Intelligent thunder-vision fusion system and method adaptive to power transmission scene

Abstract

The invention relates to the technical field of data fusion, and discloses a thunder-vision intelligent fusion system and method adapted to a power transmission scene, and the system comprises a radar module which detects the specific information of a power transmission line in real time; the visual module is used for identifying obstacles near the power transmission line, external damage and potential safety hazards; the dynamic adjustment module automatically adjusts the working mode of the sensor; the data processing module is used for comprehensively processing radar and visual data through a data fusion algorithm; the multi-dimensional perception fusion module is used for realizing omnibearing perception; the communication module transmits the detected data to a remote monitoring center in real time; the intelligent analysis and early warning module is used for automatically identifying faults and abnormal conditions and generating an early warning signal; through efficient data fusion and intelligent analysis, the accuracy and response speed of power transmission line monitoring are remarkably improved, the safety and stability of the system are enhanced, powerful support is provided for operation and maintenance personnel, and development of an intelligent power grid is promoted.

Description

Technical Field

[0001] The present invention relates to the technical field of data fusion, and in particular to a radar vision intelligent fusion system and method adapted to power transmission scenarios. Background Art

[0002] As global electricity demand continues to grow, the coverage of the transmission network is expanding, and ensuring the safe and efficient transmission of electricity has become an important task. Transmission lines not only bear a large amount of power transmission, but also need to operate in complex environments, such as extreme weather and changing terrain. This requires the monitoring and maintenance of transmission lines to be efficient and accurate in order to reduce accident risks and improve system reliability.

[0003] Traditional transmission line monitoring mainly relies on manual inspections and a small number of fixed sensors. This method has the problems of low efficiency and small coverage, especially in large-scale transmission networks, where it is difficult to achieve real-time monitoring. In addition, manual inspections are limited by factors such as weather and time, while fixed sensors can only provide local information and cannot fully reflect the status of the transmission line.

[0004] In recent years, with the rapid development of sensor technology, multimodal sensors (such as radar, camera, sound sensor and vibration sensor) have been gradually applied to power transmission line monitoring. These sensors can collect information of different dimensions at the same time, such as status, location and environmental changes, and can provide more comprehensive and accurate monitoring data. Radar technology can achieve long-range high-precision detection, while visual sensors can capture details. The combination of the two can complement blind spots. With the rapid increase in the amount of sensor data, how to effectively integrate and process this data has become the key. Summary of the invention

[0005] The present invention provides a radar-vision intelligent fusion system and method adapted to power transmission scenarios, which are used to solve the technical problems mentioned in the above background technology.

[0006] The present invention provides the following technical solution: a radar-vision intelligent fusion system adapted to power transmission scenarios, the system comprising:

[0007] Radar module, which uses multiple radars for long-distance high-precision detection, can penetrate adverse weather conditions such as rain and fog, and detect specific information of transmission lines in real time;

[0008] Vision module, which uses multiple cameras to capture images and videos to identify obstacles, external damage, and potential safety hazards near the transmission line;

[0009] Dynamic adjustment module, automatically adjusts the sensor's operating mode according to real-time changing environmental conditions such as weather, light, and radar signal strength;

[0010] The data processing module processes radar and visual data through a data fusion algorithm to improve the accuracy and robustness of detection and avoid blind spots or misjudgments of a single sensor.

[0011] The multi-dimensional perception fusion module integrates sound sensors, vibration sensors, and temperature sensors to achieve all-round perception. Through the multi-dimensional data fusion algorithm, it integrates various types of perception information to provide more detailed status monitoring and reduce false alarms and missed alarms.

[0012] The communication module uses communication technology to transmit the detected data to the remote monitoring center in real time, facilitating remote operation and maintenance and timely response;

[0013] The intelligent analysis and early warning module analyzes real-time data based on artificial intelligence algorithms, automatically identifies faults and abnormal conditions, and generates early warning signals to alert operation and maintenance personnel.

[0014] As a preferred solution of the present invention, the radar module is integrated by multiple types of radars, including:

[0015] Millimeter-wave radar: It emits millimeter-wave signals with strong penetration and good anti-interference properties. It is suitable for long-distance target detection in complex weather environments such as rain and fog. It can monitor key data such as the height, sag and swing of transmission lines in real time.

[0016] LiDAR: It has extremely high spatial resolution, suitable for accurately detecting the three-dimensional shape of power transmission lines. It can generate high-precision 3D images to assist in identifying objects and potential obstacles near power transmission lines.

[0017] Microwave radar: Suitable for large-scale monitoring and can provide stable signal feedback in the global status detection of long-distance transmission lines.

[0018] As a preferred solution of the present invention, the visual module integrates multiple types of cameras, including:

[0019] HD Camera: High-resolution cameras are used to capture high-definition images and videos of transmission lines, helping to identify subtle damage on the line surface;

[0020] Infrared camera: used to work at night or in low-light environments, and to detect abnormal heat spots and hidden safety hazards by capturing thermal images of the line and its surroundings;

[0021] Wide-angle camera: provides a wide range of monitoring angles, suitable for wide-area monitoring covering power transmission lines, capable of monitoring more areas in one capture and effectively identifying potential obstacles;

[0022] Zoom camera: It can provide close observation of specific areas of the power line, and is used to view key equipment at abnormal points, providing high-detail image analysis.

[0023] As a preferred solution of the present invention, the dynamic adjustment module includes:

[0024] Environmental perception and analysis: the built-in meteorological sensor monitors the current weather conditions in real time; the light sensor monitors the light intensity of the surrounding environment in real time; the dynamic adjustment module can monitor the signal strength of various radars to determine whether there is interference, obstruction and signal attenuation, so as to optimize the detection effect;

[0025] Automatic working mode adjustment: It can automatically select the most suitable sensor according to the current environmental conditions; automatically adjust the radar's transmission power and scanning frequency according to the signal quality; dynamically adjust the exposure time and sensitivity (ISO) parameters according to the lighting conditions to ensure accurate capture of image clarity and details;

[0026] Real-time adaptive adjustment: the module can perform multi-sensor data fusion and collaborative work based on real-time feedback from different sensors; through the built-in AI algorithm, it can learn and adapt to the sensor working mode in different environments, and automatically optimize based on historical data;

[0027] Sensor priority management can set priorities for different sensors to ensure that the most suitable sensor combination is automatically activated in a specific environment. When environmental conditions are ideal and the detection task is easier, the module will automatically reduce the power of some sensors and turn off unnecessary sensors to enter energy-saving mode to reduce system power consumption.

