Method for optimizing 5g signal coverage of a wind farm

Abstract

This invention relates to the field of wireless communication technology and discloses a method for optimizing 5G signal coverage in wind farms. The method includes: acquiring real-time data on the rotation angle and speed of the blades in the wind turbine generator, and generating a wind turbine operation feature set based on the mapping relationship between the wind turbine's motion trajectory and signal attenuation in the real-time data after time alignment; fusing the network state feature set and the wind turbine operation feature set to obtain a digital twin model of the wind farm under electromagnetic propagation; analyzing the spatiotemporal distribution characteristics of the wind farm based on the digital twin model to obtain the coverage blind zone prediction result; adjusting the beam pointing angle and beamwidth in the wind farm according to the coverage blind zone prediction result to obtain phase compensation parameters; injecting the periodic characteristics of the wind turbine's motion trajectory into the phase compensation parameters to obtain an anti-obstruction beam configuration command; and sending the anti-obstruction beam configuration command to the 5G base station. This invention can improve the effect of 5G signal coverage optimization in wind farms.

Description

Technical Field

[0001] This invention relates to the field of wireless communication technology, and in particular to a method for optimizing 5G signal coverage in wind farms. Background Technology

[0002] In the 5G signal coverage scenario of wind farms, the continuous rotation of wind turbine blades creates dynamic obstruction, and the metal structure of the wind turbine during operation is prone to multipath interference. Existing signal coverage optimization methods mostly adopt static beam configuration or parameter adjustment based on fixed scenarios, which fail to effectively correlate the dynamic mapping relationship between the wind turbine's movement trajectory and signal attenuation. This results in severe fluctuations in signal during the blade obstruction period, leading to poor coverage stability.

[0003] Meanwhile, existing technologies lack accurate modeling of the electromagnetic propagation environment of wind farms. They fail to integrate the periodic characteristics of wind turbine operation with network status data, and cannot predict in advance spatiotemporal blind spots caused by periodic blade obstruction. This results in beam adjustment lagging behind obstruction changes, making it difficult to counteract multipath interference and dynamic obstruction, ultimately leading to recurring coverage blind spots and poor signal coverage optimization. Based on the beneficial effects described in the document, the focus should be on how the technology presented can improve these effects. This should be stated in two paragraphs, avoiding examples or comparisons with existing technologies; only highlight the improvements and benefits. Summary of the Invention

[0004] This invention provides a method for optimizing 5G signal coverage in wind farms to solve the problems mentioned in the background section.

[0005] To achieve the above objectives, the present invention provides a method for optimizing 5G signal coverage in wind farms, comprising:

[0006] S1. Obtain real-time data on the blade rotation angle and rotation speed of the wind turbine, and generate a wind turbine operation feature set of the wind turbine based on the mapping relationship between the wind turbine's motion trajectory and signal attenuation in the real-time data after time alignment.

[0007] S2. By integrating the network state feature set of the wind farm with the wind turbine operation feature set, a digital twin model of the wind farm under electromagnetic propagation is obtained;

[0008] S3. Based on the digital twin model, analyze the spatiotemporal distribution characteristics of the wind farm under multipath interference and signal blockage, and obtain the coverage blind zone prediction results of the wind farm.

[0009] S4. Based on the coverage blind spot prediction results, adjust the beam pointing angle and beam width of the 5G base station in the wind farm to obtain the phase compensation parameters of the 5G base station.

[0010] S5. Inject the periodic characteristics of the wind turbine's motion trajectory into the phase compensation parameters to obtain the anti-blocking beam configuration command for the 5G base station;

[0011] S6. Send the anti-obstruction beam configuration command to the 5G base station.

[0012] In a preferred embodiment, generating the wind turbine operating feature set of the wind turbine unit based on the mapping relationship between the wind turbine trajectory and signal attenuation in the time-aligned real-time data includes:

[0013] Extract signal strength time-series data from real-time data after time alignment;

[0014] The blade rotation angle and speed data are segmented by a sliding window to obtain a motion slice of the wind turbine.

[0015] Extract the occlusion correlation factor between the change in the spatial position of the blades and the fluctuation of the signal intensity within the motion slice;

[0016] A speed-attenuation transfer function model is established based on the periodic characteristics of the rotational speed and the occlusion correlation factor.

[0017] By fusing the parameter set of the speed-deceleration transfer function model with the rate of change of the blade spatial position of the wind turbine, the wind turbine operating characteristic set of the wind turbine is obtained.

[0018] In a preferred embodiment, the speed-deceleration transfer function model includes:

[0019]

[0020] In the formula, H(*) is the wind turbine speed attenuation value, α is the shading correlation factor, e is the natural constant, β is the dynamic attenuation factor of the rotating blade shading in the wind turbine, and f r γ is the real-time rotational speed of the wind turbine in the wind turbine unit, γ is the multipath interference amplitude of the metal surface in the wind turbine unit, π is , φ is the offset of the yaw angle phase of the wind turbine in the wind turbine unit, and t is the time factor.

[0021] In a preferred embodiment, the process of fusing the network state feature set and the wind turbine operation feature set of the wind farm to obtain a digital twin model of the wind farm under electromagnetic propagation includes:

[0022] The base station signal strength data in the network state feature set are timestamped to obtain the synchronization signal strength sequence of the wind farm;

[0023] Extract the blade rotation angle data from the wind turbine operating feature set, and determine the real-time spatial position of the blade edge in the blade rotation angle data through three-dimensional spatial coordinate transformation to obtain the dynamic shading profile of the wind farm.

[0024] Using the location of the terminal in the 5G base station as the endpoint, electromagnetic wave diffraction simulation is performed on the wind farm based on the synchronization signal strength sequence and the dynamic occlusion profile to obtain the path loss distribution map of the wind farm.

[0025] A digital twin model of the wind farm is constructed based on the path loss distribution map and the channel quality index in the network state feature set.

[0026] In a preferred embodiment, the step of performing electromagnetic wave diffraction simulation on the wind farm based on the synchronization signal strength sequence and the dynamic occlusion profile, with the location of the terminal in the 5G base station as the endpoint, to obtain the path loss distribution map of the wind farm, includes:

[0027] Based on the spatial coordinate data of the dynamic occlusion contour, the spatial geometric relationship between the blade edge and the line-of-sight path of the base station-terminal is identified, and the obstacle occlusion determination result of the wind farm is obtained.

[0028] When the obstacle occlusion determination result is partial occlusion, the direct signal component in the synchronization signal intensity sequence is extracted;

[0029] Based on the terrain elevation data of the terminal, the difference in propagation distance between the ground reflection path and the diffuse reflection path is determined;

[0030] The direct signal component and the propagation distance difference are weighted and fused to obtain the composite field strength attenuation value of the wind farm;

[0031] The synthetic field strength attenuation value is mapped to the base station-terminal line-of-sight path to obtain the path loss distribution map of the wind farm.

[0032] In a preferred embodiment, the step of analyzing the spatiotemporal distribution characteristics of the wind farm under multipath interference and signal obstruction based on the digital twin model to obtain the coverage blind zone prediction result of the wind farm includes:

[0033] A heat map of the attenuation distribution of the wind farm is constructed based on the intensity value of the signal attenuation factor in the digital twin model.

[0034] The continuous regions in the attenuation distribution heatmap where the signal attenuation exceeds a predetermined threshold are marked as spatial occlusion regions.

[0035] By integrating the spatial shading region with the periodic shading time window of the wind turbine operating feature set, the spatiotemporal joint blind zone prediction result of the wind farm is obtained.

[0036] In a preferred embodiment, adjusting the beam pointing angle and beamwidth of the 5G base station in the wind farm based on the coverage blind spot prediction result to obtain the phase compensation parameters of the 5G base station includes:

[0037] The set of blind zone location coordinates in the coverage blind zone prediction results is analyzed, and the horizontal azimuth and elevation angles of the blind zone center point relative to the 5G base station are determined.

[0038] Based on the horizontal azimuth and elevation angles, the beam scanning range of the antenna array in the 5G base station is determined;

[0039] Based on the time interval characteristics in the coverage blind spot prediction results, the beam dwell time window of the 5G base station is divided.

[0040] By combining the beam scanning range with the beam dwell time window, the dynamic beam pointing sequence of the 5G base station is obtained;

[0041] Based on the interference type in the coverage blind zone prediction results, a preset beamwidth adjustment strategy is matched;

[0042] The phase compensation parameters of the 5G base station are output according to the adjusted strategy.

