Wind Speed Information Research Articles

A Remote Monitoring System for Wind Speed and Direction Based on Non-Contact Triboelectric Nanogenerator

Wind speed and wind direction sensors are crucial sensor categories in the Internet of Things (IoT), particularly vital in fields such as meteorological monitoring, construction engineering, transportation engineering, and ocean engineering. However, power supply remains a key limiting factor for the widespread application of these sensors in large-scale sensor networks. In this paper, a self-powered, non-contact wind speed and direction sensor based on the triboelectric nanogenerator (SD-TENG) is proposed. The optimized wind speed measurement structure has a start-up wind speed as low as 0.2m/s and exhibits good linearity in the range of 0.2-29m/s. Additionally, the sensor demonstrates high temperature and humidity resistance, with a voltage attenuation of 2.4% at 45°C ambient temperature and 9.8% at 95% relative humidity. For wind direction measurement, the design of non-uniform electrodes enhances the recognition capability of different channels. To meet the demands of remote monitoring, we have designed an advanced signal processing circuit that can directly convert the raw output of a wind speed sensor into wind speed information and upload it to a cloud platform via a host. This system not only records and displays wind speed data in real-time but also features historical data storage and alarm functionalities, enhancing the intelligence and automation levels of wind speed monitoring. Additionally, users can access and analyze wind speed data through both computer and mobile devices, ensuring the system's efficiency and reliability. This work provides crucial technical support for the advancement of smart cities, clean energy, and environmental monitoring, thereby promoting the application and dissemination of IoT technology in the environmental field.

Triboelectric-electromagnetic hybrid wind energy harvesting and multifunctional sensing device for self-powered smart agricultural monitoring

The power supply for numerous small electronic sensors, which are extensively used in agricultural settings, is a critical challenge for advancing the Agricultural Internet of Things (IoT) and fostering the development of smart agriculture in the upcoming era of intelligent technology. Hereby, a triboelectric-electromagnetic hybrid wind energy harvesting and sensing device (TEH-WSD) based on triboelectric-electromagnetic hybrid generator (TEHG) and multi-channel triboelectric nanogenerator (MC-TENG) for wind direction and speed sensing is proposed. Among them, the TEHG is used to collect wind energy. The instantaneous peak power of TENG units and EMG unit in TEHG can reach 5.95 mW and 5.65 mW respectively. In addition, the modified TEHG can measure the wind speed within the range of 3–5 m/s, and the output signal of MC-TEHG can be used to measure the wind direction based on the corresponding circuit and designed program continuously. Overall, TEH-WSD is capable of detection for the wind direction, wind speed, temperature and humidity information in the environment and this work provides a solution for wind energy collection and self-powered environmental monitoring in agricultural environment.

Assessing the Complementarity of Wind and Solar Energy in Kentucky

Recently, there has been a push by countries to diversify their energy mix considering various factors. In this regard, there have been several studies conducted to assess the potential for using sources such as wind and solar to generate supplemental energy to the already present energy generation setup. In this regard, this study explores the potential of wind for the Commonwealth of Kentucky. To perform this study, wind data were sourced for eight locations across Kentucky from the publicly accessible wind speed information present at Weather Underground for the years 2020–2021 (two years). An analysis was performed concerning the seasonal, monthly, and hourly variation in the wind speed so as to identify the expected times of sufficient wind energy generation. Moreover, a comparison of the collected data was performed with data from a home-based weather station as well as a deployed wind turbine to validate the variation pattern of the publicly sourced data. Finally, in order to investigate the variation patterns of wind and solar energy sources, a comparative analysis was also performed using data from a solar power generation plant in Kentucky. A seasonal and monthly complementarity was observed between the wind and solar energy. However, when considering daily patterns, the wind was found to follow solar generation with an offset. While further research is required, this analysis indicates that it is possible to deploy wind energy power generation projects in the Commonwealth of Kentucky. The seasonal complementary behavior of wind and solar energy can be used along with battery storage in conjunction with natural gas to provide a diversified electricity generation portfolio.

