A dynamic wind speed monitoring method, device and equipment based on fiber grating

Abstract

The embodiment of the specification discloses a kind of dynamic wind speed monitoring method, device and equipment based on fiber grating, method includes: by the fiber grating wind speed instrument of multiple measuring points in wind farm, each measuring point monitoring data is collected;The waveform feature of reflected spectrum data is analyzed, and first diagnostic result is obtained, and based on the physical correlation of strain wavelength data and the fan operating state data corresponding to each measuring point, second diagnostic result is determined;Based on first diagnostic result and second diagnostic result, determine the reliable measuring point and the distorted measuring point in each measuring point in wind farm;The temperature compensation of strain wavelength data of reliable measuring point is carried out by reference temperature wavelength data, and the real-time wind speed data of reliable measuring point is obtained;According to the real-time wind speed data of each reliable measuring point, the propagation characteristic that wind field disturbance in wind farm is propagated from reliable measuring point to distorted measuring point is determined, to reconstruct the position of distorted measuring point based on propagation characteristic, generate full-field wind speed data.

Description

Technical Field

[0001] This specification relates to the field of wind speed monitoring technology, and in particular to a dynamic wind speed monitoring method, device, and equipment based on fiber Bragg gratings. Background Technology

[0002] In the field of wind power generation, especially in wind farms with complex terrain such as mountains and hills, wind speed monitoring is the foundation for unit safety control and power generation optimization. Currently, the most common methods for wind speed monitoring in wind farms include: monitoring wind speed through anemometers installed on the top of each wind turbine nacelle or through wind measurement towers within the farm.

[0003] However, in real-world scenarios, anemometers are constantly exposed to outdoor environments, especially in mountainous and coastal wind farms, where they are highly susceptible to icing, lightning strikes, dust erosion, and strong turbulent vibrations. When sensors experience physical malfunctions, such as blade icing and jamming or bearing wear, their output readings may exhibit abnormal phenomena such as returning to zero, freezing, or severe drift. Current wind turbine control systems typically cannot effectively identify the authenticity of such data, often directly using abnormal readings as the true wind speed in logical judgments, leading to unstable wind speed data reliability. Furthermore, for wind farms with complex terrain, wind speed data may exhibit a highly uneven spatial distribution, with significant wake interference between multiple turbines. Wind speed data at a single turbine nacelle only represents the wind conditions at a single point at the hub height, failing to perceive the changing trends of upstream flow and the propagation path and intensity of downstream wakes. Therefore, downstream turbines cannot predict changes in the incoming flow and take proactive measures, resulting in a lack of data foundation for coordinated optimization control across the entire wind farm.

[0004] Therefore, a wind speed monitoring method that can dynamically reconstruct the dynamic wind field of the entire field is needed. Summary of the Invention

[0005] To address the aforementioned issues, this specification provides one or more embodiments of a dynamic wind speed monitoring method, apparatus, and device based on fiber Bragg gratings.

[0006] One or more embodiments of this specification employ the following technical solutions: This specification provides one or more embodiments of a dynamic wind speed monitoring method based on fiber Bragg gratings, the method comprising: The monitoring data of each measuring point is collected by fiber Bragg grating anemometers at multiple measuring points within the wind farm; wherein the monitoring data includes at least: fiber Bragg grating reflection spectrum data, strain wavelength data, and reference temperature wavelength data; The waveform characteristics of the reflection spectrum data are analyzed to obtain a first diagnostic result, and a second diagnostic result is determined based on the physical correlation between the strain wavelength data and the wind turbine operating status data corresponding to each measuring point. Based on the first diagnostic result and the second diagnostic result, reliable measuring points and distorted measuring points are determined among the measuring points in the wind farm. Temperature compensation is performed on the strain wavelength data of the reliable measuring point using the reference temperature wavelength data to obtain the real-time wind speed data of the reliable measuring point. Based on the real-time wind speed data of each trusted measuring point, the propagation characteristics of wind field disturbances in the wind farm from the trusted measuring points to the distorted measuring points are determined, and the wind field is reconstructed at the location of the distorted measuring points based on the propagation characteristics to generate full-field wind speed data.

[0007] Optionally, in one or more embodiments of this specification, the fiber Bragg grating anemometer includes at least a first fiber Bragg grating and a second fiber Bragg grating; Wherein, the first fiber grating is a strain sensing grating used to output the strain wavelength data; the second fiber grating is a temperature compensation grating, disposed adjacent to the first fiber grating and isolated from external strain, used to output the reference temperature wavelength data.

[0008] Optionally, in one or more embodiments of this specification, the plurality of measuring points include: at least two measuring points distributed along the height direction of each wind turbine tower in the wind farm, and at least one measuring point distributed along the line connecting adjacent wind turbines in the wind farm. Specifically, monitoring data is collected from multiple fiber optic anemometers at various points within the wind farm, including: The multiple fiber Bragg grating anemometers are connected in series to the same transmission fiber to construct a fiber Bragg grating sensing network; each fiber Bragg grating anemometer in the fiber Bragg grating sensing network is configured with a non-overlapping initial reflection wavelength. By transmitting broadband optical signals into the transmission optical fiber, and based on the initial reflection wavelength of each fiber optic anemometer, wavelength demodulation is performed on the returned reflection spectrum data to determine the center wavelength of the strain sensing grating and temperature compensation grating at each measuring point. The center wavelengths of the strain sensing grating and the temperature compensation grating are used as the strain wavelength data and the reference temperature wavelength data, respectively.

[0009] Optionally, in one or more embodiments of this specification, the plurality of measuring points includes: a reference measuring point and a target measuring point obtained based on the dominant wind direction of the current cycle of the wind farm; The reference measuring point is the measuring point located on the upwind side of the prevailing wind direction among the plurality of measuring points, used to provide wind field boundary condition data; the target measuring point is the measuring point located on the downwind side of the prevailing wind direction among the plurality of measuring points. The method further includes: Based on the historical wind field data within the preset period of the wind farm, the dominant wind direction of the wind farm is determined; Based on the spatial distance between the benchmark measuring point and the target measuring point and the prevailing wind direction, the direction vector and distance parameters between each measuring point are determined to establish the spatial topological relationship between each measuring point.

[0010] Optionally, in one or more embodiments of this specification, the waveform characteristics of the reflectance spectral data are analyzed to obtain a first diagnostic result, and a second diagnostic result is determined based on the physical correlation between the strain wavelength data and the wind turbine operating status data corresponding to each of the measuring points, specifically including: The waveform features of the reflectance spectral data are analyzed to extract the waveform feature parameters corresponding to the reflectance spectral data; wherein, the waveform feature parameters include: full width at half maximum (FWHM) and sidelobe suppression ratio; By comparing the waveform feature parameters with the corresponding preset threshold, it is determined that the measurement points where the waveform feature parameters exceed the preset threshold have sensor abnormalities, and a first diagnostic result is obtained. The operating status data of the wind turbine corresponding to the measuring point is obtained, and the strain wavelength data is converted into wind speed value; By comparing the wind speed value with the operating status data through aerodynamic characteristic correlation, it is determined that the measurement point data is distorted due to the deviation of the correlation between the wind speed value and the operating status data from the preset aerodynamic characteristic mapping relationship, and a second diagnostic result is obtained.

[0011] Optionally, in one or more embodiments of this specification, the propagation characteristics of wind field disturbances in the wind farm from the trusted measuring points to the distorted measuring points are determined based on the real-time wind speed data of each trusted measuring point, specifically including: Based on the spatial location and real-time wind direction of each measuring point, the spatial topology is queried to obtain the reference measuring point located on the windward side of the distorted measuring point. When the reference measuring point is a reliable measuring point, it is used as the candidate source measuring point corresponding to the distorted measuring point. Based on the spatial coordinates of each candidate source measurement point and the distortion measurement point, the spatial distance between each candidate source measurement point and the distortion measurement point is determined. Based on the spatial distance and the real-time wind speed at the candidate source measurement points, determine the time delay required for wind field disturbances to propagate from each of the candidate source measurement points to the distortion measurement point; The propagation characteristics are determined based on the time delay and the preset attenuation coefficient; the propagation characteristics include at least the time delay and the wind speed attenuation when the disturbance propagates to the distortion measurement point.

