Intelligent sampling device based on unmanned aerial vehicle and sampling control method

Abstract

The invention relates to the technical field of sampling devices, and provides an intelligent sampling device based on an unmanned aerial vehicle and a sampling control method, and the intelligent sampling device comprises the unmanned aerial vehicle and a sampling robot carried on the unmanned aerial vehicle; a first mechanical arm, a second mechanical arm, a gas storage device and a gas conveying pipe are arranged on the sampling robot; an electric wrench and a claw-shaped device are arranged at the tail end of the first mechanical arm and are used for dismounting and resetting a mounting part at the sampling opening; a first gate and a connector are arranged at the tail end of the second mechanical arm and used for being connected with the sampling port in a matched mode, and the first mechanical arm and the second mechanical arm act in a linkage mode to achieve sampling; one end of the gas delivery pipe is connected to the gas storage device, and the other end is connected to the first gate of the second mechanical arm and connected to the sampling port through the first gate and the interface. The device is suitable for the gas sampling environment of the transformer and is not limited by space, and the applicability and reliability of the device are greatly improved.

Description

Technical Field

[0001] The present disclosure relates to the technical field related to sampling devices, and in particular to an intelligent sampling device based on a drone and a sampling control method thereof. Background Art

[0002] The statements in this section merely provide background information related to the present disclosure and do not necessarily constitute prior art.

[0003] Transformers are widely used in power systems, and their fault diagnosis is a key link in ensuring the safe and stable operation of power systems. When the insulation material inside the transformer fails due to manufacturing defects, material aging or mechanical stress, it usually causes insulating media such as oil to produce gas under the action of discharge or overheating. The composition and concentration of gas can provide important clues to the type, severity and location of transformer faults. Therefore, the extraction and analysis of gas plays a key role in transformer fault diagnosis.

[0004] The inventors found in their research that, at present, the extraction of gas is usually achieved through manual operation. However, this process has many limitations and risks. At this time, overheating, partial discharge and other phenomena may have occurred inside the transformer, and there is a risk of explosion of oil or gas, which directly threatens the lives of construction workers. The traditional manual sampling method not only increases the difficulty and danger of the operation, but also limits the timeliness and accuracy of fault diagnosis to a certain extent. The gas sampling position inside the transformer is usually unique and fixed, and the transformer itself has a complex structure and a narrow sampling environment, which is very different from gas sampling in an open space. Existing automatic sampling devices are mostly designed for relatively open spaces and cannot be directly applied to the gas sampling scenarios of transformers, resulting in these devices being limited in the internal gas sampling of transformers. Summary of the invention

[0005] In order to solve the above problems, the present disclosure proposes an intelligent sampling device based on a drone and a sampling control method. The proposed intelligent sampling device is carried on a drone and can perform unmanned intelligent sampling in complex transformer environments through intelligent identification and automatic control operations.

[0006] In order to achieve the above objectives, the present disclosure adopts the following technical solutions:

[0007] One or more embodiments provide an intelligent sampling device based on a drone, including a drone and a sampling robot mounted on the drone; the sampling robot is provided with a first mechanical arm, a second mechanical arm, a gas storage device and a gas transmission pipe;

[0008] An electric wrench and a claw-shaped device are provided at the end of the first mechanical arm for removing and resetting the installation components at the sampling port; a first gate and an interface are provided at the end of the second mechanical arm for matching and connecting with the sampling port, and the first mechanical arm and the second mechanical arm are linked to realize sampling;

[0009] One end of the gas delivery pipe is connected to the gas storage device, and the other end is connected to the first gate of the second mechanical arm, and is connected to the sampling port through the first gate and the interface.

[0010] One or more embodiments provide a sampling control method based on the above-mentioned drone-based intelligent sampling device, including a drone control method and a sampling robot control method. The sampling robot control method performs linkage control on two robotic arms to achieve sampling, including the following steps:

[0011] Determine the position of the outer cover of the gas relay according to the result of identification by the sampling identification module;

[0012] Control the electric wrench on the first mechanical arm to match and connect with the mounting component on the outer cover of the gas relay, and remove the mounting component;

[0013] When it is recognized that the installation component is removed, the claw-shaped device of the first mechanical arm is controlled to open the outer cover of the Buchholz relay;

[0014] When it is recognized that the outer cover is open, the interface part of the second robot arm moves to the opened outer cover of the gas relay, connects the air inlet end of the gas pipe to the gas plug, and opens the first gate, the second gate and the gas plug for sampling.

