Remote power cruise unmanned aerial vehicle control system

Abstract

The invention discloses a remote electric power cruise unmanned aerial vehicle control system, and relates to the technical field of unmanned aerial vehicle control, and the system comprises the following components: a data collection module, a data transmission and storage module, an intelligent data analysis and prediction module, and a fault early warning and maintenance decision module. By integrating the functions of data acquisition, transmission, intelligent analysis and fault early warning, real-time monitoring and accurate prediction of the operation state of key components of the unmanned aerial vehicle are realized, the system can find and early warn potential faults in time, and unmanned aerial vehicle shutdown or accidents caused by the faults are avoided, so that the efficiency and safety of electric power cruise are remarkably improved, and the safety of electric power cruise is improved. Meanwhile, the system can also automatically adjust the flight state or task scheduling of the unmanned aerial vehicle according to a fault prediction result, thereby ensuring the continuity and accuracy of an electric power inspection task.

Description

Technical Field

[0001] The present invention relates to the technical field of unmanned aerial vehicle control, and in particular to a long-distance electric cruise unmanned aerial vehicle control system. Background Art

[0002] As power inspection tasks become increasingly complex and diverse, the requirements for the stability and reliability of drones in power patrols are becoming increasingly higher. Traditional power inspection methods have problems such as long inspection cycles, low efficiency, and major safety hazards, and can no longer meet the current power industry's needs for efficient and safe inspections.

[0003] In traditional UAV electric cruise systems, intelligent data analysis and prediction modules often have many shortcomings. On the one hand, traditional modules lack efficiency and accuracy in data collection and processing, and cannot obtain the operating status of the UAV and the wear of key components in real time and accurately. On the other hand, traditional modules are relatively simple in fault prediction algorithms, lack sufficient adaptability and robustness, and are difficult to cope with complex and changeable flight environments and failure modes. This leads to greater limitations and uncertainties in fault warning and maintenance decisions in traditional systems, and makes it impossible to promptly and effectively discover and resolve potential faults, thus affecting the efficiency and safety of electric cruise.

[0004] In summary, the traditional UAV power cruise system has obvious deficiencies in the intelligent data analysis and prediction modules, and cannot meet the current power industry's needs for efficient and safe inspections. Therefore, it is particularly important to develop a long-distance power cruise UAV control system. Summary of the invention

[0005] The purpose of the present invention is to make up for the shortcomings of the prior art and provide a long-distance electric cruise UAV control system, which can realize real-time monitoring and accurate prediction of the operating status of key components of the UAV by integrating advanced data acquisition, transmission, intelligent analysis and fault warning functions, thereby improving the efficiency and safety of electric cruise.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: a long-distance electric cruise UAV control system, the system includes the following components: a data acquisition module, a data transmission and storage module, an intelligent data analysis and prediction module, and a fault warning and maintenance decision module;

[0007] The data acquisition module: deploys sensors on key components of the drone and transmits data to the processing unit with extremely low latency through an optimized high-speed data transmission link, which uses time division multiplexing and dynamic bandwidth allocation technology;

[0008] The data transmission and storage module: uses a data compression algorithm based on information entropy and dictionary tree to compress data, uses adaptive modulation and demodulation technology to dynamically adjust parameters during wireless transmission, and the UAV local and ground control station build a distributed storage architecture based on blockchain to ensure data security, redundancy, non-tamperability and integrity;

[0009] The intelligent data analysis and prediction module: the ground control station is equipped with computing equipment to run a fault prediction algorithm based on a dynamic weighted neural network. The network structure includes an input layer, a hidden layer, and an output layer. The input layer receives multi-dimensional operation data from the drone, and the hidden layer extracts and fuses features of the data through an adaptive weight adjustment function.

[0010] The adaptive weight adjustment function is:

[0011]

[0012] in, Indicated in Moment The input node and The weights between the hidden layer nodes, and The weight adjustment coefficient is determined by cross-validation of historical data, and is dynamically adjusted according to the prediction error during the training process. express Moment The first input node data, express Moment The mean of the input node data;

[0013] The output layer outputs the fault prediction probability based on the hidden layer processing results. Through training with a large amount of historical data and real-time data, the network parameters are continuously optimized to identify the characteristic patterns of normal and abnormal operating states.

[0014] The fault warning and maintenance decision module: When the fault prediction probability output by the intelligent data analysis and prediction module exceeds the set threshold, the system immediately triggers a fault warning signal, and determines the optimal maintenance plan by quantitatively analyzing the cost, time, and task impact of different maintenance strategies.

