A control operation method and device of a rare earth metal electrolytic furnace discharging robot
By controlling the rare earth metal electrolysis furnace discharge robot with visual sensors and force sensing system, the safety risks and equipment damage problems in the rare earth metal electrolysis smelting discharge process are solved, and efficient and safe discharge operation is achieved.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- SHANGHAI JIAO TONG UNIVERSITY INNER MONGOLIA RESEARCH INSTITUTE
- Filing Date
- 2025-07-04
- Publication Date
- 2026-06-09
Smart Images

Figure CN120533711B_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of automation technology in rare earth metal electrolytic smelting, specifically to a control operation method and device for a rare earth metal electrolytic furnace discharge robot. Background Technology
[0002] With the rapid development of modern industrial automation, especially in the field of rare earth metal electrolytic smelting automation, the demand for efficient and smooth discharge technology in high-temperature and highly corrosive environments is increasing. Discharge in rare earth metal electrolytic smelting is a dangerous, laborious, and requires stability. It necessitates precisely extracting molten rare earth metal (800-1200℃) from the electrolytic furnace and smoothly transferring it to the discharge point, where it is then smoothly poured into a mold. Traditional methods rely on manual operation or semi-automated equipment. During manual operation, the splashing of high-temperature molten rare earth metal poses a safety risk, and rare earth metals release toxic gases during electrolysis, making prolonged exposure to high-temperature radiation and toxic environments unacceptable. The fluidity, oxidation level, and impurity content of the molten metal directly affect the discharge quality, but manual monitoring cannot quantify these factors in real time, easily leading to the outflow of substandard metal, causing rework in subsequent processes and wasting resources. General-purpose robots lack dynamic calibration capabilities, easily causing collisions between the extraction tool and the cathode or scratches against the furnace wall. Long-term exposure of the extraction tool to the high-temperature molten salt environment can easily lead to creep deformation. Manual operation or rigid mechanical control can easily cause tools to come into hard contact with the crucible / furnace wall, leading to equipment damage. Summary of the Invention
[0003] In view of the deficiencies in the prior art, the purpose of this application is to provide a control operation method for a rare earth metal electrolysis furnace discharge robot.
[0004] The first aspect of this application provides a control operation method for a rare earth metal electrolysis furnace discharge robot, comprising:
[0005] The discharge robot uses visual sensors to obtain the electrolysis status of the electrolytic furnace and determine whether the discharge standard has been met.
[0006] After determining that the discharge standard has been met, the discharge robot grabs the discharge tool, preheats the discharge tool, and moves it to the discharge position to prepare for discharge.
[0007] The unloading robot moves along the planned path, monitors the contact force between the robot and the electrolytic furnace through a force sensing system, and corrects the robot trajectory based on the contact force to execute the unloading process;
[0008] After the discharge process is completed, it is determined whether to continue discharging. If not, the process ends.
[0009] Optionally, the discharge robot acquires the electrolysis state of the electrolytic furnace through a vision sensor and determines whether the discharge standard has been met, including:
[0010] The visual sensor at the furnace opening position of the electrolytic furnace detects the state image data of the molten rare earth metal in the electrolytic furnace and sends it to the discharge robot.
[0011] The discharge robot processes the image data of the flame state of the molten rare earth metal in the electrolytic furnace using a depth algorithm, compares the image data with the set discharge standard image, and determines whether the image data meets the set discharge standard. If yes, it determines that the material can be discharged; otherwise, it continues to acquire image data.
[0012] Optionally, after determining that the discharge standard has been met, the discharge robot grabs the discharge tool, preheats the discharge tool, and moves it to the discharge position to prepare for discharge, including:
[0013] The location of the discharge tool is obtained and sent to the discharge robot, which then moves to the location of the discharge tool and grabs it.
[0014] After the discharge tool successfully grabs the material, it is moved into the electrolytic furnace for preheating until it reaches the same temperature as the rare earth metals inside the furnace, and then moved to the discharge position to prepare for discharge.
[0015] Optionally, the step of acquiring the location of the discharge tool and sending it to the discharge robot, and the discharge robot moving to the location of the processing tool to grasp the discharge tool, includes:
[0016] The position of the discharging tool on the tool rack is obtained and sent to the discharging robot;
[0017] The unloading robot moves to the position of the unloading tool according to the position of the unloading tool, identifies the unloading tool through the vision sensor, and verifies it with the set unloading tool. After successful verification, the unloading robot grabs the unloading tool.
[0018] After the material feeding tool is grasped, a force sensor is used to perform a secondary verification of the material feeding tool to detect whether the material feeding tool has been successfully grasped. If so, the material feeding tool is determined to have been successfully grasped, and the material feeding tool is controlled to move to the ready position. If not, the verification continues.
[0019] Optionally, after the discharge tool successfully grasps the material, moving the discharge tool into the electrolytic furnace for preheating to the same temperature as the rare earth metals inside the furnace, and then moving it to the discharge position to prepare for discharge, includes:
[0020] The position of the furnace opening of the electrolytic furnace is obtained by the visual sensor at the furnace opening and then inspected.