[0015] Compared with the prior art, the present invention has the following beneficial effects:

[0016] The sampling robot disclosed in the present invention can realize remote sampling by being carried by a drone. The electric wrench of the first robotic arm is used to loosen or remove the fixing parts of the sampling port, and the claw-shaped device is used to grab and move the cover and other covering parts of the sampling port. The gate and interface carried by the second robotic arm are used to dock the sampling port to complete the extraction and transmission of gas. The gas is pumped from the sampling port to the gas storage device through the gas pipeline for storage. The whole process does not require manual contact with the transformer, thereby avoiding potential risks in high-risk operating environments.

[0017] The present disclosure uses a drone to enable the device to easily enter the narrow space inside the transformer, and can accurately locate the sampling position in a complex structure, ensuring the accuracy and timeliness of sampling. Compared with the existing automatic sampling device, the device of this embodiment is suitable for the gas sampling environment of the transformer, is not limited by space, and greatly improves the applicability and reliability of the device.

[0018] The advantages of the present disclosure and the advantages of additional aspects will be described in detail in the following specific embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The accompanying drawings constituting a part of the present disclosure are used to provide a further understanding of the present disclosure. The exemplary embodiments of the present disclosure and the description thereof are used to explain the present disclosure but do not constitute a limitation of the present disclosure.

[0020] Figure 1 is a schematic structural diagram of a sampling device according to Embodiment 1 of the present disclosure;

[0021] Figure 2 is a flow chart of a method for controlling a drone according to Embodiment 2 of the present disclosure;

[0022] Figure 3 is a schematic diagram of a control method of a sampling robot according to Embodiment 2 of the present disclosure;

[0023] Figure 4 is a schematic diagram of the sampling process implemented by the sampling method of Example 2 of the present disclosure;

[0024] Among them: 1. Lock, 2. Gas storage device, 3. Second robotic arm, 4. First robotic arm, 4-1. Claw-shaped device, 4-2. Electric wrench, 4-3. Fixed seat, 5. First camera, 6. First gate, 7. Interface. DETAILED DESCRIPTION

[0025] The present disclosure is further described below in conjunction with the accompanying drawings and embodiments.

[0026] It should be noted that the following detailed descriptions are exemplary and are intended to provide further explanation of the present disclosure. Unless otherwise specified, all technical and scientific terms used herein have the same meanings as those commonly understood by those skilled in the art to which the present disclosure belongs.

[0027] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit exemplary embodiments according to the present disclosure. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprising" and / or "including" are used in this specification, it indicates the presence of features, steps, operations, devices, components and / or combinations thereof. It should be noted that, in the absence of conflict, the various embodiments in the present disclosure and the features in the embodiments can be combined with each other. The embodiments will be described in detail below in conjunction with the accompanying drawings.

[0028] Example 1

[0029] In the technical solutions disclosed in one or more embodiments, Figure 1 As shown, an intelligent sampling device based on a drone includes a drone and a sampling robot carried on the drone;

[0030] The sampling robot is provided with a first mechanical arm 4, a second mechanical arm 3, a gas storage device and a gas transmission pipe;

[0031] The end of the first mechanical arm 4 is provided with an electric wrench 4-2 and a claw-shaped device 4-1, which are used to remove and reset the installation components at the sampling port; the end of the second mechanical arm 3 is provided with a first gate 6 and an interface 7, which are used to achieve matching connection with the sampling port to realize sampling;

[0032] One end of the gas delivery pipe is connected to the gas storage device 2, and the other end is connected to the first gate 6 of the second mechanical arm 3, and is connected to the sampling port through the first gate 6 and the interface 7;

[0033] In this embodiment, the sampling robot can realize remote sampling by being carried by a drone. The electric wrench 4-2 of the first robotic arm 4 is used to loosen or remove the fixing parts of the sampling port, and the claw-shaped device 4-1 is used to grab and move the cover and other covering parts of the sampling port. The first gate 6 and the interface 7 carried by the second robotic arm 3 are used to dock the sampling port to complete the extraction and transmission of gas. The gas is drawn from the sampling port to the gas storage device 2 through the gas pipeline for storage. The whole process does not require manual contact with the transformer, thereby avoiding potential risks in high-risk operating environments.

