LF furnace dust removal valve adjusting method and device, medium and electronic equipment

Abstract

The application provides a LF furnace dust removal valve adjusting method, device, medium and electronic equipment, wherein the method comprises the following steps: acquiring a smoke dust image in a LF furnace; detecting the smoke dust size, smoke dust color and smoke dust diffusion direction of the smoke dust image through a pre-trained smoke dust detection model, and generating a smoke dust detection result; and controlling the dust removal valve to open to a preset opening degree according to the smoke dust detection result, so as to complete dust removal in the LF furnace. The application solves the problem that the observation of the smoke dust condition in the LF furnace is not timely, the dust removal valve adjustment lags behind, a large amount of smoke dust escapes, or the opening degree of the dust removal valve is too large, the suction force is too large, air enters the cover and a secondary oxidation accident of molten steel occurs. The scheme provided by the application realizes real-time monitoring of the smoke dust condition in the LF furnace based on image recognition, detects the smoke dust condition in the LF furnace through a smoke dust detection model, controls the dust removal valve to open to a preset opening degree according to the detection result, improves the efficiency of molten steel smelting, and ensures the quality of molten steel.

Description

Technical Field

[0001] This application relates to the field of steelmaking technology, and in particular to a method, device, medium and electronic equipment for regulating the dust removal valve of an LF furnace. Background Technology

[0002] The existing method of adjusting the opening of the dust collector valve in the LF furnace relies on the main control operator visually observing the smoke level and then inputting the corresponding value. This method has drawbacks such as untimely observation and delayed adjustment, which may lead to large plumes of smoke, adversely affecting environmental protection work, or cause excessive suction due to excessive opening of the dust collector valve, resulting in air being drawn into the ladle cover and causing secondary oxidation of the molten steel.

[0003] Therefore, how to adopt an effective method to monitor the dust situation inside the LF furnace in real time and automatically adjust the opening of the dust removal valve according to the size of the dust to prevent a large amount of dust from escaping or secondary oxidation of molten steel is an urgent technical problem to be solved. Summary of the Invention

[0004] The purpose of this application is to provide a method, device, medium, and electronic equipment for adjusting the dust removal valve of an LF furnace. This application solves the problems of untimely observation of the dust situation inside the LF furnace, delayed adjustment of the dust removal valve causing a large amount of dust to escape, or excessive suction due to excessive opening of the dust removal valve, which allows air to enter the ladle cover and cause secondary oxidation of molten steel. The solution proposed in this application is based on image recognition to monitor the dust situation inside the LF furnace in real time. The dust detection model detects the dust situation inside the LF furnace, and controls the dust removal valve to open to a preset degree according to the detection results, thereby improving the efficiency of steel smelting and ensuring the quality of molten steel.

[0005] Specifically, this application adopts the following technical solution:

[0006] According to one aspect of the embodiments of this application, a method for adjusting the dust removal valve of an LF furnace is provided. The method includes: acquiring a dust image inside the LF furnace; detecting the size, color, and diffusion direction of the dust in the dust image using a pre-trained dust detection model, and generating a dust detection result; and controlling the dust removal valve to open to a preset opening degree according to the dust detection result, so as to complete the dust removal inside the LF furnace.

[0007] In some embodiments of this application, based on the aforementioned scheme, before detecting the size, color, and diffusion direction of the smoke and dust in the smoke and dust image using a pre-trained smoke and dust detection model, the method further includes: acquiring a target smoke and dust image inside the LF furnace and processing the target smoke and dust image; constructing an initial smoke and dust detection model and training the initial smoke and dust detection model based on the processed target smoke and dust image.

[0008] In some embodiments of this application, based on the foregoing scheme, the processing of the target smoke image includes: performing data augmentation processing on the target smoke image using a mapping tool to generate a smoke image dataset.

[0009] In some embodiments of this application, the mask-rcnn model is used when constructing the initial smoke and dust detection model based on the aforementioned scheme.