[0043] In a preferred embodiment, matching a preset beamwidth adjustment strategy based on the interference type in the coverage blind zone prediction result includes:

[0044] When the interference type is multipath dominant, a narrow beam high gain strategy is adopted;

[0045] When the interference type is dominated by obstruction, a wide beam diversity strategy is adopted.

[0046] In a preferred embodiment, the step of injecting the periodic characteristics of the wind turbine's trajectory into the phase compensation parameters to obtain the anti-obstruction beam configuration command for the 5G base station includes:

[0047] Extract the angular velocity time series data of the blade rotation from the set of wind turbine operating features;

[0048] Calculate the Doppler frequency shift correlation function between the angular velocity timing data and the carrier frequency of the 5G base station;

[0049] Based on the phase change rate of the Doppler frequency shift correlation value in the Doppler frequency shift correlation function, the phase compensation gradient of the antenna array is derived.

[0050] The phase compensation gradient is mapped to the phase shifter control codeword of the antenna element in the antenna array;

[0051] Based on the time interval of the coverage blind zone prediction result, the phase shifter control codeword is injected into the phase compensation parameter to obtain the phase compensation parameter set of the 5G base station.

[0052] In a preferred embodiment, calculating the Doppler frequency shift correlation function between the angular velocity time-series data and the carrier frequency of the 5G base station includes:

[0053] The angular velocity time series data is normalized to obtain a standard angular velocity sequence;

[0054] Based on the carrier frequency parameters of the 5G base station, determine the maximum theoretical Doppler frequency shift of the blade tip motion;

[0055] Establish a time-aligned dataset of the standard angular velocity sequence and the real-time received signal carrier offset;

[0056] By fitting the angular velocity change in the angular velocity time series data with the carrier offset of the real-time received signal in the 5G base station, the linear transfer coefficient of the 5G base station is obtained.

[0057] Based on the linear transfer coefficient and the maximum theoretical Doppler frequency shift, the angular velocity-frequency shift scaling function of the 5G base station is constructed.

[0058] The phase response characteristics of the 5G base station are obtained by performing a Fourier transform on the angular velocity-frequency shift proportional function.

[0059] By integrating the linear transfer coefficient and the phase response characteristics, the Doppler frequency shift correlation function of the 5G base station is obtained, wherein the Doppler frequency shift correlation function is as follows:

[0060]

[0061] In the formula, F(*) is the Doppler frequency shift correlation value, and K d Let ω(t) be the linear transfer coefficient, θ(t) be the real-time angular velocity of the wind turbine blade, θ(t) be the real-time angle between the blade and the base station-terminal connection line, φ0 be the initial phase offset, N be the maximum harmonic order in the phase response characteristics, and A be the linear transfer coefficient. n f is the amplitude of the nth harmonic, where n is the harmonic ordinal number. b The frequency of the blades passing through is denoted by t, where t is the time factor. R represents the initial phase of the harmonics in the phase response characteristics, and R is the characteristic radius of the blade.

[0062] Compared with the prior art, the present invention has the following beneficial effects:

[0063] 1. This invention acquires real-time data on the rotation angle and speed of wind turbine blades, establishes a mapping relationship between the wind turbine's trajectory and signal attenuation, generates a wind turbine operation feature set, and integrates a network state feature set to construct a digital twin model of electromagnetic propagation in the wind farm. This model can accurately analyze the spatiotemporal distribution characteristics of multipath interference and signal obstruction, achieve accurate prediction of coverage blind spots, and provide precise decision-making basis for subsequent signal coverage optimization, fundamentally improving the targeting and effectiveness of 5G signal coverage optimization.

[0064] 2. This invention adjusts the beam pointing angle and width to obtain phase compensation parameters based on coverage blind zone prediction results, and injects the periodic characteristics of the wind turbine's trajectory to generate anti-obstruction beam configuration instructions. This approach enables the beam configuration of the 5G base station to dynamically adapt to signal changes caused by the periodic movement of the wind turbine, effectively offsetting the effects of obstruction and multipath interference, and significantly improving the stability and reliability of 5G signal coverage in wind farms. Attached Figure Description

[0065] Figure 1 A flowchart illustrating a method for optimizing 5G signal coverage in a wind farm according to an embodiment of the present invention;

[0066] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0067] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0068] This application provides a method for optimizing 5G signal coverage in a wind farm. The execution entity of this method includes, but is not limited to, at least one of the following electronic devices that can be configured to execute the method provided in this application: a server, a terminal, etc. In other words, the method for optimizing 5G signal coverage in a wind farm can be executed by software or hardware installed on a terminal device or a server device. The server includes, but is not limited to, a single server, a server cluster, a cloud server, or a cloud server cluster. The server can be an independent server or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms.

[0069] Reference Figure 1 The diagram shown is a flowchart illustrating a method for optimizing 5G signal coverage in a wind farm according to an embodiment of the present invention. In this embodiment, the method for optimizing 5G signal coverage in a wind farm includes:

[0070] S1. Obtain real-time data on the blade rotation angle and rotation speed of the wind turbine, and generate a wind turbine operation feature set of the wind turbine based on the mapping relationship between the wind turbine's motion trajectory and signal attenuation in the real-time data after time alignment.

[0071] In this embodiment of the invention, generating the wind turbine operating feature set of the wind turbine unit based on the mapping relationship between the wind turbine trajectory and signal attenuation in the real-time data after time alignment includes:

[0072] Extract signal strength time-series data from real-time data after time alignment;

[0073] The blade rotation angle and speed data are segmented by a sliding window to obtain a motion slice of the wind turbine.

[0074] Extract the occlusion correlation factor between the change in the spatial position of the blades and the fluctuation of the signal intensity within the motion slice;

[0075] A speed-attenuation transfer function model is established based on the periodic characteristics of the rotational speed and the occlusion correlation factor.

[0076] By fusing the parameter set of the speed-deceleration transfer function model with the rate of change of the blade spatial position of the wind turbine, the wind turbine operating characteristic set of the wind turbine is obtained.

[0077] The speed-deceleration transfer function model includes:

[0078]

[0079] In the formula, H(*) is the wind turbine speed attenuation value, α is the shading correlation factor, e is the natural constant, β is the dynamic attenuation factor of the rotating blade shading in the wind turbine, and f r γ is the real-time rotational speed of the wind turbine in the wind turbine unit, γ is the multipath interference amplitude of the metal surface in the wind turbine unit, π is , φ is the offset of the yaw angle phase of the wind turbine in the wind turbine unit, and t is the time factor.

[0080] Specifically, from the real-time data that has been time-aligned, real-time records related to signal strength are filtered out. These data have been synchronized in the time dimension through timestamp matching. These signal strength data are arranged in chronological order to form continuous signal strength time series data.

[0081] Furthermore, a sliding window of fixed length is set, the length of which is determined by the time it takes for the blade to rotate one revolution. The starting point of the blade rotation angle and speed data is taken as the starting point of the window. All blade rotation angle and speed data within the window range are captured. Then, the window is moved backward by a fixed step size, which is set to one-tenth of the window length. After each movement, the data within the corresponding range is captured repeatedly. The set of blade rotation angle and speed data within the window captured each time is the motion slice of the wind turbine.

[0082] Furthermore, based on the blade rotation angle data within the motion slice, the spatial position of the blade at each moment is calculated. By comparing the spatial positions at adjacent moments, the change in the blade's spatial position is obtained. Simultaneously, the change in signal intensity in the time series data of signal intensity within the motion slice is extracted as signal intensity fluctuation. The frequency and amplitude correlation of signal intensity fluctuation when the blade's spatial position changes are statistically analyzed. When the change in blade position reaches a certain amplitude, if the signal intensity shows a corresponding fluctuation, the degree of correlation between the two is recorded. This degree of correlation is the occlusion correlation factor.

[0083] Furthermore, by observing the changes in rotational speed data over time, the period length and pattern of repeated changes in rotational speed are determined as the periodicity characteristics of rotational speed. The occlusion correlation factor is mapped to the periodicity characteristics of rotational speed. When the rotational speed is in a certain stage within the period, the corresponding value of the occlusion correlation factor is recorded. The influence of the occlusion correlation factor on signal attenuation at different rotational speed stages is analyzed. For example, the signal attenuation amplitude corresponding to the occlusion correlation factor at a certain stage when the rotational speed increases is analyzed. In this way, a correspondence model between rotational speed change and signal attenuation is established, namely, the rotational speed-attenuation transfer function model.