[0034] Moreover, the use of drones enables the device to easily enter the narrow space inside the transformer and accurately locate the sampling position in a complex structure, ensuring the accuracy and timeliness of sampling. Compared with the existing automatic sampling device, the device of this embodiment is suitable for the gas sampling environment of the transformer, is not limited by space, and greatly improves the applicability and reliability of the device.

[0035] In other possible implementations, further, the drone may be equipped with multiple sampling robots to operate multiple sampling ports at the same time to improve work efficiency.

[0036] In some embodiments, the first mechanical arm 4 and the second mechanical arm 3 on the sampling robot are both multi-degree-of-freedom mechanical arms, which can realize multi-angle rotation and multi-directional movement;

[0037] In a feasible implementation manner, a fixed seat 4-3 is provided at the end of the first mechanical arm 4, a first rotating seat is provided at the end face of the fixed seat 4-3, an electric wrench 4-2 is fixedly provided at the first rotating seat, a second rotating seat is provided at the side face of the fixed seat 4-3, and a claw-shaped device 4-1 is fixedly provided at the second rotating seat;

[0038] In a feasible implementation manner, the structure of the first gate 6 on the second robot arm 3 includes a gate housing, a gas channel is formed in the gate housing, and two gate valves are arranged opposite to each other, and the two gate valves can be raised or lowered to cut off or open the gate to realize on-off control of the gas channel;

[0039] The interface on the second robot arm 3 is a gas interface, and the built-in through hole forms a gas channel and extends to the first gate 6;

[0040] Optionally, the first gate 6 and the interface 7 are arranged close to each other to ensure that there is substantially no air entering the gas transmission pipe, thereby improving the accuracy of collecting gas composition and content;

[0041] Optionally, the gas delivery pipe may be a hose, which may be a rubber tube or a PVC tube;

[0042] Optionally, in order to improve the accuracy of gas collection, a second gate is provided at the connection between the gas storage device 2 and the gas pipeline, which is used to close and open the gas pipeline to prevent gas from entering during the non-sampling period;

[0043] In a feasible implementation manner, the gas storage device 2 is used to store collected gas such as methane gas, and the gas storage device 2 is connected to a vacuum device for forming a vacuum environment in the gas storage device 2;

[0044] Specifically, the vacuum device can be detachably connected to the gas storage device 2, and can be vacuumed before sampling. After vacuuming, the gas storage device 2 is sealed, and then set on the drone through the lock 1 to start the sampling process.

[0045] Furthermore, the gas storage device 2 may be a gas storage tank, and a pressure sensor may be provided on the wall of the gas storage tank to determine whether the gas is full by the pressure.

[0046] Optionally, the gas storage device 2 may be provided with a gas outlet interface to enable gas to be taken out from the gas storage device 2 .

[0047] As a further technical solution, the sampling robot is also provided with a sampling robot identification module, including a first camera 5, a first lighting lamp, a first ultrasonic rangefinder and a sampling identification controller arranged at the front end of the sampling robot;

[0048] A sampling recognition controller controls the first camera 5 and the first lighting lamp, performs image processing according to the data collected by the first camera, and outputs a sampling position recognition result;

[0049] Ultrasonic sensor, used to measure the distance from the sampling robot to the sampling position;

[0050] During the sampling process, the first lighting lamp is turned on, and the first camera 5 captures an image of the sampling position under appropriate light illumination;

[0051] The sampling identification controller is configured to perform the following process to identify the location of the sample:

[0052] Step 21, obtaining image information and evaluation images collected by the first camera 5;

[0053] Optionally, the evaluation image is an original Buchholz relay cover model image in the database;

[0054] Step 22: Using a convolutional neural network to extract feature representation of the image;

[0055] Step 23: Based on the extracted features, a metric method such as Euclidean distance or cosine similarity is used to compare these feature representations to evaluate the similarity between the images, thereby determining the approximate position of the sampling port, and using the first ultrasonic rangefinder to determine the distance from the sampling port (air plug);

[0056] Step 24: According to the identified approximate position of the sampling port and the distance measured, a path is planned and the movement of the sampling robot is controlled so that the sampling robot is closer to the sampling position.