[0012] Optionally, in one or more embodiments of this specification, the step of reconstructing the wind field at the location of the distorted measurement point based on the propagation characteristics to generate full-field wind speed data specifically includes: Obtain real-time wind speed data of the candidate source measurement points corresponding to the distorted measurement points; Based on the time delay in the propagation characteristics, the real-time wind speed data of each candidate source measuring point is shifted along the time axis to obtain the delayed wind speed data at the distorted measuring point. Based on the wind speed attenuation in the propagation characteristics, the delayed wind speed data is amplitude corrected to obtain the corrected wind speed data of the distorted measurement point. Reconstructed wind speed data corresponding to each candidate source measurement point is obtained, and the reconstructed wind speed data of multiple reconstructed wind speed data are weighted and fused, wherein the weight of the weighted fusion is determined inversely proportional to the spatial distance between each candidate source measurement point and the distorted measurement point; The wind speed data of the wind farm is constructed by combining the reconstructed wind speed data from each distorted measuring point with the real-time wind speed data from each reliable measuring point.

[0013] Optionally, in one or more embodiments of this specification, the method further includes: Based on the first and second diagnostic results corresponding to each of the aforementioned reliable measurement points, the data confidence level of each of the aforementioned reliable measurement points is determined; The real-time wind speed data of the credible measuring points are weighted and corrected according to the confidence level, so as to determine the propagation characteristics of wind field disturbances in the wind farm from the credible measuring points to the distorted measuring points based on the corrected real-time wind speed data of each credible measuring point. After reconstructing the wind field of the distorted measuring points based on the propagation characteristics, the inference confidence of the reconstructed wind speed data is determined according to the data confidence level of the candidate source measuring points corresponding to each distorted measuring point, the spatial distance between each candidate source measuring point and the distorted measuring point, and the wind direction deviation between the real-time wind direction of each candidate source measuring point and the real-time wind direction of the distorted measuring point. If the projection confidence level is lower than the preset projection confidence level threshold, the reconstructed wind speed data of the distorted measurement point is marked as unusable.

[0014] This specification provides one or more embodiments of a dynamic wind speed monitoring device based on a fiber Bragg grating, the device comprising: The data acquisition unit is used to acquire monitoring data from multiple fiber Bragg grating anemometers within the wind farm; wherein the monitoring data includes at least: fiber Bragg grating reflection spectrum data, strain wavelength data, and reference temperature wavelength data. The diagnostic unit is used to analyze the waveform characteristics of the reflection spectrum data to obtain a first diagnostic result, and to determine a second diagnostic result based on the physical correlation between the strain wavelength data and the wind turbine operating status data corresponding to each of the measuring points. The determining unit is used to determine the reliable measuring points and distorted measuring points among the measuring points in the wind farm based on the first diagnostic result and the second diagnostic result; The compensation unit is used to perform temperature compensation on the strain wavelength data of the reliable measuring point using the reference temperature wavelength data, so as to obtain the real-time wind speed data of the reliable measuring point. The reconstruction unit is used to determine the propagation characteristics of wind field disturbances in the wind farm from the trusted measuring points to the distorted measuring points based on the real-time wind speed data of each trusted measuring point, so as to reconstruct the wind field at the location of the distorted measuring point based on the propagation characteristics and generate wind speed data for the entire field.

[0015] This specification provides one or more embodiments of a dynamic wind speed monitoring device based on a fiber Bragg grating, the device comprising: at least one processor; and, A memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform any of the methods described above.

[0016] The above-described at least one technical solution used in the embodiments of this specification can achieve the following beneficial effects: By using fiber optic anemometers at multiple measuring points within the wind farm, monitoring data from each point is collected, solving the problems of single strain signals failing to identify spectral anomalies in the sensor itself and failing to eliminate temperature interference. Two layers of diagnostics ensure the reliability of the selected data, enabling data credibility assessment. A temperature-compensated grating senses the ambient temperature in real time, compensating the strain wavelength data at reliable measuring points to obtain real-time wind speed data. This eliminates the impact of ambient temperature fluctuations on measurement accuracy. Independent compensation at each measuring point avoids the spatial representativeness problem of global compensation, improving the accuracy and environmental adaptability of wind speed measurement. Based on the physical law of wind field disturbances propagating from upstream to downstream along the wind direction, wind field reconstruction is performed on distorted measuring points, filling data gaps when sensors fail. This yields full-field wind speed data covering all measuring points, achieving full-field wind field sensing coverage without additional hardware investment. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments or prior art of this specification, the drawings used in the description of the embodiments or prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this specification. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. In the drawings: Figure 1A schematic flowchart of a dynamic wind speed monitoring method based on fiber Bragg grating provided for embodiments of this specification; Figure 2 A schematic diagram of a fiber optic grating anemometer provided in the embodiments of this specification; Figure 3 This is a schematic diagram of the structure of a fiber Bragg grating in an application scenario provided by an embodiment of this specification; Figure 4 This is a schematic diagram of the measurement point layout in an application scenario provided by an embodiment of this specification; Figure 5 A schematic diagram of a dynamic wind speed monitoring device based on a fiber Bragg grating provided in the embodiments of this specification; Figure 6 This is a schematic diagram of a dynamic wind speed monitoring device based on a fiber Bragg grating, provided as an embodiment of this specification. Detailed Implementation

[0018] This specification provides a method, apparatus, and device for dynamic wind speed monitoring based on fiber Bragg gratings.

[0019] To enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this specification, and not all embodiments. Based on the embodiments of this specification, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this specification.

[0020] Figure 1 This specification provides a dynamic wind speed monitoring method, apparatus, and device based on fiber Bragg gratings. The following detailed description of each step of the method, in conjunction with specific embodiments, includes the following steps: S101: Collect monitoring data from multiple fiber Bragg grating anemometers at various points within the wind farm; wherein the monitoring data includes at least: fiber Bragg grating reflection spectrum data, strain wavelength data, and reference temperature wavelength data.

[0021] like Figure 2 The existing anemometer shown here consists of a wind cup, a rotating shaft, a transmission gear shaft, and a fiber optic grating spring as its main components. The wind cup is as follows: Figure 2The device shown consists of three hemispherical cups used to sense wind force. The propeller, composed of the cups and a rotating shaft below, senses wind speed through rotation; the higher the wind speed, the faster the propeller rotates. A transmission gear shaft below the propeller actuates a grating spring, which is connected in series with a certain position on the same transmission fiber. When the gear of the fiber optic anemometer contacts the spring, the wavelength on the spring changes; when it doesn't contact, the wavelength returns to its original value. Therefore, the wind speed is calculated based on the magnitude of the wavelength change. However, if… Figure 2 Existing fiber Bragg grating solutions typically acquire only a single strain signal, lacking multi-dimensional data for self-diagnosis and temperature compensation. However, in practical engineering applications, the reflection spectrum of fiber Bragg gratings is easily distorted in harsh outdoor environments due to factors such as grating deformation from compression, encapsulation detachment, or drastic temperature changes. A single strain wavelength signal cannot reflect changes in the sensor's own health status. Furthermore, the strain wavelength signal itself is affected by both temperature and strain; without temperature reference data, temperature drift cannot be mitigated, leading to decreased wind speed measurement accuracy.