[0028] As a preferred solution of the present invention, the data processing module can integrate multi-source data in a complex environment to provide more accurate and comprehensive transmission line detection and monitoring results, including:

[0029] Multi-source data fusion integrates the data collected by radar and visual sensors in multiple dimensions, and realizes the deep integration of radar's long-distance high-precision detection and visual image and video data through the alignment of spatial coordinates and timestamps. The module ensures the synchronous collection of radar signals and visual images through the time synchronization mechanism, and aligns the data frames of different sensors to ensure that the fused data maintains temporal and spatial consistency.

[0030] Multi-level fusion processing directly fuses the raw data, extracts feature information from radar and visual data, performs feature-level matching and comprehensive judgment, and enhances detection accuracy; fuses independent detection results from different sensors, and makes final decisions through weight allocation and priority rules;

[0031] Blind spot compensation and misjudgment correction: Through data fusion, the visual detail capture capability and the radar long-range detection capability complement each other, avoiding the blind spot problem of a single sensor in complex terrain, obstacle occlusion, etc.; when a sensor misjudges, the data fusion module can perform cross-verification by analyzing the data of other sensors;

[0032] Real-time processing and feedback: The module has efficient processing capabilities and can receive and process massive data from radar and vision modules in real time, ensuring the system's rapid response in complex environments. The results of data fusion will be quickly fed back to the control center to achieve timely response and adjustment.

[0033] Anomaly detection and automatic alarm, through in-depth analysis of fused data, can more accurately detect abnormal conditions in transmission lines. When multiple sensors identify anomalies at the same time, the system will improve the confidence of anomaly identification through fused data; when an anomaly is detected, the data processing module will automatically generate an alarm message with detailed multi-source data analysis results.

[0034] As a preferred solution of the present invention, the data processing module improves the accuracy and robustness of sensor data through a data fusion algorithm, specifically including:

[0035] The Kalman filter is a recursive algorithm commonly used in linear systems. It uses continuous data collected by radar and visual sensors to predict and update uncertain target states. It can effectively remove noise signals in noisy environments and improve the accuracy of radar and visual data. The algorithm updates state estimates in a recursive form to ensure real-time performance. Every time new data arrives, the system generates the latest state estimate based on the previous estimate and the current sensor observation results.

[0036] Particle filter is a common method for processing nonlinear and non-Gaussian systems. It can process more complex and nonlinear systems. It describes the target state through probability distribution function and can combine the measurement data of multiple sensors to improve the judgment of complex scenes. Particle filter approximates the state distribution by generating a large number of possible target states (particles), and then updates the weights of these particles through measurement, and finally selects the most likely state as the final estimation result.

[0037] Deep learning technology, especially convolutional neural networks (CNN), has significant advantages in processing complex images and visual data. CNN can automatically extract high-level features from visual data, and combine the distance and speed information in radar data to build a more accurate object recognition and classification model. In the deep learning model, radar data can correspond to the pixels in the visual image to form a multimodal fusion network, and through joint training, the system's ability to identify obstacles, damage, and potential dangers can be improved.

[0038] As a preferred solution of the present invention, the multi-dimensional perception fusion module integrates various information such as sound, vibration, temperature, etc. to generate a comprehensive state analysis model. By fusing data of different dimensions, it can reduce false alarms or blind spots that may be caused by a single sensor; based on comparative analysis of historical data and current information, potential faults can be identified in a timely manner.

[0039] As a preferred solution of the present invention, the communication module performs signal transmission through a variety of communication technologies, specifically including:

[0040] Cellular networks (4G / 5G), modern power transmission line monitoring can use high-speed mobile networks to achieve low-latency, high-bandwidth data transmission;

[0041] LoRa (Low Power Wide Area Network), LoRa technology is suitable for long-distance, low-bandwidth communication needs of power transmission lines. It has the characteristics of low power consumption and wide coverage, and is suitable for remote areas or environments where cellular networks are difficult to use;

[0042] Satellite communication: In extreme conditions, satellite communication technology can be used to ensure that data can still be transmitted to the monitoring center in the absence of ground signals. Although satellite communication is more expensive, it can ensure that the connection is maintained in the most demanding environment;

[0043] Dedicated fiber optic network: For certain transmission lines with high security requirements, fiber optic networks deployed near transmission towers or lines can be used to ensure high-speed and secure data transmission.

[0044] As a preferred solution of the present invention, the intelligent analysis and early warning module uses supervised learning, unsupervised learning and reinforcement learning algorithms to learn the patterns of normal and abnormal states from a large amount of historical data. Once the real-time data deviates from these normal patterns, the system can identify potential faults or anomalies; the intelligent analysis and early warning module can dynamically monitor sensor data based on deep learning and statistical analysis models to discover gradually evolving abnormal situations;

[0045] Different levels of warning signals are generated according to the severity of the fault. The warning signals will be sent in real time to the remote monitoring center or the mobile device of the operation and maintenance personnel through the communication module to ensure that the warning can be conveyed in time.

[0046] The method and steps of the radar vision intelligent fusion system adapted to the power transmission scenario are as follows:

[0047] S1: Turn on the radar module and use a variety of radars to perform high-precision detection of power transmission lines over long distances. Through reflection and echo analysis of radar signals, the changes in the status of power transmission lines can be accurately captured.

[0048] S2: Synchronously start the visual module and use the cameras installed at key locations to capture images and videos; the cameras can also switch different viewing angles as needed to adapt to different scenarios;

[0049] S3: According to the changes in the external environment, the dynamic adjustment module will automatically adjust the working mode of each sensor in the system and optimize the performance of each sensor based on the data feedback from the sensor to ensure continuous monitoring in complex environments;

[0050] S4: The system sends radar, vision and other multi-modal data to the data processing module. The data processing module uses a fusion algorithm to fuse the data of multiple sensors. Through the fusion algorithm, the system can comprehensively analyze the output data of radar and vision sensors, remove noise, supplement the blind area of ​​a single sensor, and improve the robustness of the data and the accuracy of detection;

[0051] S5: The system integrates multiple multi-dimensional sensors such as sound, vibration, and temperature. Through the multi-dimensional data fusion algorithm, the collected data is deeply fused and processed. Multi-dimensional fusion effectively reduces false alarms and missed alarms, ensuring the comprehensiveness of status monitoring.

