Wind Field Information Research Articles

Significant wave height prediction in monsoon regions based on the VMD-CNN-BiLSTM model

A novel significant wave height prediction method for monsoon regions is proposed, utilizing the VMD-CNN-BiLSTM model to enhance prediction accuracy under complex meteorological conditions. Traditional numerical models exhibit limitations in managing extreme marine conditions and fail to fully integrate wind field information. Meanwhile, existing machine learning models demonstrate insufficient generalization and robustness for long-term predictions. To address these shortcomings, the predictive approach combines Variational Mode Decomposition (VMD) with a hybrid deep learning model (CNN-BiLSTM). VMD is employed to decompose the original wave height sequence and extract key features, while CNN captures the spatial features of wind field and wave height data. BiLSTM, in turn, models the temporal dependencies. Experimental results reveal that the VMD-CNN-BiLSTM model provides substantial advantages in prediction performance across all seasons, including the entire year. Compared to traditional models, the proposed method demonstrates significantly reduced Mean Absolute Error (MAE) and Root Mean Square Error (RMSE), alongside an improved coefficient of determination (R²). These findings confirm the effectiveness and reliability of the method under complex meteorological conditions such as monsoons and typhoons.

Wind Field Reconstruction Method Using Incomplete Wind Data Based on Vision Mamba Decoder Network

Accurate meteorological information is crucial for the safety of civil aviation flights. Complete wind field information is particularly helpful for planning flight routes. To address the challenge of accurately reconstructing wind fields, this paper introduces a deep learning neural network method based on the Vision Mamba Decoder. The goal of the method is to reconstruct the original complete wind field from incomplete wind data distributed along air routes. This paper proposes improvements to the Vision Mamba model to fit our mission, showing that the developed model can accurately reconstruct the complete wind field. The experimental results demonstrate a mean absolute error (MAE) of wind speed of approximately 1.83 m/s, a mean relative error (MRE) of around 7.87%, an R-square value of about 0.92, and an MAE of wind direction of 5.78 degrees.

Wind-induced response of saddle membrane structure under typhoon wind field by weather research and forecasting model and computational fluid dynamics

Currently, the research on wind-induced response of membrane structures focuses on the normal wind field, and there is little research on typhoon with greater disaster. In this paper, the wind-induced response of saddle membrane structures under typhoon is studied by numerical simulation. Firstly, the wind field information of typhoon is simulated according to the Weather Research and Forecasting model, and the information is used as the inlet boundary condition of Computational Fluid Dynamics. The vibration modal analysis is carried out, considering the influence of wind field intensity, wind direction angle, rise-span ratio, and pretension on the displacement of the membrane. The results show that the probability density curve of wind-induced response has a certain skewness. The saddle membrane structure has the largest vibration amplitude of the membrane at 0° wind direction angle, and the most unfavorable wind pressure value of the membrane is negative. In reducing the displacement of the membrane, the effect of reducing the wind-induced vibration response by increasing the rise-span ratio of the structure is better than that of the pretension. This paper reveals that the law of wind-induced response can provide a theoretical basis for the design of membrane structures against typhoons.

Reconstruction of missing wind data based on limited wind pressure measurements and machine learning

In structural health monitoring (SHM), wind field monitoring sometimes suffers from data loss owing to monitoring device failure, which inevitably creates barriers to subsequent data analysis and data mining. To this end, a novel strategy for reconstructing missing wind field data based on machine learning (ML) utilizing limited wind pressure measurements is proposed in this paper. Several ML algorithms, including decision tree, random forest, gradient boosting regression tree, support vector regression, Gaussian process regression, and backpropagation neural network, are employed to characterize potential relationships between wind pressure information (including time series and statistical parameters of wind pressures) and wind field information (e.g., wind direction and wind speed). Moreover, the effect of input information (including the type of input variables as well as the number and location of pressure transducers providing input data) on reconstruction performance and efficiency is investigated. Field measured records from an SHM system in a 600-m-high supertall building during typhoons are utilized to validate the feasibility and robustness of the proposed strategy. The results show that the presented strategy can effectively reconstruct missing wind field information in the SHM of the skyscraper during typhoons. Compared with the time series of wind pressures, selecting statistical parameters of wind pressures as input variables can effectively improve the performance and efficiency of reconstruction models. Choosing appropriate input information (e.g., using multiple input variables, adopting data from a larger number of pressure transducers, and utilizing data from pressure transducers closer to an anemometer) is beneficial for enhancing the performance of reconstruction models.