[0021] After successful verification, the discharge tool is placed into the electrolytic furnace for heating, and the temperature of the discharge tool is acquired in real time. It is determined whether the temperature of the discharge tool is consistent with the temperature of the rare earth metal in the electrolytic furnace. If so, the preheating is considered complete; otherwise, heating continues.
[0022] After the preheating is completed, the discharge tool is moved to the discharge preparation position to prepare for discharge.
[0023] Optionally, the unloading robot moves along a planned path, monitors the contact force between the robot and the electrolytic furnace through a force sensing system, and corrects the robot trajectory based on the contact force to execute the unloading process, including:
[0024] The discharge robot controls the discharge tool to move along a pre-set path, bringing the discharge tool closer to the bottom of the electrolytic furnace;
[0025] The force sensor on the discharge robot acquires the contact force between the discharge tool and the electrolytic furnace in real time.
[0026] Based on the contact force, the algorithm model set on the unloading robot corrects the trajectory of the unloading robot;
[0027] The position of the discharge tool inside the electrolysis furnace is obtained, and it is determined whether the set position has been reached. If yes, the operation is stopped; otherwise, the operation continues.
[0028] After determining that the set position has been reached, the discharge robot is controlled to stop running for a set time, and then the discharge tool is controlled to leave the electrolysis furnace.
[0029] Optionally, after controlling the discharge tool to leave the electrolytic furnace, the method further includes:
[0030] The unloading robot controls the unloading tool to move along the planned path to above the mold;
[0031] The material discharge tool is controlled to rotate around a set direction and angle to discharge the material and then stop moving.
[0032] The discharge tool stops at the set time and then returns to its original position, ending the discharge process.
[0033] Optionally, after determining that discharging will not continue, the method further includes:
[0034] The unloading robot controls the unloading tool to move to the unloading tool placement position and places the unloading tool there;
[0035] The unloading robot moves to an idle position and inspects the unloading tool to determine whether the unloading tool is deformed beyond the set size. If so, it is determined that the unloading tool needs to be replaced; otherwise, the process ends.
[0036] In a second aspect, this application provides an apparatus for performing the control operation method of the rare earth metal electrolytic furnace discharge robot, comprising: a support device, a robotic arm, and a control module;
[0037] The support device is used to support the robotic arm, and the control module is mounted on the robotic arm to control the operation of the robotic arm;
[0038] The robotic arm is equipped with a force sensor, which is used to acquire the contact force between the discharge tool and the electrolytic furnace and send it to the control module. The control module uses the contact force to execute a compliant control strategy to correct the running trajectory of the robotic arm.
[0039] Optionally, the support device includes a track bracket and an L-shaped slide rail;
[0040] The L-shaped slide rail is fixed to the rail bracket and is movable relative to the rail bracket;
[0041] The robotic arm includes a first robotic arm, a second robotic arm, a third robotic arm, a fourth robotic arm, and a fifth robotic arm;
[0042] The first robotic arm is slidably mounted on the L-shaped slide rail, the second robotic arm is movably mounted on the first robotic arm, the third robotic arm is movably mounted on the second robotic arm, the fourth robotic arm is rotatably mounted on the third robotic arm, and the fifth robotic arm is mounted at one end of the second robotic arm for controlling the movement of the third robotic arm and the fourth robotic arm.
[0043] The material discharge tool is grasped by the fourth robotic arm, and the force sensor is installed on the fourth robotic arm to obtain the contact force of the material discharge tool.
[0044] This application provides a control method for a rare earth metal electrolysis furnace discharge robot. By applying automated robot technology, force sensors, vision sensors, and compliant control strategies, it achieves automation, intelligence, stability, and reliability in the discharge operation of rare earth metal electrolysis furnaces. At the same time, it reduces labor and maintenance costs. It not only optimizes the discharge process of rare earth metal electrolysis furnaces but also provides an innovative, efficient, and safe solution for automated industrial applications, which has significant practical value and broad application prospects.
[0045] Other technical effects resulting from the additional features will be further illustrated in the corresponding embodiments. Attached Figure Description
[0046] Other features, objects, and advantages of this application will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings:
[0047] Figure 1 This is a flowchart illustrating a control operation method for a rare earth metal electrolytic furnace discharge robot according to an exemplary embodiment.
[0048] Figure 2 This is a detailed flowchart illustrating a control operation method for a rare earth metal electrolysis furnace discharge robot according to an exemplary embodiment.
[0049] Figure 3 This is a schematic diagram of the control device for a rare earth metal electrolytic furnace discharge robot according to an exemplary embodiment.
[0050] Figure 4 This is a schematic diagram of the structure of the robotic arm of a control device for a rare earth metal electrolytic furnace discharge robot, according to an exemplary embodiment.
[0051] Figure 5 This is a schematic diagram of the mechanical arm and discharge tool of a control device for a rare earth metal electrolytic furnace discharge robot according to an exemplary embodiment.
