Wind farm global wind field measurement method and system based on array wind measurement radar

Abstract

The application discloses a kind of based on array wind measurement radar's wind farm global wind field measurement method and system.The method includes the following steps: S1, wind farm early reconnaissance and array wind measurement radar deployment scheme design, S2, array wind measurement radar field deployment and system double calibration, S3, array radar global synchronous data acquisition and standardization preprocessing, S4, core parameter measurement, S5, multi-source verification and model correction, S6, achievement output;The system includes: array node module, calibration wind measurement tower module, networking communication and edge / cloud data processing module and measurement and correction module, is configured to execute the double calibration, standardization preprocessing, CFD wind field modeling, wake correction, long-term revision and multi-source verification feedback correction.The application relates to the technical field of wind farm wind resource assessment.

Description

Technical Field

[0001] This invention belongs to the field of wind farm wind resource assessment technology, specifically involving a wind farm full-area wind field calculation method and system based on array-type wind measurement radar. It is applicable to the fine calculation of full-area wind resources of wind farms in various terrains (plains, mountains, and offshore), and can serve the early feasibility study of wind farms, wind turbine micro-site selection, operation monitoring and power generation optimization. Background Technology

[0002] As the wind power industry rapidly expands towards large-scale, complex terrain, and offshore applications, accurate wind resource measurement has become a core element affecting turbine selection, micro-site selection, power generation prediction, and operational safety, directly determining the return on investment and long-term operational benefits of wind farms. Currently, traditional methods for wind farm measurement mainly rely on meteorological towers and single remote sensing wind radars. However, these methods have significant technical limitations and cannot meet the needs for refined wind measurement across the entire area.

[0003] Traditional wind measurement towers use contact observation, which requires large-scale civil construction, resulting in long construction periods, high costs, and significant terrain limitations, making them unsuitable for deployment in steep mountains, offshore areas, and other similar locations. Furthermore, a single wind measurement tower can only acquire wind parameter data at a single point, making it difficult to reflect the spatial heterogeneity of the wind field across the entire region. In particular, in wind fields with complex terrain, topographic effects such as wind speed uphill and valley winds can lead to significant differences in the spatial distribution of wind parameters. Extrapolating single-point data to the entire region can result in large calculation errors, affecting the accuracy of subsequent engineering decisions.

[0004] Although a single remote sensing wind radar (such as lidar or a single phased array radar) achieves non-contact wind measurement and is flexible in deployment, it has problems such as limited coverage, wind measurement accuracy being greatly affected by electromagnetic interference and meteorological conditions, and inability to achieve full-area monitoring without blind spots. For large wind fields or wind fields with complex terrain, a single radar is difficult to cover all the selected locations, and there is a lack of cross-validation of multi-node data, making it difficult to guarantee the reliability of the measurement results.

[0005] Furthermore, existing wind measurement methods mostly employ a single data source for wind farm calculations, failing to achieve deep fusion of multi-source data. This makes it difficult to effectively eliminate the impact of systematic errors and environmental interference, resulting in insufficient measurement accuracy and stability, and failing to meet the high standards required by the wind power industry for comprehensive and detailed wind farm measurements. Therefore, developing a wind farm measurement method that can achieve full coverage of wind farms, high accuracy, adaptability to complex terrain, and low cost has become a pressing technical challenge for the wind power industry. Summary of the Invention

[0006] This invention provides a method and system for calculating the wind field across the entire wind farm area based on an array-type wind measuring radar, in order to solve the problem of the difficulty in simultaneously achieving "coverage, accuracy, reliability, and acceptability".

