; correcting a preset second gray threshold based on the specific heat capacity, ambient temperature, and ambient light intensity to obtain the first gray threshold; among them, the smaller the specific heat capacity, the larger the gray threshold; the higher the ambient temperature, the larger the gray threshold; the stronger the ambient light intensity, the larger the gray threshold.

[0056] It can be understood that the determination subunit is specifically used for: inputting the second on-site image into a preset target detection model to obtain a second result and a third result, the second result is used to characterize whether there is slag splashing during the cutting process, the third result is used to characterize whether there is a fire risk during the cutting process, and the cutting information includes the second result and the third result.

[0057] It can be understood that the billet cutting control device 10 of the continuous casting machine further includes: a model establishment unit, used to establish a target detection model, including the following steps: obtaining training data, the training data includes a plurality of slag splashing training images and a plurality of fire training images, each image in the plurality of slag splashing training images has a slag splashing identifier, and each image in the plurality of fire training images has a fire identifier; training the YOLO model based on the training data to obtain the target detection model.

[0058] It can be understood that the determination subunit is specifically used for: based on the second on-site image, determining the current position of the billet through contour detection; based on the current position of the billet, determining the current offset of the billet, and the cutting information includes the current offset of the billet.

[0059] It should be understood that more implementation details of the billet cutting control device 10 of the continuous casting machine in the embodiments of the present invention are referred to the aforementioned billet cutting control method of the continuous casting machine. For the sake of simplicity of the specification, they will not be elaborated here.

[0060] Based on the same inventive concept, the embodiments of the present invention further provide an electronic device, such as Figure 5As shown, it includes a memory 504, a processor 502, and a computer program stored on the memory 504 and executable on the processor 502. The processor 502 executes the program to implement the steps described in any implementation manner of the billet cutting control method of the continuous casting machine.

[0061] Among them, in Figure 5 it, the bus architecture (represented by bus 500), the bus 500 may include any number of interconnected buses and bridges. The bus 500 links various circuits including one or more processors represented by the processor 502 and the memory represented by the memory 504 together. The bus 500 may also link various other circuits such as peripheral devices, voltage regulators, and power management circuits, etc., which are well known in the art, so they will not be further described herein. The bus interface 505 provides an interface between the bus 500 and the receiver 501 and the transmitter 503. The receiver 501 and the transmitter 503 may be the same element, that is, a transceiver, which provides a unit for communicating with various other devices on the transmission medium. The processor 502 is responsible for managing the bus 500 and general processing, while the memory 504 may be used to store data used by the processor 502 when performing operations.

[0062] The functions described herein may be implemented in hardware, software executed by a processor, firmware, or any combination thereof. If implemented in software executed by a processor, the functions may be stored on a computer-readable medium or transmitted via a computer-readable medium as one or more instructions or codes. Other examples and implementations are within the scope and spirit of the present invention and the appended claims. For example, due to the nature of software, the functions described above may be implemented using software executed by a processor, hardware, firmware, hardwiring, or any combination of these. In addition, each functional unit may be integrated in a processing unit, or each unit may exist physically alone, or two or more units may be integrated in one unit.

[0063] In several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are merely illustrative. For example, the division of the units may be a logical function division, and there may be other division methods in actual implementation. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Another point, the displayed or discussed coupling or direct coupling or communication connection to each other may be through some interfaces, and the indirect coupling or communication connection of units or modules may be in an electrical or other form.

[0064] The unit described as a separating component may or may not be physically separated. The components as control devices may or may not be physical units, that is, they may be located in one place or distributed to multiple units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0065] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The foregoing storage medium includes: various media such as USB flash drives, read-only memories (ROMs), random access memories (RAMs), mobile hard disks, magnetic disks or optical discs that can store program codes.

[0066] The above are only the embodiments of the present invention and are not intended to limit the present invention. For those skilled in the art, the present invention can have various changes and modifications. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included within the scope of the claims of the present invention.

