An unmanned aerial vehicle and Beidou technology-based power grid facility intelligent inspection system and method

Abstract

The application discloses a kind of power grid facilities intelligent inspection system and method based on unmanned aerial vehicle and big dipper technology, it is related to unmanned aerial vehicle inspection management field, including map signal sensing module, route planning module, signal enhancement module and signal broken chain module, the present application is combined with accurate signal coverage model, intelligent path planning, attitude fine adjustment and fault detection module, realizes efficient, accurate power grid inspection. Positioning and environmental data are obtained by big dipper technology, signal initial coverage estimation map is generated, and signal strength is corrected in real time by Gaussian process, according to image quality constraint parameters and signal coverage data, the inspection route is optimized, and the attitude fine adjustment strategy ensures that the unmanned aerial vehicle is adaptively adjusted in the weak signal area, to maximize the inspection efficiency. The automation and intelligent level of power grid inspection is improved, and the inspection quality and task execution efficiency are improved.

Description

Technical Field

[0001] This invention relates to the field of unmanned aerial vehicle (UAV) inspection and management, specifically to an intelligent inspection system and method for power grid facilities based on UAVs and BeiDou technology. Background Technology

[0002] Traditional power grid inspections rely on manual labor or fixed monitoring equipment, resulting in low efficiency, high safety risks, and limited data. Unmanned aerial vehicle (UAV) inspection technology has been widely applied in various fields, particularly in the power, telecommunications, and transportation industries. In power grid inspection, existing technologies utilize GPS positioning and automatic flight control systems to assist UAVs in performing tasks and inspecting power grid facilities. UAVs can automatically cruise along preset routes, collecting data such as images, videos, and temperature.

[0003] The BeiDou Navigation Satellite System boasts advantages such as global coverage and real-time high-precision positioning. By utilizing the real-time positioning and map information provided by the BeiDou system, unmanned aerial vehicles (UAVs) can navigate accurately in complex terrain, reducing drift errors and improving the accuracy of inspection tasks. BeiDou technology enables real-time data exchange between the UAV and the ground control system, ensuring real-time updates of positioning information during inspections and increasing the reliability of these inspections.

[0004] Existing technology, such as the invention patent with announcement number CN113554775B, is a drone power line inspection system. The system includes a cloud server, a communication module, and a drone. The cloud server receives inspection messages sent by user terminals, obtains the corresponding inspection task data to be executed and the geometric information of preset inspection points based on the task identifier carried in the inspection message, plans the target inspection task data according to the task priority of the inspection task data to be executed, and plans the drone flight path based on the geometric information of the preset inspection points. When there are multiple drone flight paths, the multiple drone flight paths are merged, and the merged drone flight path and the target inspection task data are sent to the drone through the communication module. The drone is used to perform the task of power line inspection according to the drone flight path and the target inspection task data, and sends the task execution result data to the cloud server through the communication module.

[0005] Existing technology, such as the invention patent with publication number CN113177861B, describes a method for inspecting power lines using drones. The drone includes an image acquisition module and a recognition module, which are electrically connected. The image acquisition module acquires image information of the power lines during the drone's inspection. The recognition module identifies whether a temperature warning device in the power line has activated its warning action based on the image information acquired by the image acquisition module. The temperature warning device is installed on a preset component in the power line and activates a warning action when the temperature of the preset component exceeds a temperature threshold. The drone then issues an alarm based on the activation status of the temperature warning device.

[0006] As can be seen from the above solutions, in the field of UAV power grid inspection, existing technologies typically focus on inspection image quality or flight path optimization. Flight path planning is inefficient, relying solely on task priority and ignoring the impact of environmental factors on the flight path and inspection tasks. Furthermore, they lack effective communication management, and image recognition accuracy is significantly limited by the environment. In practical applications, existing UAV inspection technologies rely on 4G / 5G or Wi-Fi communication methods, and communication quality may degrade in areas with weak signals or severe obstruction. Moreover, some communication blind spots are dynamic (e.g., temporary interference, weather factors), making it impossible to completely resolve communication quality issues through pre-planning. Summary of the Invention

[0007] To address the shortcomings of existing technologies, this invention provides an intelligent inspection system and method for power grid facilities based on unmanned aerial vehicles (UAVs) and BeiDou technology. To achieve the above objectives, this invention utilizes the following technical solution: An intelligent inspection system for power grid facilities based on unmanned aerial vehicles (UAVs) and BeiDou technology, comprising:

[0008] The map signal sensing module is used to acquire BeiDou data of the power grid facility area through BeiDou technology and input it into the signal coverage generation model. At the same time, it superimposes the additional attenuation of signal loss due to multi-source environmental data and outputs an initial signal coverage estimation map. During the UAV inspection, the initial signal coverage estimation map is corrected online through a Gaussian process to obtain a corrected signal coverage estimation map.

[0009] The route planning module is used to output several candidate routes, obtain the image quality constraint parameters of each candidate route at each power grid facility monitoring point, and retrieve the signal correction coverage estimation map. The best route is then selected from the comprehensive results to obtain the UAV inspection route.

[0010] The signal enhancement module is used to collect real-time UAV attitude data when the UAV enters a weak signal area marked in the signal correction coverage estimation map, and to determine the UAV attitude fine-tuning strategy based on the real-time signal strength fed back by the UAV radio frequency module.

[0011] The signal disconnection module is used to cache inspection data using a delay-tolerant network when the signal is interrupted, and resend it after reconnection. At the same time, it reports UAV information based on Beidou short messages at fixed intervals.

[0012] A method for intelligent inspection of power grid facilities based on unmanned aerial vehicles (UAVs) and BeiDou technology includes:

[0013] The BeiDou data of the power grid facility area is obtained by using BeiDou technology and input into the signal coverage generation model. At the same time, the additional attenuation of signal loss due to multi-source environmental data is superimposed to output the initial signal coverage estimation map. During the UAV inspection, the initial signal coverage estimation map is corrected online by a Gaussian process to obtain the corrected signal coverage estimation map.