[0007] The technical solution adopted in this invention is a method for calculating the entire wind field of a wind farm based on an array-type wind measuring radar. The method includes the following steps: S1. Preliminary survey of wind farm and design of array-type wind measuring radar deployment scheme: survey to determine the scope of wind farm, topography, electromagnetic environment and hub height of the proposed wind turbine location; select phased array wind measuring radar and design array configuration according to topography and wind farm scale, and plan the location and wind measuring parameters of calibration tower; S2. On-site deployment and dual system calibration of array-type wind measuring radar: The array radar nodes are deployed in an unobstructed area and networked for communication; the system errors of radar wind speed, wind direction and turbulence intensity are corrected by synchronous observation, calibration and correction of the array radar and calibration wind measuring tower; and the system deviation between nodes is corrected by cross-calibration of nodes, so that the data in the whole area is consistent. S3. Array radar full-domain synchronous data acquisition and standardized preprocessing: Collect wind parameters across the entire domain, and perform noise reduction, missing value completion, standardization processing and quality inspection to output valid wind measurement data that meets the wind farm industry standards. S4. Using the preprocessed array data as boundary conditions and validation data, combined with wind field topography DEM to drive CFD numerical simulation, calculate the basic wind parameters, spatial distribution of wind resources, refined wind parameters of machine positions and wake effect in the whole area, and complete the long-term wind resource correction by combining long-term meteorological station data. S5. Multi-source verification and model correction: Error judgment is made by cross-verification of array nodes and comparison of calculation results with calibrated wind tower / actual data. When the error exceeds the preset threshold, the CFD model parameters and wind parameter calculation coefficients are adjusted and iterated until the error meets the requirements. S6. Output standardized calculation results and apply them to wind turbine micro-location, unit selection, power generation prediction and wake management during operation.

[0008] Furthermore, in step S1, the array configuration is determined based on the terrain and wind field scale: for wind fields in narrow valleys or along the coastline, a linear array is used, with array nodes arranged along the prevailing wind direction and a node spacing of 500m to 1000m; for wind fields in open plains or at sea, a surface array is used, with array nodes distributed in a grid pattern and a node spacing of 1000m to 2000m; for small wind fields ≤500MW, 3 to 5 array nodes are set, and for large wind fields ≥1000MW, 6 to 10 array nodes are set.

[0009] Furthermore, in step S1, the wind sampling parameters are set as follows: wind speed and wind direction sampling frequency 1Hz, turbulence intensity and wind shear sampling frequency 10Hz, and statistics are output once every 10 minutes; the observation height layer covers the wind turbine hub height and the upper and lower limits of the impeller sweep height, with a height interval of 10m; the observation period is not less than 3 months; and 1 to 2 calibration wind measurement towers are planned to be deployed in the central area of ​​the array radar.

[0010] Furthermore, in the dual calibration of step S2, The synchronous observation calibration specifically involves: the array radar and the calibration wind tower conducting continuous synchronous observations for at least 7 days to obtain wind parameters at the same height and time period; and using linear fitting to correct the errors in radar wind speed, wind direction, and turbulence intensity, so that the relative error between radar data and wind tower data is ≤5%. The node cross-calibration specifically involves comparing the wind parameters of multiple radar nodes in the overlapping observation area and correcting the system deviation between nodes to ensure data consistency across the entire domain.

[0011] Furthermore, the standardization preprocessing in step S3 includes the following steps: S31. Denoising: Wavelet analysis and Kalman filtering are used to remove abnormal data caused by electromagnetic interference, terrain obstruction or meteorological interference. S32. Missing value completion: For short-term missing values ​​(<1h), linear interpolation is used for completion; for long-term missing values ​​(>1h), data from adjacent radar nodes is used for completion. S33. Standardization: Convert the data of each node to a unified geographic coordinate system for the wind field, unify the statistical standards for the height layer, and output wind measurement files that meet the standard format requirements; S34. Quality inspection: Remove invalid data and ensure that the effective data rate is ≥90%; if the standard is not met, extend the observation period or adjust the station layout and re-collect data.

[0012] Furthermore, step S4 specifically includes: S41. Basic wind parameter calculation: Perform statistical analysis on the preprocessed data, and output the average wind speed, average wind direction, wind energy rose diagram, wind power density, turbulence intensity, wind shear index and surface roughness length of the wind field, and analyze the spatiotemporal variation law of wind parameters. S42. Wind resource spatial distribution modeling: Combining array measured data and wind field topographic DEM data, optimize the turbulence model and wake model of the CFD model to generate a spatial distribution map of wind field speed and direction with a resolution of 10m×10m. S43. Refined wind parameter extraction for each aircraft location: Extract wind speed, wind direction, turbulence intensity, inflow angle and wind shear parameters at the hub height of each proposed aircraft location from the spatial distribution map. S44. Wake effect calculation: By monitoring the wake range and attenuation coefficient of the wind turbine through array radar, the Jensen model or Frandsen wake model is corrected and the wake loss of the wind field is calculated. S45. Long-term wind resource correction: Using linear regression and MK test methods, correlation analysis was performed on the measured data of the array radar and the data of ≥10 years provided by the surrounding long-term meteorological stations to correct the long-term annual average wind speed and annual effective wind hours of the wind field.