[0010] In some embodiments of this application, based on the foregoing scheme, the step of training the initial smoke detection model based on the processed target smoke image includes: color-coding the processed target smoke image; and inputting the coded target smoke image into the initial smoke detection model for deep learning training.

[0011] In some embodiments of this application, based on the aforementioned scheme, the step of detecting the smoke size, smoke color, and smoke diffusion direction of the smoke image using a pre-trained smoke detection model includes: inputting the smoke image into the smoke detection model, using an edge detection algorithm to identify the smoke outline, and calculating the smoke area based on the smoke outline; judging the smoke color of the smoke image using the smoke detection model, and identifying the change in the center position of the smoke outline using a CV algorithm to confirm the smoke diffusion direction.

[0012] In some embodiments of this application, based on the foregoing scheme, the step of identifying the change in the center position of the smoke and dust contour according to the CV algorithm includes: tracking the center position of the smoke and dust contour in the smoke and dust image for consecutive frames using the CV algorithm; and obtaining the smoke and dust diffusion direction in the smoke and dust image based on the movement trajectory of the center position.

[0013] According to one aspect of the embodiments of this application, an LF furnace dust removal valve adjustment device is provided. The device includes: an acquisition unit, used to acquire smoke and dust images inside the LF furnace; a detection unit, used to detect the smoke and dust size, smoke and dust color, and smoke and dust diffusion direction in the smoke and dust images using a pre-trained smoke and dust detection model, and generate smoke and dust detection results; and a control unit, used to control the dust removal valve to open to a preset opening degree according to the smoke and dust detection results, so as to complete the dust removal inside the LF furnace.

[0014] According to one aspect of the present application, a computer-readable storage medium is provided, wherein at least one piece of program code is stored in the computer-readable storage medium, the at least one piece of program code being loaded and executed by a processor to implement the operations performed by the LF furnace dust removal valve adjustment method as described above.

[0015] According to one aspect of the present application, an electronic device is provided, including a memory and a processor, the memory storing a computer program, the processor executing the computer program to perform the operations performed by the LF furnace dust removal valve adjustment method as described above.

[0016] As can be seen from the above technical solution, this application has at least the following advantages and positive effects:

[0017] The proposed solution addresses the problems of delayed human observation of dust conditions inside the LF furnace, resulting in excessive dust escape due to lagging dust removal valve adjustment, or excessive suction caused by excessively large dust removal valve opening, leading to secondary oxidation of molten steel by air entering the ladle cover. The proposed solution uses image recognition to monitor dust conditions inside the LF furnace in real time, detects dust conditions through a dust detection model, and controls the dust removal valve to open to a preset degree based on the detection results, thereby improving the efficiency of steel smelting and ensuring the quality of molten steel. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0019] Figure 1 A flowchart of an LF furnace dust removal valve adjustment method according to one embodiment of this application is shown;

[0020] Figure 2 A structural block diagram of an LF furnace dust removal valve regulating device according to one embodiment of this application is shown;

[0021] Figure 3 A schematic diagram of the structure of a computer system suitable for implementing the electronic device of the present application is shown. Detailed Implementation

[0022] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided to make this application more comprehensive and complete, and to fully convey the concept of the exemplary embodiments to those skilled in the art.

[0023] Furthermore, the described features, structures, or characteristics can be combined in any suitable manner in one or more embodiments. Numerous specific details are provided in the following description to give a thorough understanding of embodiments of this application. However, those skilled in the art will recognize that the technical solutions of this application can be practiced without one or more of the specific details, or other methods, components, apparatuses, steps, etc., can be employed. In other instances, well-known methods, apparatuses, implementations, or operations are not shown or described in detail to avoid obscuring various aspects of this application.

[0024] The flowcharts shown in the accompanying drawings are merely illustrative and do not necessarily include all content and operations / steps, nor do they necessarily have to be performed in the described order. For example, some operations / steps can be broken down, while others can be combined or partially combined; therefore, the actual execution order may change depending on the specific circumstances.

[0025] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such uses of these terms can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described.

[0026] The implementation details of the technical solutions in the embodiments of this application are described in detail below:

[0027] Reference Figure 1 , Figure 1 This is a flowchart of the LF furnace dust removal valve adjustment method in one embodiment of this application.