[0014] Output several candidate routes, obtain the image quality constraint parameters of each candidate route at each power grid facility monitoring point, and simultaneously retrieve the signal correction coverage estimation map. Select the best route from all options to obtain the UAV inspection route.

[0015] When the UAV enters the weak signal area marked in the signal correction coverage estimation map, real-time UAV attitude data is collected, and the UAV attitude fine-tuning strategy is determined based on the real-time signal strength fed back by the UAV radio frequency module.

[0016] When the signal is interrupted, the inspection data is cached using a delay-tolerant network and resent after reconnection. At the same time, UAV information is reported based on BeiDou short messages at fixed intervals.

[0017] Compared with the prior art, the embodiments of the present invention have at least the following beneficial effects:

[0018] (1) This invention provides an intelligent inspection system for power grid facilities based on UAVs and BeiDou technology. It acquires location, communication network, and environmental data of the power grid facility area using BeiDou technology, and generates an accurate signal coverage estimation map by combining this data with a signal propagation attenuation model. This initial signal coverage estimation map reflects the signal strength distribution within the power grid area, ensuring that the UAV can understand the current signal coverage during inspection. During the inspection process, the system continuously corrects the initial signal coverage estimation map using a Gaussian process model, updating the predicted signal strength in real time. This highly dynamic signal correction function ensures that the system can continuously optimize under complex environments and different weather conditions, improving the inspection efficiency of UAVs in different areas.

[0019] (2) This invention generates several candidate routes through a path planning algorithm, and combines the image quality constraint parameters of the power grid facility monitoring points with the data in the signal correction coverage estimation map to comprehensively analyze and select the most suitable inspection route. By retrieving and comparing the optimal image quality parameters in the database, and combining the correlation between multi-source environmental data and signal strength, the system can efficiently select the optimal route, ensuring that the UAV obtains clear image quality during the inspection process, while avoiding areas with weak signals, thus optimizing the execution efficiency and quality of the inspection task.

[0020] (3) This invention, through the dynamic adjustment of the signal enhancement module and the weak signal region, can collect UAV attitude data in real time and determine the attitude fine-tuning strategy based on the signal strength fed back by the UAV's radio frequency module. This process includes adjustments to multiple aspects such as flight angle, flight speed, and flight altitude. When the UAV enters the weak signal region, the system will dynamically select an appropriate flight altitude enhancement or comprehensive attitude fine-tuning strategy based on the difference between the current signal strength and the preset threshold, combined with the flight altitude. By using a multi-objective optimization model to weightedly couple different fine-tuning strategies, the system can accurately control the flight angle, speed, and altitude to complete the inspection task with the best attitude, ensuring both inspection quality and efficiency.

[0021] (4) This invention employs a strategy of first fine-tuning the altitude and then considering a comprehensive strategy. The flight altitude has a direct and significant impact on signal strength, and adjusting the altitude can quickly optimize signal quality. Therefore, adjusting the flight altitude first can solve the problem of weak signal in the shortest time and improve the system's response efficiency. Directly applying a comprehensive fine-tuning strategy may require adjusting multiple parameters, such as flight angle and speed, which increases computational complexity and adjustment time. However, fine-tuning the altitude first can effectively improve the signal, avoid unnecessary adjustments, and simplify the computation and operation process. Adjusting the flight altitude usually requires a small adjustment range, thus saving energy and reducing the consumption of UAV battery and computing resources. By adjusting the flight altitude first, the UAV will be in a position with a strong signal, which provides a more stable signal foundation for subsequent fine-tuning of angle, speed, etc., thereby improving the overall system stability and accuracy.

[0022] Of course, any product implementing this invention does not necessarily need to achieve all of the above advantages at the same time. Attached Figure Description

[0023] Figure 1 This is a schematic diagram of the system modules of the present invention.

[0024] Figure 2 This is a schematic diagram of the method flow of the present invention.

[0025] Figure 3 This is a schematic diagram of the logic flow of the present invention.

[0026] Figure 4 This is a diagram of the real-time signal monitoring interface in the unmanned aerial vehicle management system involved in this embodiment of the invention.

[0027] Figure 5 This is a diagram of the inspection path planning interface in the unmanned aerial vehicle management system involved in this embodiment of the invention.

[0028] Figure 6 This is a diagram of the drone parameter adjustment interface in the drone management system involved in this embodiment of the invention. Detailed Implementation

[0029] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0030] In the description of this invention, it should be understood that the terms "opening", "upper", "lower", "thickness", "top", "middle", "length", "inner", "around", etc., which indicate orientation or positional relationship, are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the components or elements referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as limiting this invention.

[0031] Please see Figure 1 As shown, this embodiment of the invention provides an intelligent inspection system for power grid facilities based on unmanned aerial vehicles (UAVs) and BeiDou technology, comprising:

[0032] like Figure 3The diagram shown illustrates the logic flow involved in this embodiment of the invention, demonstrating how a UAV adjusts its flight path and attitude based on real-time signal strength and environmental factors during inspection to ensure the smooth execution of the inspection mission. The system generates an initial signal coverage estimation map by acquiring BeiDou positioning data and environmental data. This map reflects the signal strength distribution in the UAV's flight area. The system monitors signal strength in real time to determine if the current signal is strong enough. If the signal strength is below a set threshold, the system enters a weak signal area and collects the UAV's attitude data. Based on this data, the system calculates a fine-tuning strategy and executes corresponding attitude adjustments. After adjustment, the system reports the signal strength and determines if the threshold has been reached. If the signal strength returns to normal, the attitude fine-tuning is completed, and the inspection mission continues. If the signal strength remains low, the system determines whether fine-tuning needs to be repeated or if the maximum number of attempts has been reached. If the maximum number of attempts has been reached, the system determines that the UAV hardware may be malfunctioning and commands it to return for repair. If the signal strength is normal, the UAV enters the path planning stage, generates candidate routes, and determines the optimal inspection route through a comprehensive selection process. Finally, the UAV executes the inspection mission according to the optimized route until the inspection is completed.

