New energy energy storage type unmanned aerial vehicle storage device and control method thereof

Abstract

The application discloses a new energy storage type unmanned aerial vehicle storage device and a control method thereof. The new energy storage type unmanned aerial vehicle storage device comprises a light energy conversion mechanism, an unfolding and folding mechanism, an energy storage and control mechanism, an angle adjusting assembly and at least one unmanned aerial vehicle. The light energy conversion mechanism comprises a solar support and a multilayer solar panel structure, and a panel temperature sensor is arranged on each solar panel. The unfolding and folding mechanism drives the multilayer solar panel structure to fold or unfold. The energy storage and control mechanism comprises a power generation box, a control system, an energy storage module and an unmanned aerial vehicle receiving and storing device. The angle adjusting assembly is used for adjusting the inclination angle of the solar support. The unmanned aerial vehicle is provided with an environment sensing module, the control system controls the unmanned aerial vehicle to take off and collect environment data, constructs a three-dimensional light field model, calculates an optimal inclination angle in combination with solar track prediction data and drives adjustment. The application increases the light receiving area through the foldable and unfoldable design of the multilayer solar panel, improves the power generation efficiency and realizes autonomous fault diagnosis.

Description

Technical Field

[0001] This application relates to the fields of new energy storage and drone applications, specifically to a new energy storage drone storage device and its control method. Background Technology

[0002] With the widespread application of drones in numerous fields such as power line inspection, environmental monitoring, and security patrols, the demand for fully automated drone helipads (storage devices) is increasing. These outdoor storage devices, equipped with solar power generation and energy storage systems, can operate independently off-grid, greatly facilitating the daily use of drones, promoting the application of drone technology in more scenarios, and driving the development of related industries.

[0003] Previously, the industry typically employed the following methods for solar-powered drone storage devices: First, fixed, rigid installation of solar panels. This method was relatively simple and direct, and could meet basic power generation needs to a certain extent. Second, the control logic relied solely on historical weather data for simple tracking to adjust the working status of the solar panels. These conventional methods solved some problems for a period of time, providing basic support for the application of solar-powered drone storage devices.

[0004] However, existing solar-powered drone storage devices have significant drawbacks. Firstly, the fixed, rigidly installed solar panels are limited by the size of the storage device, resulting in a small deployment area and an inability to dynamically adjust the optimal angle of sunlight according to the season and time of day, leading to low power generation efficiency. Secondly, in complex outdoor environments, solar panels are often blocked by trees, clouds, or other obstacles, causing actual power generation to be far lower than the theoretical value. Furthermore, when the power generation system experiences a sudden drop in power, existing devices cannot autonomously locate and diagnose the cause of the fault, relying excessively on manual on-site inspections in remote areas, resulting in high maintenance costs. Summary of the Invention

[0005] To address the technical problems in the prior art, this application provides a new energy storage type unmanned aerial vehicle (UAV) storage device and its control method.

[0006] The technical solution provided in this application for a new energy storage-type unmanned aerial vehicle (UAV) storage device and its control method is as follows:

[0007] A new energy storage-type unmanned aerial vehicle (UAV) storage device includes:

[0008] A solar energy conversion mechanism includes a solar energy support and a multi-layer solar panel structure. The multi-layer solar panel structure is mounted on the solar energy support and has multiple foldable or unfoldable solar panels. Each of the solar panels is equipped with a panel surface temperature sensor.

[0009] An unfolding mechanism is used to drive the multi-layer solar panel structure to fold or unfold.

[0010] An energy storage and control mechanism includes a power generation box, a control system, an energy storage module, and a drone transceiver and storage device. The power generation box is located at the bottom of the solar panel support, and the control system, the energy storage module, and the drone transceiver and storage device are all located inside the power generation box.

[0011] An angle adjustment component is disposed between the solar panel bracket and the power generation box, and is used to adjust the tilt angle of the solar panel bracket;

[0012] At least one drone, which is stored in the drone transceiver storage device and is equipped with an environmental perception module;

[0013] The control system is used to control the UAV to take off and collect environmental data above the multi-layer solar panel structure, construct a three-dimensional light field model based on the environmental data, and calculate the optimal tilt angle of the solar support by combining the three-dimensional light field model and solar trajectory prediction data, and drive the angle adjustment component to adjust the solar support to the optimal tilt angle.

[0014] In some embodiments, the deployment and retraction mechanism includes at least two slide rails fixed to the solar panel bracket, two slide bars slidably disposed on the slide rails, two electric push rods, and several limiting components;

[0015] The multi-layer solar panel structure includes an upper solar panel unit, a middle solar panel unit, and a lower solar panel unit. The middle solar panel unit is fixedly connected to the solar panel support. The upper and lower solar panel units are respectively fixed to two sliding bars. Two electric push rods are respectively connected between the solar panel support and the corresponding sliding bars, used to drive the upper and lower solar panel units to move away from each other to unfold or move towards each other to fold. The limiting component is used to limit the away movement of the upper and lower solar panel units. Each of the upper, middle, and lower solar panel units includes a connecting frame, a first solar panel, a second solar panel, and a third solar panel. The first solar panel is fixed to the lower end of the connecting frame. The second and third solar panels are slidably connected to the connecting frame along the width direction of the connecting frame, and the sliding directions of the second and third solar panels are opposite. The first, second, and third solar panels together constitute the solar panel.

[0016] In some embodiments, the angle adjustment assembly includes a first rotating shaft and an angle push rod; the tail end of the solar panel bracket is hinged to the top rear end of the power generation box via the first rotating shaft; the cylinder of the angle push rod is hinged to the front side wall of the power generation box via a second rotating shaft, and the piston rod end of the angle push rod is hinged to the bottom front end of the solar panel bracket; the angle push rod has a self-locking function.

[0017] In some embodiments, the drone transceiver and storage device includes a drone storage compartment, an opening and closing drive, an automatic centering mechanism, and a wireless charging module; the side wall of the drone storage compartment is provided with a door, and the opening and closing drive is disposed above the door for driving the door to open or close outward; the automatic centering mechanism includes a guide ramp and a positioning slot disposed at the bottom of the drone storage compartment; the wireless charging module is disposed at the bottom of the positioning slot.

[0018] The environmental perception module includes a light sensor array, a downward-looking camera, and a positioning module; the light sensor array is installed on the upper and lower parts of the UAV's body, including an upper hemisphere light sensor group and a lower hemisphere light sensor group; the downward-looking camera is installed on the bottom of the UAV's body; the positioning module is used to record the three-dimensional coordinates of the UAV at each sampling point.

[0019] In some embodiments, the generator box also integrates a meteorological monitoring module, a communication module, an emergency power supply interface, a zone electrical parameter monitoring module, and a string power monitoring module; the meteorological monitoring module includes a wind speed sensor, a temperature sensor, and an external meteorological cloud image data interface; the communication module includes a 4G / 5G communication unit and a Beidou short message unit; the emergency power supply interface is located on the outside of the generator box; the UAV is also equipped with a communication relay module, the zone electrical parameter monitoring module is used to monitor the real-time voltage and current of each zone of the multi-layer solar panel structure, and the string power monitoring module is used to monitor the output power of each solar panel string.

[0020] This application also provides a control method based on the aforementioned new energy storage-type unmanned aerial vehicle storage device, including the following steps:

[0021] S1 Flight Window Determination: The control system determines the optimal takeoff time window for the UAV based on the current state of charge of the energy storage module and the predicted power generation curve, and generates a flight mission plan.

[0022] S2 Environmental Data Acquisition: The UAV takes off and hovers at a preset height above the multi-layer solar panel structure along a preset path at multiple sampling points. It collects light intensity data, surface image data of each solar panel, surrounding environment image data, and three-dimensional coordinates at each sampling point through the environmental perception module.

[0023] S3 Data Processing and Modeling: The control system constructs a three-dimensional illumination field model based on the light intensity data and three-dimensional coordinates of each sampling point; determines the cleanliness coefficient of each solar panel based on the surface image data of each solar panel; identifies obstructions based on the surrounding environment image data and generates a spatiotemporal obstruction prediction map by combining it with the solar trajectory prediction model;

[0024] S4 Optimization Calculation: Taking the maximization of total power generation as the optimization objective, the optimal tilt angle adjustment time series is calculated by combining the three-dimensional illumination field model, cleanliness coefficient, spatiotemporal shading prediction map and solar trajectory prediction data.