[0052] S6: The detected multi-dimensional data is transmitted to the remote monitoring center through the communication module. The communication module uses wireless transmission to transmit real-time data to the monitoring platform. The operation and maintenance personnel can obtain the status information of the transmission line in real time through the monitoring platform, and can remotely dispatch the maintenance team to conduct troubleshooting and maintenance;

[0053] S7: Based on artificial intelligence algorithms, the intelligent analysis and early warning module automatically analyzes the data collected by sensors. The system continuously improves the accuracy of identifying abnormal situations through self-learning capabilities. The AI ​​algorithm can identify multiple failure modes and classify potential risks. When the system detects abnormalities and failures, it automatically generates early warning signals to remind operation and maintenance personnel to take timely measures.

[0054] S8: The system has predictive maintenance capabilities. Through the accumulation of historical data and trend analysis, the system can predict the occurrence of potential failures.

[0055] The present invention has the following beneficial effects:

[0056] 1. Real-time monitoring and high-precision detection, using a variety of sensors to achieve real-time monitoring of transmission lines, improve the high precision and accuracy of detection, and capture line status changes in a timely manner.

[0057] 2. Data fusion and noise suppression: Through multimodal data fusion, the output data of different sensors are comprehensively analyzed to effectively remove noise and blind spots, improve the robustness of data, and enhance the reliability of monitoring.

[0058] 3. Dynamic adaptability: The dynamic adjustment module optimizes the working mode of the sensor according to changes in the external environment, ensuring continuous and effective monitoring in complex environments and adapting to different operating scenarios.

[0059] 4. Reduce false alarms and missed alarms. The multi-dimensional perception fusion module integrates different types of data to reduce false alarms or blind spots that may be caused by a single sensor, thereby improving the comprehensiveness and accuracy of status monitoring.

[0060] 5. Efficient data transmission, through a variety of communication technologies, ensures fast and reliable transmission of monitoring data in various environments, making it convenient for operation and maintenance personnel to obtain status information in real time.

[0061] 6. Intelligent analysis and fault warning: Use artificial intelligence algorithms to conduct intelligent analysis, identify abnormal patterns, and generate early warning signals in a timely manner to ensure that operation and maintenance personnel can quickly respond to potential faults and reduce the impact of faults on system operation.

[0062] 7. Predictive maintenance capability: The system can predict the occurrence of potential failures through the accumulation of historical data and trend analysis, which helps operation and maintenance personnel intervene in advance, reduce the probability of failures, and improve the overall reliability of transmission lines.

[0063] 8. Improve operation and maintenance efficiency. Real-time monitoring, intelligent early warning and predictive maintenance functions enable operation and maintenance personnel to dispatch resources and take measures more effectively, reduce operation and maintenance costs and improve work efficiency.

[0064] 9. Safety and stability: Through comprehensive monitoring and intelligent analysis, the safety and stability of transmission lines can be improved, the risk of accidents can be reduced, and the continuity and reliability of power supply can be ensured.

[0065] Overall, the Leishi intelligent fusion system, which is adapted to power transmission scenarios, has significantly improved the accuracy and response speed of transmission line monitoring through efficient data fusion and intelligent analysis, enhanced the safety and stability of the system, provided strong support for operation and maintenance personnel, and promoted the development of smart grids. BRIEF DESCRIPTION OF THE DRAWINGS

[0066] Figure 1 It is a schematic diagram of the system flow of the present invention. DETAILED DESCRIPTION

[0067] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0068] Example

[0069] See also Figure 1 , a lightning-vision intelligent fusion system that adapts to power transmission scenarios. The system includes:

[0070] The radar module uses multiple radars to perform long-distance, high-precision detection. It can penetrate adverse weather environments such as rain and fog and detect specific information about the transmission lines in real time. The radar module can penetrate complex environmental factors such as rain, fog, dust and insufficient light at night, and still provide clear transmission line information. The radar module monitors the transmission line's suspension height, swing amplitude, line sag, and wire deformation in real time to ensure comprehensive detection of the physical state.

[0071] The vision module uses multiple cameras to capture images and videos to identify obstacles, external damage, and potential safety hazards near the transmission lines. Through image and video capture technology, the vision module can detect obstacles near the lines (such as branches, buildings), external damage (such as fractures, wear and tear), and potential safety hazards (such as bird nests, climbing objects, etc.). Combined with the deep learning image recognition algorithm, the vision module can identify and classify different types of faults, and has excellent performance under complex conditions such as low light, rain and fog.

[0072] The dynamic adjustment module automatically adjusts the operating mode of the sensor according to the real-time changing environmental conditions such as weather, lighting, and radar signal strength. For example, the system increases the operating frequency of the radar and camera before a storm arrives based on weather forecast data. When lighting conditions are good, the camera exposure settings are adjusted to reduce energy consumption. When the radar signal strength drops, the backup sensor is automatically started. This dynamic adjustment mechanism ensures that the system always maintains efficient operation in various complex scenarios, reduces energy consumption, and extends the service life of the equipment.

[0073] The data processing module comprehensively processes radar and visual data through a data fusion algorithm to improve the accuracy and robustness of detection and avoid blind spots or misjudgments of a single sensor. The Kalman filter algorithm can denoise and predict sensor input data to improve the system's ability to judge the position and state of target objects. The particle filter is used to process data fusion in complex nonlinear and non-Gaussian environments and select the optimal solution. Combined with deep learning technology, the data processing module can extract features and efficiently fuse multi-source data to avoid blind spots or misjudgments of a single sensor and improve the ability to identify faults.

[0074] The multi-dimensional perception fusion module integrates sound sensors, vibration sensors, and temperature sensors into one to achieve all-round perception. Through the multi-dimensional data fusion algorithm, it integrates various types of perception information to provide more detailed status monitoring and reduce false alarms and missed alarms. Through the multi-dimensional data fusion algorithm, the module can integrate information from different sensors.

