Power transmission line icing intelligent monitoring and cooperative deicing system and method

Abstract

The invention discloses a power transmission line icing intelligent monitoring and cooperative deicing system, and belongs to the technical field of power transmission and distribution of power systems, and the system comprises a monitoring module which is used for collecting line icing environment parameters and image data in real time; the prediction module is used for predicting an icing growth process and an icing type in a future set time period by fusing numerical weather forecast, a geographic information system and real-time image information on the basis of the monitoring data; the deicing execution module comprises a fixed deicing device and a movable deicing device, and the fixed deicing device comprises an ultrasonic deicing device and a fixed direct current deicing device which are integrated on a power transmission line windproof hammer; and the cooperative control module is used for generating a cooperative deicing strategy based on the icing prediction result and the real-time monitoring data and controlling the deicing execution module to execute deicing. Through multi-line deicing sequence optimization, energy consumption dynamic distribution and mobile device path planning, the optimization of the deicing equipment is realized, and the deicing efficiency is greatly improved.

Description

Technical Field

[0001] This invention belongs to the field of power system transmission and distribution technology, specifically relating to an intelligent monitoring and collaborative de-icing system and method for icing on transmission lines. Background Technology

[0002] In harsh winter conditions, power transmission lines are highly susceptible to icing. Icing increases the mechanical load on lines, leading to increased conductor sag and even serious accidents such as line breaks and tower collapses, threatening the safe and stable operation of the power grid. Currently, traditional de-icing methods mainly include manual de-icing, mechanical de-icing, and various ice-melting devices. Manual de-icing is inefficient, and personnel working at heights in harsh environments face extremely high risks. While technologies such as DC ice melting have some effectiveness, the equipment is usually large and complex, requiring specialized converters, control devices, and high-capacity power supplies. The procurement, transportation, installation, and commissioning costs are high, and there are strict requirements for the installation site. Furthermore, existing fixed ice melting devices suffer from slow response, high energy consumption, and a lack of precise control, making it difficult to adapt to the dynamic changes in icing caused by drastic microclimate changes. Automated ice melting methods for single lines lack effective coordinated control strategies when multiple lines are simultaneously iced, resulting in low overall ice melting efficiency, long processing times, and high energy consumption. Patent publication number CN114389223B discloses an intelligent de-icing system and method for overhead transmission lines. The de-icing system includes a transmission line icing monitoring platform, an intelligent shockwave de-icing module, and a de-icing efficiency calculation module. The transmission line icing monitoring platform includes a laser sensor, a laser reflector, a data receiver, and a data processor. The intelligent shockwave de-icing module includes a shockwave de-icing device with shockwave collection and emission ports, connected to the lowest point of the transmission line requiring de-icing. The de-icing efficiency calculation module calculates the de-icing efficiency based on the ice thickness before and after de-icing monitored by the transmission line icing monitoring platform. This achieves automatic monitoring of transmission line ice thickness, improving the efficiency of ice thickness detection. The shockwave de-icing device can release shock waves more evenly and can be easily removed by a drone. However, the de-icing technology described in this patent cannot deeply integrate comprehensive environmental monitoring, accurate icing prediction, and the coordinated control and intelligent decision-making of multiple de-icing devices, such as fixed and mobile devices. Especially when facing complex scenarios involving large areas, multiple lines, and varying icing conditions, achieving full-chain automation and optimization of monitoring, prediction, decision-making, and execution remains a pressing technical challenge. Therefore, there is an urgent need for a comprehensive solution capable of intelligent monitoring, accurate prediction, and collaborative de-icing to improve de-icing efficiency, reduce energy consumption and operating costs, and enhance the power grid's ability to cope with snow and ice disasters. Summary of the Invention

[0003] The purpose of this invention is to overcome the shortcomings of the prior art and provide an intelligent monitoring and collaborative de-icing system and method for icing of power transmission lines, thereby solving the technical problems mentioned in the background art.