[0025] S5 executes tilt angle adjustment: The control system drives the angle adjustment component to perform tilt angle adjustment on the solar support according to the optimal tilt angle adjustment time sequence;

[0026] S6 Feedback Correction: Compare the actual power generation of the previous cycle with the predicted power generation, and adjust and optimize the model parameters based on the deviation value.

[0027] In some embodiments, in step S1, based on the current state of charge of the energy storage module... Predicted power generation curve Energy consumption per drone mission and system base load power Calculate the available energy margin using the following formula. : , in, This refers to the total capacity of the battery pack. To calculate the time range, The current moment;

[0028] Calculate the solar altitude angle over time based on the current date and geographical location, and determine the takeoff time window based on the changes in the solar altitude angle; after confirmation... Greater than the safe energy margin Under the premise of [specific conditions], a flight mission plan including takeoff time, flight path and mission type is generated.

[0029] In some embodiments, step S3, determining the cleanliness coefficient, includes: combining local images captured at multiple sampling points into a complete top-view image of each solar panel through image stitching; performing grayscale and texture analysis on the surface area of ​​each solar panel; and determining the cleanliness coefficient. Defined as the ratio of the current reflectivity of the panel surface to the factory standard reflectivity, with a value range of [value missing]. ;

[0030] The generation of the spatiotemporal occlusion prediction map includes: semantic segmentation of the surrounding environment image, identification of the type and spatial location of occlusion objects, calculation of the shadow areas projected onto each solar panel by each occlusion object at different times using a solar trajectory prediction model, and output of a time series matrix. ,in For position coordinates, For time, The value is the occlusion coefficient at that location at that time. Indicates completely unobstructed. This indicates complete obstruction.

[0031] In some embodiments, in step S4, the objective function of the optimization calculation is: , in, The tilt angle of the solar panel support; This represents the total number of solar panels. For the first The nominal power of the solar panel; For the first A solar panel at temperature The efficiency coefficient at the bottom, ,in For temperature coefficient, For reference temperature, The real-time board surface temperature is obtained through a board surface temperature sensor. For the angle of inclination The lower sunlight and the first Angle of incidence of the normal to the solar panel; For the first Cleanliness coefficient of a solar panel; For the angle of inclination Next A solar panel at all times occlusion coefficient;

[0032] The optimization calculation is also subject to the following constraints:

[0033] Wind load safety constraints: when the real-time wind speed Exceeding the preset wind speed threshold hour, ,in , This is the maximum permissible tilt angle in calm weather. This is the extreme wind speed;

[0034] Temperature efficiency constraint: When the real-time plate temperature exceeds a preset temperature threshold, the objective function is modified to... , in, This is the heat dissipation weighting coefficient. Angle of inclination Estimated heat dissipation efficiency under the following conditions;

[0035] Energy storage state constraints: Introducing state of charge adjustment weights The corrected objective function is:

[0036] When the state of charge is lower than the preset lower threshold, set To prioritize power generation; when the state of charge exceeds a preset upper limit threshold, a forced setting is implemented. ,in The preset safe windward tilt angle is used to reduce power generation and avoid overcharging;

[0037] In step S6, the feedback correction includes: calculating the deviation rate of the previous period. , in, This represents the actual amount of electricity generated. To predict power generation; when the deviation rate When the preset deviation threshold is exceeded, the light attenuation coefficient and temperature influence coefficient in the optimization model are corrected using the exponential weighted moving average method.

[0038] In some embodiments, the control method further includes an abnormal event triggering step: predicting charging power Greater than the preset minimum operating power Under these conditions, the control system monitors the charging power of the energy storage module in real time and calculates the decrease in actual charging power relative to the predicted charging power. ,

[0039] when When the preset proportional threshold is exceeded and the meteorological monitoring data does not show changes in cloud cover, the control system retrieves the real-time electrical parameter data of the partition electrical parameter monitoring module and the string power monitoring module set in the device to locate the suspected faulty string or partition with abnormal power generation.

[0040] The drone is triggered to take off additionally to perform a local scanning mission. The flight path of the local scanning mission covers the solar panel area corresponding to the suspected fault string or partition located above.

[0041] The control system determines the cause of the anomaly based on a comprehensive analysis of the partial scan data and the real-time electrical parameter data.

[0042] If a newly appearing occlusion is identified in the image data of a partial scan, the spatiotemporal occlusion prediction map is updated and the optimal tilt angle is recalculated.

[0043] If the image data from a partial scan identifies surface contamination, the cleanliness coefficient is updated and a cleaning work order is generated and sent to the remote operation and maintenance center via the communication module.

[0044] If the image data does not show obvious obstruction or contamination, and the electrical parameter data of the partition electrical parameter monitoring module or string power monitoring module shows open circuit, short circuit or abnormal voltage and current drop, it is determined to be a hardware fault, and a fault alarm information is generated and sent to the operation and maintenance center through the communication module.

[0045] In summary, this application includes at least one of the following beneficial technical effects:

[0046] 1. By using a drone equipped with a light sensor array, a downward-looking camera, and a positioning module to collect environmental data at multiple spatial sampling points, a three-dimensional light field model, a cleanliness coefficient, and a spatiotemporal shading prediction map are constructed. Combined with real-time panel surface temperature obtained from a panel surface temperature sensor, real-time wind speed provided by a meteorological monitoring module, and solar trajectory prediction data, the optimal tilt angle adjustment time series is calculated under multiple constraints of wind load safety, temperature efficiency, and energy storage status. This drives the angle adjustment component to perform dynamic tilt angle adjustment on the solar panel support, achieving precise and intelligent optimization of the solar panel tilt angle. Compared with traditional fixed tilt angle or adjustment based solely on a single-point light sensor, this method can adjust the tilt angle according to the actual light environment, which is beneficial to improving power generation efficiency.

[0047] 2. Through a fully closed-loop autonomous fault diagnosis process, including charging power anomaly monitoring, preliminary positioning of electrical parameter data from the zone electrical parameter monitoring module and string power monitoring module, precise visual diagnosis via UAV local scanning, and classification decision-making and alarm, the system can automatically distinguish between three different types of abnormal causes: obstruction, pollution, and hardware failure. For each type of cause, it adopts differentiated processing strategies such as adjusting the tilt angle for autonomous resolution, generating cleaning work orders, or generating fault alarm information. This reduces the frequency of manual inspections and maintenance costs, and improves the system's long-term reliable operation capability under unattended conditions.

[0048] 3. Through the feedback correction mechanism, the actual power generation of each control cycle is compared with the predicted power generation. When the deviation rate exceeds the preset threshold, the exponential weighted moving average method is used to correct the light attenuation coefficient and temperature influence coefficient in the optimization model. This enables the system to have self-learning and self-adaptive capabilities, which is conducive to improving the prediction accuracy of the optimization model and ensuring that the device maintains a high power generation efficiency during long-term operation.

[0049] 4. By using a flight window determination mechanism, the available energy margin is pre-calculated before the UAV takes off and compared with the safe energy margin. Combined with the solar altitude angle change curve, the most valuable takeoff time window is selected. This avoids the risk of insufficient power in the energy storage module due to UAV takeoff affecting the power supply of the system's basic load, while ensuring the effectiveness and timeliness of environmental data collection, thus taking into account the needs of system energy management and environmental perception tasks. Attached Figure Description

[0050] Figure 1 This is a schematic diagram of the structure of the new energy storage drone storage device of the present invention;

[0051] Figure 2 This is a top view schematic diagram of the multi-layer solar panel structure and the unfolding mechanism of the present invention in the unfolded state;

[0052] Figure 3 This is a three-dimensional schematic diagram of the unfolded structure of the connecting frame and the first solar panel, the second solar panel, and the third solar panel in a single solar panel unit of the present invention.

