Unmanned aerial vehicle based method and system for inspecting faults of a residential photovoltaic module

Abstract

The application discloses a kind of based on unmanned aerial vehicle's household photovoltaic module fault inspection method and system, and the present application relates to the technical field of household photovoltaic module fault inspection, according to the power supply efficiency record, the distribution map of household photovoltaic module and the identification of the model of household photovoltaic module Determine multiple fault photovoltaic modules;According to the current position of multiple fault photovoltaic modules, the distribution map of residential area and the flight environment of unmanned aerial vehicle determines the fault inspection route of unmanned aerial vehicle to multiple fault photovoltaic modules, guarantees the accuracy of fault inspection route.Therefore, the next fault photovoltaic module is dynamically generated in the process of flight Sub-fault report of unmanned aerial vehicle;When unmanned aerial vehicle completes the fault inspection of last fault photovoltaic module, unmanned aerial vehicle still stays in residential area until the completion of overall fault report, realizes the synchronous performance of unmanned flight and autonomous fault detection of unmanned aerial vehicle, realizes the online fault management and control of household photovoltaic module to unmanned aerial vehicle.

Description

Technical Field

[0001] This invention relates to the technical field of fault inspection of residential photovoltaic modules, and in particular to a method and system for fault inspection of residential photovoltaic modules based on unmanned aerial vehicles (UAVs). Background Technology

[0002] With the development of technology, residential photovoltaic (PV) modules are gradually being applied to residential areas and serve as independent units for each residence. These PV modules continue to supply power to the residence even under sunlight. However, in the current technology, the testing of each PV module is independent, and the existing multiple PV modules cannot be subjected to triggered fault inspections. This makes it impossible for drones to achieve online fault management of PV modules. Summary of the Invention

[0003] The purpose of this invention is to overcome the shortcomings of the prior art. This invention provides a method and system for fault inspection of residential photovoltaic modules based on unmanned aerial vehicles (UAVs).

[0004] This invention provides a method for fault inspection of residential photovoltaic modules based on unmanned aerial vehicles (UAVs), including:

[0005] During the use of various household photovoltaic modules in residential areas, power supply efficiency records of the household photovoltaic modules are collected.

[0006] Based on the power supply efficiency record sheet, the distribution map of the residential photovoltaic modules, and the identification of the model of the residential photovoltaic modules, multiple faulty photovoltaic modules were identified.

[0007] The fault inspection route for the drone to inspect the multiple faulty photovoltaic modules is determined based on the current location of the multiple faulty photovoltaic modules, the distribution map of the residential area, and the drone's flight environment.

[0008] The drone flies along the fault inspection route, and dynamically generates a sub-fault report for the next faulty photovoltaic module during the flight.

[0009] While the drone completes its inspection of the last faulty photovoltaic module, it remains in the residential area until the overall fault report is completed. The overall fault report is generated by synthesizing multiple sub-fault reports.

[0010] This invention provides a drone-based fault inspection system for residential photovoltaic (PV) modules. The drone-based residential PV module fault inspection system is applied to the aforementioned drone-based residential PV module fault inspection method. The drone-based residential PV module fault inspection system includes:

[0011] The data acquisition module is used to collect power supply efficiency records of household photovoltaic modules during their use in residential areas.

[0012] The fault identification module is used to identify multiple faulty photovoltaic modules based on the power supply efficiency record table, the distribution map of the residential photovoltaic modules, and the model identification of the residential photovoltaic modules.

[0013] The fault inspection route module is used to determine the fault inspection route of the UAV for multiple faulty photovoltaic modules based on the current location of multiple faulty photovoltaic modules, the distribution map of the residential area, and the flight environment of the UAV.

[0014] The sub-fault reporting module is used by the UAV to fly along the fault inspection route and dynamically generate a sub-fault report for the next faulty photovoltaic module during the flight.

[0015] The overall fault report module is used so that when the drone completes the fault inspection of the last faulty photovoltaic module, the drone remains in the residential area until the overall fault report is completed. The overall fault report is generated based on the synthesis of multiple sub-fault reports.

[0016] Compared with the prior art, the beneficial effects of the present invention are:

[0017] In this embodiment of the invention, the method described herein collects power supply efficiency records of household photovoltaic modules during their use in a residential area; based on these power supply efficiency records, a distribution map of the household photovoltaic modules, and the identification of the module models, multiple faulty photovoltaic modules are identified; based on the current locations of these faulty photovoltaic modules, the distribution map of the residential area, and the drone's flight environment, a fault inspection route for the drone is determined, ensuring the accuracy of the fault inspection route, and triggering fault inspections are performed on each household photovoltaic module in the residential area using the drone.

[0018] Therefore, the drone flies along the fault inspection route, dynamically generating a sub-fault report for the next faulty photovoltaic module during flight. When the drone completes the fault inspection of the last faulty photovoltaic module, it remains in the residential area until the overall fault report is completed. The overall fault report is generated based on the synthesis of multiple sub-fault reports. By making full use of the drone's processing of the sub-fault reports of the next faulty photovoltaic module during flight, the drone's unmanned flight and autonomous fault detection are carried out simultaneously, realizing online fault management of residential photovoltaic modules by the drone. Attached Figure Description

[0019] Figure 1 This is a flowchart illustrating the method for fault inspection of residential photovoltaic modules based on unmanned aerial vehicles (UAVs) in an embodiment of the present invention.

[0020] Figure 2This is a flowchart illustrating step S11 of the method for fault inspection of residential photovoltaic modules based on unmanned aerial vehicles in an embodiment of the present invention.

[0021] Figure 3 This is a flowchart illustrating step S12 in the method for fault inspection of residential photovoltaic modules based on unmanned aerial vehicles (UAVs) in an embodiment of the present invention.

[0022] Figure 4 This is a flowchart illustrating step S13 in the method for fault inspection of residential photovoltaic modules based on drones in an embodiment of the present invention.

[0023] Figure 5 This is a flowchart illustrating step S14 of the method for fault inspection of residential photovoltaic modules based on drones in an embodiment of the present invention.

[0024] Figure 6 This is a flowchart illustrating step S15 of the method for fault inspection of residential photovoltaic modules based on drones in an embodiment of the present invention.

[0025] Figure 7 This is a schematic diagram of the structural composition of a drone-based residential photovoltaic module fault inspection system according to an embodiment of the present invention. Detailed Implementation

[0026] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention.

[0027] Please see Figures 1 to 7 A method for fault inspection of residential photovoltaic (PV) modules based on drones is applied to a scenario of fault inspection of residential PV modules based on drones. This method includes:

[0028] Step S11: During the use of each household photovoltaic module in the residential area, collect the power supply efficiency record table of the household photovoltaic modules;

[0029] Step S12: Identify multiple faulty photovoltaic modules based on the power supply efficiency record table, the distribution map of the residential photovoltaic modules, and the identification of the model of the residential photovoltaic modules;

[0030] Step S13: Determine the fault inspection route of the UAV for the multiple faulty photovoltaic modules based on the current location of the multiple faulty photovoltaic modules, the distribution map of the residential area, and the flight environment of the UAV.

[0031] Step S14: The drone flies along the fault inspection route, and dynamically generates a sub-fault report for the next faulty photovoltaic module during the flight.

[0032] Step S15: When the drone completes the fault inspection of the last faulty photovoltaic module, the drone remains in the residential area until the overall fault report is completed. The overall fault report is generated based on the synthesis of multiple sub-fault reports.

[0033] refer to Figure 2 In step S11, during the use of each household photovoltaic module in the residential area, a power supply efficiency record table of the household photovoltaic module is collected.