[0084] Furthermore, all key information describing the model characteristics is extracted from the speed-decrease transfer function model as a parameter set. The change rate of blade spatial position is calculated by the change of blade rotation angle over time. Each parameter in the parameter set is matched with the change rate of blade spatial position. For example, the decay characteristics of a certain parameter at a certain speed are combined with the change rate of blade spatial position at that speed. All the matched information is integrated to form a comprehensive information set that can reflect the operating status of the wind turbine, namely the wind turbine operating characteristic set of the wind turbine.

[0085] Specifically, α originates from the calculation results of the occlusion correlation factor between the change in the spatial position of the blades and the signal intensity fluctuation within the motion slice, i.e., the degree of correlation is obtained by statistically analyzing the frequency and amplitude correlation of signal intensity fluctuations when the spatial position of the blades changes; β is determined by analyzing the occlusion process of the rotating blades in the wind turbine under different rotation states, specifically by recording the dynamic changes in signal attenuation when the blades occlude the signal, and summarizing the degree of influence of occlusion behavior on attenuation; f rγ is the real-time wind turbine rotation speed data, which is already included in the blade rotation angle and speed data; γ is determined by measuring the strength of multipath interference formed by the reflected signal from the metal surface of the wind turbine, specifically by collecting the signal change after the superposition of the reflected signal from the metal surface and the direct signal, and statistically analyzing the amplitude of the interference part; φ is determined by measuring the phase change corresponding to the deviation between the yaw angle of the wind turbine and the reference angle, that is, recording the yaw angle offset of the wind turbine and converting it into the corresponding phase offset value; t is the time record corresponding to the real-time data, which has been used for time alignment and is directly used.

[0086] Furthermore, H(*) represents the wind turbine speed attenuation value. This formula establishes a correspondence between the wind turbine's real-time speed and the speed attenuation value by combining the shading correlation factor, the dynamic attenuation factor of the rotating blade shading, the real-time speed of the wind turbine, the multipath interference amplitude of the metal surface, the offset of the wind turbine yaw angle phase, and the time factor. That is, when the real-time speed of the wind turbine changes, the corresponding speed attenuation value is calculated through the combined effect of the various parameters in the formula, thereby reflecting the influence of speed on signal attenuation and the effect of other related factors on attenuation.

[0087] Furthermore, as the real-time speed of the wind turbine increases, the value of α multiplied by the negative β times the real-time speed of the wind turbine in the formula will gradually decrease. This is because the exponential part of the natural constant is negative and its absolute value increases with the speed, resulting in a gradual weakening of the corresponding attenuation contribution. At the same time, the value of γ multiplied by the sine function will exhibit periodic fluctuations with time and speed. This is because the sine function itself is periodic, and its fluctuation period is determined by the real-time speed of the wind turbine, the time factor, and the offset of the yaw angle phase. Overall, the speed attenuation value will gradually decrease with the increase of speed, superimposed with periodic fluctuations.

[0088] In summary, by acquiring real-time data on the rotation angle and speed of the blades in a wind turbine, accurate and timely first-hand information can be provided for monitoring the operating status of the wind turbine.

[0089] In general, in the complex operating environment of wind turbines, the rotation angle and speed of the blades are constantly changing dynamically, and real-time data acquisition ensures the keen capture of these changes.

[0090] In summary, by aligning the collected real-time data with time, the mapping relationship between the wind turbine's trajectory and signal attenuation can be accurately identified, thereby generating a set of wind turbine operating features.

[0091] In summary, this process is equivalent to building a dedicated "operational profile" for the wind turbine, which is of great significance. The wind turbine operation feature set integrates multi-dimensional key information, providing a solid data foundation for subsequent fault early warning, performance optimization, and intelligent control.

[0092] For example, through in-depth analysis of feature set data, abnormal wear that may occur in the blades can be predicted in advance, so that maintenance can be arranged in a timely manner, avoiding downtime caused by blade failure, which greatly improves the stability and reliability of wind turbine operation, and ultimately improves the overall power generation efficiency and economic benefits of wind farms.

[0093] S2. By integrating the network state feature set of the wind farm with the wind turbine operation feature set, a digital twin model of the wind farm under electromagnetic propagation is obtained;

[0094] In this embodiment of the invention, the fusion of the network state feature set and the wind turbine operation feature set of the wind farm to obtain a digital twin model of the wind farm under electromagnetic propagation includes:

[0095] The base station signal strength data in the network state feature set are timestamped to obtain the synchronization signal strength sequence of the wind farm;

[0096] Extract the blade rotation angle data from the wind turbine operating feature set, and determine the real-time spatial position of the blade edge in the blade rotation angle data through three-dimensional spatial coordinate transformation to obtain the dynamic shading profile of the wind farm.

[0097] Using the location of the terminal in the 5G base station as the endpoint, electromagnetic wave diffraction simulation is performed on the wind farm based on the synchronization signal strength sequence and the dynamic occlusion profile to obtain the path loss distribution map of the wind farm.

[0098] A digital twin model of the wind farm is constructed based on the path loss distribution map and the channel quality index in the network state feature set.

[0099] The method involves using the location of the terminal in the 5G base station as the endpoint, and performing electromagnetic wave diffraction simulation on the wind farm based on the synchronization signal strength sequence and the dynamic occlusion profile to obtain the path loss distribution map of the wind farm, including:

[0100] Based on the spatial coordinate data of the dynamic occlusion contour, the spatial geometric relationship between the blade edge and the line-of-sight path of the base station-terminal is identified, and the obstacle occlusion determination result of the wind farm is obtained.

[0101] When the obstacle occlusion determination result is partial occlusion, the direct signal component in the synchronization signal intensity sequence is extracted;

[0102] Based on the terrain elevation data of the terminal, the propagation distance difference between the ground reflection path and the diffuse reflection path is determined;

[0103] The direct signal component and the propagation distance difference are weighted and fused to obtain the composite field strength attenuation value of the wind farm;

[0104] The synthetic field strength attenuation value is mapped to the base station-terminal line-of-sight path to obtain the path loss distribution map of the wind farm.

[0105] Specifically, all base station signal strength data are selected from the network state feature set. These data contain their own timestamp records. These timestamps are matched with a preset unified time base, and the timestamps corresponding to each base station signal strength data are adjusted to make all base station signal strength data consistent in the time dimension, that is, there is a corresponding signal strength record at the same time point. Then, these base station signal strength data with completed timestamp alignment are arranged in chronological order to form the synchronization signal strength sequence of the wind farm.

[0106] Furthermore, blade rotation angle data is extracted from the wind turbine operating feature set. A three-dimensional spatial coordinate system is established with the center point of the wind turbine base as the origin. The X-axis is along the horizontal direction of the wind farm, the Y-axis is along the vertical direction of the ground, and the Z-axis is along another horizontal direction of the wind farm. Based on the blade length and rotation direction corresponding to each angle in the blade rotation angle data, the coordinate values ​​of the blade edge in the three-dimensional coordinate system are calculated. As the blade rotation angle changes, the coordinate values ​​of the blade edge are updated in real time. The continuous contour formed by these continuously updated blade edge coordinates is the dynamic shading contour of the wind farm.

[0107] Furthermore, the specific spatial location of the terminal in the 5G base station is determined as the endpoint of electromagnetic wave propagation. The initial signal strength values ​​at each time point in the synchronization signal strength sequence are obtained. Combined with the real-time spatial location of the blades in the dynamic obstruction profile, the propagation path of electromagnetic waves from the signal source to the terminal location is simulated. When the electromagnetic wave propagation path encounters the blades in the dynamic obstruction profile, the signal attenuation degree when the electromagnetic wave diffracts through the blade edge is calculated. The attenuation degree is combined with the initial signal strength in the synchronization signal strength sequence to obtain the signal loss value at different locations. The signal loss values ​​at all locations are presented in graphical form, which is the path loss distribution map of the wind farm.

[0108] Furthermore, the loss data corresponding to each location in the path loss distribution map is associated with the channel quality indicators at that location in the network state feature set. For example, the path loss value of a certain area corresponds to the channel transmission rate, bit error rate, and other indicators of that area. A virtual scene that is completely consistent with the actual geographical environment, wind turbine location, and base station location of the wind farm is constructed in the virtual space. The associated path loss data and channel quality indicators are loaded into the corresponding locations of the virtual scene, so that the virtual scene can reflect the actual signal propagation loss and channel status of the wind farm in real time. This virtual scene that can map the actual wind farm network and wind turbine related status is the digital twin model of the wind farm.