[0052] In the diagram: 100, support device; 200, robotic arm; 101, L-shaped slide rail; 102, L-shaped rail bracket; 103, first robotic arm; 104, second robotic arm; 105, third robotic arm; 106, fifth robotic arm; 107, fourth robotic arm; 108, control module; 109, material discharge tool; 110, force sensor. Detailed Implementation
[0053] The present application will now be described in detail with reference to specific embodiments. These embodiments will help those skilled in the art to further understand the present application, but do not limit the present application in any way. It should be noted that those skilled in the art can make several modifications and improvements without departing from the concept of the present application, and these all fall within the protection scope of the present application. Parts not described in detail in the following embodiments can be implemented using existing technology.
[0054] In the description of the embodiments of this application, it should be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "bottom", "inner", "outer", "clockwise", "counterclockwise", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings. They are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this application.
[0055] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined with "first" or "second" may explicitly or implicitly include one or more of that feature.
[0056] In the description of the embodiments in this application, "multiple" means two or more, unless otherwise explicitly specified. In this application, unless otherwise explicitly specified and limited, the terms "installed," "connected," "linked," "fixed," etc., should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this application according to the specific circumstances.
[0057] The terms "comprising" and "having," and any variations thereof, in the embodiments of this application are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or units is not limited to the steps or units listed, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to such processes, methods, products, or devices.
[0058] In traditional rare earth metal electrolytic smelting, relying on manual operation or semi-automated equipment, general-purpose robots lack dynamic calibration capabilities, easily leading to collisions between the digging tools and the cathode or scratches against the furnace wall. The digging tools are exposed to the high-temperature molten salt environment for extended periods, making them prone to creep deformation. Manual operation or rigid mechanical control can easily cause the tools to make hard contact with the crucible / furnace wall, resulting in equipment damage. Based on the above problems, this application provides a control method for a rare earth metal electrolytic furnace discharge robot to solve these issues.
[0059] Reference Figure 1 As shown in one embodiment of this application, a control operation method for a rare earth metal electrolysis furnace discharge robot includes:
[0060] S1. The discharge robot obtains the electrolysis status of the electrolytic furnace through a vision sensor and determines whether the discharge standard has been met.
[0061] S2. After determining that the discharge standard has been met, the discharge robot grabs the discharge tool, preheats the discharge tool, and moves it to the discharge position to prepare for discharge.
[0062] S3. The unloading robot moves along the planned path. The force sensing system monitors the contact force between the robot and the electrolytic furnace and corrects the robot trajectory based on the contact force to execute the unloading process.
[0063] S4. After the discharge process is completed, determine whether to continue discharging. If not, then end.
[0064] Specifically, firstly, the unloading robot uses visual sensors to collect real-time electrolysis status data (such as temperature, melt level, color, etc.) from the electrolysis furnace and analyzes it using built-in algorithms to determine whether the preset unloading standards have been met. If the conditions are met, the unloading robot automatically grabs the preheated unloading tool and preheats it using the electrolysis furnace to prevent damage to the tool or melt adhesion caused by rapid cooling of the high-temperature melt. Then, it precisely moves the unloading tool to a position above the discharge port of the electrolysis furnace. During the unloading process, the unloading robot moves along a preset path and simultaneously monitors the contact force between the robotic arm and the inner wall of the electrolysis furnace and the surface of the melt in real time through a force sensing system. The robot dynamically adjusts its trajectory (such as avoiding obstacles or slowing down the insertion speed) based on the contact force feedback to ensure that the unloading tool is stably immersed in the melt and completes the retrieval or clamping action. After unloading is completed, the system determines whether to continue the next round of unloading. If not, the operation is terminated, the tool is returned to the tool rack, and the unloading robot returns to its initial position. Otherwise, the unloading process is repeated.
[0065] Before the operation begins, a thorough inspection and preparation of the unloading robot is required. This includes checking the mechanical components of the robot body, force sensors, vision sensors, and control units to ensure that all systems are working properly. Then, the worker operates the robot to reset it to the initial position for the operation, ready to begin the unloading process.
[0066] It should be noted that the vision sensors include multiple sensors, which are respectively set on the tool rack where the processing tools are placed and at the furnace opening of the electrolytic furnace. They are used to detect the electrolytic state of rare earth metals in the discharge tool, the furnace opening of the electrolytic furnace, and the electrolytic furnace.
[0067] The embodiments described above in this application achieve real-time monitoring of the electrolysis status and intelligent judgment of the discharge timing through a visual sensor, avoiding errors caused by human intervention and ensuring consistent discharge quality. The force sensing system dynamically corrects the motion trajectory, which can cope with the complex working conditions of high temperature and strong corrosion environment in the electrolysis furnace, reduce the risk of mechanical collision, and extend the equipment life. The visual sensor and force sensor realize the automation level and safety of rare earth metal electrolysis discharge.