[0013] Furthermore, in step S5, the preset error thresholds are: relative wind speed error ≤ 5%, and relative power generation prediction error ≤ 10%. When these conditions are not met, the model correction includes at least adjusting one or more of the following parameters: turbulence coefficient of the CFD model, wake model attenuation coefficient, wind shear index, and surface roughness length parameter, and recalculating until the target is met.

[0014] Furthermore, in step S2, the array radar is a phased array wind measuring radar with a wind measuring range of 1 to 10 km, a wind speed measurement accuracy of ±0.3 m / s, a wind direction measurement accuracy of ±3°, and a coverage height of 50 to 300 m; for offshore wind fields, a marine version of the phased array wind measuring radar that is resistant to salt spray and typhoons is selected.

[0015] Furthermore, in step S3, the data acquisition adopts a layered architecture of edge computing local real-time preprocessing + cloud big data analysis and modeling. The local server performs real-time noise reduction and quality control, while the cloud platform performs CFD modeling and multi-source verification management, and ensures that the wind measurement data is backed up twice and not lost.

[0016] Furthermore, this method is adapted to complex terrain: for beam obstruction of wind fields in mountainous areas, a low-altitude scanning mode is adopted with the pitch angle controlled between 3° and 5° and CFD interpolation correction is combined; for wind field attenuation due to rain and fog at sea, a meteorological correction algorithm is used to correct the attenuation of wind measurement data.

[0017] A wind farm global wind field measurement system based on array-type wind measuring radar, the system being used to execute the method described above, the system comprising: The array node module consists of multiple phased array wind measurement radars and is deployed according to the array configuration; Calibrate the wind measurement tower module; Network communication and edge / cloud data processing modules; The measurement and correction module is configured to perform the dual calibration, standardized preprocessing, CFD wind field modeling, wake correction, long-term correction, and multi-source verification feedback correction. The beneficial effects of this invention are as follows: 1. Achieve precise wind measurement across the entire wind farm area, overcoming the limitations of traditional wind measurement methods: This invention employs an array-type wind measurement radar network, combined with multi-node synchronous observation, to achieve full coverage of the wind farm area without blind spots. It can accurately capture the spatial heterogeneity of the wind farm, especially the terrain effects such as wind speed ramp and valley wind in complex terrain wind farms. It avoids the errors of extrapolating single-point data from traditional wind measurement towers, significantly improving the measurement accuracy. The relative error of wind speed is ≤5%, and the relative error of power generation prediction is ≤10%.

[0018] 2. Adaptable to various terrains and wind fields, flexible deployment, and low cost: The array-type wind measuring radar is a lightweight, non-contact device that does not require large-scale civil construction and has a short installation cycle (installation of a single radar unit < 1 day, array networking < 7 days). It can adapt to various terrains and wind fields such as plains, mountains, and sea. Compared with traditional wind measuring towers, the construction cost is reduced by 40%-50%, and the operation and maintenance workload is reduced by more than 90%. Especially in offshore wind fields, it can avoid the high construction and maintenance costs of offshore wind measuring towers, resulting in significant economic advantages.

[0019] 3. Reliable calculation results from multi-source data fusion: This invention adopts a multi-source data fusion mode of "array radar + wind measurement tower + CFD numerical simulation + long-term meteorological station". Through dual calibration, multi-source verification and model correction, it effectively eliminates system errors and environmental interference, and significantly improves the reliability and stability of the calculation results. It fully complies with national wind power industry standards, has the legal effect for engineering applications, and can be directly used for wind farm project application, approval and operation and maintenance management.