[0028] According to a typical embodiment of this application, a method for adjusting the dust removal valve of an LF furnace is provided, the method comprising the following steps S1 to S3:

[0029] Step S1: Obtain an image of the smoke and dust inside the LF furnace.

[0030] Step S2: The smoke image is detected by a pre-trained smoke detection model to determine the size, color, and direction of smoke diffusion, and a smoke detection result is generated.

[0031] Step S3: Based on the dust detection results, control the dust removal valve to open to a preset opening degree to complete the dust removal in the LF furnace.

[0032] In this application, image recognition is used to monitor the emission of flue gas in real time, and the opening degree of the dust removal valve of the LF furnace is automatically controlled according to the flue gas overflow. High-temperature resistant cameras installed near the LF furnace can capture images of the flue gas inside the furnace, enabling real-time monitoring of the flue gas situation within the LF furnace.

[0033] After acquiring the smoke and dust image inside the LF furnace, a pre-trained smoke and dust detection model can be used to detect smoke and dust in the image. After inputting the smoke and dust image into the smoke and dust detection model, the size, color, and direction of smoke and dust diffusion in the image can be detected, and a smoke and dust detection result can be generated. The smoke and dust detection result can be the specific area of ​​the smoke and dust region in the image, the specific color of the smoke and dust, the specific direction of smoke and dust diffusion, etc.

[0034] After obtaining the specific area, color, and direction of smoke diffusion in the smoke image, the dust removal valve can be automatically adjusted based on the detection results using a dust removal valve adjustment module (which can be the PLC controller of the dust removal valve). This allows for timely dust removal of the smoke, effectively preventing large-scale smoke emissions or secondary oxidation of molten steel caused by excessive suction entering the ladle cover. This ensures the quality of the molten steel and the smooth progress of steel smelting, reduces the risk of manual intervention, and improves the automation level of environmental control.

[0035] In one embodiment of this application, before detecting the smoke size, color, and diffusion direction of the smoke image using a pre-trained smoke detection model, the method further includes:

[0036] Acquire target dust images inside the LF furnace and process the target dust images.

[0037] An initial smoke detection model is constructed, and the initial smoke detection model is trained based on the processed target smoke image.

[0038] In this application, the acquired smoke and dust images are detected by a smoke and dust detection model. After detection, the size, color, and diffusion direction of the smoke and dust in the images can be obtained. It is evident that the accuracy of the smoke and dust detection model is very important. In order to improve the performance of the smoke and dust detection model, increase its accuracy, and reduce false alarms and false negatives, the model needs to be trained after its establishment.

[0039] Training a model requires a large training dataset; the more training data, the better the model's performance. To train the model, images for training are first needed. During daily production, a large number of smoke and dust images can be acquired using cameras installed near the LF furnace. These images are then cleaned and processed to obtain target smoke and dust images. To ensure a sufficient number of target smoke and dust images for processing, the model's stability and detection accuracy are improved. After preparing the training dataset (the processed target smoke and dust images), an initial smoke and dust detection model is constructed. This initial model is then trained based on the processed target smoke and dust images to obtain a smoke and dust detection model with absolute stability and high detection accuracy.

[0040] In one embodiment of this application, the processing of the target smoke image includes: performing data augmentation processing on the target smoke image using a mapping tool to generate a smoke image dataset.

[0041] In this application, when training the model using the target smoke and dust images, if the number of processed target smoke and dust images is insufficient, the Stable Diffusion AI mapping tool can be used to perform data augmentation on the target smoke and dust images to obtain a large number of newly generated effective smoke and dust image data, and generate a smoke and dust image dataset (the smoke and dust image dataset is a set of target smoke and dust images used for model training), thereby increasing the number of training datasets, making the trained model perform better, and improving the stability of the model and the accuracy of the detection results.

[0042] In one embodiment of this application, the initial smoke detection model can be constructed using the mask-rcnn model.