[0068] In some specific embodiments of this application, the unloading robot acquires the electrolysis state of the electrolytic furnace through a vision sensor and determines whether the unloading standard has been met, including:
[0069] The system acquires image data of the flame state of molten rare earth metal in the electrolytic furnace detected by a vision sensor at the furnace opening position, and sends it to the discharge robot. The discharge robot acquires image data of the state of molten rare earth metal in the electrolytic furnace through a vision sensor, processes the image data using a depth algorithm, compares it with the set discharge standard image, and determines whether the image data meets the set discharge standard. If yes, it determines that the material can be discharged; otherwise, it continues to acquire image data.
[0070] Specifically, in this embodiment, the unloading robot collects real-time image data of the flame state of molten rare earth metal through a vision sensor deployed at the furnace opening of the electrolytic furnace, and transmits the image signals to the unloading robot's control system. The control system uses a pre-trained deep learning algorithm (such as a convolutional neural network CNN) to extract features and perform semantic analysis on the images to identify the physical state parameters of the melt. The analysis results are compared with a preset unloading standard image library (containing flame feature templates of qualified melts), and a similarity threshold is used to determine whether the current melt meets the unloading conditions. If the conditions are met, the robot triggers the unloading command; if the conditions are not met, image data is continuously collected in a loop until the state is qualified, ensuring that the unloading timing is accurate and controllable.
[0071] It should be noted that: the visual sensor detects the color and shape changes of the flame inside the electrolytic furnace, and then uses deep learning technology to perform temporal and semantic modeling of the dynamic flame changes inside the electrolytic furnace. By analyzing the dynamic characteristic changes of the flame in different electrolytic furnaces, it can determine whether the state of the rare earth metals inside the electrolytic furnace meets the discharge standards.
[0072] The above embodiments of this application realize intelligent and high-precision material discharge decision through closed-loop control of visual perception and deep learning. The visual sensor acquires the flame state image data of rare earth metals, and the material discharge robot uses the deep learning model to identify it and compare it with the set image to avoid product defects caused by premature material discharge or energy waste caused by delayed material discharge, thus achieving the accuracy of material discharge standards.
[0073] In some specific embodiments of this application, after determining that the discharge standard has been met, the discharge robot grabs the discharge tool, preheats the discharge tool, and moves it to the discharge position to prepare for discharge, including:
[0074] The location of the discharge tool is obtained and sent to the discharge robot. The discharge robot moves to the location of the discharge tool and grabs it. After the discharge tool is successfully grabbed, it is moved into the electrolysis furnace for preheating until it is the same temperature as the rare earth metal in the electrolysis furnace. Then it is moved to the discharge position to prepare for discharge.
[0075] In the embodiments described above, after determining that the molten rare earth metal in the electrolytic furnace has reached the discharge standard, the storage location information of the discharge tool is obtained through a vision sensor and transmitted to the discharge robot. The discharge robot plans a collision-free path based on the location information, moves to the tool storage location, and completes the precise grasping of the discharge tool by its robotic arm. After successful grasping, the robot moves the tool to a pre-set preheating area above the furnace opening of the electrolytic furnace for preheating and real-time monitoring of the tool surface temperature until it reaches the same temperature as the molten rare earth metal in the electrolytic furnace. After preheating, the robot readjusts the tool posture and moves along an optimized path to directly above the discharge port of the electrolytic furnace, completing the final positioning before discharge. This ensures that the contact angle and depth between the tool and the melt meet the process requirements, achieving high efficiency and reliability in the discharge operation and improving the purity and product stability of the extracted rare earth metal.
[0076] Specifically, the location of the unloading tool is determined, the unloading robot moves to the processing tool location, and grasps the unloading tool, including:
[0077] The robot obtains the position of the dispensing tool on the tool rack and sends it to the dispensing robot. The dispensing robot moves to the position of the dispensing tool based on its position, identifies the dispensing tool using a vision sensor, and verifies it against the set dispensing tool. If the verification is successful, the dispensing robot picks up the dispensing tool. After picking up the dispensing tool, a force sensor performs a second verification to check whether the dispensing tool was successfully picked up. If so, the dispensing tool is determined to have been successfully picked up, and the robot controls the dispensing tool to move to the ready position. If not, the verification continues.
[0078] Specifically, the precise position information of the discharging tool is obtained in real time by the vision sensor on the tool rack and transmitted to the discharging robot. The discharging robot plans the optimal movement path according to the position information and quickly moves to the top of the corresponding discharging tool on the tool rack. The vision sensor performs image recognition on the tool to confirm that the discharging tool is consistent with the preset discharging tool. After the verification is passed, the discharging robot grabs the discharging tool. After the grabbing is completed, the force sensor monitors the grabbing force and tool weight feedback in real time and performs a second verification. After the verification is successful, the discharging robot carries the discharging tool to the preparation position next to the electrolysis furnace.
[0079] It should be noted that the secondary verification of the dispensing tool includes obtaining gravity feedback from the dispensing tool through a force sensor for judgment. After success, the dispensing tool is moved to the detection switch position through the set detection switch, and the switch is touched to see if the dispensing tool is properly held, thus realizing the secondary verification of the dispensing tool.