[0022] To address the aforementioned issues, this embodiment utilizes fiber optic grating anemometers distributed across multiple measuring points within the wind farm to collect monitoring data. It should be noted that in this embodiment, the sensing fiber or grating of the fiber optic anemometer is directly bonded or embedded to the surface of the wind turbine tower or blades, requiring no rotating parts or transmission mechanisms. When wind loads act on the tower or blades, the structure experiences corresponding strain. The fiber optic grating deforms along with the structure and outputs a wavelength drift signal corresponding to this strain. This signal is then processed by a demodulator to obtain the strain wavelength data. Furthermore, to compensate for temperature drift, the fiber optic anemometer also includes a reference fiber optic grating. This reference grating is either in a free state unaffected by wind forces or only thermally coupled to the environment, used to sense real-time changes in ambient temperature and output reference temperature wavelength data.

[0023] This process solves the technical challenge of fiber Bragg gratings being sensitive to both temperature and strain by using real-time ambient temperature sensing and temperature drift compensation for strain wavelength data. This avoids wind speed measurement drift caused by ambient temperature changes, improving the accuracy and environmental adaptability of wind speed measurements. It also provides a multi-dimensional data foundation, including strain and temperature information, for subsequent reflectance spectrum self-diagnosis and spatial wind field extrapolation. This helps the entire monitoring system to promptly identify physical anomalies in sensors and fill data gaps through spatial extrapolation when sensors fail.

[0024] Preferably, in one feasible embodiment of this specification, the fiber Bragg grating anemometer includes at least a first fiber Bragg grating and a second fiber Bragg grating. The first fiber Bragg grating is a strain-sensing grating used to output strain wavelength data; the second fiber Bragg grating is a temperature-compensating grating disposed adjacent to the first fiber Bragg grating and isolated from external strain, used to output reference temperature wavelength data.

[0025] Specifically, in one feasible embodiment of this specification, such as Figure 3 The fiber Bragg grating anemometers at each measuring point employ a paired dual-grating structure. The first fiber Bragg grating, serving as a strain-sensing grating, is fixed to the surface of the wind turbine tower wall or blades via adhesive bonding or embedding. When wind loads act on the tower or blade structure, the structure experiences corresponding bending or tensile strain. The first fiber Bragg grating, fixed to the structural surface, deforms synchronously with the structure, causing a shift in its Bragg reflection wavelength. This wavelength shift reflects the structural strain level caused by the wind load, and the strain wavelength data is output after demodulation by the demodulator. The second fiber Bragg grating, serving as a temperature-compensating grating, is positioned adjacent to the first fiber Bragg grating to ensure both are in the same temperature field environment. The second fiber Bragg grating is constructed using a strain-isolated encapsulation method, such as suspending the grating within a capillary tube or placing it in a protective sleeve unaffected by external forces. Figure 3 The device is located within a strain-isolated protective sleeve, allowing it to sense only changes in ambient temperature without experiencing any structural strain. Since the second fiber grating is in a free state, the drift in its Bragg reflection wavelength is solely caused by changes in ambient temperature. Therefore, its output wavelength data serves as reference temperature wavelength data to characterize the real-time ambient temperature at the measurement point. Through this configuration of at least two gratings, real-time self-compensation for temperature drift is achieved, avoiding interference from ambient temperature changes on wind speed measurement results and improving the accuracy and environmental adaptability of wind speed measurement.

[0026] Furthermore, in one feasible embodiment of this specification, the multiple measuring points include: at least two measuring points distributed along the height direction of each wind turbine tower within the wind farm, and at least one measuring point distributed along the line connecting adjacent wind turbines within the wind farm. That is, at least two measuring points are respectively arranged along the height direction on the tower of each target wind turbine. These vertically distributed measuring points are used to capture the gradient of wind speed changes along the height direction, providing vertical data constraints for subsequent wind profile extrapolation, and can also help identify whether the incoming wind is affected by terrain or wake and undergoes vertical shear. It should be understood that the actual arrangement height can be adjusted according to the actual height of the wind turbine tower and terrain features. Arranging at least one measuring point along the line connecting adjacent wind turbines allows for direct acquisition of wind speed change data along the wake propagation path. When the wake generated by the upstream wind turbine propagates downstream along the prevailing wind direction, the measuring point located in the middle of the line can capture the wind speed attenuation and turbulence evolution process in the wake region in real time, providing reliable path constraint data for spatial extrapolation, and providing a practical basis for the correction of subsequent distorted measuring points.

[0027] Specifically, monitoring data is collected from multiple fiber optic anemometers at various points within the wind farm, including: Multiple fiber Bragg grating anemometers are connected in series to the same transmission fiber to construct a fiber Bragg grating sensing network. Each fiber Bragg grating anemometer contains at least a first fiber Bragg grating and a second fiber Bragg grating. Each fiber Bragg grating anemometer in the sensing network is configured with non-overlapping initial reflection wavelengths; that is, the Bragg wavelengths of any two fiber Bragg gratings are different throughout the entire sensing network, ensuring that the reflected signals of each grating are separated and do not interfere with each other in the wavelength domain. This process simultaneously acquires multiple sets of data through the same transmission fiber and a single wavelength scan, without increasing system complexity or hardware cost.

[0028] During the data acquisition process, a broadband light source emits a broadband optical signal covering the entire wavelength range of the gratings into the transmission optical fiber. This broadband optical signal propagates sequentially along the fiber to the fiber gratings at each measurement point. Based on the wavelength-selective reflection principle of fiber Bragg gratings, when a wavelength of the broadband light matches the Bragg wavelength of a particular grating, that wavelength is reflected back to the demodulator by that grating. Other wavelengths continue to propagate along the fiber to the next grating. The demodulator receives the reflected light signals from all gratings and analyzes the reflection spectrum. Since each grating has a non-overlapping initial reflection wavelength, the demodulator can uniquely identify which type of grating at which measurement point a reflection peak originates from based on the wavelength position of the reflection peak. Based on this, the demodulator performs wavelength demodulation on the reflection spectrum based on the initial reflection wavelength of each grating to determine the center wavelength of the strain-sensing grating and temperature-compensating grating at each measurement point. Specifically, peak detection is performed on the reflection spectrum to identify the wavelength position of each reflection peak and obtain the center wavelength value of each reflection peak. The center wavelength values ​​of each reflection peak are compared with a preset mapping table of initial wavelengths, measurement points, and grating types to determine the measurement point number and grating type to which each reflection peak belongs. Then, the current center wavelength of the strain-sensing grating and the current center wavelength of the temperature-compensated grating at each measurement point are extracted as strain wavelength data and the reference temperature wavelength data, respectively. In this process, the demodulator can simultaneously acquire the strain wavelength data and reference temperature wavelength data of each sensing grating at all measurement points in a single wavelength scan, eliminating the need for a separate demodulation channel or time-division polling for each measurement point. This ensures that the hardware cost of the fiber Bragg grating anemometer does not increase linearly with the number of measurement points. Furthermore, in one or more embodiments of this specification, the multiple measuring points include: a reference measuring point and a target measuring point. The reference measuring point is located on the windward side of the prevailing wind direction in the current cycle, and is used to provide wind field boundary condition data, i.e., inflow wind speed, wind direction, and turbulence intensity information before the wind enters the wind farm. The target measuring point is a measuring point located on the leeward side of the prevailing wind direction among the multiple measuring points, and is the measuring point whose wind speed data needs to be obtained through spatial extrapolation.

[0029] It should be noted that the prevailing wind direction of the wind farm in the current cycle refers to the wind direction corresponding to the sector with the highest frequency of occurrence in the area where the wind farm is located within the current cycle. In one feasible embodiment of this specification, the prevailing wind direction in the current cycle can be determined through the following process: obtaining the prevailing wind direction in the historical cycle corresponding to the current cycle of the wind farm, such as... Figure 3The historical wind field data recorded by the anemometer shown can be determined by those skilled in the art based on the wind climate characteristics of the wind farm's location. For example, if the period is quarterly, the quarter corresponding to the current quarter in the historical calendar can be used as the historical period to obtain the historical wind field data within that period. The wind direction data is divided into several sectors, and the frequency of wind direction data occurrence within each sector is statistically analyzed. The center direction of the sector with the highest frequency of occurrence is taken as the dominant wind direction for the wind farm in the current period.