[0004] The objective of this invention is achieved as follows: A smart monitoring and collaborative de-icing system for transmission lines, comprising: a monitoring module, including a multimodal sensor network installed on the transmission line for real-time acquisition of icing environmental parameters and image data, wherein the multimodal sensor network includes temperature sensors, humidity sensors, wind speed sensors, image sensors, and stress sensors; a prediction module, connected to the monitoring module, for predicting the icing growth process and icing type within a future set time period based on the monitoring data by fusing numerical weather forecasts, geographic information systems, and real-time image information; a de-icing execution module, including a fixed de-icing device and a mobile de-icing device, wherein the fixed de-icing device includes an ultrasonic de-icing device integrated on a wind hammer of the transmission line and a fixed DC de-icing device, and the mobile de-icing device includes a flame-throwing drone and a mobile DC de-icing vehicle; and a collaborative control module, connected to the prediction module and the de-icing execution module respectively, for generating a collaborative de-icing strategy based on the icing prediction results and real-time monitoring data and controlling the de-icing execution module to perform de-icing.

[0005] A multimodal sensor network comprehensively perceives the icing environment, predictive models proactively assess icing trends, and a collaborative control center intelligently schedules fixed and mobile de-icing resources to achieve precise and efficient collaborative de-icing. During implementation, various sensors deployed along the line (temperature, humidity, wind speed, image, stress) continuously collect data. The prediction module integrates real-time monitoring data, macro-weather forecasts, and geographic information to output a prediction of icing thickness growth and icing type (e.g., rime, frost) for a future period. The collaborative control module generates the optimal de-icing strategy based on the predicted icing severity, type, and real-time risks (e.g., excessive stress). For example, it commands the activation of fixed ultrasonic devices in lightly iced areas while simultaneously dispatching mobile de-icing vehicles to heavily iced areas. Fixed de-icing devices (ultrasonic, DC de-icing) and mobile de-icing devices (drones, de-icing vehicles) receive instructions and perform de-icing operations at designated times and locations. By deeply integrating real-time sensor data, numerical weather forecasts, and geographic information systems, the accuracy of predictions is improved, enabling intelligent, proactive, and collaborative management of power transmission line icing. This significantly improves de-icing efficiency and accuracy, reduces the risk of line breaks and tower collapses caused by icing, and saves manpower and energy costs by optimizing resource scheduling.

[0006] Furthermore, the monitoring module also includes: a micro-topographic icing observation unit, set up in typical micro-topographic areas such as high mountain watersheds, water bodies, canyons, and mountain passes, used to establish a micro-topographic icing model; and an optical fiber sensing unit, which uses optical fibers in the transmission line ground wire composite optical cable to achieve continuous monitoring of the entire line's status and establish a mapping relationship between optical fiber sensing monitoring and the line's physical status and environmental parameters. Based on the basic sensor network, enhanced monitoring is conducted on micro-topography, a key factor in icing, and existing optical fiber composite overhead ground wire (OPGW) is used to achieve continuous, distributed status perception of the entire line corridor. The optical fibers in the OPGW are used as distributed sensors, and changes in demodulated light signals (such as backscattered light) are used to invert parameters such as line vibration, strain, and temperature. Monitoring points are strategically deployed in specific geographical units to establish a more accurate micro-meteorological icing model, achieving low-cost, full-coverage line status monitoring without the need for extensive additional sensor deployment. This improves the spatial resolution and comprehensiveness of the monitoring data, making the icing prediction model more accurate, especially effective in predicting localized severe icing caused by micro-topography.

[0007] Furthermore, the prediction module includes: a dual-branch output model with a regression branch for ice growth rate and a classification branch for ice type identification; and a feature fusion unit for task-aware weighted fusion of image features, geographic features, and meteorological features. The regression branch outputs the ice growth rate over a predetermined future time period, while the classification branch outputs the probability distribution of ice type. During model execution, the input meteorological, image, and geographic features are adaptively weighted and fused. The regression branch then outputs the growth rate in millimeters per hour, and the classification branch outputs the probability of "rime, mixed rime, frost, etc." The prediction results include not only ice thickness but also ice type information crucial for de-icing strategy formulation (e.g., rime has strong adhesion, requiring higher power), making subsequent de-icing decisions more targeted.