[0033] The map signal sensing module is used to acquire BeiDou data of the power grid facility area through BeiDou technology and input it into the signal coverage generation model. At the same time, it superimposes the additional attenuation of signal loss due to multi-source environmental data and outputs an initial signal coverage estimation map. During the UAV inspection, the initial signal coverage estimation map is corrected online through a Gaussian process to obtain a corrected signal coverage estimation map.

[0034] The BeiDou data of the power grid facility area is obtained through BeiDou technology, including positioning data, communication network data, DEM terrain data and power grid facility location data. This data is then input into the signal coverage generation model, which is a model based on the signal propagation attenuation theory. This model is used to generate a visual signal coverage image of the signal strength in a specified area. The original signal coverage estimation map of the signal strength is then mapped into the signal coverage generation model.

[0035] It's important to note that the BeiDou Navigation Satellite System's positioning function is used to obtain the precise location of the area where power grid facilities are located. Positioning data typically includes latitude and longitude information to determine the geographical location of the power grid facilities and surrounding areas. Communication network data includes the network type for signal transmission and the distribution of base stations. A Digital Elevation Model (DEM) describes ground elevation information. It provides topographic features such as geographical relief, mountains, and buildings, helping to account for the impact of terrain during signal propagation, as mountains or buildings can shield or attenuate signal transmission. Power grid facility location data describes the specific geographical location of power grid equipment (such as substations and transmission lines), which is crucial for calculating signal coverage.

[0036] Signal propagation attenuation theory states that in the real world, signals are attenuated by various factors when propagating through the air, such as terrain obstruction, airborne substances (e.g., water vapor, rain, snow), building obstruction, and distance. These factors cause signal strength to gradually weaken as the propagation distance increases. The signal coverage generation model, based on this principle, simulates the attenuation process of a signal propagating from its source to its target area. The input data for the signal coverage generation model includes the aforementioned location information, communication network data, and terrain data. This data is used to simulate the signal strength changes as it travels from the power grid's transmission source, through various terrain features, buildings, and climatic conditions, and finally reaches a specific area. The initial coverage estimation map represents the signal strength values ​​at different locations, but it is only a preliminary result generated based on the existing model and input data, and does not consider actual flight data, environmental changes, or other dynamic factors.

[0037] It should be noted that the specific process of generating a visual image from signal strength includes:

[0038] Signal strength at different locations is acquired in real time using sensors. A signal coverage generation model suitable for the target area environment is selected. These models are based on the physics theory of propagation attenuation, and common ones include: free space propagation model, secondary path loss model, multipath attenuation model, and environmental attenuation model. Based on the selected signal propagation model, the signal attenuation at each location is calculated. By inputting known signal propagation parameters (such as transmit power, frequency, path loss index, etc.), the model can predict the signal strength at a specified location. To generate a signal strength image in areas without direct measurement points, spatial interpolation of the signal strength between known data points is usually required. Common interpolation methods include Kriging interpolation and inverse distance weighting. In other embodiments of the invention, the appropriate signal coverage generation model and interpolation method can be selected according to the actual situation. In this embodiment of the invention, the multipath attenuation model and Kriging interpolation method are selected. The interpolated data is used to plot the signal strength data into an image. Common visualization methods include heatmaps and contour lines.

[0039] Multi-source environmental data, including historical meteorological data, electromagnetic interference data, and basic environmental data, is acquired through external sensing devices and input into the signal coverage generation model. The additional attenuation of the signal strength corresponding to the multi-source environmental data is mapped into the signal coverage generation model and superimposed on the original coverage estimation map to output the initial signal coverage estimation map.

[0040] It should be noted that historical meteorological data refers to meteorological information over a past period, and meteorological conditions have a significant impact on signal propagation. For example, signals are more attenuated during rainy or snowy weather, or when humidity is high. Therefore, meteorological data can help assess signal strength under different meteorological conditions.

[0041] Electromagnetic interference (EMI) data refers to electromagnetic waves generated by various devices (such as power equipment, communication base stations, and industrial equipment) that can interfere with the propagation of wireless signals. EMI degrades signal quality, therefore it is necessary to acquire EMI data and consider its impact on signal propagation.

[0042] Basic environmental data includes other features of the ground and environment, such as building density, roads, and vegetation cover. These factors also affect the propagation of wireless signals because they can cause effects such as signal obstruction, reflection, and scattering.

[0043] The acquired multi-source environmental data (meteorological, electromagnetic interference, and basic environmental data) are input into the signal coverage generation model. This model is trained based on signal propagation attenuation theory, therefore it not only considers the physical laws of signal propagation but also needs to integrate external environmental factors to correct for signal strength. During signal propagation, environmental factors additionally affect signal strength. The signal coverage generation model calculates the additional attenuation of signal strength caused by these factors based on the input environmental data.

[0044] Based on the relationship between environmental data and signal strength, the signal coverage generation model maps the impact of these environmental factors onto the original signal coverage estimation map. For example, if an area has high electromagnetic interference, the signal strength will be additionally attenuated; if the area has high humidity, the signal will also be affected accordingly. The original coverage estimation map is a preliminary signal strength map generated based on basic signal propagation attenuation theory and regional topography. Building upon this, after incorporating multi-source environmental data, the signal coverage generation model adds the additional attenuation from environmental factors to the signal strength in the original estimation map. After environmental data correction, the resulting initial signal coverage estimation map, which comprehensively considers factors such as topography, weather, and electromagnetic interference, provides a more accurate reflection of the actual signal coverage situation.

[0045] like Figure 4 The diagram shows the real-time signal monitoring interface of the UAV management system involved in this embodiment of the invention. It includes a signal sensing module and a signal enhancement module, marking the real-time signal data of each area, and listing areas with weak signals and UAV attitudes. This helps the management system monitor the signal status of the area where the UAV is located in real time, ensuring that effective adjustment measures are taken promptly in environments with poor signal, thus guaranteeing the flight safety of the UAV and the smooth progress of inspection tasks.