[0057] Among them, the first robotic arm 4 and the second robotic arm 3 realize sampling, and the specific process is as follows:

[0058] Air plug cover operation stage: the sampling robot uses the front-end first camera 5 to collect images for identification, locates the sampling position, and uses the first ultrasonic rangefinder to continuously approach the cover. After approaching, the first mechanical arm 4 aligns the electric wrench 4-2 with the installation components such as screws at the sampling position, then docks and uses the electric wrench 4-2 to unscrew the external screws in a counterclockwise direction, and then uses the claw-shaped device 4-1 to open the cover;

[0059] Gas collection stage: after the outer cover is opened, the interface 6 of the second robotic arm 3 connects the gas inlet end of the gas pipe to the gas plug, and the first gate 6 is opened to allow the gas to enter the gas pipe; after the gas pipe is connected, the electric wrench 4-2 is used to open the gas plug for gas sampling. After the gas is filled, the first robotic arm 4 will close the gas plug, close the outer cover, and finally tighten the screws in reverse.

[0060] Among them, the gas plug on the transformer is provided with an opening device, which can be opened to release gas through the through hole inside the gas plug; the second robotic arm 3 aligns the interface 7 with the gas plug and makes the gas pipe mouth tightly fasten the gas plug to prevent air from flowing in, then the first gate 6 of the robotic arm and the second gate on the gas storage device 2 are opened, and then the first robotic arm 4 begins to slowly open the gas plug, and gas begins to continuously enter the gas storage tank.

[0061] In some embodiments, the sampling robot is also provided with a walking device, which can adopt a crawler-type walking mechanism; the connection method between the sampling robot and the drone can adopt a lock connection, and the sampling robot is also provided with a lock 1 for matching and connecting with the carrying component on the drone; the carrying component can be automatically controlled to connect or release the lock 1;

[0062] In some embodiments, the drone carries the sampling robot to achieve transportation, thereby achieving remote sampling; the drone is also provided with a drone identification module, which includes:

[0063] Signal receiving device: used to receive navigation signals sent from the ground to help the drone determine its approximate location. These signals may include GPS signals or other radio wave signals to guide the drone to fly to a predetermined location;

[0064] Second camera: used to capture the environment image around the drone and transmit the image data to the drone processing unit for analysis;

[0065] Secondary lighting: Provides sufficient light source for the camera to ensure the clarity of image data

[0066] Second ultrasonic rangefinder: used to measure the distance between the drone and the target location. These measurement data are used to adjust the position of the drone in real time to ensure that the drone can safely and accurately approach the sampling location;

[0067] UAV processing unit: It is configured to control the UAV to accurately reach the sampling location through standard image comparison, fusion navigation and real-time distance measurement. The specific process is as follows:

[0068] Step 11: Start the drone, obtain the information of the accident departure point to determine the sampling area, and use the signal receiving device to navigate the drone to the approximate location of the accident point;

[0069] Step 12: After the drone approaches the sampling area, the second camera is enabled to start acquiring real-time image data of the scene, and the image data is transmitted to the drone processing unit for analysis.

[0070] Step 13: The drone processing unit analyzes the real-time image data acquired by the camera based on the neural network, and compares the acquired image with the standard image information stored in the original data until the comparison similarity reaches the set threshold; the drone speed is reduced;

[0071] Optionally, the setting threshold of the similarity can be set to above 90%, and the drone identifies the exact location of the accident point.

[0072] Step 14: In the slow flight state, enable the second ultrasonic rangefinder to measure the distance between the drone and surrounding objects, and adjust the flight path and position in real time to accurately reach the sampling location;

[0073] After sampling, the sampling robot is transported to a safe area through identification technology. The above drone identification module can ensure that the drone can accurately locate in complex environments and perform accurate operations when the sampling robot needs to be deployed or recovered.