[0043] In order to ensure the stability of the model and the accuracy of smoke detection, some important training parameters need to be set when training the mask-rcnn model. The following are the specific parameter settings for model training.

[0044] Learning rate: This is an important parameter in optimization algorithms, determining the speed at which model parameters are updated. A large learning rate may lead to unstable model training, while a small learning rate may make the training process too slow. An initial learning rate of 0.0005 can be set.

[0045] Batch Size: This is the number of samples the model processes in each forward and backward propagation iteration. A large batch size may lead to memory shortages, while a small batch size may slow down the training process. A batch size of 4 can be set.

[0046] Number of iterations (Epochs): This is the number of times the model performs forward and backward propagation on the training set. A smaller number of iterations may lead to underfitting, while a larger number of iterations may lead to overfitting. The number of iterations can be set to 100.

[0047] Weight Decay: This is a regularization method used to prevent model overfitting. It is achieved by adding a term to the loss function that is proportional to the square of the model parameters. Weight Decay can be set to 0.0005.

[0048] Momentum: This is an optimization method used to accelerate model convergence. It achieves this by accumulating the direction of previous gradients. The momentum value can be set to 0.9.

[0049] Anchor scales and ratios: These are parameters used to generate prior bounding boxes. You can set the scale value to 16 and the ratio value to 0.5.

[0050] In this application, it is important to note that the model training parameters described above can be adjusted based on the specific task and dataset. When adjusting the parameters, cross-validation is typically used to evaluate the performance of different parameter combinations and select the best-performing combination.

[0051] In one embodiment of this application, training the initial smoke detection model based on the processed target smoke image includes:

[0052] The processed target smoke image is then labeled with smoke color.

[0053] The labeled target smoke and dust image is input into the initial smoke and dust detection model for deep learning training.

[0054] In this application, when training the initial smoke detection model based on the processed target smoke image, in order for the initial smoke detection model to correctly identify the smoke color, the processed target smoke image can be analyzed to determine the smoke color and label it, such as labeling the target smoke image as yellow smoke, black smoke, etc. These labeled target smoke images are then input into the initial smoke detection model for deep learning training. The training objective is to enable the model to correctly identify the smoke color and smoke diffusion direction based on the input smoke image.

[0055] During the training process for smoke and dust color recognition, the model learns the correlation between smoke and dust colors and image pixels (different smoke and dust colors correspond to different image pixels), thereby achieving color recognition and enabling the smoke and dust detection model to more accurately identify smoke and dust. By adding color recognition to the smoke and dust detection model, the model can more accurately determine the intensity and direction of smoke and dust diffusion, thus more accurately controlling the opening of the dust removal valve.

[0056] In one embodiment of this application, the step of detecting the smoke size, color, and diffusion direction of the smoke image using a pre-trained smoke detection model includes:

[0057] The smoke and dust image is input into the smoke and dust detection model, an edge detection algorithm is used to identify the smoke and dust outline, and the area of ​​the smoke and dust region is calculated based on the smoke and dust outline.

[0058] The smoke and dust detection model is used to determine the color of the smoke and dust in the image, and the CV algorithm is used to identify the changes in the center position of the smoke and dust outline to confirm the direction of smoke and dust diffusion.

[0059] In this application, when detecting smoke and dust using the smoke and dust detection model, image processing techniques, such as edge detection algorithms (e.g., Sobel algorithm, Canny algorithm), can be used to find the smoke and dust contours from the smoke and dust image, thus obtaining the boundary of the smoke and dust area. After identifying the smoke and dust contours, the area of ​​the smoke and dust area can be calculated based on the contours. The area of ​​the smoke and dust area reflects the size of the smoke and dust inside the LF furnace. When the calculated smoke and dust area exceeds the standard, the opening of the dust removal valve needs to be increased (the opening can be adjusted from 0% to 100% based on the size of the smoke and dust). For example, if the smoke and dust detection model detects that the size of the smoke and dust inside the LF furnace is a large area of ​​smoke and dust, and there is a possibility of escape, it may be necessary to open the dust removal valve to 90% to promote rapid smoke and dust discharge. If the opening of the dust removal valve is less than 9% at this time, the dust removal valve will be automatically adjusted to avoid a large amount of smoke and dust escaping and polluting the environment.

