Photovoltaic module disassembling system and method based on cooperation of multiple mechanical arms and intelligent sensing

Abstract

The invention discloses a photovoltaic module disassembling system and method based on cooperation of multiple mechanical arms and intelligent perception, and the system is characterized in that a mechanical arm autonomous grabbing and sorting module completes the autonomous grabbing and sorting of stacked photovoltaic modules through mechanical arms, and enables the back surfaces of the photovoltaic modules to face upwards; the junction box disassembling module is used for separating the junction box on the back surface of the photovoltaic module and tidying the disassembled junction box; the aluminum frame disassembling module completely separates an aluminum frame from a back plate, the multi-mechanical-arm collaborative shearing and classifying module conducts collaborative grabbing and shearing on the aluminum frame, complete disassembling of the aluminum frame is completed, and the photovoltaic back plate and the aluminum frame which are disassembled are recognized and sorted to the corresponding positions through the sorting mechanical arm vision module. According to the scheme, based on cooperation of multiple mechanical arms, disassembly of a back plate, a frame and a junction box of the assembly is efficiently completed, compared with a traditional manual production line, the sorting and disassembly efficiency is improved by 5 times, and meanwhile the accident rate is reduced by 100% compared with manual operation.

Description

Technical Field

[0001] The present invention belongs to the field of photovoltaic module recycling, and specifically relates to a photovoltaic module disassembly system and method based on multi-robotic arm collaboration and intelligent perception. Background Art

[0002] Against the backdrop of profound changes in the global energy system, the overexploitation and consumption of traditional fossil fuels has led to increasingly severe resource shortages. Crystalline silicon photovoltaic power generation technology, with its unique advantages, stands out among numerous clean energy technologies. Global demand for photovoltaic power continues to grow by more than 20% annually, gradually transitioning from a supplementary energy source to a primary one.

[0003] However, with the continuous development of photovoltaic power generation, the number of discarded crystalline silicon photovoltaic cell modules has increased dramatically, and the problem of waste pollution has gradually become prominent. my country's photovoltaic module scrapping peaks in 2035, and by 2050, the amount of scrapped photovoltaic panels is expected to reach 20 million tons. However, discarded modules have high recycling value, and their recycling can significantly reduce environmental pollution. Secondary use can also reduce unnecessary mining of the resources needed for them, further protecting the ecological environment.

[0004] Photovoltaic modules are mainly composed of key parts such as laminates, aluminum frames and junction boxes. Different parts have their own recycling value according to their composition, so it is very important to separate different parts for processing.

[0005] In the field of recycling retired photovoltaic modules, the Chinese market is still in its infancy. Existing technologies mostly focus on component recovery, using chemical or thermal methods to recycle the different components of the backsheet. However, the disassembly process still uses traditional disassembly methods that rely on manual or semi-automatic tools, resulting in low efficiency (manual disassembly accounts for more than 80%), low material recovery rates (silicon wafer recovery rates are less than 90%), and poor safety. Existing technologies, such as some that use semi-automatic disassembly and rely on handheld tools, have high flexibility but limited efficiency; physical separation methods require multiple steps and cannot handle non-standardized modules. As a result, the recycling efficiency in the actual production process is reduced, the overall disassembly cost is high, and it cannot be put into use well. Therefore, a complete intelligent disassembly system is needed to further improve efficiency through intelligent and automatic disassembly. Summary of the Invention

[0006] In response to the above problems, the purpose of the present invention is to provide a photovoltaic module disassembly system and method based on multi-robotic arm collaboration and intelligent perception.