[0074] Specifically, for transformers, the drone identification process is as follows: the drone starts to transport the sampling robot, first takes a photo through the second camera, uses the transformer structure diagram in the database, performs image recognition to find the approximate position of the gas plug cover, and automatically plans the approach route. After flying above, the second camera uses the second lighting to assist in finding a suitable landing position, and uses ultrasonic ranging to continuously adjust the position. Finally, the drone is lowered until the ultrasonic ranging value matches the appropriate distance value in the database. The lock between the drone and the robot opens, and the sampling robot lands smoothly.

[0075] In the above technical solution, an intelligent sampling device based on drones is provided. When a fault occurs, the gas relay issues a fault warning. The staff will arrange the system at a safe distance. After the drone carrying the sampling robot arrives at the designated location, the robot will be deployed. Then the unmanned sampling robot will start autonomous sampling and rely on sensors to determine the position of the gas plug. Then the gas pipeline will be connected and gas sampling will begin. When the sampling is completed, a signal will be sent to the drone. After the drone arrives at the designated location, the robot position will be locked and the robot will be brought back to the safe area. The staff will remove the gas pipeline of the sampling robot and then transport the gas to the laboratory for testing and fault determination.

[0076] Example 2

[0077] Based on Example 1, this embodiment provides a sampling control method based on the intelligent sampling device based on the drone described in Example 1, including a control method for the drone and a control method for the sampling robot, wherein the control method for the drone is as follows: Figure 2 As shown, the following steps are included:

[0078] Step S1, controlling the drone to carry the sampling robot to the approximate location of the sampling area;

[0079] Step S2, obtaining images around the drone, the drone navigation information and distance measurement data, integrating the image comparison method, navigation information and distance recognition, determining the sampling landing position, and placing the sampling robot at the sampling position;

[0080] After the sampling robot lands safely, the drone releases the lock and completes the sampling robot's transportation mission.

[0081] Before step S1, a preparation stage is also included, including:

[0082] Suck out the gas in the gas storage device 2 until it reaches the vacuum device, then connect the interface of the gas storage device 2 to the vacuum gas pipe, and close the gas inlet end of the gas pipe with a gate to prevent air from entering, then connect the gas pipe port to the interface of the second mechanical arm 3, and then install the sampling robot to the lock 1 under the drone. And check whether the lock 1 can be opened by unmanned operation;

[0083] Specifically, for the transformer, in step S1, the drone starts to start the transport sampling robot, first takes a photo with the second camera, uses the transformer structure diagram in the database to perform image recognition to find the approximate position of the air plug cover, then rises to one meter above the oil tank, and uses infrared ranging to fly straight above the oil tank while maintaining safety. The second camera uses the lighting to assist in finding a suitable landing position, flies above and uses the ultrasonic ranging results to continuously approach the landing point. Finally, the drone is lowered until the ultrasonic ranging value matches the appropriate distance value in the database. The lock between the drone and the sampling robot is opened, and the sampling robot lands smoothly.

[0084] In step S2, the method of obtaining the image around the drone, the navigation information of the drone and the distance measurement data, integrating the image comparison method, the navigation information and the distance recognition, and determining the sampling landing position includes the following steps:

[0085] Step S21, activating the second camera to start acquiring real-time image data of the scene, and transmitting the image data to the drone processing unit;

[0086] Step S22: The drone processing unit performs similarity analysis on the real-time image data acquired by the second camera based on the neural network model, and compares the acquired image with the standard image information stored in the original data until the comparison similarity reaches a set threshold; the drone speed is reduced;

[0087] Optionally, the setting threshold of the similarity can be set to above 90%, and the drone identifies the exact location of the accident point.