[0046] During the UAV inspection process, real-time location signal data is collected, including signal strength, real-time location, and flight status. Based on the initial signal coverage estimation map, a Gaussian process model is initialized, and the covariance function of the Gaussian process is set. The covariance function is used to define the correlation of signal strength at different locations.

[0047] It should be noted that the initial signal coverage estimation map is a preliminary signal strength map calculated by the signal coverage generation model, representing the signal strength distribution in a certain area. This map is only a theoretically calculated coverage map and does not consider dynamic changes in the actual environment, therefore further correction is needed. The Gaussian process is a powerful machine learning method used to handle uncertainties and dynamic changes. It models signal strength using training data and provides predictions of signal strength in unobserved areas. The initial coverage estimation map can be used as a preliminary model of the Gaussian process. Specifically, in the Gaussian process model, the signal strength value at each location is correlated with the signal values ​​at other locations, forming a continuous signal strength prediction process. The core of the Gaussian process is the covariance function, often called the kernel function. It is used to define the correlation of signal strength between different locations. The covariance function describes the relationship between the signal strength at one location and the signal strength at other locations. The covariance function uses a mathematical formula to describe the degree of correlation between two points (locations) in space. For example, during electromagnetic wave propagation, signal strength usually decreases with increasing distance, so the covariance function reflects that the signal correlation between two distant locations is low, and the signal correlation between two nearby locations is high. Signal propagation in space typically exhibits spatial correlation. The covariance function quantifies this spatial correlation, enabling signal strength prediction in areas where it hasn't been measured. For example, if a drone receives a signal strength at its current location, the covariance function predicts the signal strength in areas closer to that location. This allows for prediction using a Gaussian process model even in areas without direct signal measurements. As the drone continues to fly and collect real-time data, the signal coverage generation model is continuously updated. Each time new signal strength and location data is collected, the Gaussian process updates its signal strength prediction based on this data, further refining the initial estimate. Flight state typically refers to the drone's current operational state during flight, including cruise, hovering, descent, and acceleration.

[0048] Real-time location signal data is input into a Gaussian process model, which is a machine learning method based on Bayesian theory. The Gaussian process model outputs the signal prediction update value and variance of the initial signal coverage estimation map. The initial signal coverage estimation map is then corrected to obtain the corrected signal coverage estimation map.

[0049] It's important to note that variance represents the uncertainty of the prediction or the reliability of the signal strength prediction. The Gaussian process not only provides the predicted signal value but also the prediction variance at each location. A large variance at a given location indicates that the Gaussian process model's prediction of the signal strength at that location is less reliable; conversely, a smaller variance indicates that the Gaussian process model's prediction of the signal strength is more accurate.

[0050] Extract the signal prediction update value at the location where the prediction variance exceeds the preset variance threshold, record the location as the pre-update location, and replace the initial signal strength value with the signal prediction update value at the pre-update location.

[0051] like Figure 5 The diagram shows the inspection path planning interface in the UAV management system according to an embodiment of the present invention. It includes modules for automatic inspection path planning and manual route planning. The UAV management system provides a flexible path planning and control mechanism for inspection tasks, offering two path planning modes—automatic planning and manual setting—which can optimize the path based on real-time conditions to ensure efficient and safe task execution.

[0052] The route planning module is used to output several candidate routes, obtain the image quality constraint parameters of each candidate route at each power grid facility monitoring point, and retrieve the signal correction coverage estimation map. The best route is then selected from the comprehensive results to obtain the UAV inspection route.

[0053] The path planning algorithm is used to output several candidate routes, and the image quality constraint parameters of each candidate route at each power grid facility monitoring point are obtained, including the drone shooting angle, drone shooting distance, drone flight altitude and drone flight speed.

[0054] The optimal image quality parameters for each power grid facility monitoring point are obtained from the database, including the optimal drone shooting angle, optimal drone shooting distance, optimal drone flight altitude, and optimal drone flight speed.

[0055] The image quality constraint parameters of each candidate route at each power grid facility monitoring point are compared one-to-one with the corresponding optimal image quality parameters, and then weighted and coupled to obtain the image quality constraint values ​​of each candidate route at each power grid facility monitoring point. Specifically, these include:

[0056] ;

[0057] Among them, Pic i,p For the image quality constraint value of the i-th candidate route at the p-th power grid facility monitoring point, view i,p Let l be the UAV shooting perspective of the i-th candidate route at the p-th power grid facility monitoring point. i,p h represents the distance h between the drone and the monitoring point of the power grid facility at the i-th candidate flight path. i,p Let v be the flight altitude of the UAV at the p-th power grid facility monitoring point for the i-th candidate flight path. i,p Let the view be the flight speed of the UAV at the p-th power grid facility monitoring point for the i-th candidate route. p,0 For the optimal UAV shooting perspective of the p-th power grid facility monitoring point, l p,0 h is the optimal drone shooting distance for the p-th power grid facility monitoring point.p,0 Let v be the optimal UAV flight altitude for the p-th power grid facility monitoring point. p,0 Let α1 be the optimal UAV flight speed for the p-th power grid facility monitoring point, α2 be the weighting factor for the UAV's shooting angle, α3 be the weighting factor for the UAV's shooting distance, α4 be the weighting factor for the UAV's flight altitude, i be the candidate route number (i=1,2,3,...,n), n be the total number of candidate routes, and p be the power grid facility monitoring point number (p=1,2,3,...,m), m be the total number of power grid facility monitoring points.

[0058] It should be noted that the drone shooting perspective weighting factor, drone shooting distance weighting factor, drone flight altitude weighting factor, and drone flight speed weighting factor are used to adjust the weight of each parameter in the image quality constraint value based on the actual situation during the calculation. They can be obtained through comprehensive analysis of historical task data.