[0075] The communication module uses communication technology to transmit the detected data to the remote monitoring center in real time, which is convenient for remote operation and maintenance and timely response. The data can be sent to the remote monitoring center safely and reliably through encrypted transmission technology, which supports monitoring personnel to view the operating status of the transmission line in real time. The communication module also supports interconnection with the intelligent dispatching platform of the transmission system, ensuring that the operation and maintenance personnel can remotely control the equipment, monitor the line status and automatically generate fault warnings based on the data.

[0076] The intelligent analysis and early warning module analyzes real-time data based on artificial intelligence algorithms, automatically identifies faults and abnormal conditions, and generates early warning signals to alert operation and maintenance personnel. Through learning and training from historical data, the module can automatically identify faults and abnormal conditions of transmission lines, such as broken wires, insulation damage, and tower tilt, and generate accurate early warning signals. When potential risks are detected, the system will send early warning signals to operation and maintenance personnel in real time through the communication module to help them respond in a timely manner.

[0077] The radar-based intelligent fusion system adapted to power transmission scenarios is a comprehensive solution that integrates a variety of advanced sensor technologies and artificial intelligence algorithms. It can achieve all-round monitoring of transmission lines under complex environmental conditions. The system provides strong guarantees for the safe and stable operation of transmission lines through high-precision monitoring of radar, vision and multi-dimensional perception modules, combined with data fusion algorithms and intelligent analysis and early warning functions.

[0078] Specifically, the radar module is integrated with multiple types of radars, including:

[0079] Millimeter-wave radar: By emitting millimeter-wave signals, it has the characteristics of strong penetration and good anti-interference. It is suitable for long-distance target detection in complex weather environments such as rain and fog, and can monitor the key data of height, sag and swing of transmission lines in real time; millimeter-wave radar is suitable for long-distance target detection and can provide stable output data, playing an important role in monitoring key physical parameters of transmission lines (such as height and sag).

[0080] LiDAR: It has extremely high spatial resolution, which is suitable for accurately detecting the three-dimensional shape of transmission lines. It can generate high-precision 3D images to assist in identifying objects and potential obstacles near transmission lines. The high resolution of LiDAR makes it particularly suitable for detailed shape detection and obstacle identification, providing strong support for accurate analysis in complex terrain and environments.

[0081] Microwave radar: Suitable for large-scale monitoring. It can provide stable signal feedback in the global status detection of long-distance transmission lines. Microwave radar performs well in large-scale monitoring and can provide stable signal feedback to ensure that the global status of the line is within the controllable range.

[0082] Through the combination of millimeter-wave radar, lidar and microwave radar, the system can give full play to their respective advantages in different detection tasks to form a comprehensive and accurate monitoring system:

[0083] Long-distance precise detection, the long-range detection capabilities of millimeter-wave radar and microwave radar ensure blind-spot monitoring in long-distance transmission lines, which is especially suitable for detection under severe weather conditions; three-dimensional morphological modeling, lidar provides the system with detailed three-dimensional image information to help better identify and process the complex environment around the transmission lines; multi-level perception, by integrating different radar data, the system can achieve multi-level perception of the transmission lines, monitor the physical status in real time, identify obstacles and detect potential safety hazards.

[0084] The integration of radar modules ensures the safety of transmission lines in various complex environments and significantly improves the detection accuracy and robustness of the system.

[0085] Specifically, the vision module integrates multiple types of cameras, including:

[0086] HD Camera: High-resolution cameras are used to capture HD images and videos of transmission lines, helping to identify subtle damage on the line surface. HD images and videos help improve the efficiency of manual inspections, facilitate quick analysis and decision-making by operation and maintenance personnel, and reduce the frequency and cost of manual inspections.

[0087] Infrared camera: used to work at night or in low-light environments. By capturing thermal images of the lines and their surroundings, abnormal heat spots and hidden safety hazards can be discovered. The infrared camera's ability to work effectively in low-light environments greatly improves the timeliness of monitoring, allowing effective monitoring at night or in rainy weather.

[0088] Wide-angle camera: provides a wide range of monitoring angles, suitable for wide-area monitoring covering power transmission lines, capable of monitoring more areas in one capture and effectively identifying potential obstacles; wide-angle cameras reduce the number of cameras required and the overall cost of the system, while providing comprehensive monitoring of a wide range of environments, suitable for comprehensive monitoring of power transmission lines.

[0089] Zoom camera: It can provide close observation of specific areas of the transmission line, check key equipment at abnormal points, and provide high-detail image analysis; the zoom camera can capture subtle feature information in detail, such as joint status, insulator damage, etc., which helps further fault diagnosis and maintenance decision-making.

[0090] By combining high-definition cameras, infrared cameras, wide-angle cameras and zoom cameras, the vision module can provide multi-dimensional monitoring capabilities: all-round monitoring, the coverage capability of the wide-angle camera is combined with the detail capture of the high-definition camera to achieve all-round and meticulous line monitoring; environmental adaptability, the introduction of infrared cameras ensures continuous monitoring at night or in low-light environments, and enhances the adaptability and robustness of the system; precise analysis, the characteristics of the zoom camera enable operation and maintenance personnel to conduct in-depth analysis of specific problems, effectively improving the accuracy of fault diagnosis.

[0091] Specifically, the dynamic adjustment module includes:

[0092] Environmental perception and analysis: the built-in meteorological sensor monitors the current weather conditions in real time; the light sensor monitors the light intensity of the surrounding environment in real time; the dynamic adjustment module can monitor the signal strength of various radars to determine whether there is interference, obstruction and signal attenuation, so as to optimize the detection effect;

[0093] Automatic working mode adjustment: It can automatically select the most suitable sensor according to the current environmental conditions; automatically adjust the radar's transmission power and scanning frequency according to the signal quality; dynamically adjust the exposure time and sensitivity (ISO) parameters according to the lighting conditions to ensure accurate capture of image clarity and details;

[0094] Real-time adaptive adjustment: the module can perform multi-sensor data fusion and collaborative work based on real-time feedback from different sensors; through the built-in AI algorithm, it can learn and adapt to the sensor working mode in different environments, and automatically optimize based on historical data;

[0095] Sensor priority management can set priorities for different sensors to ensure that the most suitable sensor combination is automatically activated in a specific environment. When environmental conditions are ideal and the detection task is easier, the module will automatically reduce the power of some sensors and turn off unnecessary sensors to enter energy-saving mode to reduce system power consumption.