[0088] Step S23: After the drone slows down, the second ultrasonic rangefinder is activated to measure the distance between the drone and surrounding objects, and the flight path and position are adjusted in real time to accurately reach the sampling location;

[0089] In some embodiments, the control method of the sampling robot performs linkage control on two robotic arms to achieve sampling, such as Figure 3 As shown, the following steps are included:

[0090] Step A1, determining the position of the gas relay cover according to the result of identification by the sampling identification module;

[0091] Step A2, controlling the electric wrench 4-2 on the first mechanical arm to match and connect with the mounting component on the outer cover of the gas relay, and removing the mounting component;

[0092] Specifically, the mounting component on the outer cover of the gas relay may be a screw, and the outer cover can be opened by removing the screw;

[0093] Furthermore, it also includes a process of judging whether the installed component is removed, specifically: obtaining the action time of the electric wrench 4-2 and the current recognition result of the sampling recognition module, and identifying the action state of the electric wrench 4-2. Specifically, the fusion judgment process is as follows:

[0094] 1) If the action time of the electric wrench 4-2 is within the set threshold range and is finally in a non-action state, it is determined that the removal action of the installed component is completed and the removal is successful; otherwise, it is determined that the removal fails, which is the result 1;

[0095] 2) The sampling and recognition module acquires images in real time, compares the images acquired after the electric wrench 4-2 is moved with the images acquired before the electric wrench 4-2 is moved, and performs target recognition on the position of the installation component. If there is no installation component in the image after the electric wrench 4-2 is moved, it is determined that the pickup is successful, otherwise, it is determined that the pickup fails, which is the second result;

[0096] 3) When both result 1 and result 2 indicate that the removal is successful, it is determined that the installed component is removed; otherwise, the image is sent back to wait for manual instructions;

[0097] When the backend operator sees the returned image, he / she checks whether the item has been taken down and sends a manual instruction to determine whether the item has been taken down successfully;

[0098] Step A3: when it is recognized that the installation component is removed, the claw-shaped device 4 - 1 of the first mechanical arm 4 is controlled to open the outer cover;

[0099] Step A4: when it is recognized that the outer cover is opened, the interface part of the second robot arm 3 moves to the opened outer cover, and connects the air inlet end of the air pipe to the air plug, and opens the first gate, the second gate and the air plug to take samples;

[0100] Under the condition that the outside air cannot enter, the gas is controlled to enter the gas pipeline through the first gate 6 and the second gate.

[0101] Furthermore, after sampling is completed, the sampling port is restored, and when the gas sampling is completed, the first mechanical arm 4 operates the electric wrench 4-2 again to close the gas plug. Then, the interface part of the second mechanical arm 3 releases the connection between the gas pipe and the gas plug. At the same time, the first mechanical arm 4 uses the claw-shaped device to reset the outer cover of the gas relay, and the first mechanical arm 4 continues to tighten the screws on the outer cover through the electric wrench 4-2 to ensure the sealing and safety of the entire system.

[0102] In this embodiment, through the above linkage control, the second mechanical arm 3 and the first mechanical arm 4 can work together to realize the entire process of automatic opening, sampling, closing and resetting the outer cover of the gas plug. This not only improves the accuracy and efficiency of the operation, but also reduces the risk of manual intervention.

[0103] The sampling process achieved by the above method is as follows: Figure 4 As shown, it includes the following stages:

[0104] Preparation stage: The storage module of the sampling robot is mainly used. The staff will first check the equipment as a whole, and suck out the gas in the gas storage device 2 until it is vacuum, then connect the gas storage device 2 to the vacuum gas pipe, and close the gas inlet end of the gas pipe with a gate to prevent air from entering, and then connect the gas pipe port to the interface of the second mechanical arm 3, and then install the sampling robot to the lock under the drone.

[0105] Drone transportation stage: mainly uses drone identification module and drone transportation; the drone starts to transport the sampling robot, first takes a photo with the second camera, uses the transformer structure diagram in the database to perform image recognition to find the approximate position of the gas plug cover, then rises to one meter above the fuel tank, and uses infrared ranging to fly straight to the top of the fuel tank while maintaining safety. The second camera uses the lighting to assist in finding a suitable landing position, flies above and uses the second ultrasonic ranging result to continuously approach the landing point. Finally, the drone is lowered until the ultrasonic ranging value matches the appropriate distance value in the database. The lock between the drone and the robot opens, and the sampling robot lands smoothly.