[0015] Furthermore, the sensor in the data acquisition module adopts a combination of three-axis acceleration and frequency analysis for the vibration sensor of the motor. When the motor is running, the frequency analysis of the vibration signal is performed through an improved fast Fourier transform algorithm, and a phase compensation factor is introduced to eliminate the spectrum leakage problem caused by the non-stable operation of the motor. By analyzing the vibration amplitude and phase changes in different frequency bands, the hidden dangers of bearing wear and rotor imbalance of the motor can be more accurately judged, providing richer and more accurate data support for subsequent fault prediction.

[0016] Furthermore, the distributed storage architecture in the data transmission and storage adopts an improved consensus algorithm based on Byzantine fault tolerance. In the distributed network composed of the local storage node of the drone and the storage node of the ground control station, each node verifies the data consistency through the improved consensus algorithm after receiving the data. The node reputation evaluation mechanism is introduced into the algorithm. The weight of the node in the consensus process is dynamically adjusted according to the historical data transmission accuracy and response time indicators of the node. For nodes with high reputation, a higher voting weight is given to ensure that data consistency can be achieved quickly and accurately in the presence of malicious nodes or network failures, thereby ensuring the reliability and security of data storage.

[0017] Furthermore, the dynamic weighted neural network in the intelligent data analysis and prediction module adopts an adaptive learning rate adjustment strategy during the training process. Dynamic adjustment according to the change of prediction error, the formula is:

[0018]

[0019] in, is the initial learning rate, is the adjustment factor, express Moment The actual fault state of the samples, It indicates the predicted fault state. When the prediction error decreases, the learning rate decreases slowly, making the model convergence more stable. When the prediction error increases, the learning rate increases, speeding up the model parameter update and improving the model's adaptability to new data features.

[0020] Furthermore, the maintenance decision model in the fault warning and maintenance decision module takes into account the urgency of the power patrol mission. For UAVs that are performing power inspection missions in key areas, if a minor fault is predicted, a real-time remote adjustment strategy is preferentially adopted. Adjustment instructions are sent to the UAV through the ground control station to reduce the load of the faulty components and maintain mission execution. At the same time, the development of the fault is monitored in real time. After the mission is completed or reaches a safe area, it will arrange to return to the base for maintenance. For serious faults, the emergency return procedure is immediately initiated, and the backup UAV is coordinated to take over the mission to ensure the continuity and safety of the power patrol mission.

[0021] Furthermore, the data acquisition module also includes an environmental parameter acquisition submodule, which installs temperature and humidity sensors, air pressure sensors, and wind speed and direction sensors on the UAV to collect the temperature, humidity, air pressure, wind speed and wind direction parameters of the UAV flight environment in real time. These environmental parameters are integrated with the operating data of the UAV components for analysis. In high temperature and high humidity environments, the performance of the battery will be affected. Through the correlation analysis of environmental parameters and battery operating data, battery failures can be predicted more accurately. At the same time, under complex meteorological conditions, environmental parameters are used to assist the flight control and fault diagnosis of the UAV, thereby improving the adaptability and reliability of the UAV in harsh environments.

[0022] Furthermore, the adaptive weight adjustment function in the intelligent data analysis and prediction module is:

[0023]

[0024] in, Indicated in Moment The input node and The weights between the hidden layer nodes, and The weight adjustment coefficient is determined by cross-validation of historical data, and is dynamically adjusted according to the prediction error during the training process. express Moment The first input node data, express Moment The mean of the input node data.

[0025] Furthermore, the intelligent data analysis and prediction module also has the function of updating the fault mode library. When the drone actually fails and is repaired, the operating data, fault phenomenon and repair result information before the failure are added to the fault mode library as new samples. By learning the new samples, the parameters and structure of the dynamic weighted neural network will be adaptively adjusted. For newly emerging fault modes, the model will automatically generate new feature recognition rules, continuously enrich and improve the fault prediction capability, and improve the system's recognition and prediction accuracy for various potential faults.

[0026] Furthermore, the fault warning and maintenance decision module exchanges data with the asset management system, power dispatching system and geographic information system of the power company. When interacting with the asset management system, when it is determined that the drone needs maintenance, the system automatically sends maintenance requirements, including parts and personnel skill requirements. The asset management system feeds back maintenance records and updates the archives, interacts with the power dispatching system, adjusts the power cruise mission scheduling in real time based on faults and maintenance decisions, interacts with the geographic information system, optimizes fault handling and mission execution path planning based on geographic location, and realizes multi-system collaboration to ensure efficient power cruise missions.

