Wind power plant thunder and lightning sensing and defending system

Abstract

The invention discloses a wind power plant thunder and lightning sensing and defending system, and relates to the technical field of thunder and lightning defending, and a thunder and lightning sensing and electric field monitoring module specifically comprises the steps that electric field intensity data and charge distribution data are collected based on a sensor set 5min time interval obtained by an electric field sensor and a thunder and lightning sensor; after the data is obtained, the system stores electric field intensity and charge distribution data collected in real time in a digital format and summarizes the electric field intensity and charge distribution data into electric field intensity fluctuation data according to a set time interval, the electric field fluctuation data is sorted according to a time sequence and serves as a basic record of electric field environment change in a monitoring area, and a data file is stored in a system database; and the electric field intensity data and the direction change characteristic data in the atmosphere are recorded through high-frequency sampling, the numerical values of the data are converted into path trend data, a time sequence of lightning activities is formed, and the system analyzes the electric field intensity change collected by the lightning detector in real time.

Description

Technical Field

[0001] The present invention relates to the technical field of lightning protection, and in particular to a lightning sensing and protection system for a wind farm. Background Art

[0002] With the development of the wind energy industry, wind farms are being used more and more widely around the world. Wind farms are usually located in open areas such as coasts and plateaus. These areas are high in terrain and lack other obstacles, so they are more vulnerable to the threat of lightning activities. The impact of lightning on wind farms is mainly reflected in the damage to wind turbines and electrical equipment. Especially in areas with dense lightning strikes, the equipment failure rate of wind farms is relatively high, and may even cause serious accidents such as fires and power outages. Therefore, lightning protection measures for wind farms are particularly important.

[0003] Traditional lightning rods and grounding grids are mainly static protection with limited range of action. Especially in the case of large wind farms, they cannot effectively cover all equipment, resulting in some equipment being in blind spots and extremely vulnerable to lightning strikes. Grounding measures do not take into account the dispersion of lightning energy, resulting in lightning energy concentrated on a few devices or grounding points, increasing the load risk of a single device and grounding point. They lack the ability to predict and dynamically induce lightning paths, cannot be adjusted based on real-time monitoring data, and are difficult to adapt to the complex and changeable characteristics of lightning activity. Summary of the invention

[0004] The purpose of the present invention is to address the deficiencies in the prior art and provide a wind farm lightning perception and defense system. Through an intelligent lightning defense system that combines lightning path prediction, airflow induction and multi-point grounding, real-time prediction of lightning paths, decentralized guidance of energy and adjustment of dynamic defense strategies are achieved.

[0005] To this end, the present application provides a wind farm lightning perception and protection system, comprising the following steps:

[0006] Lightning sensing and electric field monitoring module, used to obtain the electric field strength, charge distribution and lightning strike signals around the wind farm and to provide early warning of lightning signals;

[0007] The lightning path prediction and wind farm layout module is used for the electric field data of the lightning perception and electric field monitoring module to predict the lightning path and generate an adaptive layout of the wind farm;

[0008] The airflow induction and wind turbine linkage module adjusts the wind turbine speed and airflow control device to guide the airflow in the wind farm through the lightning path prediction and the layout results of the wind farm layout module;

[0009] Lightning energy dispersion and multi-point grounding module: through multi-point grounding arrangement, lightning energy is guided to multiple grounding points in the wind farm;

[0010] The defense effect monitoring module obtains airflow induction data, electric field changes, ground current and fan status information in real time during the operation of each module, which serves as the core data source for system monitoring;

[0011] Feedback and adjustment module, used to dynamically adjust the airflow induction strategy and fan speed control according to the data feedback from the defense effect monitoring module;

[0012] Data recording and adaptive optimization module, used to store and analyze data from each module, and optimize lightning perception and defense strategies;

[0013] In some specific embodiments:

[0014] The lightning sensing and electric field monitoring module specifically includes:

[0015] Among them, the electric field data intensity is obtained based on the electric field sensor;

[0016] Among them, the charge distribution data is obtained based on the lightning sensor;

[0017] Among them, the sensors obtained based on the electric field sensor and the lightning sensor set a 5-min time interval to collect the electric field strength data and charge distribution data;

[0018] After acquiring the data, the system stores the real-time collected electric field strength and charge distribution data in digital format, and summarizes them into electric field strength fluctuation data at set time intervals. The electric field fluctuation data are sorted in chronological order as the basic record of the changes in the electric field environment in the monitoring area. The data files will be stored in the system database for subsequent analysis of the accumulation of cloud charges;

[0019] Lightning signal detection: detecting lightning signals based on the electric field strength data and charge distribution data obtained by the data collection module;