[0034] In the specific implementation of this invention, the specific steps are as follows:

[0035] S111: Collect the location of the residential area and determine the individual household photovoltaic modules in the residential area based on the location of the residential area, the distribution map of the residential area, and the photovoltaic power supply path;

[0036] S112: Monitor the usage of each household photovoltaic module in real time, collect the power supply efficiency of each household photovoltaic module in the residential area, and determine the power supply efficiency record table of the household photovoltaic modules based on the power supply efficiency of each household photovoltaic module in the residential area, the corresponding house, and the real-time power consumption of the house.

[0037] In the embodiments of this application, the location of the residential area is collected, and each household photovoltaic module in the residential area is determined based on the location of the residential area, the distribution map of the residential area, and the photovoltaic power supply path. This takes into account the overall consideration of the location of the residential area, the distribution map of the residential area, and the photovoltaic power supply path, ensuring the accuracy of each household photovoltaic module in the residential area.

[0038] At this point, a Geographic Information System (GIS) is used to obtain the geographic coordinates of the residential area to collect the regional location of the residential area. At the same time, the regional location of the residential area, the distribution map of the residential area, and the photovoltaic power supply path are introduced. Combined with the distribution map of the residential area, the location and layout of each residence are identified. Using the photovoltaic power supply path map or design drawings, the number of household photovoltaic modules, installation location, and connection method of each residence are determined.

[0039] Specifically, assuming the residential area is a community consisting of 10 residential buildings, each with 6 floors and 2 households per floor, totaling 120 households, a distribution map of the residential area obtained from the urban planning department clearly shows the location and layout of each building. Next, a photovoltaic power supply path map is obtained from the photovoltaic system integrator, which details the installation location (e.g., rooftop), quantity (e.g., how many photovoltaic panels are installed in each building), and how they are connected to the grid or energy storage system for each residential building. During a site survey, it was found that the roof of one building was under repair, resulting in the temporary removal of some photovoltaic modules. Therefore, the record sheet indicates a reduction in the number of photovoltaic modules in this building, and the photovoltaic power supply path map is updated. The regional location of the residential area was collected, and based on the distribution map and the photovoltaic power supply path map, the specific location and connection of each household photovoltaic module within the residential area were determined.

[0040] Furthermore, the system monitors the usage of each household photovoltaic module in real time and collects the power supply efficiency of each household photovoltaic module in the residential area. Based on the power supply efficiency of each household photovoltaic module in the residential area, the corresponding residence, and the real-time power consumption of that residence, a power supply efficiency record table for the household photovoltaic modules is determined. This system takes into account the overall power supply efficiency of each household photovoltaic module in the residential area, the corresponding residence, and the real-time power consumption of that residence, ensuring the accuracy of the power supply efficiency record table for the household photovoltaic modules.

[0041] At this time, real-time operating data of photovoltaic modules, such as current, voltage, and power, are acquired to monitor their working status and performance. Simultaneously, smart sensors or data acquisition devices are installed on each photovoltaic module to monitor its operating parameters in real time. Using Internet of Things (IoT) technology, the monitoring data is transmitted in real time to a central monitoring system or cloud platform. The monitoring system or cloud platform processes and analyzes the received data to identify abnormalities or faults.

[0042] The power supply efficiency of a photovoltaic (PV) module, which is its ability to convert solar energy into electrical energy, is calculated based on the real-time monitored output power of the PV module (such as DC power or AC power) and the input solar radiation (such as measured by an irradiance sensor). The power supply efficiency is expressed as a percentage, and the calculation formula is: Power supply efficiency = Output power / (Input radiation × Conversion efficiency coefficient), where the conversion efficiency coefficient is an inherent performance parameter of the PV module.

[0043] Specifically, suppose there is a community with 100 households, each of which is equipped with photovoltaic modules. In order to monitor the usage of these photovoltaic modules in real time and collect power supply efficiency data, the following measures are taken: smart sensors are installed on each photovoltaic module to monitor current, voltage and power parameters in real time. These sensors transmit data to a central monitoring system via wireless networks (such as LoRa, Zigbee or 4G / 5G).

[0044] A central monitoring system was established within the community. This system can receive and process data from all sensors. It can display the real-time operating status of each photovoltaic module, including power output, voltage, and current. Simultaneously, an irradiance sensor is integrated into the system to measure the amount of solar radiation received by each photovoltaic module. Based on the real-time monitored output power and input radiation, the system automatically calculates the power supply efficiency of each photovoltaic module. For example, if a household's photovoltaic modules receive 1000 W·h / m² of radiation in a day, and its real-time output power is 150 W (average), then the power supply efficiency is 15% (assuming a conversion efficiency coefficient of 1).

[0045] In addition, a comprehensive assessment of the power supply capacity of photovoltaic modules and the power consumption of residences is conducted to generate a detailed power supply efficiency record table; the power supply efficiency of each photovoltaic module is associated with its corresponding residence; real-time power consumption data of each residence is collected through smart meters or energy management systems; and the power supply efficiency of photovoltaic modules, the power consumption of residences, and timestamp information are integrated into a power supply efficiency record table.

[0046] Specifically, in the central monitoring system, each residence is assigned a unique identifier and associated with the corresponding photovoltaic module; by connecting with the community's smart meter system, the power consumption data of each residence can be obtained in real time; the system automatically integrates the power supply efficiency of each photovoltaic module, the power consumption of the corresponding residence, and timestamp information into a power supply efficiency record table.

[0047] For example, a record shows that the photovoltaic module power supply efficiency of a certain household on a certain day is 15%, while the household's electricity consumption on that day is 10 kWh; by comparing these two data, the effectiveness of the photovoltaic module in meeting the household's electricity demand can be evaluated.

[0048] In one embodiment of this application, to clearly demonstrate the relationship between the power supply efficiency of photovoltaic modules and residential power consumption, a power supply efficiency matching table is created; the power supply efficiency matching table is shown in Table 1:

[0049] Table 1 Power Supply Efficiency Matching Table

[0050]

[0051] In this power supply efficiency matching table, each row represents a specific point in time, including the house number, photovoltaic module number, power supply efficiency, real-time power consumption, and timestamp.

[0052] refer to Figure 3 In step S12, multiple faulty photovoltaic modules are identified based on the power supply efficiency record table, the distribution map of the household photovoltaic modules, and the identification of the model of the household photovoltaic modules.

[0053] In the specific implementation of this invention, the specific steps are as follows:

[0054] S121: In this power supply efficiency record table, multiple abnormal parameters are determined based on the autonomous detection of the power supply efficiency record table, and each abnormal parameter combination is constructed on the basis of a residence. The corresponding abnormal level is determined based on the identification of the abnormal parameter combination of the residence.

[0055] S121: Based on the identification of the anomaly level and the corresponding photovoltaic module model, a first fault parameter is determined; based on the identification of the anomaly level and the distribution map of the residential photovoltaic modules, a second fault parameter is determined; and multiple faulty photovoltaic modules are determined according to the first fault parameter, the second fault parameter, and a preset fault parameter matching table.

[0056] In the embodiments of this application, multiple abnormal parameters are determined based on the autonomous detection of the power supply efficiency record table, and each abnormal parameter combination is constructed on a residential basis. The corresponding abnormal level is determined based on the identification of the abnormal parameter combination of the residential, thus ensuring the accuracy of the abnormal level.

[0057] At this point, the power supply efficiency record table is scanned and analyzed to identify parameters that differ significantly from normal or expected values. These significant differences manifest as a sudden drop in power supply efficiency, increased fluctuations, or prolonged periods below a certain threshold.

[0058] The identified anomalous parameters are categorized according to residences. For multiple anomalous parameters within the same residence, they are combined to generate an anomalous parameter set. Such combinations help to understand the correlation and common causes between anomalies. At the same time, the severity of the anomalies is assessed based on the number, type, severity, and correlation between the anomalous parameters. Different anomalous levels (such as mild, moderate, and severe) are set, and clear judgment criteria are defined for each level. Based on these criteria, a corresponding anomalous level is assigned to the combination of anomalous parameters for each residence.

