Transmission, transformation and distribution integrated unmanned aerial vehicle inspection method
By combining GPS positioning algorithm and visually assisted navigation, dynamic path planning and edge computing technology, the problems of inaccurate positioning and poor flight stability in drone inspections are solved, the inspection efficiency and coverage are improved, and in severe weather are achieved stable flight and efficient data processing.
Patent Information
- Application Number
- CN202410980476.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-22
- Publication Date
- 2024-11-12
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
The existing inspection methods for integrated UAVs with transmission and transformation have problems such as inaccurate positioning, poor flight stability, and low patrol coverage and patrol efficiency, especially in bad weather conditions.
The GPS positioning algorithm and visually assisted navigation system are used to accurately navigate the drone, and the inspection route is generated by dynamic path planning algorithm, and the edge computing equipment is used to perform real-time data processing and flight parameter adjustment. The weighted fusion algorithm and sound recognition algorithm are used to generate patrols. inspection reports and transmit data through 5G or satellite communication technology.
It realizes accurate arrival of drones under visual occlusion conditions, improves patrol coverage and efficiency, reduces the impact of bad weather on flight stability, and accelerates data processing through edge computing, reducing the burden on ground control stations.
Smart Images

Figure CN118938949A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of unmanned aerial vehicle inspection technology, and in particular to a transmission, transformation and distribution integrated unmanned aerial vehicle inspection method. Background Art
[0002] The integrated transmission, transformation and distribution drone inspection method uses drones to conduct automatic or semi-automatic inspections of transmission lines, substations and distribution networks. The core of this method is to combine the high efficiency and flexibility of drones with advanced sensors and image processing technology to achieve efficient monitoring of key equipment in the power system.
[0003] Drones can quickly cover large areas, especially in complex or inaccessible terrain. Drones can capture images of transmission lines and equipment in real time through their high-definition cameras and sensors, and diagnose potential problems by analyzing these images. Compared with traditional manual inspections, drone inspections can save a lot of manpower and time costs. In addition, the various sensors carried by drones, such as infrared cameras and high-resolution cameras, can be used to capture detailed images and data, thereby more accurately identifying and analyzing problems. In addition, this inspection method does not require staff to directly contact high-voltage power facilities, greatly reducing the risks of personnel inspections. Therefore, the method of integrating transmission, transformation and distribution with drone inspections is becoming more and more widely used.
[0004] However, the existing transmission, transformation and distribution integrated drone inspection method has the following defects: 1. The drone cannot accurately reach the designated power system inspection area under visual occlusion conditions; 2. The inspection route cannot be adjusted in real time according to the images and sounds of the power system inspection area collected by the drone, resulting in low coverage and inspection efficiency of the drone inspection; 3. Under severe weather conditions (such as strong winds, rain, snow and other low visibility conditions), the flight stability of the drone will be affected by the weather, thereby affecting the accuracy of the inspection results. Summary of the invention
[0005] To this end, the technical problem to be solved by the present invention is to overcome the problems of inaccurate drone positioning, poor flight stability, and low inspection coverage and efficiency in the prior art transmission and distribution integrated drone inspection method.
[0006] In order to solve the above technical problems, the present invention provides a transmission, transformation and distribution integrated UAV inspection method, comprising:
[0007] Based on preset coordinates, the drone uses GPS positioning algorithm and visual assisted navigation system to reach the power system inspection area, collects image information and sound information of the power system inspection area in real time, and transmits the image information and sound information to the ground control station;
[0008] The ground control station uses a dynamic path planning algorithm to generate an inspection route based on the sound information and the image information and sends it to the drone;
[0009] The drone performs inspections based on the inspection route, collects air temperature, humidity, wind speed, and image signals, temperature signals, and sound signals of power facilities in real time, and adjusts the flight altitude and flight speed based on the air temperature, humidity, and wind speed;
[0010] The edge computing device on the drone uses a weighted fusion algorithm and a sound recognition algorithm to generate an inspection report based on the image signal, temperature signal and sound signal of the power facility, and transmits the inspection report to a ground control station.