[0109] Specifically, the spatial coordinates of the base station and the terminal are obtained, and these two coordinates are connected to form a straight line segment of the line-of-sight path between the base station and the terminal. All spatial coordinate points of the blade edges in the dynamic occlusion profile are extracted, and the vertical distance from each blade edge coordinate point to the straight line segment of the line-of-sight path is calculated. At the same time, it is determined whether the blade edge coordinate points are within the range between the two endpoints of the straight line segment of the line-of-sight path. If the vertical distance from all blade edge coordinate points to the line-of-sight path is greater than a preset threshold and is not within the endpoint range of the line-of-sight path, it is determined to be unobstructed. If the vertical distance from some blade edge coordinate points to the line-of-sight path is less than or equal to the preset threshold and is within the endpoint range of the line-of-sight path, it is determined to be partially obstructed. If all blade edge coordinate points completely cover the straight line segment of the line-of-sight path and the vertical distance is less than or equal to the preset threshold, it is determined to be completely obstructed. These determination results together constitute the obstacle occlusion determination results of the wind farm.

[0110] Furthermore, when the obstacle occlusion determination result is partial occlusion, the synchronization signal strength sequence contains direct signal components, reflected signal components, and scattered signal components caused by occlusion. Since the direct signal component is not completely blocked, its signal strength fluctuation amplitude is small and maintains a relatively stable propagation characteristic. By observing the change pattern of signal strength in the synchronization signal strength sequence, signal segments with fluctuation amplitude within a preset range and duration conforming to the propagation characteristics of direct signals are selected. The signal strength values ​​corresponding to these signal segments are extracted, which are the direct signal components in the synchronization signal strength sequence.

[0111] Furthermore, topographic elevation data of the terminal's location and surrounding area are collected. This data includes altitude information of various points on the ground. The ground reflection path from the base station to the terminal is determined. This path is the propagation path from the base station through a certain ground reflection point to the terminal. The actual length of this path is measured. At the same time, the diffuse reflection path is determined. This path is the propagation path from the base station through multiple ground diffuse reflection points to the terminal. The actual length of this path is measured. The propagation length of the ground reflection path is subtracted from the propagation length of the diffuse reflection path. The resulting value is the difference in propagation distance between the ground reflection path and the diffuse reflection path.

[0112] Furthermore, weighting coefficients are determined based on the propagation distance difference. The greater the propagation distance difference, the more significant the impact of ground reflection and diffuse reflection on the signal, and the larger the corresponding reflection-related weighting coefficient, the smaller the weighting coefficient of the direct signal component. Conversely, the weighting coefficients are adjusted. The direct signal component is multiplied by its corresponding weighting coefficient to obtain the contribution value of the direct signal. Then, the attenuation effect of reflection and diffuse reflection on the signal is calculated in combination with the propagation distance difference. The direct signal contribution value and the attenuation effect value are added together to obtain the composite field strength attenuation value of the wind farm.

[0113] Furthermore, using the straight segment of the line-of-sight path from the base station to the terminal as a reference, the line-of-sight path is divided into several equidistant spatial points, each spatial point corresponding to a specific location coordinate. The synthetic field strength attenuation value is matched to the corresponding spatial point on the line-of-sight path according to the spatial coordinates corresponding to the calculation. Each spatial point has a corresponding synthetic field strength attenuation value. The attenuation values ​​of all spatial points are presented on a two-dimensional or three-dimensional map using color gradients or numerical labels to form a graphic that can intuitively show the signal attenuation at different locations on the line-of-sight path. This graphic is the path loss distribution map of the wind farm.

[0114] In summary, by integrating the network status feature set of a wind farm with the wind turbine operation feature set to construct a digital twin model, it is possible to deeply correlate the real-time status of the 5G network in the wind farm (such as signal strength, channel quality, etc.) with the dynamic operation status of the wind turbine (such as blade position, rotation period, etc.), and accurately reproduce the propagation process of electromagnetic signals in the wind farm under the influence of wind turbine movement.

[0115] In summary, this model breaks through the information gap between network status and wind turbine operating status, and can realistically simulate the electromagnetic propagation law under complex scenarios such as multipath interference and signal blockage, providing a high-fidelity simulation platform for subsequent analysis of the spatiotemporal characteristics of signal coverage.

[0116] In summary, this digital twin model can dynamically reflect the real-time impact of wind turbine movement on signal propagation, upgrading the characterization of the electromagnetic environment of wind farms from static description to dynamic simulation.

[0117] In summary, this model can more accurately capture the dynamic changes in signal attenuation, providing reliable model support for coverage blind spot prediction. This lays a scientific foundation for subsequent beam adjustment and anti-blocking configuration, fundamentally improving the ability to control the propagation characteristics of 5G signals in wind farms and ensuring the accuracy and foresight of signal coverage optimization.

[0118] S3. Based on the digital twin model, analyze the spatiotemporal distribution characteristics of the wind farm under multipath interference and signal blockage, and obtain the coverage blind zone prediction results of the wind farm.

[0119] In this embodiment of the invention, the step of analyzing the spatiotemporal distribution characteristics of the wind farm under multipath interference and signal obstruction based on the digital twin model to obtain the coverage blind zone prediction result of the wind farm includes:

[0120] A heat map of the attenuation distribution of the wind farm is constructed based on the intensity value of the signal attenuation factor in the digital twin model.

[0121] The continuous regions in the attenuation distribution heatmap where the signal attenuation exceeds a predetermined threshold are marked as spatial occlusion regions.

[0122] By integrating the spatial shading region with the periodic shading time window of the wind turbine operating feature set, the spatiotemporal joint blind zone prediction result of the wind farm is obtained.

[0123] Specifically, the intensity values ​​of all signal attenuation factors are extracted from the digital twin model. These intensity values ​​are bound to specific spatial coordinates within the wind farm, with each coordinate point corresponding to a unique intensity value. These coordinate points are arranged on a virtual map according to the actual geographical layout of the wind farm, and color gradient rules are set. The area with the smallest signal attenuation factor intensity value is represented by light blue, and as the intensity value gradually increases, the color transitions to blue and then dark blue. The area with the largest intensity value is represented by dark red. The color corresponding to the intensity value of each coordinate point is filled into the corresponding position on the virtual map, ultimately forming an attenuation distribution heat map that can intuitively display the degree of signal attenuation at different locations within the wind farm.

[0124] Furthermore, based on the minimum signal quality required for normal data transmission by terminal equipment within the wind farm, a predetermined threshold for signal attenuation is determined. This threshold is a fixed attenuation strength value. When the signal attenuation reaches this value, the communication quality of the terminal equipment will drop to a level where normal data transmission is impossible. The signal attenuation strength value of each coordinate point is checked one by one in the attenuation distribution heatmap. When the strength value of a coordinate point exceeds the predetermined threshold, the position of that coordinate point is recorded. The adjacent coordinate points are then checked. If the strength values ​​of adjacent coordinate points also exceed the predetermined threshold, these coordinate points are determined to constitute a continuous area. This continuous area is completely encircled with a black border, and gray is filled inside the area as a marker. The marked area is the spatial occlusion area.

[0125] Furthermore, time records related to blade shading were filtered from the wind turbine operation feature set. These records include the start and end times of each shading of the base station-terminal signal path during blade rotation. Since blade rotation is periodic, the same shading phenomenon will repeat at fixed time intervals. The shading time interval that repeats more than three times consecutively is determined as a periodic shading time window. Each window contains a clear start and end time. The coordinate range of the spatial shading area on the virtual map is associated with these periodic shading time windows, that is, it is determined in which time windows a certain spatial shading area will cause signal attenuation to exceed a predetermined threshold due to blade shading. The coordinate range of all spatial shading areas and their corresponding periodic shading time windows are compiled into a table. Each row in the table records the location information of a spatial area and all corresponding shading time windows. The information presented in this table is the spatiotemporal joint blind zone prediction result of the wind farm.

[0126] In summary, analyzing the spatiotemporal distribution characteristics of wind farms under multipath interference and signal blockage based on digital twin models and obtaining coverage blind zone prediction results can accurately capture the pattern of signal propagation being affected by the dynamic operation of wind turbines.