[0096] Through the combination of environmental perception and analysis, automatic working mode adjustment, real-time adaptive adjustment and sensor priority management, the dynamic adjustment module ensures that the Leishi intelligent fusion system can automatically optimize the working status in a changing environment. Intelligent management not only improves the accuracy and reliability of monitoring, but also enhances the system's adaptability under various climatic conditions, making it an ideal solution for modern transmission line monitoring.

[0097] Specifically, the data processing module can integrate multi-source data in complex environments to provide more accurate and comprehensive transmission line detection and monitoring results, including:

[0098] Multi-source data fusion integrates the data collected by radar and visual sensors in multiple dimensions, and realizes the deep combination of radar's long-distance high-precision detection and visual image and video data through the alignment of spatial coordinates and timestamps. This integration enables the advantages of the two data types to complement each other and improves the overall monitoring capability. The module ensures the synchronous collection of radar signals and visual images through the time synchronization mechanism, and aligns the data frames of different sensors to ensure that the fused data maintains temporal and spatial consistency. This mechanism makes the relationship between data clearer during the analysis and decision-making process, avoiding misjudgment caused by time delays.

[0099] Multi-level fusion processing directly fuses the raw data, extracts feature information from radar and visual data, and performs feature-level matching and comprehensive judgment. This deep fusion not only enhances the accuracy of detection, but also improves the intelligence level of the system. It fuses the independent detection results of different sensors, makes the final decision through weight allocation and priority rules, and gives priority to high-credibility data sources, making the detection results more reliable.

[0100] Blind spot compensation and misjudgment correction: Through data fusion, the visual detail capture capability and the radar's long-range detection capability complement each other, avoiding the blind spot problem of a single sensor in complex terrain, obstacle obstruction, etc. For example, in areas that are difficult for some sensors to detect, the high-precision data of another sensor can fill these blind spots; when a sensor makes a misjudgment, the data fusion module can cross-validate by analyzing the data of other sensors to confirm the accuracy of the results. This redundant design improves the robustness of the system.

[0101] Real-time processing and feedback. The module has efficient processing capabilities and can receive and process massive data from radar and vision modules in real time, ensuring the system's rapid response in complex environments. The results of data fusion will be quickly fed back to the control center to achieve timely response and adjustment. The real-time feedback mechanism is the key to dealing with emergencies.

[0102] Anomaly detection and automatic alarm, through in-depth analysis of fused data, can more accurately detect abnormal conditions in transmission lines. When multiple sensors identify anomalies at the same time, the system will improve the confidence of anomaly identification through fused data. The synergy of multiple sensors greatly improves the sensitivity of anomaly detection. When an anomaly is detected, the data processing module will automatically generate an alarm message, accompanied by detailed multi-source data analysis results. This automated function can quickly notify operation and maintenance personnel, reduce human delays, and ensure timely handling of potential risks.

[0103] The data processing module integrates multi-source data and makes full use of the characteristics of radar and visual sensors to improve the detection accuracy of transmission lines; data fusion can effectively compensate for the blind spots and misjudgments of a single sensor, and improve the robustness and reliability of the overall system; powerful data processing capabilities and fast feedback mechanisms ensure that the system can respond in real time in complex environments; through deep learning and data analysis, the module can continuously optimize the working mode and improve the intelligence level of the system; it automatically detects anomalies and generates alarms, which improves operation and maintenance efficiency and safety and reduces the need for human intervention.

[0104] Specifically, the data processing module improves the accuracy and robustness of sensor data through data fusion algorithms, including:

[0105] The Kalman filter is a recursive algorithm commonly used in linear systems. It uses continuous data collected by radar and visual sensors to predict and update uncertain target states. It can effectively remove noise signals in noisy environments and improve the accuracy of radar and visual data. The algorithm updates state estimates in a recursive form to ensure real-time performance. Every time new data arrives, the system generates the latest state estimate based on the previous estimate and the current sensor observation results.

[0106] Particle filter is a common method for processing nonlinear and non-Gaussian systems. It can process more complex and nonlinear systems. It describes the target state through probability distribution function and can combine the measurement data of multiple sensors to improve the judgment of complex scenes. Particle filter approximates the state distribution by generating a large number of possible target states (particles), and then updates the weights of these particles through measurement, and finally selects the most likely state as the final estimation result.

[0107] Deep learning technology, especially convolutional neural networks (CNN), has significant advantages in processing complex images and visual data. CNN can automatically extract high-level features from visual data, and combine the distance and speed information in radar data to build a more accurate object recognition and classification model. In the deep learning model, radar data can correspond to the pixels in the visual image to form a multimodal fusion network, and through joint training, the system's ability to identify obstacles, damage, and potential dangers can be improved.

[0108] By using Kalman filters, particle filters and deep learning techniques, the data processing module can effectively improve the accuracy and robustness of sensor data in complex environments. The combination of these algorithms enables the system to achieve:

[0109] Precise monitoring: The use of Kalman filters and particle filters improves the reliability and real-time nature of data, ensuring accurate monitoring of the status of transmission lines; adapting to complex environments: The ability of particle filters to adapt to nonlinear and complex scenarios enables the system to maintain good performance under various environmental conditions; intelligent identification: The application of deep learning technology improves the system's ability to automatically identify obstacles and potential dangers, enhancing the intelligence level of safety monitoring.

[0110] Overall, the use of these fusion algorithms has greatly improved the performance of the transmission line monitoring system and provided more accurate and reliable data support for operation and maintenance.

[0111] Specifically, the multi-dimensional perception fusion module integrates various information such as sound, vibration, and temperature to generate a comprehensive state analysis model. By fusing data from different dimensions, it can reduce false alarms or blind spots that may be caused by a single sensor. It can also conduct comparative analysis based on historical data and current information to identify potential faults in a timely manner.