[0011] Preferably, the GPS positioning algorithm uses Kalman filtering to predict the position of the drone, and the formula of Kalman filtering is expressed as:
[0012] x k|k =x k|k-1 +K k (y k -H k x k|k-1 ),
[0013] Among them, x k|k represents the predicted position of the drone at time k; x k|k-1 K represents the predicted position of the drone at time k; k represents the Kalman gain; y k H represents the measured value of the drone position at time k; k Represents the measurement mapping matrix.
[0014] Preferably, the positioning accuracy of the GPS positioning algorithm is [x-0.5, x+0.5], where x represents a preset coordinate.
[0015] Preferably, the dynamic path planning algorithm is a genetic algorithm and a particle swarm optimization algorithm, the crossover probability of the genetic algorithm is 0.7, the mutation probability is 0.1, the number of particles of the particle swarm optimization algorithm is 50, and the number of iterations is 100.
[0016] Preferably, the formula of the weighted fusion algorithm is:
[0017]
[0018] Where D represents the decision output; ω i represents the weight of the i-th sensor carried by the drone; d i Represents the signal of the power facility collected by the i-th sensor carried by the drone.
[0019] Preferably, the formula of the sound recognition algorithm is:
[0020] f t =σ(W f ·[h t-1 ,x t ]+b f ),
[0021] Among them, f t represents the forget gate; σ represents the sigmoid activation function; W f 、b f represents network parameters; h t-1 represents the hidden state at the previous moment; x t Represents the input at the current moment.
[0022] Preferably, after transmitting the inspection report to the ground control station, the method further includes: the ground control station predicting and classifying faults of the inspection report based on a random forest fault prediction model.
[0023] Preferably, the random forest fault prediction model is expressed as:
[0024]
[0025] Where P(y|x) represents the output of the random forest fault prediction model; B represents the number of trees in the random forest fault model; P b (y|x) represents the prediction result of the bth tree; x represents the inspection report.
[0026] Preferably, the UAV adopts 5G communication technology or satellite communication technology to transmit the image information and the sound information to the ground control station; wherein the data transmission speed of the 5G communication technology is greater than or equal to 1Gbps; the data transmission speed of the satellite communication technology is greater than or equal to 100Mbps.
[0027] Preferably, the drone is equipped with an automatic obstacle avoidance system, which uses ultrasonic and lidar sensors to detect obstacles.
[0028] The transmission, transformation and distribution integrated drone inspection method provided in the present application accurately navigates the drone by combining the GPS positioning algorithm and the visual assisted navigation system, so that the drone can still accurately reach the power system inspection area even in the case of visual obstruction; the ground control station uses a dynamic path planning algorithm to generate a drone inspection route based on the image information and sound information of the power system inspection area collected by the drone in real time, optimizes the coverage of the drone inspection, and thus improves the drone inspection efficiency; during the inspection process based on the inspection route, the drone collects air temperature, humidity, wind speed and image signals, temperature signals and sound signals of power facilities in real time, and adjusts the flight altitude and flight speed based on the air temperature, humidity and wind speed, and avoids the problem of unstable flight of the drone caused by bad weather by adjusting the flight mode in real time according to weather conditions; in addition, the present application uses the edge computing equipment on the drone to process the signals of the power facilities collected by the drone nearby and generate an inspection report, which not only improves the inspection efficiency but also reduces the burden on the ground control station. BRIEF DESCRIPTION OF THE DRAWINGS
[0029] In order to make the content of the present invention more clearly understood, the present invention is further described in detail below according to specific embodiments of the present invention in conjunction with the accompanying drawings, wherein:
[0030] Figure 1 Flowchart of the transmission, transformation and distribution integrated drone inspection method provided in this application. DETAILED DESCRIPTION
[0031] The present invention is further described below in conjunction with the accompanying drawings and specific embodiments so that those skilled in the art can better understand the present invention and implement it, but the embodiments are not intended to limit the present invention.
[0032] The drone is equipped with a high-resolution optical camera, an infrared camera and a directional microphone array. The high-resolution optical camera and infrared camera are used to capture visual images of power facilities. The optical camera is used for visual recognition during the day, while the infrared camera is used at night or in low-light conditions to detect thermal images of equipment. The directional microphone array is used to capture sounds from power facilities, such as arc sounds, noise or other abnormal sounds.
