A machine vision-based waste power battery disassembly device and a control method thereof

Abstract

This invention discloses a machine vision-based dismantling equipment and control method for waste power batteries. The equipment includes a casing dismantling device comprising a first conveying mechanism, a first clamping assembly, a first gantry and a second gantry, a first camera, a bolt recognition model for receiving images captured by the first camera and identifying the bolt positions on the battery cover based on the images, a punching assembly, a suction cup assembly, and a waste cover bin; a battery freezing device comprising a second conveying mechanism, a freezing chamber, and a dry ice nozzle for spraying dry ice to freeze the batteries; a battery cutting device comprising a third conveying mechanism, a second clamping assembly, a third gantry, a second camera, a cell recognition model for receiving images captured by the second camera and identifying the battery cell positions based on the images, and a cutting tool assembly; and a controller connected to the casing dismantling device, the battery freezing device, and the battery cutting device. This invention facilitates the dismantling and recycling of cylindrical batteries.

Description

Technical Field

[0001] This invention relates to the field of lithium battery recycling, and in particular to a machine vision-based dismantling device for waste power batteries and its control method. Background Technology

[0002] The new energy vehicle industry is developing rapidly, but as the electric vehicle market continues to expand, the power batteries of new energy vehicles will also face obsolescence. Theoretically, valuable metal elements such as nickel, cobalt, manganese, and lithium salts can be extracted from retired power batteries, and these materials can be used in the manufacture of new batteries. However, if a large number of retired and scrapped batteries are not properly recycled and disposed of, they will pose serious safety hazards and pollute the environment.

[0003] For retired power batteries, dismantling and recycling, and secondary utilization are two complementary methods widely recognized in the industry. Currently, the dismantling of used power batteries in the market is mainly divided into two categories: traditional manual dismantling and mechanical dismantling. Traditional manual dismantling involves manually breaking down the battery and recycling materials. This method typically requires a large amount of manpower and time, is inefficient, and poses certain safety risks. Mechanical dismantling utilizes automated and mechanized equipment to dismantle used power batteries; mechanical dismantling methods can effectively improve dismantling efficiency and safety, reducing human intervention, but may require specific equipment and processes for different types of batteries.

[0004] For example, common battery types in new energy vehicles include cylindrical batteries and blade batteries, which differ in structure: Cylindrical batteries: Cylindrical batteries are one of the most commonly used battery types. They are cylindrical in shape, similar to common pen batteries. Cylindrical batteries typically consist of one or more individual cells, and in vehicles, multiple cells are usually connected in series to achieve the required voltage and capacity. Blade batteries: Blade batteries, also known as thin batteries or corrugated paper batteries, are flat rectangular or blade-shaped. The design of blade batteries allows for the stacking of individual cells and electrical connection via current collectors.

[0005] Therefore, the above problems urgently need to be solved. Summary of the Invention

[0006] Purpose of the invention: The first purpose of this invention is to provide a machine vision-based waste power battery dismantling and recycling device for cylindrical batteries.

[0007] The second objective of this invention is to provide a control method for a waste power battery dismantling equipment based on machine vision.

[0008] Technical Solution: To achieve the above objectives, this invention discloses a machine vision-based dismantling device for waste power batteries, comprising:

[0009] The casing disassembly device includes a first conveying mechanism, a clamping assembly located on the first conveying mechanism for clamping the battery, a first gantry and a second gantry erected sequentially above the first conveying mechanism, a first camera located on the first gantry and facing the first conveyor belt, a bolt recognition model for receiving images captured by the first camera and recognizing the bolt positions on the battery cover based on the images, a punching assembly located on the first gantry for punching holes at the bolt positions, a suction cup assembly located on the second gantry for picking up the battery cover, and a waste cover box for collecting the battery cover.

[0010] A battery freezing device includes a second conveying mechanism connected to a first conveying mechanism, a freezing chamber located on the second conveying mechanism, and a dry ice nozzle located in the freezing chamber for spraying dry ice to freeze the battery.

[0011] The battery cutting device includes a third conveying mechanism connected to the second conveying mechanism, a clamping assembly located on the third conveying mechanism for clamping the battery, a third gantry mounted on the third conveying mechanism, a second camera located on the third gantry and facing the third conveyor belt, a cell recognition model for receiving images captured by the second camera and recognizing the position of the battery cell based on the images, and a cutting tool assembly for cutting the battery cell.