False Alarm Causes and Wind Field Sensitivity Analysis of a Severe Rainfall Event in the Guangdong-Hong Kong-Macao Greater Bay Area Urban Cluster

On May 11, 2022, despite the favorable upper and lower-level circulation patterns of the high-altitude trough, shear line, and southwest jet stream, the urban cluster of the Guangdong-Hong Kong-Macao Greater Bay Area experienced light to moderate rainfall, deviating significantly from the forecasted heavy rain and local heavy rainstorm. This study explores the reasons for false alarms and predictability using ground observation data, radar data, ECMWF-ERA5 reanalysis field data, and ECMWF and CMA-TRAMS forecast data. The results indicate that the warm and moist airflow transported by the low-level jet stream was intercepted by the upstream MCS (mesoscale convective system) along the coastal area of western Guangdong, and inadequate conditions of negative vorticity dynamics led to insufficient moisture, thermodynamic, and dynamic conditions over the urban cluster, preventing the triggering of heavy precipitation. In addition, the 700 hPa westerly flow guiding the airflow and the stable low-level shear line, coupled with surface convergence lines, influenced the northward or southward movement of MCSs along the coastal and inland regions of western Guangdong. The weak and discontinuous intensity of echoes in the upstream Zhaoqing region further hindered the influence of surrounding echoes on the urban cluster. Numerical forecast models ECMWF and CMA-TRAMS overestimated the 850 hPa windspeed and 925 hPa meridional windspeed, resulting in the forecasted urban cluster experiencing heavy rain. Sensitivity tests of wind fields indicate that the 850 hPa wind field information is more sensitive to precipitation in the urban cluster. In this process, weak signal correction can be achieved in strong precipitation forecasts using the distinct signal of lower 850 hPa water vapor flux divergence compared to 925 hPa. Therefore, in the future, when the Guangdong-Hong Kong-Macao Greater Bay Area encounters similar warm-sector heavy rainfall events, adjustments to model forecasts can be made using specific 850 hPa elements such as wind speed, water vapor flux divergence, or specific humidity to enhance predictive accuracy.

Estimating Winter Arctic Sea Ice Motion Based on Random Forest Models

Sea ice motion (SIM) plays a crucial role in setting the distribution of the ice cover in the Arctic. Limited by images’ spatial resolution and tracking algorithms, challenges exist in obtaining coastal sea ice motion (SIM) based on passive microwave satellite sensors. In this study, we developed a method based on random forest (RF) models to obtain Arctic SIM in winter by incorporating wind field and coastal geographic location information. These random forest models were trained using Synthetic Aperture Radar (SAR) SIM data. Our results show good consistency with SIM data retrieved from satellite imagery and buoy observations. With respect to the SAR data, compared with SIM estimated with RF model training using reanalysis surface wind, the results by additional coastal information input had a lower root mean square error (RMSE) and a higher correlation coefficient by 31% and 14% relative improvement, respectively. The latter SIM result also showed a better performance for magnitude, especially within 100 km of the coastline in the north of the Canadian Arctic Archipelago. In addition, the influence of coastline on SIM is quantified through variable importance calculation, at 22% and 28% importance of all RF variables for east and north SIM components, respectively. These results indicate the great potential of RF models for estimating SIM over the whole Arctic Ocean in winter.

Mapping future offshore wind resources in the South China Sea under climate change by regional climate modeling

Cost-effective deployment of wind energy systems benefits from a thorough understanding of wind resources. Therefore, it is imperative to figure out how the spatiotemporal patterns of wind resources will evolve in a warmer future world. Based on a regional climate model system named RegCM4.7, this study investigates the future offshore wind resources under climate change in a tropical ocean, namely the South China Sea (SCS). First, 25-km-resolution wind field information over the SCS is retrieved from a regional climate model driven by 22 global climate models. Then, the simulated wind fields are validated against the reanalysis dataset ERA5 and the Cross-Calibrated Multi-Platform satellite dataset. Subsequently, the spatial distribution and temporal variation of wind resources under climate change are comprehensively assessed, and the inter-annual and intra-annual wind resource variability is quantified by the robust coefficient of determination. Last, the significance, robustness, and uncertainty of climate change signals are evaluated. It is projected that the wind power density will increase by 10–20 % in the northern SCS but decrease by 10–30 % in the southern SCS in 2081–2100 (compared to 1985–2004) under the Representative Concentration Pathway 8.5.