[0008] Furthermore, the ultrasonic de-icing device includes: a wind hammer counterweight system for realizing the original wind hammer's physical wind-resistant resonance function; an inductive energy storage and charging system, consisting of a high-voltage current transformer (CT) power-taking module, a rectifier charging module, and a battery management system, for obtaining electrical energy in an insulated, non-contact manner; a temperature and humidity sensing system for activating an icing warning signal when natural icing conditions are met; and an ultrasonic generation system for continuous or intermittent operation based on the working instructions issued by the temperature and humidity sensing system. In implementation, the high-voltage CT draws power non-contactly from the transmission line and charges the battery. The temperature and humidity sensor determines whether the environment meets icing conditions. Once met, the ultrasonic generation system activates, converting electrical energy into mechanical vibrations that act on the wind hammer and the line. It automatically activates when icing conditions are met, using high-frequency ultrasonic vibrations to generate stress at the ice-conductor interface, causing the ice layer to break and fall off. This cleverly utilizes the space and mass of the existing wind hammer component on the line, achieving functional integration without altering the main structure of the line, making it easy to promote.

[0009] Furthermore, the collaborative control module includes: a data fusion unit for fusing and analyzing temperature distribution data and wind speed and direction data to determine the risk level of conductor torsional load; a region division unit for dividing the transmission line into several de-icing device control areas based on the conductor ice adhesion change curve; a power allocation unit for allocating power to each de-icing device control area to obtain the first control strategy for the corresponding area; and a path planning unit for planning the operating path of the mobile de-icing device using an optimization algorithm based on the region ranking results and the first control strategy. For different sections of a single line, refined risk assessment and region division are performed to achieve differentiated power allocation and resource scheduling. During implementation, temperature and wind speed data are fused to determine which sections have a high risk of torsional load due to "wind load and asymmetric icing". For these high-risk sections, their ice adhesion changes are further analyzed, and they are divided into different control areas. The optimal de-icing power (first strategy) is calculated for each area, and the optimal travel path (second strategy) is planned for the mobile device within the area. The control objective is not only to remove ice, but also to prevent more dangerous situations such as conductor twisting and jumping. It achieves progressively refined control from macro-risk identification to micro-resource scheduling, accurately deploying de-icing resources to the most dangerous and needed sections, thereby improving de-icing efficiency while ensuring safety.

[0010] Furthermore, the collaborative control module also includes: a multi-line coordination unit, used to acquire real-time environmental data and ice thickness data of multiple de-icing lines based on the line sensor network; a de-icing index calculation unit, used to calculate the line de-icing index and determine the real-time icing status based on the real-time environmental data and ice thickness data; a de-icing sequence optimization unit, used to calculate a comprehensive evaluation index and determine the line de-icing sequence based on the moving distance and cost of de-icing equipment and the line de-icing index; and an energy consumption allocation unit, used to construct an energy consumption allocation optimization function to determine the line de-icing energy consumption of each de-icing line. When multiple lines need de-icing simultaneously, limited mobile de-icing equipment (such as de-icing trucks) is considered as a schedulable resource, and the optimal disposal sequence and energy consumption allocation for each line are determined through optimization algorithms. In the face of regional icing disasters, this maximizes the utilization of limited de-icing resources, shortens the overall power grid recovery time, and improves the emergency response capability of the power grid.

[0011] A method for intelligent monitoring and collaborative de-icing of transmission lines, applied to the aforementioned system, includes: a monitoring step: real-time acquisition of transmission line icing environmental parameters and image data via a multimodal sensor network; a prediction step: based on the monitoring data, by fusing numerical weather prediction, geographic information system, and real-time image information, predicting the icing growth process and icing type within a future set time period; a strategy generation step: generating a collaborative de-icing strategy based on the icing prediction results and real-time monitoring data; and a de-icing execution step: controlling fixed de-icing devices and mobile de-icing devices to perform collaborative de-icing operations.

[0012] Furthermore, the prediction step includes: weighted fusion of image features, geographic features, and meteorological features for task perception through a dual-branch output model; outputting the icing growth rate through the regression branch, and obtaining the trend of icing thickness change within a future set time period based on multiplying the icing growth rate by the corresponding time interval; outputting the probability distribution of icing type through the classification branch, and taking the icing type with the highest probability as the prediction result.

[0013] Furthermore, the strategy generation step includes: performing fusion analysis on temperature distribution data and wind speed and direction data to determine the conductor torsional load risk level; when the conductor torsional load risk level is high, performing correlation calculation on temperature distribution data and wind speed and direction data to obtain the conductor ice adhesion change curve; dividing the transmission line into several ice melting device control areas based on the conductor ice adhesion change curve, and allocating power to each area to obtain a first control strategy; and combining the first control strategy to perform operation planning for each ice melting device control area to obtain a second control strategy.