[0106] Gas plug cover operation stage: mainly uses the sampling identification module of the sampling robot; the sampling robot uses the first camera at the front end to accurately locate the position of the gas plug cover, and uses the first ultrasonic rangefinder to continuously approach the cover. After approaching, the first mechanical arm 4 aims the electric wrench 4-2 at the screw, then docks and uses the electric wrench to unscrew the external screw counterclockwise, and then uses the claw-shaped device 4-1 to open the cover.

[0107] Gas collection stage: the second robotic arm 3 aligns the interface 7 with the gas plug and makes the gas pipe mouth tightly fasten the gas plug to prevent air from flowing in. Then the first gate 6 and the second gate of the robotic arm are opened, and then the first robotic arm 4 begins to slowly open the gas plug, and the gas begins to continuously enter the gas storage device 2.

[0108] Pressure monitoring stage: gas continuously enters the gas storage device 2 from the interface 7 and gradually squeezes the outer wall of the storage bin. The pressure of the pressure sensor on the outer wall of the storage bin gradually increases, and the sensing value increases accordingly. A signal is sent out after reaching the trigger shutdown value.

[0109] Closing operation stage: mainly uses the sampling robot identification module; the first robot arm 4 starts to close the gas plug, the second robot arm 3 closes the gate, and then closes the second gate connected to the gas storage device 2, and the gas stops entering the gas storage device 2. The second robot arm 3 removes the interface from the gas plug, the first robot arm uses the first camera to identify, and uses the claw hook component to close the outer cover, and then uses the first camera to identify the outer cover screws, and after positioning, uses the first robot arm 4 to tighten the screws clockwise.

[0110] Recovering the robot: The drone identification module and the transport module are mainly used. The drone uses the camera to locate the sampling robot, slowly lands and continuously adjusts its position until it is accurately docked and locked. The drone then transports the sampling robot back to the safe area.

[0111] Sample processing stage: After the staff opens the storage bin of the sampling robot, they take out the gas in the gas storage device 2 through the port, and then take out the gas for experimental testing.

[0112] The above description is only a preferred embodiment of the present disclosure and is not intended to limit the present disclosure. For those skilled in the art, the present disclosure may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present disclosure shall be included in the protection scope of the present disclosure.

[0113] Although the above describes the specific implementation methods of the present disclosure in conjunction with the accompanying drawings, it is not intended to limit the scope of protection of the present disclosure. Technical personnel in the relevant field should understand that on the basis of the technical solution of the present disclosure, various modifications or variations that can be made by those skilled in the art without creative work are still within the scope of protection of the present disclosure.

Claims

1. The intelligent sampling device based on drone is characterized by: It includes a drone and a sampling robot mounted on the drone; the sampling robot is provided with a first mechanical arm, a second mechanical arm, a gas storage device and a gas transmission pipe; An electric wrench and a claw-shaped device are provided at the end of the first mechanical arm for removing and resetting the installation components at the sampling port; a first gate and an interface are provided at the end of the second mechanical arm for matching and connecting with the sampling port, and the first mechanical arm and the second mechanical arm are linked to realize sampling; One end of the gas delivery pipe is connected to the gas storage device, and the other end is connected to the first gate of the second mechanical arm, and is connected to the sampling port through the first gate and the interface.

2. The intelligent sampling device based on drone according to claim 1, characterized in that: A fixed seat is arranged at the end of the first mechanical arm, a rotatable first rotating seat is arranged at the end face of the fixed seat, an electric wrench is fixedly arranged on the first rotating seat, a rotatable second rotating seat is arranged at the side face of the fixed seat, and a claw-shaped device is fixedly arranged on the second rotating seat.

3. The intelligent sampling device based on drone according to claim 1, characterized in that: The first gate on the second robot arm includes a gate housing, a gas channel is formed in the gate housing, and two gate valves are relatively arranged in the gate housing. The two gate valves can be raised and lowered to achieve on-off control of the gas channel.

4. The intelligent sampling device based on drone according to claim 1, characterized in that: The gas transmission pipe adopts a hose, which is a rubber pipe or a PVC pipe; Alternatively, the gas storage device is connected to a vacuum pumping device for forming a vacuum environment inside the gas storage device.