[0059] Specifically, suppose there is a community with 10 residential units, each of which is equipped with photovoltaic (PV) modules, and a power efficiency record table has been established to monitor the performance of these modules. When analyzing the power efficiency record table, it was found that the power efficiency of the PV modules in residential unit 003 has been declining continuously over the past week, from the normal 16% to 10%. At the same time, the power efficiency of the modules in residential unit 005 fluctuates significantly, sometimes high and sometimes low, without a stable trend.

[0060] For residential building 003, the abnormal parameter "continuous decline in power supply efficiency" is listed separately to generate an abnormal parameter set; for residential building 005, the abnormal parameter "increased fluctuation in power supply efficiency" is listed separately to generate another abnormal parameter set.

[0061] For the abnormal parameter set of Residential 003, its severity is assessed as "moderate" because the continuous decline in power supply efficiency indicates some kind of fault or aging of the components, requiring further investigation. For the abnormal parameter set of Residential 005, its severity is assessed as "slight" because although the increased fluctuations indicate unstable component performance, they have not yet reached the level of seriously affecting power supply efficiency. Therefore, at the end of step S121, a result containing two residential units and their corresponding abnormality levels is obtained: the abnormality level of Residential 003 is "moderate," and the abnormal parameter is "continuous decline in power supply efficiency"; the abnormality level of Residential 005 is "slight," and the abnormal parameter is "increased fluctuations in power supply efficiency." This information will provide important reference for subsequent steps (such as fault inspection and maintenance plan development).

[0062] Furthermore, a first fault parameter is determined based on the anomaly level and the corresponding photovoltaic module model, and a second fault parameter is determined based on the anomaly level and the distribution map of the residential photovoltaic modules. Multiple faulty photovoltaic modules are then identified according to the first fault parameter, the second fault parameter, and a preset fault parameter matching table. This approach takes into account the first fault parameter, the second fault parameter, and the overall considerations of the residential building, ensuring the accuracy of identifying multiple faulty photovoltaic modules.

[0063] At this point, a first fault parameter and a second fault parameter are introduced. For the first fault parameter, it is determined based on the anomaly level and the corresponding photovoltaic module model. Simultaneously, the anomaly level determined in step S121 is reviewed to understand the severity of faults in each residential photovoltaic module. Different fault modes and characteristics of different photovoltaic module models are considered; some models are more susceptible to specific environmental factors or have specific failure rates. Combining the anomaly level and photovoltaic module model information, a matching process using a photovoltaic module database is employed to determine the first fault parameter, which requires referencing historical fault data.

[0064] For the second fault parameter, it is determined based on the anomaly level and the distribution map of the residential photovoltaic modules. At the same time, the anomaly level information is obtained from step S121. The distribution map is used to view the installation location, orientation, surrounding environment (such as obstructions, trees, buildings), and the connection relationships between them of the photovoltaic modules. Combining the anomaly level and distribution map information, the source of the fault is analyzed. For example, if the photovoltaic modules in a certain area generally experience a decrease in efficiency, it is because the sunlight in that area is blocked or the modules are contaminated. Based on the analysis results, the second fault parameter most relevant to the anomaly level and distribution map is determined.

[0065] The first and second fault parameters are combined with residential information for a comprehensive assessment. Based on the assessment results, it is determined which photovoltaic modules are most likely to be faulty, which involves further inspection or testing of multiple modules. Finally, all identified faulty photovoltaic modules are listed, including their serial numbers, locations, models, and causes of failure.

[0066] Specifically, suppose there is a community with 5 residential houses, each with the same type of photovoltaic modules installed, and step S121 has been completed, determining the anomaly level of each house. For the first fault parameter, the anomaly level of houses 001 and 002 is "moderate", house 003 is "slight", and houses 004 and 005 are "normal". All houses are equipped with XYZ type photovoltaic modules. According to historical data, XYZ type modules are prone to efficiency degradation under prolonged high-temperature conditions. Considering that the anomaly level of houses 001 and 002 is "moderate" and they are located in areas of the community with direct sunlight and poor ventilation, it is initially determined that the first fault parameter is the degradation of module efficiency caused by high temperature.

[0067] Regarding the second fault parameter, the anomaly level of residences 001 and 002 is "moderate", residence 003 is "slight", and residences 004 and 005 are "normal". When reviewing the distribution map, it was found that the photovoltaic modules of residence 003 were partially blocked by a tall tree. Combining the anomaly level (slight for residence 003) and the distribution map information, it was determined that the second fault parameter was caused by the reduction in module efficiency due to tree shading.

[0068] Based on the combined information from the first and second fault parameters, it was determined that the modules in residences 001 and 002 failed due to high temperatures, while the module in residence 003 failed due to tree shading. Further inspection revealed that the modules in residences 001 and 002 did indeed exhibit excessively high temperatures, while the module in residence 003 showed a significant decrease in efficiency in the shaded areas. Finally, the following faulty photovoltaic modules were listed:

[0069] Residential Unit 001: XYZ model photovoltaic modules, the cause of failure is reduced efficiency due to high temperature;

[0070] Residential Unit 002: XYZ model photovoltaic modules, the cause of failure is reduced efficiency due to high temperature;

[0071] Residential Unit 003: XYZ model photovoltaic module (partially obscured by trees), the cause of failure is reduced efficiency due to tree obstruction.

[0072] In one embodiment of this application, the anomaly level weight is 0.5 (because the anomaly level directly reflects the severity of the problem); the photovoltaic module model weight is 0.3 (because different models have different failure modes); the distribution map feature weight is 0.2 (because the distribution map provides intuitive information about the module installation environment and conditions); taking residential unit 001 as an example, the corresponding fault parameter scores are calculated as follows:

[0073] For the first fault parameter "cell aging or damage", since the abnormality level is "medium" (high score) and the photovoltaic module model is A, a high comprehensive score is assigned to it;

[0074] For the second fault parameter "local shadowing effect", since the anomaly level is "medium" (high score) and the distribution map shows that the component is slightly occluded, it is also assigned a high comprehensive score;

[0075] Finally, based on the overall score, the system determines which fault parameters have the highest probability of failure and further identifies which photovoltaic modules are faulty. This method provides a more refined evaluation tool, which helps to more accurately identify and resolve photovoltaic module failures.

[0076] refer to Figure 4 In step S13, the fault inspection route of the UAV for the multiple faulty photovoltaic modules is determined based on the current location of the multiple faulty photovoltaic modules, the distribution map of the residential area, and the flight environment of the UAV.

[0077] In the specific implementation of this invention, the specific steps are as follows:

[0078] S131: Collect the corresponding fault types based on the traversal of multiple faulty photovoltaic modules, and determine multiple fault distribution areas based on the current location of multiple faulty photovoltaic modules, the floor of the residence, and the corresponding fault types. The fault distribution area is a multi-dimensional space.

[0079] S132: Collect multiple flight environment parameters based on the UAV's initial flight over multiple fault distribution areas, and determine the UAV's flight environment based on the multiple flight environment parameters and the aerial images detected by the UAV.

[0080] S133: Based on multiple fault distribution areas and the flight environment of the UAV, determine the first inspection route for each fault distribution area; based on the distribution map of multiple fault distribution areas and residential areas, determine the second inspection route for each fault distribution area; and based on the synthesis of the first inspection route and the second inspection route, determine the fault inspection route of the UAV for multiple faulty photovoltaic modules.

[0081] In the embodiments of this application, the corresponding fault types are collected by traversing multiple faulty photovoltaic modules, and multiple fault distribution areas are determined based on the current location of multiple faulty photovoltaic modules, the floor of the residence, and the corresponding fault type. The fault distribution area is a multi-dimensional space that takes into account the overall consideration of the current location of multiple faulty photovoltaic modules, the floor of the residence, and the corresponding fault type, thus ensuring the accuracy of the multiple fault distribution areas.