Reconstruction of environmental site conditions by the integration of SCADA and reanalysis data

For the operational optimization of wind farms, AEP estimation and other tasks, high quality data of environmental conditions at the site are necessary. However, such data is often not available or has insufficient quality. This work tries to fill this gap, by integrating two data sources: the (usually available) operational data from the SCADA (Supervisory Control and Data Acquisition) system, and reanalysis data. SCADA data streams contain measurements from each wind turbine in the farm, but they are affected by various sources of uncertainty (including local flow effects, miscalibration, etc.), and might contain gaps. Meteorological reanalysis datasets can be used to fill gaps and complement SCADA data. However, modelled data can contain a wide range of biases and errors, due to limited model fidelity, coarse spatial and temporal resolution, inaccuracies in the input data feeding the model, etc. This study considers various methods to extract and merge wind speed and direction information from these diverse data sources. The analysis is based on field data measured at two experimental test sites, an offshore site equipped with 111 multi-MW turbines and a lidar buoy, and an onshore site equipped with 14 multi-MW wind turbines and a lidar. The methods are evaluated in the spectral and temporal domains by comparing the reconstructed wind characteristics with measurements from the lidars.

Performance similarities between standard and retrofit LiDAR-assisted control for wind turbines

LiDAR-assisted control has proven to be a highly effective method for mitigating rotor speed deviations and reducing loads in wind turbines. This effectiveness stems from the ability of feedforward controllers to utilize incoming wind speed information obtained from LiDARs, enabling advanced blade pitch actions before the wind disturbs the turbine. However, the standard implementation of LiDAR-assisted control often necessitates modifications to the existing feedback controller, where the feedforward pitch rate is typically integrated into the feedback controller. This process can be challenging in terms of accessibility and may be limited to specific stakeholders, such as turbine manufacturers.A retrofit design provides an ideal solution, where the retrofit LiDAR-assisted controller modifies the rotor speed measurement to induce pitch actions without requiring alterations to the existing feedback controller. This paper aims to demonstrate the performance similarities between the standard LiDAR-assisted controller and its retrofit counterpart. Specifically, we establish that the retrofit LiDAR-assisted controller, with appropriate tuning, is equivalent to its standard counterpart. This equivalence implies that architecturally dissimilar controllers can yield the same performance in terms of rotor speed deviations and tower load reductions. The presented findings are supported by results obtained from high-fidelity closed-loop turbine simulations.

U-Plume: automated algorithm for plume detection and source quantification by satellite point-source imagers

Abstract. Current methods for detecting atmospheric plumes and inferring point-source rates from high-resolution satellite imagery are labor-intensive and not scalable with regard to the growing satellite dataset available for methane point sources. Here, we present a two-step algorithm called U-Plume for automated detection and quantification of point sources from satellite imagery. The first step delivers plume detection and delineation (masking) with a U-Net machine learning architecture for image segmentation. The second step quantifies the point-source rate from the masked plume using wind speed information and either a convolutional neural network (CNN) or a physics-based integrated mass enhancement (IME) method. The algorithm can process 62 images (each measuring 128 pixels × 128 pixels) per second on a single 2.6 GHz Intel Core i7-9750H CPU. We train the algorithm using large-eddy simulations of methane plumes superimposed on noisy and variable methane background scenes from the GHGSat-C1 satellite instrument. We introduce the concept of point-source observability, Ops=Q/(UWΔB), as a single dimensionless number to predict plume detectability and source rate quantification error from an instrument as a function of source rate Q, wind speed U, instrument pixel size W, and instrument-dependent background noise ΔB. We show that Ops can powerfully diagnose the ability of an imaging instrument to observe point sources of a certain magnitude under given conditions. U-Plume successfully detects and masks plumes from sources as small as 100 kg h−1 in GHGSat-C1 images over surfaces with low background noise and successfully handles larger point sources over surfaces with substantial background noise. We find that the IME method for source quantification is unbiased over the full range of source rates, while the CNN method is biased towards the mean of its training range. The total error in source rate quantification is dominated by wind speed at low wind speeds and by the masking algorithm at high wind speeds. A wind speed of 2–4 m s−1 is optimal for detection and quantification of point sources from satellite data.