[0080] The embodiments described above in this application use a vision sensor on the tool rack and a force sensor on the unloading robot to locate and identify the unloading tool, ensuring that the unloading robot accurately grasps the target unloading tool, avoiding process accidents caused by mishandling tools, achieving dual verification protection, and improving the accuracy of grasping the unloading tool.
[0081] In some specific embodiments of this application, after the discharge tool successfully grasps the material, it is moved into the electrolytic furnace for preheating until it reaches the same temperature as the rare earth metals inside the furnace, and then moved to the discharge position to prepare for discharge, including:
[0082] The position of the furnace opening is obtained by a visual sensor at the furnace opening and verified. After successful verification, the discharge tool is placed into the furnace for heating, and the temperature of the discharge tool is monitored in real time. It is determined whether the temperature of the discharge tool is consistent with the temperature of the rare earth metal in the furnace. If so, preheating is considered complete; otherwise, heating continues. After preheating is determined to be complete, the discharge tool is moved to the discharge preparation position to prepare for discharge.
[0083] Among these measures, by detecting the furnace opening position and the downward position of the output tool, it is possible to prevent the output tool from colliding with the cathode or anode of the electrolytic furnace during its downward movement.
[0084] It should be noted that there is a movable cylindrical cathode inside the electrolytic furnace, and the anode is attached to the furnace wall.
[0085] In the above embodiments of this application, the position of the furnace opening and the downward path of the discharge tool are detected in real time by a visual sensor at the furnace opening. Combined with the cathode and anode inside the electrolytic furnace, the movement trajectory of the tool is planned to avoid mechanical collisions caused by positioning deviations or cathode displacement, thereby improving the safety of rare earth metal electrolytic discharge operations.
[0086] In some specific embodiments of this application, the unloading robot moves along a planned path, monitors the contact force between the robot and the electrolytic furnace through a force sensing system, and corrects the robot trajectory based on the contact force to execute the unloading process, including:
[0087] The unloading robot controls the unloading tool to move along a pre-planned path, bringing the tool closer to the bottom of the electrolytic furnace. Force sensors on the robot continuously monitor the contact force between the tool and the furnace. Based on this force, an algorithm model on the robot corrects its trajectory. The robot then determines the position of the tool within the furnace and checks if it has reached the set position. If so, it stops; otherwise, it continues. After determining that the set position has been reached, the robot stops for a set time before the tool leaves the furnace.
[0088] Based on the three-dimensional coordinates and attitude angles of the target pose, the pose transformation matrix of the end effector is calculated, and the robot joints are controlled by inverse kinematics to complete high-precision trajectory motion.
[0089] Specifically, the unloading robot first generates a planned path for the unloading tool based on the preset task, and calculates the motion parameters of each joint using three-dimensional coordinates and attitude angles, driving the robotic arm to move the unloading tool towards the bottom of the electrolytic furnace. During the movement, force sensors (such as six-dimensional force / torque sensors) installed at the end of the tool collect the contact force between the tool and the inner wall of the electrolytic furnace, the surface of the melt, or the electrodes in real time, and transmit the signals to the robot control module. The control module triggers a trajectory correction algorithm based on the contact force threshold, dynamically avoiding obstacles by adjusting the joint angles or speeds, while simultaneously updating the spatial position of the end effector in real time using a pose transformation matrix. When the unloading tool reaches the preset unloading position inside the electrolytic furnace, the unloading robot remains stationary for a set time to ensure that the melt is fully adhered or solidified. Finally, the robot controls the unloading tool to smoothly exit the electrolytic furnace at a preset speed, completing the unloading process.
[0090] It should be noted that the force information of the interaction between the unloading robot and the electrolytic furnace is obtained in real time based on the force sensor. The force information includes the magnitude and direction of the contact force and is fed back to the compliant controller on the unloading robot. The compliant controller constructs a compliant control process based on the reinforcement learning algorithm. It takes the feedback data of the force sensor and the joint position and speed information of the unloading robot as input, processes them through a deep neural network, and outputs the adjustment strategy of the unloading robot. The adjustment strategy includes motion direction correction and force fine adjustment, so as to realize the flexible operation of simulating manual unloading and correcting the action in real time according to the contact force, thereby enhancing the robot's adaptability to complex environments.
[0091] In some specific embodiments of this application, after controlling the discharge tool to leave the electrolytic furnace, the following steps are also included:
[0092] The unloading robot controls the unloading tool to move along the planned path to the top of the mold; it controls the unloading tool to rotate around the set unloading direction and angle, unloads the material and stops moving; the unloading tool returns to the center after stopping for a set time, ending the unloading process.
[0093] The set pouring direction and angle are used to control the robotic arm to rotate the discharge tool around the rotation center to the set angle position to achieve pouring.
[0094] In some specific embodiments of this application, after determining that material discharge will not continue, the method further includes:
[0095] The unloading robot controls the unloading tool to move to the unloading tool placement position and places the unloading tool; the unloading robot moves to an idle position, inspects the unloading tool, and determines whether the unloading tool is deformed beyond the set size. If so, it is determined that the unloading tool needs to be replaced; otherwise, the process ends.