[0007] The specific technical solutions for achieving the purpose of the present invention are as follows:

[0008] A photovoltaic module disassembly system based on multi-manipulator collaboration and intelligent perception, including a robot arm autonomous grasping and sorting module, a junction box disassembly module, an aluminum frame disassembly module, and a multi-manipulator collaborative shearing and sorting module;

[0009] The robotic arm autonomous grasping and sorting module uses the robotic arm to identify the stacked photovoltaic modules through a visual recognition model and a deep learning recognition model, and then uses multiple robotic arms to collaboratively complete the autonomous grasping and sorting of the stacked photovoltaic modules and place the photovoltaic modules with the back side facing upwards;

[0010] The junction box disassembly module uses a double spatula to separate the junction box on the back of the photovoltaic module and put the disassembled junction box into place;

[0011] The aluminum frame disassembly module determines the center position of the back of the photovoltaic module through a visual recognition model, and then uses an electric push rod device to fit the back plate of the photovoltaic module. The electric push rod synchronously pushes out the frame and disassembles the aluminum frame through thrust, so that the aluminum frame and the back plate are completely separated;

[0012] The multi-robot collaborative shearing and classification module uses the visual perception model on the robot arm to identify the aluminum frame. After the identification is completed, two sorting robot arms are used to collaboratively grab the aluminum frame, and then the cutting robot arm is used to cut the corners of the aluminum frame to complete the disassembly of the aluminum frame. Finally, the disassembled photovoltaic backplane and aluminum frame are identified by the sorting robot arm visual module and sorted to the corresponding position.

[0013] Furthermore, the robotic arm autonomous grasping and sorting module includes a visual recognition unit, a path planning unit and a collaborative grasping unit;

[0014] The visual recognition unit uses a pre-trained convolutional neural network to identify the photovoltaic module backsheet stacked on the top layer and select its complete outline;

[0015] The path planning unit plans the motion path, grasping position and angle of the robot arm according to the frame selection position of the visual recognition unit;

[0016] The collaborative grasping unit adopts multiple robotic arms to collaboratively grasp the photovoltaic module backboard when the backboard size exceeds a threshold or the bending degree is greater than a threshold.

[0017] Furthermore, the junction box disassembly module includes a double blade unit and a push rod unit;

[0018] The double scraper units are arranged horizontally and symmetrically, and cut inward from both sides of the photovoltaic module junction box to sever the connection between the junction box and the photovoltaic module backplane, and the push rod unit pushes the disassembled junction box to the specified position.

[0019] Furthermore, the aluminum frame disassembly module includes a visual recognition unit, a synchronous motor unit and a robust control unit;

[0020] The visual recognition unit determines the center position of the photovoltaic module back plate through the visual recognition model and adjusts the electric push rod device to the aluminum frame;

[0021] The synchronous motor unit controls multiple sets of electric push rods to push against the frame in multiple directions to perform thrust disassembly;

[0022] The robust control unit adopts a robust controller based on a nonlinear disturbance observer to compensate for external disturbances during the disassembly process.

[0023] Furthermore, the multi-manipulator collaborative shearing and sorting module includes a collaborative shearing unit and a sorting and recycling unit;

[0024] The collaborative shearing unit controls the sorting robot arm to grab the aluminum frame and controls the shearing robot arm to shear from the corner;

[0025] The classification and recycling unit identifies the type of sheared material through a trained convolutional neural network and sorts it to a designated location.

[0026] The present invention also provides a photovoltaic module disassembly method based on multi-manipulator collaboration and intelligent perception, comprising the following steps:

[0027] Step 1: Autonomous grasping and sorting by the robotic arm: Using a visual recognition model, the robotic arm identifies and perceives the stacked photovoltaic modules. Multiple robotic arms collaborate to autonomously grasp and sort the stacked photovoltaic modules, placing the modules with the backs of the photovoltaic panels facing upwards on the conveyor belt.

[0028] Step 2: Disassemble the junction box of the photovoltaic module. Use a double spatula to separate the junction box on the back of the photovoltaic module and put the disassembled junction box into place.