Assessment of land use/ land cover change derived catchment hydrologic response: An integrated parsimonious hydrological modeling and alteration analysis based approach

Land use/land cover (LULC) change, often a consequence of natural or anthropogenic drivers, plays a decisive role in governing global catchment dynamics, and subsequent impact on regional hydrology. Insight into the complex relationship between the drivers of LULC change and catchment hydrology is of utmost importance to decision makers. Contemplating the dynamic rainfall-runoff response of the Indian catchments, this study proposes an integrated modeling-based approach to identify the drivers and relative contribution to catchment hydrology. The proposed approach was evaluated in the tropical climate Nagavali River Basin (NRB) (9512 km2) of India. The Soil and Water Assessment Tool (SWAT) hydrological model, which uses daily-scale rainfall, temperature, wind speed, relative humidity, solar radiation, and streamflow information was integrated with the Indicators of Hydrologic Alteration (IHA) technique to characterize the plausible changes in the flow regime of the NRB. Subsequently, the Partial Least Squares Regression (PLSR) based modeling analysis was performed to quantify the relative contribution of individual LULC components on the catchment water balance. The outcomes of the study revealed that forest land has been significantly converted to agricultural land (45–59%) across the NRB resulting in mean annual streamflow increase of 3.57 m3/s during the monsoon season. The affinity between land use class and streamflow revealed that barren land (CN = 83–87) exhibits the maximum positive response to streamflow followed by the built-up land (CN = 89–91) and fallow land (CN = 88–93). The period 1985–1995 experienced an increased ET scenario (911–1050 mm), while the recent period (2005–2020) experienced reduced ET scenario owing to conversion of forest to agricultural land. Certainly, the study endorses adopting the developed methodology for understanding the complex land use and catchment-scale hydrologic interactions across global-scales for early watershed management planning.

On-site wind speed recovery from smartphone audio: Time domain deep learning approach, laboratory validation and outdoor field test

On-site wind speed information is paramount in various engineering domains. Conventional methods, reliant on expensive equipment and skilled operators, present challenges for practical uses. This study presents a novel time domain approach to recover wind speed from smartphone audio time series using deep learning. Leveraging the ubiquity of smartphones, the proposed approach offers a cost-effective and accessible alternative for wind speed measurement based on audio signals, allowing personnels to obtain on-site wind speed using smartphones. The model is built upon muti-layer architecture of one-dimensional convolutional neural network, and is trained and fine-tuned with high-fidelity wind tunnel laboratory data. The model’s performance is tested across different smartphone models and outdoor environments to assess its generalizability. The presented approach is generally effective, providing an overall accuracy with mean squared error of 0.2 to 0.9 (m/s)2. Although the model’s accuracy diminishes slightly in outdoor settings, it still shows favorable agreement with specialized wind gauges.

Point and interval wind speed forecasting of multivariate time series based on dual-layer LSTM

Accurate prediction of wind speed is significant importance in various applications such as renewable energy management, weather prediction, and aviation safety. However, more papers only focus on the time series of wind speed, ignoring the impact of other factors on wind speed, which will reduce the effectiveness of wind speed prediction. Therefore, an innovative methodology is suggested for both point and interval forecasts of multivariate wind speed time series using a Dual-layer Long Short-Term Memory (NVDL) in this paper. The proposed multivariate forecasting model takes into account the dependencies and correlations among different variables, which are essential for capturing the complex dynamics of wind speed variations. The first layer of the Dual-layer LSTM is responsible for capturing the temporal dependencies within each variable individually, while the second layer captures the interdependencies among the variables. By incorporating this dual-layer architecture, our model effectively captures the complex spatiotemporal patterns present for multivariate wind speed information. The results obtained from both interval and point prediction demonstrate that the proposed methodology outperforms all comparative models in the precision and stability of wind speed forecasting. Therefore, the proposed forecasting methodology, characterized by minimal prediction errors and exceptional generalization ability, can be a reliable tool for smart grid programming.

An image processing approach to reconstruct wind using long-range wind lidars

Lidar-based wind sensing technology, originally used in the wind energy sector, is now being utilized in wind engineering to monitor wind action for designing fjord-crossing infrastructure like long-span bridges. Accurate estimation of design wind loads is crucial for the design process of such bridges. This paper examines wind data from two pairs of long-range lidars positioned along one side of the Sulafjorden, Norway, in a measurement campaign initiated by the Norwegian Public Roads Administration (NPRA). Two different scanning modes, the Plan Position Indicator (PPI) mode and the staring mode, with the fixed line-of-sight (LOS) orientation, are used to determine the line-of-sight wind speeds, the wind direction, and the related along wind speed. An image processing approach is used to compute along wind velocity information from all range gates based on the LOS wind speeds from two near-parallel lidar beams. The estimated along wind velocity is validated through wind data calculated at the intersection point of the two lidars, which provides velocity data with two horizontal components. The image-based reconstruction is found to produce reasonable wind speed estimates with a goodness-of-fit coefficient of R2 = 0.903. The mean wind direction estimate is smoothened but still comparable to the actual wind direction. The image processing approach shows promising potential to provide wind speed and direction information for all range gates for the lidar setup used, which can supplement traditional single-point wind velocity characterization by dual lidars.