[0082] At this point, the multiple faulty photovoltaic modules identified in the previous steps are inspected one by one. This is done through on-site inspection, remote monitoring data, or historical maintenance records. During the process, the specific fault type of each faulty photovoltaic module is recorded. Fault types include, but are not limited to, cell aging, loose wiring, inverter failure, glass breakage, and dust accumulation. The collected fault type information is organized into a table or database for subsequent analysis and processing.

[0083] Obtain the precise location information of each faulty photovoltaic module, including latitude and longitude coordinates, distance and direction relative to a certain point; for multi-story residential buildings, record the floor where each faulty module is located, which helps to understand whether the fault is related to factors such as floor height and angle of sunlight; analyze the geographical distribution characteristics of different types of faults; for example, some faults are more common on lower floors or in areas prone to shading.

[0084] Based on location, floor, and fault type information, the faulty photovoltaic modules are divided into multiple fault distribution areas using a Geographic Information System (GIS) or similar tools. These areas are multi-dimensional spaces that consider not only geographical location but also floor and fault type dimensions. Each fault distribution area is named and marked on a map using color, shape, or other visual elements for subsequent analysis and processing.

[0085] Specifically, suppose a multi-story residential area has 10 residential buildings, each with 6 floors, and each floor is equipped with photovoltaic modules. Through the previous steps, multiple faulty photovoltaic modules were identified, and information on their fault types was collected. Ten modules failed due to cell aging, and they were distributed across different floors of 5 buildings. In addition, 5 modules failed due to loose wiring, and they were mainly concentrated in the lower floors of 2 buildings. There were also 3 modules whose efficiency decreased due to dust accumulation, and they were located on the top floors of several buildings near the road.

[0086] The fault distribution areas were divided into three categories: first, the aging area of ​​the battery cells, involving different floors of 5 buildings; second, the area of ​​loose wiring, mainly concentrated on the lower floors of 2 buildings; and third, the dust-covered area, located on the top floors of several buildings near the road. At the same time, each area was further subdivided, for example, according to the floor height or the location of the specific faulty component.

[0087] On the GIS map, different colors or shapes are used to represent the three fault distribution areas, and the specific location of each faulty component is marked on the map. In this way, the distribution of faults in geographic space can be seen intuitively, providing an important reference for subsequent handling and maintenance. Multiple faulty photovoltaic modules have been successfully divided into multiple fault distribution areas, and detailed descriptions and labels have been provided for each area. This will help in the planning and optimization of drone inspection routes in subsequent steps.

[0088] Furthermore, based on the initial flight of the UAV to multiple fault distribution areas, multiple flight environment parameters are collected, and the UAV's flight environment is determined according to the multiple flight environment parameters and the aerial images detected by the UAV. This overall consideration of multiple flight environment parameters and aerial images detected by the UAV ensures the accuracy of the UAV's flight environment.

[0089] Before the initial flight, ensure the drone is in good condition, the battery is fully charged, and the camera and sensor devices are working properly. At the same time, depending on the characteristics of the flight area, specific sensors need to be installed to collect specific environmental parameters. Develop a detailed flight plan for each fault distribution area, including takeoff point, flight altitude, flight speed, and flight path. The flight plan should ensure that the drone can safely and efficiently cover the entire area.

[0090] According to the flight plan, the drone will conduct an initial flight over each fault distribution area. During the flight, the sensors on the drone will collect flight environment parameters in real time, such as wind speed, wind direction, temperature, air pressure, humidity, and visibility. These parameters are crucial for assessing the safety of the flight environment and the flight performance of the drone. The collected flight environment parameters will be recorded in the drone's built-in storage device or a remote server for subsequent analysis and processing.

[0091] Data analysis of collected flight environment parameters is performed to assess the safety of the flight environment and the flight performance of the UAV. For example, wind speed and direction are analyzed to determine the risk of turbulence or sudden changes in wind direction; temperature and air pressure are analyzed to predict weather changes; visibility is analyzed to determine the field of view of the UAV camera; at the same time, aerial images taken by the UAV are used to analyze the terrain, landforms, buildings, and vegetation features of the flight area, which are crucial for identifying potential flight obstacles and planning safe flight routes.

[0092] By analyzing the combined flight environment parameters and aerial images, the flight environment of the UAV is determined. This includes a safety assessment of the flight area, a prediction of the UAV's flight performance, and information on potential flight obstacles and risk factors. The assessment results of the flight environment are compiled into a report, which includes the safety level of the flight environment, the prediction of the UAV's flight performance, and information on potential flight obstacles and risk factors. This report will provide an important reference for subsequent UAV inspection route planning and flight operations.

[0093] Specifically, suppose a drone is being used for inspection in a multi-story residential area. This area has multiple fault distribution areas, and the drone needs to conduct an initial flight over these areas to collect flight environment parameters. The drone is equipped with an anemometer, thermometer, barometer, and camera equipment, and a detailed flight plan has been developed, including the takeoff point, flight altitude (set to 10 meters), flight speed (set to 5 meters per second), and flight path.

[0094] The drone conducted its initial flight over each fault distribution area according to the flight plan. During the flight, sensors collected environmental parameters such as wind speed, wind direction, temperature, air pressure, and humidity in real time, and captured aerial images. This data was recorded in the drone's built-in storage. Data analysis of the collected environmental parameters revealed that some areas had high wind speeds, posing a risk of turbulence. Simultaneously, aerial images showed protruding antennas or billboards on the roofs of certain buildings, posing flight obstacles. Based on the comprehensive data analysis, the drone's flight environment was determined. In areas with high wind speeds, the drone needed to reduce its flight speed to minimize the impact of turbulence. In areas with potential flight obstacles, the drone needed to plan safe flight routes to avoid these obstacles. Finally, the flight environment assessment results were compiled into a report, providing important reference for subsequent drone inspection route planning and flight operations. The successful collection of multiple flight environment parameters and the determination of the drone's flight environment will help in the planning and optimization of drone inspection routes in subsequent steps, ensuring that the drone can perform inspection tasks safely and efficiently.

[0095] Therefore, based on multiple fault distribution areas and the UAV's flight environment, a first inspection route is determined for each fault distribution area. Based on the distribution map of multiple fault distribution areas and residential areas, a second inspection route is determined for each fault distribution area. The UAV's fault inspection route for multiple faulty photovoltaic modules is determined by combining the first and second inspection routes. This achieves the synthesis of the first and second inspection routes, further enabling precise control of the UAV's fault inspection route for multiple faulty photovoltaic modules, ensuring the accuracy of the fault inspection route, and enabling triggered fault inspection of each household photovoltaic module in the residential area by the UAV.

[0096] At this point, a first inspection route and a second inspection route are introduced. For the first inspection route, based on multiple fault distribution areas and the UAV's flight environment, the first inspection route for each fault distribution area is determined. At the same time, the multiple fault distribution areas identified in the previous steps are reviewed to understand the location, size, shape, and number of faulty photovoltaic modules contained in each area. Combining the UAV's flight environment parameters (such as wind speed, wind direction, and visibility) and aerial image analysis results, the flight difficulty and safety risks of different fault distribution areas are assessed. Based on the characteristics of the fault distribution areas and the assessment results of the flight environment, a preliminary first inspection route is planned for each fault distribution area. The first inspection route should ensure that the UAV can safely and efficiently cover all faulty photovoltaic modules while avoiding potential flight obstacles and risky areas.

[0097] For the second inspection route, based on the distribution maps of multiple fault distribution areas and residential areas, the second inspection route for each fault distribution area is determined. At this time, a detailed distribution map of the residential area is obtained to understand the layout of buildings, road networks, and information on people's activity areas. The fault distribution area and residential area distribution maps are overlaid and analyzed to identify the spatial relationship between faulty photovoltaic modules and key elements of residential buildings and roads. When planning the inspection route, the safety and privacy of residents are fully considered. Drones are avoided from staying or flying low over densely populated or sensitive areas for extended periods. Based on the analysis results of the residential area distribution map and the consideration of residents' safety and privacy, a second inspection route that is more in line with the actual situation is planned. The second inspection route should minimize the impact on residents' lives while ensuring that drones can successfully reach each fault distribution area.