[0053] Figure 4 This is a schematic diagram of the angle adjustment component, solar panel support, and power generation box of the present invention;

[0054] Figure 5 This is a schematic diagram of the internal structure of the UAV transceiver and storage device of the present invention and the UAV storage status;

[0055] Figure 6 This is a top-view schematic diagram of the flight path and hovering sampling point distribution of the UAV during the environmental data acquisition task of the present invention;

[0056] Figure 7 This is a schematic diagram illustrating the construction process of the three-dimensional illumination field model of the present invention;

[0057] Figure 8 This is a schematic diagram illustrating the generation principle of the spatiotemporal occlusion prediction map of the present invention;

[0058] Figure 9 This is a schematic diagram of the overall process and abnormal triggering mechanism of the control method of the present invention;

[0059] Figure 10 This is a schematic diagram illustrating the calculation logic and takeoff time window selection for the flight window determination step of this invention.

[0060] Explanation of reference numerals in the attached drawings: 1. Photovoltaic energy conversion mechanism; 11. Solar panel support; 12. Multi-layer solar panel structure; 121. Upper solar panel unit; 122. Middle solar panel unit; 123. Lower solar panel unit; 124. Connecting frame; 125. First solar panel; 126. Second solar panel; 127. Third solar panel; 2. Deployment and retraction mechanism; 21. Slide rail; 22. Slide bar; 23. Electric push rod; 24. Limiting component; 3. Energy storage and control mechanism; 31. Power generation box; 32. Control system; 33. Energy storage module; 34. UAV transceiver and storage device; 34 1. Drone storage compartment; 342. Door; 343. Opening and closing drive; 344. Automatic centering mechanism; 3441. Guide ramp; 3442. Positioning slot; 345. Wireless charging module; 35. Weather monitoring module; 36. Communication module; 37. Emergency power supply interface; 38. Zone electrical parameter monitoring module; 39. String power monitoring module; 4. Angle adjustment assembly; 41. First rotating shaft; 42. Angle push rod; 43. Second rotating shaft; 5. Drone; 51. Environmental perception module; 511. Light sensor array; 512. Downward-looking camera; 513. Positioning module. Detailed Implementation

[0061] This application mainly adopts a multi-layer solar panel deployment and reception, UAV perception and algorithm optimization scheme, which achieves the effects of improving power generation efficiency and realizing autonomous fault diagnosis. The following is a further detailed description of this application.

[0062] Example 1

[0063] Please refer to Figures 1 to 5 The new energy storage-type drone storage device provided in this application includes a solar energy conversion mechanism 1, a deployment and retraction mechanism 2, an energy storage and control mechanism 3, an angle adjustment component 4, and at least one drone 5. The solar energy conversion mechanism 1 is mounted on a solar panel support 11. The deployment and retraction mechanism 2 is used to drive the multi-layer solar panel structure 12 to fold or unfold. The power generation box 31 in the energy storage and control mechanism 3 is located at the bottom of the solar panel support 11. The angle adjustment component 4 is located between the solar panel support 11 and the power generation box 31. The drone 5 is stored in the drone transceiver storage device 34 of the energy storage and control mechanism 3. This achieves the effects of increasing the photovoltaic area in a limited space, accurately adjusting the tilt angle of the solar panel support 11 to improve power generation efficiency, and realizing autonomous fault diagnosis. The reason is that the foldable and unfoldable design of the multi-layer solar panel structure 12 increases the light-receiving area. The drone 5 collects environmental data and combines it with algorithms to accurately calculate the optimal tilt angle. At the same time, each monitoring module can assist in locating faults.

[0064] Please refer to Figure 1Specifically, the solar energy conversion mechanism 1 includes a solar panel support 11 and a multi-layer solar panel structure 12. The solar panel support 11, serving as the supporting foundation for the entire solar energy conversion mechanism 1, is typically made of high-strength metal materials, such as aluminum alloy, which is lightweight, high-strength, and corrosion-resistant. The multi-layer solar panel structure 12 is mounted on the solar panel support 11 and contains multiple foldable or unfoldable solar panels. Monocrystalline silicon solar panels can be used, as they have high conversion efficiency. Each solar panel is equipped with a surface temperature sensor for real-time monitoring of its temperature. The surface temperature sensor can be a thermocouple sensor, which can quickly and accurately measure temperature. In addition to thermocouple sensors, a thermistor sensor can also be used, which has higher sensitivity. The surface temperature sensor is installed close to the back of the solar panel, directly acquiring the real-time surface temperature of the solar panel under actual operating conditions. This real-time surface temperature data is transmitted to the control system 32 for subsequent temperature efficiency correction in optimization calculations, eliminating measurement errors between ambient air temperature and the actual temperature rise of the photovoltaic panel surface.

[0065] Please refer to Figure 2 and Figure 3 Specifically, the unfolding mechanism 2 includes at least two slide rails 21 fixed to the solar panel support 11, two sliders 22 slidably mounted on the slide rails 21, two electric push rods 23, and several limiting components 24. The slide rails 21 are generally made of stainless steel with a smooth surface, reducing friction when the sliders 22 slide. The sliders 22 cooperate with the slide rails 21 to slide smoothly on them. The electric push rods 23 drive the sliders 22 to move, thereby realizing the folding or unfolding of the multi-layer solar panel structure 12. The electric push rods 23 can be electro-hydraulic push rods, which have high thrust and good stability. The limiting components 24 limit the divergent movement of the upper solar panel unit 121 and the lower solar panel unit 123 to prevent over-unfolding. The limiting components 24 can be limiting blocks, usually made of rubber, which can act as cushioning. Alternatively, photoelectric sensors can be used as limiting components 24; when a solar panel unit moves to a specific position, the photoelectric sensor sends a signal to stop the movement of the electric push rods 23.

[0066] The multi-layer solar panel structure 12 includes an upper solar panel unit 121, a middle solar panel unit 122, and a lower solar panel unit 123. The middle solar panel unit 122 is fixedly connected to the solar support 11, and the upper solar panel unit 121 and the lower solar panel unit 123 are respectively fixed to two sliders 22. Two electric push rods 23 are respectively connected between the solar support 11 and the corresponding sliders 22, used to drive the upper solar panel unit 121 and the lower solar panel unit 123 to move away from each other to unfold or move towards each other to fold. Figure 3As shown, the upper solar panel unit 121, the middle solar panel unit 122, and the lower solar panel unit 123 each include a connecting frame 124, a first solar panel 125, a second solar panel 126, and a third solar panel 127. The first solar panel 125 is fixed to the lower end of the connecting frame 124. The second solar panel 126 and the third solar panel 127 are slidably connected to the connecting frame 124 along the width direction of the connecting frame 124, and the sliding directions of the second solar panel 126 and the third solar panel 127 are opposite. The first solar panel 125, the second solar panel 126, and the third solar panel 127 together constitute a solar panel. This design allows the solar panel to increase the light-receiving area when unfolded and reduce the space occupied when folded.

[0067] The working process of the unfolding and retracting mechanism 2 is as follows: When the control system 32 issues an unfolding command, the two electric push rods 23 act simultaneously, pushing the upper solar panel unit 121 and the lower solar panel unit 123 to move in opposite directions along the slide rail 21, achieving the first dimension of longitudinal unfolding, until the limiting component 24 triggers the limiting signal and stops; at the same time or subsequently, the second solar panel 126 and the third solar panel 127 in each solar panel unit slide out to both sides along the width direction of the connecting frame 124, achieving the second dimension of lateral unfolding. When the control system 32 issues a folding command, the above process is executed in reverse. The second solar panel 126 and the third solar panel 127 first retract into the connecting frame 124, and then the two electric push rods 23 drive the upper solar panel unit 121 and the lower solar panel unit 123 to move towards each other, and finally the three layers of solar panel units are stacked and retracted on the solar bracket 11. Through the above two-dimensional unfolding and retracting design, the solar panel light-receiving area is expanded several times within a limited device volume, significantly improving the power generation capacity per unit area.