An Autoencoder‐Based Deep‐Learning Method for Augmenting the Sensing Capability of Piezoelectric Microelectromechanical System Sensors in a Fluid‐Dynamic System

Herein, an innovative deep‐learning architecture is proposed to enhance the sensing capabilities of a microelectromechanical system (MEMS) used in fluid dynamic applications. The MEMS sensor comprises a polyvinylidene fluoride flexible (PVDF) piezoelectric flag and a bluff body, with vortex generation influenced not only by the bluff body's geometry but also by the input fluid speed. As a result, mechanical vibrations are induced in the piezoelectric flag, leading to charge displacement and the generation of electrical voltage signals. Through the developed deep learning method, accurate extraction of wind speed and successful classification of turbulence are achieved. Experimental tests in a wind tunnel, involving various wind speeds and bluff body geometries, demonstrate the robust correlation between the extracted continuous manifold in Fourier spectra and wind speed. By incorporating a feed‐forward network alongside the autoencoder, wind speed information even under strong turbulence is extracted. Moreover, the deep learning method's ability to classify different bluff bodies, independent of wind speed, is investigated. The findings reveal a unique capability to fingerprint turbulence and distinguish them for various applications. This research showcases the potential of our deep learning‐based MEMS systems for enhancing fluid dynamic sensing and classification tasks.

PSVIII-A-3 Activity Behavior and Growth Performance During Summer of Grazing Beef Heifers with Divergent Residual Feed Intake

Abstract Concerns about sustainability of beef production systems generate an interest in improved feed efficiency; however, there is a lack of research evaluating cattle activity budgets in response to the interaction between residual feed intake (RFI) and environment while grazing. Lying behavior and activity can provide insight into how animals interact with the environment and serve as an indicator of animal comfort. Furthermore, exposure to severe environmental factors can change behavioral patterns and impair animal performance. Therefore, this study evaluated activity budgets and performance in grazing beef heifers with divergent residual feed intake (RFI) during the summer season. From July to August 2022, forty-four crossbred beef heifers [358 ± 4.78 kg of body weight (BW); approximately 14 months of age] previously tested for RFI in drylot and classified as more (n = 21; LOW-RFI = -0.9 ± 0.70) or less feed efficient (n = 23; HIGH-RFI = 1.3 ± 1.00) were grazed at 2.72 AUM/ha over 7 wk in Alberta, Canada. IceRoboticTM pedometers (IceQube+) were used to track 24-hr heifer activity budgets [n = 43; total steps and lying and standing time (min/d and min/h) for 36 d]. Full BW was obtained on d -1, 0, 14, 28, 42, and 43 while fat scan on rib and rump were measured by ultrasound (Aloka 500 V diagnostic real-time) on d 0 and 42. Air temperature, relative humidity, wind speed and solar radiation information were collected within 1 km of the grazed area to calculate the Comprehensive Climate Index (CCI). Based on CCI, weather conditions were considered to impose risk to cause mild, moderate, severe, and extreme stress for 5, 18, 7, and 1 days, respectively. For BW, average daily gain (ADG), rump (RF) and rib fat (RiF), the data were analyzed as a completely randomized design, while behavior activity included repeated measures. An RFI x day interaction was observed for lying and standing times (P = 0.02) and total steps (P = 0.001). Greater number of steps (P < 0.01) and an increased standing time (P < 0.01) were observed in HIGH-RFI heifers. RFI × hour interaction was observed for lying and standing times (P = 0.006), where LOW-RFI heifers spent more time lying at 10:00 am (P < 0.01). Furthermore, LOW-RFI heifers had decreased number of steps per hour throughout the study (P = 0.03; 178 vs 191 ± 4.1). No effects were observed for ADG, BW, RF, and RiF (P > 0.24). In summary, selected activity behaviors differed between beef heifers with divergent residual feed intake while summer grazing. Further studies are needed to investigate the effects of continuous changes in weather conditions to better understand the environmental impacts on animal behavior while selecting for more efficient beef cattle.

