Wind turbine lightning strike detection method based on radio frequency electromagnetic detection

Abstract

The application belongs to the technical field of radio frequency lightning positioning, and specifically discloses a wind turbine lightning stroke and lightning detection method based on radio frequency electromagnetic detection. The application plans the initial layout of the directional antenna by means of signal-to-noise ratio, elevation angle and site quantity constraints, overcomes the defects of low signal-to-noise ratio and easy interference of the omnidirectional antenna in the wind farm, and realizes selective focusing of the target airspace and high signal-to-noise ratio data acquisition. Further, the station arrangement parameters are iteratively optimized through simulation of lightning signals, and the monitoring capability of the lightning stroke area of the key components of the wind turbine is enhanced. The time parameters are accurately extracted by using the high gain characteristics of the directional antenna, and combined with the unified time base alignment technology, the high precision measurement of the time difference of arrival is ensured. Finally, the high-resolution three-dimensional lightning stroke channel form is generated based on the spherical propagation model, the detection interference problem in the complex environment of the wind farm is systematically solved, and accurate three-dimensional observation basis is provided for lightning protection design and accident analysis.

Description

Technical Field

[0001] This application belongs to the field of radio frequency lightning location technology, and more specifically, relates to a method for detecting lightning strikes on wind turbines based on radio frequency electromagnetic detection. Background Technology

[0002] Lightning is a common natural phenomenon, often occurring during severe convective weather, and is characterized by strong spatiotemporal randomness and suddenness. Multiple physical processes typically accompany lightning discharges, resulting in complex channel structures and rapid evolution, making precise observation challenging. Existing lightning detection methods can be categorized by observation method into contact and non-contact types; and by frequency band into very low frequency (VLF) and radio frequency (RF) methods. Non-contact methods offer advantages such as flexibility and high efficiency in on-site data acquisition. Compared to VLF, RF radiation features high intensity and is less affected by ground environmental parameters, persisting almost throughout the entire discharge process. By receiving and analyzing RF radiation, high temporal resolution and high-precision location of cloud-to-ground lightning can be achieved within a certain range.

[0003] In recent years, with the rapid development of wind power, the single-unit capacity of wind turbine generators has been continuously increasing. To capture more wind energy, the hub height and rotor diameter of the generators have also increased accordingly. Wind turbine blades have a large exposed area and high tip speed during thunderstorms, making them more susceptible to triggering upward leaders and more prone to negative polarity cloud flashes compared to the surrounding environment. Furthermore, wind farms are also more likely to trigger lightning when cloud levels are low and electric fields are strong.

[0004] Therefore, it is evident that combining radio frequency radiation observation in wind farms can provide a more direct way to observe lightning strikes. However, in non-contact radio frequency lightning location systems, existing solutions mostly use omnidirectional antennas to receive radiated signals, which has significant limitations in wind farm scenarios with dense equipment and complex electromagnetic environments. When the source point is at the edge of the observation network or at a large horizontal distance from the observation station, positioning errors will occur due to geometric limitations; at the same time, the omnidirectional antenna will simultaneously receive radiation from distant clouds and environmental noise, resulting in insufficient signal-to-noise ratio and an inability to accurately capture the near-ground details of cloud-to-ground lightning.

[0005] Therefore, existing systems have difficulty in accurately locating the near-ground lightning development channel during wind turbine lightning strikes, which limits the observation and analysis of lightning channels in wind farms. Summary of the Invention

[0006] To address the shortcomings of existing technologies, the purpose of this application is to provide a method for detecting lightning strikes on wind turbines based on radio frequency electromagnetic detection. This method aims to solve the problem that existing positioning systems, when facing wind farm scenarios, are unable to achieve accurate observation of lightning channels in wind farms due to the influence of antenna structure and arrangement methods.

[0007] To achieve the above objectives, in a first aspect, this application provides a method for detecting lightning strikes on wind turbines based on radio frequency electromagnetic detection, comprising: obtaining the initial positions and initial directional antenna directions of multiple radio frequency observation stations based on signal-to-noise ratio constraints, elevation angle constraints, and number constraints; iterating based on simulated lightning radiation source signals, initial positions, and initial directional antenna directions to obtain target positions and target directional antenna directions, and arranging each radio frequency observation station according to the target positions and target directional antenna directions; aligning multiple radio frequency observation stations with a unified time base, and extracting the arrival times of radio frequency pulses of multiple radio frequency observation stations by detecting rising edges; obtaining the arrival time differences of multiple radio frequency observation stations based on the arrival times, obtaining a three-dimensional source point set through a spherical propagation time difference positioning model, and obtaining the near-ground channel morphology of lightning strikes based on the three-dimensional source point set.

[0008] In one embodiment, based on the signal-to-noise ratio (SNR) constraints, elevation angle constraints, and number constraints of multiple RF observation stations, the initial positions and initial directional antenna directions of multiple RF observation stations are obtained. Prior to this, the method further includes: constructing a received power model of the RF observation station in any direction according to the Fries transmission equation; obtaining the SNR constraints of the RF observation station based on the received power model and noise power; obtaining the altitude range of the RF observation station and obtaining the elevation angle constraint based on the maximum elevation angle difference and the altitude range of the RF observation station; and obtaining the number constraints based on the geometric relationship and signal quality of multiple RF observation stations.

[0009] In one embodiment, based on the signal-to-noise ratio (SNR) constraint, elevation angle constraint, and number constraint of the radio frequency (RF) observation stations, the initial positions and initial directional antenna pointing of multiple RF observation stations are obtained, including: delineating a feasible deployment area for RF observation stations according to the geographical location of the wind turbine, surrounding topography, and distribution of potential obstacles; determining the total number of RF observation stations based on the number constraint; within the feasible deployment area, with the goal of satisfying the SNR and elevation angle constraints and optimizing the spatial geometric distribution of the stations as a strategy, initial position placement planning is performed to generate an initial position coordinate set for multiple RF observation stations; based on the RF observation stations in the initial position coordinate set, the initial pointing angle of their directional antennas is calculated according to their relative orientation to the wind turbine to ensure that the main lobe of the antenna can cover the wind turbine.

[0010] In one embodiment, the process involves iteratively obtaining the target position and target directional antenna direction based on the simulated lightning radiation source signal, initial position, and initial directional antenna direction, and then deploying each radio frequency (RF) observation station using the target position and target directional antenna direction. This includes: obtaining the arrival time uncertainty of multiple RF observation stations based on the signal-to-noise ratio of the simulated lightning radiation source signal, and determining it as a time delay error; obtaining the positioning accuracy of multiple RF observation stations based on the time delay error and geometric precision factor; building an optimization index model based on the positioning accuracy, coverage, and number of RF observation stations; iterating the initial position and initial directional antenna direction based on the optimization index model to obtain the target position and target directional antenna direction, and then deploying each RF observation station using the target position and target directional antenna direction.

