Robot material supplementing production line for aluminum smelting

Abstract

The invention discloses a robot material supplementing production line for aluminum smelting. The robot material supplementing production line comprises a pressing machine, a conveying and storing mechanism, a discharging mechanism and a smelting furnace which are arranged in sequence. According to the method, a detector is configured to detect and obtain the composition proportion of an aluminum-iron solution, a robot is configured to compare current composition proportion information, current aluminum block weight information and a preset aluminum-iron formula to obtain material supplementing data, and according to the material supplementing data, an iron block is grabbed from an iron material area and put into a smelting furnace, and then an aluminum block is grabbed from an aluminum material area and put into the smelting furnace; and automatic detection and feeding of the ferro-aluminum solution are realized. The multiple material block types are arranged in multiple layers, the material blocks are digitally conveyed and stored, the weight and position information of the material blocks is accurately recorded, and the corresponding material blocks can be conveniently selected for follow-up smelting operation.

Description

Technical Field

[0001] This invention relates to the field of aluminum smelting technology, and in particular to a robotic feeding production line for aluminum smelting. Background Technology

[0002] Aluminum smelting technology is divided into multiple stages based on purity requirements: primary aluminum is purified into refined aluminum through a three-layer electrolysis method. High-purity aluminum is produced by using refined aluminum as raw material and achieving purity through zone melting or organic solution electrolysis. Recycled aluminum is produced by remelting scrap aluminum, which requires less energy. Scrap aluminum is mainly used to produce aluminum alloys. Before remelting, the scrap aluminum must be sorted, classified, and appropriately blended and processed to achieve the desired alloy composition. Currently, scrap aluminum is collected, compacted into blocks, and then conveyed to the feeding location via a conveyor. It is then fed into the scrap aluminum smelting furnace for melting as needed using grab buckets or conveyors. The automation of aluminum molten metal inspection and replenishment in scrap aluminum smelting is not high. Summary of the Invention

[0003] The purpose of this invention is to address the aforementioned problems by providing a robotic feeding production line for aluminum smelting, which automates the detection and feeding of molten aluminum and iron, and employs multiple layered arrangements of various material block types to digitally transport and store the material blocks, accurately recording the weight and position information of each material block.

[0004] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0005] A robotic feeding production line for aluminum smelting includes a press, a conveying and storage mechanism, a feeding mechanism, and a smelting furnace arranged in sequence.

[0006] The conveying and storage mechanism includes a conveyor line, a storage line, and a main controller. The conveyor line includes a front conveyor unit and a layered conveyor unit. The storage line includes at least two sliding lines. The front conveyor unit is connected to the at least two sliding lines through the layered conveyor units, and the sliding lines are arranged horizontally in a straight line. The main controller is connected to the layered conveyor units and performs the following conveying and storage steps to convey the material block through the front conveyor unit and the layered conveyor units to the sliding lines and allow it to slide on them:

[0007] Construct a conveying and storage model, set a sliding line and all storage positions, sliding distances, and material block information. With the sliding distance of the material block on the sliding line as the target, based on the distance and height difference between the front conveyor unit and the sliding line, and according to the weight of the material block and the front speed of the last section of the front conveyor unit, adjust the layering speed of the last section of the layered conveyor unit to obtain the storage mapping relationship between the sliding distance and the weight, front speed, and layering speed of the material block.

[0008] The system transports and stores material blocks, obtains the weight of the block to be transported, selects the corresponding sliding line, obtains the storage position and sliding distance of the sliding line, obtains the corresponding forward speed, and obtains the corresponding layering speed according to the storage mapping relationship. The layering speed is then transmitted to the layering conveyor unit corresponding to the sliding line, so that the layering conveyor unit can execute the conveying task at the layering speed and provide feedback. The system then updates the storage position, sliding distance, and material block information of the sliding line based on the feedback results.

[0009] The feeding mechanism includes a detector and robot, a storage line, and a stirring mechanism. The detector can detect the composition ratio of the aluminum-iron molten metal in the smelting furnace. The robot is equipped with a moving mechanism and a gripping mechanism, enabling it to grip, feed, and move materials. The stirring mechanism is located in the smelting furnace. The storage line includes an iron material area for storing iron blocks and an aluminum material area for storing aluminum blocks. The robot is equipped with a feeding model to receive the current composition ratio information from the detector, obtain the weight information of all current aluminum blocks, and compare the current composition ratio information, current aluminum block weight information, and preset aluminum-iron formula to obtain feeding data. This feeding data includes the selected aluminum and iron blocks. Based on this feeding data, the robot first grabs the iron block from the iron material area and puts it into the smelting furnace, and then grabs the aluminum block from the aluminum material area and puts it into the smelting furnace. At the same time, the stirring mechanism is triggered to run and perform stirring operations.