[0014] Furthermore, the strategy generation step also includes: acquiring real-time environmental data and ice thickness data of multiple de-icing lines based on the line sensor network; constructing a de-icing index calculation model, calculating the line de-icing index based on the real-time environmental data and ice thickness data, and determining the real-time ice status; calculating a comprehensive evaluation index and determining the line de-icing sequence based on the moving distance and cost of the de-icing equipment, combined with the line de-icing index; constructing an energy consumption allocation optimization function, determining the line de-icing energy consumption of each de-icing line, and generating a multi-line collaborative de-icing strategy.

[0015] The beneficial effects of this invention are as follows: By combining a multimodal sensor network with micro-topography observation and fiber optic sensing, comprehensive and high-precision real-time monitoring of the icing status and environmental parameters of transmission lines is achieved, providing a reliable data foundation for subsequent prediction and decision-making. Employing a prediction model that integrates multiple data sources, it can not only predict the increase in icing thickness but also identify icing types, enabling forward-looking judgments on icing trends and transforming de-icing decisions from passive response to proactive intervention. Combining fixed de-icing devices (such as wind-resistant hammer integrated ultrasonic de-icers) with mobile de-icing devices (such as flamethrower drones and ice-melting trucks) through unified scheduling via a collaborative control module forms a highly efficient de-icing mode combining "fixed-point removal" and "mobile support," overcoming the limitations of single de-icing methods. Through multi-line de-icing sequence optimization, dynamic energy consumption allocation, and mobile device path planning, the optimal allocation of de-icing resources (energy and equipment) is achieved, significantly improving overall de-icing efficiency. Attached Figure Description

[0016] Figure 1 This is a schematic diagram of the system architecture of the present invention; Figure 2 This is a schematic diagram of the method flow of the present invention. Detailed Implementation

[0017] The present invention will now be described in further detail with reference to the accompanying drawings. It should be noted that this is only for the purpose of more clearly illustrating and explaining the present invention.

[0018] like Figure 1 and 2As shown in the figure, this embodiment discloses an intelligent monitoring and collaborative de-icing system for transmission lines, comprising: a monitoring module, including a multimodal sensor network installed on the transmission line for real-time acquisition of icing environmental parameters and image data, the multimodal sensor network including temperature sensors, humidity sensors, wind speed sensors, image sensors, and stress sensors; a prediction module, connected to the monitoring module, for predicting the icing growth process and icing type within a future set time period based on the monitoring data by fusing numerical weather forecasts, geographic information systems, and real-time image information; a de-icing execution module, including a fixed de-icing device and a mobile de-icing device, the fixed de-icing device including an ultrasonic de-icing device integrated on the transmission line wind hammer and a fixed DC de-icing device, the mobile de-icing device including a flame-throwing drone and a mobile DC de-icing vehicle; and a collaborative control module, connected to the prediction module and the de-icing execution module respectively, for generating a collaborative de-icing strategy based on the icing prediction results and real-time monitoring data and controlling the de-icing execution module to perform de-icing.

[0019] A multimodal sensor network comprehensively perceives the icing environment, predictive models proactively assess icing trends, and a collaborative control center intelligently schedules fixed and mobile de-icing resources to achieve precise and efficient collaborative de-icing. During implementation, various sensors deployed along the line (temperature, humidity, wind speed, image, stress) continuously collect data. The prediction module integrates real-time monitoring data, macro-weather forecasts, and geographic information to output a prediction of icing thickness growth and icing type (e.g., rime, frost) for a future period. The collaborative control module generates the optimal de-icing strategy based on the predicted icing severity, type, and real-time risks (e.g., excessive stress). For example, it commands the activation of fixed ultrasonic devices in lightly iced areas while simultaneously dispatching mobile de-icing vehicles to heavily iced areas. Fixed de-icing devices (ultrasonic, DC de-icing) and mobile de-icing devices (drones, de-icing vehicles) receive instructions and perform de-icing operations at designated times and locations. By deeply integrating real-time sensor data, numerical weather forecasts, and geographic information systems, the accuracy of predictions is improved, enabling intelligent, proactive, and collaborative management of power transmission line icing. This significantly improves de-icing efficiency and accuracy, reduces the risk of line breaks and tower collapses caused by icing, and saves manpower and energy costs by optimizing resource scheduling.