Simulation of Storm Surge Heights Based on Reconstructed Historical Typhoon Best Tracks Using Expanded Wind Field Information

A numerical model integrating tides, waves, and surges can accurately evaluate the surge height (SH) risks of tropical cyclones. Furthermore, incorporating the external forces exerted by the storm’s wind field can help to accurately reproduce the SH. However, the lack of long-term typhoon best track (BT) data degrades the SH evaluations of past events. Moreover, archived BT data (BTD) for older typhoons contain less information than recent typhoon BTD. Thus, herein, the wind field structure, specifically its relationship with the central air pressure, maximum wind speed, and wind radius, are augmented. Wind formulae are formulated with empirically adjusted radii and the maximum gradient wind speed is correlated with the central pressure. Furthermore, the process is expanded to four quadrants through regression analyses using historical asymmetric typhoon advisory data. The final old typhoon BTs are converted to a pseudo automated tropical cyclone forecasting format for consistency. Validation tests of the SH employing recent BT and reconstructed BT (rBT) indicate the importance of the nonlinear interactions of tides, waves, and surges for the macrotidal west and microtidal south coasts of Korea. The expanded wind fields—rBT—based on the historical old BT successfully assess the return periods of the SH. The proposed process effectively increases typhoon population data by incorporating actual storm tracks.

Digital twin of wind farms via physics-informed deep learning

The spatiotemporal flow field in a wind farm determines the wind turbines’ energy production and structural fatigue. However, it is not obtainable by the current measurement, modeling, and prediction tools in wind industry. Here we propose a novel data and knowledge fusion approach to create the first digital twin for onshore/offshore wind farm flow system, which can predict the in situ spatiotemporal wind field covering the entire wind farm. The digital twin is developed by integrating the Lidar measurements, the Navier–Stokes equations, and the turbine modeling using actuator disk method, via physics-informed neural networks. The design enables the seamless integration of Lidar measurements and turbine operating data for real-time flow characterization, and the fusion of flow physics for retrieving unmeasured wind field information. It thus addresses the limitations of existing wind prediction approaches based on supervised machine learning, which cannot achieve such prediction because the training targets are not available. Case studies of a wind farm under typical operating scenarios (i.e. a greedy case, a wake-steering case, and a partially-operating case) are carried out using high-fidelity numerical experiments, and the results show that the developed digital twin achieves very accurate mirroring of the physical wind farm, capturing detailed flow features such as wake interaction and wake meandering. The prediction error for the flow fields, on average, is just 4.7% of the value range. With the accurate flow field information predicted, the digital twin is expected to enable brand new research across wind farm lifecycle including monitoring, control, and load assessment.

A UAV Wind Field Perception System Inspired by Biological Perception

People have raised their expectations for UAV performance due to the widespread use of UAVs in both military and non-military settings. One of the most significant fields of research right now is how to enhance UAVs’ endurance capabilities. Many birds in the wild have the ability to fly for extended periods of time or great distances using the wind, which is called energy harvesting. A seabird called an albatross can use the wind’s horizontal gradient on the water’s surface to generate energy. By gliding, they may fly for a very long time without flapping their wings, thus lowering their own energy consumption. Due to the albatrosses’ nostrils sensitive sensory nerves and sensitivity to environmental information, such as airflow, the albatrosses can modify its flight attitude. Similar to this, real-time dynamic planning of the trajectory can only be done for UAVs in order to realize energy-capturing flying if correct and real-time wind field information is obtained. As a result, developing wind field sensing technology is crucial to the realization of energy-capturing gliding. In this study, we built a 3D wind field sensing system with wind vane sensor and pitot tube. Wind tunnel tests were used to calibrate and alter it. The system’s operation is initially validated by real flight, which may give environmental information, enabling UAVs to utilize the wind field as a reference for planning their flight paths.

Developing a deep learning-based storm surge forecasting model

Storm surge is the anomalous rising of the sea surface induced by intense atmospheric disturbances. The storm surge caused by tropical cyclones often causes great socio-economic, human activity, and life and property hazards to coastal areas. In terms of research resource consumption and computational time, machine learning algorithms that depend on data-driven strong nonlinear mapping skills outperform standard numerical model forecasting. To obtain a lighter and faster storm surge shortcoming forecast, we use a deep learning-based single-station water level prediction model for a storm surge at several locations in this work. In contrast to earlier research, this study employs convolutional neural networks to extract two-dimensional wind field information and merge them with local water level features to produce a more time-efficient intelligent forecast.

Измерение поля атмосферного ветра с помощью полностью волоконного лидара с длиной волны генерации 1.5 μm

In this paper, we propose a 1.5 µm all-fiber lidar system for atmospheric wind field measurements. The lidar emission source adopts the master oscillator power amplifier 1532 nm single-frequency all-fiber structure with a repetition frequency of 10 kHz. The lidar output energy is 427 µJ, and the maximum peak power is 610 W at the pulse width of 700 ns. The system utilizes Doppler wind speed coherent measurement technology, and the lidar source modulates an 80 MHz intermediate frequency carrier. The maximum longitudinal height signal distance of the lidar system is 4.5 km for atmospheric wind field, and the measurement accuracy can be within the range of 10-2 level. The lidar system can be used to detect high-altitude or long-distance high-precision wind field information. At the same time, the system can play an important role in future atmospheric environment engineering research.