[0011] In one embodiment, the radio frequency (RF) observation station includes: a directional antenna, a filtering and amplification module, a sampling and timing module, and a trigger slicing module. The system performs unified time base alignment on multiple RF observation stations and extracts the arrival times of RF pulses from multiple RF observation stations by detecting rising edges to obtain the arrival time difference of multiple RF observation stations. This includes: performing unified time base alignment on multiple RF observation stations based on the sampling and timing module; simultaneously acquiring lightning radiation source signals based on the directional antenna and transmitting them to the filtering and amplification module; performing bandpass filtering and low-noise processing on the received signals based on the filtering and amplification module to obtain the processed signals and transmitting them to the sampling and timing module; converting the processed signals into digital signals based on the sampling and timing module and providing GNSS timing; and detecting the acquired digital signals based on the trigger slicing module, and when the rising edge of the pulse meets a slope threshold and exceeds a set threshold, extracting the data waveforms before and after the corresponding time from the circular buffer as an event slice output.

[0012] In one embodiment, the arrival time difference (OTD) of multiple radio frequency observation stations is obtained based on the arrival time, and a three-dimensional source point set is obtained through a spherical propagation OTD localization model. The near-ground channel morphology of the lightning strike is then obtained based on the three-dimensional source point set. This includes: establishing a hyperbolic equation constrained by the OTD based on a spherical wave propagation model, identifying it as a spherical propagation OTD localization model; eliminating the nonlinear terms of the spherical propagation OTD localization model, obtaining a relationship model between the radiation source location and the initial radiation time, and solving for the initial estimate of the radiation source location; using the initial estimate as the starting point for iteration, constructing an objective function with the goal of minimizing the distance difference residual, and solving it using a nonlinear least squares method to obtain the three-dimensional source point set of the radiation source; and obtaining the near-ground channel morphology of the lightning strike based on the three-dimensional source point set.

[0013] A lightning strike detection system for wind turbines based on radio frequency electromagnetic detection includes:

[0014] The computing unit is used to obtain the initial position and initial directional antenna pointing of multiple radio frequency observation stations based on the signal-to-noise ratio constraint, elevation angle constraint and number constraint of the radio frequency observation stations; it is also used to iterate based on the simulated lightning radiation source signal, the initial position and the initial directional antenna pointing to obtain the target position and the target directional antenna pointing.

[0015] Multiple radio frequency observation stations are used to extract the arrival time of radio frequency pulses from multiple radio frequency observation stations by detecting the rising edge after unified time base alignment.

[0016] The detection device is used to obtain the time difference of arrival of multiple radio frequency observation stations based on the arrival time, obtain a three-dimensional source point set through the time difference of arrival positioning model of spherical propagation, and obtain the near-ground channel morphology of lightning strike based on the three-dimensional source point set.

[0017] A method for detecting lightning strikes on wind turbines based on radio frequency electromagnetic detection includes: obtaining the initial positions and initial directional antenna directions of multiple radio frequency observation stations based on signal-to-noise ratio constraints, elevation angle constraints, and number constraints; iterating based on simulated lightning radiation source signals, initial positions, and initial directional antenna directions to obtain target positions and target directional antenna directions, and arranging each radio frequency observation station according to the target positions and target directional antenna directions; aligning multiple radio frequency observation stations with a unified time base, and extracting the arrival times of radio frequency pulses of multiple radio frequency observation stations by detecting rising edges; obtaining the arrival time difference of multiple radio frequency observation stations based on the arrival times, obtaining a three-dimensional source point set through a spherical propagation time difference of arrival positioning model, and obtaining the near-ground channel morphology of lightning strikes based on the three-dimensional source point set.

[0018] In one embodiment, based on the signal-to-noise ratio (SNR) constraints, elevation angle constraints, and number constraints of multiple RF observation stations, the initial positions and initial directional antenna directions of multiple RF observation stations are obtained. Prior to this, the method further includes: constructing a received power model of the RF observation station in any direction according to the Fries transmission equation; obtaining the SNR constraints of the RF observation station based on the received power model and noise power; obtaining the altitude range of the RF observation station and obtaining the elevation angle constraint based on the maximum elevation angle difference and the altitude range of the RF observation station; and obtaining the number constraints based on the geometric relationship and signal quality of multiple RF observation stations.

[0019] In one embodiment, based on the signal-to-noise ratio (SNR) constraint, elevation angle constraint, and number constraint of the radio frequency (RF) observation stations, the initial positions and initial directional antenna pointing of multiple RF observation stations are obtained, including: delineating a feasible deployment area for RF observation stations according to the geographical location of the wind turbine, surrounding topography, and distribution of potential obstacles; determining the total number of RF observation stations based on the number constraint; within the feasible deployment area, with the goal of satisfying the SNR and elevation angle constraints and optimizing the spatial geometric distribution of the stations as a strategy, initial position placement planning is performed to generate an initial position coordinate set for multiple RF observation stations; based on the RF observation stations in the initial position coordinate set, the initial pointing angle of their directional antennas is calculated according to their relative orientation to the wind turbine to ensure that the main lobe of the antenna can cover the wind turbine.

[0020] In one embodiment, the process involves iteratively obtaining the target position and target directional antenna direction based on the simulated lightning radiation source signal, initial position, and initial directional antenna direction, and then deploying each radio frequency (RF) observation station using the target position and target directional antenna direction. This includes: obtaining the arrival time uncertainty of multiple RF observation stations based on the signal-to-noise ratio of the simulated lightning radiation source signal, and determining it as a time delay error; obtaining the positioning accuracy of multiple RF observation stations based on the time delay error and geometric precision factor; building an optimization index model based on the positioning accuracy, coverage, and number of RF observation stations; iterating the initial position and initial directional antenna direction based on the optimization index model to obtain the target position and target directional antenna direction, and then deploying each RF observation station using the target position and target directional antenna direction.

[0021] In one embodiment, the radio frequency (RF) observation station includes: a directional antenna, a filtering and amplification module, a sampling and timing module, and a trigger slicing module. The system performs unified time base alignment on multiple RF observation stations and extracts the arrival times of RF pulses from multiple RF observation stations by detecting rising edges to obtain the arrival time difference of multiple RF observation stations. This includes: performing unified time base alignment on multiple RF observation stations based on the sampling and timing module; simultaneously acquiring lightning radiation source signals based on the directional antenna and transmitting them to the filtering and amplification module; performing bandpass filtering and low-noise processing on the received signals based on the filtering and amplification module to obtain the processed signals and transmitting them to the sampling and timing module; converting the processed signals into digital signals based on the sampling and timing module and providing GNSS timing; and detecting the acquired digital signals based on the trigger slicing module, and when the rising edge of the pulse meets a slope threshold and exceeds a set threshold, extracting the data waveforms before and after the corresponding time from the circular buffer as an event slice output.