[0012] The controller is connected to the casing disassembly device, the battery freezing device, and the battery cutting device.

[0013] The punching assembly includes a first fixed base connected to the first gantry frame, an upper fixed base located on the first fixed base, a main connecting rod evenly distributed around the upper fixed base, a suction cup located at the lower end of the main connecting rod, a punching cylinder located at the center of the upper fixed base, a lower fixed base sleeved on the front end of the punching cylinder, a hydraulic support rod with one end connected to the lower fixed base and the other end connected to the main connecting rod, and a punch located at the end of the punching cylinder.

[0014] Preferably, the suction cup assembly includes a second fixed base connected to the second gantry, a fixed rod connected to the second fixed base, and a double-layer suction cup connected to the fixed rod by an adjustable bolt.

[0015] Furthermore, the clamping assembly includes a mounting base connected to the first or third transmission mechanism, a cylinder located on the mounting base, a push rod connected to the extended end of the cylinder, a left clamping plate located at the front end of one push rod, and a right clamping plate located at the front end of the other push rod.

[0016] Furthermore, the tool assembly includes a third fixed base connected to the third gantry, a first connecting rod connected to the third fixed base and symmetrically arranged, a second connecting rod and a third connecting rod sequentially connected to the first connecting rod, a tool connected to the end of the third connecting rod, an angle adjusting connecting rod whose two ends are respectively connected to the ends of the left and right third connecting rods, a tool motor located on the third fixed base, and an angle adjusting bolt whose one end is connected to the output shaft of the tool motor and whose other end passes through the angle adjusting connecting rod.

[0017] Preferably, the first gantry includes a first column, a first horizontal beam arranged in parallel to the left and right, a first X-axis guide rail located on the first horizontal beam, a first longitudinal beam that can move back and forth along the first X-axis guide rail, and a first Y-axis guide rail located on the first longitudinal beam, wherein the punching assembly is located on the first Y-axis guide rail and can move back and forth along the first Y-axis guide rail.

[0018] Furthermore, the second gantry includes a second column, a second horizontal beam arranged in parallel to the left and right, a second longitudinal beam connected to the second horizontal beam, a second Y-guide rail located on the second longitudinal beam, and a suction cup vertical rod that can move back and forth along the second Y-guide rail and is used to connect the suction cup assembly.

[0019] Furthermore, the third gantry includes a third column, a third longitudinal beam arranged in parallel front and rear, a third Y-axis guide rail located on the third longitudinal beam, a third crossbeam that can move back and forth along the third Y-axis guide rail, a third X-axis guide rail located on the third crossbeam, and a camera crossbeam mounted on the third longitudinal beam for mounting a second camera, wherein the tool assembly is located on the third X-axis guide rail and can move back and forth along the third X-axis guide rail.

[0020] Preferably, both the bolt identification model and the cell identification model adopt the YOLOv5 framework, which includes a backbone component, a neck component, and a prediction component.

[0021] This invention discloses a control method for a waste power battery dismantling equipment based on machine vision, comprising the following steps:

[0022] The first conveyor mechanism is activated to transport used power batteries. The first camera captures real-time images of the used power batteries transported by the first conveyor mechanism. When the used power batteries reach the center position of the clamping assembly, the controller controls the first conveyor mechanism to pause. The clamping assembly clamps the used power batteries. The first camera continuously captures images of the used power batteries and transmits them to the bolt recognition module. The bolt recognition module identifies and marks the bolt positions and outputs the bolt position information to the controller. The controller controls the punching assembly to remove the bolts from the top cover of the used power batteries based on the bolt position information. The first conveyor mechanism is then activated to transport the used power batteries to the area below the suction cup assembly. The suction cup assembly is then controlled to pick up the top cover of the used power batteries and place it in the corresponding waste cover box.

[0023] The used power batteries with the top cover removed are transported to the refrigeration chamber through the second conveyor mechanism. The second conveyor mechanism is paused, and the dry ice nozzle is controlled to spray dry ice to cool and freeze the batteries. Then the second conveyor mechanism is restarted to transfer the used power batteries to the third conveyor mechanism.

[0024] The second camera captures real-time images of the frozen waste power batteries on the third conveyor mechanism. When the frozen waste power batteries reach the center position of the clamping assembly, the controller controls the third conveyor mechanism to pause, and the clamping assembly clamps the frozen waste power batteries. The second camera continuously captures images of the frozen waste power batteries and transmits them to the cell identification module. The cell identification module identifies and marks the position of the battery cells and outputs the cell position information to the controller. The controller controls the cutting tool assembly to cut and grab the battery cells based on the cell position information.