[0027] Furthermore, the data transmission and storage module adopts multiple encryption technologies during the data transmission process. First, the collected data is encrypted by AES to ensure the confidentiality of the data during transmission. Then, a message authentication code (MAC) based on the hash algorithm is added to the data packet header. An improved hash algorithm is used to combine the data timestamp and the unique identification of the drone to generate the MAC. The receiving end verifies the MAC to ensure the integrity and source reliability of the data. In areas with weak wireless network signals or large interference, it automatically switches to a backup encryption channel. The channel uses quantum key distribution technology to further improve the security of data transmission and prevent data from being stolen or tampered with.

[0028] Compared with the existing technology, this long-distance electric cruise drone control system has the following beneficial effects:

[0029] 1. The system integrates data collection, transmission, intelligent analysis and fault warning functions to achieve real-time monitoring and accurate prediction of the operating status of key components of the UAV. The system can promptly detect and warn of potential faults, avoiding UAV shutdowns or accidents caused by faults, thereby significantly improving the efficiency and safety of power cruising. At the same time, the system can automatically adjust the flight status or task scheduling of the UAV according to the fault prediction results to ensure the continuity and accuracy of power inspection tasks.

[0030] 2. The system ensures the secure redundant storage and confidentiality of the collected data during transmission by adopting a distributed storage architecture based on blockchain and multiple encryption technologies. This design not only prevents the data from being stolen or tampered with, but also ensures the data's non-tamperability and integrity, providing reliable data support for subsequent fault analysis and processing. In addition, the system further improves the stability and security of data transmission through adaptive modulation and demodulation technology and backup encryption channels, providing a strong guarantee for the smooth progress of the electric cruise mission.

[0031] Other advantages, objectives and features of the present invention will be set forth in part in the following description and, in part, will be apparent to those skilled in the art based on an examination of the following or may be taught from the practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0033] Figure 1 Provide a flow chart for the function implementation of the control system of a long-distance electric cruise UAV;

[0034] Figure 2 This is the overall architecture flow chart of the long-distance electric cruise UAV control system. DETAILED DESCRIPTION

[0035] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the specific implementation mode, structure, characteristics and effects of the present invention are described in detail below in combination with the accompanying drawings and preferred embodiments.

[0036] Example 1: This example describes that in remote mountainous areas, where the terrain is complex and transportation is inconvenient, the inspection of power lines is difficult. A power company uses a long-distance power cruise drone control system to conduct line inspections.

[0037] When the drone flies in mountainous areas, the data acquisition module works fully. The sensors deployed in its key components, such as the vibration sensor for the motor, adopt a combination of three-axis acceleration and frequency analysis. When the motor is running, the vibration signal is frequency analyzed using an improved fast Fourier transform algorithm, and a phase compensation factor is introduced to eliminate the spectrum leakage problem and accurately obtain the motor vibration condition. At the same time, the temperature and humidity sensors, air pressure sensors, and wind speed and direction sensors in the environmental parameter acquisition submodule collect the harsh environmental parameters of the mountainous area, such as low temperature and strong wind, and transmit them together with the operating data of the drone components to the processing unit with extremely low latency through an optimized high-speed data transmission link (using time division multiplexing and dynamic bandwidth allocation technology).

[0038] In order to adapt to the complex signal environment in mountainous areas, the data transmission and storage module uses a data compression algorithm based on information entropy and dictionary tree to reduce the amount of data and facilitate transmission. During wireless transmission, adaptive modulation and demodulation technology is used to dynamically adjust parameters. The local drone and ground control station build a distributed storage architecture based on blockchain, adopt an improved consensus algorithm based on Byzantine fault tolerance, and introduce a node reputation evaluation mechanism. After each node receives the data, it performs consistency verification. Nodes with high reputation have higher voting weights to ensure that the data is secure, redundant, tamper-proof and complete. Multiple encryption technologies are also used for data transmission. AES encryption is performed first, and a message authentication code (MAC) generated based on an improved hash algorithm (combined with timestamp and drone unique identification) is added to the data packet header. In areas with weak signals or large interference, it automatically switches to a backup encryption channel using quantum key distribution technology.

[0039] The ground control station is equipped with computing equipment, the intelligent data analysis and prediction module runs the fault prediction algorithm, the input layer of the dynamic weighted neural network receives the multi-dimensional operation data of the drone, and the hidden layer uses the adaptive weight adjustment function Perform feature extraction and fusion. During the training process, an adaptive learning rate adjustment strategy is adopted ( , dynamically adjust the learning rate according to the prediction error, optimize the network parameters, and identify the characteristic patterns of normal and abnormal operating states.