[0020] 4. Serving the entire life cycle of wind farms, with high application value: The calculation results of this invention can not only meet the needs of early feasibility studies and turbine site selection for wind farms, but also be applied to wake management, power prediction and fault early warning during operation. It can increase wind farm power generation by 3%-5%, reduce operational risks, provide accurate data support for the refined management of wind farms throughout their entire life cycle, and has broad prospects for promotion and application. Attached Figure Description

[0021] Figure 1 This is a simplified flowchart of the method of the present invention; Figure 2 This is a simplified structural block diagram of the system of the present invention; Figure 3 This is a schematic diagram of the array configuration of an array-type wind-measuring radar (linear array along the prevailing wind direction / gridded area array) and a schematic diagram of the wake monitoring range; Figure 4 This is a spatial distribution map of wind speed / direction across the entire wind field (10m×10m) and a schematic diagram of turbine location parameter extraction (impeller sweep height layer interval, hub height extraction point). Detailed Implementation

[0022] The following example of an offshore wind farm application illustrates how this invention can be implemented, but it is not limited to this example.

[0023] Those skilled in the art can reproduce this by combining well-known CFD / wind resource software platforms (RANS type or corresponding solvers) and conventional radar station deployment specifications.

[0024] Example 1: Offshore wind field S1. Design of Preliminary Survey and Array-type Wind Measurement Radar Deployment Scheme for Wind Farm The wind field spans approximately 30km × 12km, with a proposed turbine hub height of 140m. Ten salt-spray resistant phased array wind radars will be selected as array nodes, arranged in a planar array configuration with a node spacing of approximately 1800m, and deployed on an offshore platform. The observation altitude will be 100-180m, with a spacing of 10m. Wind speed / direction will be measured at 1Hz, turbulence / wind shear at 10Hz, and statistics will be output every 10 minutes. The observation period will be 4 months. One to two calibration wind towers are planned to be deployed at representative locations in the central area of ​​the array.

[0025] S2, Array-type wind measurement radar on-site deployment and system dual calibration After the radar is installed, the elevation angle is adjusted to about 5°, the horizontal reference and height are calibrated and networked; the radar and the wind tower are started to observe synchronously for at least 7 days, and the wind speed / wind direction / turbulence sequence at the same height and time period is linearly fitted to correct the radar system error so that the relative error between it and the wind tower is controlled to ≤5%; then the data of the overlapping area between nodes is used for cross-calibration to eliminate the deviation between nodes and ensure that the data of the whole area can be stitched together.

[0026] S3, Array Radar Full-Domain Synchronous Data Acquisition and Standardized Preprocessing The array continuously collected wind parameters across the entire region for four months; wavelet analysis was used to locate abnormal structures caused by obstruction / rain and fog, and Kalman filtering was used for smoothing estimation to remove anomalies; short-term missing data (<1h) was linearly interpolated, and long-term missing data (>1h) was spatially filled by data from adjacent nodes; the data was unified to the wind field geographic coordinate system and the statistical caliber of the 10m height layer; the effective data rate reached 92.1% after quality inspection (meeting the availability criterion of ≥90%), forming a standardized wind measurement dataset.

[0027] S4, Core Parameter Calculation The statistics show that the average wind speed across the entire wind field is approximately 9.2 m / s, the wind power density is approximately 680 W / m², and the turbulence intensity is approximately 0.10. The array data and the seabed / island DEM are substituted into CFD to optimize the turbulence model and wake-related parameters, generating a 10m×10m wind speed / direction spatial distribution map. The hub height wind parameters of the proposed aircraft location are extracted from the distribution map. The wake loss is corrected by combining the Frandsen-type wake processing with the array's observation of the wake area (in this example, the wake loss is estimated to be about 7.5%). Based on the data from the surrounding long-term marine meteorological stations (12 years), long-term correction is performed using MK / linear regression, resulting in a long-term annual average wind speed of approximately 9.3 m / s and an annual effective wind duration of approximately 3240 h.

[0028] S5, Multi-source Validation and Model Correction Internal verification: The relative error in the overlapping area of ​​10 nodes is approximately 2.7% (<3% of the consistency threshold), requiring no recalibration; External verification: The relative error between the wind speed calculated by the wind station and the actual measurement by the wind tower is about 4.2%, and the relative error of the power generation prediction is about 9.1%. Both are lower than the preset thresholds (wind speed ≤ 5%, power generation ≤ 10%), so they are deemed qualified and will not enter the recalculation cycle.

[0029] S6. Output of Results Outputs a global distribution map, turbine location parameter table, wind rose diagram, wind power density / turbulence distribution and long-term correction report; after turbine location optimization, it can be used for wake management and power prediction support during operation; wind farm power generation is increased by 3.8%, and operational safety is significantly improved.