[0033] See also Figure 1 , Figure 1 The flow chart of the transmission, transformation and distribution integrated drone inspection method provided in this application specifically includes:
[0034] S10: The drone uses the GPS positioning algorithm and the visual assisted navigation system to reach the power system inspection area based on the preset coordinates, collects image information and sound information of the power system inspection area in real time, and transmits the image information and sound information to the ground control station;
[0035] S20: The ground control station uses a dynamic path planning algorithm to generate an inspection route based on sound and image information and sends it to the drone;
[0036] S30: The drone conducts inspections based on the inspection route, collects air temperature, humidity, wind speed, and image signals, temperature signals, and sound signals of power facilities in real time, and adjusts the flight altitude and flight speed based on the air temperature, humidity, and wind speed;
[0037] S40: The edge computing device on the drone uses a weighted fusion algorithm and a sound recognition algorithm to generate an inspection report based on the image signal, temperature signal, and sound signal of the power facility, and transmits the inspection report to the ground control station.
[0038] Specifically, edge computing devices can process 100MB of data per second, greatly improving the data processing speed. Processing data locally on edge computing devices can reduce the burden on ground control stations.
[0039] The transmission, transformation and distribution integrated drone inspection method provided in the present application accurately navigates the drone by combining the GPS positioning algorithm and the visual assisted navigation system, so that the drone can still accurately reach the power system inspection area even in the case of visual obstruction; the ground control station uses a dynamic path planning algorithm to generate a drone inspection route based on the image information and sound information of the power system inspection area collected by the drone in real time, optimizes the coverage of the drone inspection, and thus improves the drone inspection efficiency; during the inspection process based on the inspection route, the drone collects air temperature, humidity, wind speed and image signals, temperature signals and sound signals of power facilities in real time, and adjusts the flight altitude and flight speed based on the air temperature, humidity and wind speed, and avoids the problem of unstable flight of the drone caused by bad weather by adjusting the flight mode in real time according to weather conditions; in addition, the present application uses the edge computing equipment on the drone to process the signals of the power facilities collected by the drone nearby and generate an inspection report, which not only improves the inspection efficiency but also reduces the burden on the ground control station.
[0040] Specifically, the GPS positioning algorithm in step S10 uses Kalman filtering to predict the position of the drone. The formula of Kalman filtering is expressed as:
[0041] x k|k =x k|k-1 +K k (y k -H k x k|k-1 ),
[0042] Among them, x k|k represents the predicted position of the drone at time k; x k|k-1K represents the predicted position of the drone at time k; k represents the Kalman gain; y k H represents the measured value of the drone position at time k; k Represents the measurement mapping matrix.
[0043] Among them, the positioning accuracy of the GPS positioning algorithm is [x-0.5, x+0.5], where x represents the preset coordinates; and the visually assisted navigation system can recognize specific signs; the GPS positioning algorithm can provide positioning accuracy within a wide range, and the visually assisted navigation system provides more detailed navigation support when the UAV approaches the target area. Therefore, this application uses a dual guarantee mechanism to ensure that the UAV can accurately navigate to the power system inspection area, and can perform accurate navigation even under visual occlusion conditions.
[0044] Specifically, in some embodiments of the present application, the dynamic path planning algorithm in step S20 is a genetic algorithm and a particle swarm optimization algorithm, whose objective function is the shortest inspection path covering the maximum inspection area, the crossover probability of the genetic algorithm is 0.7, the mutation probability is 0.1, the number of particles in the particle swarm is 50, and the number of iterations is 100.
[0045] Specifically, in some embodiments of the present application, the air temperature range of the air collected by the drone in step S30 is -20°C to 50°C, the humidity range is 10% to 95%, and the wind speed range is 0 to 20 m / s;
[0046] Optionally, in some embodiments of the present application, the UAV adjusts the flight altitude and flight speed based on air temperature, humidity and wind speed, including:
[0047] If the wind speed sensor reports a sudden increase in wind speed to 15m / s, which is greater than the standard wind speed for single-circle flight (usually 10m / s), the drone automatically lowers its flight altitude to reduce wind resistance and automatically reduces its flight speed to a speed that is more suitable for crosswinds; for example, the drone's flight altitude is reduced from 120m to 80m, and the flight speed is reduced from 15m / s to 10m / s;
[0048] If the temperature sensor reports that the temperature has dropped to 5°C, the drone will increase the frequency of battery and important component inspections and increase the flight speed to reduce the impact of low temperature on battery performance.