[0020] For better results, the monitoring module also includes: a micro-topography icing observation unit, set up in typical micro-topographic areas such as high mountain watersheds, water bodies, canyons, and mountain passes, used to establish a micro-topography icing model; and an optical fiber sensing unit, which uses optical fibers in the transmission line ground wire composite optical cable to continuously monitor the status of the entire line and establish a mapping relationship between optical fiber sensing monitoring and the line's physical status and environmental parameters. Based on the basic sensor network, enhanced monitoring is conducted on micro-topography, a key factor in icing, and existing optical fiber composite overhead ground wire (OPGW) is used to achieve continuous, distributed status sensing of the entire line corridor. The optical fibers in the OPGW are used as distributed sensors, and changes in demodulated light signals (such as backscattered light) are used to invert parameters such as line vibration, strain, and temperature. Monitoring points are strategically deployed in specific geographical units to establish a more accurate micro-meteorological icing model, achieving low-cost, full-coverage line status monitoring without the need for extensive additional sensor deployment. This improves the spatial resolution and comprehensiveness of the monitoring data, making the icing prediction model more accurate, especially effective in predicting localized severe icing caused by micro-topography.

[0021] To achieve better results, the prediction module includes: a dual-branch output model with a regression branch for ice growth rate and a classification branch for ice type identification; and a feature fusion unit for task-aware weighted fusion of image features, geographic features, and meteorological features. The regression branch outputs the ice growth rate over a predetermined future time period, while the classification branch outputs the probability distribution of ice type. During model execution, the input meteorological, image, and geographic features are adaptively weighted and fused. The regression branch then outputs the growth rate in millimeters per hour, and the classification branch outputs the probability of ice types such as "rime ice," "mixed rime ice," and "frost ice." The prediction results include not only ice thickness but also ice type information crucial for de-icing strategy formulation (e.g., rime ice has strong adhesion and requires higher power), making subsequent de-icing decisions more targeted.

[0022] For better results, the ultrasonic de-icing device includes: a wind hammer counterweight system to achieve the original wind hammer's physical wind-resistant resonance function; an inductive energy storage and charging system, consisting of a high-voltage CT power-taking module, a rectifier charging module, and a battery management system, to obtain electrical energy in an insulated, non-contact manner; a temperature and humidity sensing system to activate an icing warning signal when natural icing conditions are met; and an ultrasonic generation system to operate continuously or intermittently based on the working instructions issued by the temperature and humidity sensing system. During implementation, the high-voltage CT draws power non-contactly from the transmission line and charges the battery. The temperature and humidity sensor determines whether the environment meets icing conditions. Once met, the ultrasonic generation system activates, converting electrical energy into mechanical vibrations that act on the wind hammer and the line. It automatically activates when icing conditions are met, using high-frequency ultrasonic vibrations to generate stress at the ice-conductor interface, causing the ice layer to break and fall off. This cleverly utilizes the space and mass of the existing wind hammer component on the line, achieving functional integration without altering the main line structure, making it easy to promote.

[0023] To achieve better results, the collaborative control module includes: a data fusion unit for fusing and analyzing temperature distribution data and wind speed and direction data to determine the torsional load risk level of the conductor; a region division unit for dividing the transmission line into several de-icing device control areas based on the conductor ice adhesion change curve; a power allocation unit for allocating power to each de-icing device control area to obtain the corresponding first control strategy; and a path planning unit for planning the operating path of the mobile de-icing device using an optimization algorithm based on the region ranking results and the first control strategy. For different sections of a single line, refined risk assessment and region division are performed to achieve differentiated power allocation and resource scheduling. During implementation, temperature and wind speed data are fused to determine which sections have a high risk of torsional load due to "wind load and asymmetric icing". For these high-risk sections, the ice adhesion change is further analyzed, and they are divided into different control areas. The optimal de-icing power (first strategy) is calculated for each area, and the optimal travel path (second strategy) is planned for the mobile device within the area. The control objective is not only to remove ice, but also to prevent more dangerous situations such as conductor twisting and jumping. It achieves progressively refined control from macro-risk identification to micro-resource scheduling, accurately deploying de-icing resources to the most dangerous and needed sections, thereby improving de-icing efficiency while ensuring safety.