[0127] In summary, the digital twin model integrates key features of network status and wind turbine operation, and can realistically reproduce electromagnetic propagation scenarios. Based on this, it analyzes features such as the signal superposition effect of multipath interference and the spatiotemporal range of signal blockage. It can directly link the formation of coverage blind spots with factors such as the periodicity of wind turbine movement and path loss of signal propagation. This allows the prediction results to not only include the spatial location information of the blind spots, but also to simultaneously output the time interval and duration of the blind spots, achieving accurate spatiotemporal prediction of coverage blind spots.

[0128] Overall, these predictions provide clear optimization targets for subsequent 5G base station beam adjustments.

[0129] In summary, by understanding the spatiotemporal distribution characteristics of blind spots, beam adjustment can be specifically matched to the time window and spatial location of blind spot occurrences, avoiding aimless generalization adjustments, significantly improving the accuracy of beam configuration, and providing a reliable basis for generating anti-obstruction beam configuration instructions, thereby ensuring the effectiveness of 5G signal coverage optimization measures from the source.

[0130] S4. Based on the coverage blind spot prediction results, adjust the beam pointing angle and beam width of the 5G base station in the wind farm to obtain the phase compensation parameters of the 5G base station.

[0131] In this embodiment of the invention, adjusting the beam pointing angle and beamwidth of the 5G base station in the wind farm based on the coverage blind spot prediction result to obtain the phase compensation parameters of the 5G base station includes:

[0132] The set of blind zone location coordinates in the coverage blind zone prediction results is analyzed, and the horizontal azimuth and elevation angles of the blind zone center point relative to the 5G base station are determined.

[0133] Based on the horizontal azimuth and elevation angles, the beam scanning range of the antenna array in the 5G base station is determined;

[0134] Based on the time interval characteristics in the coverage blind spot prediction results, the beam dwell time window of the 5G base station is divided.

[0135] By combining the beam scanning range with the beam dwell time window, the dynamic beam pointing sequence of the 5G base station is obtained;

[0136] Based on the interference type in the coverage blind zone prediction results, a preset beamwidth adjustment strategy is matched;

[0137] The phase compensation parameters of the 5G base station are output according to the adjusted strategy.

[0138] The step of matching a preset beamwidth adjustment strategy based on the interference type in the coverage blind zone prediction result includes:

[0139] When the interference type is multipath dominant, a narrow beam high gain strategy is adopted;

[0140] When the interference type is dominated by obstruction, a wide beam diversity strategy is adopted.

[0141] Specifically, the horizontal azimuth and elevation angles of the antenna array are obtained from the equipment parameters of the 5G base station. The horizontal azimuth angle represents the pointing angle of the antenna array in the horizontal plane, and the elevation angle represents the tilt angle of the antenna array in the vertical plane. A three-dimensional spatial coordinate system is established with the base station as the origin, with the horizontal direction as the X-axis and Y-axis, and the vertical direction as the Z-axis. The scanning angle range of the antenna array in the XY plane is determined based on the horizontal azimuth angle. This range extends to the left and right of the horizontal azimuth angle as the center, and this angle value is determined by the hardware characteristics of the antenna array, usually ±60 degrees. The scanning angle range of the antenna array in the XZ plane is determined based on the elevation angle. This range extends to the top and bottom of the elevation angle as the center, and this angle value is also determined by the hardware characteristics of the antenna array, usually ±30 degrees. The scanning angle ranges in the horizontal and vertical directions are combined to form a three-dimensional conical spatial region. This region is the beam scanning range of the antenna array in the 5G base station.

[0142] Furthermore, all time interval features are extracted from the coverage blind zone prediction results. These features include the start time, end time, and duration of each coverage blind zone. The time axis is divided into several equal time segments at fixed intervals, with each time segment having a length of 5 milliseconds. The number of times the coverage blind zone appears in each time segment is counted. If the number of times the coverage blind zone appears in a certain time segment reaches a preset threshold, the time segment is marked as a high-priority time segment. If the number of occurrences is less than the threshold, it is marked as a low-priority time segment. Consecutive high-priority time segments are merged into a larger time window, which is the beam dwell time window. Within this window, the antenna array needs to focus the beam on a specific area to cover the blind zone, while within the low-priority time segment, the antenna array can perform a wider scan.

[0143] Furthermore, the beam scanning range is gridded in a three-dimensional spatial coordinate system, dividing it into several equal-sized grid cells. Each grid cell corresponds to a specific spatial direction, and each grid cell is assigned a unique identifier. Based on the division of the beam dwell time window, the grid cells that need to be scanned in each time window are determined. These grid cells are determined based on the spatial location information in the coverage blind zone prediction results. At the beginning of each beam dwell time window, the beam points to these key grid cells in a pre-set order, pointing to one grid cell at a time and staying for a certain period of time. This time length is determined by the total length of the beam dwell time window and the number of grid cells to be scanned. The beam pointing order and the corresponding time points in each time window are recorded to form an ordered list. This list is the dynamic beam pointing sequence of the 5G base station.

[0144] Furthermore, all interference type information is extracted from the coverage blind zone prediction results. These interference types include multipath fading, co-channel interference, and Doppler shift. For each interference type, the corresponding adjustment strategy is found in a preset table of interference type and beamwidth adjustment strategy. This table is based on a large amount of experimental data and actual operating experience. For example, when the interference type is multipath fading, the corresponding adjustment strategy is to narrow the beamwidth to enhance the signal directionality and reduce the influence of reflected signals. When the interference type is co-channel interference, the corresponding strategy is to appropriately widen the beamwidth and increase the beam's transmit power to improve the signal-to-interference ratio. The adjustment strategies corresponding to each interference type in the coverage blind zone prediction results are extracted to form a strategy set. This set is the matched beamwidth adjustment strategy.

[0145] Furthermore, based on the matched beamwidth adjustment strategy, the beam parameters that need to be adjusted within each beam dwell time window are determined. These parameters include the horizontal beamwidth, vertical beamwidth, and phase offset. For each beam parameter that needs adjustment, the corresponding phase compensation value is looked up in a preset parameter adjustment table. This parameter adjustment table is pre-calculated based on the characteristics of the antenna array and the signal processing algorithm. For example, when it is necessary to narrow the beamwidth, the corresponding phase compensation value is looked up. These values ​​are used to adjust the signal phase of each antenna element in the antenna array, so that the synthesized beamwidth is narrowed. All phase compensation values ​​within each beam dwell time window are arranged in chronological order to form a sequence. This sequence is the phase compensation parameter of the 5G base station. These parameters are input into the beamforming controller of the 5G base station. The controller adjusts the signal phase of the antenna array according to these parameters to achieve dynamic control of the beamwidth and direction.

[0146] Specifically, when the interference type is determined to be multipath-dominated based on the coverage blind zone prediction results, a narrow beam high-gain strategy is activated. During operation, the horizontal beamwidth of the antenna array is reduced from the conventional 60 degrees to 30 degrees, and the vertical beamwidth is reduced from the conventional 30 degrees to 15 degrees. By adjusting the signal phase of each antenna element in the antenna array, the beam energy is concentrated in a narrower spatial range. At the same time, the transmit power of the beam is increased by 20% on the original basis, which enhances the signal strength on the main propagation path, reduces the interference of multipath signals such as reflection and refraction on the main signal, and ensures that the signal on the direct path can stably reach the terminal.

[0147] Furthermore, when the interference type is determined to be obstruction-dominated based on the coverage blind zone prediction results, a wide-beam diversity strategy is activated. During operation, the horizontal beamwidth of the antenna array is increased from the conventional 60 degrees to 90 degrees, and the vertical beamwidth is increased from the conventional 30 degrees to 45 degrees. At the same time, the diversity transmission mode of the antenna array is enabled, transmitting three independent beams from three different spatial angles. The coverage areas of these three beams overlap and include the possible gaps in the obstruction area. Even if one or two beams are blocked by obstructions such as propellers, the remaining beams can still transmit the signal to the terminal through the unobstructed path, ensuring that the terminal can always receive a valid signal.

[0148] In summary, adjusting the beam pointing angle and beamwidth of 5G base stations based on coverage blind spot prediction results to obtain phase compensation parameters allows the beam configuration to be directly pointed to areas with weak signal coverage, achieving targeted optimization.

[0149] In summary, the coverage blind zone prediction results clearly identify the spatial locations and spatiotemporal distribution characteristics of severe signal attenuation in wind farms. Based on this, adjusting the beam pointing angle can enable the beam to be precisely focused in the direction of the blind zone; adjusting the beam width can adapt to the size of the blind zone—using a wide beam for large blind zones and a narrow beam to enhance signal strength for small, concentrated blind zones, effectively improving the matching degree between the beam and the blind zone.