[0030] It is understandable that the role of the same measuring point may change in cycles with different prevailing wind directions. For example, when the prevailing wind direction for the wind farm in the current cycle is westerly, the measuring point located on the west side is the baseline measuring point, and the measuring point located on the east side is the target measuring point. If the prevailing wind direction changes in the next cycle, such as turning to easterly, the role of the same measuring point may be interchanged in the next cycle. Therefore, the division of baseline and target measuring points is relative to the prevailing wind direction determined in the current cycle, and is not an absolute limitation on the function of each measuring point.

[0031] After determining the prevailing wind direction for the current period and establishing the distinction between reference and target measuring points, the spatial topology relationships between the measuring points are established based on their spatial coordinates. Specifically, for any pair of reference and target measuring points, the direction vector and distance parameters are calculated based on their spatial coordinates. That is, the direction vector from the reference measuring point to the target measuring point is obtained by subtracting the spatial coordinates of the reference measuring point from the spatial coordinates of the target measuring point. Based on this, and combined with the prevailing wind direction vector for the current period, the angle between the prevailing wind direction vector and the obtained direction vector is calculated. Simultaneously, the spatial distance between the reference and target measuring points is obtained. This angle and spatial distance together constitute the spatial topology parameters between the pair of measuring points.

[0032] The spatial topology parameters corresponding to each benchmark measuring point and each target measuring point are summarized to construct the spatial topology relationship between each measuring point. This spatial topology relationship database can be stored in the form of key-value pairs or data tables.

[0033] Those skilled in the art should understand that the above-mentioned number of measurement points, specific structural parameters of fiber optic gratings, and demodulation methods are merely illustrative examples and do not constitute a limitation on the scope of protection of this invention. For example, the number of measurement points and spatial distribution density can be flexibly adjusted according to the actual area and terrain conditions of the wind farm, and the initial wavelength interval of each grating can be flexibly configured according to the parameters of fiber optic gratings from different manufacturers and the wavelength bandwidth used.

[0034] S102: Analyze the waveform characteristics of the reflection spectrum data to obtain a first diagnostic result, and determine a second diagnostic result based on the physical correlation between the strain wavelength data and the wind turbine operating status data corresponding to each of the measurement points.

[0035] In actual wind farm operation, fiber Bragg grating anemometers are exposed to harsh outdoor environments for extended periods. Their reflectance spectra may become distorted due to the following reasons: non-axial stress on the grating causing side lobes or secondary peaks; encapsulation delamination leading to full width at half maximum (FWHM) broadening; connector contamination causing a decrease in reflectivity; and drastic temperature changes causing abnormal wavelength drift. If the system relies solely on a single value from the strain wavelength data output, it cannot detect these spectral anomalies and will continue to use data from damaged sensors, resulting in erroneous wind speed measurements being continuously input into the control system. More seriously, when the anemometer's readings become stuck due to icing or drift due to mechanical failure, conventional systems lack cross-validation mechanisms to identify anomalies. This can ultimately lead to two serious consequences: failure to shut down the system in time under extreme wind conditions causing safety accidents, or accidental shutdown under normal wind conditions resulting in power generation losses.

[0036] Therefore, this embodiment employs a two-layer diagnostic approach, using both sensor hardware malfunction and measured data reasonableness to jointly determine the data reliability of each measuring point, providing a reliable data source for subsequent steps. Specifically, after obtaining monitoring data in step S101, this embodiment analyzes the waveform characteristics of the reflectance spectrum data to obtain a first diagnostic result, and determines a second diagnostic result based on the physical correlation between strain wavelength data and the corresponding wind turbine operating status data for each measuring point. This diagnostic process allows subsequent spatial extrapolation to select only reliable measuring points as the extrapolation source, fundamentally avoiding the inclusion of sensor malfunctions or physical anomalies in the extrapolation process, thus ensuring that the reconstructed full-field wind speed data has a reliable foundation.

[0037] Specifically, in one feasible embodiment of this specification, the waveform characteristics of the reflectance spectral data are analyzed to obtain a first diagnostic result, and a second diagnostic result is determined based on the physical correlation between the strain wavelength data and the wind turbine operating status data corresponding to each measuring point, specifically including: The waveform characteristics of the reflectance spectral data are analyzed to extract corresponding waveform feature parameters, including the full width at half maximum (FWHM) and the sidelobe suppression ratio (SPR). The FWHM is the wavelength width corresponding to half the maximum reflectance of the main peak of the reflectance spectrum. Under normal sensor conditions, the reflectance spectrum of a fiber optic grating exhibits a standard Gaussian distribution, and its FWHM is determined by the grating's design parameters and is fixed at the factory; it should remain essentially constant under normal operating conditions. When the grating is subjected to non-axial stress compression, chirping, or encapsulation delamination, the FWHM will significantly broaden. Furthermore, when the grating is partially broken or annealed, the FWHM may also exhibit abnormal changes. The SPR is the ratio of the peak value of the main peak to the peak value of the largest sidelobe in the reflectance spectrum. For a uniform fiber optic grating, theoretically, there are several symmetrically distributed sidelobes on both sides of the main peak, but the sidelobe intensity is much lower than that of the main peak. When the grating is subjected to a non-uniform stress field, the sidelobe intensity will increase significantly, and the SPR will decrease accordingly. When the sidelobe suppression ratio is lower than the preset threshold, it means that the grating is subjected to non-uniform stress distribution along the length of the grating region, which may result in local debonding or abnormal stress, affecting the accuracy and reliability of strain measurement.

[0038] During the diagnostic process, waveform feature parameters are compared with corresponding preset thresholds to identify measurement points where waveform feature parameters exceed the preset thresholds, indicating sensor malfunction and obtaining a first diagnostic result. Specifically, when any waveform feature parameter exceeds the corresponding preset threshold range, such as when the full width at half maximum (FWHM) is greater than or equal to a preset width threshold or the sidelobe suppression ratio (SPR) is less than or equal to a preset suppression ratio threshold, the measurement point is determined to have a sensor malfunction, and a first diagnostic result is generated marking the measurement point as sensor malfunctioning. It should be understood that the specific thresholds for FWHM and SPR can be reasonably set by those skilled in the art based on the selected fiber grating model, specifications, and engineering experience in practical applications; this specification does not specifically limit these settings in the embodiments.

[0039] Furthermore, it should be noted that, in the specific implementation process, other waveform characteristic parameters such as the number of peaks, the main peak wavelength shift, and the spectral symmetry can be further extracted as supplementary diagnostic results to the first diagnostic result, depending on the actual diagnostic needs.

[0040] Based on the above, the first diagnostic result determines whether the grating is in physical health from the sensor hardware level. However, even if the sensor hardware is normal, the output strain wavelength data may deviate from the actual wind conditions due to icing, contamination, loose installation, or other reasons. Therefore, in this embodiment, the operating status data of the wind turbine corresponding to the measuring point is further obtained, and the strain wavelength data is converted into wind speed values. Specifically, the strain wavelength data can be converted into wind speed measurement values ​​according to preset strain and wind speed calibration coefficients. These calibration coefficients can be obtained through on-site calibration before the wind farm is put into operation. For example, wind speed data can be collected simultaneously at the anemometer tower and the fiber optic grating measuring point, and the correspondence between strain wavelength and wind speed can be established through regression analysis.