[0025] Beneficial Effects: Compared with existing technologies, this invention has the following significant advantages: By combining a casing disassembly device, a battery freezing device, and a battery cutting device, this invention achieves an efficient and safe disassembly and processing process for cylindrical batteries. The system offers advantages such as automated operation, precise identification and positioning, and material recycling, contributing to improved processing efficiency, resource recovery, and environmental protection of waste batteries. Attached Figure Description

[0026] Figure 1 This is a schematic diagram of the structure of the present invention;

[0027] Figure 2 This is a schematic diagram of the outer shell disassembly device in this invention;

[0028] Figure 3 This is a schematic diagram of the punching assembly in this invention;

[0029] Figure 4 This is a schematic diagram of the structure of the first clamping assembly in this invention;

[0030] Figure 5 This is a schematic diagram of the suction cup assembly in this invention;

[0031] Figure 6 This is a schematic diagram of the battery refrigeration device in this invention;

[0032] Figure 7 This is a schematic diagram of the battery cutting device in this invention;

[0033] Figure 8 This is a schematic diagram of the tool assembly in this invention;

[0034] Figure 9 This is a flowchart of bolt position identification in this invention;

[0035] Figure 10This is a flowchart of the cell location identification process in this invention;

[0036] Figure 11 This is a flowchart of the bolt identification model in this invention;

[0037] Figure 12 This is a flowchart of the battery cell identification model in this invention;

[0038] Figure 13 This is a schematic diagram of the YOLOv5 framework in this invention;

[0039] Figure 14 This is a schematic diagram of YOLOv5 framework training in this invention. Detailed Implementation

[0040] The technical solution of the present invention will be further described below with reference to the accompanying drawings.

[0041] like Figure 1 As shown, the present invention discloses a machine vision-based waste power battery dismantling device, including a shell dismantling device 100, a battery freezing device 200, a battery cutting device 300 and a controller, which are respectively connected to the shell dismantling device 100, the battery freezing device 200 and the battery cutting device 300.

[0042] like Figure 2 As shown, the casing disassembly device 100 includes a first conveying mechanism 110, a punching assembly 120, a first gantry frame 130, a first camera 140, a first clamping assembly 150, a second gantry frame 160, a suction cup assembly 170, and a waste cover box 180. The first conveying mechanism 110 can be a conveyor belt, such as... Figure 3 As shown, the punching assembly 120 is located on the first gantry 130. The punching assembly 120 includes a suction cup 121, a main connecting rod 122, an upper fixed base 123, a first fixed seat 124, a punching cylinder 125, a hydraulic support rod 126, a punching punch 127, and a lower fixed base 128. The first fixed seat 124 is connected to the first gantry 130, and the upper fixed base 123 is located on the first fixed base 124. Multiple main connecting rods 122 are evenly distributed around the upper fixed base 123. The suction cup 121 is located at the lower end of the main connecting rod 122 and sucks the upper cover plate of the battery to facilitate punching. The punching cylinder 125 is located at the center of the upper fixed base 123, and the lower fixed base 128 is sleeved on the front end of the punching cylinder 125. Each main connecting rod 122 is equipped with a hydraulic support rod 126, one end of which is connected to the lower fixed base 128 and the other end to the main connecting rod 122. The punching punch 127 is located at the end of the punching cylinder 125. The hydraulic support rod 126 controls the opening and closing angle of the main connecting rod. When operation begins, the hydraulic support rod 126 first adjusts the opening and closing angle to allow the suction cup 121 to adhere to the battery casing, thereby stabilizing the punching mechanism.

[0043] The first gantry 130 and the second gantry 160 are sequentially mounted above the first conveying mechanism 110. The first gantry 130 includes a first column, a first crossbeam, a first X-axis guide rail, a first longitudinal beam, and a first Y-axis guide rail. The two first crossbeams are arranged parallel to each other. The first X-axis guide rail is located on the first crossbeam, and the first longitudinal beam can move back and forth along the first X-axis guide rail. The first Y-axis guide rail is located on the first longitudinal beam, and the punching assembly 120 is located on the first Y-axis guide rail and can move back and forth along the first Y-axis guide rail. Figure 4 As shown, the first clamping assembly 150 is located on the first conveying mechanism 110 and is used to clamp the battery. The first clamping assembly 150 includes a mounting base 151, a left clamping plate 152, a right clamping plate 153, a push rod 154, and a cylinder 155. The mounting base 151 is connected to the first conveying mechanism, the cylinder 155 is located on the mounting base 151, the push rod 154 is connected to the extended end of the cylinder 155, the left clamping plate 152 is located at the front end of one side of the push rod 154, and the right clamping plate 153 is located at the front end of the other side of the push rod 154. When operation begins, the push rod 154 on the cylinder 155 pushes the left clamping plate 152 and the right clamping plate 153 to complete the clamping of the battery pack.