[0049] If the optical sensor reports reduced visibility, the drone will automatically activate the infrared navigation system and reduce the flight speed to 8m / s to ensure a safe flight distance.
[0050] Specifically, in some embodiments of the present application, the formula of the weighted fusion algorithm in step S40 is:
[0051]
[0052] Where D represents the decision output; ω i represents the weight of the i-th sensor carried by the drone; d i Represents the signal of the power facility collected by the i-th sensor carried by the drone.
[0053] The formula for the sound recognition algorithm is:
[0054] f t =σ(W f ·[h t-1 ,x t ]+b f ),
[0055] Among them, f t represents the forget gate; σ represents the sigmoid activation function; W f 、b f represents network parameters; h t-1 represents the hidden state at the previous moment; x t Represents the input at the current moment.
[0056] Inspection reports include information such as GPS positioning, timestamps, key images, and detected abnormal problems; inspection reports include continuous data points within at least 3 seconds to ensure accurate recording of abnormal events. The subsequent process is in-depth data analysis: the continuous data points included in the report allow for more in-depth data analysis. Through a detailed review of these data, the nature, duration, and possible causes of abnormal events can be more accurately determined. This analysis helps diagnose faults, identify potential safety hazards, and specify maintenance or repair strategies. Trend identification and prediction. The collection and analysis of long-term series data is crucial to identifying trends and patterns in facility operations. This information can be used to predict possible problems in the future, thereby achieving preventive maintenance, reducing sudden failures and downtime events, and real-time response and decision support. When an abnormal event is confirmed, the system can notify the operation and maintenance team in real time, allowing the operation and maintenance team to respond quickly. This timely feedback mechanism helps to take necessary measures to avoid potential damage or further system failures. The recording of continuous data points also provides a rich source of information for historical data archiving. These historical data are of great value for long-term facility management and performance optimization, and can be used to improve inspection strategies and technical upgrades.
[0057] Therefore, in some embodiments of the present application, after the inspection report is transmitted to the ground control station, it also includes: the ground control station predicts and classifies faults in the inspection report based on the random forest fault prediction model; by detecting and classifying faults in the inspection report, it is possible to perform long-term trend analysis and maintenance recommendations on the inspection data.
[0058] Specifically, the random forest fault prediction model is expressed as:
[0059]
[0060] Where P(y|x) represents the output of the random forest fault prediction model; B represents the number of trees in the random forest fault model; P b (y|x) represents the prediction result of the bth tree; x represents the inspection report.
[0061] Preferably, in some embodiments of the present application, the drone uses 5G communication technology or satellite communication technology to transmit the collected image information and sound information to the ground control station to ensure the stability and high speed of data transmission in various environments. The data transmission speed of 5G communication technology is greater than or equal to 1Gbps, and the data transmission speed of satellite communication technology is greater than or equal to 100Mbps.
[0062] Preferably, the drone is also equipped with an automatic obstacle avoidance system, which uses ultrasonic and lidar sensors to detect obstacles.
[0063] For example, as shown in Table 1, the inspection report provided in the embodiment of the present application is used to predict and classify the faults:
[0064] Table 1
[0065]
[0066] Among them, temperature, humidity, wind speed, thermal imager temperature difference, sound decibels and image damage index are all characteristic data collected during drone inspections. The number of historical failures indicates the number of failures that have occurred in the power facility in the past, and the actual failure type is the predicted failure category of the power facility.
[0067] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.
[0068] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.
[0069] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.
[0070] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.
[0071] Obviously, the above embodiments are merely examples for the purpose of clear explanation and are not intended to limit the implementation methods. For those skilled in the art, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to list all the implementation methods here. The obvious changes or modifications derived therefrom are still within the scope of protection of the present invention.