[0041] Then, the operating status data of the wind turbine corresponding to the measuring point is acquired, including but not limited to: generator speed, generator power, blade pitch angle, and nacelle vibration acceleration. Based on this, the calculated wind speed value is compared with the operating status data using aerodynamic characteristics. It should be noted that under normal operating conditions, there is a definite aerodynamic characteristic mapping relationship between the wind turbine's power generation and speed and the incoming wind speed. This mapping relationship is usually pre-stored in the control system or database in the form of aerodynamic characteristic curves, generally based on the wind turbine's factory design parameters or on-site calibration. During the diagnostic process, the current measured wind speed value is substituted into the aerodynamic characteristic curve to retrieve the corresponding theoretical power value or theoretical speed value, and then compared with the operating status data. When the deviation between the theoretical and actual values ​​exceeds a preset tolerance range, the wind speed data at that measuring point is determined to be distorted, and a second diagnostic result is generated, marking the measuring point as having data distortion.

[0042] S103: Based on the first diagnostic result and the second diagnostic result, determine the reliable measuring points and the distorted measuring points among the measuring points in the wind farm.

[0043] In the embodiments of this specification, after obtaining the first and second diagnostic results for each measuring point, the two diagnostic results are comprehensively evaluated to determine the final data reliability status of each measuring point within the wind farm. Specifically, in one feasible embodiment of this specification, when the first diagnostic result determines that the measuring point has a sensor malfunction, or the second diagnostic result determines that the measuring point's data is distorted, the measuring point is identified as a distorted measuring point. Conversely, when both the first and second diagnostic results are normal, the measuring point is identified as a reliable measuring point. That is, if either the first or second diagnostic result is determined to be abnormal, the measuring point is identified as a distorted measuring point; only when both are determined to be normal is the measuring point identified as a reliable measuring point, thereby ensuring the reliability of subsequent wind farm reconstruction.

[0044] In a preferred embodiment, after completing the diagnosis of all measuring points, the method further includes: summarizing the diagnosis results of all measuring points in the wind farm, generating a correspondence table between each measuring point and the reliable state or the distortion state, and updating it in real time to the data cache area for subsequent use.

[0045] In this process, by comprehensively judging the first and second diagnostic results, the reliability of the measurement point data was rigorously screened. This dual-dimensional judgment enabled the differentiation between two different fault modes, and also helped maintenance personnel to take differentiated measures for different fault types, optimizing the allocation of maintenance resources.

[0046] S104: Temperature compensation is performed on the strain wavelength data of the reliable measuring point using the reference temperature wavelength data to obtain the real-time wind speed data of the reliable measuring point.

[0047] As described in the background section, the strain wavelength data of fiber Bragg gratings is affected by both temperature and strain. Changes in ambient temperature cause thermal expansion and thermo-optic effects in the fiber Bragg grating, resulting in an additional drift in the Bragg wavelength. If this wavelength drift is directly attributed entirely to structural strain caused by wind load, the wind speed measurement will have a significant error due to fluctuations in ambient temperature. Therefore, before converting the strain wavelength data into a wind speed value, the temperature-induced component must first be subtracted from the total wavelength drift. Specifically, in one feasible embodiment of this specification, the fiber Bragg grating sensing unit at each reliable measurement point includes a first fiber Bragg grating, i.e., a strain sensing grating, and a second fiber Bragg grating, i.e., a temperature compensation grating.

[0048] In step S101 above, the strain wavelength data and reference temperature wavelength data of the measuring point have been obtained through wavelength demodulation. The strain sensing grating, because it directly bears the structural strain, outputs a current wavelength that includes both the strain drift caused by wind load and the thermal drift caused by ambient temperature, i.e., the total drift. The temperature compensation grating, on the other hand, uses strain-isolated encapsulation, so its output current wavelength only drifts with changes in ambient temperature and does not include a strain component; therefore, its drift is the pure temperature drift.

[0049] When performing temperature compensation using the aforementioned fiber Bragg grating anemometer settings, the total drift of the strain-sensing grating and the temperature drift of the temperature-compensating grating are first determined separately. Specifically, in one feasible approach, the wavelength drift of each grating is obtained by comparing its currently measured wavelength value with its initial wavelength value. Then, the temperature drift of the temperature-compensating grating is subtracted from the total drift of the strain-sensing grating to eliminate the influence of ambient temperature on the measurement results, yielding the net strain wavelength drift caused solely by wind load.

[0050] After obtaining the net strain wavelength shift caused solely by wind load, the net strain shift is converted into the real-time wind speed value at the current moment for the reliable measuring point, based on a pre-calibrated physical correlation. This mapping relationship, as described in step S102, can be pre-obtained through synchronous acquisition and data analysis of the wind measurement tower and fiber optic grating measuring points before the wind farm is put into operation.

[0051] In this process, the ambient temperature changes are sensed in real time by a temperature-compensated grating, and the temperature component is subtracted from the total wavelength drift. This ensures that the obtained real-time wind speed value only reflects the structural strain caused by wind load, and the measurement value is no longer affected by ambient temperature fluctuations such as sunshine and daily temperature variations, which helps improve the accuracy of wind speed measurement. The temperature-compensated grating at each measuring point independently senses the real-time ambient temperature at that point. Different measuring points can be independently compensated according to the microclimate differences of their respective locations, avoiding the problem of insufficient spatial representativeness caused by using a single temperature sensor for global compensation. Moreover, the temperature-compensated grating, which is also the second fiber optic grating, shares the same transmission fiber and demodulator as the strain sensing grating, eliminating the need for additional temperature sensors such as thermocouples or resistance temperature detectors.

[0052] S105: Based on the real-time wind speed data of each of the trusted measuring points, determine the propagation characteristics of wind field disturbances in the wind farm from the trusted measuring points to the distorted measuring points, and reconstruct the wind field at the location of the distorted measuring points based on the propagation characteristics to generate full-field wind speed data.

[0053] This specification's embodiments have obtained real-time wind speed data from all reliable measuring points, but wind speed data for distorted measuring point locations is still missing. Directly using these missing or erroneous data points for wind turbine control would result in incomplete data support for control decisions. Therefore, for each measuring point marked as distorted, the physical characteristic of wind field disturbances having a definite propagation law in space is utilized to spatially extrapolate and reconstruct the wind speed at the distorted measuring point location based on the real-time wind speed data from reliable measuring points, generating full-field wind speed data.

[0054] This refers to the spatiotemporal characteristics of airflow velocity changes caused by factors such as topography, wind turbine wake effects, and atmospheric turbulence as wind flows through a wind farm. In fiber optic wind measurement, this manifests as structural strain changes on the surface of the wind turbine tower or blades. It propagates along the wind direction within the wind farm and is sequentially sensed and tracked by sensing gratings deployed at different spatial locations. In some feasible embodiments, wind farm disturbances include at least one of the following: wind speed acceleration or deceleration caused by topography, wind speed attenuation and turbulence enhancement caused by upstream wind turbine wakes, and wind speed fluctuations caused by atmospheric thermal or dynamic effects.

[0055] Propagation characteristics are a set of parameters used to quantify the physical process of wind field disturbances propagating from candidate source measurement points to distorted measurement points.

[0056] The full-field wind speed data is a dataset covering all measuring points within the wind farm. The wind speed values ​​at reliable measuring points are the measured wind speed values ​​after temperature compensation, while the wind speed values ​​at distorted measuring points are the extrapolated wind speed values ​​obtained through wind field reconstruction.

[0057] Specifically, in one or more embodiments of this specification, the propagation characteristics of wind field disturbances from reliable measuring points to distorted measuring points are determined based on real-time wind speed data from each reliable measuring point, including: like Figure 4 The method involves querying the spatial topology based on the spatial location of each measuring point and the real-time wind direction to obtain a reference measuring point located upwind of the distorted measuring point. If the reference measuring point is a reliable measuring point, it is used as a candidate source measuring point corresponding to the distorted measuring point. Because wind disturbances propagate from upstream to downstream along the wind direction in a wind field, only reliable measuring points located upwind of the distorted measuring point have physical deduction value. Therefore, for any combination of a reference measuring point and a distorted measuring point, the direction vector from the reference measuring point to the distorted measuring point, stored in the spatial topology, is compared with the current real-time wind direction. When the direction vector is consistent with the real-time wind direction, it indicates that the wind first blows from the reference measuring point to the distorted measuring point, and the reference measuring point is located upwind of the distorted measuring point. If the reference measuring point is a reliable measuring point, it is listed as a candidate source measuring point corresponding to the distorted measuring point; otherwise, the reference measuring point is not suitable for the subsequent reconstruction process of the distorted measuring point.