[0068] Please refer to Figure 1 Specifically, the energy storage and control mechanism 3 includes a generator box 31, a control system 32, an energy storage module 33, and a drone transceiver and storage device 34. The generator box 31 is located at the bottom of the solar panel bracket 11 and typically uses a robust metal casing for good protection. The control system 32, energy storage module 33, and drone transceiver and storage device 34 are all housed within the generator box 31. The control system 32 can be a microcontroller-based control system, capable of precisely controlling the operation of each component. The energy storage module 33 can be a lithium battery, which has high energy density and long lifespan.

[0069] The generator box 31 also integrates a meteorological monitoring module 35, a communication module 36, an emergency power supply interface 37, a zoned electrical parameter monitoring module 38, and a string power monitoring module 39. The meteorological monitoring module 35 includes a wind speed sensor, a temperature sensor, and an external meteorological cloud image data interface. The wind speed sensor collects real-time wind speed data at the device's location, providing input for wind load safety constraints in subsequent optimization calculations. The temperature sensor collects ambient temperature data. The external meteorological cloud image data interface receives cloud distribution and weather forecast data pushed by an external meteorological service platform, assisting in determining whether changes in sunlight are caused by cloud cover. The communication module 36 includes a 4G / 5G communication unit and a BeiDou short message unit. The 4G / 5G communication unit enables high-speed data transmission in network coverage areas, while the BeiDou short message unit enables emergency communication and status reporting in remote areas without network coverage. The emergency power supply interface 37 is located on the outside of the generator box 31 and is used to provide power output to external devices or connect to an external power source to replenish power to the energy storage module 33 in emergencies. The partition electrical parameter monitoring module 38 is used to monitor the real-time voltage and current of each partition of the multi-layer solar panel structure 12, and the string power monitoring module 39 is used to monitor the output power of each solar panel string. When used together, the two can quickly locate the specific faulty string or partition when the power generation drops abnormally, providing a precise target area for the subsequent UAV 5 local scanning mission.

[0070] The UAV 5 is also equipped with a communication relay module. When the UAV 5 is performing a mission in the air, the communication relay module can act as a signal relay node between the device and the remote operation and maintenance center, expanding the communication coverage and improving the reliability of data transmission in remote areas.

[0071] Please refer to Figure 5 The drone transceiver and storage device 34 includes a drone storage compartment 341, an opening and closing drive 343, an automatic centering mechanism 344, and a wireless charging module 345. The drone storage compartment 341 has a door 342 on its side wall, facing outwards from the generator box 31, allowing the drone 5 to fly horizontally in or out through the side entrance after the door 342 is opened. The opening and closing drive 343 is located above the door 342 and can be an electric push rod. Its fixed end is hinged to the upper part of the side wall of the drone storage compartment 341, and its piston rod end is hinged to the upper edge of the door 342. The extension and retraction of the piston rod drives the door 342 to open outwards or close inwards around its lower edge. When the door 342 is closed, it fits tightly against the side wall of the drone storage compartment 341, providing protection for the drone 5 inside. When the door 342 is open, it flips outwards to a horizontal or near-horizontal position, providing an unobstructed lateral entry and exit passage for the drone 5.

[0072] The automatic centering mechanism 344 includes a guide ramp 3441 and a positioning slot 3442 disposed inside the UAV storage compartment 341. For example... Figure 5 As shown, a guide ramp 3441 is installed on the inner wall of the UAV storage compartment 341, forming a sloping structure that slopes from top to bottom towards the center of the compartment. After the UAV 5 flies into the compartment through the side door 342, if its attitude deviates during landing, the guide ramp 3441 guides the UAV 5 to slide along the ramp towards the center of the compartment floor. A positioning slot 3442 is located at the center of the bottom of the UAV storage compartment 341, and its shape matches the bottom of the UAV 5's landing gear. After being guided by the guide ramp 3441, the UAV 5 finally lands in the positioning slot 3442, achieving precise positioning. A wireless charging module 345 is located at the bottom of the positioning slot 3442. After the UAV 5 lands in the positioning slot 3442, its bottom charging receiving coil aligns with the transmitting coil of the wireless charging module 345, enabling wireless charging using the principle of electromagnetic induction.

[0073] The operation of the UAV transceiver and storage device 34 is as follows: When the control system 32 issues a takeoff command for the UAV 5, the piston rod of the opening and closing drive component 343 extends, pushing the hatch 342 to open outward around its lower edge. The hatch 342 stops after flipping to a horizontal or near-horizontal position, and the side entrance of the UAV storage compartment 341 is fully open. Subsequently, the UAV 5 starts its rotors, rises inside the compartment to the height of the side entrance, and then flies horizontally out of the compartment. After flying out of the compartment, the UAV 5 continues to climb to a preset height above the multi-layer solar panel structure 12 and begins to perform environmental data collection tasks.

[0074] When UAV 5 completes its mission and returns, it flies to the side entrance of the UAV storage compartment 341, adjusts its attitude, and flies horizontally into the compartment through the side entrance, then slowly descends inside. During the descent, even if the landing attitude of UAV 5 deviates slightly, the guide ramp 3441 will guide UAV 5 to slide along the ramp towards the center of the compartment floor, eventually landing in the positioning slot 3442, achieving precise alignment. This ensures that the charging receiving coil on the bottom of UAV 5 is precisely aligned with the transmitting coil of the wireless charging module 345 at the bottom of the positioning slot 3442, and then automatically begins wireless charging to store energy for the next mission. After charging is complete, the piston rod of the opening and closing drive 343 retracts, pulling the door 342 to flip inward around its lower edge and close. The door 342 fits tightly against the side wall of the UAV storage compartment 341, forming a sealed protection for UAV 5 and preventing rainwater, dust, and other contaminants from entering.

[0075] The environmental perception module 51 includes a light sensor array 511, a downward-facing camera 512, and a positioning module 513. The light sensor array 511 is installed on the upper and lower parts of the UAV 5, comprising an upper hemisphere light sensor group and a lower hemisphere light sensor group, capable of comprehensively sensing light intensity. The upper hemisphere light sensor group collects incident light intensity from the sky hemisphere, while the lower hemisphere light sensor group collects light intensity reflected from the ground and solar panel surface. The two sensor groups work together to establish complete spatial distribution data of light intensity at each sampling point, providing fundamental data for subsequent construction of a three-dimensional light field model. The downward-facing camera 512 is installed on the bottom of the UAV 5 and is used to capture images of the solar panel surface and the surrounding environment. The positioning module 513 records the three-dimensional coordinates of the UAV 5 at each sampling point; a GPS positioning module can be used. Alternatively, an RTK differential positioning module can be used to achieve centimeter-level positioning accuracy, ensuring a precise correspondence between the three-dimensional coordinate data and the light intensity data at each sampling point.

[0076] Please refer to Figure 4 Specifically, the angle adjustment assembly 4 includes a first rotating shaft 41 and an angle push rod 42. The tail end of the solar bracket 11 is hinged to the top rear end of the generator box 31 via the first rotating shaft 41, allowing the solar bracket 11 to rotate around the first rotating shaft 41. The cylinder of the angle push rod 42 is hinged to the front side wall of the generator box 31 via a second rotating shaft 43, and the piston rod end of the angle push rod 42 is hinged to the bottom front end of the solar bracket 11. The angle push rod 42 has a self-locking function, which can maintain stability after being adjusted to a suitable tilt angle. The angle push rod 42 can be an electrically driven push rod, which is easy to operate and has high control precision.

[0077] The angle adjustment component 4 works as follows: When the control system 32 outputs a tilt angle adjustment command, the piston rod of the angle push rod 42 extends or retracts, pushing the front end of the solar bracket 11 to rise upward or fall downward around the first rotating shaft 41, thereby changing the tilt angle of the solar bracket 11 relative to the horizontal plane. Once the piston rod reaches the target position, the self-locking function of the angle push rod 42 automatically activates, locking the solar panel bracket 11 at the current tilt angle. Even under external forces such as wind loads, angle drift will not occur, ensuring stable operation of the solar panel at the optimal tilt angle. This hinged structure design allows for a wide tilt angle adjustment range and smooth operation of the solar panel bracket 11. Furthermore, the cylinder of the angle push rod 42 is hinged to the front wall of the generator box 31 via the second rotating shaft 43, forming a stable triangular support structure with excellent wind load resistance.