Evaluation of two observation operator schemes for wind profiler radar data assimilation and its impacts on short-term forecasting

The assimilation of wind profiler (profiler) observations has the potential of improving the model initial condition of dynamical field with its high spatial and temporal wind field information. In this study, the impacts of two kinds of wind observation operators, i.e., eastward and northward velocities (u and v; uv_scheme) and wind speed and direction (spd and dir; sd_scheme), on analyses and forecasts are discussed to determine the optimal observation operator for profiler data assimilation. To compare the performance of the two observation operators on profiler data assimilation, we carry out a series of single wind tests and a two-week period cycling assimilation and forecasting experiments. The single wind tests show that the adjustment of the two observation operators to the analyses of wind can be significantly different with large innovation (OMB) of dir. The uv_scheme adjusts the corresponding model variables by ingesting profiler information of u and v winds, so that the analyzed spd is indirectly affected by the observed dir. By contrast, the changes in observed dir will not affect the analyzed spd in sd_scheme, as the observed spd and dir are independently assimilated. The case diagnoses illustrate that the sd_scheme generates the best analysis of wind in RMSE when validating against either the profiler or the pilot observations. The improved analysis of winds in turn leads to better wind forecasts and precipitation prediction, confirming the advantage of sd_scheme. Results from the cycling experiments further demonstrate that, compared with the uv_scheme, the sd_scheme shows decent improvements in wind forecasts, thus leading to better precipitation forecast skills.

Research and Optimization of Meteo-Particle Model for Wind Retrieval

As the aviation industry has entered a critical period of development, the demand for Automatic Dependent Surveillance Broadcast (ADS-B) technology is becoming increasingly urgent. Real-time detection of aviation wind field information and the early warning of wind field shear by atmospheric sounding system are two important factors related to the safe operation of aviation and airport. According to the advantages of ADS-B and Mode S data, this paper uses the Meteo-Particle (MP) model proposed by Sun et al., in their previous research to retrieve high-altitude wind field. Comparing the precision and accuracy of wind field retrieved results, and the optimization parameters of MP model suitable for meteorological model are further studied. To solve the problem of incomplete wind field coverage obtained by retrieval, an extrapolation algorithm of wind field is proposed. The results show that: (1) a comprehensive evaluation index is introduced, which can more effectively evaluate the comprehensive difference of wind field retrieval results in wind speed and direction. (2) The adaptability results of MP model in different periods and altitudes provide some reference for the research of other scholars. (3) The new parameter setting can improve the accuracy of the retrieved results, and the appropriate extrapolation of wind field fills in the blank part of aviation and meteorology.

Spatiotemporal wind field prediction based on physics-informed deep learning and LIDAR measurements

Spatiotemporal wind field information is of great interest in wind industry e.g. for wind resource assessment and wind turbine/farm monitoring & control. However, its measurement is not feasible because only sparse point measurements are available with the current sensor technology such as LIDAR. This work fills the gap by developing a method that can achieve spatiotemporal wind field predictions by combining LIDAR measurements and flow physics. Specifically, a deep neural network is constructed and the Navier–Stokes equations, which provide a good description of atmospheric flows, are incorporated in the deep neural network by employing the physics-informed deep learning technique. The training of this physics-incorporated deep learning model only requires the sparse LIDAR measurement data while the spatiotemporal wind field in the whole domain (which cannot be measured) can be predicted after training. This study, which can discover complex wind patterns that do not present in the training dataset, is totally distinct from previous machine learning based wind prediction studies which treat machine learning models as “black-box” and require the corresponding input and target values to learn complex relations. The numerical results on the prediction of the wind field in front of a wind turbine show that the proposed method predicts the spatiotemporal flow velocity (including both downwind and crosswind components) in the whole domain very well for a wide range of scenarios (including various measurement noises, resolutions, LIDAR look directions, and turbulence levels), which is promising given that only line-of-sight wind speed measurements at sparse locations are used.