Output Feedback Adaptive Pitch Angle Control for Variable Speed Wind Turbine

This paper proposes a simple adaptive controller for pitch angle control of the variable speed wind turbine. The aim of the controller is to keep the speed of the generator at the rated value when the wind speed is above the nominal value. The controller contains two components: the output feedback component for stable purpose and the adaptive component to cancel the effect of nonlinearities, system uncertainties, and external disturbances. The controller does not need information of system parameters and wind speed then it is robust to the system uncertainties as well as wind speed varying. Moreover, this controller is quite simple, therefore it is easy for implementation. The stability of the closed loop and the convergence of the adaptive law is mathematically proven via the Lyapunov theory. The effectiveness of the given scheme is verified via simulation under two scenarios: step wind speed and random wind speed. Also, the comparisons are done between proposed adaptive controller, the corresponding nonadaptive controller, and the PID controller. The simulation results show that, for both cases, the given adaptive controller has the best responses with low steady state error of rotor speed, low overshoot, and smooth output power.

An optimal sensor placement scheme for wind flow and pressure field monitoring

Sensing of wind pressures on wind-sensitive structures and wind speeds in densely populated cities is crucial to ensure structure safety and monitoring urban wind environment. However, these wind pressure or wind speed sensors are often sparsely distributed due to their high cost, and consequently the collected wind pressure or speed information is limited. Under these circumstances, it is particularly important to place these limited number of sensors in the optimal locations with the capacity of extracting the most information. In this study, an optimal sensor placement scheme based on multi-resolution dynamic mode decomposition (mrDMD), a space post-processing function: signed distance function (SDF) and particle swarm optimization-random forest (PSO-RF) is proposed, called mrDMD-SDF-PSO-RF (mrDSPR). The effectiveness of the proposed scheme is verified using two case studies of reconstruction based on sparse sensors: wind flow field of a city block and wind pressure field of a single building. The result indicates that the proposed scheme significantly improves the efficiency of the sensors. Therefore, based on the proposed optimal sensor placement scheme, the locations of placing sensors can be well-designed for various wind engineering problems, which builds a solid foundation of super-resolution reconstruction of wind pressure on buildings and wind flow field in urban environment.

Impact of Mode-S Enhanced Surveillance Weather Observations on Weather Forecasts over the MetCoOp Northern European Model Domain

Abstract Meteorological Cooperation on Operational Numeric Weather Prediction (MetCoOp) is a northern European collaboration on operational numerical weather prediction based on a common limited-area, kilometer-scale ensemble system. The initial states of this model are produced using a three-dimensional, variational, data assimilation scheme utilizing a large number of observations from conventional in situ measurements, weather radars, global navigation satellite systems, advanced scatterometer data, and satellite radiances. Since 2019, the MetCoOp system was enhanced by utilization of observations based on selective mode (Mode-S) enhanced surveillance (EHS) reports that are broadcast by aircraft in response to interrogation from air traffic control radar. These observations, obtained from the European Meteorological Aircraft Derived Data Center, are used to derive indirect information of atmospheric wind speed and temperature. The use of these observations compensated for the considerably reduced number of direct aircraft observations encountered as an effect of the COVID-19 pandemic. The MetCoOp observation handling system is described, with emphasis on Mode-S EHS data. The quality of these observations is evaluated, and we show that they are suitable to be used in MetCoOp data assimilation. The impact on average forecast verification scores of the additional Mode-S EHS data is slightly positive. The benefit of using Mode-S EHS was demonstrated for an individual case and also a demonstration of utilizing the Stockholm Arlanda receiver data in assimilation has been performed.

A novel approach to Weibull distribution for the assessment of wind energy speed

The latest power sources trend is renewable energy. Numerous researchers, innovators, and technicians work diligently to exploit renewable power. The effectiveness of wind speed data is crucial when estimating wind energy production from a wind turbine. For evaluating wind possibilities, there are several studies on the four generally utilized wind speed distribution models. However, there isn’t enough research to determine how sensitive these models are to the observed wind speed data quality. The current research aims to demonstrate wind speed data distribution using Weibull distribution. Researchers discover that the statistical data of the changes among consecutive points of the anticipated wind speed information may drastically deviate from the characteristics of the observed wind. Firstly, the variables of Weibull distribution are estimated using modified maximum likelihood (MML) methods, Energy Pattern Factor and method of moment (MOM). The suggested approach is based on an algorithm that compares the statistical characteristics of the produced and actual wind speeds to get a more precise estimation. Therefore, the findings show that using Weibull distribution to represent wind speed modelling is feasible, precise, and efficient.