[0020] Among them, the electric field strength data and direction change characteristic data in the atmosphere are recorded through high-frequency sampling, and their values ​​are converted into path trend data to form a time series of lightning activities. The system conducts real-time analysis on the changes in electric field strength collected by lightning detectors to capture the rapid changes in the direction of the electric field, thereby inferring the possible direction and landing area of ​​lightning. The distribution data of multiple point detectors are compared in real time to generate a trend diagram of the lightning path, marking the potential striking location as a reference for subsequent defense measures;

[0021] High electric field threshold warning, based on analysis of electric field strength data, charge distribution data, and lightning signal detection data:

[0022] Among them: the lightning warning threshold is set according to historical lightning data and actual environmental conditions, based on the average electric field intensity of lightning activities in the area and the electric field rise rate before a typical lightning strike, and the threshold range is set to 30kV / m;

[0023] Among them: when it is detected that the electric field strength is close to or exceeds the preset lightning warning threshold of 30kV / m, the system will determine whether the warning condition is met based on the time series trend of the electric field rise rate and the electric field strength. If the electric field strength increases rapidly within a unit time and stably exceeds the threshold, the system determines that the lightning risk has increased and triggers the warning signal generation mechanism;

[0024] Once the triggering conditions are met, the system automatically activates the warning signal generation module, first storing the electric field changes, rising rates and values ​​reaching the threshold in the real-time database. The warning signal generated by the system includes data such as timestamp, specific electric field strength value and its rate of change. After the warning signal is generated, the system continues to monitor whether the defense response level needs to be further improved according to the subsequent electric field changes.

[0025] Based on the lightning path prediction and wind farm layout module and the wind farm layout data integration module, the spatial layout information of each wind turbine and facility in the wind farm is collected and integrated;

[0026] Among them: collecting detailed information on the topography and landforms in and around the wind farm, including mainstream wind direction, slope, and geological characteristics;

[0027] Among them: all the collected spatial coordinate data of wind turbines, facilities and surrounding environment are converted into a unified coordinate system;

[0028] Among them: Based on the actual boundaries of the wind farm and the collected geographic data, a three-dimensional model of the wind farm is established through GIS software. The spatial position, relative height and surrounding terrain characteristics of each wind turbine and facility are integrated into a three-dimensional spatial layout map in the model to intuitively display the spatial distribution of wind farm facilities;

[0029] The lightning path prediction module uses the LSTM model to predict the lightning path based on the data features of the wind farm layout data integration module. The LSTM model predicts the direction of lightning travel by modeling the trend of electric field changes and the spatial features of the layout.

[0030] Airflow induction layout generation, based on the path data provided by the lightning path prediction module, determines the key areas of the airflow induction layout;

[0031] Among them: through the control of the speed and blade angle of the wind turbine, a low-pressure area and an airflow vortex area are formed in the wind farm, so that the airflow induction effect is enhanced. The setting of the low-pressure area is based on the analysis of the path and facility distribution, and the vortex area is generated through the coordinated control of adjacent wind turbines;

[0032] Based on lightning energy dispersion and multi-point grounding module;

[0033] Grounding device layout: multiple grounding points are evenly set up in the wind farm, and a widely covered conductive network is formed with carbon fiber materials. This network structure ensures that the grounding points are evenly distributed, so that the lightning current can be quickly conducted and dispersed among multiple grounding points, avoiding concentrated load at a single location and achieving effective lightning energy release;

[0034] Each grounding point is in close contact with the ground through highly conductive materials to form a low impedance path, allowing the lightning current to be conducted between the grounding points and diffuse to the ground. The carbon fiber grounding layer acts as a highly conductive path in the overall structure, quickly introducing lightning energy into the ground and avoiding overload of concentrated current at a single location.

[0035] Low pressure areas and airflow vortex areas generated based on the airflow induction layout in the wind farm layout data integration module;

[0036] Among them, the wind turbine speed and blade angle are adjusted to form a specific airflow direction to enhance the attraction of the low-pressure area and deflect lightning when it approaches the wind farm facilities;

[0037] The energy dispersion of multi-point grounding is based on the multiple grounding points arranged by the grounding device and the low-pressure area generated by the airflow induction layout. The speed and blade angle are adjusted through the coordinated control of the fan to form an airflow path, which gradually guides the lightning to the area with dense distribution of grounding points, so that it gradually approaches the location of the grounding point during the deflection process.