[0024] To achieve better results, the collaborative control module further includes: a multi-line coordination unit, used to acquire real-time environmental data and ice thickness data of multiple de-icing lines based on the line sensor network; a de-icing index calculation unit, used to calculate the line de-icing index and determine the real-time icing status based on the real-time environmental data and ice thickness data; a de-icing sequence optimization unit, used to calculate a comprehensive evaluation index and determine the line de-icing sequence based on the moving distance and cost of de-icing equipment and the line de-icing index; and an energy consumption allocation unit, used to construct an energy consumption allocation optimization function to determine the line de-icing energy consumption for each de-icing line. When multiple lines need de-icing simultaneously, limited mobile de-icing equipment (such as de-icing trucks) is considered as a schedulable resource, and the optimal disposal sequence and energy consumption allocation for each line are determined through optimization algorithms. In the face of regional icing disasters, this maximizes the utilization of limited de-icing resources, shortens the overall power grid recovery time, and improves the power grid's emergency response capability.

[0025] A method for intelligent monitoring and collaborative de-icing of transmission lines, applied to the aforementioned system, includes: a monitoring step: real-time acquisition of transmission line icing environmental parameters and image data via a multimodal sensor network; a prediction step: based on the monitoring data, by fusing numerical weather prediction, geographic information system, and real-time image information, predicting the icing growth process and icing type within a future set time period; a strategy generation step: generating a collaborative de-icing strategy based on the icing prediction results and real-time monitoring data; and a de-icing execution step: controlling fixed de-icing devices and mobile de-icing devices to perform collaborative de-icing operations.

[0026] To achieve better results, the prediction steps include: weighted fusion of image features, geographic features, and meteorological features using a dual-branch output model; outputting the icing growth rate through the regression branch, and multiplying the icing growth rate by the corresponding time interval to obtain the trend of icing thickness change within a set future time period; and outputting the probability distribution of icing types through the classification branch, and taking the icing type with the highest probability as the prediction result.

[0027] To achieve better results, the strategy generation steps include: fusing and analyzing temperature distribution data and wind speed and direction data to determine the conductor torsional load risk level; when the conductor torsional load risk level is high, performing correlation calculations on temperature distribution data and wind speed and direction data to obtain the conductor ice adhesion change curve; dividing the transmission line into several ice melting device control areas based on the conductor ice adhesion change curve, and allocating power to each area to obtain a first control strategy; and combining the first control strategy to perform operation planning for each ice melting device control area to obtain a second control strategy.

[0028] To achieve better results, the strategy generation steps further include: acquiring real-time environmental data and ice thickness data for multiple de-icing lines based on the line sensor network; constructing a de-icing index calculation model to calculate the line de-icing index and determine the real-time ice status based on the real-time environmental data and ice thickness data; calculating a comprehensive evaluation index and determining the line de-icing sequence based on the moving distance and cost of the de-icing equipment and the line de-icing index; and constructing an energy consumption allocation optimization function to determine the line de-icing energy consumption for each de-icing line and generate a multi-line collaborative de-icing strategy.

[0029] During implementation, this system first requires the deployment of a dense multimodal sensor network along the target transmission line, including sensors for temperature, humidity, wind speed, images (visible light / infrared), and stress. Additional observation points will be set up in micro-topographical areas such as high mountain passes with severe icing. Simultaneously, a distributed sensing system will be constructed using the optical fibers in the existing OPGW (optical fiber composite overhead ground wire) to monitor line vibration, strain, and other information. Data collected by the monitoring module will be transmitted via wireless communication networks (such as 4G / 5G, LoRa) to the prediction and collaborative control modules located at the regional monitoring center. The prediction module will train a model based on historical and real-time data, periodically outputting predictions of icing growth and icing type (such as frost, ice, etc.) for each line section over the next 6-72 hours. Upon receiving the prediction results, the collaborative control module will combine them with real-time monitoring information (such as a sudden increase in stress at a certain point indicating risk) to make decisions. For example, when it is predicted that the ice thickness in a certain section will exceed a threshold and is of the type of rime ice that easily leads to uneven icing, the module will activate the following collaborative strategy: First, it instructs the fixedly installed wind hammer integrated ultrasonic de-icing device in that section to operate intermittently at a specific power to perform preliminary de-icing or suppress ice growth. Second, it dispatches a nearby mobile DC de-icing vehicle to that section and calculates the optimal de-icing current and energy input based on the line resistance, length, and ice thickness. Simultaneously, it controls a flamethrower drone to precisely de-ice complex areas that are difficult to cover by fixed devices such as tower guide wires and insulator strings. Throughout the process, the operating status and de-icing effect of each de-icing device are fed back through the monitoring module, forming a closed-loop control.