[0058] In a preferred embodiment, an angle threshold can be further set. When the angle between the spatial direction of the reliable measuring point relative to the distorted measuring point and the real-time wind direction is less than the preset angle threshold, the reliable measuring point is determined to be located on the upwind side of the distorted measuring point. It should be noted that the specific value of the above-mentioned angle threshold can be set by those skilled in the art according to the terrain characteristics of the wind farm and actual needs. For example, it can be set to ±20° in open and flat terrain, and can be appropriately relaxed to ±45° in complex terrain. This specification does not specifically limit the specific value of the angle threshold.

[0059] It should also be noted that in actual wind farms, one distorted measuring point may correspond to multiple candidate source measuring points, and one reliable measuring point may simultaneously serve as a candidate source measuring point for multiple distorted measuring points. When multiple candidate source measuring points exist, the propagation characteristics of each candidate source measuring point can be calculated separately, and the results of each inference can be combined through a weighted fusion method to improve the reliability of the reconstructed wind speed data.

[0060] After obtaining the candidate source measurement points, for each pair of candidate source measurement points and distortion measurement points, the spatial distance between each candidate source measurement point and distortion measurement point is determined according to the spatial coordinates of each candidate source measurement point and distortion measurement point.

[0061] Wind field disturbances require a certain amount of time to propagate from upstream to downstream. Specifically, a candidate source measuring point first senses a wind field disturbance, which then propagates along the wind direction at a certain speed and reaches the location of the distorted measuring point after a period of time. This propagation time is the time delay. In the embodiments of this specification, the time delay required for wind field disturbances to propagate from each candidate source measuring point to the distorted measuring point is determined based on the spatial distance and the real-time wind speed of the candidate source measuring points. Specifically, the time delay required for wind field disturbances to propagate from the candidate source measuring points to the distorted measuring points is obtained by dividing the spatial distance by the real-time wind speed of the candidate source measuring points.

[0062] The propagation characteristics are determined based on the time delay and a preset attenuation coefficient. These propagation characteristics include at least the time delay and the wind speed attenuation when the disturbance propagates to the distorted measuring point. In the embodiments of this specification, the wind speed attenuation is determined based on the preset attenuation coefficient and the propagation distance. Specifically, the attenuation coefficient characterizes the degree of wind speed attenuation per unit propagation distance; a larger value indicates that the wind field disturbance attenuates faster during propagation. Multiplying the attenuation coefficient by the spatial distance yields the total wind speed attenuation during the propagation of the wind field disturbance from the candidate source measuring point to the distorted measuring point. The aforementioned wind speed attenuation coefficient can be pre-calibrated based on the topographic features of the wind farm area. In an optional embodiment, the wind speed attenuation coefficient can be further corrected based on atmospheric stability. The above calibration and correction methods can be selected and implemented by those skilled in the art based on the actual engineering environment, and this specification does not specifically limit them.

[0063] In the above process, the propagation characteristics of wind field disturbances from each candidate source measurement point to the distorted measurement point are determined. These propagation characteristics include at least two parameters: time delay and wind speed attenuation. The time delay is used to shift the real-time wind speed data of the candidate source measurement points along the time axis during subsequent wind field reconstruction, aligning the historical wind speed data of the candidate source measurement points with the current time of the distorted measurement point. The wind speed attenuation is used to correct the amplitude of the shifted wind speed data to restore the true wind speed amplitude when the disturbance propagates to the location of the distorted measurement point. When a distorted measurement point corresponds to multiple candidate source measurement points, the propagation characteristics of each candidate source measurement point are determined separately. Subsequently, the inference results of each candidate source measurement point can be combined using a weighted fusion method.

[0064] Specifically, in one or more feasible embodiments of this specification, wind field reconstruction is performed on the locations of distorted measurement points based on propagation characteristics to generate full-field wind speed data, specifically including: For each distorted measuring point, the real-time wind speed data of each candidate source measuring point corresponding to that distorted measuring point are first acquired. When a distorted measuring point corresponds to multiple candidate source measuring points, the real-time wind speed data of each candidate source measuring point are acquired separately. Based on the time delay in the propagation characteristics, the real-time wind speed data of each candidate source measuring point is shifted along the time axis to obtain the delayed wind speed data at the distorted measuring point. Specifically, for the real-time wind speed data of the candidate source measuring point at the current time T, it is shifted to the position of the distorted measuring point at the current time T for use. That is, the wind field disturbance sensed by the candidate source measuring point at time (T-ΔT) propagates to the position of the distorted measuring point at time T after a time delay of ΔT. After the above time axis shift, the delayed wind speed data at the distorted measuring point can be obtained.

[0065] During the propagation of wind field disturbances, the wind speed amplitude attenuates due to factors such as turbulent dissipation and ground friction. Therefore, the delayed wind speed data obtained after time alignment cannot be directly used as the final wind speed value for the distorted measurement point; it needs to be corrected based on the wind speed attenuation in the propagation characteristics. The correction process is as follows: subtract the wind speed attenuation in the propagation characteristics from the delayed wind speed data to obtain the corrected wind speed data. Since the attenuation represents the wind speed amplitude lost during the propagation of the wind field disturbance from the candidate source measurement point to the distorted measurement point, after the above amplitude correction, the true wind speed amplitude at the location of the disturbance propagating to the distorted measurement point can be recovered, thus obtaining the corrected wind speed data for that distorted measurement point.

[0066] Following the aforementioned time delay shift and amplitude correction process, the real-time wind speed data of each candidate source measuring point are independently extrapolated to obtain multiple reconstructed wind speed data. These multiple reconstructed wind speed data are then weighted and fused, with the fused result serving as the final reconstructed wind speed data for the distorted measuring point. The weight of each candidate source measuring point is determined inversely proportional to its spatial distance from the distorted measuring point. In practice, the reciprocal of the distance between each candidate source measuring point and the distorted measuring point is first calculated. Then, the reciprocal of the distance between each candidate source measuring point and the distorted measuring point is divided by the sum of the reciprocals of the distances of all candidate source measuring points to obtain a normalized weight value. The reconstructed wind speed data of each candidate source measuring point is multiplied by its corresponding weight value and then summed to obtain the final weighted fused reconstructed wind speed data.

[0067] The reconstructed wind speed data from each distorted measuring point is merged with the real-time wind speed data from each reliable measuring point to form a full-field wind speed data covering all measuring points within the wind farm. In this full-field wind speed data, the wind speed value at the reliable measuring point is the measured wind speed value after temperature compensation, while the wind speed value at the distorted measuring point is the extrapolated wind speed value obtained through the wind field reconstruction described above. This achieves dynamic wind speed data coverage across the entire wind farm without blind spots. Furthermore, by using inverse spatial distance weighted fusion, the cumulative error that may be introduced by long-distance extrapolation is effectively suppressed, improving the overall accuracy and robustness of the reconstructed wind speed data after fusion.