[0038] Defense effect monitoring module, including:

[0039] Among them, the airflow induction effect data, electric field strength change data, current distribution data of each grounding point and fan speed data are obtained based on the electric field sensor, lightning sensor and fan status monitor;

[0040] Real-time defense effect evaluation is performed based on the airflow induction effect data, electric field strength change data, current distribution data of each grounding point and fan speed data obtained by the sensor;

[0041] Among them, the intensity threshold of airflow induction is set according to the meteorological conditions of the wind farm and the requirements of lightning protection. The specific value depends on the maximum wind speed that the wind turbine can provide and the blade adjustment range. The airflow induction intensity in the low-pressure area can be set in the range of 10-15 m / s to ensure that the airflow is sufficient to guide the deflection of lightning;

[0042] Among them, the electric field strength change data is used to identify the potential path of lightning activity. When the electric field strength change exceeds 30kV / m and the increase rate reaches the set threshold of 5kV / m per second, the system will determine it as a high lightning risk;

[0043] The upper limit of the current load of the grounding point is set according to the material conductivity and grounding resistance, and is set within the range of 500-1000A to ensure that the current of each grounding point will not be overloaded. If the current of the grounding point exceeds this range, the system will regard it as an abnormal load.

[0044] Feedback and adjustment module, including:

[0045] Based on the data of airflow induction effect, electric field strength change, current distribution of each grounding point and fan speed data;

[0046] Among them, the data based on airflow induction effect, electric field strength change data, current distribution data of each grounding point and fan speed data are preprocessed, and the preprocessing includes: data dimension reduction and feature extraction;

[0047] Among them, data dimension reduction and feature extraction use principal component analysis to reduce data dimension, retain main features, reduce data redundancy, and extract the airflow intensity change rate, electric field gradient, and grounding point current peak value from the processed data;

[0048] Based on the obtained fan speed data, the fan speed is adjusted according to the airflow induction requirements to increase or decrease the local airflow intensity. The specific speed setting is determined by the results of the deviation analysis. When the airflow intensity is insufficient, the system increases the fan speed to the set upper limit to enhance the low-pressure area effect.

[0049] The resistance value of the carbon fiber material at the grounding point is analyzed based on the electric field strength change data. The conductivity of a certain grounding point is low, and the grounding point is marked as a conductivity adjustment target;

[0050] Data recording and adaptive optimization module, including:

[0051] Based on the data obtained by the defense effectiveness monitoring module, the information is classified and stored according to the data source and data type, and the data is attached with timestamps, locations and specific tags to ensure the accuracy and traceability of subsequent analysis;

[0052] Among them, a convolutional neural network is used to process the data obtained by the monitoring module, build an environmental model, and define the state space of airflow, electric field strength, and fan speed;

[0053] Among them, after each defense strategy is executed, the data recording and adaptive optimization module will obtain feedback information, including the number of lightning strikes, lightning protection effects, and environmental changes;

[0054] Based on the convolutional neural network reinforcement learning algorithm, the strategy model is updated, and the fan speed and grounding point distribution are adjusted according to the algorithm update strategy model.

[0055] In summary, the present application provides a wind farm lightning perception and defense system, which realizes real-time prediction of lightning paths, decentralized energy guidance and adjustment of dynamic defense strategies through an intelligent lightning defense system that combines lightning path prediction, airflow induction and multi-point grounding. It monitors electric field changes in real time through electric field sensors and lightning detectors, and warns of lightning strike risks in advance. It uses AI algorithms to predict paths, and effectively guides lightning paths through airflow induction and wind turbine linkage to avoid concentrated lightning strikes. The multi-point grounding design can effectively disperse lightning energy and avoid the risk of overload at a single grounding point. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 It is an overall flow chart of a wind farm lightning perception and protection system provided in an embodiment of the present application. DETAILED DESCRIPTION

[0057] Please refer to Figure 1 , which shows the process of an embodiment of a wind farm lightning perception and protection system according to the present disclosure.

[0058] like Figure 1 As shown, a wind farm lightning sensing and protection system includes the following steps:

[0059] Lightning sensing and electric field monitoring module, used to obtain the electric field strength, charge distribution and lightning strike signals around the wind farm and to provide early warning of lightning signals;

[0060] The lightning path prediction and wind farm layout module is used for the electric field data of the lightning perception and electric field monitoring module to predict the lightning path and generate an adaptive layout of the wind farm;

[0061] The airflow induction and wind turbine linkage module adjusts the wind turbine speed and airflow control device to guide the airflow in the wind farm through the lightning path prediction and the layout results of the wind farm layout module;

[0062] Lightning energy dispersion and multi-point grounding module: through multi-point grounding arrangement, lightning energy is guided to multiple grounding points in the wind farm;

[0063] The defense effect monitoring module obtains airflow induction data, electric field changes, ground current and fan status information in real time during the operation of each module, which serves as the core data source for system monitoring;