[0078] The entire working process of the device is as follows:

[0079] After the device is deployed to the target location in the field, the control system 32 first drives the deployment mechanism 2 to unfold the multi-layer solar panel structure 12, so that the upper solar panel unit 121, the middle solar panel unit 122 and the lower solar panel unit 123 unfold away from each other along the slide rail 21. At the same time, the second solar panel 126 and the third solar panel 127 in each unit slide out to the sides, completing the deployment of all solar panels, starting photovoltaic power generation and charging the energy storage module 33.

[0080] In daily operation, the control system 32 periodically assesses whether the takeoff conditions of the UAV 5 are met: first, it calculates the available energy margin. It was confirmed that its energy margin was greater than the safe energy margin. Then, the optimal takeoff time window is determined based on the solar altitude angle change curve. When the conditions are met, the control system 32 drives the opening and closing drive component 343 to open the hatch 342 outward. The UAV 5 flies horizontally out of the side entrance of the UAV storage cabin 341 and climbs to a preset height above the multi-layer solar panel structure 12. It hovers at multiple sampling points along a preset path, collects light intensity data at each sampling point through the light sensor array 511, captures surface images of each solar panel and surrounding environment images through the downward-looking camera 512, and records the three-dimensional coordinates of each sampling point through the positioning module 513. After the data collection is completed, the UAV 5 returns to the front of the side entrance of the UAV storage cabin 341, flies horizontally into the cabin through the side entrance, descends, and is guided by the guide ramp 3441 in the automatic centering mechanism 344 to fall into the positioning slot 3442 for positioning. It is then automatically charged by the wireless charging module 345.

[0081] After receiving the environmental data transmitted back by the UAV 5, the control system 32 performs data processing and modeling: a three-dimensional illumination field model is constructed based on the light intensity data and three-dimensional coordinates of each sampling point; the cleanliness coefficient of each solar panel is determined through grayscale analysis and texture analysis based on the solar panel surface image data. Based on surrounding environmental image data, occlusions are identified through semantic segmentation, and a spatiotemporal occlusion prediction map is generated by combining it with a solar trajectory prediction model. .

[0082] Subsequently, the control system 32 optimizes the total power generation by taking the maximization of total power generation as the goal, and integrates the three-dimensional illumination field model and cleanliness coefficient. Spatiotemporal occlusion prediction map Real-time board surface temperature obtained by board surface temperature sensor and the real-time wind speed provided by the meteorological monitoring module 35. Using multi-dimensional data, and under the constraints of wind load safety, temperature efficiency, and energy storage status, the optimal tilt angle adjustment time series for a future period is calculated, and the angle push rod 42 is driven to perform dynamic tilt angle adjustment on the solar support 11 according to the time series.

[0083] At the end of each control cycle, the control system 32 calculates the actual power generation of that cycle. Compared with the predicted power generation Compare the results and calculate the deviation rate:

[0084] When the deviation rate exceeds the preset deviation threshold, the exponential weighted moving average method is used to correct the light attenuation coefficient and temperature influence coefficient in the optimization model, so that the prediction accuracy of the system will continue to improve with the accumulation of running time.

[0085] In addition, during normal system operation, the control system 32 continuously monitors the actual charging power of the energy storage module 33. Once an abnormal drop in charging power is detected, the abnormal event handling process will be triggered to realize the automatic detection, location and diagnosis of faults.

[0086] When the meteorological monitoring module 35 detects that the wind speed is close to or exceeds the limit wind speed In extreme weather conditions, the control system 32 will drive the angle push rod 42 to adjust the solar panel bracket 11 to a horizontal position. It also drives the unfolding mechanism 2 to fold and retract the multi-layer solar panel structure 12 to minimize the windward area and resist severe weather, thus ensuring the physical safety of the device.

[0087] The implementation principle of this embodiment is as follows: This new energy storage drone storage device, through the foldable and unfoldable design of the multi-layer solar panel structure 12, increases the light-receiving area of ​​the solar panels within a limited space, thereby improving power generation efficiency. The drone 5 is equipped with an environmental perception module 51, which can collect environmental data. The control system 32 combines this data with solar trajectory prediction data to calculate the optimal tilt angle, and adjusts the tilt angle of the solar bracket 11 through the angle adjustment component 4, so that the solar panels are always in the best light-receiving state. At the same time, the monitoring modules in the power generation box 31 can monitor the operating status of the solar panels in real time. When a fault occurs, it can quickly locate and diagnose the cause, realizing autonomous fault diagnosis, reducing operation and maintenance costs, and representing a significant improvement and enhancement compared to existing technologies.

[0088] Example 2

[0089] Please refer to Figures 6 to 10 The control method based on a new energy storage-type unmanned aerial vehicle (UAV) storage device provided in this application includes the following steps:

[0090] Please refer to Figure 9 and Figure 10S1 Flight Window Determination: The control system 32 determines the optimal takeoff time window for the UAV 5 based on the current state of charge of the energy storage module 33 and the predicted power generation curve, and generates a flight mission plan. Specifically, the control system 32 will determine the optimal takeoff time window for the UAV 5 based on the current state of charge of the energy storage module 33. Predicted power generation curve Energy consumption of a single drone mission and system base load power According to the formula Calculate the available energy margin, where This refers to the total capacity of the battery pack. To calculate the time range, This is the current time. Then, based on the current date and geographical location, the curve of the solar altitude angle changing over time is calculated, and the takeoff time window is determined by combining this change in solar altitude angle. For example... Figure 10 As shown, the input parameter pool includes the current state of charge (SOC) and the power curve. The calculation engine sequentially performs three steps: calculating available energy margin, analyzing the maximum region of altitude angle derivative, and verifying safety boundaries, based on energy consumption, baseline parameters, and solar ephemeris data. The final output is a flight plan that includes the optimal takeoff time, flight trajectory matrix, and mission mode. Figure 10 The curve below shows the rate of change of solar altitude angle. Based on the distribution over time, the system selects the periods when the rate of change reaches its maximum value as window 1 and window 2. After confirmation... Greater than the safe energy margin Under the premise of [specific conditions], a flight mission plan including takeoff time, flight path and mission type is generated.

[0091] The technical benefits of this step are as follows: by pre-calculating the energy margin, the risk of power outage to the system's basic load is avoided when the UAV 5 takes off when the energy storage module 33 is low on power; by combining the solar altitude angle change curve to determine the takeoff time window, the UAV 5 performs environmental data collection during the period when the solar altitude angle changes most significantly, and the obtained light distribution data is the most valuable reference for tilt angle optimization.

[0092] Please refer to Figure 6 S2 Environmental Data Acquisition: The UAV 5 takes off and hovers at a preset height above the multi-layer solar panel structure 12 along a preset path at multiple sampling points. The environmental perception module 51 collects light intensity data, surface image data of each solar panel, surrounding environment image data, and three-dimensional coordinates at each sampling point. The light sensor array 511 collects light intensity data, the downward-looking camera 512 captures images of the solar panel surface and the surrounding environment, and the positioning module 513 records the three-dimensional coordinates of the UAV 5 at each sampling point.

[0093] Specifically, UAV 5 hovers sequentially above multiple sampling points on the multi-layer solar panel structure 12, following a pre-set path in its flight mission plan. For example... Figure 6 As shown, the unfolded multi-layer solar panel structure 12 forms an array of nine solar panels, from panel 1 to panel 9. The flight path of the UAV 5 covers the area directly above each solar panel and the outer scanning boundary area, and collects data at each hovering sampling point Pi. Figure 6 The diagram also shows the landing base located on one side of the power generation unit 31, which is the exit position of the side hatch 342 of the UAV storage compartment 341. The UAV 5 flies horizontally out of this side entrance and climbs to a preset height, then flies horizontally above the multi-layer solar panel structure 12 to perform data collection. At each sampling point, the upper hemisphere light sensor group collects the incident light intensity from the sky, and the lower hemisphere light sensor group collects the reflected light intensity from the ground and the solar panel surface. The two sets of data, combined with the three-dimensional coordinates of the sampling point, constitute the complete spatial distribution information of the light intensity at that point. At the same time, the downward-looking camera 512 takes high-resolution pictures of the surface of the solar panel directly below, acquiring images of the panel surface for subsequent cleanliness analysis; the downward-looking camera 512 also takes pictures of the surrounding environment of the device, acquiring image data of surrounding buildings, trees and other potential obstructions.