[0010] The material blocks are divided into at least two types according to weight and size, and at least two sliding lines are arranged in upper and lower layers, with each sliding line storing one type of material block. Alternatively, they can be arranged horizontally side by side, and the sliding lines can be passive roller lines or chutes.

[0011] The layered conveyor unit consists of at least two sorting and lifting conveyors, each paired with a sliding line. The front conveyor unit is equipped with a sorting module to sort the material blocks to the corresponding sorting and lifting conveyor.

[0012] The robot is equipped with a feeding model to receive the current component ratio information from the detector, obtain the weight information of all current aluminum blocks, and obtain feeding data by comparing the current component ratio information, the current aluminum block weight information and the preset aluminum-iron formula. The feeding data includes the selected aluminum blocks and iron blocks. Based on the feeding data, the robot first grabs the iron block from the iron material area and puts it into the smelting furnace, and then grabs the aluminum block from the aluminum material area and puts it into the smelting furnace. At the same time, the stirring mechanism is triggered to run the stirring operation.

[0013] The robot is equipped with an explosion-proof vision system to identify the location of each aluminum block on the storage line.

[0014] As mentioned above, the configuration of the detector is used to detect and obtain the composition ratio of the aluminum-iron solution. The configuration of the robot is used to obtain the replenishment data by comparing the current composition ratio information, the current aluminum block weight information and the preset aluminum-iron formula. Based on the replenishment data, the robot first grabs the iron block from the iron material area and puts it into the smelting furnace, and then grabs the aluminum block from the aluminum material area and puts it into the smelting furnace, thereby realizing the automated detection and replenishment of the aluminum-iron solution.

[0015] Based on the aforementioned solution, in an improved solution, to detect and identify the timing of feeding materials into the smelting furnace, the robotic automatic feeding system for aluminum smelting is equipped with an image recognition model. An explosion-proof vision mechanism acquires video data of the aluminum-iron molten metal surface. The image recognition module processes the image of the aluminum-iron molten metal surface and outputs a judgment that the slag amount meets the feeding requirements when the slag area change is less than or equal to the slag amount threshold. The feeding model obtains the judgment that the slag amount meets the feeding requirements, then monitors and controls the temperature of the aluminum-iron molten metal in the smelting furnace to stabilize within a preset temperature range (e.g., 660~700℃), and then runs the stirring mechanism to perform stirring operations for 5-10 minutes. Afterwards, the composition ratio of the aluminum-iron molten metal at different locations in the smelting furnace is obtained by a detector. When the difference in the composition ratio of the aluminum-iron molten metal at each location is less than the error value, it is determined that the aluminum-iron molten metal is uniform, triggering the robot to perform feeding operations. The image recognition model, built upon deep learning, periodically identifies slag on the surface of the molten aluminum-iron solution during slag removal intervals. It calculates the current slag area percentage and compares it to the previous percentage to obtain a numerical change in slag area. This change is then compared to a preset slag quantity threshold. If the change is less than or equal to the threshold, the slag quantity is deemed sufficient for replenishment. Thus, by recognizing slag on the molten aluminum-iron solution surface through image analysis and preliminarily determining replenishment timing based on slag area changes, and further considering the temperature control and uniformity of the molten aluminum-iron solution, automated detection and identification of replenishment timing can be achieved.

[0016] Based on the aforementioned solution, in an improved version, to promptly detect and trigger the stirring operation, the robotic automatic feeding system for aluminum smelting uses an image recognition module to process the image of the aluminum-iron molten metal surface during the process of placing iron or aluminum blocks into the smelting furnace. The module determines whether an iron or aluminum block is present in the current image; if not, it determines that an iron or aluminum block has been placed and triggers the stirring mechanism to perform the stirring operation. This automates the detection and triggering of the stirring operation.