[0064] Feedback and adjustment module, used to dynamically adjust the airflow induction strategy and fan speed control according to the data feedback from the defense effect monitoring module;

[0065] Data recording and adaptive optimization module, used to store and analyze data from each module, and optimize lightning perception and defense strategies;

[0066] In some specific embodiments:

[0067] The lightning sensing and electric field monitoring module specifically includes:

[0068] Among them, the electric field data intensity is obtained based on the electric field sensor;

[0069] Among them, the charge distribution data is obtained based on the lightning sensor;

[0070] Among them, the sensors obtained based on the electric field sensor and the lightning sensor set a 5-min time interval to collect the electric field strength data and charge distribution data;

[0071] After acquiring the data, the system stores the real-time collected electric field strength and charge distribution data in digital format, and summarizes them into electric field strength fluctuation data at set time intervals.

[0072] The data are sorted in chronological order as the basic record of the changes in the electric field environment in the monitoring area. The data files will be stored in the system database for subsequent analysis of the accumulation of cloud charges;

[0073] Lightning signal detection: detecting lightning signals based on the electric field strength data and charge distribution data obtained by the data collection module;

[0074] Among them, the electric field strength data and direction change characteristic data in the atmosphere are recorded through high-frequency sampling, and their values ​​are converted into path trend data to form a time series of lightning activities. The system conducts real-time analysis on the changes in electric field strength collected by lightning detectors to capture the rapid changes in the direction of the electric field, thereby inferring the possible direction and landing area of ​​lightning. The distribution data of multiple point detectors are compared in real time to generate a trend diagram of the lightning path, marking the potential striking location as a reference for subsequent defense measures;

[0075] High electric field threshold warning, based on analysis of electric field strength data, charge distribution data, and lightning signal detection data:

[0076] Among them: the lightning warning threshold is set according to historical lightning data and actual environmental conditions, based on the average electric field intensity of lightning activities in the area and the electric field rise rate before a typical lightning strike, and the threshold range is set to 30kV / m;

[0077] Among them: when it is detected that the electric field strength is close to or exceeds the preset lightning warning threshold of 30kV / m, the system will determine whether the warning condition is met based on the time series trend of the electric field rise rate and the electric field strength. If the electric field strength increases rapidly within a unit time and stably exceeds the threshold, the system determines that the lightning risk has increased and triggers the warning signal generation mechanism;

[0078] Once the triggering conditions are met, the system automatically activates the warning signal generation module, first storing the electric field changes, rising rates and values ​​reaching the threshold in the real-time database. The warning signal generated by the system includes data such as timestamp, specific electric field strength value and its rate of change. After the warning signal is generated, the system continues to monitor whether the defense response level needs to be further improved according to the subsequent electric field changes.

[0079] Based on the lightning path prediction and wind farm layout module and the wind farm layout data integration module, the spatial layout information of each wind turbine and facility in the wind farm is collected and integrated;

[0080] Among them: collecting detailed information on the topography and landforms in and around the wind farm, including mainstream wind direction, slope, and geological characteristics;

[0081] Among them: all the collected spatial coordinate data of wind turbines, facilities and surrounding environment are converted into a unified coordinate system;

[0082] Among them: Based on the actual boundaries of the wind farm and the collected geographic data, a three-dimensional model of the wind farm is established through GIS software. The spatial position, relative height and surrounding terrain characteristics of each wind turbine and facility are integrated into a three-dimensional spatial layout map in the model to intuitively display the spatial distribution of wind farm facilities;

[0083] The lightning path prediction module uses the LSTM model to predict the lightning path based on the data features of the wind farm layout data integration module. The LSTM model predicts the direction of lightning travel by modeling the trend of electric field changes and the spatial features of the layout.

[0084] Airflow induction layout generation, based on the path data provided by the lightning path prediction module, determines the key areas of the airflow induction layout;

[0085] Among them: through the control of the speed and blade angle of the wind turbine, a low-pressure area and an airflow vortex area are formed in the wind farm, so that the airflow induction effect is enhanced. The setting of the low-pressure area is based on the analysis of the path and facility distribution, and the vortex area is generated through the coordinated control of adjacent wind turbines;

[0086] Based on lightning energy dispersion and multi-point grounding module;

[0087] Grounding device layout: multiple grounding points are evenly set up in the wind farm, and a widely covered conductive network is formed with carbon fiber materials. This network structure ensures that the grounding points are evenly distributed, so that the lightning current can be quickly conducted and dispersed among multiple grounding points, avoiding concentrated load at a single location and achieving effective lightning energy release;

[0088] Each grounding point is in close contact with the ground through highly conductive materials to form a low impedance path, allowing the lightning current to be conducted between the grounding points and diffuse to the ground. The carbon fiber grounding layer acts as a highly conductive path in the overall structure, quickly introducing lightning energy into the ground and avoiding overload of concentrated current at a single location.