[0022] In one embodiment, the arrival time difference (OTD) of multiple radio frequency observation stations is obtained based on the arrival time, and a three-dimensional source point set is obtained through a spherical propagation OTD localization model. The near-ground channel morphology of the lightning strike is then obtained based on the three-dimensional source point set. This includes: establishing a hyperbolic equation constrained by the OTD based on a spherical wave propagation model, identifying it as a spherical propagation OTD localization model; eliminating the nonlinear terms of the spherical propagation OTD localization model, obtaining a relationship model between the radiation source location and the initial radiation time, and solving for the initial estimate of the radiation source location; using the initial estimate as the starting point for iteration, constructing an objective function with the goal of minimizing the distance difference residual, and solving it using a nonlinear least squares method to obtain the three-dimensional source point set of the radiation source; and obtaining the near-ground channel morphology of the lightning strike based on the three-dimensional source point set.

[0023] Secondly, this application provides a lightning strike detection system for wind turbines based on radio frequency electromagnetic detection, comprising: a computing device for acquiring the initial position and initial directional antenna pointing of multiple radio frequency observation stations based on signal-to-noise ratio constraints, elevation angle constraints, and number constraints of the radio frequency observation stations; and for iterating based on simulated lightning radiation source signals, initial positions, and initial directional antenna pointing to acquire the target position and target directional antenna pointing; multiple radio frequency observation stations for extracting the arrival time of radio frequency pulses of multiple radio frequency observation stations by detecting rising edges after unified time base alignment; and a detection device for acquiring the arrival time difference of multiple radio frequency observation stations based on the arrival time, acquiring a three-dimensional source point set through a spherical propagation time difference of arrival positioning model, and acquiring the near-ground channel morphology of lightning strikes based on the three-dimensional source point set.

[0024] Thirdly, this application provides an electronic device including a memory and one or more processors; the memory is coupled to the one or more processors, the memory is used to store computer program code, the computer program code including computer instructions; the one or more processors invoke the computer instructions to cause the electronic device to perform the method described in the first aspect.

[0025] Fourthly, this application provides a computer-readable storage medium including instructions that, when executed on an electronic device, cause the electronic device to perform the method described in the first aspect.

[0026] Fifthly, this application provides a computer program product, including a computer program or instructions that, when run on an electronic device, cause the electronic device to perform the method described in the first aspect.

[0027] It is understood that the beneficial effects of the second to fifth aspects mentioned above can be found in the relevant descriptions in the first aspect mentioned above, and will not be repeated here.

[0028] Overall, the technical solutions conceived in this application have the following beneficial effects compared with the prior art:

[0029] This application addresses the challenges of low signal-to-noise ratio (SNR) and susceptibility to multipath effects in wind farm environments with strong interference by employing SNR, elevation angle, and site number constraints to plan the initial positions and directional antenna pointing of multiple directional radio frequency (RF) observation sites. This achieves selective focusing on the target airspace, laying a high SNR data foundation for subsequent high-precision detection. Furthermore, iterative optimization using simulated lightning radiation source signals yields the final parameters of the site and target directional antennas, overcoming the difficulty of adaptive optimization in focusing on key areas using traditional site deployment methods. This enhances monitoring of lightning-prone areas of critical wind turbine components. Next, by unifying time base alignment and utilizing the high gain and strong anti-interference characteristics of directional antennas, the application accurately extracts the pulse arrival time, solving the problem of large timing errors in low SNR environments and providing reliable time data for high-precision positioning. Finally, based on the precise time difference of arrival, a spherical propagation model is used to invert the three-dimensional source point set and reconstruct the near-ground channel morphology of lightning strikes.

[0030] Compared with existing technologies, this system systematically solves the problem of interference from the complex electromagnetic and terrain environment of wind farms on lightning detection by adopting a complete technology chain from site optimization and high-precision time-frequency measurement to positioning algorithms. The final result is not isolated point positioning, but high-resolution, high-confidence three-dimensional imaging of the near-ground channel morphology of lightning strikes, thus providing unprecedentedly detailed observational data for lightning protection design, risk assessment, and accident analysis of wind turbines. Attached Figure Description

[0031] Figure 1 This is one of the flowcharts of the lightning strike detection method for wind turbines based on radio frequency electromagnetic detection provided in the embodiments of this application;

[0032] Figure 2 This application provides 300MHz band radiation patterns for omnidirectional and directional antennas in embodiments of this application.

[0033] Figure 3 This is the second flowchart of the lightning strike detection method for wind turbine generators based on radio frequency electromagnetic detection provided in the embodiments of this application;

[0034] Figure 4 These are two-dimensional and three-dimensional schematic diagrams of the radio frequency observation site layout provided in the embodiments of this application;

[0035] Figure 5 This is the third flowchart of the lightning strike detection method for wind turbine generators based on radio frequency electromagnetic detection provided in the embodiments of this application;

[0036] Figure 6This is an internal flowchart of the radio frequency observation station provided in the embodiments of this application;

[0037] Figure 7 This is the fourth flowchart of the lightning strike detection method for wind turbine generators based on radio frequency electromagnetic detection provided in the embodiments of this application;

[0038] Figure 8 This is a structural block diagram of a wind turbine lightning strike detection system based on radio frequency electromagnetic detection provided in an embodiment of this application;

[0039] Figure 9 This is a schematic diagram of the structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0040] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0041] In this article, the term "and / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. The symbol " / " in this article indicates that the related objects are in an "or" relationship; for example, A / B means A or B.

[0042] The terms "first" and "second," etc., used in the specification and claims herein are used to distinguish different objects, not to describe a specific order of objects. For example, "first response message" and "second response message," etc., are used to distinguish different response messages, not to describe a specific order of response messages.

[0043] In the embodiments of this application, the terms "exemplary" or "for example" are used to indicate that something is an example, illustration, or description. Any embodiment or design that is described as "exemplary" or "for example" in the embodiments of this application should not be construed as superior or more advantageous than other embodiments or designs. Specifically, the use of the terms "exemplary" or "for example" is intended to present the relevant concepts in a specific manner.

[0044] In the description of the embodiments of this application, unless otherwise stated, "multiple" means two or more, for example, multiple processing units means two or more processing units, multiple elements means two or more elements, etc.

[0045] Current lightning monitoring of wind farms largely focuses on wide-area monitoring, using omnidirectional antennas with relatively high elevation angles. The pursuit of bandwidth and directional balance in omnidirectional antennas results in insufficient signal-to-noise ratios in local observation directions, making it difficult to receive radio frequency signals from near-ground units (wind turbines). When the signal source is located at the edge of the detection range, it also increases the positioning error. The complex environment of wind farms can also affect angle calculations, impacting subsequent analysis.