[0044] The second gantry 160 includes a second column, a second horizontal beam, a second longitudinal beam, a second Y-guide rail, and a suction cup vertical rod. The two second horizontal beams are arranged parallel to each other. The two ends of the second longitudinal beam are connected to the two second horizontal beams respectively. The second Y-guide rail is located on the second longitudinal beam. The suction cup vertical rod can move back and forth along the second Y-guide rail and is used to connect to the suction cup assembly 170. Figure 5 As shown, the suction cup assembly 170 includes a double-layer suction cup 171, an adjustable bolt 172, a fixing rod 173, and a second fixing seat 174. The second fixing seat 174 is connected to the second gantry frame, and the fixing rod 173 is connected to the second fixing seat 174. The double-layer suction cup 171 is connected to the fixing rod 173 through the adjustable bolt 172.

[0045] A first camera 140 is located on a first gantry 130 and directly faces the center of the first clamping assembly on the first conveyor belt. A bolt recognition model receives images captured by the first camera 140 and identifies the bolt positions on the battery cover based on these images. A punching assembly 120 is located on the first gantry 130 and is used to punch holes at the bolt positions. A suction cup assembly 170 is located on a second gantry 160 and is used to pick up the battery cover. A waste cover bin 180 collects the battery cover. The bolt recognition model uses the YOLOv5 framework, which includes a backbone assembly, a neck assembly, and a prediction assembly. Figure 9As shown, when the used power battery approaches the first camera, the first camera detects its position. When it reaches the center of the first clamping assembly, it controls the first conveying mechanism to stop, the first clamping assembly clamps the battery, and the first camera takes five photos of the used power battery. The bolt recognition model marks and scores the bolt positions in the five photos, selects the position information with the highest confidence, and transmits the position information to the system. The punching assembly on the first gantry uses the position information from the system to remove the bolts from the outer casing. Figure 11 As shown, the processing method first removes shake from the captured photos and performs foreground detection. During the detection process, shadows and light are suppressed to improve image quality. Then, the battery pack is detected first, followed by the identification of bolt positions, and the position information is uploaded to the system.

[0046] The battery refrigeration device 200 includes a second conveying mechanism, a refrigeration chamber, and a dry ice nozzle 230. The second conveying mechanism is connected to the first conveying mechanism, such as... Figure 6 As shown, the second conveying mechanism is a roller conveyor structure 260, which is driven by a sprocket conveyor structure 250. The freezer compartment includes a left compartment 210, a top cover 220, and a right compartment 240. Batteries with the top cover removed are conveyed by the roller conveyor structure 260 into the freezer compartment consisting of the left compartment 210, the top cover 220, and the right compartment 240. A dry ice nozzle 230 is installed on the top cover 220, which sprays dry ice to cool and freeze the polyurethane foam on the surface of the battery cells. The dry ice nozzle 230 is connected to a device for storing dry ice. The freezing process makes the polyurethane foam brittle, facilitating subsequent cutting and disassembly operations.

[0047] like Figure 7 As shown, the battery cutting device 300 includes a third conveying mechanism 350, a second clamping assembly 340, a third gantry 320, a second camera 330, a tool assembly 310, and a cell identification model. The third conveying mechanism 350 is connected to the second conveying mechanism and can be a conveyor belt type conveying mechanism. The second clamping assembly 340 is located on the third conveying mechanism and is used to clamp the battery. The second clamping assembly 340 includes a mounting base 151, a left clamping plate 152, a right clamping plate 153, a push rod 154, and a cylinder 155. The mounting base 151 is connected to the first conveying mechanism, the cylinder 155 is located on the mounting base 151, the push rod 154 is connected to the extended end of the cylinder 155, the left clamping plate 152 is located at the front end of one side of the push rod 154, and the right clamping plate 153 is located at the front end of the other side of the push rod 154. After being clamped by the second clamping component 340, the frozen battery is first identified and its cell position is located by the cell identification model. Then it is cut by the cutting tool component 310 and clamped and placed into the storage box. The cutting tool component 310 can flexibly adjust its opening angle.