[0089] Low pressure areas and airflow vortex areas generated based on the airflow induction layout in the wind farm layout data integration module;

[0090] Among them, the wind turbine speed and blade angle are adjusted to form a specific airflow direction to enhance the attraction of the low-pressure area and deflect lightning when it approaches the wind farm facilities;

[0091] The energy dispersion of multi-point grounding is based on the multiple grounding points arranged by the grounding device and the low-pressure area generated by the airflow induction layout. The speed and blade angle are adjusted through the coordinated control of the fan to form an airflow path, which gradually guides the lightning to the area with dense distribution of grounding points, so that it gradually approaches the location of the grounding point during the deflection process.

[0092] Defense effect monitoring module, including:

[0093] Among them, the airflow induction effect data, electric field strength change data, current distribution data of each grounding point and fan speed data are obtained based on the electric field sensor, lightning sensor and fan status monitor;

[0094] Real-time defense effect evaluation is performed based on the airflow induction effect data, electric field strength change data, current distribution data of each grounding point and fan speed data obtained by the sensor;

[0095] Among them, the intensity threshold of airflow induction is set according to the meteorological conditions of the wind farm and the requirements of lightning protection. The specific value depends on the maximum wind speed that the wind turbine can provide and the blade adjustment range. The airflow induction intensity in the low-pressure area can be set in the range of 10-15 m / s to ensure that the airflow is sufficient to guide the deflection of lightning;

[0096] Among them, the electric field strength change data is used to identify the potential path of lightning activity. When the electric field strength change exceeds 30kV / m and the increase rate reaches the set threshold of 5kV / m per second, the system will determine it as a high lightning risk;

[0097] The upper limit of the current load of the grounding point is set according to the material conductivity and grounding resistance, and is set within the range of 500-1000A to ensure that the current of each grounding point will not be overloaded. If the current of the grounding point exceeds this range, the system will regard it as an abnormal load.

[0098] Feedback and adjustment module, including:

[0099] Based on the data of airflow induction effect, electric field strength change, current distribution of each grounding point and fan speed data;

[0100] Among them, the data based on airflow induction effect, electric field strength change data, current distribution data of each grounding point and fan speed data are preprocessed, and the preprocessing includes: data dimension reduction and feature extraction;

[0101] Among them, data dimension reduction and feature extraction use principal component analysis to reduce data dimension, retain main features, reduce data redundancy, and extract the airflow intensity change rate, electric field gradient, and grounding point current peak value from the processed data;

[0102] Based on the obtained fan speed data, the fan speed is adjusted according to the airflow induction requirements to increase or decrease the local airflow intensity. The specific speed setting is determined by the results of the deviation analysis. When the airflow intensity is insufficient, the system increases the fan speed to the set upper limit to enhance the low-pressure area effect.

[0103] The resistance value of the carbon fiber material at the grounding point is analyzed based on the electric field strength change data. The conductivity of a certain grounding point is low, and the grounding point is marked as a conductivity adjustment target;

[0104] Data recording and adaptive optimization module, including:

[0105] Based on the data obtained by the defense effectiveness monitoring module, the information is classified and stored according to the data source and data type, and the data is attached with timestamps, locations and specific tags to ensure the accuracy and traceability of subsequent analysis;

[0106] Among them, a convolutional neural network is used to process the data obtained by the monitoring module, build an environmental model, and define the state space of airflow, electric field strength, and fan speed;

[0107] Among them, after each defense strategy is executed, the data recording and adaptive optimization module will obtain feedback information, including the number of lightning strikes, lightning protection effects, and environmental changes;

[0108] Based on the convolutional neural network reinforcement learning algorithm, the strategy model is updated, and the fan speed and grounding point distribution are adjusted according to the algorithm update strategy model.