[0040] When the UAV motor vibrates abnormally and the fault prediction probability output by the intelligent data analysis and prediction module exceeds the set threshold, the fault warning and maintenance decision module triggers a warning signal. Since the UAV is performing a key area inspection mission, the system judges it to be a minor fault and gives priority to the real-time remote adjustment strategy. The ground control station sends adjustment instructions to the UAV to reduce the motor load. At the same time, it monitors the development of the fault in real time, and arranges it to return to the base for maintenance after the mission is completed. It also interacts with the power company's asset management system, power dispatching system and geographic information system for data, such as sending maintenance requirements to the asset management system, coordinating with the power dispatching system to adjust task scheduling, and using the geographic information system to optimize the return path.

[0041] Example 2: This example describes a coastal area where numerous power facilities provide stable power support for local economic development and residents' lives. However, the area is attacked by sea breeze and salt spray all year round. The operating environment of the power facilities is extremely harsh and the risk of failure remains high. In order to ensure the stability and reliability of power supply, the local power company uses a long-distance power cruise drone control system for routine inspections.

[0042] The drone's sensors closely monitor the operating conditions of components, the motor vibration sensors accurately collect data, and the environmental parameter collection submodule focuses on humidity, wind speed and direction, because high humidity and salt mist carried by strong winds are extremely harmful to power equipment. For example, salt mist may cause line corrosion and short circuit failures. These environmental data are integrated with component operation data to provide key information for analyzing equipment health.

[0043] Under the influence of sea breeze and salt spray, data transmission faces challenges. The system ensures stable data transmission through data compression and adaptive modulation and demodulation technology. Distributed storage architecture and multiple encryption technologies ensure data security and resist interference from complex coastal electromagnetic environments and potential data theft risks.

[0044] The dynamic weighted neural network is trained based on the operating characteristics of power facilities and drones in coastal areas. The adaptive weight adjustment function and adaptive learning rate strategy enable the model to be continuously optimized and accurately identify normal and abnormal operating conditions. For example, by learning the long-term accumulated equipment operation data under the influence of salt spray erosion, the model can predict in advance possible equipment failures due to salt spray corrosion.

[0045] Once the fault prediction triggers an early warning, for drones performing inspection missions in important areas, if a minor fault occurs, such as abnormal battery consumption but the mission can still be maintained, the ground control station remotely adjusts the equipment parameters, reduces energy consumption and monitors in real time. If it is a serious fault, such as a communication system failure, the emergency return is immediately initiated and the backup drone is coordinated to take over the mission. This module interacts with related systems to optimize maintenance and mission execution plans based on coastal geographic information and power dispatch requirements.

[0046] The above description is only a preferred embodiment of the present invention and does not limit the present invention in any form. Although the present invention has been disclosed as a preferred embodiment as above, it is not used to limit the present invention. Any technical personnel in this field can make some changes or modify the technical contents disclosed above into equivalent embodiments without departing from the scope of the technical solution of the present invention. However, any brief modifications, equivalent changes and modifications made to the above embodiments based on the technical essence of the present invention without departing from the content of the technical solution of the present invention are still within the scope of the technical solution of the present invention.

Claims

1. A long-distance electric cruise UAV control system, characterized in that: The system includes the following components: data acquisition module, data transmission and storage module, intelligent data analysis and prediction module and fault warning and maintenance decision module; the data acquisition module: deploy sensors on key components of the drone, and transmit data to the processing unit with extremely low latency through an optimized high-speed data transmission link. The link adopts time division multiplexing and dynamic bandwidth allocation technology; the data transmission and storage module: compresses data using a data compression algorithm based on information entropy and dictionary tree, and dynamically adjusts parameters using adaptive modulation and demodulation technology during wireless transmission. The local drone and ground control station build a distributed storage architecture based on blockchain to ensure data security, redundancy, non-tamperability and integrity; the intelligent data analysis and prediction module: the ground control station is equipped with a computer Equipment, runs a fault prediction algorithm, which is based on a dynamic weighted neural network. The network structure includes an input layer, a hidden layer and an output layer. The input layer receives multi-dimensional operation data from the drone, and the hidden layer extracts and fuses the data through an adaptive weight adjustment function; the output layer outputs the fault prediction probability according to the processing result of the hidden layer, and continuously optimizes the network parameters through training with a large amount of historical data and real-time data to identify the characteristic patterns of normal and abnormal operation states; the fault warning and maintenance decision module: when the fault prediction probability output by the intelligent data analysis and prediction module exceeds the set threshold, the system immediately triggers a fault warning signal, and determines the optimal maintenance plan by quantitatively analyzing the cost, time, and task impact of different maintenance strategies.