[0150] In summary, the phase compensation parameters generated in this process are the result of quantitative optimization of beamform, which can be directly used to correct the signal transmission status of base station antennas, reduce signal waste and insufficient coverage caused by beam pointing deviation or width mismatch, and lay a precise parameter foundation for subsequent injection of wind turbine motion periodic characteristics and generation of anti-obstruction beam configuration instructions, further improving the effectiveness of 5G signal coverage.

[0151] S5. Inject the periodic characteristics of the wind turbine's motion trajectory into the phase compensation parameters to obtain the anti-blocking beam configuration command for the 5G base station;

[0152] In this embodiment of the invention, the step of injecting the periodic characteristics of the wind turbine's trajectory into the phase compensation parameters to obtain the anti-obstruction beam configuration command for the 5G base station includes:

[0153] Extract the angular velocity time series data of the blade rotation from the set of wind turbine operating features;

[0154] Calculate the Doppler frequency shift correlation function between the angular velocity timing data and the carrier frequency of the 5G base station;

[0155] Based on the phase change rate of the Doppler frequency shift correlation value in the Doppler frequency shift correlation function, the phase compensation gradient of the antenna array is derived.

[0156] The phase compensation gradient is mapped to the phase shifter control codeword of the antenna element in the antenna array;

[0157] Based on the time interval of the coverage blind zone prediction result, the phase shifter control codeword is injected into the phase compensation parameter to obtain the phase compensation parameter set of the 5G base station.

[0158] The calculation of the Doppler frequency shift correlation function between the angular velocity time-series data and the carrier frequency of the 5G base station includes:

[0159] The angular velocity time series data is normalized to obtain a standard angular velocity sequence;

[0160] Based on the carrier frequency parameters of the 5G base station, determine the maximum theoretical Doppler frequency shift of the blade tip motion;

[0161] Establish a time-aligned dataset of the standard angular velocity sequence and the real-time received signal carrier offset;

[0162] By fitting the angular velocity change in the angular velocity time series data with the carrier offset of the real-time received signal in the 5G base station, the linear transfer coefficient of the 5G base station is obtained.

[0163] Based on the linear transfer coefficient and the maximum theoretical Doppler frequency shift, the angular velocity-frequency shift scaling function of the 5G base station is constructed.

[0164] The phase response characteristics of the 5G base station are obtained by performing a Fourier transform on the angular velocity-frequency shift proportional function.

[0165] By integrating the linear transfer coefficient and the phase response characteristics, the Doppler frequency shift correlation function of the 5G base station is obtained, wherein the Doppler frequency shift correlation function is as follows:

[0166]

[0167] In the formula, F(*) is the Doppler frequency shift correlation value, and K d Let ω(t) be the linear transfer coefficient, θ(t) be the real-time angular velocity of the wind turbine blade, θ(t) be the real-time angle between the blade and the base station-terminal connection line, φ0 be the initial phase offset, N be the maximum harmonic order in the phase response characteristics, and A be the linear transfer coefficient. n f is the amplitude of the nth harmonic, where n is the harmonic ordinal number. b The frequency of the blades passing through is denoted by t, where t is the time factor. R represents the initial phase of the harmonics in the phase response characteristics, and R is the characteristic radius of the blade.

[0168] Specifically, data related to blade rotation is extracted from the wind turbine operating feature set. This data includes the blade rotation angle at each time point. The angle difference between two adjacent time points is calculated and then divided by the time interval between these two time points to obtain the average angular velocity within each time interval. The average angular velocities of all time intervals are arranged in chronological order to form the angular velocity time series data of blade rotation. Each angular velocity value corresponds to a precise timestamp.

[0169] Furthermore, the carrier frequency of the 5G base station is fixed to a known reference value. Each angular velocity value in the time-series data of the propeller rotation angular velocity is substituted into the Doppler frequency shift calculation formula, which describes the frequency change relationship caused by the relative motion between the wave source and the observer. The Doppler frequency shift value corresponding to each time point is calculated. The Doppler frequency shift value at each time point is correlated with the angular velocity value at that time point to form a series of data pairs. The functional relationship between these data pairs is analyzed. A function that can accurately describe the relationship between angular velocity and Doppler frequency shift is determined by fitting method. This function is the Doppler frequency shift correlation function.

[0170] Furthermore, the Doppler frequency shift correlation value in the Doppler frequency shift correlation function is analyzed. The phase difference between the Doppler frequency shift correlation values ​​at two adjacent time points is calculated and then divided by the time interval between these two time points to obtain the phase change rate within each time interval. The phase change rates of all time intervals are arranged in chronological order to form a phase change rate sequence. Based on the geometry of the antenna array and the signal propagation principle, the correspondence between the phase change rate and the phase compensation gradient of the antenna array is determined. Through this correspondence, the phase change rate sequence is converted into a phase compensation gradient sequence of the antenna array. Each phase compensation gradient value corresponds to a specific time point and antenna array position.

[0171] Furthermore, the antenna array is divided into several independently controlled antenna array elements, each containing multiple antenna elements. A unique identifier is assigned to each antenna array element, and a mapping table is established between phase compensation gradient values ​​and phase shifter control codes. This mapping table maps different ranges of phase compensation gradient values ​​to different phase shifter control codes. These codes are binary codes used to control the operating state of the phase shifter. Based on each gradient value in the phase compensation gradient sequence, the corresponding phase shifter control code is searched in the mapping table. The found control codes are arranged in chronological order to form a phase shifter control code sequence. Each control code corresponds to a specific antenna array element and time point.

[0172] Furthermore, all time interval information is extracted from the coverage blind zone prediction results. This information includes the start and end times of each coverage blind zone. The parameter segment corresponding to each time interval is found in the phase compensation parameter sequence. The control codewords of the corresponding time points in the phase shifter control codeword sequence are inserted into the corresponding parameter segments of the phase compensation parameter sequence to ensure that the phase compensation parameters contain the corresponding phase shifter control codewords within the time period of each coverage blind zone. The phase compensation parameter sequence with the inserted phase shifter control codewords is then organized and verified to ensure the continuity and correctness of the parameters. The final sequence obtained is the phase compensation parameter set of the 5G base station.

[0173] Specifically, the maximum and minimum values ​​of all angular velocity values ​​are found from the angular velocity time series data. The difference between the maximum and minimum values ​​is calculated. The minimum value is subtracted from each angular velocity value, and the result is divided by the difference between the maximum and minimum values. Through this calculation, all angular velocity values ​​are converted to the range of 0 to 1. These converted values ​​are then arranged in the original time order to obtain the standard angular velocity sequence.

[0174] Furthermore, the carrier frequency parameters of the 5G base station are obtained. These parameters are the fixed frequency values ​​of the signals transmitted by the base station. Given the length of the wind turbine blades, the maximum linear velocity of the blade tip is obtained by multiplying the blade length by the maximum angular velocity value in the angular velocity time series data. Based on the carrier frequency, the speed of light, and the maximum linear velocity of the blade tip, when the direction of the blade tip's motion is consistent with the direction of electromagnetic wave propagation, the resulting Doppler frequency shift value is calculated. This value is the maximum theoretical Doppler frequency shift of the blade tip motion.

[0175] Furthermore, the timestamp corresponding to each angular velocity value in the standard angular velocity sequence is extracted, and the carrier offset of the real-time received signal from the 5G base station and its corresponding timestamp are also extracted. The timestamps of the two sequences are compared, and angular velocity values ​​and carrier offsets with completely identical timestamps are retained. For cases where the timestamps are close but not completely identical, value pairs with a time difference within 0.001 seconds are selected. Using the timestamp of the angular velocity value as a reference, the corresponding carrier offset is retained. These matched angular velocity values ​​and carrier offsets are combined into data pairs in chronological order to form a time-aligned dataset.

[0176] Furthermore, the angular velocity difference between two adjacent time points is calculated from the angular velocity time-series data to obtain the angular velocity change. At the same time, the carrier offset difference between corresponding adjacent time points is calculated from the real-time received signal from the 5G base station to obtain the carrier offset change. All angular velocity changes are used as the horizontal axis data and the corresponding carrier offset changes are used as the vertical axis data. All data points are marked in the coordinate system, and a straight line that best approximates all data points is drawn. The inclination of this straight line is the linear transfer coefficient, which reflects the proportional relationship between the angular velocity change and the carrier offset change.