[0046] Specifically, when the source point is located at the edge of the observation network or at a large horizontal distance from the observation station, it affects the geometric conditions for calculating the Time Difference of Arrival (TDOA) / Time of Arrival (TOA), significantly amplifying the Geometric Dilution of Precision (GDOP) and leading to positioning errors. Using an omnidirectional antenna simultaneously receives radiation from distant clouds and environmental noise, failing to highlight the near-ground radio frequency pulse during wind turbine lightning strikes, resulting in insufficient signal-to-noise ratio and difficulty in accurately acquiring the radiation point set near the wind turbine, thus hindering the accurate understanding of near-ground details of cloud-to-ground lightning. Furthermore, the wind turbine tower, nacelle, and blades may reflect, diffract, and scatter radio frequency signals, which, when received by the omnidirectional antenna, can affect positioning and even produce false radiation sources.

[0047] Therefore, for complex wind farm environments with dense equipment and varied terrain, it is necessary to improve the resolution and positioning accuracy of the near-ground end of the ground flash channel to provide a theoretical basis for the lightning protection design and physical process research of wind turbine units.

[0048] Based on this, this application proposes a method for detecting lightning strikes on wind turbines based on radio frequency electromagnetic detection. Please refer to... Figure 1 , Figure 1 This is one of the flowcharts of the lightning strike detection method for wind turbine generators based on radio frequency electromagnetic detection provided in the embodiments of this application.

[0049] In this embodiment, the method for detecting lightning strikes on wind turbines based on radio frequency electromagnetic detection includes steps S10 to S40.

[0050] Step S10: Based on the signal-to-noise ratio constraints, elevation angle constraints, and number constraints of the radio frequency observation stations, obtain the initial positions and initial directional antenna directions of multiple radio frequency observation stations.

[0051] It should be noted that the initial location of the radio frequency (RF) observation station refers to the geographical coordinates of the station, determined during the initial planning and deployment phase of the RF observation system, based on a comprehensive consideration of various factors, and before any fine-tuning or optimization. The RF observation station must include an antenna for receiving radio frequency electromagnetic signals from lightning radiation sources.

[0052] It should be noted that the directional antenna chosen for the radio frequency observation station in this embodiment is for near-ground lightning observation purposes. Specifically, strong radio frequency radiation is generated during lightning flashes. Using radio frequency radiation for positioning allows for a more detailed description of the discharge process. When monitoring lightning at wind farms, the signal frequency band is wide and the positioning accuracy requirement is high. If an omnidirectional antenna is used, it will receive radiation from distant clouds and environmental noise along with the radio frequency radiation pulses. During a lightning strike on a wind turbine, the signal-to-noise ratio at the near-ground end is insufficient, making it difficult to accurately obtain the radiation point set near the wind turbine and to grasp the details of near-ground lightning strikes effectively.

[0053] For example, the radio frequency observation station of this application uses a single log-periodic dipole array (LPDA) antenna as the directional receiving antenna. An LPDA is a broadband directional antenna composed of a series of dipoles of different lengths, arranged sequentially along the main support rod. The dipoles of different lengths have different resonant frequencies, thus enabling the reception of broadband short-pulse radiated signals ranging from tens to hundreds of MHz. Since these dipoles are arranged in the same direction, phase differences in other directions will cause the signals to cancel each other out, thus only receiving signals in the direction of the main lobe, achieving directional reception.

[0054] For a detailed comparison, please refer to [link / reference]. Figure 2 , Figure 2 This is the radiation pattern of the omnidirectional and directional antennas in the 300MHz band provided in the embodiments of this application. Figure 2 In the antenna pattern, it can be clearly seen that compared with omnidirectional antennas, directional antennas can increase the gain in the main lobe direction, making the signal-to-noise ratio higher in the target direction; at the same time, they can reduce the gain in other directions, reduce external interference, and meet the requirements for near-ground observation.

[0055] Understandably, a crucial factor determining the sampling performance of a directional antenna is its pointing direction. Therefore, in addition to acquiring the initial location of the radio frequency (RF) observation station, it is also necessary to obtain the pointing direction of the directional antenna. The initial pointing direction refers to the pointing direction of the directional antenna during the initial planning and deployment phase of the RF observation system, before it has undergone fine-tuning and optimization.

[0056] Understandably, in this embodiment, the initial positions and initial directional antenna directions of multiple radio frequency (RF) observation stations are obtained through signal-to-noise ratio (SNR), elevation angle, and quantity constraints. This is because SNR, elevation angle, and quantity constraints impose requirements on RF electromagnetic observation from three dimensions: signal quality, spatial coverage, and system performance. Specifically, the SNR constraint directly corresponds to the energy detectability threshold of lightning radiation signals; the elevation angle constraint is essentially a geometric selection of the height layer of the lightning discharge channel; and the quantity constraint reflects the trade-off between system necessity and economy.

[0057] Therefore, signal-to-noise ratio constraints, elevation angle constraints, and quantity constraints need to be obtained before performing step S10.

[0058] In one specific implementation, the steps for obtaining signal-to-noise ratio constraints, elevation angle constraints, and quantity constraints are as follows: Based on the Friesian transmission equation, a received power model of the radio frequency observation station in any direction is constructed; based on the received power model and noise power, the signal-to-noise ratio constraints of the radio frequency observation station are obtained; the altitude range of the radio frequency observation station is obtained, and the elevation angle constraint is obtained based on the maximum elevation angle difference and the altitude range of the radio frequency observation station; based on the geometric relationship and signal quality of multiple radio frequency observation stations, quantity constraints are obtained.

[0059] It is understandable that the lightning flash process that emits radio frequency radiation can be equated to a signal source, which can radiate the equivalent isotropically radiated power (EIRP). According to the Friesian transmission equation, when the assumptions of far-field, polarization matching, and line-of-sight propagation are satisfied, the detection station in the direction... Received power model P r As shown in equation (1) below.

[0060] (1)

[0061] Among them, P t G t This is the equivalent isotropic radiation power; This indicates the gain of the probe station in that direction; Free space loss factor, R represents the wavelength. i This indicates the distance of the radiation source from the station.

[0062] As is understandable, the signal-to-noise ratio (SNR) is the ratio of signal power to noise power. Signal power can be equivalently represented by the antenna's received power, and the noise power can be expressed as P. n , will P r Substituting the values, the signal-to-noise ratio constraint is obtained as shown in equation (2).

[0063] (2)

[0064] Therefore, it can be seen that by aligning the antenna's main lobe obliquely with the wind farm, the antenna's gain in that direction will increase. This can be improved. The antenna gain is very low in directions other than the main lobe (lateral and rearward). By pointing the main lobe towards the wind farm, the antenna is essentially aligning its low-gain lateral and rearward directions with other potential sources of interference (such as industrial noise from wind turbine operation, irrelevant communication signals, etc.). This significantly attenuates the background noise power P received from these non-target directions. n This reduces side and rear signal interference, thereby achieving a higher SNR in the target direction.