[0048] The third gantry 320 is mounted on the third conveyor mechanism. The third gantry 320 includes a third column, a third longitudinal beam, a third Y-axis guide rail, a third crossbeam, a third X-axis guide rail, and a camera crossbeam. The third longitudinal beam is arranged parallel to each other. The third Y-axis guide rail is located on the third longitudinal beam. The third crossbeam can move back and forth along the third Y-axis guide rail. The third X-axis guide rail is located on the third crossbeam. The camera crossbeam is mounted on the third longitudinal beam and is used to mount a second camera. The cutting tool assembly is located on the third X-axis guide rail and can move back and forth along it. The second camera 330 is located on the third gantry and faces the center of the second clamping assembly on the third conveyor belt. The battery cell recognition model is used to receive images captured by the second camera and identify the position of the battery cells based on the images. The cutting tool assembly 310 is used to cut the battery cells. Figure 8 As shown, the tool assembly 310 includes a tool 311, a third connecting rod 312, a second connecting rod 313, a first connecting rod 314, a tool motor 315, an angle adjusting bolt 316, an angle adjusting connecting rod 317, and a third fixed base 318. The third fixed base 318 is connected to the third gantry frame, and the first connecting rod 314 is connected to the third fixed base. The first connecting rod 314, the second connecting rod 313, and the third connecting rod 312 are connected sequentially and symmetrically arranged on the third fixed base 318. The tool 311 is connected to the end of the third connecting rod 312. The two ends of the angle adjusting connecting rod 317 are connected to the ends of the left and right third connecting rods, respectively. The tool motor 315 is located on the third fixed base. One end of the angle adjusting bolt 316 is connected to the output shaft of the tool motor, and the other end passes through the angle adjusting connecting rod. During operation, the opening and closing angle of the tool is adjusted by the angle adjusting bolt 316 and the adjusting connecting rod 317 to achieve battery cutting.

[0049] The battery cell identification model uses the YOLOv5 framework, which includes backbone, neck, and prediction components, such as... Figure 10 As shown, when the cooled and frozen battery approaches the second camera, the second camera detects its position. When it reaches the center of the second clamping assembly, it controls the third conveying mechanism to stop, the second clamping assembly clamps the battery, and the second camera takes five photos of the cooled and frozen battery. The cell recognition model marks and scores the positions of the battery cells in the five photos, selects the position with the highest confidence level, and transmits this position information to the system. The cutting tool assembly on the third gantry, relying on the system's position information, cuts and clamps the battery cells, placing them into a storage box. Figure 12 As shown, the process first involves shaving the captured photos and performing foreground detection. During the detection process, shadows and light are suppressed to improve image quality. Then, the battery pack is detected first, followed by the identification of the battery cell positions, and the position information is uploaded to the system.

[0050] like Figure 13 As shown, YOLOv5 is a single-segment visual detection algorithm characterized by its high speed and accuracy. The Backbone component is primarily responsible for image feature extraction, including a Focus operation. This operation reduces the number of parameters, network layers, gradients, and computations. The feature extraction part consists of a D-ELAN module, which is a highly efficient feature extraction network composed of CBS layers of varying sizes. The CBS layer comprises Conv layers, BN layers, and the SiLU activation function. Conv is mainly responsible for feature extraction, BN layers accelerate convergence and prevent gradient explosion, and SiLU is an activation function that introduces non-linearity to aid optimization. Finally, the SPP module, a spatial pyramid pooling structure, effectively avoids image distortion caused by image region cropping and scaling, and solves the problem of repetitive feature extraction related to graphs in neural networks, significantly improving the speed of candidate box generation and saving computational costs. The Neck component is responsible for multi-scale feature fusion of the feature maps and passing these features to the prediction layer. It includes the D-ELAN-A module, which is based on the D-ELAN module but with some parameter adjustments to better suit the application requirements of the Neck. Finally, the Prediction component performs the final regression prediction to achieve the purpose of classification and detection. Figure 14 As shown, the algorithm training process is explained as follows: First, take photos of the battery and label the positions of the bolts or the positions of the battery cells in the photos. Divide the data into a validation set and a dataset at a ratio of 1:4. Use the dataset to train the algorithm, and use the validation set to verify the algorithm's accuracy and detection speed as a reference. Finally, save the model with the highest accuracy for actual production.