[0017] By adopting the above technical solution, the present invention has the following beneficial effects:

[0018] 1. The present invention configures a detector to detect and obtain the composition ratio of aluminum-iron solution, and configures a robot to obtain replenishment data by comparing the current composition ratio information, the current aluminum block weight information and the preset aluminum-iron formula. Based on the replenishment data, the robot first grabs the iron block from the iron material area and puts it into the smelting furnace, and then grabs the aluminum block from the aluminum material area and puts it into the smelting furnace, thereby realizing automated detection and replenishment of aluminum-iron solution.

[0019] 2. This invention identifies slag on the surface of an aluminum-iron solution through image recognition, and then preliminarily determines the timing of material replenishment based on changes in slag area. Furthermore, based on the temperature control and uniformity of the aluminum-iron solution, it can achieve automated detection and identification of the timing of material replenishment.

[0020] 3. This invention employs multiple layered arrangements of various material block types, digitally conveying and storing the blocks, accurately recording the weight and location information of each block for subsequent smelting operations. It utilizes an active conveyor line and a passive roller or chute storage mode, based on a conveying and storage model mapping relationship, to transport the blocks to the target storage location. Attached Figure Description

[0021] Figure 1 This is a layout diagram of the conveying and storage of the present invention.

[0022] Figure 2 This is a side view of the layout of the conveyor line and the sliding line of the present invention.

[0023] Figure 3 This is the material replenishment operation process of the present invention. Detailed Implementation

[0024] The specific implementation of the invention will be further described below with reference to the accompanying drawings.

[0025] As mentioned above, this application includes basic solutions and improved solutions. For example, an improved solution configures a separator module for the spacing between material blocks; an improved solution configures an explosion-proof vision mechanism for identifying the position of each aluminum block on the storage line; an improved solution automatically identifies the timing of feeding material into the smelting furnace, etc. The feature combinations of each application example can be combined according to actual needs. The following will use a preferred example of all feature combinations as an example to illustrate.

[0026] like Figures 1-3 As shown, the robotic feeding production line for aluminum smelting in this application includes a press 1, a conveying and storage mechanism, a feeding mechanism, and a smelting furnace 4 arranged in sequence.

[0027] The press 1 outlet is equipped with a weighing component to realize the weighing and detection of the material block. The press, smelting furnace and smelting furnace process parameters are all implemented using existing technologies. This application is to improve the feeding and conveying storage technology used therein, as follows.

[0028] The conveying and storage mechanism includes a conveyor line, a storage line, and a main controller. The conveyor line includes a front conveyor unit 2 and a layered conveyor unit. The storage line includes at least two sliding lines. The front conveyor unit is connected to the at least two sliding lines through the layered conveyor units, and the sliding lines are arranged horizontally in a straight line. The main controller is connected to the layered conveyor units and performs the following conveying and storage steps to convey the material block 100 to the sliding lines via the front conveyor unit and the layered conveyor units, and allow it to slide on the sliding lines:

[0029] First, a conveying and storage model is constructed, as follows: a sliding line and all storage positions 101 on it are set, along with the sliding distance and material block information. Taking the sliding distance of the material block on the sliding line as the target, based on the distance and height difference between the front conveyor unit and the sliding line, and according to the weight of the material block and the front speed of the last section of the front conveyor unit, the layering speed of the last section of the layered conveyor unit is adjusted to obtain the storage mapping relationship between the sliding distance and the weight, front speed, and layering speed of the material block.

[0030] Then, based on the conveying and storage model, the material blocks are conveyed and stored. Specifically, the weight of the material block to be conveyed is obtained, and the corresponding sliding line is selected accordingly. The storage position and sliding distance of the sliding line are obtained, and the corresponding forward speed is obtained. The corresponding layer speed is obtained according to the storage mapping relationship. The layer speed is transmitted to the layer conveyor unit corresponding to the sliding line so that the layer conveyor unit can execute the conveying task of the layer speed and feed back the result. Then, the storage position, sliding distance and material block information of the sliding line are updated according to the feedback result.

[0031] The optimization of the conveying and storage model mapping relationship is achieved using existing technologies such as target optimization algorithms. The material blocks are divided into at least two weight types, such as 50kg and 60kg, and at least two sliding lines are arranged in upper and lower layers, with each sliding line storing one type of material block. Alternatively, a horizontal parallel arrangement can be used, spreading out in multiple rows, which may affect the clamping arm length and transfer distance in subsequent clamping operations. The sliding lines are either passive roller lines or chutes.