[0029] Step 3: Dismantle the aluminum frame. Use the visual recognition model to determine the center position of the back of the photovoltaic module. Then use the electric push rod device to fit the back plate of the photovoltaic module. Then use the electric push rod to synchronously push out the frame and dismantle the aluminum frame through the thrust, so that the aluminum frame and the back plate are completely separated.

[0030] Step 4: Multiple robotic arms collaborate in shearing and autonomous planning and classification. The visual perception model on the robotic arm is used to identify the aluminum frame. After identification, two sorting robotic arms are used to collaboratively grab the aluminum frame. The cutting robotic arm then cuts the corners of the aluminum frame to complete the disassembly of the aluminum frame. Finally, the disassembled photovoltaic backsheet and aluminum frame are identified by the visual module of the sorting robotic arm and sorted to the corresponding positions.

[0031] Compared with the prior art, the present invention has the following beneficial effects:

[0032] (1) The solution of the present invention is aimed at the pre-disassembly task of retired photovoltaic modules. Based on a multi-manipulator collaborative operation system, it innovatively proposes a pre-disassembly production line solution and an intelligent multi-task collaborative planning method. Through a task scheduling algorithm based on heuristic search and an improved trajectory planning algorithm, it achieves efficient collaboration between the manipulator and the disassembly equipment in task allocation and path planning, and efficiently completes the disassembly of the module backplane, frame and junction box.

[0033] (2) The solution of the present invention breaks through the limitations of traditional single-visual recognition, innovatively integrates multiple sensor data, and combines it with advanced deep learning algorithms to achieve accurate recognition and perception of photovoltaic modules in non-standard and complex scenarios. During the component sorting stage, the recognition accuracy rate is ≥90%, which will significantly improve the integrity of the components and the quality of the recycled materials during the disassembly process. Compared with traditional manual production lines, this solution improves sorting and disassembly efficiency by 5 times, while reducing the accident rate by 100% compared to manual operations.

[0034] The present invention will be further described below with reference to specific embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 Schematic diagram of the overall architecture of a photovoltaic module disassembly system based on multi-robotic arm collaboration and intelligent perception in an embodiment of the present invention.

[0036] Figure 2 Schematic diagram of autonomous grasping and sorting by a robotic arm in an embodiment of the present invention.

[0037] Figure 3 This is a schematic diagram of the disassembly of a junction box in an embodiment of the present invention.

[0038] Figure 4 This is a schematic diagram of the disassembly of an aluminum frame in an embodiment of the present invention.

[0039] Figure 5 Schematic diagram of multi-manipulator collaborative shearing and classification in an embodiment of the present invention. DETAILED DESCRIPTION

[0040] Example

[0041] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. The described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts are within the scope of protection of the present invention.

[0042] As used in this application and the claims, the words "a," "an," "an," and / or "the" are not intended to refer to the singular but may include the plural unless the context clearly indicates otherwise. Generally speaking, the terms "comprises" and "include" only indicate the inclusion of the steps and elements specifically identified, and these steps and elements do not constitute an exclusive list. A method or apparatus may also include other steps or elements.

[0043] Unless otherwise specifically stated, the relative arrangement of the parts and steps, numerical expressions and numerical values ​​set forth in these embodiments do not limit the scope of the present application. At the same time, it should be understood that, for ease of description, the sizes of the various parts shown in the drawings are not drawn according to actual proportional relationships. The techniques, methods and equipment known to those of ordinary skill in the relevant art may not be discussed in detail, but where appropriate, the techniques, methods and equipment should be considered as part of the authorization specification. In all examples shown and discussed here, any specific values ​​should be interpreted as being merely exemplary and not as limitations. Therefore, other examples of the exemplary embodiments may have different values. It should be noted that similar numbers and letters represent similar items in the following figures, and therefore, once an item is defined in one figure, it does not need to be further discussed in subsequent figures.

[0044] Combine Figure 1 , a photovoltaic module disassembly system and method based on multi-robotic arm collaboration and intelligent perception.