[0059] It's also important to note that the shooting angle refers to the angular relationship between the drone's camera and the power grid infrastructure target, typically involving yaw angle (horizontal angle) and pitch angle (vertical angle). It determines image quality and the range of information acquired. The shooting angle is closely related to flight altitude, shooting distance, and flight speed. At higher flight altitudes, the shooting angle needs to be adjusted to ensure the equipment is clearly visible, avoiding image distortion or loss of detail. The shooting distance refers to the spatial distance between the drone and the monitored target, affecting image clarity and resolution. Too close a distance may limit the shooting area, while too far a distance will result in a lack of image detail. The shooting distance is directly related to flight altitude, flight speed, and shooting angle, and usually needs to be adjusted according to the flight altitude and inspection task requirements to ensure image quality. Flight altitude directly affects the drone's shooting range and image quality. Lower flight altitudes help improve image resolution but may limit the field of view; higher flight altitudes can expand the field of view but may sacrifice image detail. Adjusting the flight altitude is closely related to shooting distance and shooting angle. Higher flight altitudes may require an appropriate increase in shooting distance, while simultaneously adjusting the shooting angle to ensure full equipment coverage. Flight speed is another key factor affecting image stability and inspection efficiency. Excessive flight speed can lead to blurry images, while excessively slow speeds will prolong mission execution time and affect inspection efficiency. Flight speed is interrelated with flight altitude, shooting distance, and viewing angle; therefore, it is generally necessary to optimize the execution speed of inspection missions while ensuring image stability. These four parameters are closely linked and collectively affect the inspection effectiveness of the drone.

[0060] Simultaneously, the signal prediction update values ​​of each power grid facility monitoring point in the signal correction coverage estimation map are retrieved, and the image quality constraint values ​​of each candidate route at each power grid facility monitoring point are correlated with the signal strength. Then, the image quality constraint values ​​of each candidate route at each power grid facility monitoring point are statistically processed to obtain the comprehensive image quality constraint value of each candidate route.

[0061] It should be noted that when evaluating candidate flight routes, the system needs to consider the signal quality of each route as it passes through each power grid facility monitoring point. Specifically, this step combines the updated signal prediction values ​​(i.e., the latest signal strength data for each monitoring point) in the signal correction coverage estimation map with the image quality constraint values ​​of the flight route. The image quality constraint values ​​consider the impact of factors such as shooting angle, shooting distance, and flight altitude on image quality. Stronger signal strength may result in better image quality; conversely, weaker signal strength may compromise image quality. Therefore, the correlation between signal strength and image quality can help evaluate the performance of each flight route during actual inspections. After correlating the signal strength and image quality constraint values, the system performs a comprehensive analysis of the image quality of each candidate flight route at all monitoring points. Through statistical processing, in this embodiment of the invention, a weighted average method is used to integrate the image quality constraint values ​​of each candidate flight route into a comprehensive image quality constraint value. The comprehensive image quality constraint value reflects the overall performance of the candidate flight route throughout the entire inspection process, including a comprehensive evaluation of multiple factors such as signal strength and image quality.

[0062] The image quality constraint comprehensive values ​​of each candidate route are sorted from high to low, and the candidate route with the highest image quality constraint comprehensive value is selected as the UAV inspection route.

[0063] The signal enhancement module is used to collect real-time UAV attitude data when the UAV enters a weak signal area marked in the signal correction coverage estimation map, and to determine the UAV attitude fine-tuning strategy based on the real-time signal strength fed back by the UAV radio frequency module.

[0064] like Figure 6 The diagram shows the drone parameter adjustment interface in the drone management system according to an embodiment of the present invention, including flight parameter adjustment and parameter configuration management. An integrated operating platform is provided to help operators adjust the drone's flight parameters and system configuration in real time, ensuring that the drone can complete tasks stably, safely, and efficiently in complex environments. By adjusting parameters such as flight altitude, speed, and attitude, operators can optimize flight performance; through parameter configuration management, the system's hardware and software can operate in optimal condition, thereby improving task execution and drone reliability.

[0065] The system monitors the signal strength of drones in real time. When the signal strength of a drone is lower than the preset signal strength threshold, the location of the drone is marked as a weak signal area, and the signal coverage generation model is updated accordingly.

[0066] During the collection of real-time UAV attitude data, the influence of electromagnetic interference on the real-time UAV attitude data is corrected by combining the signal correction coverage estimation map. The corrected real-time UAV attitude data is then output, which includes the UAV attitude angle, flight altitude, flight speed, and flight acceleration.

[0067] By establishing an electromagnetic interference model, interference patterns in sensor data can be identified, such as periodic fluctuations or sudden changes. Common interference identification methods include sensor calibration and dynamic filtering techniques (such as Kalman filtering or extended Kalman filtering). During the correction process, the sensor data is corrected by adjusting the signal strength data in the signal correction overlay estimation map. Signal strength serves as a reference for the external environment and helps assess the reliability of sensor data. For example, in areas with weak signal strength (i.e., areas with strong electromagnetic interference), the reliability of sensor data is lower. In such cases, the weight of sensor data should be reduced, and the correction strength of the external signal correction map should be increased. Conversely, in areas with strong signal strength, there is less signal interference, and the sensor data is more accurate, so its weight can be increased.

[0068] Specifically, electromagnetic interference (EMI) correction involves four main attitude parameters of the UAV: ​​attitude angle, flight altitude, flight speed, and flight acceleration. In attitude angle correction, Kalman filtering is used to correct gyroscope and magnetometer data, incorporating signal strength information to reduce deviations caused by EMI. Flight altitude correction is achieved by comparing data with GPS or altimeter data, utilizing signal strength for correction, especially in areas with weak signals, increasing reliance on external data. Flight speed and acceleration data correction is achieved by weighting data from multiple sensors (such as accelerometers, gyroscopes, and external signals) to ensure accuracy. These methods effectively eliminate the impact of EMI on UAV attitude data, outputting more accurate real-time attitude data.