[0065] Specifically, noise power P n The background noise power can be obtained by measuring it with a receiver during system initialization or during periods of no signal. The average power value actually measured and recorded through the same receiving channel during periods without lightning activity or other significant signal sources is the noise power of the site.

[0066] In addition, constructing a high-resolution near-ground region requires selecting appropriate elevation angle constraints. Let the wind turbine nacelle height be H and the rotor radius be R; the elevation range for a single-station observation can be approximated as... The site antenna height is Hs, and the horizontal distance between it and the wind turbine is Ri. The elevation angle for aligning the LPDA main lobe with the midpoint of the interval is... As shown in equation (3) below; maximum elevation angle difference It is shown in the following formula (4).

[0067] (3)

[0068] (4)

[0069] It is understandable that, since it is necessary to ensure that the half-power beamwidth (HPBW) of a single site's LPDA covers the entire altitude range, the elevation angle constraint is as shown in equation (5).

[0070] (5)

[0071] For example, taking a 10MW offshore wind turbine generator set in Fujian as an example, the hub center of the generator set is approximately 115 meters above sea level, and the turbine rotor diameter is 185 meters. The typical vertical boom angle (HPBW) for an LPDA (Low Power Distribution Aspect Ratio) is 60 to 80 degrees. If the distance between wind turbines is 800 to 1500 meters, the elevation angle can be estimated. It is approximately 15 to 20 degrees Celsius.

[0072] Furthermore, constructing a high-resolution near-ground region requires a sufficient number of detection stations to achieve full coverage of the target. By arranging stations around the wind farm at as dispersed azimuth angles as possible on the horizontal plane, a higher synthetic SNR can be obtained. In this case, the TDOA of the signal will simultaneously constrain the radiation source position from multiple directions, better reducing GDOP, resulting in a more stable and accurate three-dimensional position.

[0073] For example, when using a spherical wave model to solve the problem, for each arrival time difference... A possible solution location exists on a hyperboloid. The intersection of two hyperboloids forms a space curve, and at least three hyperboloids are required for a solution; therefore, the number of stations is constrained to be no less than four. Specifically, if other models are used for the solution, corresponding number constraints can be obtained.

[0074] For example, after obtaining the three constraints, the regional electromagnetic environment can be analyzed through signal-to-noise ratio modeling. Combining lightning radiation characteristics and background noise distribution, candidate site areas that meet the signal detection threshold can be screened. Next, elevation angle geometric constraints are implemented. Based on the target observation height layer range, trigonometric relationships are used to determine the compliant elevation angle range of the antenna's main beam, eliminating candidate directions that deviate from this range. Finally, an incremental coverage method is used to control the number of sites, iteratively selecting sites from the candidate set that maximize the overall coverage quality until the regional coverage requirements are met or the preset maximum number of sites is reached.

[0075] In one specific implementation, this embodiment provides a detailed method for executing step S10. Please refer to... Figure 3 , Figure 3 This is the second schematic flowchart of the lightning strike detection method for wind turbines based on radio frequency electromagnetic detection provided in this application embodiment. In this embodiment, step S10 includes steps S11 to S14.

[0076] Step S11: Delineate the feasible deployment area of ​​radio frequency observation stations based on the geographical location of the wind turbine, the surrounding topography and the distribution of potential obstacles.

[0077] Step S12: Determine the total number of radio frequency observation stations based on quantity constraints.

[0078] Understandably, delineating feasible deployment areas for radio frequency observation stations requires constructing a multi-dimensional geospatial framework. First, using wind turbine coordinates as the core, a high-precision digital elevation model and land cover data are integrated, along with 3D spatial information on potential obstacles, forming a comprehensive database encompassing terrain features, land cover distribution, and electromagnetic environment characteristics. Then, geometric calculations assess the obstruction risk of the line-of-sight channel, automatically selecting basic areas that meet propagation conditions. Finally, considering electromagnetic compatibility requirements, the impact range of interference sources to be avoided is marked, ultimately generating a continuous and deployable spatial block scheme.

[0079] Understandably, as mentioned above, after determining the solution for the spherical wave model, the number constraint is no less than 4. It is advisable to prioritize setting it to 4 radio frequency observation stations, and confirm the final number according to the incremental coverage principle.

[0080] Step S13: Within the feasible deployment area, with the goal of satisfying the signal-to-noise ratio constraint and elevation angle constraint, and with the strategy of optimizing the spatial geometric distribution of the stations, initial location planning is carried out to generate an initial location coordinate set for multiple radio frequency observation stations.

[0081] Understandably, within a feasible deployment area, the primary consideration is the signal-to-noise ratio (SNR) constraint. The received signal strength at each candidate point is calculated using an electromagnetic propagation model to filter out areas that meet the minimum SNR threshold. Simultaneously, considering the elevation angle constraint, geometric relationships are used to determine the minimum elevation angle range required for the antenna's main lobe to cover the wind turbine, eliminating candidate points that do not meet the line-of-sight requirements. Based on this, a mathematical model is constructed with the goal of optimizing the spatial geometric distribution of sites, aiming to achieve coverage uniformity, minimize redundant coverage, and shorten baseline length. Then, a particle swarm optimization algorithm is used for global search, iteratively adjusting site positions until the objective function converges to the optimal solution. Finally, an initial set of site coordinates is generated, containing the final number of sites. Each coordinate satisfies SNR ≥ the threshold, elevation angle meets observation requirements, and the spatial distribution between sites exhibits an approximately uniform triangular grid characteristic, providing a basic configuration for high coverage efficiency in subsequent precise deployment.

[0082] Step S14: Based on the radio frequency observation stations in the initial position coordinate set, calculate the initial pointing angle of their directional antennas according to their relative orientation to the wind turbines, so as to ensure that the main lobe of the antenna can cover the wind turbines.

[0083] Understandably, based on the initial position coordinate set generated in step S13, the initial pointing angle of the antenna at each site is determined using a three-dimensional spatial vector analysis method: First, a spatial vector is established with the site location as the origin and the wind turbine location as the endpoint. The azimuth angle of this vector projected onto the horizontal plane is calculated using the arctangent function, serving as the reference value for the antenna's horizontal pointing. Simultaneously, the elevation angle of the line connecting the site and the turbine is calculated using the vector dot product formula, and the pointing offset required for main lobe coverage is determined in conjunction with the antenna beamwidth parameters. Finally, the pointing angle parameter set for each site is output, ensuring that the main lobe of all site antennas can stably cover the target wind turbine, and that the pointing angle deviation is controlled within ±1 degree, providing an accurate initial pointing reference for subsequent beamforming optimization. Please refer to... Figure 4 , Figure 4 These are two-dimensional and three-dimensional schematic diagrams of the radio frequency observation site layout provided in the embodiments of this application. One possible layout is illustrated.