[0032] In one embodiment, the layered conveyor unit consists of at least two sorting and lifting conveyors, as shown in the figure, sorting and lifting conveyor I 21 and sorting and lifting conveyor II 22. Each sorting and lifting conveyor is paired with a sliding line, as shown in the figure, sliding line I 3 and sliding line II 31. The front conveyor unit is equipped with a sorting module to sort the material blocks to the corresponding sorting and lifting conveyor.

[0033] In another embodiment, unlike the previous embodiment, the robotic feeding production line for aluminum smelting uses a liftable conveyor unit. This liftable conveyor is equipped with a lifting module to adjust its connection to the corresponding sliding line based on the weight of the material block to be conveyed; this minimizes the size of the liftable conveyor. Preferably, in this conveying and storage system, the front section of the front conveyor unit is equipped with a separating module to adjust the distance between the current material block and the previous material block based on the weight of the material block to be conveyed, the sliding line, the sliding distance, the forward speed, the layering speed, and the lifting time of the lifting module. This ensures that the current material block can be conveyed to the corresponding storage position on the sliding line within the corresponding lifting time. This allows for real-time adjustment of the distance based on the required lifting time of the sliding line, improving conveying efficiency while ensuring layered arrangement.

[0034] The feeding mechanism includes a detector, a robot, a storage line, and a stirring mechanism. The detector can detect the composition ratio of the molten aluminum and iron in the smelting furnace. The robot is equipped with a moving mechanism and a gripping mechanism, enabling it to grip, feed, and move materials. The stirring mechanism is located in the smelting furnace. The storage line includes an iron material area for storing iron blocks and an aluminum material area for storing aluminum blocks (the aluminum material area includes sliding line I and sliding line II). Its material replenishment process is as follows: Figure 1 As shown, the following will be explained in detail with reference to specific components.

[0035] The robot is equipped with a feeding model to receive the current component ratio information from the detector, obtain the weight information of all current aluminum blocks, and obtain feeding data by comparing the current component ratio information, the current aluminum block weight information and the preset aluminum-iron formula. The feeding data includes the selected aluminum blocks and iron blocks. Based on the feeding data, the robot first grabs the iron block from the iron material area and puts it into the smelting furnace, and then grabs the aluminum block from the aluminum material area and puts it into the smelting furnace. At the same time, the stirring mechanism is triggered to run the stirring operation.

[0036] The robot is equipped with an autonomous driving system and a robotic arm, which, in conjunction with laser navigation and vision, navigates to the storage line. The robotic arm's gripper handles material handling, and all of these are existing components. The detector, robot, smelting furnace, and its mixing mechanism are briefly described here and will not be elaborated upon in detail.

[0037] To adapt to the working environment of the smelting furnace, the front end of the robotic arm is equipped with an explosion-proof vision mechanism to identify the position of each aluminum block on the storage line; moreover, the explosion-proof camera (explosion-proof vision mechanism) monitors the temperature of the smelting furnace or aluminum-iron molten metal in real time through a built-in thermal imaging sensor. For example, Hikvision's DS-2TD2528T-3 / Q model supports dual-spectrum temperature measurement (-20℃~150℃ or 0℃~550℃ range) with an accuracy of ±2℃ and has a temperature abnormality alarm function.

[0038] The stirring mechanism can be a stirring cart or an electromagnetic stirrer to thoroughly stir the molten aluminum and iron in the melting furnace, thereby ensuring a uniform mixing of the internal concentration gradient and promoting the diffusion of alloying elements. The stirring time needs to be long enough (usually 5-10 minutes) to ensure that forced convection covers the entire melting furnace.

[0039] The smelting process requires slag removal and stirring. During the feeding process, the timing of adding scrap aluminum blocks depends on the amount of slag. A decrease in slag may indicate a reduction in the degree of oxidation of the melt or the initial removal of impurities. If there is too much slag at the beginning of smelting, slag removal or refining by blowing air (nitrogen / argon treatment for 10-15 minutes) should be used to reduce impurities before considering feeding. Then, the melt temperature needs to be stabilized within the appropriate range, such as 660-700℃ (flame furnace) or 680-750℃ (medium frequency furnace). Furthermore, the melt state needs to be assessed. After the initial charge has completely melted, stirring should be used to confirm uniform composition before feeding in stages and stirring evenly to avoid bridging of unmelted aluminum blocks that could lead to localized overheating. When initially adding material to the furnace, iron blocks should be added first, placed low and gently to avoid excessive impact on the furnace lining and potential cracking. A separate stirring mechanism should be installed to perform stirring operations as needed.