[0069] The signal strength difference of the drone is obtained by subtracting the signal strength from the signal strength threshold. At the same time, the drone's flight altitude is extracted and compared with the preset signal influence altitude threshold. If the drone's flight altitude is lower than the signal influence altitude threshold, the signal strength difference is entered into the preset mapping set of signal strength difference - flight altitude increase value in the database for mapping and matching to obtain the drone's flight altitude increase value. At the same time, the flight altitude increase threshold is extracted from the database and the flight altitude increase value is compared with the flight altitude increase threshold. If the flight altitude increase value is less than or equal to the flight altitude increase threshold, the flight altitude fine-tuning sub-strategy is selected. The flight altitude fine-tuning sub-strategy is the flight altitude increase value, and the drone is controlled to execute the flight altitude increase.

[0070] If the drone's flight altitude is greater than or equal to the signal influence altitude threshold, or the flight altitude increase value is greater than the flight altitude increase threshold, then the drone's comprehensive attitude fine-tuning strategy is selected to adjust the drone.

[0071] Controlling the drone to perform altitude increases also includes post-altitude feedback, specifically including:

[0072] After controlling the drone to increase its flight altitude, the signal strength of the drone is monitored again. If the signal strength of the drone is greater than or equal to the signal strength threshold, it is determined that the attitude fine-tuning is completed and the power grid facility inspection task continues.

[0073] If the signal strength of the drone is still less than the signal strength threshold, it is determined that the attitude needs to be readjusted. After updating the drone attitude data, the drone's comprehensive attitude fine-tuning strategy is initiated.

[0074] A comprehensive attitude fine-tuning strategy for the UAV was selected for adjustment, with the specific processing conditions as follows:

[0075] The comprehensive attitude fine-tuning strategy for UAVs includes flight angle fine-tuning sub-strategies, flight speed fine-tuning strategies, and flight altitude fine-tuning strategies.

[0076] The flight angle fine-tuning sub-strategy includes yaw angle fine-tuning and pitch angle fine-tuning. It retrieves the location of the signal base station, uses the current position of the UAV and the known position of the base station, and obtains the relative angle between the UAV's radio frequency module and the signal base station's transmitting module through three-dimensional geometric calculation. At the same time, it obtains the preset optimal signal angle, and the target adjustment angle is obtained by subtracting the relative angle from the optimal signal angle.

[0077] The flight speed fine-tuning sub-strategy involves inputting the signal strength difference into a pre-stored mapping set of signal strength difference-flight speed reduction values ​​in the database and performing mapping matching to obtain the flight speed reduction value of the UAV.

[0078] The flight angle fine-tuning strategy, flight speed fine-tuning strategy, and flight altitude fine-tuning strategy are integrated and input into a multi-objective optimization model. Angle unit weighting factor, speed unit weighting factor, and altitude unit weighting factor are extracted, and after weighted coupling, the integrated output is obtained to obtain the UAV's comprehensive attitude fine-tuning strategy and control the UAV to execute it.

[0079] It should be noted that the basic idea of ​​multi-objective optimization is to use a mathematical model to handle the relationships between multiple objectives and obtain the most suitable solution. The angle unit weighting factor represents the importance of the flight angle to the overall fine-tuning strategy. If the signal strength is highly sensitive to changes in the flight angle, a higher weight may be needed to ensure that the flight angle fine-tuning maximizes signal reception. The velocity unit weighting factor reflects the impact of flight speed on the final fine-tuning result. If speed has a significant impact on task completion time and image quality, the velocity weighting factor may be larger. The altitude unit weighting factor represents the contribution of flight altitude to image quality and signal strength. When flight altitude has a significant impact on the signal, it should be assigned a larger weight. By multiplying the adjustment amount of the sub-fine-tuning strategy with its corresponding unit weighting factor, the contribution value of each fine-tuning parameter to the overall optimization result is obtained. Then, these contribution values ​​are combined through addition to output the final overall attitude fine-tuning strategy.

[0080] The output yields a comprehensive attitude fine-tuning strategy for the UAV and controls the UAV to execute it. It also includes signal feedback, specifically:

[0081] After the drone performs its mission, it provides signal feedback. If the signal strength of the drone is greater than or equal to the signal strength threshold, it is determined that the attitude fine-tuning is completed and the power grid facility inspection task continues.

[0082] If the signal strength of the drone is still less than the signal strength threshold, it is determined that the attitude needs to be readjusted. After updating the drone attitude data, the drone's comprehensive attitude fine-tuning strategy is entered. If the number of times the comprehensive attitude fine-tuning strategy is executed reaches the upper limit, but the signal strength is still less than the signal strength threshold, it is determined that there is a hardware fault in the drone's radio frequency module, and a return-to-warehouse repair command is issued to the drone.

[0083] It should be noted that, in this embodiment of the invention, the logical flow is to first determine the height fine-tuning strategy and then consider the comprehensive attitude fine-tuning strategy, which has the following advantages:

[0084] Flight altitude has a direct and significant impact on signal strength. Adjusting the altitude can quickly optimize signal quality; therefore, adjusting the flight altitude first can resolve weak signal issues in the shortest time, improving system response efficiency. Secondly, directly applying a comprehensive fine-tuning strategy might require adjusting multiple parameters, such as flight angle and speed, increasing computational complexity and adjustment time. Fine-tuning the altitude first not only effectively improves the signal but also avoids unnecessary adjustments, simplifying computation and operation. Furthermore, adjusting flight altitude typically requires a smaller adjustment range, thus saving energy, reducing UAV battery consumption and computational resource consumption, and leaving more operational space for subsequent tasks. Finally, by adjusting the flight altitude first, the UAV will be in a position with a stronger signal, providing a more stable signal foundation for subsequent fine-tuning of angle, speed, etc., thereby improving the overall system stability and accuracy. In short, the strategy of fine-tuning the altitude first can quickly resolve signal problems and provide a more stable environment for subsequent comprehensive fine-tuning strategies, optimizing the entire inspection process.