[0112] The multi-dimensional perception fusion module integrates various information such as sound, vibration, temperature, etc. to form an efficient comprehensive status analysis system. The advantages of this system are:

[0113] Comprehensive monitoring, through the integration of multi-dimensional data, provides a comprehensive monitoring perspective, effectively supplementing the shortcomings of a single sensor; accurate identification, based on the comparative analysis of historical data and real-time data, timely identifies potential faults, reduces false alarms, and improves the reliability of detection; early warning capability, the intelligent early warning mechanism improves the response speed of operation and maintenance, so that potential problems can be dealt with in a timely manner, thereby ensuring the safe and stable operation of the transmission line.

[0114] Overall, the multi-dimensional perception fusion module significantly enhances the intelligence level of the transmission line monitoring system and promotes efficient and accurate operation and maintenance management.

[0115] Specifically, the communication module transmits signals through a variety of communication technologies, including:

[0116] Cellular network (4G / 5G), modern transmission line monitoring can use high-speed mobile networks to achieve low-latency, high-bandwidth data transmission; cellular networks support the mobility of equipment, and can flexibly deploy sensors and monitoring equipment in the monitoring area, enhancing the adaptability of the system.

[0117] LoRa (Low Power Wide Area Network), LoRa technology is suitable for the long-distance, low-bandwidth communication needs of power transmission lines. It has the characteristics of low power consumption and wide coverage, and is suitable for remote areas or environments where cellular networks are difficult to use. Due to the low power consumption characteristics of LoRa, it can reduce the maintenance frequency and operating costs of equipment, extend battery life, and is suitable for long-term unattended monitoring environments.

[0118] Satellite communication: Under extreme conditions, satellite communication technology can be used to ensure that data can still be transmitted to the monitoring center in the absence of ground signals. Although satellite communication is more expensive, it can ensure connectivity in the most demanding environments. Satellite communication can cover the globe, providing a good solution for global transmission line monitoring.

[0119] Dedicated fiber optic networks: For certain transmission lines with high security requirements, fiber optic networks deployed near transmission towers or lines can be used to ensure high-speed and secure data transmission. The physical properties of fiber optic networks make them less vulnerable to hacker attacks, which enhances data security and is very suitable for monitoring important infrastructure.

[0120] The communication module uses a variety of communication technologies to ensure that the transmission line monitoring system can achieve efficient and reliable data transmission in various environments. The main advantages include:

[0121] Diversity and flexibility: the ability to select appropriate communication methods based on different environments and needs to enhance the adaptability of the system; efficiency and reliability: the combination of cellular networks and satellite communications and other technologies ensures data transmission speed and stability under complex conditions; security: the deployment of fiber optic networks enhances data security and is suitable for application scenarios with high security requirements.

[0122] Through the integration of these technologies, the communication module not only improves the real-time and accuracy of transmission line monitoring, but also provides a solid foundation for remote operation and maintenance and fault response.

[0123] Specifically, the intelligent analysis and early warning module uses supervised learning (by using labeled historical data, the supervised learning algorithm can identify the pattern between normal and abnormal states), unsupervised learning (in the absence of labeled data, the unsupervised learning algorithm can discover the potential structure and abnormal pattern in the data, which is particularly suitable for detecting new types of faults that have never been seen) and reinforcement learning (using reinforcement learning algorithms, the system can continuously learn and optimize decision-making strategies in a dynamic environment, making it more flexible in dealing with complex situations) and other algorithms to learn the patterns of normal and abnormal states from a large amount of historical data. Once the real-time data deviates from these normal patterns, the system can identify potential faults or anomalies; the intelligent analysis and early warning module can dynamically monitor sensor data based on deep learning and statistical analysis models to discover gradually evolving abnormal situations;

[0124] Different levels of warning signals are generated according to the severity of the fault. The warning signals will be sent in real time to the remote monitoring center or the mobile device of the operation and maintenance personnel through the communication module to ensure that the warning can be conveyed in time.

[0125] The comprehensive advantages of the intelligent analysis and early warning module are reflected in the following aspects:

[0126] Efficient fault identification: through deep learning and a variety of machine learning technologies, the module can quickly and accurately identify potential faults, improving the sensitivity and accuracy of detection; real-time response capability, dynamic monitoring and real-time early warning mechanism ensure that operation and maintenance personnel are quickly notified when problems occur, reducing the impact of faults on system operation; flexible response to different situations, whether it is a new fault or a traditional problem, the module can provide flexible solutions through self-learning and optimization algorithms; decision support, the data analysis and early warning signals provided can provide strong support for operation and maintenance decisions, optimize maintenance strategies, and reduce unnecessary maintenance costs.

[0127] The implementation of the intelligent analysis and early warning module provides a strong guarantee for the safe and stable operation of the transmission line monitoring system, making preventive maintenance and fault handling more efficient.

[0128] The method and steps of the radar vision intelligent fusion system adapted to the power transmission scenario are as follows:

[0129] S1: Turn on the radar module and use a variety of radars to perform high-precision detection of power transmission lines over long distances. Through reflection and echo analysis of radar signals, the changes in the status of power transmission lines can be accurately captured.

[0130] S2: Synchronously start the visual module and use the cameras installed at key locations to capture images and videos; the cameras can also switch different viewing angles as needed to adapt to different scenarios;

[0131] S3: According to the changes in the external environment, the dynamic adjustment module will automatically adjust the working mode of each sensor in the system and optimize the performance of each sensor based on the data feedback from the sensor to ensure continuous monitoring in complex environments;

[0132] S4: The system sends radar, vision and other multi-modal data to the data processing module. The data processing module uses a fusion algorithm to fuse the data of multiple sensors. Through the fusion algorithm, the system can comprehensively analyze the output data of radar and vision sensors, remove noise, supplement the blind area of ​​a single sensor, and improve the robustness of the data and the accuracy of detection;

[0133] S5: The system integrates multiple multi-dimensional sensors such as sound, vibration, and temperature. Through the multi-dimensional data fusion algorithm, the collected data is deeply fused and processed. Multi-dimensional fusion effectively reduces false alarms and missed alarms, ensuring the comprehensiveness of status monitoring.