[0109] In the above content, in actual application, the real-time collected electric field strength and charge distribution data are stored in digital format and summarized as electric field strength fluctuation data at set time intervals. The electric field fluctuation data are sorted in chronological order as the basic record of changes in the electric field environment in the monitoring area. The data file will be stored in the system database for subsequent analysis of the accumulation of cloud charges. Based on the above data, lightning signals are detected, and the electric field strength data and direction change characteristic data in the atmosphere are recorded through high-frequency sampling, and their values ​​are converted into path trend data to form a time series of lightning activities. The system performs real-time analysis on the changes in electric field strength collected by lightning detectors to capture rapid changes in the direction of the electric field, thereby inferring the possible direction and landing area of ​​lightning. The distribution data of multiple point detectors are compared in real time to generate a trend diagram of the lightning path, marking potential strike locations as a reference for subsequent defense measures;

[0110] The data obtained above are analyzed: the lightning warning threshold is set according to historical lightning data and actual environmental conditions, based on the average electric field strength of lightning activities in the area and the electric field rise rate before a typical lightning strike. The threshold range is set to 30kV / m. When the electric field strength is detected to be close to or exceeds the preset lightning warning threshold of 30kV / m, the system will determine whether the warning condition is met based on the time series trend of the electric field rise rate and the electric field strength. The electric field strength increases rapidly within a unit time and stably exceeds the threshold. The system determines that the lightning risk is increasing and triggers the warning signal generation mechanism. When the triggering conditions are met, the system automatically activates the warning signal generation module. First, the electric field changes, the rise rate and the value reaching the threshold are stored in the real-time database. The warning signal generated by the system includes timestamps, specific electric field strength values ​​and their change rates. After the warning signal is generated, the system continues to monitor whether the defense response level needs to be further improved according to the subsequent electric field changes; according to the wind power The actual boundary of the field and the collected geographic data are used to establish a three-dimensional model of the wind farm through GIS software. The spatial position, relative height and surrounding terrain characteristics of each wind turbine and facility are integrated into a three-dimensional spatial layout diagram in the model to intuitively display the spatial distribution of wind farm facilities. Based on the data characteristics of the wind farm layout data integration module, the LSTM model is used to predict the lightning path. The LSTM model predicts the direction of lightning travel by modeling the trend of electric field changes and the spatial characteristics of the layout. Based on the path data provided by the lightning path prediction module, the key areas of the airflow induction layout are determined. The speed and blade angle of the wind turbine are used to control the formation of low-pressure areas and airflow vortex areas in the wind farm to enhance the airflow induction effect. The setting of the low-pressure area is based on the analysis of the path and facility distribution. The vortex area is generated through the coordinated control of adjacent wind turbines. The wind turbine speed and blade angle are adjusted to form a specific airflow direction to enhance the attraction of the low-pressure area and deflect lightning when approaching wind farm facilities.

[0111] Multiple grounding points are evenly set up in the wind farm, and a widely covered conductive network is formed with carbon fiber materials. This network structure ensures that the grounding points are evenly distributed, so that the lightning current can be quickly conducted and dispersed between multiple grounding points, avoiding concentrated load at a single location, and achieving effective lightning energy release. Each grounding point is in close contact with the ground by highly conductive materials to form a low-impedance path, so that the lightning current is conducted between the grounding points and diffused to the ground. The carbon fiber grounding layer acts as a highly conductive path in the overall structure, quickly introducing lightning energy into the ground, avoiding overload of concentrated current at a single location;

[0112] The energy dispersion of multi-point grounding is based on the multiple grounding points arranged by the grounding device and the low-pressure area generated by the airflow induction layout. The speed and blade angle are adjusted through the coordinated control of the fan to form an airflow path, which gradually guides the lightning to the area with dense distribution of grounding points, so that it gradually approaches the location of the grounding point during the deflection process.

[0113] The airflow induction effect data, electric field strength change data, current distribution data of each grounding point and wind turbine speed data are obtained based on the electric field sensor, lightning sensor and wind turbine status monitor. The airflow induction intensity threshold is set according to the meteorological conditions and lightning protection requirements of the wind farm. The specific value depends on the maximum wind speed and blade adjustment range that the wind turbine can provide. The airflow induction intensity in the low-pressure area is set in the range of 10-15 m / s to ensure that the airflow is sufficient to guide the lightning deflection. The electric field strength change data is used to identify the potential path of lightning activity. When the electric field strength changes by more than 30kV / m and the increase rate reaches the set threshold of 5kV / m per second, the system will determine it as a high lightning risk. The current load upper limit of the grounding point is set according to its material conductivity and grounding resistance, and is set in the range of 500-1000A to ensure that the current of each grounding point will not be overloaded. If the current of the grounding point exceeds this range, the system will regard it as a load abnormality.

[0114] Preprocessing is performed on the data based on the airflow induction effect, the electric field strength change data, the current distribution data of each grounding point and the fan speed data. The preprocessing includes: data dimension reduction and feature extraction. The data dimension reduction and feature extraction are reduced by the principal component analysis method to retain the main features and reduce data redundancy. The airflow intensity change rate, electric field gradient and grounding point current peak are extracted from the processed data. Based on the obtained fan speed data, the fan speed is adjusted according to the airflow induction requirements to increase or decrease the local airflow intensity. The specific speed setting is determined by the results of the deviation analysis. When the airflow intensity is insufficient, the system increases the fan speed to the set upper limit to enhance the low-pressure zone effect. The conductivity of a certain grounding point is low, and the grounding point is marked as the conductivity adjustment target.