Claims
1. A transmission, transformation and distribution integrated drone inspection method, characterized in that: include: Based on preset coordinates, the drone uses GPS positioning algorithm and visual assisted navigation system to reach the power system inspection area, collects image information and sound information of the power system inspection area in real time, and transmits the image information and sound information to the ground control station; The ground control station uses a dynamic path planning algorithm to generate an inspection route based on the sound information and the image information and sends it to the drone; The drone performs inspections based on the inspection route, collects air temperature, humidity, wind speed, and image signals, temperature signals, and sound signals of power facilities in real time, and adjusts the flight altitude and flight speed based on the air temperature, humidity, and wind speed; The edge computing device on the drone uses a weighted fusion algorithm and a sound recognition algorithm to generate an inspection report based on the image signal, temperature signal and sound signal of the power facility, and transmits the inspection report to a ground control station.
2. The transmission, transformation and distribution integrated drone inspection method according to claim 1 is characterized in that: The GPS positioning algorithm uses Kalman filtering to predict the position of the drone. The formula of Kalman filtering is expressed as: x k|k =x k|k-1 +K k (y k -H k x k|k-1 ), Among them, x k|k represents the predicted position of the drone at time k; x k|k-1 K represents the predicted position of the drone at time k; k represents the Kalman gain; y k H represents the measured value of the drone position at time k; k Represents the measurement mapping matrix.
3. The transmission, transformation and distribution integrated drone inspection method according to claim 1 is characterized in that: The positioning accuracy of the GPS positioning algorithm is [x-0.5, x+0.5], where x represents the preset coordinates.
4. The transmission, transformation and distribution integrated drone inspection method according to claim 1 is characterized in that: The dynamic path planning algorithm is a genetic algorithm and a particle swarm optimization algorithm. The crossover probability of the genetic algorithm is 0.7, the mutation probability is 0.1, the number of particles of the particle swarm optimization algorithm is 50, and the number of iterations is 100.
5. The transmission, transformation and distribution integrated drone inspection method according to claim 1 is characterized in that: The formula of the weighted fusion algorithm is: Where D represents the decision output; ω i represents the weight of the i-th sensor carried by the drone; d i Represents the signal of the power facility collected by the i-th sensor carried by the drone.
6. The transmission, transformation and distribution integrated drone inspection method according to claim 1 is characterized in that: The formula of the sound recognition algorithm is: f t =σ(W f ·[h t-1 ,x t ]+b f ), Among them, f t represents the forget gate; σ represents the sigmoid activation function; W f , b f represents network parameters; h t-1 represents the hidden state at the previous moment; x t Represents the input at the current moment.
7. The transmission, transformation and distribution integrated drone inspection method according to claim 1 is characterized in that: After the inspection report is transmitted to the ground control station, the method further includes: the ground control station performs fault prediction and classification on the inspection report based on a random forest fault prediction model.
8. The transmission, transformation and distribution integrated drone inspection method according to claim 7 is characterized in that: The random forest fault prediction model is expressed as: Where P(y|x) represents the output of the random forest fault prediction model; B represents the number of trees in the random forest fault model; P b (y|x) represents the prediction result of the bth tree; x represents the inspection report.
9. The transmission, transformation and distribution integrated drone inspection method according to claim 1 is characterized in that: The drone uses 5G communication technology or satellite communication technology to transmit the image information and the sound information to a ground control station; wherein the data transmission speed of the 5G communication technology is greater than or equal to 1Gbps; the data transmission speed of the satellite communication technology is greater than or equal to 100Mbps.
10. The transmission, transformation and distribution integrated drone inspection method according to claim 1 is characterized in that: The drone is equipped with an automatic obstacle avoidance system that uses ultrasonic and lidar sensors to detect obstacles.
Citation Information
Patent Citations
Transformer substation equipment sound fault detection and positioning method based on deep learning
CN112183647A
Method and system for comprehensive inspection of power equipment
CN112731086A
Unmanned aerial vehicle remote power transmission line intelligent inspection system and method
CN112731960A
Inspection method and device for power transmission line
CN113485433A
Distribution network routing inspection method combining high-precision positioning of the unmanned aerial vehicle and visual tracking technology
CN113485441A
Cited By
Beidou positioning and AI fused intelligent line dynamic monitoring system
CN119787623A
Inspection path determination method and device, equipment, storage medium and program product
CN120066089A