[0051] This invention discloses a control method for a waste power battery dismantling equipment based on machine vision, comprising the following steps:

[0052] The first conveyor mechanism is activated to transport used power batteries. The first camera captures real-time images of the used power batteries transported by the first conveyor mechanism. When the used power batteries reach the center position of the clamping assembly, the controller controls the first conveyor mechanism to pause. The clamping assembly clamps the used power batteries. The first camera continuously captures images of the used power batteries and transmits them to the bolt recognition module. The bolt recognition module identifies and marks the bolt positions and outputs the bolt position information to the controller. The controller controls the punching assembly to remove the bolts from the top cover of the used power batteries based on the bolt position information. The first conveyor mechanism is then activated to transport the used power batteries to the area below the suction cup assembly. The suction cup assembly is then controlled to pick up the top cover of the used power batteries and place it in the corresponding waste cover box.

[0053] The used power batteries with the top cover removed are transported to the refrigeration chamber through the second conveyor mechanism. The second conveyor mechanism is paused, and the dry ice nozzle is controlled to spray dry ice to cool and freeze the batteries. Then the second conveyor mechanism is restarted to transfer the used power batteries to the third conveyor mechanism.

[0054] The second camera captures real-time images of the frozen waste power batteries on the third conveyor mechanism. When the frozen waste power batteries reach the center position of the clamping assembly, the controller controls the third conveyor mechanism to pause, and the clamping assembly clamps the frozen waste power batteries. The second camera continuously captures images of the frozen waste power batteries and transmits them to the cell identification module. The cell identification module identifies and marks the position of the battery cells and outputs the cell position information to the controller. The controller controls the cutting tool assembly to cut and grab the battery cells based on the cell position information.

[0055] The preferred embodiments of the present invention have been described in detail above, but the design concept of the present invention is not limited thereto. Within the scope of the technical concept of the present invention, various equivalent transformations can be made to the technical solution of the present invention, and these equivalent transformations all fall within the protection scope of the present invention.

Claims

1. A machine vision-based dismantling device for waste power batteries, characterized in that, include: The casing disassembly device (100) includes a first conveying mechanism (110), a first clamping assembly (150) located on the first conveying mechanism and used to clamp the battery, a first gantry frame (130) and a second gantry frame (160) erected sequentially above the first conveying mechanism (110), a first camera (140) located on the first gantry frame (130) and facing the first conveying mechanism (110), a bolt recognition model for receiving images captured by the first camera (140) and recognizing the bolt positions on the battery cover based on the images, a punching assembly (120) located on the first gantry frame and used to punch holes at the bolt positions, a suction cup assembly (170) located on the second gantry frame for picking up the battery cover, and a waste cover box (180) for collecting the battery cover. The punching assembly (120) includes a first fixed base (124) connected to the first gantry (130), an upper fixed base (123) located on the first fixed base, a main connecting rod (122) evenly distributed around the upper fixed base, a suction cup (121) located at the lower end of the main connecting rod, a punching cylinder (125) located at the center of the upper fixed base, a lower fixed base (128) sleeved on the front end of the punching cylinder, a hydraulic support rod (126) with one end connected to the lower fixed base and the other end connected to the main connecting rod, and a punching punch (127) located at the end of the punching cylinder. The battery freezing device (200) includes a second conveying mechanism connected to the first conveying mechanism, a freezing chamber located on the second conveying mechanism, and a dry ice nozzle (230) located in the freezing chamber for spraying dry ice to freeze the battery. The battery cutting device (300) includes a third conveying mechanism (350) connected to the second conveying mechanism, a second clamping assembly (340) located on the third conveying mechanism and used to clamp the battery, a third gantry (320) mounted on the third conveying mechanism, a second camera (330) located on the third gantry and facing the third conveyor belt, a cell recognition model for receiving images captured by the second camera and recognizing the position of the battery cell based on the images, and a cutting tool assembly (310) for cutting the battery cell. The tool assembly (310) includes a third fixed base (318) connected to the third gantry, a first connecting rod (314) connected to the third fixed base and symmetrically arranged, a second connecting rod (313) and a third connecting rod (312) connected in sequence to the first connecting rod, a tool (311) connected to the end of the third connecting rod, an angle adjusting connecting rod (317) with both ends connected to the ends of the left and right third connecting rods respectively, a tool motor (315) located on the third fixed base, and an angle adjusting bolt (316) with one end connected to the output shaft of the tool motor and the other end passing through the angle adjusting connecting rod. The controller is connected to the housing disassembly device (100), the battery freezing device (200), and the battery cutting device (300), respectively.