An all-day lidar for detecting soft targets over 100 km based on superconducting nanowire single-photon detectors

It is a big challenge for lidar to detect soft targets over long distances in the atmosphere due to the low reflection of soft targets and the strong atmospheric attenuation. In this paper, we propose an all-day lidar system based on the 4 pixel array superconducting nanowire single-photon detectors. This significantly improves the detection efficiency of the aerosol and other targets by utilizing the advantages of high sensitivity, low dark count rate, wide dynamic range and photon number resolution. The system detects both soft targets and hard targets 100 km away in the atmosphere. In experiments, based on the photon-number resolving detection method, the lidar detects and distinguishes soft and hard targets simultaneously 50 km away during the day. Furthermore, the system obtains the wind field information in the atmosphere by monitoring clouds at a distance exceeding 86 km. The detection results indicate that the system is promising for applications as a long distance all-day lidar.

Long-distance mobile MAX-DOAS observations of NO2 and SO2 over the North China Plain and identification of regional transport and power plant emissions

In this study, the spatiotemporal distribution of the NO2 and SO2 Vertical Columns Densities (VCDs) in the North China Plain (NCP) region was measured by long-distance mobile measurements using a mobile Multi-AXis Differential Optical Absorption Spectroscopy (MAX-DOAS) instrument. The mobile observations were performed in both summer (July 2017) and winter (January and February 2018) and the total driving mileage exceeded 3000 km. The concentrations of NO2 and SO2 in different seasons and areas were significantly different. During winter, serious NO2 and SO2 pollution was observed in northern Anhui Province, central Shandong Province, and the Beijing-Tianjin-Hebei Region. The evolution and transportation processes of three typical heavy pollution cases were discussed in detail. Combined with the Weather Research and Forecasting with Chemistry model simulated wind field information, the NO2 transportation flux from northern Jiangsu Province to northern Anhui Province is quantified as 7.12 kg s−1. Finally, we estimated the NO2 and SO2 emissions from the Dezhou and Hengshui power plants by plume cross section scanning and encircled observations, respectively. The NO2 and SO2 emission fluxes of the Dezhou Power Plant are 0.79 kg s−1 and 1.11 kg s−1, respectively, while the NO2 and SO2 emission fluxes of the Hengshui Power Plant are 0.12 kg s−1 and 0.36 kg s−1, respectively. This study quantitatively analyzed the transportation of atmospheric pollutants and power plant emissions, which is helpful for understanding the occurrence and evolution of pollution and is also useful for the government to introduce policies to protect and control the atmospheric environment.

Retrieval of Sea Surface Wind Fields Using Multi-Source Remote Sensing Data

Timely and accurate sea surface wind field (SSWF) information plays an important role in marine environmental monitoring, weather forecasting, and other atmospheric science studies. In this study, a piecewise linear model is proposed to retrieve SSWF information based on the combination of two different satellite sensors (a microwave scatterometer and an infrared scanning radiometer). First, the time series wind speed dataset, extracted from the HY-2A satellite, and the brightness temperature dataset, extracted from the FY-2E satellite, were matched. The piecewise linear regression model with the highest R2 was then selected as the best model to retrieve SSWF information. Finally, experiments were conducted with the Usagi, Fitow, and Nari typhoons in 2013 to evaluate accuracy. The results show that: (1) the piecewise linear model is successfully established for all typhoons with high R2 (greater than 0.61); (2) for all three cases, the root mean square error () and mean bias error (MBE) are smaller than 2.2 m/s and 1.82 m/s, which indicates that it is suitable and reliable for SSWF information retrieval; and (3) it solves the problem of the low temporal resolution of HY-2A data (12 h), and inherits the high temporal resolution of the FY-2E data (0.5 h). It can provide reliable and high temporal SSWF products.

Real-time wind field measurements using all-fiber mobile Doppler wind lidar

Mobile, all-fiber Doppler wind lidar based on a real-time motion wind retrieval algorithm method is demonstrated. The mobile wind lidar can negate the influences of attitude, velocity, and other factors on the Doppler frequency and is capable of obtaining more accurate real-time mid- to low-altitude wind field information over a large space. In the moving platform test, the measurement error of the wind speed is 0.33 m / s, and the measurement error of the wind direction is 5.32 deg.

In-flight Wind Field Identification and Prediction of Parafoil Systems

The wind field is an essential factor that affects accurate homing and flare landing of parafoil systems. In order to obtain the ambient wind field during the descent of a parafoil system, a combination method of in-flight wind field identification and prediction is proposed. First, a wind identification method only using global position system information is derived based on the flight dynamics of parafoil systems. Then a wind field prediction model is constructed using the atmospheric dynamics, and the low-altitude wind field is predicted based on the identified wind field of high-altitude. Finally, simulations of wind field identification and prediction are conducted. The results demonstrate that the proposed method can identify the wind fields precisely and also predict the wind fields reasonably. This method can potentially be applied in practical parafoil systems to provide wind field information for homing tasks.