[0096] Specifically, a vision sensor mounted on the tool rack detects the unloading tool after unloading. The vision sensor takes a picture of the unloading tool and compares the image with a pre-defined CAD ideal model image to determine if the unloading tool needs to be replaced. Specifically, using an RGB-D image as input, a UNet network extracts features and outputs a keypoint heatmap. Post-processing optimization obtains high-precision 3D keypoints. Subsequently, rigid body registration and TPS elastic registration algorithms are used to match the detected point cloud with the CAD ideal model, calculate the keypoint displacement vectors, and quantify the local deformation of the sampling spoon in dimensions such as length, width, and height.
[0097] It should be noted that during the preheating, discharging and unloading processes, the compliant control function is activated to correct the robot trajectory based on the end contact force, ensuring that the tool end maintains no contact or low contact force with the inner wall of the electrolysis furnace.
[0098] Reference Figure 2As shown, firstly, the discharge robot starts and initializes. Then, the electrolytic furnace discharges material to check if the discharge standard is met. If so, the discharge process is prepared. The discharge robot receives the position information of the tool spoon identified by the vision sensor and moves to the tool spoon (discharge tool) position. The vision sensor performs the first verification of the discharge tool. After successful verification, the discharge robot grabs the tool spoon. The force sensor on the discharge robot performs the second verification of the tool spoon. After successful verification, it moves to the preparation position. Next, another vision sensor detects and verifies the furnace opening of the electrolytic furnace. After successful verification, the discharge robot moves the tool spoon to the preheating position for preheating. After preheating is completed, it moves to the preheating completion position and prepares for discharge.
[0099] The second verification also includes controlling the discharge tool to move to the pre-set detection switch position after the force sensor detects the tool's gravity, and touching the switch to determine whether the tool is being held properly.
[0100] Then, the unloading robot moves along the planned path, its tool spoon gradually approaching the bottom of the electrolytic furnace to dig out molten rare earth metal. During the approach, the end effector force sensing system continuously monitors the contact force between the unloading robot and the electrolytic furnace. The algorithm model corrects the robot's trajectory based on the end effector contact force, ensuring that the tool tip maintains no contact or low contact force with the inner wall of the electrolytic furnace. When the tool tip reaches the lowest position set by the trajectory or when the bottom contact force is too large to penetrate further, the unloading robot stops for a period of time and then moves upward (i.e., moves to the unloading completion position after digging). During the movement, the system detects the contact force and adjusts the trajectory according to the algorithm model. Moving to the pouring point, the unloading operation process is executed: the unloading robot moves along the planned path to above the mold, the center of the spoon's bottom rotates around the pre-set tool rotation center to a set angle, begins to pour out the molten rare earth metal, stops for a period of time and then returns to center (i.e., returns to center after pouring), and then moves to the pouring completion position.
[0101] Finally, determine whether to continue discharging. If not, the operation is complete. After completion, the discharging robot moves to the tool rack, places the tool, and returns to its initial position, awaiting the next discharging cycle. Simultaneously, system maintenance: regularly inspect and maintain the discharging robot and discharging tools to ensure long-term stable operation. If the discharging tool deforms beyond the set dimensions or shape, a fault message is sent, and the process ends.
[0102] Throughout the process, operators monitor the operation status in real time through the user interface and intervene manually when necessary.
[0103] Reference Figures 3 to 5 As shown, in a second aspect of this application, an apparatus for controlling the operation of a rare earth metal electrolysis furnace discharge robot is provided, comprising: a support device 100, a robotic arm 200, and a control module 108.
[0104] The support device 100 is used to support the robotic arm 200. The control module 108 is installed on the robotic arm 200 and is used to control the operation of the robotic arm 200. The robotic arm 200 is equipped with a force sensor 110 and a vision sensor to identify the discharge tool 109, the furnace opening position of the electrolytic furnace, and the state of the molten rare earth metal. The force sensor 110 is used to obtain the contact force between the discharge tool 109 and the electrolytic furnace and send it to the control module 108. The control module 108 corrects the running trajectory of the robotic arm 200 by executing a compliant control strategy based on the contact force.
[0105] Specifically, firstly, the visual image processing technology of the vision sensor accurately identifies the tool, furnace opening position, and state of the molten rare earth metal. Secondly, it controls the discharge robot to move to the target position. Finally, based on the end effector force sensor 110 and the set control strategy, it adjusts the target pose of the operation, which not only improves the automation level of rare earth metal electrolysis furnace discharge but also significantly enhances the efficiency and stability of the operation, showing good prospects for industrial application.
[0106] Among them, heat insulation and heat dissipation measures are taken for the electronic components and connecting lines of the force sensor 110; high-temperature resistant materials are selected to create a protective shell, and the circuit layout is optimized to ensure that key components operate stably under the high-temperature radiation of the electrolysis furnace.
[0107] In some specific embodiments of this application, the support device 100 includes an L-shaped track bracket 102 and an L-shaped slide rail 101.