[0045] The specific technical solutions for achieving the purpose of the present invention are as follows:

[0046] A photovoltaic module disassembly system based on multi-manipulator collaboration and intelligent perception, including a robot arm autonomous grasping and sorting module, a junction box disassembly module, an aluminum frame disassembly module, and a multi-manipulator collaborative shearing and sorting module;

[0047] The robotic arm autonomous grasping and sorting module uses the robotic arm to identify the stacked photovoltaic modules through a visual recognition model and a deep learning recognition model, and then uses multiple robotic arms to collaboratively complete the autonomous grasping and sorting of the stacked photovoltaic modules and place the photovoltaic modules with the back side facing upwards;

[0048] The junction box disassembly module uses a double spatula to separate the junction box on the back of the photovoltaic module and put the disassembled junction box into place;

[0049] The aluminum frame disassembly module determines the center position of the back of the photovoltaic module through a visual recognition model, and then uses an electric push rod device to fit the back plate of the photovoltaic module. The electric push rod synchronously pushes out the frame and disassembles the aluminum frame through thrust, so that the aluminum frame and the back plate are completely separated;

[0050] The multi-robot collaborative shearing and classification module uses the visual perception model on the robot arm to identify the aluminum frame. After the identification is completed, two sorting robot arms are used to collaboratively grab the aluminum frame, and then the cutting robot arm is used to cut the corners of the aluminum frame to complete the disassembly of the aluminum frame. Finally, the disassembled photovoltaic backplane and aluminum frame are identified by the sorting robot arm visual module and sorted to the corresponding position.

[0051] Combine Figure 2 , the robotic arm autonomous grasping and sorting module includes a visual recognition unit, a path planning unit and a collaborative grasping unit;

[0052] The visual recognition unit uses a pre-trained convolutional neural network to identify the photovoltaic module backsheet stacked on the top layer and select its complete outline;

[0053] The path planning unit plans the motion path, grasping position and angle of the robot arm according to the frame selection position of the visual recognition unit;

[0054] The collaborative grasping unit adopts multiple robotic arms to collaboratively grasp the photovoltaic module backboard when the backboard size exceeds a threshold value (>2m×1m) or the curvature is greater than a threshold value (curvature>15°).

[0055] In this embodiment, the movement path, grasping position and angle of the sorting robot 1 are planned based on the frame selection position identified by the identification unit. When the backboard is large or partially bent, the sorting robot 2 is used to cooperate with the sorting robot 1 to grasp it. The photovoltaic module backboard is then rotated by the robot arm so that the junction box is in an upward position and is then placed on the conveyor belt.

[0056] Combine Figure 3 , the junction box disassembly module includes a double blade unit and a push rod unit;

[0057] The double scraper units are arranged horizontally and symmetrically, and cut inward from both sides of the photovoltaic module junction box to sever the connection between the junction box and the photovoltaic module backplane, and the push rod unit pushes the disassembled junction box to the specified position.

[0058] Combine Figure 4 , the aluminum frame disassembly module includes a visual recognition unit, a synchronous motor unit and a robust control unit;

[0059] The visual recognition unit determines the center position of the photovoltaic module back plate through the visual recognition model and adjusts the electric push rod device to the aluminum frame;

[0060] The synchronous motor unit controls multiple sets of electric push rods to push against the frame in multiple directions to perform thrust disassembly;

[0061] The robust control unit adopts a robust controller based on a nonlinear disturbance observer to compensate for external disturbances during the disassembly process.

[0062] Combine Figure 5 , the multi-manipulator collaborative shearing and classification module includes a collaborative shearing unit and a classification and recycling unit;

[0063] The collaborative shearing unit controls the sorting robot arm to grab the aluminum frame and controls the shearing robot arm to shear from the corner;

[0064] The classification and recycling unit identifies the type of sheared material through a trained convolutional neural network and sorts it to a designated location.