[0094] Please refer to Figure 7 and Figure 8 S3 Data Processing and Modeling: The control system 32 constructs a three-dimensional illumination field model based on the light intensity data and three-dimensional coordinates of each sampling point; determines the cleanliness coefficient of each solar panel based on the surface image data of each solar panel; identifies obstructions based on the surrounding environment image data and generates a spatiotemporal obstruction prediction map by combining the solar trajectory prediction model.

[0095] like Figure 7 As shown, the construction process of the three-dimensional illumination field model is as follows: the physical sensing end uses an omnidirectional illumination sensor array 511 to collect illumination intensity values ​​Li at each sampling point. After data fusion, a spatial label dataset is formed, where each sampling point Pi corresponds to a set of three-dimensional coordinates. The illumination intensity value Li is used to generate a continuous model matrix through Kriging spatial interpolation. This matrix describes the distribution of light intensity at any position and in any direction in the space above the device.

[0096] The cleanliness coefficient is determined by image stitching, which combines local images taken from multiple sampling points into a complete top-view image of each solar panel. Grayscale and texture analysis are then performed on the surface areas of each solar panel to determine the cleanliness coefficient. Defined as the ratio of the current reflectivity of the panel surface to the factory standard reflectivity, with a value range of [value missing]. .

[0097] like Figure 8 As shown, the generation principle of the spatiotemporal occlusion prediction map is as follows: semantic segmentation of the surrounding environment image is performed to identify the type of occlusion and its spatial location, and the shadow area of ​​each occlusion projected onto each solar panel at different times is calculated by combining the solar trajectory prediction model. Figure 8 The diagram shows the shadows cast by the same object on a solar panel at four times: 10:00, 12:00, 14:00, and 16:00, with different positions and areas. Based on this, the system outputs a time series matrix. ,in For position coordinates, For time, The value is the occlusion coefficient at that location at that time. Indicates completely unobstructed. This indicates complete obstruction.

[0098] The technical advantage of this step is that the three-dimensional illumination field model overcomes the limitation of traditional single-point illumination sensors, which can only acquire the illumination intensity of a single location. By collecting data from multiple spatial sampling points using a UAV, a three-dimensional distribution of illumination intensity in the space above the device is constructed, accurately reflecting the differences in illumination at different heights and locations. Cleanliness coefficient. The introduction of this feature allows the optimization calculation to account for the actual power generation capacity reduction of each solar panel due to pollution such as dust and bird droppings. Spatiotemporal shading prediction map. By combining static spatial information about shading objects with dynamic solar trajectories, it is possible to predict the shading status of each solar panel at different times in the future, providing a time-dimensional decision-making basis for tilt angle optimization.

[0099] Please refer to Figure 9 S4 optimization calculation: Taking the maximization of total power generation as the optimization objective, and combining a three-dimensional solar field model, cleanliness coefficient, spatiotemporal shading prediction map, and solar trajectory prediction data, the optimal tilt angle adjustment time series is calculated. For example... Figure 9 As shown, this step corresponds to the "Multi-objective PSO Optimization Calculation" stage in the flowchart. The objective function for the optimization calculation is:

[0100] in, The tilt angle of solar panel bracket 11; This represents the total number of solar panels. For the first The nominal power of the solar panel; For the first A solar panel at temperature The efficiency coefficient at the bottom,

[0101] in, For temperature coefficient, For reference temperature, The real-time board surface temperature is obtained through a board surface temperature sensor. For the angle of inclination The lower sunlight and the first Angle of incidence of the normal to the solar panel; For the first Cleanliness coefficient of a solar panel; For the angle of inclination Next A solar panel at all times The occlusion coefficient. The optimization calculation is also subject to the following constraints:

[0102] Wind load safety constraint: When the wind speed sensor in the meteorological monitoring module 35 detects the real-time wind speed Exceeding the preset wind speed threshold At that time, the tilt angle of the solar panel bracket 11 Subject to upper limit constraints ,in: , in, This is the maximum permissible tilt angle in calm weather. The limit wind speed is defined by this piecewise function. This function ensures that the solar panel support 11 can be freely adjusted to the optimal tilt angle when the wind speed is low, the maximum allowable tilt angle is linearly reduced as the wind speed gradually increases to reduce wind load, and the solar panel support 11 is forcibly adjusted to a horizontal position when the wind speed reaches the limit value to ensure structural safety.

[0103] Temperature efficiency constraint: When the real-time surface temperature of the solar panel, obtained by the panel temperature sensor, exceeds the preset temperature threshold, it indicates that the power generation efficiency of the solar panel has decreased significantly due to high temperature. At this time, the objective function is modified to... , in, This is the heat dissipation weighting coefficient. Angle of inclination The estimated heat dissipation efficiency is given. The physical meaning of this constraint is that appropriately increasing the tilt angle is beneficial to the natural convection heat dissipation of the solar panel surface. Therefore, under high temperature conditions, the optimization algorithm will seek a balance between power generation and heat dissipation efficiency, and select a tilt angle that can both ensure high light incidence and facilitate cooling.

[0104] Energy storage state constraints: Introducing state of charge adjustment weights The corrected objective function is:

[0105] When the state of charge of the energy storage module 33 is lower than the preset lower threshold, the setting is... To prioritize power generation, the optimization algorithm is made to favor tilt angles with higher power output; when the state of charge exceeds a preset upper limit threshold, it indicates that the energy storage module 33 is close to full charge, at which point a forced setting is implemented. ,in The preset safe windward tilt angle is used to reduce power generation and avoid overcharging, prioritizing the structural safety of the device.

[0106] S5 executes tilt angle adjustment: The control system 32 drives the angle adjustment component 4 to perform tilt angle adjustment on the solar bracket 11 according to the optimal tilt angle adjustment time sequence. The angle push rod 42 in the angle adjustment component 4, according to the instructions of the control system 32, pushes the solar bracket 11 to rotate around the first rotating shaft 41, adjusting it to a suitable tilt angle. After the angle push rod 42 reaches the target position, the self-locking function automatically activates, stably locking the solar bracket 11 at the current tilt angle. The control system 32 sequentially drives the angle push rod 42 to perform the corresponding tilt angle adjustment action according to each time node in the optimal tilt angle adjustment time sequence, realizing dynamic tracking and adjustment of the tilt angle of the solar bracket 11 throughout the day.

[0107] S6 Feedback Correction: Compares the actual power generation of the previous cycle with the predicted power generation, and adjusts and optimizes the model parameters based on the deviation value. Calculates the deviation rate of the previous cycle.

[0108] in, This represents the actual amount of electricity generated. To predict power generation. When the deviation rate When the preset deviation threshold is exceeded, the light attenuation coefficient and temperature influence coefficient in the optimization model are corrected using the exponential weighted moving average method.

[0109] The technical advantage of this step lies in the fact that, through a closed-loop feedback mechanism, the system can automatically identify and correct parameter deviations in the optimization model. For example, with the long-term use of solar panels, their photoelectric conversion efficiency will gradually decrease, resulting in actual power generation being lower than the predicted value based on initial parameters. By using the exponentially weighted moving average method, the system can smoothly correct the light attenuation coefficient and temperature influence coefficient by using recent data as the primary source and historical data as a supplement, which helps improve the prediction accuracy of the optimization model and achieves the system's self-learning and self-adaptation.

[0110] The implementation principle of this embodiment is as follows: This control method determines the takeoff time window of the UAV 5, allowing the UAV 5 to collect accurate environmental data. This data is then processed and modeled, and an optimization algorithm is used to calculate the optimal tilt angle adjustment time series, enabling the solar panel bracket 11 to be adjusted to the optimal tilt angle, thereby improving power generation efficiency. Simultaneously, feedback correction continuously optimizes the model parameters, making the system more accurate and reliable. Compared to traditional control methods, this method introduces more environmental factors and constraints, better adapting to complex field environments and achieving intelligent control and management of the solar panels.