[0098] Furthermore, the first and second inspection routes are compared and analyzed to identify their differences and commonalities. The first and second inspection routes are then subjected to multiple interactions and integrations. This process aims to comprehensively consider factors such as flight environment, fault distribution, residential area layout, and resident safety and privacy protection to determine the optimal inspection route. After multiple interactions and integrations, a fault inspection route for multiple faulty photovoltaic modules by the UAV is determined. This route should be both efficient and safe, ensuring that the UAV can successfully reach each fault distribution area and complete the inspection task.

[0099] Specifically, suppose we are conducting drone inspections in a multi-story residential area. This area has multiple fault distribution zones, and the drone's flight environment parameters and the distribution map of the residential area are known. The first inspection route planning is as follows: Based on the characteristics of the fault distribution zones and the assessment results of the flight environment, we initially plan the inspection route in each fault distribution zone. For example, in areas with high wind speeds, we plan a lower altitude flight route to reduce the impact of turbulence; in areas with dense buildings, we plan a route that bypasses the buildings to avoid collisions.

[0100] The second inspection route planning involves identifying the spatial relationship between faulty photovoltaic modules and key elements such as residential buildings and roads, based on the distribution map of the residential area. When planning the route, drones are avoided flying over areas with frequent resident activity to reduce the impact on residents' lives. At the same time, considering that some roads serve as take-off and landing points for drones, the route is planned to be closer to these roads.

[0101] After comparing and analyzing the first and second inspection routes, it was found that they conflicted or overlapped in some areas. By adjusting the methods or through manual intervention, the routes in these areas were integrated and optimized. Finally, an inspection route was determined that took into account the flight environment, fault distribution, residential area layout, and the protection of residents' safety and privacy. This route will serve as the basis for the UAV to perform inspection tasks. The UAV successfully planned a fault inspection route for multiple faulty photovoltaic modules. This route comprehensively considers multiple factors to ensure that the UAV can perform inspection tasks safely and efficiently.

[0102] refer to Figure 5 In step S14, the UAV flies along the fault inspection route and dynamically generates a sub-fault report for the next faulty photovoltaic module during the flight.

[0103] In the specific implementation of this invention, the specific steps are as follows:

[0104] S141: Based on the fault inspection route and the current location of multiple faulty photovoltaic modules, determine multiple fault review nodes, and determine the dynamic review method of the UAV at each fault review node according to the multiple fault review nodes, the fault type of the faulty photovoltaic module, the surrounding environment of the faulty photovoltaic module, and the preset review method matching table.

[0105] S142: In a fault review node, the UAV reviews the faulty photovoltaic module based on the corresponding dynamic review method and collects multiple images of the faulty photovoltaic module;

[0106] S143: Based on multiple images of the faulty photovoltaic module, the power supply efficiency of the faulty photovoltaic module, and the power supply path of the faulty photovoltaic module, determine the sub-fault report of the faulty photovoltaic module. The drone completes the output of the sub-fault report of the faulty photovoltaic module while flying on the path to the next faulty photovoltaic module.

[0107] In the embodiments of this application, multiple fault re-inspection nodes are determined based on the fault inspection route and the current location of multiple faulty photovoltaic modules. The dynamic re-inspection method of the UAV at each fault re-inspection node is determined according to the multiple fault re-inspection nodes, the fault type of the faulty photovoltaic module, the surrounding environment of the faulty photovoltaic module, and a preset re-inspection method matching table. This takes into account the overall consideration of multiple fault re-inspection nodes, the fault type of the faulty photovoltaic module, and the surrounding environment of the faulty photovoltaic module, and ensures the accuracy of the dynamic re-inspection method of the UAV at each fault re-inspection node.

[0108] At this point, reviewing the previously determined fault inspection route, which has taken into account the drone's flight environment, fault distribution area, and residential area factors, ensures that the drone can safely and efficiently reach the location of each faulty photovoltaic module. Based on the latest monitoring data or on-site survey results, confirm the current location of multiple faulty photovoltaic modules. This location information is updated due to changes in the layout of the photovoltaic power station, module movement, or the occurrence of new faults.

[0109] By matching the location of the faulty photovoltaic module with the inspection route, the key nodes that the drone needs to stop and re-examine during the inspection process can be identified. These key nodes are located above or near the faulty photovoltaic module, which facilitates detailed observation and image acquisition by the drone.

[0110] Furthermore, a detailed analysis of the fault types of each faulty photovoltaic module is conducted to understand the specific manifestations, causes, and impacts on power plant operation. Common fault types include module damage, shading, and poor wiring. The surrounding environment of the faulty photovoltaic module is examined, including terrain, climate, buildings, and vegetation, as these factors affect the UAV's flight safety, image acquisition quality, and the selection of review methods. Based on the analysis of fault types and the surrounding environment, dynamic review methods are developed for the UAV at each review node, including adjustments to flight altitude, speed, heading, and image acquisition parameters (such as exposure time, focal length, and resolution). The review methods should ensure that the UAV can safely and accurately acquire detailed images and information of the faulty photovoltaic module.

[0111] Specifically, suppose that in a residential area, a drone has flown along a preset inspection route and discovered three faulty photovoltaic modules, located in different areas of the power station. Through GPS positioning and on-site investigation, the precise locations of these three faulty modules are confirmed. Subsequently, three re-inspection nodes are marked on the inspection route, corresponding to the locations of the three faulty modules.

[0112] Of the three previously identified inspection points, the first point involved a photovoltaic module with a shading issue. On-site investigation revealed the obstruction to be a tall tree. Considering the tree's impact on drone flight safety and the need for image acquisition, the drone hovered at low altitude at the first inspection point, adjusting the camera's focus and exposure time to ensure clear images of the obstructed photovoltaic module. For the second and third points, different inspection methods were chosen based on the type of fault and the surrounding environment. For example, the second point involved a faulty module with poor wiring, so the drone hovered at close range and activated infrared thermal imaging. The third point, due to high-voltage power lines nearby, required a higher flight altitude and a safer heading for inspection.

[0113] Furthermore, at a fault review node, the UAV reviews the faulty photovoltaic module based on the corresponding dynamic review method and collects multiple images of the faulty photovoltaic module, thus introducing multiple images of the faulty photovoltaic module.

[0114] At this point, the drone accurately flies to the designated fault review node location according to the preset inspection route and review node plan. This process requires the drone to have accurate navigation and positioning capabilities. After arriving at the node, the drone begins to perform the review task according to the previously determined dynamic review method. This includes adjusting the flight altitude, speed, and direction, as well as activating specific image acquisition parameters. While performing the review method, the drone prepares the image acquisition equipment to ensure that the camera or sensor is in the best working condition so as to capture high-quality images.

[0115] To ensure a comprehensive and accurate reflection of the condition of faulty photovoltaic modules, drones need to collect images from multiple angles, including front, side, and top views. When collecting images, drones need to ensure that the image clarity, contrast, and color reproduction meet the required quality standards, which requires adjusting the camera's exposure time, focal length, and aperture parameters. The collected images need to be stored in the drone's internal memory or cloud storage system in a timely manner.

[0116] Specifically, suppose in a photovoltaic power station, a drone has flown to a designated fault re-inspection node according to a preset inspection route; this node corresponds to a photovoltaic module with a crack; based on the previous analysis, the drone chose a dynamic re-inspection method of hovering at low altitude and adjusting the camera focus and exposure time.

[0117] The drone flew precisely above the faulty component and began performing the pre-set inspection. It lowered its altitude, hovering about 5 meters above the ground, while adjusting the camera's focus to concentrate on the crack and appropriately increasing the exposure time to ensure the crack was clearly visible in the image.