[0030] Comparative example (used to illustrate the source of error when "array closed loop is missing") Comparative Example 1: Extrapolation Scheme for Single-Measuring Wind Tower Only one anemometer tower is erected at the center of the wind field, and the wind parameters at other locations are extrapolated from the tower data according to power law / empirical wind shear.

[0031] The results are acceptable under uniform roughness in plains; however, in coastal / topographical undulation scenarios, extrapolation cannot reproduce spatial heterogeneity such as "channel acceleration, leeward deceleration, and local turbulent uplift", which causes the wind parameters at the turbine location to deviate from the actual measurements, and thus the power generation prediction is too optimistic.

[0032] Comparative conclusion: Compared with the "array multi-point constraint → CFD distribution → machine position extraction" of this invention, Comparative Example 1 lacks spatial sampling and dual calibration anchoring, resulting in a significant decrease in the reliability and acceptability of the calculation.

[0033] Comparative Example 2: Single Remote Sensing Radar Scheme (without array cross-calibration, without wind tower synchronous calibration closed loop) A single radar covers the central area, while the edge positions are still pushed outwards; and the radar system errors (calibration drift, near-shore atmospheric attenuation, loss of low SNR layer) lack the correction of the same source standard benchmark, and there is no consistency verification for the node overlap area. After preprocessing, the unknown bias is more likely to "infiltrate" the CFD boundary conditions.

[0034] Comparative conclusion: Without the closed loop of "dual calibration + standardized preprocessing + multi-source verification feedback correction" described in this invention, the wind field scale calculation results often exhibit the problem of "local reasonableness but unreliability across the entire domain", making it difficult to form auditable standardized deliverables.

[0035] Finally, it should be emphasized that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention can have various changes and modifications. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for calculating the entire wind field of a wind farm based on an array-type wind-measuring radar, characterized in that, The method includes the following steps: S1. Preliminary survey of wind farm and design of array-type wind measuring radar deployment scheme: survey to determine the scope of wind farm, topography, electromagnetic environment and hub height of the proposed wind turbine location; select phased array wind measuring radar and design array configuration according to topography and wind farm scale, and plan the location and wind measuring parameters of calibration tower; S2. On-site deployment and dual system calibration of array-type wind measuring radar: The array radar nodes are deployed in an unobstructed area and networked for communication; the system errors of radar wind speed, wind direction and turbulence intensity are corrected by synchronous observation, calibration and correction of the array radar and calibration wind measuring tower; and the system deviation between nodes is corrected by cross-calibration of nodes, so that the data in the whole area is consistent. S3. Array radar full-domain synchronous data acquisition and standardized preprocessing: Collect wind parameters across the entire domain, and perform noise reduction, missing value completion, standardization processing and quality inspection to output valid wind measurement data that meets the wind farm industry standards. S4. Using the preprocessed array data as boundary conditions and validation data, combined with wind field topography DEM to drive CFD numerical simulation, calculate the basic wind parameters, spatial distribution of wind resources, refined wind parameters of machine positions and wake effect in the whole area, and complete the long-term wind resource correction by combining long-term meteorological station data. S5. Multi-source verification and model correction: Error judgment is made by cross-verification of array nodes and comparison of calculation results with calibrated wind tower / actual data. When the error exceeds the preset threshold, the CFD model parameters and wind parameter calculation coefficients are adjusted and iterated until the error meets the requirements. S6. Output standardized calculation results and apply them to wind turbine micro-location, unit selection, power generation prediction and wake management during operation.

2. The method according to claim 1, characterized in that, In step S1, the array configuration is determined based on the terrain and wind field scale: for wind fields in narrow valleys or along the coastline, a linear array is used, with array nodes arranged along the prevailing wind direction and a node spacing of 500m to 1000m; for wind fields in open plains or at sea, a surface array is used, with array nodes distributed in a grid pattern and a node spacing of 1000m to 2000m; for small wind fields ≤500MW, 3 to 5 array nodes are set, and for large wind fields ≥1000MW, 6 to 10 array nodes are set.

3. The method according to claim 1, characterized in that, In step S1, the wind sampling parameters are set as follows: wind speed and wind direction sampling frequency is 1Hz, turbulence intensity and wind shear sampling frequency is 10Hz, and statistics are output once every 10 minutes; the observation height layer covers the wind turbine hub height and the upper and lower limits of the impeller sweep height, with a height interval of 10m; the observation period is not less than 3 months; and 1 to 2 calibration wind measurement towers are planned to be deployed in the central area of ​​the array radar.