[0115] Based on the data obtained by the defense effect monitoring module, the information is classified and stored according to the data source and data type, and the data is attached with timestamps, locations and specific tags to ensure the accuracy and traceability of subsequent analysis. The convolutional neural network is used to process the data obtained by the monitoring module, build an environmental model, and define the state space of airflow, electric field strength, and fan speed. After each defense strategy is executed, the data recording and adaptive optimization module will obtain feedback information, such as the number of lightning strikes, lightning protection effects, and environmental changes. The strategy model is updated based on the convolutional neural network reinforcement learning algorithm, and the fan speed and grounding point distribution are adjusted according to the algorithm-updated strategy model.

Claims

1. A wind farm lightning sensing and protection system, characterized in that: Includes the following modules: Lightning sensing and electric field monitoring module, used to obtain the electric field strength, charge distribution and lightning strike signals around the wind farm and to provide early warning of lightning signals; The lightning path prediction and wind farm layout module is used for the electric field data of the lightning perception and electric field monitoring module to predict the lightning path and generate an adaptive layout of the wind farm; The airflow induction and wind turbine linkage module adjusts the wind turbine speed and airflow control device to guide the airflow in the wind farm through the lightning path prediction and the layout results of the wind farm layout module; Lightning energy dispersion and multi-point grounding module: through multi-point grounding arrangement, lightning energy is guided to multiple grounding points in the wind farm; The defense effect monitoring module obtains airflow induction data, electric field changes, ground current and fan status information in real time during the operation of each module, which serves as the core data source for system monitoring; Feedback and adjustment module, used to dynamically adjust the airflow induction strategy and fan speed control according to the data feedback from the defense effect monitoring module; The data recording and adaptive optimization module is used to store and analyze the data of each module and optimize the lightning perception and defense strategies.

2. A wind farm lightning sensing and protection system according to claim 1, characterized in that: Lightning sensing and electric field monitoring module include: Among them, the electric field data intensity is obtained based on the electric field sensor; Among them, the charge distribution data is obtained based on the lightning sensor; Among them, the sensors obtained based on the electric field sensor and the lightning sensor are set to collect the electric field strength data and the charge distribution data at a 5-min time interval; Lightning signal detection: detecting lightning signals based on the electric field strength data and charge distribution data obtained by the data collection module; Among them, the electric field strength data and direction change characteristic data in the atmosphere are recorded through high-frequency sampling, and their values ​​are converted into path trend data to form a time series of lightning activities. The system conducts real-time analysis on the changes in electric field strength collected by lightning detectors to capture the rapid changes in the direction of the electric field, thereby inferring the possible direction and landing area of ​​lightning. The distribution data of multiple point detectors are compared in real time to generate a trend diagram of the lightning path and mark the potential strike location; High electric field threshold warning, based on analysis of electric field strength data, charge distribution data, and lightning signal detection data: Among them: the lightning warning threshold is set according to historical lightning data and actual environmental conditions, based on the average electric field intensity of lightning activities in the area and the electric field rise rate before a typical lightning strike, and the threshold range is set to 30kV / m; Among them: when it is detected that the electric field strength is close to or exceeds the preset lightning warning threshold of 30kV / m, the system will determine whether the warning conditions are met based on the time series trend of the electric field rise rate and the electric field strength. If the electric field strength increases rapidly per unit time and stably exceeds the threshold, the system determines that the lightning risk has increased and triggers the warning signal generation mechanism.

3. A wind farm lightning sensing and protection system according to claim 1, characterized in that: Lightning path prediction and wind field layout module, including: Wind farm layout data integration module, which collects and integrates the spatial layout information of each wind turbine and facility in the wind farm; Among them: collecting detailed information on the topography and landforms in and around the wind farm, including mainstream wind direction, slope, and geological characteristics; Among them: all the collected spatial coordinate data of wind turbines, facilities and surrounding environment are converted into a unified coordinate system; Among them: Based on the actual boundaries of the wind farm and the collected geographic data, a three-dimensional model of the wind farm is established through GIS software. The spatial position, relative height and surrounding terrain characteristics of each wind turbine and facility are integrated into a three-dimensional spatial layout diagram in the model; The lightning path prediction module uses the LSTM model to predict the lightning path based on the data features of the wind farm layout data integration module. The LSTM model predicts the direction of lightning travel by modeling the trend of electric field changes and the spatial features of the layout. Airflow induction layout generation, based on the path data provided by the lightning path prediction module, determines the airflow induction layout area; Among them: low-pressure areas and airflow vortex areas are formed in the wind farm by controlling the speed and blade angle of the wind turbine, so that the airflow induction effect is enhanced. The setting of the low-pressure area is based on the analysis of the path and facility distribution, and the vortex area is generated through the coordinated control of adjacent wind turbines.