[0118] The drone captured images of the faulty component from three angles: front, side, and top. Multiple photos were taken from each angle to ensure that different details of the crack could be captured. The captured images were of high clarity, with the details of the crack clearly visible and good color reproduction. The captured images were promptly stored in the drone's internal memory.

[0119] Therefore, based on multiple images of the faulty photovoltaic module, its power supply efficiency, and its power supply path, a sub-fault report for the faulty photovoltaic module is determined. The drone completes the output of the sub-fault report for the faulty photovoltaic module while flying on its way to the next faulty photovoltaic module. This comprehensive approach, which considers multiple images of the faulty photovoltaic module, its power supply efficiency, and its power supply path, ensures the accuracy of the sub-fault report for the faulty photovoltaic module.

[0120] At this point, the images of multiple faulty photovoltaic modules collected by the UAV are preprocessed to improve image quality and facilitate subsequent analysis; image recognition technology is used to identify fault features in the images, including cracks, damage, stains, and obstructions; based on the identified fault features, the severity of the fault is assessed, which involves quantitative analysis of the size, quantity, and distribution of the fault features.

[0121] The power supply efficiency data of faulty photovoltaic modules are obtained from the monitoring system of the photovoltaic power plant. This data includes key parameters such as power output, current, and voltage. The obtained power supply efficiency data is compared with that of normal modules to analyze the impact of the fault on the module's power supply efficiency. This involves calculating power loss and efficiency reduction percentage indicators. Combined with the results of image analysis, the correlation between fault characteristics and power supply efficiency reduction is explored, which helps to understand the impact of the fault on module performance more deeply.

[0122] The analysis involves determining the location and role of the faulty photovoltaic (PV) module within the power supply path of the power plant. This includes identifying whether the faulty module is on the critical path and assessing its impact on the overall power supply system. Based on the power supply path analysis, the potential risks posed by the fault are evaluated, considering grid stability and the risk of cascading failures of other components. A detailed sub-fault report is compiled, integrating the results of image analysis, power supply efficiency analysis, and power supply path analysis. The report should include a fault description, fault severity, impact on power supply efficiency, potential risks, and recommended remedial measures. As the drone flies to the next faulty PV module, wireless communication technology is used to transmit the sub-fault report to the ground control station or data analysis center in real time, ensuring the timeliness and accuracy of the information.

[0123] Specifically, suppose that in a photovoltaic power station, a drone has completed image acquisition and power supply efficiency data acquisition of a faulty photovoltaic module; through image processing technology, the drone identifies multiple cracks on the module, and the length and width of the cracks both exceed the preset thresholds, indicating that the fault is relatively serious; compared with normal modules, the power output of the faulty module is reduced by about 20% and the efficiency is reduced by about 15%, indicating that the fault has a significant impact on the power supply efficiency of the module.

[0124] The faulty component was located on a critical power supply path of the power plant. Failure to repair it promptly would lead to a decrease in the overall stability of the power supply system and could even trigger a cascading failure of other components. Based on this analysis, the drone generated a detailed sub-fault report. The report indicated that the faulty component had multiple severe cracks, resulting in a significant reduction in power output and efficiency, and that the fault was located on a critical power supply path, posing a potential risk. Immediate repair measures were recommended, such as replacing the faulty component or performing necessary maintenance. While the drone was flying towards the next faulty component, the sub-fault report was transmitted in real-time to the ground control station. Staff immediately received the report and took appropriate action. Through this series of steps, the drone was able to accurately identify the sub-fault report of the faulty photovoltaic component and complete the report output while flying towards the next faulty component, ensuring the timeliness and accuracy of the information.

[0125] In one embodiment of this application, a fault image matching table is introduced to match fault image features, power supply efficiency changes, and power supply path information with predefined fault types and severity levels; the image matching table is shown in Table 2:

[0126] Table 2 Image Matching Table

[0127]

[0128] Suppose that the image of the faulty photovoltaic module captured by the drone shows obvious cracks, a 25% reduction in power supply efficiency, and that the module is located on the critical power supply path of the power plant; according to the matching table, the fault type is determined to be "module damage," and the severity is "severe"; based on the above matching results, the drone generates the following sub-fault report:

[0129] Sub-fault report

[0130] Faulty photovoltaic module ID: XXXXXX;

[0131] Fault type: Component damage;

[0132] Severity: Severe;

[0133] Fault description: There are obvious cracks on the surface of the image display component, the power supply efficiency is reduced by 25%, and it is located on the critical power supply path;

[0134] Recommended action: Replace the faulty component immediately to ensure stable operation of the power plant;

[0135] As the drone flies toward the next faulty photovoltaic module, it transmits this sub-fault report to the ground control station in real time via wireless communication technology.

[0136] refer to Figure 6 In step S15, when the drone completes the fault inspection of the last faulty photovoltaic module, the drone remains in the residential area until the overall fault report is completed. The overall fault report is generated based on the synthesis of multiple sub-fault reports.

[0137] In the specific implementation of this invention, the specific steps are as follows:

[0138] S151: The drone flies along the fault inspection route and completes the fault inspection of each faulty photovoltaic module in turn until the fault inspection of the last faulty photovoltaic module is completed.

[0139] S152: Determine the final location of the drone based on the location of the last faulty photovoltaic module, the location of the surrounding drones, and the remaining power of the drone. The final location of the drone is within a residential area.

[0140] S153: The UAV synthesizes multiple sub-fault reports, generates an overall fault report based on the synthesis of multiple sub-fault reports, and determines the UAV's optimized action events based on the optimization events of the overall fault report, the UAV's functions, and the usage status of each faulty photovoltaic module.

[0141] In the embodiments of this application, the drone flies along the fault inspection route and sequentially completes the fault inspection of each faulty photovoltaic module until the fault inspection of the last faulty photovoltaic module is completed, thus ensuring the completeness of the fault inspection of each faulty photovoltaic module.

[0142] At this time, the drone flies along the preset fault inspection route and conducts detailed fault inspections on each faulty photovoltaic module in sequence until all fault points are inspected. Before executing the mission, one or more inspection routes are planned in advance based on the layout of the photovoltaic power station, the distribution of faulty photovoltaic modules, and the flight capabilities of the drone. The route should cover all fault points as much as possible, while taking into account the flight safety, efficiency, and obstacle avoidance requirements of the drone.

[0143] The drone takes off from the designated launch point and enters the inspection state according to the preset flight altitude and speed parameters; using GPS or other positioning technology, it ensures that the drone can fly accurately along the planned route; at the same time, the drone uses its onboard camera and infrared sensor equipment to identify and locate each faulty photovoltaic module; through image processing technology, the drone can distinguish between normal modules and faulty modules and record the location information of the faulty modules.

[0144] Upon reaching each fault point, the drone conducts a detailed inspection, including acquiring images, measuring power supply efficiency, and recording power supply path information. The collected data is analyzed to assess the severity and cause of the fault. The drone records the inspection results of each fault point in its internal memory or cloud storage system. Based on the inspection results, the drone updates the status information of the faulty photovoltaic module for subsequent analysis and processing.

[0145] After completing the inspection of one fault point, the drone locates and flies to the next fault point according to a preset route or a route adjusted in real time. During the flight, the drone needs to make minor adjustments to the route according to the actual situation (such as weather changes and flight obstacles). Furthermore, when the drone completes the inspection of the last faulty photovoltaic module, it marks the completion of the entire fault inspection mission. At this time, the drone needs to perform power checks and data uploads for subsequent operations to prepare for the next mission.

[0146] Specifically, suppose there are 10 faulty photovoltaic modules in a residential area that need to be inspected. A drone takes off according to a preset inspection route and flies to each fault point in turn. At the first fault point, the drone identifies obvious cracks on the surface of the module and collects images and power efficiency data. At the second fault point, the drone finds that the module is covered by an obstruction, resulting in reduced power efficiency. It records the location and size of the obstruction and takes clear images. And so on, the drone conducts detailed inspections and data collection for each fault point.