[0060] The dust detection model enables real-time monitoring of dust conditions within the LF furnace. When the dust area is smaller than the standard dust area, the opening of the dust removal valve is automatically reduced to prevent excessive suction caused by an overly large valve opening, which could lead to secondary oxidation of the molten steel due to air entering the ladle cover. For example, if the dust detection model detects a small area of ​​dust within the LF furnace, the dust removal valve only needs to be opened to 15%. If the opening is greater than or less than 15%, the valve will automatically adjust, ensuring the quality of the molten steel and the smooth progress of steelmaking.

[0061] After identifying the smoke and dust outline, the centroid of the outline, i.e., the geometric center of the smoke and dust region (referred to as the center position), can be calculated. The centroid can be calculated by finding the average position of all pixels within the smoke and dust outline. After obtaining the center position (centroid), the changes in the center position of the smoke and dust outline can be identified using the CV algorithm to determine the diffusion direction and diffusion speed (the speed at which the smoke and dust move) of the smoke and dust in the image.

[0062] In this application, the purpose of detecting the color and direction of smoke diffusion is as follows: If the detected smoke is yellow, it generally moves relatively quickly. If the detected smoke is yellow and its direction of movement is first diffusion and then convergence, the dust collector valve opening does not need adjustment, indicating that the current opening is sufficient to draw away the smoke and prevent further diffusion or escape. If the detected smoke converges very quickly, the dust collector valve opening will be appropriately reduced (a fast convergence speed indicates excessive suction). If the detected smoke is white (pale white) and covers a small area, white smoke has little impact on the environment and process, so the dust collector valve opening can be appropriately reduced to decrease suction. If the detected smoke diffuses upwards at a very fast speed, the dust collector valve opening needs to be increased to increase suction.

[0063] To better control the opening of the dust collector valve based on factors such as smoke size, color, and diffusion direction, a Region of Interest (ROI) can be set above the LF furnace lid. This ROI is a sensitive area; if smoke reaches it, the dust collector may fail to draw it back. A condition is added to the ROI: if the smoke entering the ROI is pale white and moves slowly (low speed indicates a small amount of smoke), the dust collector valve opening remains unchanged. If white smoke enters the ROI and moves quickly, the opening needs to be increased appropriately. If yellow smoke enters the area, the opening is immediately increased regardless of speed. The overall control logic is: white smoke is not very harmful, but a large quantity is, requiring adjustment of the dust collector valve opening based on actual conditions; yellow smoke must be prevented—this is the purpose of color detection; fast smoke movement indicates intense smoke emission and a large amount of smoke—this is the purpose of speed detection.

[0064] In one embodiment of this application, identifying the change in the center position of the smoke and dust contour using a CV algorithm includes:

[0065] The center position of the smoke and dust contour in consecutive frames of the smoke and dust image is tracked using the CV algorithm.

[0066] The direction of smoke diffusion in the smoke image is obtained based on the movement trajectory of the center position.

[0067] In this application, when identifying the changes in the center position of the smoke and dust contour according to the CV algorithm, the CV algorithm can be used to track the center position of the smoke and dust contour in the smoke and dust images of consecutive frames. By comparing the movement trajectory of the centroid position (center position) in two or more consecutive frames, the direction of smoke and dust diffusion and the speed of smoke and dust diffusion can be determined.

[0068] The specific implementation of this application will be further illustrated by specific embodiments below, but the specific implementation of this application is not limited to the following embodiments.

[0069] Figure 2 This is a structural block diagram of an LF furnace dust removal valve regulating device according to an embodiment of this application.

[0070] Reference Figure 2 As shown, an LF furnace dust removal valve regulating device 200 according to an embodiment of this application includes: an acquisition unit 201, a detection unit 202, and a control unit 203.

[0071] The acquisition unit 201 is used to acquire images of smoke and dust inside the LF furnace.