[0177] Furthermore, based on the linear transfer coefficient, the proportional relationship between angular velocity change and Doppler frequency shift is determined. The maximum theoretical Doppler frequency shift is taken as the upper limit of the function. That is, when the angular velocity reaches the maximum value, the frequency shift value output by the function is equal to the maximum theoretical Doppler frequency shift, and when the angular velocity is 0, the frequency shift value output by the function is 0. Within the range of 0 to the maximum angular velocity, according to the proportional relationship corresponding to the linear transfer coefficient, the correspondence between the angular velocity value and the Doppler frequency shift value is established. This correspondence is the angular velocity-frequency shift proportional function.

[0178] Furthermore, the angular velocity-frequency shift proportional function is sampled in the time domain at fixed time intervals to obtain a series of discrete function values. These function values ​​correspond to the frequency shift prediction results at different time points. By using Fourier transform, these time domain function values ​​are converted to the frequency domain to obtain the amplitude and phase corresponding to different frequency components in the frequency domain. The law of phase change with frequency is extracted, that is, the phase value and change trend at different frequencies. These laws are the phase response characteristics.

[0179] Furthermore, the linear transfer coefficient is used as the amplitude parameter, which determines the strength of the influence of angular velocity changes on Doppler frequency shift. The phase response characteristics are used as the phase parameter, which reflects the phase change characteristics of the frequency shift signal at different frequencies. By combining the amplitude and phase parameters, a function that can reflect both the amplitude correlation between angular velocity and frequency shift and the phase correlation is formed. This function is the Doppler frequency shift correlation function of 5G base stations.

[0180] Specifically, K d The linear transfer coefficient, derived from the angular velocity change in the fitted angular velocity time series data and the carrier offset of the real-time received signal from the 5G base station, is determined by fitting the inclination of the straight line using the angular velocity change and carrier offset change as data points; ω(t) is the real-time angular velocity value extracted from the blade rotation angular velocity time series data from the wind turbine operation feature set, which changes dynamically over time; θ(t) is calculated by determining the real-time position of the blade through the blade rotation angle data and combining it with the position connection between the base station and the terminal, reflecting the real-time angle between the blade and the line-of-sight path; φ0 is the phase offset value at the initial moment in the phase response feature, determined by the starting phase of the phase response feature after Fourier transform; N is the maximum harmonic order determined when performing harmonic analysis on the phase response feature, determined by the highest order of the effective harmonics in the phase response feature; A n It is the amplitude value of the nth harmonic in the phase response characteristics, determined by the amplitudes of each harmonic obtained through Fourier transform decomposition; f b The frequency is the number of times the blade rotates through the base station-terminal line-of-sight path in one revolution divided by the rotation period, and is calculated and determined by the blade rotation period and the number of blades; t is the time factor used in time-aligned data, which corresponds to the timestamp of real-time data. is the initial phase of the nth harmonic in the phase response characteristics, which is determined by the initial phase of each harmonic obtained through Fourier transform decomposition; R is the characteristic radius of the wind turbine blade, which comes from the inherent parameters of the wind turbine and is half the blade length.

[0181] Furthermore, F(*) represents the Doppler frequency shift correlation value. This formula describes the effect of blade rotation on the Doppler frequency shift through two parts, where K dThe part composed of ω(t), R, and cos(θ(t)+φ0) reflects the linear relationship between real-time angular velocity, blade position, and Doppler frequency shift, demonstrating the immediate influence of blade motion on frequency shift; while the sum of each harmonic term reflects the periodic fluctuation in phase response characteristics, demonstrating the periodic change in frequency shift caused by the periodicity of blade rotation. The overall function combines linear real-time changes with periodic fluctuations, fully relating blade motion to Doppler frequency shift.

[0182] Furthermore, as ω(t) increases, if θ(t) remains constant and cos(θ(t)+φ0) is positive, K d The component ·ω(t)·R·cos(θ(t)+φ0) increases with increasing angular velocity; if the cosine value is negative, it decreases with increasing angular velocity. When θ(t) changes, cos(θ(t)+φ0) fluctuates between -1 and 1 with the angle, causing this linear component to exhibit corresponding fluctuations. Simultaneously, Part of the frequency fluctuates periodically over time according to integer multiples of the blade's passing frequency. The superposition of each harmonic forms a periodic change. Overall, the Doppler frequency shift correlation value is a superposition of linear change trend and periodic fluctuation, which dynamically changes with angular velocity and included angle and exhibits periodic fluctuation characteristics.

[0183] In summary, injecting the periodic characteristics of the wind turbine's trajectory into the phase compensation parameters to obtain anti-blocking beam configuration commands enables the beam adjustment of the 5G base station to dynamically adapt to the wind turbine's movement.

[0184] In general, the rotation of wind turbine blades has a clear periodicity, and the impact of their motion trajectory on signal obstruction also exhibits a periodic pattern. By incorporating this characteristic into the phase compensation parameters, the beam configuration can predict the timing and spatial range of blade obstruction in advance, enabling the beam to make adaptive adjustments to its direction and width before obstruction occurs, thereby reducing the interference of obstruction on signal coverage from the root.

[0185] In summary, this process extracts the angular velocity time-series data of the blade rotation, derives the Doppler frequency shift correlation function and phase compensation gradient, and integrates dynamic compensation logic into beam configuration commands. This allows the commands to not only meet current signal coverage requirements but also continuously adapt to real-time changes caused by the periodic movement of the wind turbines. This predictive beam configuration significantly improves the ability of 5G base stations to resist dynamic obstruction, ensuring the stability and continuity of 5G signal coverage within the wind farm.

[0186] S6. Send the anti-obstruction beam configuration command to the 5G base station.

[0187] Specifically, the generated anti-shading beam configuration instructions are extracted from the system. These instructions include specific parameters such as beam pointing angle, beamwidth, transmit power, and effective time window. These parameters are determined based on the coverage blind spot prediction results and wind turbine operation characteristic set, and are used to guide 5G base stations to adjust their beams to resist the effects of shading.

[0188] Furthermore, the anti-obstruction beam configuration command is converted into a command format supported by the 5G base station. This format is a binary command code preset inside the base station, which includes a command header, a parameter segment, and a check segment. The command header is used to identify the command type as anti-obstruction beam configuration, the parameter segment is filled with the encoded values ​​of beam-related parameters, and the check segment is generated by calculating the checksum of the parameter segment to ensure that no errors occur during command transmission.

[0189] Furthermore, the anti-obstruction beam configuration command is transmitted via a wired Ethernet link between the 5G base station and the control center. The link uses the TCP / IP protocol for data transmission. The control center, as the sending end, first establishes a network connection with the 5G base station. After the connection is successful, the command data is encapsulated into data packets according to the protocol format and sent to the command receiving port of the 5G base station.

[0190] Furthermore, the instruction receiving module of the 5G base station listens to a designated port. When it receives a data packet, it first decapsulates the data packet, extracts the instruction header, parameter segment, and check segment, calculates the checksum of the parameter segment and compares it with the value of the check segment. If they match, the instruction is determined to be complete and valid. If they do not match, a retransmission request is sent to the control center until a complete and valid instruction is received.

[0191] Furthermore, after confirming the validity of the anti-obstruction beam configuration command, the 5G base station parses the parameters in the command into control signals that can be recognized by the base station beam control module. At the same time, it records the reception time and effective time window of the command. When the effective time window arrives, the beam control module adjusts the beam parameters of the antenna array according to the control signal, completes the beam configuration, and sends a response message of successful configuration back to the control center.

[0192] In summary, sending anti-obstruction beam configuration commands to 5G base stations can transform the optimization strategies generated based on the actual wind farm scenario into actual base station operations, achieving a closed loop from theoretical optimization to actual signal coverage improvement. This command integrates coverage blind spot prediction results, beam adjustment parameters, and the periodic characteristics of wind turbine movement, providing a precise configuration solution for the dynamic electromagnetic environment of wind farms. After receiving and executing the command, the base station can adjust the beam direction, width, and phase compensation status in real time.

[0193] In summary, this process ensures that 5G base stations can dynamically adapt their signal transmission strategies based on changes in shading and multipath interference caused by wind turbine movement in the wind farm. This effectively counteracts the periodic shading caused by blade rotation, reduces signal attenuation and coverage blind spots caused by shading and interference, and ultimately achieves a stable improvement in the quality of 5G signal coverage in the wind farm, ensuring the continuity and reliability of communication within the wind farm.