[0134] S6: The detected multi-dimensional data is transmitted to the remote monitoring center through the communication module. The communication module uses wireless transmission to transmit real-time data to the monitoring platform. The operation and maintenance personnel can obtain the status information of the transmission line in real time through the monitoring platform, and can remotely dispatch the maintenance team to conduct troubleshooting and maintenance;

[0135] S7: Based on artificial intelligence algorithms, the intelligent analysis and early warning module automatically analyzes the data collected by sensors. The system continuously improves the accuracy of identifying abnormal situations through self-learning capabilities. The AI ​​algorithm can identify multiple failure modes and classify potential risks. When the system detects abnormalities and failures, it automatically generates early warning signals to remind operation and maintenance personnel to take timely measures.

[0136] S8: The system has a predictive maintenance function. Through the accumulation of historical data and trend analysis, the system can predict the occurrence of potential failures. This function helps operation and maintenance personnel intervene in advance, reduce the probability of failures, and improve the overall reliability of the transmission line.

[0137] The radar intelligent fusion system significantly improves the ability to monitor the status of transmission lines through the integration of multiple sensors, real-time data processing and intelligent analysis. Through data fusion and artificial intelligence technology, the system can not only efficiently identify and warn of potential faults, but also provide predictive maintenance functions to ensure the safe and stable operation of transmission lines.

[0138] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0139] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the technical principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. The intelligent fusion system of lightning vision adapted to power transmission scenarios is characterized by: The system includes: Radar module, which uses multiple radars for long-distance high-precision detection, can penetrate adverse weather conditions such as rain and fog, and detect specific information of transmission lines in real time; Vision module, which uses multiple cameras to capture images and videos to identify obstacles, external damage, and potential safety hazards near the transmission line; Dynamic adjustment module, automatically adjusts the sensor's operating mode according to real-time changing environmental conditions such as weather, light, and radar signal strength; The data processing module processes radar and visual data through a data fusion algorithm to improve the accuracy and robustness of detection and avoid blind spots or misjudgments of a single sensor. The multi-dimensional perception fusion module integrates sound sensors, vibration sensors, and temperature sensors to achieve all-round perception. Through the multi-dimensional data fusion algorithm, it integrates various types of perception information to provide more detailed status monitoring and reduce false alarms and missed alarms. The communication module uses communication technology to transmit the detected data to the remote monitoring center in real time, facilitating remote operation and maintenance and timely response; The intelligent analysis and early warning module analyzes real-time data based on artificial intelligence algorithms, automatically identifies faults and abnormal conditions, and generates early warning signals to alert operation and maintenance personnel.

2. The radar vision intelligent fusion system adapted to power transmission scenarios according to claim 1 is characterized in that: The radar module is integrated with multiple types of radars, including: Millimeter-wave radar: It emits millimeter-wave signals with strong penetration and good anti-interference properties. It is suitable for long-distance target detection in complex weather environments such as rain and fog. It can monitor key data such as the height, sag and swing of transmission lines in real time. LiDAR: It has extremely high spatial resolution, suitable for accurately detecting the three-dimensional shape of power transmission lines. It can generate high-precision 3D images to assist in identifying objects and potential obstacles near power transmission lines. Microwave radar: Suitable for large-scale monitoring and can provide stable signal feedback in the global status detection of long-distance transmission lines.

3. The radar vision intelligent fusion system adapted to power transmission scenarios according to claim 1 is characterized in that: The visual module integrates multiple types of cameras, including: HD Camera: High-resolution cameras are used to capture high-definition images and videos of transmission lines, helping to identify subtle damage on the line surface; Infrared camera: used to work at night or in low-light environments, and to detect abnormal heat spots and hidden safety hazards by capturing thermal images of the line and its surroundings; Wide-angle camera: provides a wide range of monitoring angles, suitable for wide-area monitoring covering power transmission lines, capable of monitoring more areas in one capture and effectively identifying potential obstacles; Zoom camera: It can provide close observation of specific areas of the power line, and is used to view key equipment at abnormal points, providing high-detail image analysis.

4. The radar vision intelligent fusion system adapted to power transmission scenarios according to claim 1 is characterized in that: The dynamic adjustment module comprises: Environmental perception and analysis: the built-in meteorological sensor monitors the current weather conditions in real time; the light sensor monitors the light intensity of the surrounding environment in real time; the dynamic adjustment module can monitor the signal strength of various radars to determine whether there is interference, obstruction and signal attenuation, so as to optimize the detection effect; Automatic working mode adjustment: It can automatically select the most suitable sensor according to the current environmental conditions; automatically adjust the radar's transmission power and scanning frequency according to the signal quality; dynamically adjust the exposure time and sensitivity (ISO) parameters according to the lighting conditions to ensure accurate capture of image clarity and details; Real-time adaptive adjustment: the module can perform multi-sensor data fusion and collaborative work based on real-time feedback from different sensors; through the built-in AI algorithm, it can learn and adapt to the sensor working mode in different environments, and automatically optimize based on historical data; Sensor priority management can set priorities for different sensors to ensure that the most suitable sensor combination is automatically activated in a specific environment. When environmental conditions are ideal and the detection task is easier, the module will automatically reduce the power of some sensors and turn off unnecessary sensors to enter energy-saving mode to reduce system power consumption.

5. The radar vision intelligent fusion system adapted to power transmission scenarios according to claim 1 is characterized in that: The data processing module can integrate multi-source data in complex environments to provide more accurate and comprehensive transmission line detection and monitoring results, including: Multi-source data fusion integrates the data collected by radar and visual sensors in multiple dimensions, and realizes the deep integration of radar's long-distance high-precision detection and visual image and video data through the alignment of spatial coordinates and timestamps. The module ensures the synchronous collection of radar signals and visual images through the time synchronization mechanism, and aligns the data frames of different sensors to ensure that the fused data maintains temporal and spatial consistency. Multi-level fusion processing directly fuses the raw data, extracts feature information from radar and visual data, performs feature-level matching and comprehensive judgment, and enhances detection accuracy; fuses independent detection results from different sensors, and makes final decisions through weight allocation and priority rules; Blind spot compensation and misjudgment correction: Through data fusion, the visual detail capture capability and the radar long-range detection capability complement each other, avoiding the blind spot problem of a single sensor in complex terrain, obstacle occlusion, etc.; when a sensor misjudges, the data fusion module can perform cross-verification by analyzing the data of other sensors; Real-time processing and feedback: The module has efficient processing capabilities and can receive and process massive data from radar and vision modules in real time, ensuring the system's rapid response in complex environments. The results of data fusion will be quickly fed back to the control center to achieve timely response and adjustment. Anomaly detection and automatic alarm, through in-depth analysis of fused data, can more accurately detect abnormal conditions in transmission lines. When multiple sensors identify anomalies at the same time, the system will improve the confidence of anomaly identification through fused data; when an anomaly is detected, the data processing module will automatically generate an alarm message with detailed multi-source data analysis results.