2. A long-distance electric cruise UAV control system according to claim 1, characterized in that: The sensor in the data acquisition module adopts a combination of three-axis acceleration and frequency analysis for the vibration sensor of the motor. When the motor is running, the vibration signal is frequency analyzed through an improved algorithm of fast Fourier transform, and a phase compensation factor is introduced to eliminate the spectrum leakage problem caused by the non-steady operation of the motor, by analyzing the vibration amplitude and phase changes in different frequency bands.

3. A long-distance electric cruise UAV control system according to claim 1, characterized in that: The distributed storage architecture in the data transmission and storage adopts an improved consensus algorithm based on Byzantine fault tolerance. In the distributed network composed of the local storage nodes of the drone and the storage nodes of the ground control station, each node verifies the data consistency through the improved consensus algorithm after receiving the data. The algorithm introduces a node reputation evaluation mechanism, and dynamically adjusts the weight of the node in the consensus process according to the historical data transmission accuracy and response time indicators of the node, and assigns a higher voting weight to nodes with high reputation.

4. A long-distance electric cruise UAV control system according to claim 1, characterized in that: The dynamic weighted neural network in the intelligent data analysis and prediction module adopts an adaptive learning rate adjustment strategy during the training process. Dynamic adjustment according to the change of prediction error, the formula is: ; in, is the initial learning rate, is the adjustment factor, express Moment The actual fault state of the samples, Indicates the predicted fault state. When the prediction error decreases, the learning rate decreases slowly, making the model convergence more stable. When the prediction error increases, the learning rate increases to speed up the model parameter update.

5. The long-distance electric cruise UAV control system according to claim 1 is characterized in that: The maintenance decision model in the fault warning and maintenance decision module takes into account the urgency of the power patrol mission. For UAVs that are performing power patrol missions in key areas, if a minor fault is predicted, a real-time remote adjustment strategy is preferentially adopted. Adjustment instructions are sent to the UAV through the ground control station to reduce the load of the faulty components and maintain mission execution. At the same time, the development of the fault is monitored in real time. After the mission is completed or the safe area is reached, the UAV is arranged to return to the base for maintenance. For serious faults, the emergency return procedure is immediately initiated, and the backup UAV is coordinated to take over the mission.

6. A long-distance electric cruise UAV control system according to claim 1, characterized in that: The data acquisition module also includes an environmental parameter acquisition submodule, which installs a temperature and humidity sensor, an air pressure sensor, and a wind speed and direction sensor on the drone to collect the temperature and humidity, air pressure, wind speed and wind direction parameters of the drone's flight environment in real time. These environmental parameters are integrated with the operating data of the drone components for analysis. At the same time, under complex meteorological conditions, environmental parameters are used to assist the drone's flight control and fault diagnosis.

7. The long-distance electric cruise UAV control system according to claim 1 is characterized in that: The adaptive weight adjustment function in the intelligent data analysis and prediction module is: ; in, Indicated in Moment The input node and The weights between the hidden layer nodes, and The weight adjustment coefficient is determined by cross-validation of historical data, and is dynamically adjusted according to the prediction error during the training process. express Moment The first input node data, express Moment The mean of the input node data.

8. The long-distance electric cruise UAV control system according to claim 1 is characterized in that: The intelligent data analysis and prediction module also has the function of updating the fault mode library. When the drone actually fails and is repaired, the operating data, fault phenomenon and repair result information before the failure are added to the fault mode library as new samples. By learning the new samples, the parameters and structure of the dynamic weighted neural network will be adaptively adjusted. For newly emerging fault modes, the model will automatically generate new feature recognition rules to continuously enrich and improve the fault prediction capabilities.

9. The long-distance electric cruise UAV control system according to claim 1 is characterized in that: The fault warning and maintenance decision module exchanges data with the asset management system, power dispatching system and geographic information system of the power company. When interacting with the asset management system, when it is determined that the drone needs maintenance, the system automatically sends maintenance requirements, including parts and personnel skill requirements. The asset management system feeds back maintenance records and updates the archives, interacts with the power dispatching system, and adjusts the power cruise mission scheduling in real time based on faults and maintenance decisions. It interacts with the geographic information system and optimizes fault handling and mission execution path planning based on geographic location.

10. The long-distance electric cruise UAV control system according to claim 1 is characterized in that: The data transmission and storage module adopts multiple encryption technologies during the data transmission process. First, the collected data is encrypted with AES, and then a message authentication code (MAC) based on a hash algorithm is added to the data packet header. An improved hash algorithm is used to combine the data timestamp and the unique identification of the drone to generate a MAC. The receiving end verifies the MAC and automatically switches to a backup encryption channel in areas with weak wireless network signals or large interference. This channel uses quantum key distribution technology to further improve the security of data transmission and prevent data from being stolen or tampered with.

Images