[0072] The detection unit 202 is used to detect the size, color, and direction of smoke diffusion in the smoke image using a pre-trained smoke detection model, and to generate smoke detection results.

[0073] The control unit 203 is used to control the dust removal valve to open to a preset degree according to the dust detection results, so as to complete the dust removal in the LF furnace.

[0074] Reference Figure 3 , Figure 3 A schematic diagram of the structure of a computer system suitable for implementing the electronic device of the present application is shown.

[0075] like Figure 3 As shown, the computer system 300 includes a Central Processing Unit (CPU) 301, which can perform various appropriate actions and processes based on programs stored in Read-Only Memory (ROM) 302 or programs loaded from storage portion 308 into Random Access Memory (RAM) 303, such as performing the methods described in the above embodiments. The RAM 303 also stores various programs and data required for system operation. The CPU 1101, ROM 302, and RAM 303 are interconnected via a bus 304. An Input / Output (I / O) interface 305 is also connected to the bus 304.

[0076] The following components are connected to I / O interface 305: an input section 306 including a keyboard, mouse, etc.; an output section 307 including a cathode ray tube (CRT), liquid crystal display (LCD), etc., and speakers, etc.; a storage section 308 including a hard disk, etc.; and a communication section 309 including a network interface card such as a LAN (Local Area Network) card, modem, etc. The communication section 309 performs communication processing via a network such as the Internet. A drive 310 is also connected to I / O interface 305 as needed. Removable media 311, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., are installed on drive 310 as needed so that computer programs read from them can be installed into storage section 308 as needed.

[0077] Specifically, according to embodiments of this application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of this application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication section 309, and / or installed from removable medium 311. When the computer program is executed by central processing unit (CPU) 301, it performs various functions defined in the system of this application.

[0078] It should be noted that the computer-readable medium shown in the embodiments of this application can be a computer-readable signal medium or a computer-readable storage medium, or any combination of the two. A computer-readable storage medium can be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), flash memory, optical fiber, portable compact disc read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this application, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In this application, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such transmitted data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. The computer-readable signal medium can also be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to wireless, wired, etc., or any suitable combination thereof.

[0079] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. Each block in a flowchart or block diagram may represent a module, segment, or portion of code, which contains one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram or flowchart, and combinations of blocks in a block diagram or flowchart, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0080] The units described in the embodiments of this application can be implemented in software or hardware, and the described units can also be located in a processor. The names of these units do not necessarily limit the specific unit itself.

[0081] According to a typical embodiment of this application, this application also proposes a computer-readable storage medium storing at least one piece of program code, which is loaded and executed by a processor to implement the operations performed by the LF furnace dust removal valve adjustment method as described above.

[0082] According to a typical embodiment of this application, this application also proposes an electronic device, which includes a memory and a processor, wherein the memory stores a computer program, characterized in that the processor executes the computer program to implement the operations performed by the LF furnace dust removal valve adjustment method as described above.

[0083] It should be noted that although several modules or units for the device used to perform actions have been mentioned in the detailed description above, this division is not mandatory. In fact, according to the embodiments of this application, the features and functions of two or more modules or units described above can be embodied in one module or unit. Conversely, the features and functions of one module or unit described above can be further divided and embodied by multiple modules or units.

[0084] As can be seen from the above technical solution, this application has at least the following advantages and positive effects:

[0085] Firstly, the proposed solution solves the problems of untimely observation of dust conditions inside the LF furnace, delayed adjustment of the dust removal valve leading to a large amount of dust escaping, or excessive suction caused by excessive opening of the dust removal valve, resulting in air entering the ladle cover and causing secondary oxidation of molten steel. The proposed solution uses image recognition to monitor the dust conditions inside the LF furnace in real time, detects the dust conditions inside the LF furnace through a dust detection model, and controls the dust removal valve to open to a preset degree based on the detection results, thereby improving the efficiency of steel smelting and ensuring the quality of molten steel.

[0086] Secondly, by adopting the scheme proposed in this application, the color recognition technology of the smoke and dust detection model can more accurately determine the intensity and diffusion direction of the smoke and dust, thereby more accurately controlling the opening of the dust removal valve.