[0065] Specifically in this embodiment, the collaborative shearing unit controls the sorting robot arm 3 and the sorting robot arm 4 to jointly grab a section of separated aluminum frame, and controls the shearing robot arm to shear from the corner; if there is no corner, it is directly sorted to the corresponding recycling box;

[0066] If there is only a single corner in the aluminum frame segment, the corner is cut by the shearing robot arm, and then the sorting robot arm 3 and the sorting robot arm 4 sort the cut aluminum frame segments into corresponding recycling bins respectively;

[0067] If the aluminum frame segment has two corners, the corner close to the sorting robot arm 3 is preferentially cut. After the cutting is completed, the sorting robot arm 3 first recycles the aluminum frame segment, and then re-identifies and grabs the aluminum frame with a single corner through the visual perception module, and repeats the cutting process of the single corner for cutting and recycling;

[0068] After the aluminum frame segment is recycled, the photovoltaic backboard on the conveyor belt is re-identified through the visual perception module. If there is still aluminum frame on the conveyor belt, the above steps are repeated for sorting, cutting and recycling until all aluminum frame segments are recycled. Finally, the sorting robot arm 3 and the sorting robot arm 4 work together to grab the photovoltaic module backboard and sort it into the corresponding recycling box.

[0069] In addition, the present invention also provides a photovoltaic module disassembly method based on multi-manipulator collaboration and intelligent perception, comprising the following steps:

[0070] Step 1: The robotic arm autonomously grasps and sorts the stacked photovoltaic modules using a visual recognition model. Multiple robotic arms collaborate to autonomously grasp and sort the stacked modules, placing the modules with the backs of the photovoltaic panels facing upwards on the conveyor belt.

[0071] Step 1-1: Use the pre-trained convolutional neural network to identify the photovoltaic module backsheet stacked on the top layer and select its complete outline;

[0072] Step 1-2: Plan the robot's motion path, grasping position, and angle based on the frame position of the visual recognition unit.

[0073] When the backsheet size exceeds a threshold or the curvature is greater than a threshold, multiple robotic arms are used to collaboratively grasp the photovoltaic module backsheet.

[0074] Step 2: Disassemble the junction box of the photovoltaic module. Use a double spatula to separate the junction box on the back of the photovoltaic module and put the disassembled junction box into place.

[0075] During disassembly, use a horizontally set double scraper unit to cut inward from both sides of the photovoltaic module junction box, sever the connection between the junction box and the photovoltaic module backplane, and use the push rod device to push the disassembled junction box to the designated position.

[0076] Step 3: Dismantle the aluminum frame. Use the visual recognition model to determine the center position of the back of the photovoltaic module. Then use the electric push rod device to fit the back plate of the photovoltaic module. Then use the electric push rod to synchronously push out the frame and dismantle the aluminum frame through the thrust, so that the aluminum frame and the back plate are completely separated.

[0077] During disassembly, the center position of the photovoltaic module backplane is determined through the visual recognition model, and the electric push rod device is adjusted to the aluminum frame;

[0078] Multiple sets of electric push rods are controlled to synchronously push against the frame in multiple directions for thrust disassembly. At the same time, a robust controller based on nonlinear disturbance observer is used to compensate for external disturbances during the disassembly process.

[0079] Step 4: Multiple robotic arms collaborate in shearing and autonomous planning and classification. The visual perception model on the robotic arm is used to identify the aluminum frame. After identification, two sorting robotic arms are used to collaboratively grab the aluminum frame. The cutting robotic arm then cuts the corners of the aluminum frame to complete the disassembly of the aluminum frame. Finally, the disassembled photovoltaic backsheet and aluminum frame are identified by the visual module of the sorting robotic arm and sorted to the corresponding positions.

[0080] During this process, the sorting robot arm is controlled to grab the aluminum frame, and the cutting robot arm is controlled to cut it from the corner. Then, the trained convolutional neural network is used to identify the type of cut material and sort it to the designated location.