[0108] The L-shaped slide rail 101 is fixed on the L-shaped track bracket 102 and is movable relative to the L-shaped track bracket 102; the robotic arm 200 includes a first robotic arm 103, a second robotic arm 104, a third robotic arm 105, a fourth robotic arm 107 and a fifth robotic arm 106.
[0109] The first robotic arm 103 is slidably mounted on the L-shaped slide rail 101, the second robotic arm 104 is movably mounted on the first robotic arm 103, the third robotic arm 105 is movably mounted on the second robotic arm 104, the fourth robotic arm 107 is rotatably mounted on the third robotic arm 105, and the fifth robotic arm 106 is mounted at one end of the second robotic arm 104 for controlling the movement of the third robotic arm 105 and the fourth robotic arm 107.
[0110] The material discharge tool 109 is gripped by the fourth robotic arm 107, and the force sensor 110 is installed on the fourth robotic arm 107 to obtain the contact force of the material discharge tool 110.
[0111] Reference Figure 3As shown, specifically, by using the L-shaped track bracket 102 as the basic support structure, the installed L-shaped slide rail 101 enables bidirectional movement in both horizontal and vertical directions, i.e. (X-axis and Y-axis movement); the bottom of the first robotic arm 103 is nested in the L-shaped slide rail 101, enabling movement in the X-axis and Y-axis directions, and its upper part is connected to the second robotic arm 104 through a linear guide rail, on which the second robotic arm 104 moves; the third robotic arm 105 is connected to the second robotic arm 104 through a rotary joint, enabling flexible swinging around the axis; the fourth robotic arm 107 is connected to the third robotic arm 105 through another rotary joint, forming the main body of the end effector; the fifth robotic arm 106 is set at the end of the second robotic arm 104, controlling the movement of the third robotic arm 105 and the fourth robotic arm 107 on the Z-axis to ensure the end effector's pose accuracy. During discharge, the first robotic arm 103 quickly positions itself above the target electrolytic furnace along the L-shaped slide rail 101. The second robotic arm 104 to the fourth robotic arm 107 move to deliver the end of the fourth robotic arm 107 to the tool rack, grabbing and removing the material tool 109. The force sensor 110 monitors the grabbing force in real time and feeds it back to the control module 108. Subsequently, the robotic arm 200 assembly inserts the tool into the electrolytic furnace melt through trajectory planning. The force sensor 110 continuously collects contact force data, and the control module 108 dynamically adjusts the torque of each joint to achieve smooth obstacle avoidance and precise material discharge.
[0112] It should be noted that the unloading robot also includes a compliant controller (not shown in the figure). The unloading robot is embedded with a high-precision six-dimensional force sensor 110 to acquire force information of the interaction between the unloading robot and the electrolytic furnace in real time, covering the magnitude and direction of the contact force, and providing data support for subsequent compliant control.
[0113] Among them, the compliant controller is constructed based on the reinforcement learning algorithm. It takes the feedback data from the force sensor 110 and the joint position and speed information of the unloading robot as input, processes them through a deep neural network, generates the adjustment strategy of the unloading robot, and outputs it to the control module 108 of the unloading robot. The adjustment strategy includes motion direction correction and force fine adjustment, realizing flexible operation of real-time correction of action based on contact force when simulating manual unloading, and enhancing the adaptability of the unloading robot to complex environments.
[0114] The preferred features in the above embodiments can be used individually in any embodiment, or in any combination thereof, provided they do not conflict with each other. Furthermore, parts not described in detail in the embodiments can be implemented using existing technologies.
[0115] The foregoing has described some specific embodiments of this application. It should be understood that this application is not limited to the specific embodiments described above, and those skilled in the art can make various modifications or variations within the scope of the claims, which do not affect the substantive content of this application. The above-described preferred features can be used in any combination without conflict.
Claims
1. A control method for a rare earth metal electrolytic furnace discharge robot, characterized in that, include: The discharge robot uses visual sensors to obtain the electrolysis status of the electrolytic furnace and determine whether the discharge standard has been met. After determining that the discharge standard has been met, the discharge robot grabs the discharge tool, preheats the discharge tool, and moves it to the discharge position to prepare for discharge. The unloading robot moves along the planned path, monitors the contact force between the robot and the electrolytic furnace through a force sensing system, and corrects the robot trajectory based on the contact force to execute the unloading process; After the discharge process is completed, it is determined whether to continue discharging; if not, the process ends. The discharge robot acquires the electrolysis state of the electrolytic furnace through a vision sensor and determines whether the discharge standard has been met, including: The visual sensor at the furnace opening position of the electrolytic furnace detects the state image data of the molten rare earth metal in the electrolytic furnace and sends it to the discharge robot. The discharge robot uses a depth algorithm to process the image data of the flame state of molten rare earth metal in the electrolytic furnace, compares it with the set discharge standard image, and determines whether the image data meets the set discharge standard. If it does, it determines that the material can be discharged; otherwise, it continues to acquire image data. After determining that the discharge standard has been met, the discharge robot grabs the discharge tool, preheats the discharge tool, and moves it to the discharge position to prepare for discharge, including: The location of the discharge tool is obtained and sent to the discharge robot, which then moves to the location of the discharge tool and grabs it. After the discharge tool successfully grabs the material, it is moved into the electrolytic furnace for preheating until it reaches the same temperature as the rare earth metals inside the furnace, and then moved to the discharge position to prepare for discharge. The process of acquiring the location of the discharge tool and sending it to the discharge robot, followed by the discharge robot moving to the location of the discharge tool and grabbing it, includes: The position of the discharging tool on the tool rack is obtained and sent to the discharging robot; The unloading robot moves to the position of the unloading tool according to the position of the unloading tool, identifies the unloading tool through the vision sensor, and verifies it with the set unloading tool. After successful verification, the unloading robot grabs the unloading tool. After the material feeding tool is grasped, a force sensor is used to perform a secondary verification of the material feeding tool to detect whether the material feeding tool has been successfully grasped. If so, the material feeding tool is determined to have been successfully grasped, and the material feeding tool is controlled to move to the ready position. If not, the verification continues.