6. The radar vision intelligent fusion system adapted to power transmission scenarios according to claim 5 is characterized in that: The data processing module improves the accuracy and robustness of sensor data through a data fusion algorithm, specifically including: The Kalman filter is a recursive algorithm commonly used in linear systems. It uses continuous data collected by radar and visual sensors to predict and update uncertain target states. It can effectively remove noise signals in noisy environments and improve the accuracy of radar and visual data. The algorithm updates state estimates in a recursive form to ensure real-time performance. Every time new data arrives, the system generates the latest state estimate based on the previous estimate and the current sensor observation results. Particle filter is a common method for processing nonlinear and non-Gaussian systems. It can process more complex and nonlinear systems. It describes the target state through probability distribution function and can combine the measurement data of multiple sensors to improve the judgment of complex scenes. Particle filter approximates the state distribution by generating a large number of possible target states (particles), and then updates the weights of these particles through measurement, and finally selects the most likely state as the final estimation result. Deep learning technology, especially convolutional neural networks (CNN), has significant advantages in processing complex images and visual data. CNN can automatically extract high-level features from visual data, and combine the distance and speed information in radar data to build a more accurate object recognition and classification model. In the deep learning model, radar data can correspond to the pixels in the visual image to form a multimodal fusion network, and through joint training, the system's ability to identify obstacles, damage, and potential dangers can be improved.

7. The radar vision intelligent fusion system adapted to power transmission scenarios according to claim 1 is characterized in that: The multi-dimensional perception fusion module integrates various information such as sound, vibration, temperature, etc. to generate a comprehensive state analysis model. By fusing data of different dimensions, it can reduce false alarms or blind spots that may be caused by a single sensor; based on comparative analysis of historical data and current information, potential faults can be identified in a timely manner.

8. The radar vision intelligent fusion system adapted to power transmission scenarios according to claim 1 is characterized in that: The communication module transmits signals through a variety of communication technologies, including: Cellular networks (4G / 5G), modern power transmission line monitoring can use high-speed mobile networks to achieve low-latency, high-bandwidth data transmission; LoRa (Low Power Wide Area Network), LoRa technology is suitable for long-distance, low-bandwidth communication needs of power transmission lines. It has the characteristics of low power consumption and wide coverage, and is suitable for remote areas or environments where cellular networks are difficult to use; Satellite communication: In extreme conditions, satellite communication technology can be used to ensure that data can still be transmitted to the monitoring center in the absence of ground signals. Although satellite communication is more expensive, it can ensure that the connection is maintained in the most demanding environment; Dedicated fiber optic network: For certain transmission lines with high security requirements, fiber optic networks deployed near transmission towers or lines can be used to ensure high-speed and secure data transmission.

9. The radar vision intelligent fusion system adapted to power transmission scenarios according to claim 1 is characterized in that: The intelligent analysis and early warning module uses supervised learning, unsupervised learning, and reinforcement learning algorithms to learn the patterns of normal and abnormal states from a large amount of historical data. Once the real-time data deviates from these normal patterns, the system can identify potential faults or anomalies; The intelligent analysis and early warning module can dynamically monitor sensor data based on deep learning and statistical analysis models to detect gradually evolving abnormal situations; Different levels of warning signals are generated according to the severity of the fault. The warning signals will be sent in real time to the remote monitoring center or the mobile device of the operation and maintenance personnel through the communication module to ensure that the warning can be conveyed in time.

10. The radar vision intelligent fusion system adapted to power transmission scenarios according to claims 1-9 is characterized in that: The steps are as follows: S1: Turn on the radar module and use a variety of radars to perform high-precision detection of power transmission lines over long distances. Through reflection and echo analysis of radar signals, the changes in the status of power transmission lines can be accurately captured. S2: Synchronously start the visual module and use the cameras installed at key locations to capture images and videos; the cameras can also switch different viewing angles as needed to adapt to different scenarios; S3: According to the changes in the external environment, the dynamic adjustment module will automatically adjust the working mode of each sensor in the system and optimize the performance of each sensor based on the data feedback from the sensor to ensure continuous monitoring in complex environments; S4: The system sends radar, vision and other multi-modal data to the data processing module. The data processing module uses a fusion algorithm to fuse the data of multiple sensors. Through the fusion algorithm, the system can comprehensively analyze the output data of radar and vision sensors, remove noise, supplement the blind area of ​​a single sensor, and improve the robustness of the data and the accuracy of detection; S5: The system integrates multiple multi-dimensional sensors such as sound, vibration, and temperature. Through the multi-dimensional data fusion algorithm, the collected data is deeply fused and processed. Multi-dimensional fusion effectively reduces false alarms and missed alarms, ensuring the comprehensiveness of status monitoring. S6: The detected multi-dimensional data is transmitted to the remote monitoring center through the communication module. The communication module uses wireless transmission to transmit real-time data to the monitoring platform. The operation and maintenance personnel can obtain the status information of the transmission line in real time through the monitoring platform, and can remotely dispatch the maintenance team to conduct troubleshooting and maintenance; S7: Based on artificial intelligence algorithms, the intelligent analysis and early warning module automatically analyzes the data collected by sensors. The system continuously improves the accuracy of identifying abnormal situations through self-learning capabilities. The AI ​​algorithm can identify multiple failure modes and classify potential risks. When the system detects abnormalities and failures, it automatically generates early warning signals to remind operation and maintenance personnel to take timely measures. S8: The system has predictive maintenance capabilities. Through the accumulation of historical data and trend analysis, the system can predict the occurrence of potential failures.

Images