[0111] Example 3

[0112] Please refer to Figure 9 This embodiment adds an abnormal event triggering step to the existing embodiment 2. For example... Figure 9 The anomaly triggering mechanism, shown in the dashed box on the left, triggers a local anomaly scan process when the system detects a sudden drop in power. Specifically: during the prediction of charging power... Greater than the preset minimum operating power Under these conditions, the control system 32 monitors the charging power of the energy storage module 33 in real time and calculates the decrease in actual charging power relative to the predicted charging power:

[0113] in, The charging power is predicted based on the current optimized model. Set the actual charging power of energy storage module 33. Greater than the preset minimum operating power The prerequisite is to eliminate false triggering caused by normal low-power operation during periods of extremely weak light, such as sunrise and sunset.

[0114] when When the preset proportional threshold is exceeded and the meteorological monitoring data of the meteorological monitoring module 35 does not show changes in cloud cover, the control system 32 immediately retrieves the real-time electrical parameter data of the partition electrical parameter monitoring module 38 and the string power monitoring module 39 to locate the suspected faulty string or partition with abnormally low power generation.

[0115] Specifically, the partition electrical parameter monitoring module 38 provides real-time voltage and current data for each partition, the string power monitoring module 39 provides output power data for each string, and the control system 32 identifies strings or partitions with significantly low output power by comparing the current output power of each string or partition with its historical normal power range, and marks them as suspected fault areas.

[0116] Then, the drone 5 is triggered to fly out from the side entrance of the drone storage compartment 341 to perform a local scanning task. The flight path of the local scanning task covers the solar panel area corresponding to the suspected faulty string or partition located above. The drone 5 takes a high-resolution image of the area through the downward-looking camera 512. The control system 32 comprehensively judges the cause of the anomaly based on the local scanning data and the real-time electrical parameter data.

[0117] If newly appearing occlusions, such as fallen branches or flying plastic bags, are identified in the locally scanned image data, the spatiotemporal occlusion prediction map is updated. And re-execute the optimization calculation in step S4, recalculate the optimal tilt angle, and drive the angle push rod 42 to adjust the tilt angle of the solar bracket 11 to avoid the shading area as much as possible;

[0118] If the image data from a partial scan identifies surface contamination, such as large areas of bird droppings or mud splatter, then the cleanliness coefficient of the corresponding solar panel is updated. It generates a cleaning work order containing the coordinates of the contaminated location, the degree of contamination, and cleaning suggestions, and sends it to the remote operation and maintenance center via the communication module 36, where operation and maintenance personnel arrange on-site cleaning operations.

[0119] If the image data from the partial scan does not show obvious obstruction or contamination, and the electrical parameter data from the partition electrical parameter monitoring module 38 or the string power monitoring module 39 exhibit typical hardware fault characteristics such as open circuit, abnormally high voltage and zero current, short circuit, zero voltage and abnormal current, or abnormal voltage and current drop, then it is determined to be a hardware fault. A fault alarm message containing the fault string number, fault type, and abnormal electrical parameter data is generated and sent to the operation and maintenance center through the communication module 36. The operation and maintenance personnel then bring the corresponding spare parts to the site for repair.

[0120] The technical effect of the above-mentioned abnormal event triggering steps is that it realizes a fully closed-loop autonomous fault diagnosis process of "abnormal charging power monitoring → preliminary positioning of electrical parameter data → precise visual diagnosis of UAV → classification decision and alarm". The system can automatically distinguish between three different types of abnormal causes: obstruction, pollution and hardware failure, and take different handling strategies for each type of cause: for obstruction problems, it can be resolved autonomously by adjusting the tilt angle; for pollution and hardware failure, it generates accurate work orders or alarm information and pushes them to the operation and maintenance center, which greatly reduces the frequency of manual inspections and operation and maintenance costs, and improves the long-term reliable operation capability of the system under unattended conditions.

[0121] The implementation principle of this embodiment is as follows: After adding an abnormal event triggering step, the system can promptly detect abnormal drops in charging power and, through the cooperation of various monitoring modules and UAV 5, quickly locate the cause of the fault and take corresponding measures. This achieves automatic fault diagnosis and handling, reduces manual intervention, lowers maintenance costs, and improves the reliability and stability of the system.

[0122] The specific embodiments described above do not constitute a limitation on the scope of protection of this application. Any other corresponding changes and modifications made based on the technical concept of this application should be included within the scope of protection of this application.

Claims

1. A new energy storage-type unmanned aerial vehicle (UAV) storage device, characterized in that, include: The light energy conversion mechanism (1) includes a solar energy support (11) and a multi-layer solar panel structure (12). The multi-layer solar panel structure (12) is disposed on the solar energy support (11) and has multiple foldable or unfoldable solar panels. Each of the solar panels is provided with a panel surface temperature sensor. The unfolding mechanism (2) is used to drive the multi-layer solar panel structure (12) to fold or unfold; The energy storage and control mechanism (3) includes a power generation box (31), a control system (32), an energy storage module (33), and a drone transceiver and storage device (34). The power generation box (31) is located at the bottom of the solar support (11), and the control system (32), the energy storage module (33), and the drone transceiver and storage device (34) are all located inside the power generation box (31). An angle adjustment component (4) is disposed between the solar bracket (11) and the power generation box (31) for adjusting the tilt angle of the solar bracket (11); At least one drone (5), the drone (5) is stored in the drone transceiver storage device (34), and the drone (5) is equipped with an environmental perception module (51); The control system (32) is used to control the UAV (5) to take off and collect environmental data above the multi-layer solar panel structure (12). The environmental data includes light intensity data, surface image data of each solar panel, surrounding environment image data and three-dimensional coordinates. The control system (32) constructs a three-dimensional light field model based on the light intensity data and three-dimensional coordinates of each sampling point, determines the cleanliness coefficient of each solar panel based on the surface image data of each solar panel, identifies obstructions based on the surrounding environment image data and generates a spatiotemporal obstruction prediction map in combination with the solar trajectory prediction model, and calculates the optimal tilt angle of the solar support (11) in combination with the three-dimensional light field model, the cleanliness coefficient, the spatiotemporal obstruction prediction map and the solar trajectory prediction data, and drives the angle adjustment component (4) to adjust the solar support (11) to the optimal tilt angle.

2. The new energy storage type unmanned aerial vehicle storage device according to claim 1, characterized in that: The deployment and take-up mechanism (2) includes at least two slide rails (21) fixed on the solar panel bracket (11), two slide bars (22) slidably disposed on the slide rails (21), two electric push rods (23) and several limiting components (24); The multi-layer solar panel structure (12) includes an upper solar panel unit (121), a middle solar panel unit (122), and a lower solar panel unit (123). The middle solar panel unit (122) is fixedly connected to the solar support bracket (11), and the upper solar panel unit (121) and the lower solar panel unit (123) are respectively fixed to two sliders (22). Two electric push rods (23) are respectively connected between the solar support bracket (11) and the corresponding sliders (22) to drive the upper solar panel unit (121) and the lower solar panel unit (123) to move away from each other to unfold or move towards each other to fold. The limiting component (24) is used to control the away movement of the upper solar panel unit (121) and the lower solar panel unit (123). Limiting is performed; the upper solar panel unit (121), the middle solar panel unit (122) and the lower solar panel unit (123) each include a connecting frame (124), a first solar panel (125), a second solar panel (126) and a third solar panel (127). The first solar panel (125) is fixed to the lower end of the connecting frame (124). The second solar panel (126) and the third solar panel (127) are slidably connected to the connecting frame (124) along the width direction of the connecting frame (124), and the sliding directions of the second solar panel (126) and the third solar panel (127) are opposite. The first solar panel (125), the second solar panel (126) and the third solar panel (127) together constitute the solar panel.

3. The new energy storage type unmanned aerial vehicle storage device according to claim 1, characterized in that: The angle adjustment assembly (4) includes a first rotating shaft (41) and an angle push rod (42); the tail end of the solar bracket (11) is hinged to the top rear end of the generator box (31) through the first rotating shaft (41); the cylinder of the angle push rod (42) is hinged to the front side wall of the generator box (31) through the second rotating shaft (43), and the piston rod end of the angle push rod (42) is hinged to the bottom front end of the solar bracket (11); the angle push rod (42) has a self-locking function.