[0147] After completing the inspection of all 10 fault points, the drone returns to its takeoff point or designated landing area. During the return process, it uploads the inspection results to the cloud storage system or ground control station for subsequent analysis and processing by staff. This example demonstrates the efficiency and accuracy of drones in fault inspection tasks. It can complete the inspection of a large number of fault points in a short time and provide valuable data support for subsequent analysis and repair work.

[0148] Furthermore, based on the location of the last faulty photovoltaic module, the surrounding drone landing locations, and the drone's remaining battery power, the drone's final landing location was determined. This final landing location was within a residential area, ensuring the accuracy of the drone's final landing location.

[0149] At this point, after the drone has completed the inspection of all the faulty photovoltaic modules, it needs to determine a suitable final landing location based on the location of the last fault point, the available drone landing locations in the vicinity, and its own remaining power. This location needs to ensure the safe landing of the drone and subsequent maintenance, while also taking into account the potential impact on the surrounding environment, especially when the landing location is located in a residential area.

[0150] When the drone completes the inspection of the last faulty photovoltaic module, it accurately records the geographical location information of the fault point. This is achieved through GPS or other positioning technologies to ensure the accuracy of the location data. At the same time, the drone uses its built-in map data or real-time query service to search for available landing locations in the surrounding area. These locations include dedicated drone landing pads, open spaces, and building rooftops.

[0151] The drone monitors its battery level in real time and assesses the remaining power based on flight history data and current mission consumption. This includes considering the flight distance required to return to the landing location, backup power requirements, and safety margin. At the same time, the drone comprehensively evaluates the location of the last failure point, information on surrounding landing locations, and the remaining power. Using a usage method or decision tree, the drone selects the most suitable landing location. When the landing location is located in a residential area, the drone needs to give special consideration to noise, privacy, and the impact on residents.

[0152] The drone flies safely to its final selected stop location according to the planned route; during the flight, the drone continuously monitors its battery level and flight status to ensure a safe landing; furthermore, upon reaching the stop location, the drone executes the landing procedure to ensure a smooth landing; if necessary, the drone needs to perform subsequent operations such as data uploading, battery replacement, or charging.

[0153] Specifically, suppose that after the drone completes the inspection of the 10th faulty photovoltaic module, it is located in the northeast corner of the photovoltaic power station. At this time, the drone has 30% of its remaining power, which is enough to support its flight to a nearby landing location. The 10th fault point recorded by the drone is located in an area in the northeast corner of the photovoltaic power station. The drone finds two available landing locations nearby: one is a dedicated helipad inside the photovoltaic power station, and the other is an open space in a residential park about 500 meters away from the photovoltaic power station.

[0154] The drone assessed that its remaining battery power was sufficient to fly to an open space in a residential park, but considering the safety margin and backup power requirements, choosing a closer helipad seemed safer; however, since the helipad was already occupied by other drones, the drone decided to go to an open space in the residential park; after comprehensively considering the location's safety, accessibility, and impact on residents, the drone chose a relatively secluded open space in the residential park that was far from the residents' activity area as its final landing location.

[0155] The drone safely flew along its planned route to an open space in a residential park, continuously monitoring its battery level and flight status throughout the flight. Upon reaching its designated landing location, the drone landed smoothly and immediately began uploading inspection data to the cloud storage system. Simultaneously, the ground control station received the drone's landing notification and dispatched personnel to the location for subsequent battery replacement and maintenance. This example demonstrates that drones need to consider multiple factors when determining their final landing location, including location security, accessibility, remaining battery power, and impact on the surrounding environment. In practical applications, these factors vary depending on the scenario and task, but drones need to be able to adapt flexibly and make optimal decisions.

[0156] Therefore, the UAV interacts with multiple sub-fault reports, synthesizing them to generate an overall fault report. Based on the optimized events of the overall fault report, the UAV's functions, and the usage status of each faulty photovoltaic module, the UAV's optimized action events are determined, ensuring the accuracy of these optimized action events. Simultaneously, by fully utilizing the UAV's processing of sub-fault reports from the next faulty photovoltaic module during flight, unmanned flight and autonomous fault detection are performed concurrently, enabling online fault management of residential photovoltaic modules by the UAV.

[0157] At this point, after the drone completes the inspection of all faulty photovoltaic modules, it will generate multiple sub-fault reports, which contain detailed information about each fault point. Next, the drone needs to interact with these sub-fault reports multiple times to generate a comprehensive overall fault report. Based on this report, the drone will combine its functions, the usage status of the faulty components, and optimization events to determine subsequent optimization actions.

[0158] After completing its inspection mission, the drone automatically collects and organizes all generated sub-fault reports. These reports include the location of the fault point, the type of fault, the severity of the fault, and image data. At the same time, it performs multiple interactions and comprehensive analyses on the multiple sub-fault reports. This includes comparing information from different fault points, identifying the correlation between faults, and assessing the impact of the faults on the overall operation of the photovoltaic power station. Based on the results of the multiple interactions and comprehensive analyses, the drone generates a comprehensive overall fault report. This report includes a fault overview, a fault point distribution map, a fault severity assessment, a fault cause, and recommended optimization measures.

[0159] The drone identifies priority optimization events from the overall fault report. These optimization events include emergency repair of critical fault points, early warning of fault trends, and adjustment of power plant operation strategies. At the same time, the drone determines specific optimization action events based on its functions (such as monitoring, maintenance, and alarm) and the usage status of faulty components (such as whether they are online or have been repaired). These actions include sending alarm information to the ground control station, automatically dispatching maintenance personnel to the fault point, and adjusting the power plant operation strategy to mitigate the impact of the fault.

[0160] Specifically, suppose that after completing its inspection mission, the drone generates five sub-fault reports, each corresponding to a different faulty photovoltaic module. The drone collects these reports and notices that two of them indicate severe module damage, while the other three indicate a slight decrease in power supply efficiency. The drone conducts in-depth analysis of these reports and finds that the severely damaged modules are located on the critical power supply path of the power plant, while the modules with decreased power supply efficiency are scattered in different areas of the power plant. The drone also notes that these faults are related to recent severe weather conditions.

[0161] Based on these analyses, the drone generated an overall fault report, detailing information on all fault points and identifying component damage on the critical power supply path as the primary issue. The drone identified immediate optimization events requiring repair of the damaged components on the critical power supply path and monitoring the failure trends of other components. The drone decided to utilize its alarm function to immediately send an alarm message to the ground control station, indicating the component damage on the critical power supply path and recommending emergency repairs. Simultaneously, the drone planned to periodically monitor the status of other components and adjust the power station's operating strategy based on weather conditions. The drone executed the alarm action and received confirmation from the ground control station. Subsequently, the drone continued to monitor the power station's status and provide real-time feedback to the control station as needed.

[0162] In one embodiment of this application, the optimized event matching table is shown in Table 3:

[0163] Table 3 Optimization Event Matching Table

[0164]

[0165] The drone collected the following sub-fault reports: Report 1: Component A is damaged; Report 2: Component B's power supply efficiency has decreased; Report 3: Component C's connection cable is loose; Through multiple interactions and matching, the drone generated the following overall fault report: Overall Fault Report: Component A requires emergency repair (offline), Component B requires monitoring + early warning (online but inefficient), Component C requires dispatching maintenance personnel for inspection (offline / online but abnormal).

[0166] Based on the overall fault report, the UAV determined the following optimized action events: immediately alert the ground control station and request dispatch maintenance personnel to urgently repair component A; begin continuous monitoring of component B and set an early warning threshold for power efficiency degradation; dispatch maintenance personnel to the location of component C for inspection and repair.

[0167] Please see Figure 7 , Figure 7 This is a schematic diagram of the structural composition of a drone-based residential photovoltaic module fault inspection system according to an embodiment of the present invention; the drone-based residential photovoltaic module fault inspection system includes:

[0168] The data acquisition module 21 is used to collect the power supply efficiency record table of each household photovoltaic module during its use in the residential area.