Wind resource modelling of entire sites using Large Eddy Simulation

Accurate wind speed, direction, shear and turbulence information are key inputs for wind farm energy yield and turbine suitability assessment. Met-mast measurements offer limited coverage both in horizontal and vertical directions, with few locations on the site and heights typically limited to below turbine hub-height. Simulation becomes an increasingly appealing option to mitigate these shortcomings. Current flow models used in wind resource assessment (WRA) rely heavily on assumptions to simplify complex atmospheric physics and reduce computing expense. As a result, they cannot resolve turbulence, and output average or annualized statistics. A new generation of models utilizing high-performance-computing now allows for high-fidelity simulation of sites with fewer modelling assumptions. In this work, we demonstrate and validate a computationally feasible large eddy simulation (LES) approach to generate high-fidelity time series of relevant quantities for WRA - including resolved turbulence - at sites in the order of 1000 square kilometres. The capability of the LES to produce insights in complex terrain is presented. A validation of the approach is performed by comparison of raw model results against met-mast measurements in a wide range of sites with varied complexities. Results show good agreement with measurements, with further error reduction expected through downstream adjustment of the model results with met-mast data on sites. Overall, the approach enables high-fidelity time-series based modelling for wind resource characterization and value addition to the WRA tool chain.

Wind Speed Measurement via Visual Recognition of Wind-Induced Waving Light Stick Target

Wind measurement in confined spaces is a challenge due to the influence of the dimensions of anemometers in intrusive flow-field measurements where the anemometer probes directly contact and influence the near-probe flow field. In this work, a new wind speed detection methodology is proposed based on wind-induced motion of a stick via vision-based recognition. The target’s displacement in pixel coordinates is mapped to its angular displacement in world coordinates to derive wind speed and direction information by applying the calibration coefficients. Simulation experiments were carried out to validate the model, the error of which was within an angular displacement of 4.0° and 3.0° for wind speed and direction detections, respectively. When applied to the measurement of wind speed in the inner equipment cabin of a stationary high-speed train, the error was within ±1.1 m/s in terms of average RMSE. Thus, the proposed method provides an accurate and economic option for monitoring 2D wind in a confined space.

Wind-sea and swell separation of 1D wave spectrum by deep learning

In this study, three existing methods to separate wind-sea and swell from 1D wave spectra, namely the Pierson–Moskowitz (PM) method, the steepness method (HP method), and the overshooting method (JP method), were evaluated statistically using data from 38 meteorological buoys from National Data Buoy Center (NDBC) over the period 2010–2020. Among the three methods, the PM method shows the best agreement with the 2D separation method, because it uses the wind speed information as an input term. Using these buoy data, a new 1D wind-sea-swell separation method based on deep learning is proposed. This new method can directly compute the wind-sea or swell significant wave height (SWH) from a 1D wave spectrum with or without wind speed data as input. When the wind speed is used as an input term, the overall root-mean-square error (RMSE) of the method for wind-sea/swell SWH is 0.27/0.36 m compared to the 2D separation method, which outperforms all existing 1D separation methods. When there is no wind speed data, the RMSE of wind-sea/swell SWH can still reach 0.36/0.41 m, which is similar to the accuracy of the PM method uses wind speed as input. Both methods are robust and have no significant geographical dependence.

A wind speed vector-wind power curve modeling method based on data denoising algorithm and the improved Transformer

• A wind power curve modeling method based on improved Transformer is proposed. • The CEEMDAN algorithm is introduced for data denoising. • Multivariate combined with DTW analysis optimizes the network input. • Convolution combined with multi-head attention enhances information mining. • A multi-dimensional evaluation mechanism is established to evaluate the model. The complex nonlinear relationship between the wind speed and the wind power, and the singularity of wind speed information leads to the lack of accuracy of current wind power curve modeling. To address the problem, this paper presents a high-precision wind power curve modeling method based on the wind speed vector, including the wind speeds and wind directions at different heights of the wind measuring tower. First, considering the stochastic fluctuation of the wind speed vectors and wind power sequences, complementary ensemble empirical modal decomposition with adaptive noise (CEEMDAN) is used to decompose and reconstruct the high-noise data. Second, based on the reconstructed data, dynamic time warping (DTW) is adopted to analyze the lagged causality between current wind power and historical wind speed series. In order to better mine the rules, the improved Transformer network is proposed with two convolutional layers and multi-head attention mechanisms to develop the wind speed vector-wind power curve model. Finally, through comparative experiments with the mainstream methods, the advancement of the proposed wind power curve modeling method is verified from the perspective of modeling error and its distribution