[0081] This embodiment employs two coordinated multi-manipulator systems, along with junction box and aluminum frame disassembly devices. Using multi-sensor fusion technology and adaptive control algorithms, this system implements a photovoltaic module disassembly system based on multi-manipulator collaboration and intelligent sensing. This system further improves the efficiency of disassembling retired photovoltaic module backsheets and the recovery rate of material components, while reducing labor costs. The system is highly adaptable, capable of handling complex scenarios such as stacking and damage, and is compatible with over 80% of modules from mainstream manufacturers.

[0082] The above-described embodiments merely represent several implementation methods of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that a person skilled in the art could make various modifications and improvements without departing from the spirit of the present application, all of which fall within the scope of protection of the present application. Therefore, the scope of protection of the present patent application shall be determined by the appended claims.

Claims

1. A photovoltaic module disassembly system based on multi-manipulator collaboration and intelligent perception, characterized in that: It includes a robotic arm autonomous grasping and sorting module, a junction box disassembly module, an aluminum frame disassembly module, and a multi-robotic arm collaborative shearing and sorting module; The robotic arm autonomous grasping and sorting module uses the robotic arm to identify the stacked photovoltaic modules through a visual recognition model and a deep learning recognition model, and then uses multiple robotic arms to collaboratively complete the autonomous grasping and sorting of the stacked photovoltaic modules and place the photovoltaic modules with the back side facing upwards; The junction box disassembly module uses a double spatula to separate the junction box on the back of the photovoltaic module and put the disassembled junction box into place; The aluminum frame disassembly module determines the center position of the back of the photovoltaic module through a visual recognition model, and then uses an electric push rod device to fit the back plate of the photovoltaic module. The electric push rod synchronously pushes out the frame and disassembles the aluminum frame through thrust, so that the aluminum frame and the back plate are completely separated; The multi-robot collaborative shearing and classification module uses the visual perception model on the robot arm to identify the aluminum frame. After the identification is completed, two sorting robot arms are used to collaboratively grab the aluminum frame, and then the cutting robot arm is used to cut the corners of the aluminum frame to complete the disassembly of the aluminum frame. Finally, the disassembled photovoltaic backplane and aluminum frame are identified by the sorting robot arm visual module and sorted to the corresponding position.

2. The photovoltaic module disassembly system based on multi-manipulator collaboration and intelligent perception according to claim 1 is characterized in that: The robotic arm autonomous grasping and sorting module includes a visual recognition unit, a path planning unit and a collaborative grasping unit; The visual recognition unit uses a pre-trained convolutional neural network to identify the photovoltaic module backsheet stacked on the top layer and select its complete outline; The path planning unit plans the motion path, grasping position and angle of the robot arm according to the frame selection position of the visual recognition unit; The collaborative grasping unit adopts multiple robotic arms to collaboratively grasp the photovoltaic module backboard when the backboard size exceeds a threshold or the bending degree is greater than a threshold.

3. The photovoltaic module disassembly system based on multi-manipulator collaboration and intelligent perception according to claim 1 is characterized in that: The junction box disassembly module includes a double scraper unit and a push rod unit; The double scraper units are arranged horizontally and symmetrically, and cut inward from both sides of the photovoltaic module junction box to sever the connection between the junction box and the photovoltaic module backplane, and the push rod unit pushes the disassembled junction box to the specified position.

4. The photovoltaic module disassembly system based on multi-manipulator collaboration and intelligent perception according to claim 1 is characterized in that: The aluminum frame disassembly module includes a visual recognition unit, a synchronous motor unit and a robust control unit; The visual recognition unit determines the center position of the photovoltaic module back plate through the visual recognition model and adjusts the electric push rod device to the aluminum frame; The synchronous motor unit controls multiple sets of electric push rods to push against the frame in multiple directions to perform thrust disassembly; The robust control unit adopts a robust controller based on a nonlinear disturbance observer to compensate for external disturbances during the disassembly process.