Claims

1. A continuous casting machine casting slab cutting control method, characterized in that: include: Acquiring a first on-site image during the process of the continuous casting machine's cutting gun cutting the ingot; Based on the first on-site image, determining cutting information, the cutting information including a first result for indicating whether the casting slab is cut successfully; Based on the cutting information, the cutting process of the ingot is controlled.

2. The continuous casting machine casting slab cutting control method according to claim 1, characterized in that: The determining cutting information based on the first on-site image includes: Preprocessing the first scene image to obtain a second scene image, wherein the preprocessing includes grayscale processing; Based on the second on-site image, the cutting information is determined.

3. The continuous casting machine casting slab cutting control method according to claim 2, characterized in that: The determining the cutting information based on the second on-site image includes: Determine a first region of interest on the second on-site image, wherein the first region of interest includes a region corresponding to a cut section of the ingot; If the cutting section exists in the first region of interest, determining the first result is that the cutting is successful; If the cutting section does not exist in the first region of interest, it is determined that the first result is that the cutting is unsuccessful.

4. The continuous casting machine casting slab cutting control method according to claim 3, characterized in that: Also includes: Get a first grayscale threshold; Based on the first grayscale threshold, binarizing the first region of interest to obtain a second region of interest; Obtain the length and width of the white sub-region of the second region of interest; Determining the aspect ratio of the white sub-region based on the length and width of the white sub-region; If the deviation between the aspect ratio of the white sub-region and the preset ratio is less than a preset deviation threshold, it is determined that the cutting section exists in the first region of interest; otherwise, it is determined that the cutting section does not exist in the first region of interest.

5. The slab cutting control method of the continuous casting machine according to claim 4, characterized in that: The obtaining of the first grayscale threshold comprises: Obtaining the steel type of the ingot, the ambient temperature and ambient light intensity of the ingot's environment; Determining the specific heat capacity of the ingot based on the steel type of the ingot; The preset second grayscale threshold is corrected based on the specific heat capacity, the ambient temperature and the ambient light intensity to obtain the first grayscale threshold; wherein, the smaller the specific heat capacity, the larger the grayscale threshold; the higher the ambient temperature, the larger the grayscale threshold; the greater the ambient light intensity, the larger the grayscale threshold.

6. The continuous casting machine casting slab cutting control method according to any one of claims 2 to 5, characterized in that: The determining the cutting information based on the second on-site image includes: The second scene image is input into a preset target detection model to obtain a second result and a third result, wherein the second result is used to characterize whether there is slag splashing during the cutting process, and the third result is used to characterize whether there is a fire risk during the cutting process. The cutting information includes the second result and the third result.

7. The continuous casting machine casting slab cutting control method according to claim 6, characterized in that: The target detection model is established through the following steps: Acquire training data, wherein the training data includes a plurality of slag splash training images and a plurality of fire training images, each of the plurality of slag splash training images having a slag splash mark, and each of the plurality of fire training images having a fire mark; The YOLO model is trained based on the training data to obtain the target detection model.

8. The slab cutting control method for a continuous casting machine according to any one of claims 2 to 5, characterized in that: The determining the cutting information based on the second on-site image includes: Based on the second on-site image, determining the current position of the casting billet by contour detection; Based on the current position of the casting strand, a current offset of the casting strand is determined, and the cutting information includes the current offset of the casting strand.

9. A casting slab cutting control device for a continuous casting machine, characterized in that: include: An image acquisition unit, used for acquiring a first on-site image during the process of the continuous casting machine's cutting gun cutting the ingot; an information determining unit, configured to determine cutting information based on the first on-site image, wherein the cutting information includes a first result for indicating whether the casting billet is cut successfully; A cutting control unit is used to control the cutting process of the ingot based on the cutting information.

10. An electronic device, characterized in that: include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the method according to any one of claims 1 to 8 when executing the computer program.

Images