5. The intelligent sampling device based on drone according to claim 1, characterized in that: The sampling robot is also provided with a sampling robot identification module, including a first camera, a first lighting lamp, a first ultrasonic rangefinder and a sampling identification controller arranged at the front end of the sampling robot; Ultrasonic sensor, used to measure the distance from the sampling robot to the sampling position; A first camera, used to capture an image of a sampling position; A first lighting lamp, used for providing lighting for image acquisition by the first camera; The sampling identification controller is configured to perform the following process to identify the location of the sample: Obtaining image information collected by the first camera and an evaluation image; A convolutional neural network is used to extract feature representations of the acquired images; Based on the extracted features, the similarity between the images is evaluated by comparing the feature representations using a Euclidean distance or cosine similarity measurement method, thereby determining the approximate location of the sampling port, and using a first ultrasonic rangefinder to determine the distance to the air plug; According to the identified approximate position of the sampling port and the measured distance, the path is planned and the movement of the sampling robot is controlled so that the sampling robot is closer to the sampling position.

6. The intelligent sampling device based on drone according to claim 1, characterized in that: The drone is also provided with a drone identification module, including: a signal receiving device, a second camera, a second lighting lamp, a second ultrasonic rangefinder and a drone processing unit; UAV processing unit: configured to perform standard image comparison, fusion navigation and real-time distance measurement based on the image collected by the second camera and the distance measured by the second ultrasonic rangefinder, so as to control the UAV to accurately reach the sampling location.

7. The sampling control method of the intelligent sampling device based on a drone according to any one of claims 1 to 6, characterized in that: The control method of the sampling robot includes a control method of a drone and a control method of a sampling robot, and the control method of the sampling robot performs linkage control on two mechanical arms to realize sampling, and includes the following steps: Determine the position of the outer cover of the gas relay according to the result of identification by the sampling identification module; Control the electric wrench on the first mechanical arm to match and connect with the mounting component on the outer cover of the gas relay, and remove the mounting component; When it is recognized that the installation component is removed, the claw-shaped device of the first mechanical arm is controlled to open the outer cover of the Buchholz relay; When it is recognized that the outer cover is open, the interface part of the second robot arm moves to the opened outer cover of the gas relay, connects the air inlet end of the gas pipe to the gas plug, and opens the first gate, the second gate and the gas plug for sampling.

8. The sampling control method according to claim 7, characterized in that: The process of judging whether the installed component is removed is to obtain the action time of the electric wrench and the current recognition result of the sampling recognition module, and to identify the action state of the electric wrench. Specifically: If the action time of the electric wrench is within the set threshold range and is finally in a non-action state, it is determined that the removal action of the installed component is completed and the removal is successful; otherwise, it is determined that the removal fails, which is the result one; The sampling and recognition module acquires images in real time, compares the images acquired after the electric wrench is moved with the images acquired before the movement, and performs target recognition on the position of the installed components. If there is no installed component in the image after the electric wrench is moved, it is determined that the pickup is successful, otherwise, it is determined that the pickup fails, which is the second result; When both result 1 and result 2 indicate that the removal is successful, it is determined that the installed component is removed; otherwise, the image is sent back to wait for manual instructions.

9. The sampling control method according to claim 7, characterized in that: The control method of the drone comprises the following steps: Control the drone carrying the sampling robot to the approximate location of the sampling area; Acquire images around the drone, drone navigation information, and distance measurement data, integrate image comparison methods, navigation information, and distance recognition, determine the sampling landing position, and place the sampling robot at the sampling position.

10. The sampling control method according to claim 9, characterized in that: The method for determining the sampling landing position comprises the following steps: Enable the second camera to start acquiring real-time image data of the scene; Based on the neural network model, the real-time image data obtained by the second camera is analyzed for similarity, and the obtained image is compared with the standard image information stored in the original data until the comparison similarity reaches the set threshold; the speed of the drone is reduced; After the drone slows down, the second ultrasonic rangefinder is activated to measure the distance between the drone and surrounding objects, and the flight path and position are adjusted in real time to accurately reach the sampling location.

Images