[0040] The amount of slag on the surface of the molten aluminum can be determined by using an explosion-proof camera and human observation, along with slag removal and changes in slag quantity. To achieve automated identification, this application employs an AI slag quantity identification model. Based on existing artificial intelligence and image recognition technologies such as neural networks and deep learning algorithms, a slag quantity identification model is constructed for preliminary judgment, automatically calculating a value called the slag area change value. This slag area change value is used as a correlation value for the feeding operation, serving as a preliminary judgment for the timing of feeding into the smelting furnace. The temperature of the molten aluminum can be monitored using an explosion-proof camera, and a smelting furnace temperature control system is configured to maintain the molten aluminum in the furnace within a preset temperature range, achieving stable temperature control and meeting temperature requirements. The uniformity of the molten aluminum can be detected at various locations using a detector and a robotic arm. The obtained data from each location is compared with error data. If the data is within the error range, the molten aluminum is considered uniform; otherwise, it is considered non-uniform, and the process is repeated for stirring and further evaluation of uniformity.

[0041] Specifically, in order to detect and identify the timing of feeding materials into the smelting furnace, the robotic automatic feeding system for aluminum smelting is equipped with an image recognition model. The explosion-proof vision mechanism acquires video data of the aluminum-iron molten metal surface. The image recognition module processes the image of the aluminum-iron molten metal surface and outputs a judgment that the slag amount meets the feeding requirements when the slag area change is less than or equal to the slag amount threshold. The feeding model obtains the judgment that the slag amount meets the feeding requirements, and then monitors and controls the temperature of the aluminum-iron molten metal in the smelting furnace to stabilize within a preset temperature range (e.g., 660~700℃). Then, the stirring mechanism is run to stir for 5-10 minutes. After that, the composition ratio of the aluminum-iron molten metal at different locations in the smelting furnace is obtained by the detector. When the difference in the composition ratio of the aluminum-iron molten metal at each location is less than the error value, it is determined that the aluminum-iron molten metal is uniform, and the robot is triggered to perform the feeding operation. The image recognition model, built upon deep learning, periodically identifies slag on the surface of the molten aluminum-iron solution during slag removal intervals. It calculates the current slag area percentage and compares it to the previous percentage to obtain a numerical change in slag area. This change is then compared to a preset slag quantity threshold. If the change is less than or equal to the threshold, the slag quantity is deemed sufficient for replenishment. Thus, by recognizing slag on the molten aluminum-iron solution surface through image analysis and preliminarily determining replenishment timing based on slag area changes, and further considering the temperature control and uniformity of the molten aluminum-iron solution, automated detection and identification of replenishment timing can be achieved.

[0042] To promptly detect and trigger stirring operations, this robotic automatic feeding system for aluminum smelting employs an image recognition module that processes images of the molten aluminum-iron molten metal surface during the process of placing iron or aluminum blocks into the furnace. The module determines the presence of either iron or aluminum blocks in the current image; if none are found, it indicates that either has been placed, triggering the stirring mechanism to begin stirring. This automates the detection and triggering of stirring operations.

[0043] As described above, a detector is configured to obtain the aluminum-iron molten metal composition ratio. A robot, based on the current composition ratio, the current aluminum block weight, and a preset aluminum-iron formula, obtains replenishment data. Based on this data, it first grabs an iron block from the iron block area and places it into the smelting furnace, then grabs an aluminum block from the aluminum block area and places it into the smelting furnace, thus automating the detection and replenishment of the aluminum-iron molten metal. Aluminum blocks are pressed by a press, resulting in significant weight variations, while the weight variations of iron blocks are controllable. Selecting the appropriate iron block based on the known aluminum block weight allows for faster replenishment. Multiple layers of various block types are arranged, and the blocks are digitally transported and stored, accurately recording the weight and location information of each block for subsequent smelting operations. An active conveyor line and a passive roller or chute storage mode are used, based on a storage mapping relationship in the transport and storage model, to transport the blocks to the target storage location.

[0044] It should be noted that the examples of the above embodiments can preferably be combined with one or more of each other according to actual needs, and the accompanying drawings of multiple examples adopt a set of combined technical features, which will not be described in detail here.

[0045] The above description is a detailed explanation and illustration of the preferred embodiments of the present invention. However, these descriptions are not intended to limit the scope of protection claimed by the present invention. All equivalent changes or modifications made under the technical teachings of the present invention should fall within the patent protection scope covered by the present invention.