[0030] When multiple lines experience simultaneous icing hazards, the collaborative control module executes a multi-line collaborative de-icing process. First, it acquires real-time environmental data (temperature, humidity, wind speed) and ice thickness data for each de-icing line based on a sensor network. Second, it constructs a de-icing index calculation model, which comprehensively considers real-time environmental factors, historical data, and ice thickness to calculate the de-icing index (RBZS) for each line, thereby quantifying the severity of icing and the urgency of de-icing. Then, combining the travel distance and cost of the de-icing equipment to each line with the calculated de-icing index, a comprehensive evaluation index is calculated, and the lines are prioritized for de-icing based on this index, ensuring that the lines with the most severe hazards and highest efficiency are treated first. Finally, it constructs an energy consumption allocation optimization function, aiming to minimize total energy consumption or maximize overall efficiency while meeting the minimum de-icing power requirements of each line, and uses optimization algorithms (such as particle swarm optimization) to solve for the optimal de-icing current or power configuration for each line.

[0031] For the high-altitude mountain pass sections of ultra-high voltage power transmission lines with complex terrain and climate, collaborative control based on data fusion is implemented. First, temperature distribution data and wind speed and direction data for this section are acquired and fused for analysis. For example, asymmetric icing distribution is calculated based on the principle of thermal-mass balance, and combined with conductor vibration data analysis, the torsional load risk level of the conductor is assessed. If a high risk is determined, temperature and wind speed data are further correlated to obtain the change curve of conductor ice adhesion. Based on the ice adhesion change curve, the long line is divided into several control zones with similar de-icing needs. For each zone, power density per unit length is allocated using methods such as deep reinforcement learning, thereby determining the optimal power setpoint for each fixed DC de-icing device within the zone (first control strategy). Next, combined with the power deployment of the fixed devices, ant colony optimization algorithms are used to plan the optimal operating path and operation sequence for mobile de-icing devices (such as flamethrower drone swarms) within their respective zones (second control strategy), ensuring effective spatiotemporal complementarity with the fixed devices. The advantages of this invention are that it does not require changes to the original line structure (utilizing the space integration of wind hammer), can be deployed in a distributed manner, and has a fast response, making it particularly suitable for suppressing the initial growth of icing or clearing light icing.

[0032] The above are merely preferred embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A smart monitoring and collaborative de-icing system for icing on power transmission lines, characterized in that, include: The monitoring module includes a multimodal sensor network installed on the transmission line for real-time acquisition of line icing environmental parameters and image data. The multimodal sensor network includes temperature sensors, humidity sensors, wind speed sensors, image sensors, and stress sensors. The prediction module, connected to the monitoring module, is used to predict the ice growth process and ice type within a future set time period based on monitoring data by fusing numerical weather forecasts, geographic information systems, and real-time image information. The de-icing execution module includes a fixed de-icing device and a mobile de-icing device. The fixed de-icing device includes an ultrasonic de-icing device integrated on the wind hammer of the power transmission line and a fixed DC de-icing device. The mobile de-icing device includes a flame-spraying drone and a mobile DC de-icing vehicle. The collaborative control module is connected to the prediction module and the de-icing execution module respectively. It is used to generate a collaborative de-icing strategy based on the icing prediction results and real-time monitoring data, and control the de-icing execution module to perform de-icing.

2. The intelligent monitoring and collaborative de-icing system for transmission line icing according to claim 1, characterized in that, The monitoring module also includes: a micro-topography icing observation unit, set up in typical micro-topography areas such as high mountain watersheds, water bodies, canyons and passes, used to establish a micro-topography icing model; and an optical fiber sensing unit, which uses optical fibers in the transmission line ground wire composite optical cable to realize continuous monitoring of the status of the entire line and establish a mapping relationship between optical fiber sensing monitoring and the status of the line body and environmental parameters.

3. The intelligent monitoring and collaborative de-icing system for transmission line icing according to claim 1, characterized in that, The prediction module includes: a dual-branch output model, which has a regression branch for ice growth rate and a classification branch for ice type identification; and a feature fusion unit, which performs task-aware weighted fusion of image features, geographic features and meteorological features; wherein, the regression branch outputs the ice growth rate within a set future time period, and the classification branch outputs the probability distribution of ice type.