[0068] Furthermore, in one or more feasible embodiments of this specification, reliable and distorted measurement points among each measurement point have been determined through the first and second diagnostic results. However, the reliability of data may still differ among measurement points deemed reliable. Some measurement points may have spectral waveforms of high reliability, while others, although passing the diagnostic threshold, may have indicators at a critical state, belonging to medium-to-low reliability data. If all reliable measurement points are treated equally and used indiscriminately for subsequent reconstruction and deduction, low-reliability data may affect accuracy. Therefore, to solve this problem, the method further includes: Based on the first and second diagnostic results corresponding to each reliable measurement point, the data confidence level of each reliable measurement point is determined. Specifically, the determination of the confidence level is based on the following criteria: the degree of deviation of the reflection spectrum waveform characteristic parameters in the first diagnostic result from their respective preset thresholds. For example, the closer the waveform characteristic parameters are to the ideal state, the better the sensor hardware health is, and the higher the reliability of its measurement data; conversely, if the waveform characteristic parameters, although not exceeding the preset threshold, have significantly deviated from the initial calibration value, it indicates that the sensor, although not failed, has already experienced a certain degree of performance degradation, and its data confidence is correspondingly reduced. It also includes the degree of agreement between the aerodynamic characteristic correlation comparison in the second diagnostic result. For example, the more the correlation between the wind speed value and the wind turbine operating status data conforms to the preset aerodynamic characteristic curve, the more the wind speed data of that measurement point conforms to physical laws, and the higher the reliability; conversely, if the deviation between the two, although not exceeding the distortion judgment threshold, has significantly deviated from the ideal agreement state, the confidence is correspondingly reduced.

[0069] In this embodiment, the confidence level of each reliable measurement point is divided into several levels according to a preset classification standard. In one exemplary implementation, the confidence level can be divided into three levels: high confidence, medium confidence, and low confidence. Measurement points whose reflectance spectrum waveform characteristic parameters are consistent with the factory calibration and whose wind speed values ​​show a high degree of aerodynamic correlation with the wind turbine operating data are rated as high confidence level. Measurement points whose waveform characteristic parameters deviate but are still within the normal range, or whose aerodynamic correlation consistency shows a certain degree of deviation, are rated as medium confidence level. Measurement points that pass the diagnostic threshold but whose waveform characteristic parameters or aerodynamic correlation consistency are at a critical state are rated as low confidence level.

[0070] It should be understood that the above three levels and the specific numerical judgment criteria for each level can be set by those skilled in the art based on the specific model of the fiber optic anemometer and the actual operating conditions and engineering experience of the wind farm. This specification does not make any specific limitations on this.

[0071] The real-time wind speed data of reliable measuring points are weighted and corrected according to the confidence level. After the above confidence-weighted correction, the real-time wind speed data of each reliable measuring point is adjusted to a corrected value with confidence weight. Based on the corrected real-time wind speed data of each reliable measuring point, the propagation characteristics of wind field disturbances in the wind farm from reliable measuring points to distorted measuring points are determined. The process of determining these propagation characteristics is consistent with the above embodiment and will not be repeated here.

[0072] After reconstructing the wind field at the distorted measuring points based on the propagation characteristics, the inference confidence of the reconstructed wind speed data is determined according to the data confidence level of the candidate source measuring points corresponding to each distorted measuring point, the spatial distance between each candidate source measuring point and the distorted measuring point, and the wind direction deviation between the real-time wind direction of each candidate source measuring point and the real-time wind direction of the distorted measuring point. In a specific implementation, the inference confidence is calculated as follows: first, the individual scores of the above three dimensions are calculated separately; then, each individual score is multiplied by its corresponding weight coefficient; finally, the weighted scores are summed to obtain a comprehensive inference confidence score.

[0073] After determining the projection confidence level, it is compared with a preset projection confidence level threshold. This preset threshold can be pre-set by those skilled in the art based on the wind farm's safe operation requirements and the control system's fault tolerance capabilities. If the projection confidence level is lower than the preset projection confidence level threshold, the reconstructed wind speed data for the distorted measurement points is unavailable.

[0074] In a preferred embodiment, for distorted measurement points marked as unusable, the control system can trigger corresponding safety response measures, such as reducing the operating load of the wind turbine corresponding to the measurement point or executing a safety shutdown command, to prevent control decision errors due to unreliable data. Reconstructed wind speed data with a projection confidence level reaching or exceeding a preset threshold is marked as usable for subsequent wind turbine control strategy formulation.

[0075] By extending the reliable binary judgment to a multi-level confidence level, fine-grained differentiation and hierarchical management of data quality are achieved, improving the overall quality of the input data source for the inference. Through confidence-weighted correction, the interference of low-quality data on the inference results is suppressed at the source, making the determination of propagation characteristics more accurate and reliable, thereby improving the accuracy of subsequent wind farm reconstruction. Based on the first and second diagnostic results, the reliability of the reconstruction results obtained from reliable data inference is further evaluated, achieving a two-level progression from data quality assessment to reconstruction quality assessment. This differs from existing technologies that rely solely on a single threshold judgment, providing more comprehensive data quality information for wind turbine control. By comparing the inference confidence level with a preset threshold, the misuse of low-reliability reconstruction data for wind turbine control is prevented, eliminating safety risks caused by unreliable data.

[0076] like Figure 5 The diagram shown is a structural schematic of a dynamic wind speed monitoring device based on a fiber Bragg grating provided in an embodiment of this specification. The device includes: The acquisition unit 501 is used to acquire monitoring data from multiple fiber Bragg grating anemometers at various points within the wind farm; wherein the monitoring data includes at least: fiber Bragg grating reflection spectrum data, strain wavelength data, and reference temperature wavelength data. The diagnostic unit 502 is used to analyze the waveform characteristics of the reflection spectrum data to obtain a first diagnostic result, and to determine a second diagnostic result based on the physical correlation between the strain wavelength data and the wind turbine operating status data corresponding to each of the measuring points. The determining unit 503 is used to determine the reliable measuring points and the distorted measuring points among the measuring points in the wind farm based on the first diagnostic result and the second diagnostic result; The compensation unit 504 is used to perform temperature compensation on the strain wavelength data of the reliable measuring point using the reference temperature wavelength data, so as to obtain the real-time wind speed data of the reliable measuring point. The reconstruction unit 505 is used to determine the propagation characteristics of wind field disturbances in the wind farm from the trusted measuring points to the distorted measuring points based on the real-time wind speed data of each trusted measuring point, so as to reconstruct the wind field at the location of the distorted measuring point based on the propagation characteristics and generate wind speed data for the entire field.

[0077] like Figure 6 The diagram shown is a structural schematic of a dynamic wind speed monitoring device based on a fiber Bragg grating provided in an embodiment of this specification. The device includes: At least one processor; and, A memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform any of the methods described above.

[0078] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the embodiments of apparatus, devices, and non-volatile computer storage media are basically similar to the method embodiments, so the descriptions are relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0079] The foregoing has described specific embodiments of this specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims may be performed in a different order than that shown in the embodiments and may still achieve the desired result. Furthermore, the processes depicted in the drawings do not necessarily require the specific or sequential order shown to achieve the desired result. In some embodiments, multitasking and parallel processing are possible or may be advantageous.

[0080] The above description is merely one or more embodiments of this specification and is not intended to limit this specification. Various modifications and variations can be made to the one or more embodiments of this specification by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principle of one or more embodiments of this specification should be included within the scope of the claims of this specification.

Claims

1. A dynamic wind speed monitoring method based on fiber Bragg gratings, characterized in that, The method includes: The monitoring data of each measuring point is collected by fiber Bragg grating anemometers at multiple measuring points within the wind farm; wherein the monitoring data includes at least: fiber Bragg grating reflection spectrum data, strain wavelength data, and reference temperature wavelength data; The waveform characteristics of the reflection spectrum data are analyzed to obtain a first diagnostic result, and a second diagnostic result is determined based on the physical correlation between the strain wavelength data and the wind turbine operating status data corresponding to each measuring point. Based on the first diagnostic result and the second diagnostic result, reliable measuring points and distorted measuring points are determined among the measuring points in the wind farm. Temperature compensation is performed on the strain wavelength data of the reliable measuring point using the reference temperature wavelength data to obtain the real-time wind speed data of the reliable measuring point. Based on the real-time wind speed data of each trusted measuring point, the propagation characteristics of wind field disturbances in the wind farm from the trusted measuring points to the distorted measuring points are determined, and the wind field is reconstructed at the location of the distorted measuring points based on the propagation characteristics to generate full-field wind speed data.