4. The method according to claim 1, characterized in that, In the dual calibration of step S2, The synchronous observation calibration specifically involves: the array radar and the calibration wind tower conducting continuous synchronous observations for at least 7 days to obtain wind parameters at the same height and time period; and using linear fitting to correct the errors in radar wind speed, wind direction, and turbulence intensity, so that the relative error between radar data and wind tower data is ≤5%. The node cross-calibration specifically involves comparing the wind parameters of multiple radar nodes in the overlapping observation area and correcting the system deviation between nodes to ensure data consistency across the entire domain.

5. The method according to claim 1, characterized in that, The standardized preprocessing in step S3 includes the following steps: S31. Denoising: Wavelet analysis and Kalman filtering are used to remove abnormal data caused by electromagnetic interference, terrain obstruction or meteorological interference. S32. Missing value completion: For short-term missing values ​​(<1h), linear interpolation is used for completion; for long-term missing values ​​(>1h), data from adjacent radar nodes is used for completion. S33. Standardization: Convert the data of each node to a unified geographic coordinate system for the wind field, unify the statistical standards for the height layer, and output wind measurement files that meet the standard format requirements; S34. Quality inspection: Remove invalid data and ensure that the effective data rate is ≥90%; if the standard is not met, extend the observation period or adjust the station layout and re-collect data.

6. The method according to claim 1, characterized in that, The S4 step specifically includes: S41. Basic wind parameter calculation: Perform statistical analysis on the preprocessed data, and output the average wind speed, average wind direction, wind energy rose diagram, wind power density, turbulence intensity, wind shear index and surface roughness length of the wind field, and analyze the spatiotemporal variation law of wind parameters. S42. Wind resource spatial distribution modeling: Combining array measured data and wind field topographic DEM data, optimize the turbulence model and wake model of the CFD model to generate a spatial distribution map of wind field speed and direction with a resolution of 10m×10m. S43. Refined wind parameter extraction for each aircraft location: Extract wind speed, wind direction, turbulence intensity, inflow angle and wind shear parameters at the hub height of each proposed aircraft location from the spatial distribution map. S44. Wake effect calculation: By monitoring the wake range and attenuation coefficient of the wind turbine through array radar, the Jensen model or Frandsen wake model is corrected and the wake loss of the wind field is calculated. S45. Long-term wind resource correction: Using linear regression and MK test methods, correlation analysis was performed on the measured data of the array radar and the data of ≥10 years provided by the surrounding long-term meteorological stations to correct the long-term annual average wind speed and annual effective wind hours of the wind field.

7. The method according to claim 1, characterized in that, In step S5, the preset error thresholds are: relative wind speed error ≤ 5%, and relative power generation prediction error ≤ 10%. If these conditions are not met, the model correction includes at least adjusting one or more of the following parameters: turbulence coefficient of the CFD model, wake model attenuation coefficient, wind shear index, and surface roughness length parameter, and recalculating until the target is met.

8. The method according to claim 1, characterized in that, In step S2, the array radar is a phased array wind measuring radar with a wind measuring range of 1 to 10 km, a wind speed measurement accuracy of ±0.3 m / s, a wind direction measurement accuracy of ±3°, and a coverage height of 50 to 300 m; for offshore wind fields, a marine version of the phased array wind measuring radar that is resistant to salt spray and typhoons is selected.

9. The method according to claim 1, characterized in that, In step S3, data acquisition adopts a layered architecture of edge computing local real-time preprocessing + cloud big data analysis and modeling. The local server performs real-time noise reduction and quality control, while the cloud platform performs CFD modeling and multi-source verification management, and ensures that the wind measurement data is backed up twice and not lost.

10. A wind farm global wind field measurement system based on array-type wind measuring radar, characterized in that, The system is used to perform the method as described in any one of claims 1 to 9, the system comprising: The array node module consists of multiple phased array wind measurement radars and is deployed according to the array configuration; Calibrate the wind measurement tower module; Network communication and edge / cloud data processing modules; The measurement and correction module is configured to perform the dual calibration, standardized preprocessing, CFD wind field modeling, wake correction, long-term correction, and multi-source verification feedback correction.

Images