[0084] Step S20: Based on the simulated lightning radiation source signal, initial position and initial directional antenna pointing, iterate to obtain the target position and target directional antenna pointing, and arrange each radio frequency observation station according to the target position and target directional antenna pointing.

[0085] Understandably, after determining the initial site location and antenna orientation, the actual site layout and site selection need to take into account environmental conditions and continuously improve through iterative optimization to enhance resolution and accuracy, and determine the optimal configuration of each site, namely the target location and the target directional antenna orientation.

[0086] In one specific implementation, this embodiment provides a detailed method for executing step S20. Please refer to... Figure 5 , Figure 5 This is the third flowchart of the lightning strike detection method for wind turbine generators based on radio frequency electromagnetic detection provided in the embodiments of this application.

[0087] In this embodiment, step S20 includes steps S21 to S24.

[0088] Step S21: Based on the signal-to-noise ratio of the simulated lightning radiation source signal, obtain the uncertainty of the arrival time of multiple radio frequency observation stations and determine it as a time delay error.

[0089] Understandably, for potential lightning radiation sources within the wind farm, the Monte Carlo method is introduced to randomly sample lightning sources within the target area, simulating the signal strength and arrival time received by each observation station when lightning radiates at different locations. The uncertainty of the arrival time can be estimated based on the signal-to-noise ratio of the received signal, i.e., the time delay error is shown in equation (6) below.

[0090] (6)

[0091] Where B represents the effective bandwidth.

[0092] Step S22: Based on the time delay error and geometric accuracy factor, obtain the positioning accuracy of multiple radio frequency observation stations.

[0093] It is understandable that the latency error of each station is considered. After substituting into the positioning equation, the overall positioning accuracy can be calculated using the geometric precision factor. As shown in equation (7) below.

[0094] (7)

[0095] Step S23: Based on positioning accuracy, coverage, and the number of radio frequency observation stations, build an optimization index model.

[0096] Step S24: Iterate the initial position and initial directional antenna pointing based on the optimized index model to obtain the target position and target directional antenna pointing, and arrange each radio frequency observation station according to the target position and target directional antenna pointing.

[0097] It is understandable that optimizing the index model requires considering positioning accuracy, coverage and the number of radio frequency observation stations to achieve the optimal result. Therefore, the optimization index model J is shown in equation (8).

[0098] (8)

[0099] in, C represents the weighting coefficient; C represents the coverage rate, which is the proportion that can be detected by four or more sites; D represents the number of sites, deployment costs, etc.

[0100] Based on this, the azimuth and elevation angles of each station are adjusted within a small range, and J is recalculated after each adjustment. If the new parameter combination can reduce the value of J, the adjustment is recorded and retained. As the number of iterations increases, a stable optimal configuration is obtained.

[0101] Step S30: Perform unified time base alignment on multiple radio frequency observation stations, and extract the arrival time of radio frequency pulses of multiple radio frequency observation stations by detecting rising edges.

[0102] For specific details regarding the sampling process at the radio frequency observation site, please refer to [link / reference]. Figure 6 , Figure 6 This is an internal flowchart of a radio frequency (RF) observation station provided in an embodiment of this application. Based on the flowchart, the RF observation station should include: a directional antenna, a filtering and amplification module, a sampling and timing module, and a trigger slicing module.

[0103] Understandably, before sampling begins, it is necessary to perform unified time base alignment of multiple radio frequency observation stations based on the sampling timing module to avoid initial errors in the arrival times sampled by multiple radio frequency observation stations.

[0104] Understandably, multiple radio frequency observation stations, based on directional antennas, simultaneously collect lightning radiation source signals and transmit them to the filtering and amplification module. In near-ground lightning detection at wind farms and other locations, the 30MHz to 80MHz frequency band is often chosen as the receiving frequency. Signal strength is relatively high in this band, and hardware design and implementation are less complex, with less external electromagnetic interference.

[0105] Understandably, the receiving signal is processed by band-pass filtering and low-noise methods using the filtering and amplification module to obtain the processed signal, which is then transmitted to the sampling and timing module. Upon receiving the signal, a band-pass filter (BPF) and a low-noise amplifier (LNA) are sequentially connected to ensure that out-of-band electromagnetic interference is suppressed without distortion, thereby improving the signal-to-noise ratio and waveform amplitude.

[0106] Based on the sampling and timing module, the processed signal is converted into a digital signal and timed via GNSS. A high-speed analog-to-digital converter converts the analog signal into a digital signal. To achieve nanosecond-level accuracy of TOA and precise voltage quantization, the sampling rate is set to 100MHz and the quantization bit depth to 12 bits. A 10MHz reference and 1PPS pulse are output from the GNSS, laying the foundation for subsequent nanosecond-level synchronization between multiple stations and ensuring the accuracy of TDOA.

[0107] Understandably, based on the trigger slicing module, the acquired digital signal is detected. When the rising edge of the pulse meets the slope threshold and exceeds a set threshold, the data waveform before and after the corresponding moment of the digital signal is extracted from the circular buffer as an event slice output. This not only minimizes the possibility of data loss but also reduces the pressure on data backhaul and prevents the consumption of more bandwidth.

[0108] Step S40: Based on the arrival time, obtain the arrival time difference of multiple radio frequency observation stations, obtain a three-dimensional source point set through the spherical propagation time difference of arrival positioning model, and obtain the near-ground channel morphology of lightning strike based on the three-dimensional source point set.

[0109] It is understandable that multiple stations simultaneously sample the same radio frequency radiation, and the arrival time t of each station is obtained by detecting the rising edge. i . Select radio frequency observation station No. 1 as a reference, and the arrival time difference is shown in the following formula (9).

[0110] (9)

[0111] Understandably, after obtaining the time difference of arrival, a nonlinear positioning equation can be established based on the spherical wave propagation model, and the three-dimensional spatial coordinates of the lightning discharge source can be calculated using an iterative optimization algorithm. Finally, spatial clustering and curve fitting are performed on the positioning points of the continuous time series to reconstruct the topology of the lightning channel containing branch structures.

[0112] In one specific implementation, this embodiment provides a detailed method for executing step S40. Please refer to... Figure 7 , Figure 7This is the fourth flowchart illustrating the lightning strike detection method for wind turbines based on radio frequency electromagnetic detection provided in this application embodiment. In this embodiment, step S40 includes steps S41 to S44.

[0113] Step S41: Based on the spherical wave propagation model, establish a hyperbolic equation constrained by the time difference of arrival, and determine it as a spherical propagation time difference of arrival positioning model.