2. The dynamic wind speed monitoring method based on fiber Bragg grating according to claim 1, characterized in that, The fiber Bragg grating anemometer includes at least a first fiber Bragg grating and a second fiber Bragg grating; Wherein, the first fiber grating is a strain sensing grating used to output the strain wavelength data; the second fiber grating is a temperature compensation grating, disposed adjacent to the first fiber grating and isolated from external strain, used to output the reference temperature wavelength data.

3. The dynamic wind speed monitoring method based on fiber Bragg grating according to claim 2, characterized in that, The plurality of measuring points include: at least two measuring points distributed along the height direction of each wind turbine tower in the wind farm, and at least one measuring point distributed along the line connecting adjacent wind turbines in the wind farm. Specifically, monitoring data is collected from multiple fiber optic anemometers at various points within the wind farm, including: The multiple fiber Bragg grating anemometers are connected in series to the same transmission fiber to construct a fiber Bragg grating sensing network; each fiber Bragg grating anemometer in the fiber Bragg grating sensing network is configured with a non-overlapping initial reflection wavelength. By transmitting broadband optical signals into the transmission optical fiber, and based on the initial reflection wavelength of each fiber optic anemometer, wavelength demodulation is performed on the returned reflection spectrum data to determine the center wavelength of the strain sensing grating and temperature compensation grating at each measuring point. The center wavelengths of the strain sensing grating and the temperature compensation grating are used as the strain wavelength data and the reference temperature wavelength data, respectively.

4. The dynamic wind speed monitoring method based on fiber Bragg grating according to claim 3, characterized in that, The multiple measuring points include: benchmark measuring points and target measuring points obtained based on the prevailing wind direction of the current cycle of the wind farm; The reference measuring point is the measuring point located on the upwind side of the prevailing wind direction among the plurality of measuring points, used to provide wind field boundary condition data; the target measuring point is the measuring point located on the downwind side of the prevailing wind direction among the plurality of measuring points. The method further includes: Based on the historical wind field data within the preset period of the wind farm, the dominant wind direction of the wind farm is determined; Based on the spatial distance between the benchmark measuring point and the target measuring point and the prevailing wind direction, the direction vector and distance parameters between each measuring point are determined to establish the spatial topological relationship between each measuring point.

5. The dynamic wind speed monitoring method based on fiber Bragg grating according to claim 1, characterized in that, Analyzing the waveform characteristics of the reflectance spectral data to obtain a first diagnostic result, and based on the physical correlation between the strain wavelength data and the wind turbine operating status data corresponding to each measuring point, determining a second diagnostic result, specifically including: The waveform features of the reflectance spectral data are analyzed to extract the waveform feature parameters corresponding to the reflectance spectral data; wherein, the waveform feature parameters include: full width at half maximum (FWHM) and sidelobe suppression ratio; By comparing the waveform feature parameters with the corresponding preset threshold, it is determined that the measurement points where the waveform feature parameters exceed the preset threshold have sensor abnormalities, and a first diagnostic result is obtained. The operating status data of the wind turbine corresponding to the measuring point is obtained, and the strain wavelength data is converted into wind speed value; By comparing the wind speed value with the operating status data through aerodynamic characteristic correlation, it is determined that the measurement point data is distorted due to the deviation of the correlation between the wind speed value and the operating status data from the preset aerodynamic characteristic mapping relationship, and a second diagnostic result is obtained.

6. The dynamic wind speed monitoring method based on fiber Bragg grating according to claim 4, characterized in that, Based on the real-time wind speed data from each of the trusted measuring points, the propagation characteristics of wind field disturbances within the wind farm from the trusted measuring points to the distorted measuring points are determined, specifically including: Based on the spatial location and real-time wind direction of each measuring point, the spatial topology is queried to obtain the reference measuring point located on the windward side of the distorted measuring point. When the reference measuring point is a reliable measuring point, it is used as the candidate source measuring point corresponding to the distorted measuring point. Based on the spatial coordinates of each candidate source measurement point and the distortion measurement point, the spatial distance between each candidate source measurement point and the distortion measurement point is determined. Based on the spatial distance and the real-time wind speed at the candidate source measurement points, determine the time delay required for wind field disturbances to propagate from each of the candidate source measurement points to the distortion measurement point; The propagation characteristics are determined based on the time delay and the preset attenuation coefficient; the propagation characteristics include at least the time delay and the wind speed attenuation when the disturbance propagates to the distortion measurement point.

7. The dynamic wind speed monitoring method based on fiber Bragg grating according to claim 6, characterized in that, The step of reconstructing the wind field at the location of the distorted measurement point based on the propagation characteristics to generate full-field wind speed data specifically includes: Obtain real-time wind speed data of the candidate source measurement points corresponding to the distorted measurement points; Based on the time delay in the propagation characteristics, the real-time wind speed data of each candidate source measuring point is shifted along the time axis to obtain the delayed wind speed data at the distorted measuring point. Based on the wind speed attenuation in the propagation characteristics, the delayed wind speed data is amplitude corrected to obtain the corrected wind speed data of the distorted measurement point. Reconstructed wind speed data corresponding to each candidate source measurement point is obtained, and the reconstructed wind speed data of multiple reconstructed wind speed data are weighted and fused, wherein the weight of the weighted fusion is determined inversely proportional to the spatial distance between each candidate source measurement point and the distorted measurement point; The wind speed data of the wind farm is constructed by combining the reconstructed wind speed data from each distorted measuring point with the real-time wind speed data from each reliable measuring point.

8. The dynamic wind speed monitoring method based on fiber Bragg grating according to claim 6, characterized in that, The method further includes: Based on the first and second diagnostic results corresponding to each of the aforementioned reliable measurement points, the data confidence level of each of the aforementioned reliable measurement points is determined; The real-time wind speed data of the credible measuring points are weighted and corrected according to the confidence level, so as to determine the propagation characteristics of wind field disturbances in the wind farm from the credible measuring points to the distorted measuring points based on the corrected real-time wind speed data of each credible measuring point. After reconstructing the wind field of the distorted measuring points based on the propagation characteristics, the inference confidence of the reconstructed wind speed data is determined according to the data confidence level of the candidate source measuring points corresponding to each distorted measuring point, the spatial distance between each candidate source measuring point and the distorted measuring point, and the wind direction deviation between the real-time wind direction of each candidate source measuring point and the real-time wind direction of the distorted measuring point. If the projection confidence level is lower than the preset projection confidence level threshold, the reconstructed wind speed data of the distorted measurement point is marked as unusable.

9. A dynamic wind speed monitoring device based on fiber Bragg gratings, characterized in that, The device includes: The data acquisition unit is used to acquire monitoring data from multiple fiber Bragg grating anemometers within the wind farm; wherein the monitoring data includes at least: fiber Bragg grating reflection spectrum data, strain wavelength data, and reference temperature wavelength data. The diagnostic unit is used to analyze the waveform characteristics of the reflection spectrum data to obtain a first diagnostic result, and to determine a second diagnostic result based on the physical correlation between the strain wavelength data and the wind turbine operating status data corresponding to each of the measuring points. The determining unit is used to determine the reliable measuring points and distorted measuring points among the measuring points in the wind farm based on the first diagnostic result and the second diagnostic result; The compensation unit is used to perform temperature compensation on the strain wavelength data of the reliable measuring point using the reference temperature wavelength data, so as to obtain the real-time wind speed data of the reliable measuring point. The reconstruction unit is used to determine the propagation characteristics of wind field disturbances in the wind farm from the trusted measuring points to the distorted measuring points based on the real-time wind speed data of each trusted measuring point, so as to reconstruct the wind field at the location of the distorted measuring point based on the propagation characteristics and generate wind speed data for the entire field.

10. A dynamic wind speed monitoring device based on fiber Bragg gratings, characterized in that, The device includes: at least one processor; and, A memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform the method described in any one of claims 1-8.

Images