4. The intelligent monitoring and collaborative de-icing system for transmission line icing according to claim 1, characterized in that, The ultrasonic de-icing device includes: The anti-wind hammer counterweight system is used to realize the original anti-wind hammer physical windproof resonance function; The inductive energy storage and charging system consists of a high-voltage current transformer power extraction module, a rectifier charging module, and a battery management system, and is used to obtain electrical energy in an insulated, non-contact manner. Temperature and humidity sensing system, used to activate icing warning signal when natural icing conditions are met; An ultrasonic generating system is used to operate continuously or intermittently based on working instructions issued by a temperature and humidity sensing system.

5. The intelligent monitoring and collaborative de-icing system for transmission line icing according to claim 1, characterized in that, The collaborative control module includes: The data fusion unit is used to fuse and analyze temperature distribution data and wind speed and direction data to determine the risk level of conductor torsional load. The area division unit is used to divide the transmission line into several de-icing device control areas based on the conductor ice adhesion change curve. The power distribution unit is used to distribute power to the control area of ​​each ice melting device to obtain the first control strategy for the corresponding area. The path planning unit is used to plan the operating path of the mobile de-icing device based on the regional sorting results and the first control strategy, using an optimization algorithm.

6. The intelligent monitoring and collaborative de-icing system for transmission line icing according to claim 5, characterized in that, The collaborative control module also includes: A multi-line coordination unit is used to acquire real-time environmental data and ice thickness data of multiple de-icing lines based on the line sensor network; The ice melting index calculation unit is used to calculate the ice melting index of the line and determine the real-time icing status based on real-time environmental data and ice thickness data. The de-icing sequence optimization unit is used to calculate a comprehensive evaluation index and determine the de-icing sequence of the line based on the moving distance and cost of the de-icing equipment and the line de-icing index. The energy consumption allocation unit is used to construct the energy consumption allocation optimization function and determine the line ice melting energy consumption of each ice melting line.

7. A method for intelligent monitoring and collaborative de-icing of icing on transmission lines, applied to the system described in any one of claims 1-6, characterized in that, include: Monitoring steps: Real-time acquisition of icing environmental parameters and image data of transmission lines through a multimodal sensor network; Prediction steps: Based on monitoring data, by integrating numerical weather prediction, geographic information system and real-time image information, predict the ice growth process and ice type within a future set time period; Strategy generation steps: Generate a collaborative de-icing strategy based on icing prediction results and real-time monitoring data; De-icing procedure: Control the fixed de-icing device and the mobile de-icing device to perform a coordinated de-icing operation.

8. The intelligent monitoring and collaborative de-icing method for transmission line icing according to claim 7, characterized in that, The prediction steps include: A task-aware weighted fusion of image features, geographic features, and meteorological features is performed using a dual-branch output model. The icing growth rate is output by the regression branch. Based on the icing growth rate and the corresponding time interval, the trend of icing thickness change within a set future time period is obtained. The probability distribution of icing types is output through the classification branch, and the icing type with the highest probability is taken as the prediction result.

9. The intelligent monitoring and collaborative de-icing method for transmission line icing according to claim 7, characterized in that, The strategy generation steps include: By integrating and analyzing temperature distribution data and wind speed and direction data, the risk level of torsional load on the conductor can be determined. When the risk level of conductor torsional load is high, the temperature distribution data and wind speed and direction data are correlated and calculated to obtain the conductor ice adhesion force change curve. Based on the change curve of ice adhesion on the conductor, the transmission line is divided into several control areas for ice melting devices, and power is allocated to each area to obtain the first control strategy. By combining the first control strategy with the operational planning of the control area of ​​each ice-melting device, a second control strategy is obtained.

10. The intelligent monitoring and collaborative de-icing method for transmission line icing according to claim 9, characterized in that, The strategy generation step also includes: Real-time environmental data and ice thickness data of multiple de-icing lines are acquired based on the line sensor network; Construct an ice melt index calculation model to calculate the ice melt index of the line and determine the real-time icing status based on real-time environmental data and ice thickness data; Based on the travel distance and cost of the de-icing equipment, combined with the line de-icing index, a comprehensive evaluation index is calculated and the line de-icing sequence is determined. Construct an energy consumption allocation optimization function to determine the line ice melting energy consumption of each ice melting line and generate a multi-line collaborative ice melting strategy.

Images