[0087] Third, by adopting the solution proposed in this application, the problem of insufficient smoke and dust image data is addressed by using the StableDiffusion AI mapping tool to train the smoke and dust model, thereby obtaining a large amount of newly produced effective smoke and dust data and increasing the amount of dataset.

[0088] Fourth, adopting the solution proposed in this application simplifies the operation process, reduces the risk of manual intervention, and improves the level of automation in environmental protection control.

[0089] Although this application has been described with reference to several typical embodiments, it should be understood that the terminology used is descriptive and exemplary, and not restrictive. Since this application can be embodied in many forms without departing from the spirit or substance of the application, it should be understood that the above embodiments are not limited to any of the foregoing details, but should be interpreted broadly within the spirit and scope defined by the appended claims. Therefore, all variations and modifications falling within the scope of the claims or their equivalents should be covered by the appended claims.

Claims

1. A method for adjusting the dust removal valve of an LF furnace, characterized in that, The method includes: Acquire images of smoke and dust inside the LF furnace; The smoke image is detected using a pre-trained smoke detection model to determine the size, color, and diffusion direction of smoke, and a smoke detection result is generated. This detection includes: inputting the smoke image into the smoke detection model; using an edge detection algorithm to identify the smoke outline; calculating the area of ​​the smoke region based on the smoke outline; determining the color of the smoke image using the smoke detection model; and using a color-changing algorithm (CV) to identify changes in the center position of the smoke outline to confirm the diffusion direction. The step of identifying changes in the center position of the smoke outline using the CV algorithm includes: tracking the center position of the smoke outline in consecutive frames of the smoke image using the CV algorithm; and obtaining the diffusion direction of the smoke in the image based on the movement trajectory of the center position. Based on the dust detection results, the dust removal valve is controlled to open to a preset degree to complete the dust removal in the LF furnace.

2. The method according to claim 1, characterized in that, Before detecting the smoke size, color, and diffusion direction of the smoke image using a pre-trained smoke detection model, the method further includes: Acquire target dust images inside the LF furnace and process the target dust images; An initial smoke detection model is constructed, and the initial smoke detection model is trained based on the processed target smoke image.

3. The method according to claim 2, characterized in that, The processing of the target smoke image includes: The target smoke and dust image is augmented using a mapping tool to generate a smoke and dust image dataset.

4. The method according to claim 2, characterized in that, The initial smoke and dust detection model was constructed using the mask-rcnn model.

5. The method according to claim 2, characterized in that, The step of training the initial smoke detection model based on the processed target smoke image includes: The processed target smoke image is then labeled with smoke color. The labeled target smoke and dust image is input into the initial smoke and dust detection model for deep learning training.

6. A dust removal valve regulating device for an LF furnace, characterized in that, The device includes: The acquisition unit is used to acquire images of smoke and dust inside the LF furnace; The detection unit is used to detect the size, color, and diffusion direction of smoke in the smoke image using a pre-trained smoke detection model, and generate smoke detection results. The detection of smoke size, color, and diffusion direction using the pre-trained smoke detection model includes: inputting the smoke image into the smoke detection model, using an edge detection algorithm to identify the smoke outline, and calculating the area of ​​the smoke region based on the smoke outline; judging the color of the smoke in the smoke image using the smoke detection model, and identifying the change in the center position of the smoke outline using a CV algorithm to confirm the diffusion direction of the smoke; wherein, identifying the change in the center position of the smoke outline using the CV algorithm includes: tracking the center position of the smoke outline in consecutive frames of the smoke image using the CV algorithm; and obtaining the diffusion direction of the smoke in the smoke image based on the movement trajectory of the center position. The control unit is used to control the dust removal valve to open to a preset degree based on the dust detection results, so as to complete the dust removal in the LF furnace.

7. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores at least one piece of program code, which is loaded and executed by a processor to perform the operations performed by the method as described in any one of claims 1 to 5.

8. An electronic device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it performs the operations described in any one of claims 1 to 5.

Images