5. The photovoltaic module disassembly system based on multi-manipulator collaboration and intelligent perception according to claim 1 is characterized in that: The multi-manipulator collaborative shearing and sorting module includes a collaborative shearing unit and a sorting and recycling unit; The collaborative shearing unit controls the sorting robot arm to grab the aluminum frame and controls the shearing robot arm to shear from the corner; The classification and recycling unit identifies the type of sheared material through a trained convolutional neural network and sorts it to a designated location.

6. A photovoltaic module disassembly method based on multi-manipulator collaboration and intelligent perception, characterized in that: The following steps are involved: Step 1: Autonomous grasping and sorting by the robotic arm: Using a visual recognition model, the robotic arm identifies and perceives the stacked photovoltaic modules. Multiple robotic arms collaborate to autonomously grasp and sort the stacked photovoltaic modules, placing the modules with the backs of the photovoltaic panels facing upwards on the conveyor belt. Step 2: Disassemble the junction box of the photovoltaic module. Use a double spatula to separate the junction box on the back of the photovoltaic module and put the disassembled junction box into place. Step 3: Dismantle the aluminum frame. Use the visual recognition model to determine the center position of the back of the photovoltaic module. Then use the electric push rod device to fit the back plate of the photovoltaic module. Then use the electric push rod to synchronously push out the frame and dismantle the aluminum frame through the thrust, so that the aluminum frame and the back plate are completely separated. Step 4: Multiple robotic arms collaborate in shearing and autonomous planning and classification. The visual perception model on the robotic arm is used to identify the aluminum frame. After identification, two sorting robotic arms are used to collaboratively grab the aluminum frame. The cutting robotic arm then cuts the corners of the aluminum frame to complete the disassembly of the aluminum frame. Finally, the disassembled photovoltaic backsheet and aluminum frame are identified by the visual module of the sorting robotic arm and sorted to the corresponding positions.

7. The photovoltaic module disassembly method based on multi-manipulator collaboration and intelligent perception according to claim 6, characterized in that: The autonomous grasping and sorting by the robotic arm in step 1 is specifically as follows: Step 1-1: Use the pre-trained convolutional neural network to identify the photovoltaic module backsheet stacked on the top layer and select its complete outline; Step 1-2: Plan the robot's motion path, grasping position, and angle based on the frame position of the visual recognition unit. When the backsheet size exceeds a threshold or the curvature is greater than a threshold, multiple robotic arms are used to collaboratively grasp the photovoltaic module backsheet.

8. The photovoltaic module disassembly method based on multi-manipulator collaboration and intelligent perception according to claim 6, characterized in that: When disassembling the photovoltaic module junction box in step 2, a horizontally arranged double scraper unit is used to cut inward from both sides of the photovoltaic module junction box to sever the connection between the junction box and the photovoltaic module backplane, and a push rod device is used to push the disassembled junction box to a designated position.

9. The photovoltaic module disassembly method based on multi-manipulator collaboration and intelligent perception according to claim 6, characterized in that: When disassembling the aluminum frame in step 3, the center position of the photovoltaic module backplane is determined by the visual recognition model, and the electric push rod device is adjusted to the inside of the aluminum frame; Multiple sets of electric push rods are controlled to synchronously push against the frame in multiple directions for thrust disassembly. At the same time, a robust controller based on nonlinear disturbance observer is used to compensate for external disturbances during the disassembly process.

10. The photovoltaic module disassembly method based on multi-manipulator collaboration and intelligent perception according to claim 6, characterized in that: During the multi-robotic arm collaborative shearing and autonomous planning classification in step 4, the aluminum frame is grasped by controlling the sorting robotic arm, and the cutting robotic arm is controlled to cut from the corner. Then, the type of the sheared material is identified through the trained convolutional neural network and sorted to the designated location.

Images