[0085] The signal disconnection module is used to cache inspection data using a delay-tolerant network when the signal is interrupted, and resend it after reconnection. At the same time, it reports UAV information based on Beidou short messages at fixed intervals.

[0086] The BeiDou Navigation Satellite System not only provides positioning, navigation, and timing functions, but also offers short message communication services. It allows users to send short messages via satellite to ground stations or other satellite systems; the short message format is typically concise. Compared to traditional wireless communication, BeiDou short messages can maintain communication even in remote areas without ground communication base stations or in areas with weak signals. At preset fixed intervals, drones automatically transmit their flight status and location to the drone management base station via the BeiDou system. The drone information includes drone location information, flight status information, and battery level information.

[0087] In this embodiment, as Figure 2 The present invention provides an intelligent inspection method for power grid facilities based on unmanned aerial vehicles (UAVs) and BeiDou technology, specifically including:

[0088] The BeiDou data of the power grid facility area is obtained by using BeiDou technology and input into the signal coverage generation model. At the same time, the additional attenuation of signal loss due to multi-source environmental data is superimposed to output the initial signal coverage estimation map. During the UAV inspection, the initial signal coverage estimation map is corrected online by a Gaussian process to obtain the corrected signal coverage estimation map.

[0089] Output several candidate routes, obtain the image quality constraint parameters of each candidate route at each power grid facility monitoring point, and simultaneously retrieve the signal correction coverage estimation map. Select the best route from all options to obtain the UAV inspection route.

[0090] When the UAV enters the weak signal area marked in the signal correction coverage estimation map, real-time UAV attitude data is collected, and the UAV attitude fine-tuning strategy is determined based on the real-time signal strength fed back by the UAV radio frequency module.

[0091] When the signal is interrupted, the inspection data is cached using a delay-tolerant network and resent after reconnection. At the same time, UAV information is reported based on BeiDou short messages at fixed intervals.

[0092] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0093] The preferred embodiments of the present invention disclosed above are merely illustrative of the invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the invention to specific implementation methods. Clearly, many modifications and variations can be made based on the content of this specification. The selection and detailed description of these embodiments in this specification are intended to better explain the principles and practical applications of the invention, thereby enabling those skilled in the art to better understand and utilize the invention. Any modifications or variations that do not deviate from the structure of the invention or exceed the scope defined by the invention should fall within the protection scope of the invention.

Claims

1. A smart inspection system for power grid facilities based on unmanned aerial vehicles (UAVs) and BeiDou technology, characterized in that, include: The map signal sensing module is used to acquire BeiDou data of the power grid facility area through BeiDou technology and input it into the signal coverage generation model. At the same time, it superimposes the additional attenuation of signal loss due to multi-source environmental data and outputs an initial signal coverage estimation map. During the UAV inspection, the initial signal coverage estimation map is corrected online through a Gaussian process to obtain a corrected signal coverage estimation map. The route planning module is used to output several candidate routes, obtain the image quality constraint parameters of each candidate route at each power grid facility monitoring point, and simultaneously retrieve the signal correction coverage estimation map to comprehensively select the best UAV inspection route. The comprehensive selection of optimal drone inspection routes specifically includes: The path planning algorithm is used to output several candidate routes, and the image quality constraint parameters of each candidate route at each power grid facility monitoring point are obtained, including the drone shooting angle, drone shooting distance, drone flight altitude and drone flight speed. The optimal image quality parameters for each power grid facility monitoring point are obtained from the database, including the optimal drone shooting angle, optimal drone shooting distance, optimal drone flight altitude, and optimal drone flight speed. The image quality constraint parameters of each candidate route at each power grid facility monitoring point are compared one by one with the corresponding optimal image quality parameters, and then weighted and coupled to obtain the image quality constraint value of each candidate route at each power grid facility monitoring point. Simultaneously, the signal prediction update values ​​of each power grid facility monitoring point in the signal correction coverage estimation map are retrieved. After the signal strength correlation is performed on the image quality constraint values ​​of each candidate route at each power grid facility monitoring point, the image quality constraint values ​​of each candidate route at each power grid facility monitoring point are statistically processed to obtain the comprehensive image quality constraint value of each candidate route. The image quality constraint comprehensive values ​​of each candidate route are sorted from high to low, and the candidate route with the highest image quality constraint comprehensive value is selected as the UAV inspection route. The signal enhancement module is used to collect real-time UAV attitude data when the UAV enters the weak signal area marked in the signal correction coverage estimation map, and to determine the UAV attitude fine-tuning strategy based on the real-time signal strength fed back by the UAV radio frequency module. The signal disconnection module is used to cache inspection data using a delay-tolerant network when the signal is interrupted, and resend it after reconnection. At the same time, it reports UAV information based on Beidou short messages at fixed intervals.

2. The intelligent inspection system for power grid facilities based on UAV and BeiDou technology according to claim 1, characterized in that: The initial coverage estimation map of the output signal is obtained through the following process: The BeiDou data of the power grid facility area is obtained through BeiDou technology, including positioning data, communication network data, DEM terrain data and power grid facility location data. The data is then input into the signal coverage generation model, which is a model based on the signal propagation attenuation theory. This model is used to generate a visual signal coverage image of the signal strength in a specified area. The original signal coverage estimation map of the signal strength is then mapped in the signal coverage generation model. Acquire multi-source environmental data, including historical meteorological data, electromagnetic interference data, and basic environmental data, and input them into the signal coverage generation model. The additional attenuation of the signal strength corresponding to the multi-source environmental data is mapped into the signal coverage generation model. This data is then superimposed on the original coverage estimation map to output the initial signal coverage estimation map.

3. The intelligent inspection system for power grid facilities based on UAV and BeiDou technology according to claim 1, characterized in that: The process of obtaining the corrected signal coverage estimate by online correction of the initial signal coverage estimate using a Gaussian process is as follows: During the UAV inspection process, real-time location signal data is collected, including signal strength, real-time location and flight status. Based on the initial signal coverage estimation map, a Gaussian process model is initialized, and the covariance function of the Gaussian process is set. The covariance function is used to define the correlation of signal strength at different locations. The real-time location signal data is input into the Gaussian process model, and the signal prediction update value and variance of the initial signal coverage estimation map are output. The initial signal coverage estimation map is then corrected to obtain the corrected signal coverage estimation map.