[0169] The fault identification module 22 is used to identify multiple faulty photovoltaic modules based on the power supply efficiency record table, the distribution map of the household photovoltaic modules, and the identification of the model of the household photovoltaic modules.

[0170] The fault inspection route module 23 is used to determine the fault inspection route of the UAV for multiple faulty photovoltaic modules based on the current location of multiple faulty photovoltaic modules, the distribution map of the residential area and the flight environment of the UAV.

[0171] The sub-fault reporting module 24 is used for the UAV to fly along the fault inspection route and dynamically generate a sub-fault report for the next faulty photovoltaic module during the flight.

[0172] The overall fault report module 25 is used to keep the drone in the residential area until the overall fault report is completed when the drone completes the fault inspection of the last faulty photovoltaic module. The overall fault report is generated based on the synthesis of multiple sub-fault reports.

[0173] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

Claims

1. A method for fault inspection of residential photovoltaic modules based on unmanned aerial vehicles (UAVs), characterized in that, include: During the use of various household photovoltaic modules in residential areas, power supply efficiency records of the household photovoltaic modules are collected. Based on the power supply efficiency record sheet, the distribution map of the residential photovoltaic modules, and the identification of the model of the residential photovoltaic modules, multiple faulty photovoltaic modules were identified. The fault inspection route for the drone to inspect the multiple faulty photovoltaic modules is determined based on the current location of the multiple faulty photovoltaic modules, the distribution map of the residential area, and the drone's flight environment. The drone flies along a fault inspection route, dynamically generating sub-fault reports for faulty photovoltaic modules during its flight. This includes: determining multiple fault review nodes based on the fault inspection route and the current locations of multiple faulty photovoltaic modules; determining the dynamic review method of the drone at each fault review node based on the multiple fault review nodes, the fault type of the faulty photovoltaic module, the surrounding environment of the faulty photovoltaic module, and a preset review method matching table; at a fault review node, the drone reviews the faulty photovoltaic module based on the corresponding dynamic review method and collects multiple images of the faulty photovoltaic module; and determining the sub-fault report of the faulty photovoltaic module based on the multiple images of the faulty photovoltaic module, the power supply efficiency of the faulty photovoltaic module, and the power supply path of the faulty photovoltaic module. The drone completes the output of the sub-fault report of the faulty photovoltaic module as it flies to the next faulty photovoltaic module. While the drone completes its inspection of the last faulty photovoltaic module, it remains in the residential area until the overall fault report is completed. The overall fault report is generated by synthesizing multiple sub-fault reports.

2. The method for fault inspection of residential photovoltaic modules based on unmanned aerial vehicles (UAVs) according to claim 1, characterized in that, The process of using various household photovoltaic modules in residential areas includes collecting power supply efficiency records for these modules, including: Collect the location of the residential area, and determine the individual household photovoltaic modules in the residential area based on the location of the residential area, the distribution map of the residential area, and the photovoltaic power supply path; The system monitors the usage of each household photovoltaic module in real time and collects the power supply efficiency of each household photovoltaic module in the residential area. Based on the power supply efficiency of each household photovoltaic module in the residential area, the corresponding residential building, and the real-time power consumption of the residential building, a power supply efficiency record table for the household photovoltaic modules is determined.

3. The method for fault inspection of residential photovoltaic modules based on unmanned aerial vehicles (UAVs) according to claim 1, characterized in that, The method of identifying multiple faulty photovoltaic modules based on the power supply efficiency record table, the distribution map of residential photovoltaic modules, and the model identification of residential photovoltaic modules includes: In this power supply efficiency record table, multiple abnormal parameters are determined based on the autonomous detection of the power supply efficiency record table, and various abnormal parameter combinations are constructed on a residential basis. The corresponding abnormal level is determined based on the identification of the abnormal parameter combinations of the residential. The first fault parameter is determined based on the anomaly level and the corresponding photovoltaic module model. The second fault parameter is determined based on the anomaly level and the distribution map of the residential photovoltaic modules. Multiple faulty photovoltaic modules are identified according to the first fault parameter, the second fault parameter, and a preset fault parameter matching table.

4. The method for fault inspection of residential photovoltaic modules based on unmanned aerial vehicles (UAVs) according to claim 1, characterized in that, The process of determining the drone's fault inspection route for multiple faulty photovoltaic modules based on their current locations, the distribution map of the residential area, and the drone's flight environment includes: The corresponding fault types are collected by traversing multiple faulty photovoltaic modules, and multiple fault distribution areas are determined based on the current location of multiple faulty photovoltaic modules, the floor of the residence, and the corresponding fault type. These fault distribution areas are multi-dimensional spaces. The flight environment of the UAV is determined based on the initial flight of the UAV to multiple fault distribution areas and the aerial images detected by the UAV.

5. The method for fault inspection of residential photovoltaic modules based on unmanned aerial vehicles (UAVs) according to claim 4, characterized in that, The method of determining the drone's fault inspection route for multiple faulty photovoltaic modules based on the current location of the multiple faulty photovoltaic modules, the distribution map of the residential area, and the drone's flight environment also includes: Based on multiple fault distribution areas and the flight environment of the UAV, a first inspection route is determined for each fault distribution area. Based on the distribution map of multiple fault distribution areas and residential areas, a second inspection route is determined for each fault distribution area. The fault inspection route of the UAV for multiple faulty photovoltaic modules is determined by combining the first inspection route and the second inspection route.

6. The method for fault inspection of residential photovoltaic modules based on unmanned aerial vehicles (UAVs) according to claim 1, characterized in that, When the drone completes the fault inspection of the last faulty photovoltaic module, it remains in the residential area until the overall fault report is completed. The overall fault report is generated based on the synthesis of multiple sub-fault reports, including: The drone flies along the fault inspection route and completes the fault inspection of each faulty photovoltaic module in turn until the fault inspection of the last faulty photovoltaic module is completed. Based on the location of the last faulty photovoltaic module, the landing locations of nearby drones, and the drone's remaining battery power, the drone's final landing location was determined to be within a residential area.

7. The method for fault inspection of residential photovoltaic modules based on unmanned aerial vehicles (UAVs) according to claim 6, characterized in that, When the drone completes the fault inspection of the last faulty photovoltaic module, it remains in the residential area until the overall fault report is completed. The overall fault report is generated based on the synthesis of multiple sub-fault reports and also includes: The drone interacts with multiple sub-fault reports, generates an overall fault report based on the synthesis of multiple sub-fault reports, and determines the drone's optimized action events based on the optimization events of the overall fault report, the drone's functions, and the usage status of each faulty photovoltaic module.

8. A drone-based fault inspection system for residential photovoltaic modules, characterized in that, The drone-based residential photovoltaic module fault inspection system is applied to the drone-based residential photovoltaic module fault inspection method as described in any one of claims 1-7, wherein the drone-based residential photovoltaic module fault inspection system comprises: The data acquisition module is used to collect power supply efficiency records of household photovoltaic modules during their use in residential areas. The fault identification module is used to identify multiple faulty photovoltaic modules based on the power supply efficiency record table, the distribution map of the residential photovoltaic modules, and the model identification of the residential photovoltaic modules. The fault inspection route module is used to determine the fault inspection route of the UAV for multiple faulty photovoltaic modules based on the current location of multiple faulty photovoltaic modules, the distribution map of the residential area, and the flight environment of the UAV. The sub-fault reporting module is used for the UAV to fly along the fault inspection route and dynamically generate sub-fault reports of faulty photovoltaic modules during the flight. The overall fault report module is used so that when the drone completes the fault inspection of the last faulty photovoltaic module, the drone remains in the residential area until the overall fault report is completed. The overall fault report is generated based on the synthesis of multiple sub-fault reports.

Images