4. A new energy storage-type unmanned aerial vehicle storage device according to claim 1, characterized in that: The UAV transceiver and storage device (34) includes a UAV storage compartment (341), an opening and closing drive (343), an automatic centering mechanism (344), and a wireless charging module (345). The side wall of the UAV storage compartment (341) is provided with a door (342). The opening and closing drive (343) is located above the door (342) and is used to drive the door (342) to open or close outward. The automatic centering mechanism (344) includes a guide ramp (3441) and a positioning slot (3442) located at the bottom of the UAV storage compartment (341). The wireless charging module (345) is located at the bottom of the positioning slot (3442). The environmental perception module (51) includes a light sensor array (511), a downward-facing camera (512), and a positioning module (513). The light sensor array (511) is installed on the upper and lower parts of the UAV (5), including an upper hemisphere light sensor group and a lower hemisphere light sensor group. The downward-facing camera (512) is installed on the bottom of the UAV (5). The positioning module (513) is used to record the three-dimensional coordinates of the UAV (5) at each sampling point.

5. A new energy storage-type unmanned aerial vehicle storage device according to claim 1, characterized in that: The generator box (31) also integrates a meteorological monitoring module (35), a communication module (36), an emergency power supply interface (37), a zone electrical parameter monitoring module (38), and a string power monitoring module (39); the meteorological monitoring module (35) includes a wind speed sensor, a temperature sensor, and an external meteorological cloud map data interface; the communication module (36) includes a 4G / 5G communication unit and a Beidou short message unit; the emergency power supply interface (37) is located on the outside of the generator box (31); the UAV (5) is also equipped with a communication relay module, the zone electrical parameter monitoring module (38) is used to monitor the real-time voltage and current of each zone of the multi-layer solar panel structure (12), and the string power monitoring module (39) is used to monitor the output power of each solar panel string.

6. A control method based on the device according to any one of claims 1-5, characterized in that, Includes the following steps: S1 Flight window determination: The control system (32) determines the optimal take-off time window of the UAV (5) based on the current state of charge and the predicted power generation curve of the energy storage module (33) and generates a flight mission plan; S2 Environmental Data Acquisition: The UAV (5) takes off to a preset height above the multi-layer solar panel structure (12), hovers at multiple sampling points along a preset path, and collects light intensity data, solar panel surface image data, surrounding environment image data and three-dimensional coordinates at each sampling point through the environmental perception module (51); S3 Data Processing and Modeling: The control system (32) constructs a three-dimensional illumination field model based on the light intensity data and three-dimensional coordinates of each sampling point; determines the cleanliness coefficient of each solar panel based on the surface image data of each solar panel; identifies obstructions based on the surrounding environment image data and generates a spatiotemporal obstruction prediction map in combination with the solar trajectory prediction model; S4 Optimization Calculation: Taking the maximization of total power generation as the optimization objective, the optimal tilt angle adjustment time series is calculated by combining the three-dimensional illumination field model, cleanliness coefficient, spatiotemporal shading prediction map and solar trajectory prediction data. S5 performs tilt adjustment: The control system (32) drives the angle adjustment component (4) to perform tilt adjustment on the solar bracket (11) according to the optimal tilt adjustment time sequence; S6 Feedback Correction: Compare the actual power generation of the previous cycle with the predicted power generation, and adjust and optimize the model parameters based on the deviation value.

7. The control method according to claim 6, characterized in that: In step S1, based on the current state of charge of the energy storage module (33) Predicted power generation curve (5) Energy consumption of a single mission for unmanned aerial vehicles (UAVs) and system base load power Calculate the available energy margin using the following formula. : ， in, This refers to the total capacity of the battery pack. To calculate the time range, The current moment; Calculate the solar altitude angle over time based on the current date and geographical location, and determine the takeoff time window based on the changes in the solar altitude angle; after confirmation... Greater than the safe energy margin Under the premise of [specific conditions], a flight mission plan including takeoff time, flight path and mission type is generated.

8. The control method according to claim 6, characterized in that: In step S3, determining the cleanliness coefficient includes: combining local images taken at multiple sampling points into a complete top-view image of each solar panel through image stitching; performing grayscale and texture analysis on the surface area of ​​each solar panel; and determining the cleanliness coefficient. Defined as the ratio of the current reflectivity of the panel surface to the factory standard reflectivity, with a value range of [value missing]. ; The generation of the spatiotemporal occlusion prediction map includes: semantic segmentation of the surrounding environment image, identification of the type and spatial location of occlusion objects, calculation of the shadow areas projected onto each solar panel by each occlusion object at different times using a solar trajectory prediction model, and output of a time series matrix. ,in For position coordinates, For time, The value is the occlusion coefficient at that location at that time. Indicates completely unobstructed. This indicates complete obstruction.

9. The control method according to claim 8, characterized in that: In step S4, the objective function for the optimization calculation is: ， in, The tilt angle of the solar panel bracket (11); This represents the total number of solar panels. For the first The nominal power of the solar panel; For the first A solar panel at temperature The efficiency coefficient at that point, ,in For temperature coefficient, For reference temperature, The real-time board surface temperature is obtained through a board surface temperature sensor. For the angle of inclination The lower sunlight and the first Angle of incidence of the normal to the solar panel; For the first Cleanliness coefficient of a solar panel; For the angle of inclination Next A solar panel at all times occlusion coefficient; The optimization calculation is also subject to the following constraints: Wind load safety constraints: when the real-time wind speed Exceeding the preset wind speed threshold hour, ,in ， This is the maximum permissible tilt angle in calm weather. This is the extreme wind speed; Temperature efficiency constraint: When the real-time plate temperature exceeds a preset temperature threshold, the objective function is modified to... ， in, This is the heat dissipation weighting coefficient. Inclination angle Estimated heat dissipation efficiency under the following conditions; Energy storage state constraints: Introducing state of charge adjustment weights The corrected objective function is ； When the state of charge is lower than the preset lower threshold, set To prioritize power generation; when the state of charge exceeds a preset upper limit threshold, a forced setting is implemented. ,in The preset safe windward tilt angle is used to reduce power generation and avoid overcharging; In step S6, the feedback correction includes: calculating the deviation rate of the previous period. ， in, This represents the actual amount of electricity generated. To predict power generation; when the deviation rate When the preset deviation threshold is exceeded, the light attenuation coefficient and temperature influence coefficient in the optimization model are corrected using the exponential weighted moving average method.

10. The control method according to claim 6, characterized in that: It also includes abnormal event triggering steps: in predicting charging power Greater than the preset minimum operating power Under the given conditions, the control system (32) monitors the charging power of the energy storage module (33) in real time and calculates the decrease in actual charging power relative to the predicted charging power. ； when When the preset ratio threshold is exceeded and the meteorological monitoring data does not show changes in cloud cover, the control system (32) retrieves the real-time electrical parameter data of the partition electrical parameter monitoring module (38) and string power monitoring module (39) set in the device to locate the suspected fault string or partition with abnormal power generation. Trigger the UAV (5) to take off additionally to perform a local scanning mission. The flight path of the local scanning mission covers the solar panel area corresponding to the suspected fault string or partition located above. The control system (32) comprehensively judges the cause of the abnormality based on the partial scan data and the real-time electrical parameter data: If a newly appearing occlusion is identified in the image data of a partial scan, the spatiotemporal occlusion prediction map is updated and the optimal tilt angle is recalculated. If the image data from a partial scan identifies contamination on the board surface, the cleanliness coefficient is updated and a cleaning work order is generated and sent to the remote operation and maintenance center via the communication module (36). If the image data does not show obvious obstruction and pollution, and the electrical parameter data of the partition electrical parameter monitoring module (38) or the string power monitoring module (39) shows open circuit, short circuit or abnormal voltage and current drop, then it is determined to be a hardware fault, and a fault alarm information is generated and sent to the operation and maintenance center through the communication module (36).

Images