[0114] Understandably, 3D localization is performed using a spherical propagation model under the near-Earth propagation assumption. This can be achieved by setting the location of the unknown source. The coordinates of each station are r i If the speed of light is c, then the hyperbolic equation, i.e. the time difference of arrival positioning model for spherical propagation, is shown in equation (10).

[0115] (10)

[0116] Step S42: Eliminate the nonlinear term in the time difference of arrival positioning model for spherical propagation, obtain the relationship model between the radiation source location and the initial time of radiation, and calculate the initial estimate of the radiation source location.

[0117] Understandably, since the equations involve nonlinear constraints, initial values ​​need to be determined before nonlinear optimization can be performed. By squaring and subtracting the original equations, the constraints can be eliminated. Related items, resulting in a set The linear equations, i.e. the relationship between the location of the radiation source and the initial moment of radiation, are shown in equation (11) below.

[0118] (11)

[0119] Understandably, an approximate solution, namely the initial estimate of the source position, can be obtained through the least squares method, which serves as the starting point for the nonlinear iteration.

[0120] Step S43: Starting from the initial estimate, construct an objective function with the goal of minimizing the distance difference residual, and solve it using the nonlinear least squares method to obtain the three-dimensional source point set of the radiation source.

[0121] It is understandable to write each constraint as a residual. The expression is shown in equation (12).

[0122] (12)

[0123] The objective function is constructed using the equal-weight least squares method as shown in equation (13).

[0124] (13)

[0125] Understandably, in order to minimize To make the geometric distance difference consistent with the measured time difference multiplied by the speed of light, a nonlinear least squares method (Levenberg-Marquardt, LM) is introduced. The Jacobian matrix H is written as shown in equation (14). The i-th row is the difference between the unit direction vectors in the two directions, as shown in equation (15). The final LM iterative update formula is shown in equation (16).

[0126] (14)

[0127] (15)

[0128] (16)

[0129] in, f is the position increment for this iteration; f is the residual vector, derived from... The composition is determined by directly substituting the current solution s into the distance difference formula. After multiple iterations, the stable solution s obtained is the three-dimensional spatial location of the lightning radiation source. Finally, the rationality of the layout can be evaluated by GDOP, and the calculation formula is shown in the following formula (17). Finally, the three-dimensional source point set of the radiation source is obtained.

[0130] (17)

[0131] Step S44: Obtain the near-ground channel morphology of lightning strike based on the three-dimensional source point set.

[0132] Understandably, by sorting the radiation sources by time and drawing a three-dimensional set of radiation sources, we can obtain the near-ground end channel morphology of ground flashes, which facilitates subsequent analysis of the flashover points and channel evolution characteristics.

[0133] This application addresses the challenges of low signal-to-noise ratio (SNR) and susceptibility to multipath effects in wind farm environments with strong interference by employing SNR, elevation angle, and site number constraints to plan the initial positions and directional antenna pointing of multiple directional radio frequency (RF) observation sites. This achieves selective focusing on the target airspace, laying a high SNR data foundation for subsequent high-precision detection. Furthermore, iterative optimization using simulated lightning radiation source signals yields the final parameters of the site and target directional antennas, overcoming the difficulty of adaptive optimization in focusing on key areas using traditional site deployment methods. This enhances monitoring of lightning-prone areas of critical wind turbine components. Next, by unifying time base alignment and utilizing the high gain and strong anti-interference characteristics of directional antennas, the application accurately extracts the pulse arrival time, solving the problem of large timing errors in low SNR environments and providing reliable time data for high-precision positioning. Finally, based on the precise time difference of arrival, a spherical propagation model is used to invert the three-dimensional source point set and reconstruct the near-ground channel morphology of lightning strikes.

[0134] Compared with existing technologies, this system systematically solves the problem of interference from the complex electromagnetic and terrain environment of wind farms on lightning detection by adopting a complete technology chain from site optimization and high-precision time-frequency measurement to positioning algorithms. The final result is not isolated point positioning, but high-resolution, high-confidence three-dimensional imaging of the near-ground channel morphology of lightning strikes, thus providing unprecedentedly detailed observational data for lightning protection design, risk assessment, and accident analysis of wind turbines.

[0135] Furthermore, based on the above embodiments, this application also proposes an embodiment of a wind turbine lightning strike detection system based on radio frequency electromagnetic detection, please refer to... Figure 8 The lightning strike detection device for wind turbines based on radio frequency electromagnetic detection described below can be referred to in correspondence with the lightning strike detection method for wind turbines based on radio frequency electromagnetic detection described above.

[0136] In this embodiment, the wind turbine lightning strike detection system based on radio frequency electromagnetic detection includes:

[0137] The computing unit is used to obtain the initial position and initial directional antenna pointing of multiple radio frequency observation stations based on the signal-to-noise ratio constraint, elevation angle constraint and number constraint of the radio frequency observation stations; it is also used to iterate based on the simulated lightning radiation source signal, the initial position and the initial directional antenna pointing to obtain the target position and the target directional antenna pointing.

[0138] Multiple radio frequency observation stations are used to extract the arrival time of radio frequency pulses from multiple radio frequency observation stations by detecting the rising edge after unified time base alignment.

[0139] The detection device is used to obtain the time difference of arrival of multiple radio frequency observation stations based on the arrival time, obtain a three-dimensional source point set through the time difference of arrival positioning model of spherical propagation, and obtain the near-ground channel morphology of lightning strike based on the three-dimensional source point set.

[0140] It is understood that the detailed functional implementation of the above-mentioned devices can be found in the description of the aforementioned method embodiments, and will not be repeated here.

[0141] It should be understood that the above-described device is used to execute the methods in the above embodiments. The implementation principle and technical effect of the corresponding program modules in the device are similar to those described in the above methods. The working process of the device can be referred to the corresponding process in the above methods, and will not be repeated here.

[0142] Based on the methods in the above embodiments, this application provides an electronic device, please refer to... Figure 9The electronic device may include a processor 10, a communications interface 20, a memory 30, and a communication bus 40, wherein the processor 10, the communications interface 20, and the memory 30 communicate with each other via the communication bus 40. The processor 10 may call logical instructions in the memory 30 to execute the methods described in the above embodiments.

[0143] Furthermore, the logical instructions in the aforementioned memory 30 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application.

[0144] Based on the methods in the above embodiments, this application provides a computer-readable storage medium storing a computer program that, when run on a processor, causes the processor to execute the methods in the above embodiments.

[0145] Based on the methods in the above embodiments, this application provides a computer program product that, when run on a processor, causes the processor to execute the methods in the above embodiments.

[0146] It is understood that the processor in the embodiments of this application can be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. A general-purpose processor can be a microprocessor or any conventional processor.