4. A wind farm lightning sensing and protection system according to claim 1, characterized in that: Specifically include: Lightning energy dispersion and multi-point grounding module, including: Grounding device layout: multiple grounding points are evenly set up in the wind farm, and a widely covered conductive network is formed with carbon fiber materials. This network structure ensures that the grounding points are evenly distributed, so that the lightning current can be quickly conducted and dispersed among multiple grounding points, avoiding concentrated load at a single location and achieving effective lightning energy release; Low pressure areas and airflow vortex areas generated based on the airflow induction layout in the wind farm layout data integration module; Among them, the wind turbine speed and blade angle are adjusted to form a specific airflow direction to enhance the attraction of the low-pressure area and deflect lightning when it approaches the wind farm facilities; The energy dispersion of multi-point grounding is based on the low-pressure area generated by the multiple grounding points arranged by the grounding device and the airflow induction layout. The speed and blade angle are adjusted through the coordinated control of the fan to form an airflow path, which gradually guides the lightning to the area with dense grounding points, so that it gradually approaches the location of the grounding point during the deflection process.

5. A wind farm lightning sensing and protection system according to claim 1, characterized in that: Defense effect monitoring module, specifically include: Among them, the airflow induction effect data, electric field strength change data, current distribution data of each grounding point and fan speed data are obtained based on the electric field sensor, lightning sensor and fan status monitor; Real-time defense effect evaluation is performed based on the airflow induction effect data, electric field strength change data, current distribution data of each grounding point and fan speed data obtained by the sensor; Among them, the intensity threshold of airflow induction is set according to the meteorological conditions of the wind farm and the requirements of lightning protection. The specific value depends on the maximum wind speed that the wind turbine can provide and the blade adjustment range. The airflow induction intensity in the low-pressure area can be set in the range of 10-15 m / s to ensure that the airflow is sufficient to guide the deflection of lightning; Among them, the electric field strength change data is used to identify the potential path of lightning activity. When the electric field strength change exceeds 30kV / m and the increase rate reaches the set threshold of 5kV / m per second, the system will determine it as a high lightning risk; Among them, the current load upper limit of the grounding point is set according to its material conductivity and grounding resistance, and is set within the range of 500-1000A to ensure that the current of each grounding point will not be overloaded. If the current of the grounding point exceeds this range, the system will regard it as an abnormal load.

6. A wind farm lightning sensing and protection system according to claim 1, characterized in that: Feedback and adjustment module, including: Based on the data of airflow induction effect, electric field strength change, current distribution of each grounding point and fan speed data; Among them, the data based on airflow induction effect, electric field strength change data, current distribution data of each grounding point and fan speed data are preprocessed, and the preprocessing includes: data dimension reduction and feature extraction; Among them, data dimension reduction and feature extraction use principal component analysis to reduce data dimension, retain main features, reduce data redundancy, and extract the airflow intensity change rate, electric field gradient, and grounding point current peak value from the processed data; Based on the obtained fan speed data, the fan speed is adjusted according to the airflow induction requirements to increase or decrease the local airflow intensity. The specific speed setting is determined by the results of the deviation analysis. When the airflow intensity is insufficient, the system increases the fan speed to the set upper limit to enhance the low-pressure area effect. The resistance value of the carbon fiber material at the grounding point is analyzed based on the electric field strength change data. The conductivity of a certain grounding point is low, and the grounding point is marked as a conductivity adjustment target.

7. A wind farm lightning sensing and protection system according to claim 1, characterized in that: Data recording and adaptive optimization module, including: Based on the data obtained by the defense effectiveness monitoring module, the information is classified and stored according to the data source and data type, and the data is attached with timestamps, locations and specific tags to ensure the accuracy and traceability of subsequent analysis; Among them, a convolutional neural network is used to process the data obtained by the monitoring module, build an environmental model, and define the state space of airflow, electric field strength, and fan speed; Among them, after each defense strategy is executed, the data recording and adaptive optimization module will obtain feedback information, including the number of lightning strikes, lightning protection effects, and environmental changes; Based on the convolutional neural network reinforcement learning algorithm, the strategy model is updated, and the fan speed and grounding point distribution are adjusted according to the algorithm update strategy model.

Images