4. The intelligent inspection system for power grid facilities based on UAV and BeiDou technology according to claim 1, characterized in that: The step of collecting real-time UAV attitude data when the UAV enters a weak signal area marked in the signal correction coverage estimation map specifically includes: The system monitors the signal strength of the drone in real time. When the signal strength of the drone is lower than the preset signal strength threshold, the location of the drone is marked as a weak signal area, and the signal coverage generation model is updated accordingly. During the collection of real-time UAV attitude data, the influence of electromagnetic interference on the real-time UAV attitude data is corrected by combining the signal correction coverage estimation map. The corrected real-time UAV attitude data is then output, which includes the UAV attitude angle, flight altitude, flight speed, and flight acceleration.

5. The intelligent inspection system for power grid facilities based on UAV and BeiDou technology according to claim 1, characterized in that: The strategy for determining the attitude fine-tuning of the UAV specifically includes: The signal strength difference is obtained by subtracting the signal strength of the drone from the signal strength threshold. At the same time, the drone's flight altitude is extracted and compared with the preset signal influence altitude threshold. If the drone's flight altitude is lower than the signal influence altitude threshold, the signal strength difference is entered into the preset mapping set of signal strength difference - flight altitude increase value in the database for mapping and matching to obtain the drone's flight altitude increase value. Meanwhile, the flight altitude increase threshold is extracted from the database and the flight altitude increase value is compared with the flight altitude increase threshold. If the flight altitude increase value is less than or equal to the flight altitude increase threshold, the flight altitude fine-tuning sub-strategy is selected and the drone is controlled to execute the flight altitude increase. If the drone's flight altitude is greater than or equal to the signal influence altitude threshold, or the flight altitude increase value is greater than the flight altitude increase threshold, then the drone's comprehensive attitude fine-tuning strategy is selected to adjust the drone.

6. The intelligent inspection system for power grid facilities based on UAV and BeiDou technology according to claim 5, characterized in that: The control of the drone to increase its flight altitude also includes feedback after the altitude increase, specifically including: After the drone has increased its flight altitude, the signal strength of the drone is monitored again. If the signal strength of the drone is greater than or equal to the signal strength threshold, the attitude fine-tuning is completed and the power grid facility inspection task continues. If the signal strength of the drone is still less than the signal strength threshold, it is determined that the attitude needs to be readjusted. After updating the drone attitude data, the drone's comprehensive attitude fine-tuning strategy is initiated.

7. The intelligent inspection system for power grid facilities based on UAV and BeiDou technology according to claim 5, characterized in that: The selected UAV comprehensive attitude fine-tuning strategy adjusts the UAV, and the specific processing conditions are as follows: The comprehensive attitude fine-tuning strategy for UAVs includes flight angle fine-tuning sub-strategies, flight speed fine-tuning strategies, and flight altitude fine-tuning strategies. The flight angle fine-tuning sub-strategy includes yaw angle fine-tuning and pitch angle fine-tuning. It retrieves the location of the signal base station, obtains the relative angle between the UAV radio frequency module and the signal base station transmission module, and obtains the optimal signal angle. The target adjustment angle is obtained by subtracting the relative angle from the optimal signal angle. The flight speed fine-tuning sub-strategy includes inputting the signal strength difference into a pre-stored mapping set of signal strength difference-flight speed reduction values ​​in the database and performing mapping matching to obtain the flight speed reduction value of the UAV; The flight angle fine-tuning strategy, flight speed fine-tuning strategy, and flight altitude fine-tuning strategy are integrated and input into a multi-objective optimization model. Angle unit weighting factor, speed unit weighting factor, and altitude unit weighting factor are extracted, and after weighted coupling, the integrated output is obtained to obtain the UAV's comprehensive attitude fine-tuning strategy and control the UAV to execute it.

8. The intelligent inspection system for power grid facilities based on UAV and BeiDou technology according to claim 7, characterized in that: The output yields a comprehensive attitude fine-tuning strategy for the UAV and controls the UAV to execute it. It also includes signal feedback, specifically: After the drone performs its mission, it will provide signal feedback. If the signal strength of the drone is greater than or equal to the signal strength threshold, it will be determined that the attitude fine-tuning is completed and the power grid facility inspection mission will continue. If the signal strength of the drone is still less than the signal strength threshold, it is determined that the attitude needs to be readjusted. After updating the drone attitude data, the drone's comprehensive attitude fine-tuning strategy is entered. If the number of times the comprehensive attitude fine-tuning strategy is executed reaches the upper limit, it is determined that there is a hardware fault in the drone's radio frequency module, and a return-to-warehouse repair command is issued to the drone.

9. A method applied to the intelligent inspection system for power grid facilities based on UAV and BeiDou technology as described in any one of claims 1-8, characterized in that: The BeiDou data of the power grid facility area is obtained by using BeiDou technology and input into the signal coverage generation model. At the same time, the additional attenuation of signal loss due to multi-source environmental data is superimposed to output the initial signal coverage estimation map. During the UAV inspection, the initial signal coverage estimation map is corrected online by Gaussian process to obtain the signal correction coverage estimation map. Output several candidate routes, obtain the image quality constraint parameters of each candidate route at each power grid facility monitoring point, and retrieve the signal correction coverage estimation map. Select the best route from all options to obtain the UAV inspection route. When the UAV enters the weak signal area marked in the signal correction coverage estimation map, real-time UAV attitude data is collected, and the UAV attitude fine-tuning strategy is determined based on the real-time signal strength fed back by the UAV radio frequency module. When the signal is interrupted, the inspection data is cached using a delay-tolerant network and resent after reconnection. At the same time, UAV information is reported based on BeiDou short messages at fixed intervals.

Images