2. The machine vision-based waste power battery dismantling equipment according to claim 1, characterized in that: The suction cup assembly (170) includes a second fixed base (174) connected to the second gantry, a fixed rod (173) connected to the second fixed base, and a double-layer suction cup (171) connected to the fixed rod by an adjustable bolt (172).

3. The machine vision-based waste power battery dismantling equipment according to claim 1, characterized in that: The first clamping assembly (150) and the second clamping assembly (340) both include a mounting base (151), a cylinder (155) located on the mounting base, a push rod (154) connected to the extended end of the cylinder, a left clamping plate (152) located at the front end of the push rod on one side, and a right clamping plate (153) located at the front end of the push rod on the other side. The mounting base of the first clamping assembly (150) is connected to the first conveying mechanism, and the mounting base of the second clamping assembly (340) is connected to the third conveying mechanism.

4. The machine vision-based waste power battery dismantling equipment according to claim 1, characterized in that: The first gantry (130) includes a first column, a first horizontal beam arranged in parallel to the left and right, a first X-guide rail located on the first horizontal beam, a first longitudinal beam that can move back and forth along the first X-guide rail, and a first Y-guide rail located on the first longitudinal beam, wherein the punching assembly is located on the first Y-guide rail and can move back and forth along the first Y-guide rail.

5. The machine vision-based waste power battery dismantling equipment according to claim 1, characterized in that: The second gantry (160) includes a second column, a second horizontal beam arranged in parallel to the left and right, a second longitudinal beam connected to the second horizontal beam, a second Y-guide rail located on the second longitudinal beam, and a suction cup vertical rod that can move back and forth along the second Y-guide rail and is used to connect the suction cup assembly.

6. The machine vision-based waste power battery dismantling equipment according to claim 1, characterized in that: The third gantry (320) includes a third column, a third longitudinal beam arranged in parallel front and rear, a third Y-guide rail located on the third longitudinal beam, a third crossbeam that can move back and forth along the third Y-guide rail, a third X-guide rail located on the third crossbeam, and a camera crossbeam mounted on the third longitudinal beam for mounting a second camera, wherein the tool assembly is located on the third X-guide rail and can move back and forth along the third X-guide rail.

7. The machine vision-based waste power battery dismantling equipment according to claim 1, characterized in that: Both the bolt identification model and the cell identification model adopt the YOLOv5 framework, which includes a backbone component, a Neck component, and a Prediction component.

8. A control method for a machine vision-based waste power battery dismantling equipment according to any one of claims 1 to 7, characterized in that, Includes the following steps: The first conveyor mechanism is activated to transport used power batteries. The first camera captures real-time images of the used power batteries transported by the first conveyor mechanism. When the used power batteries reach the center position of the clamping assembly, the controller controls the first conveyor mechanism to pause. The clamping assembly clamps the used power batteries. The first camera continuously captures images of the used power batteries and transmits them to the bolt recognition module. The bolt recognition module identifies and marks the bolt positions and outputs the bolt position information to the controller. The controller controls the punching assembly to remove the bolts from the top cover of the used power batteries based on the bolt position information. The first conveyor mechanism is then activated to transport the used power batteries to the area below the suction cup assembly. The suction cup assembly is then controlled to pick up the top cover of the used power batteries and place it in the corresponding waste cover box. The used power batteries with the top cover removed are transported to the refrigeration chamber through the second conveyor mechanism. The second conveyor mechanism is paused, and the dry ice nozzle is controlled to spray dry ice to cool and freeze the batteries. Then the second conveyor mechanism is restarted to transfer the used power batteries to the third conveyor mechanism. The second camera captures real-time images of the frozen waste power batteries on the third conveyor mechanism. When the frozen waste power batteries reach the center position of the clamping assembly, the controller controls the third conveyor mechanism to pause, and the clamping assembly clamps the frozen waste power batteries. The second camera continuously captures images of the frozen waste power batteries and transmits them to the cell identification module. The cell identification module identifies and marks the position of the battery cells and outputs the cell position information to the controller. The controller controls the cutting tool assembly to cut and grab the battery cells based on the cell position information.

Images