Claims

1. A robotic make-up line for aluminum smelting, characterized by: The device comprises a press, a conveying and storing mechanism, a discharging mechanism and a smelting furnace arranged in sequence. The conveying and storing mechanism comprises a conveying line and a storing line, a main controller, the conveying line comprises a front conveying unit and a layered conveying unit, the storing line comprises at least two slidable lines, the front conveying unit is connected to the at least two slidable lines through the layered conveying unit, and the slidable lines are arranged in a horizontal straight line, and the main controller is connected to the layered conveying unit and performs the following conveying and storing steps to convey the material block to the slidable line through the front conveying unit and the layered conveying unit and make it slide on the slidable line: A conveying and storing model is constructed, and the slidable line and all the storing positions, sliding distances and material block information thereon are set, the sliding distance of the material block on the slidable line is taken as a target, the distance and height difference between the front conveying unit and the slidable line are based on, the layered speed of the layered conveying unit at the end is adjusted according to the weight of the material block, the front speed of the front conveying unit at the end, and the storing mapping relationship between the sliding distance and the weight of the material block and the front speed and the layered speed is obtained; The weight of the material block to be conveyed is obtained, the corresponding slidable line is selected according to the weight, the storing positions and the sliding distances of the slidable line are obtained, the corresponding front speed is obtained, the corresponding layered speed is obtained according to the storing mapping relationship, the layered speed is transmitted to the layered conveying unit corresponding to the slidable line, the layered conveying unit performs the conveying task at the layered speed and feeds back the result, and then the storing positions, the sliding distances and the material block information of the slidable line are updated according to the feedback result; The discharging mechanism comprises a detector and a robot, a storing line and a stirring mechanism, the detector can detect the component ratio of the aluminum-iron solution in the smelting furnace, the robot is provided with a moving mechanism and a grabbing mechanism to have the functions of grabbing and discharging, the stirring mechanism is arranged in the smelting furnace, and the storing line comprises an iron material area for storing iron blocks and an aluminum material area for storing aluminum blocks; wherein the robot is provided with a replenishment model for receiving the current component ratio information of the detector, obtaining all the current aluminum block weight information, obtaining the replenishment data according to the comparison between the current component ratio information, the current aluminum block weight information and a preset aluminum-iron formula ratio, the replenishment data comprising selected aluminum blocks and iron blocks, grabbing the iron blocks from the iron material area and putting them into the smelting furnace, grabbing the aluminum blocks from the aluminum material area and putting them into the smelting furnace, and triggering the stirring mechanism to operate and stir at the same time.

2. The robotic make-up line for aluminum smelting of claim 1, wherein: The robot is provided with an anti-explosion visual mechanism for identifying the positions of the aluminum blocks on the storing line.

3. The robotic make-up line for aluminum smelting of claim 2, wherein: The robot is configured with an image recognition model, an explosion-proof vision mechanism acquires aluminum-iron solution surface video data, an image recognition module processes the aluminum-iron solution surface image, according to the slag area change of the current image, when the slag area change is less than or equal to the slag amount threshold, the slag amount meets the feeding requirement is output; the feeding model obtains the slag amount meets the feeding requirement, then monitors and controls the aluminum-iron solution temperature of the smelting furnace to be stable in the preset temperature range, then runs the stirring mechanism to perform stirring operation for 5-10 minutes, and then detects the composition ratio of the aluminum-iron solution at different positions of the smelting furnace through the detector, when the composition ratio difference of the aluminum-iron solution at each position is less than the error value, it is determined that the aluminum-iron solution is uniform, and the robot is triggered to perform feeding operation.

4. The robotic make-up line for aluminum smelting of claim 3, wherein: The image recognition model is constructed based on deep learning, identifies the slag on the surface of the aluminum-iron solution at the slag removal interval period, calculates the current slag area ratio, compares the current slag area ratio with the previous slag area ratio, obtains the slag area change value, compares the slag area change value with the preset slag amount threshold, and when the slag area change value is less than or equal to the slag amount threshold, it is determined that the slag amount meets the feeding requirement.

5. The robotic make-up line for aluminum smelting of claim 3, wherein: The image recognition module processes the aluminum-iron solution surface image during the process of putting the iron block or aluminum block into the smelting furnace, judges whether there is an iron block or aluminum block in the current image, if not, it is determined that the iron block or aluminum block has been put in, and the stirring mechanism is triggered to perform stirring operation.

Images