[0147] The method steps in this application embodiment can be implemented in hardware or by a processor executing software instructions. The software instructions can consist of corresponding software modules, which can be stored in random access memory (RAM), flash memory, read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), registers, hard disks, portable hard disks, CD-ROMs, or any other form of storage medium known in the art. An exemplary storage medium is coupled to the processor, enabling the processor to read information from and write information to the storage medium. Of course, the storage medium can also be a component of the processor. The processor and the storage medium can reside in an ASIC.

[0148] In the above embodiments, implementation can be achieved entirely or partially through software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented entirely or partially as a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted through the computer-readable storage medium. The computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid-state disk (SSD)).

[0149] It is understood that the various numerical designations used in the embodiments of this application are merely for the convenience of description and are not intended to limit the scope of the embodiments of this application.

[0150] Those skilled in the art will readily understand that the above description is merely a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A method for detecting lightning strikes on wind turbine generators based on radio frequency electromagnetic detection, characterized in that, include: Based on the signal-to-noise ratio (SNR), elevation angle, and number constraints of radio frequency (RF) observation stations, the initial positions and initial directional antenna directions of multiple RF observation stations are obtained: Based on the geographical location of the wind turbine, surrounding topography, and distribution of potential obstacles, a feasible deployment area for RF observation stations is delineated; based on the number constraints, the total number of RF observation stations is determined; within the feasible deployment area, with the goal of satisfying the SNR and elevation angle constraints and optimizing the spatial geometric distribution of the stations, initial position placement planning is performed to generate a set of initial position coordinates for multiple RF observation stations. Based on the radio frequency observation stations in the initial position coordinate set, the initial pointing angle of their directional antennas is calculated according to their relative orientation to the wind turbine, so as to ensure that the main lobe of the antenna can cover the wind turbine. Based on the simulated lightning radiation source signal, initial position, and initial directional antenna pointing, the target position and target directional antenna pointing are obtained through iteration. Each radio frequency observation station is then deployed based on the target position and target directional antenna pointing. Based on the signal-to-noise ratio of the simulated lightning radiation source signal, the arrival time uncertainty of multiple radio frequency observation stations is obtained and determined as time delay error. Based on the time delay error and geometric accuracy factor, the positioning accuracy of multiple radio frequency observation stations is obtained. Based on the positioning accuracy, coverage, and number of radio frequency observation stations, an optimization index model is built; the initial position and initial directional antenna direction are iterated based on the optimization index model to obtain the target position and target directional antenna direction, and each radio frequency observation station is arranged according to the target position and the target directional antenna direction. Multiple radio frequency (RF) observation stations are aligned to a unified time base, and the arrival times of RF pulses from these stations are extracted by detecting rising edges. Each RF observation station includes a directional antenna, a filtering and amplification module, a sampling and timing module, and a trigger slicing module. Based on the sampling and timing module, unified time base alignment is performed on the multiple RF observation stations. Simultaneously, lightning radiation source signals are acquired using the directional antenna and transmitted to the filtering and amplification module. The received signals are bandpass filtered and amplified with low noise using the filtering and amplification module, and the processed signals are transmitted to the sampling and timing module. The processed signals are converted into digital signals and synchronized via GNSS. The acquired digital signals are detected using the trigger slicing module; when the rising edge of the pulse meets a slope threshold and exceeds a set threshold, the data waveforms before and after the corresponding time point of the digital signal are extracted from the circular buffer and output as an event slice. Based on the arrival time, the arrival time differences of multiple radio frequency observation stations are obtained, and a three-dimensional source point set is obtained through a spherical propagation time difference of arrival positioning model. Based on the three-dimensional source point set, the near-ground channel morphology of lightning strike is obtained: Based on the spherical wave propagation model, a hyperbolic equation constrained by the arrival time difference is established, which is determined to be a spherical propagation time difference of arrival positioning model; the nonlinear terms of the spherical propagation time difference of arrival positioning model are eliminated, and a relationship model between the radiation source location and the initial radiation time is obtained, and the initial estimate of the radiation source location is calculated; using the initial estimate as the starting point of iteration, an objective function is constructed with the goal of minimizing the distance difference residual, and solved by the nonlinear least squares method to obtain the three-dimensional source point set of the radiation source; based on the three-dimensional source point set, the near-ground channel morphology of lightning strike is obtained.

2. The method for detecting lightning strikes on wind turbines based on radio frequency electromagnetic detection as described in claim 1, characterized in that, Based on the signal-to-noise ratio constraints, elevation angle constraints, and number constraints of radio frequency observation stations, the initial positions and initial directional antenna pointing of multiple radio frequency observation stations are obtained. This process also includes: Based on the Friesian transmission equation, a model of the received power of the radio frequency observation station in any direction is constructed. Based on the received power model and noise power, the signal-to-noise ratio constraints of the radio frequency observation station are obtained; Obtain the altitude range of the radio frequency observation station, and obtain the elevation angle constraint based on the maximum elevation angle difference and the altitude range of the radio frequency observation station; Quantitative constraints are obtained based on the geometric relationships and signal quality of multiple radio frequency observation sites.

3. A lightning strike detection system for wind turbine generators based on radio frequency electromagnetic detection, characterized in that, The method for detecting lightning strikes on wind turbines based on radio frequency electromagnetic detection as described in claim 1 or 2 includes: The computing unit is used to obtain the initial position and initial directional antenna pointing of multiple radio frequency observation stations based on the signal-to-noise ratio constraint, elevation angle constraint and number constraint of the radio frequency observation stations; it is also used to iterate based on the simulated lightning radiation source signal, the initial position and the initial directional antenna pointing to obtain the target position and the target directional antenna pointing. Multiple radio frequency observation stations are used to extract the arrival time of radio frequency pulses from multiple radio frequency observation stations by detecting the rising edge after unified time base alignment; The detection device is used to obtain the time difference of arrival of multiple radio frequency observation stations based on the arrival time, obtain a three-dimensional source point set through a spherical propagation time difference of arrival positioning model, and obtain the near-ground channel morphology of lightning strike based on the three-dimensional source point set.

4. An electronic device, characterized in that, Includes memory and one or more processors; The memory is coupled to the one or more processors, and the memory is used to store computer program code, the computer program code including computer instructions; The one or more processors invoke the computer instructions to cause the electronic device to execute the lightning strike detection method for wind turbine generators based on radio frequency electromagnetic detection as described in claim 1 or 2.

5. A computer-readable storage medium comprising instructions, characterized in that: When the instruction is executed on an electronic device, the electronic device performs the lightning strike detection method for wind turbine generators based on radio frequency electromagnetic detection as described in claim 1 or 2.

6. A computer program product, comprising a computer program or instructions, characterized in that: When the computer program or instructions are run on the electronic device, the electronic device performs the lightning strike detection method for wind turbine generators based on radio frequency electromagnetic detection as described in claim 1 or 2.

Images