2. The control operation method for a rare earth metal electrolytic furnace discharge robot according to claim 1, characterized in that, After the discharge tool successfully grabs the material, it is moved into the electrolysis furnace for preheating until it reaches the same temperature as the rare earth metals inside the furnace, and then moved to the discharge position to prepare for discharge. This includes: The position of the furnace opening of the electrolytic furnace is obtained by the visual sensor at the furnace opening and then inspected. After successful verification, the discharge tool is placed into the electrolytic furnace for heating, and the temperature of the discharge tool is acquired in real time. It is determined whether the temperature of the discharge tool is consistent with the temperature of the rare earth metal in the electrolytic furnace. If so, the preheating is considered complete; otherwise, heating continues. After the preheating is completed, the discharge tool is moved to the discharge preparation position to prepare for discharge.
3. The control operation method for a rare earth metal electrolytic furnace discharge robot according to claim 1, characterized in that, The unloading robot moves along a planned path, monitors the contact force between the robot and the electrolytic furnace through a force sensing system, and corrects the robot's trajectory based on the contact force to execute the unloading process, including: The discharge robot controls the discharge tool to move along a pre-set path, bringing the discharge tool closer to the bottom of the electrolytic furnace; The force sensor on the discharge robot acquires the contact force between the discharge tool and the electrolytic furnace in real time. Based on the contact force, the algorithm model set on the unloading robot corrects the trajectory of the unloading robot; The position of the discharge tool inside the electrolysis furnace is obtained, and it is determined whether the set position has been reached. If yes, the operation is stopped; otherwise, the operation continues. After determining that the set position has been reached, the discharge robot is controlled to stop running for a set time, and then the discharge tool is controlled to leave the electrolysis furnace.
4. The control operation method for a rare earth metal electrolytic furnace discharge robot according to claim 3, characterized in that, After controlling the discharge tool to leave the electrolytic furnace, the following is also included: The unloading robot controls the unloading tool to move along the planned path to above the mold; The discharge tool is controlled to rotate around a set discharge direction and angle to discharge material and then stop moving. The discharge tool stops at the set time and then returns to its original position, ending the discharge process.
5. The control operation method for a rare earth metal electrolytic furnace discharge robot according to claim 1, characterized in that, After determining that no further material output should be produced, the following also applies: The unloading robot controls the unloading tool to move to the unloading tool placement position and places the unloading tool there; The unloading robot moves to an idle position and inspects the unloading tool to determine whether the unloading tool is deformed beyond the set size. If so, it is determined that the unloading tool needs to be replaced; otherwise, the process ends.
6. An apparatus for controlling the operation of a rare earth metal electrolytic furnace discharge robot according to any one of claims 1-5, characterized in that, include: Support device, robotic arm, and control module; The support device is used to support the robotic arm, and the control module is mounted on the robotic arm to control the operation of the robotic arm; The robotic arm is equipped with a force sensor, which is used to acquire the contact force between the discharge tool and the electrolytic furnace and send it to the control module. The control module uses the contact force to execute a compliant control strategy to correct the running trajectory of the robotic arm.
7. The apparatus according to claim 6, characterized in that, The support device includes a track bracket and an L-shaped slide rail; The L-shaped slide rail is fixed to the rail bracket and is movable relative to the rail bracket; The robotic arm includes a first robotic arm, a second robotic arm, a third robotic arm, a fourth robotic arm, and a fifth robotic arm; The first robotic arm is slidably mounted on the L-shaped slide rail, the second robotic arm is movably mounted on the first robotic arm, the third robotic arm is movably mounted on the second robotic arm, the fourth robotic arm is rotatably mounted on the third robotic arm, and the fifth robotic arm is mounted at one end of the second robotic arm for controlling the movement of the third robotic arm and the fourth robotic arm. The material discharge tool is grasped by the fourth robotic arm, and the force sensor is installed on the fourth robotic arm to obtain the contact force of the material discharge tool.