[0194] In the several embodiments provided by this invention, it should be understood that the disclosed method can be implemented in other ways.

[0195] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0196] The embodiments of this application can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence is the theory, method, and technology that uses digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results.

[0197] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for 5G signal coverage optimization of a wind farm, characterized in that, The method includes: S1. Acquire real-time data on the blade rotation angle and speed of the wind turbine, and generate a wind turbine operation feature set for the wind turbine based on the mapping relationship between the wind turbine's motion trajectory and signal attenuation in the real-time data after time alignment, including: Extract signal strength time-series data from real-time data after time alignment; The blade rotation angle and speed data are segmented by a sliding window to obtain a motion slice of the wind turbine. Extract the occlusion correlation factor between the change in the spatial position of the blades and the fluctuation of the signal intensity within the motion slice; A speed-attenuation transfer function model is established based on the periodic characteristics of the rotational speed and the occlusion correlation factor. By fusing the parameter set of the speed-deceleration transfer function model with the rate of change of the blade spatial position of the wind turbine, the wind turbine operating characteristic set of the wind turbine is obtained. S2. By fusing the network state feature set and the wind turbine operation feature set of the wind farm, a digital twin model of the wind farm under electromagnetic propagation is obtained, including: The base station signal strength data in the network state feature set are timestamped to obtain the synchronization signal strength sequence of the wind farm; Extract the blade rotation angle data from the wind turbine operating feature set, and determine the real-time spatial position of the blade edge in the blade rotation angle data through three-dimensional spatial coordinate transformation to obtain the dynamic shading profile of the wind farm. Using the location of the terminal in the 5G base station as the endpoint, electromagnetic wave diffraction simulation is performed on the wind farm based on the synchronization signal strength sequence and the dynamic occlusion profile to obtain the path loss distribution map of the wind farm. Based on the path loss distribution map and the channel quality index in the network state feature set, a digital twin model of the wind farm is constructed. The method involves using the location of the terminal in the 5G base station as the endpoint, and performing electromagnetic wave diffraction simulation on the wind farm based on the synchronization signal strength sequence and the dynamic occlusion profile to obtain the path loss distribution map of the wind farm, including: Based on the spatial coordinate data of the dynamic occlusion contour, the spatial geometric relationship between the blade edge and the line-of-sight path of the base station-terminal is identified, and the obstacle occlusion determination result of the wind farm is obtained. When the obstacle occlusion determination result is partial occlusion, the direct signal component in the synchronization signal intensity sequence is extracted; Based on the terrain elevation data of the terminal, the difference in propagation distance between the ground reflection path and the diffuse reflection path is determined; The direct signal component and the propagation distance difference are weighted and fused to obtain the composite field strength attenuation value of the wind farm; The composite field strength attenuation value is mapped to the base station-terminal line-of-sight path to obtain the path loss distribution map of the wind farm; S3. Based on the digital twin model, analyze the spatiotemporal distribution characteristics of the wind farm under multipath interference and signal blockage, and obtain the coverage blind zone prediction results of the wind farm. S4. Based on the coverage blind spot prediction results, adjust the beam pointing angle and beam width of the 5G base station in the wind farm to obtain the phase compensation parameters of the 5G base station. S5. Inject the periodic characteristics of the wind turbine's motion trajectory into the phase compensation parameters to obtain the anti-blocking beam configuration command for the 5G base station; S6. Send the anti-obstruction beam configuration command to the 5G base station.

2. The method for optimizing 5G signal coverage in a wind farm as described in claim 1, characterized in that, The speed-deceleration transfer function model includes: In the formula, This represents the wind turbine speed decay value. For the occlusion correlation factor, It is a natural constant. This refers to the dynamic attenuation factor of the rotating blades blocking the wind turbine. The real-time rotational speed of the wind turbine in the aforementioned wind turbine unit. The amplitude of multipath interference on the metal surface of the wind turbine unit. for, This represents the offset of the yaw angle phase in the wind turbine unit. The time factor.

3. The 5G signal coverage optimization method for wind farms as described in claim 1, characterized in that, Based on the digital twin model, the spatiotemporal distribution characteristics of the wind farm under multipath interference and signal obstruction are analyzed to obtain the coverage blind zone prediction results of the wind farm, including: A heat map of the attenuation distribution of the wind farm is constructed based on the intensity value of the signal attenuation factor in the digital twin model. The continuous regions in the attenuation distribution heatmap where the signal attenuation exceeds a predetermined threshold are marked as spatial occlusion regions. By integrating the spatial shading region with the periodic shading time window of the wind turbine operating feature set, the spatiotemporal joint blind zone prediction result of the wind farm is obtained.

4. The method for optimizing 5G signal coverage in a wind farm as described in claim 1, characterized in that, The step of adjusting the beam pointing angle and beamwidth of the 5G base station in the wind farm based on the coverage blind zone prediction results to obtain the phase compensation parameters of the 5G base station includes: The set of blind zone location coordinates in the coverage blind zone prediction results is analyzed, and the horizontal azimuth and elevation angles of the blind zone center point relative to the 5G base station are determined. Based on the horizontal azimuth and elevation angles, the beam scanning range of the antenna array in the 5G base station is determined; Based on the time interval characteristics in the coverage blind spot prediction results, the beam dwell time window of the 5G base station is divided. By combining the beam scanning range with the beam dwell time window, the dynamic beam pointing sequence of the 5G base station is obtained; Based on the interference type in the coverage blind zone prediction results, a preset beamwidth adjustment strategy is matched; The phase compensation parameters of the 5G base station are output according to the adjusted strategy.

5. The method for optimizing 5G signal coverage in a wind farm as described in claim 4, characterized in that, The step of matching a preset beamwidth adjustment strategy based on the interference type in the coverage blind zone prediction result includes: When the interference type is multipath dominant, a narrow beam high gain strategy is adopted; When the interference type is dominated by obstruction, a wide beam diversity strategy is adopted.

6. The method for optimizing 5G signal coverage in a wind farm as described in claim 5, characterized in that, The step of injecting the periodic characteristics of the wind turbine's trajectory into the phase compensation parameters to obtain the anti-obstruction beam configuration command for the 5G base station includes: Extract the angular velocity time series data of the blade rotation from the set of wind turbine operating features; Calculate the Doppler frequency shift correlation function between the angular velocity timing data and the carrier frequency of the 5G base station; Based on the phase change rate of the Doppler frequency shift correlation value in the Doppler frequency shift correlation function, the phase compensation gradient of the antenna array is derived. The phase compensation gradient is mapped to the phase shifter control codeword of the antenna element in the antenna array; Based on the time interval of the coverage blind zone prediction result, the phase shifter control codeword is injected into the phase compensation parameter to obtain the phase compensation parameter set of the 5G base station.

7. The method for optimizing 5G signal coverage in a wind farm as described in claim 6, characterized in that, The calculation of the Doppler frequency shift correlation function between the angular velocity time-series data and the carrier frequency of the 5G base station includes: The angular velocity time series data is normalized to obtain a standard angular velocity sequence; Based on the carrier frequency parameters of the 5G base station, determine the maximum theoretical Doppler frequency shift of the blade tip motion; Establish a time-aligned dataset of the standard angular velocity sequence and the real-time received signal carrier offset; By fitting the angular velocity change in the angular velocity time series data with the carrier offset of the real-time received signal in the 5G base station, the linear transfer coefficient of the 5G base station is obtained. Based on the linear transfer coefficient and the maximum theoretical Doppler frequency shift, the angular velocity-frequency shift scaling function of the 5G base station is constructed. The phase response characteristics of the 5G base station are obtained by performing a Fourier transform on the angular velocity-frequency shift proportional function. By integrating the linear transfer coefficient and the phase response characteristics, the Doppler frequency shift correlation function of the 5G base station is obtained, wherein the Doppler frequency shift correlation function is as follows: In the formula, This is the correlation value for Doppler frequency shift. The linear transfer coefficient is... This represents the real-time angular velocity of the wind turbine blades. This is the real-time angle between the blade and the base station-terminal connection. For the initial phase shift, The maximum harmonic order in the phase response characteristics is given. For the first Second harmonic amplitude For harmonic ordinal numbers, The frequency at which the blades pass through. As a time factor, The initial phase of the harmonics in the phase response characteristics, The characteristic radius of the blade.

Images