This dataset is a unique compilation of field-based meteorological observations and wind power generation data, collected directly from one of our company's operational sites. The dataset represents a detailed hourly record, starting from January 2, 2017. This rich dataset provides real-world insights into the interplay between various weather conditions and wind energy production.

Context and Inspiration: The dataset was conceived out of the necessity to understand the dynamic relationship between meteorological variables and their impact on wind power generation. By collecting data directly from the field and the wind turbine installations, we aim to provide a comprehensive and authentic dataset that can be instrumental for industry-specific research, operational optimization, and academic purposes.

Data Collection: Data was meticulously gathered using state-of-the-art equipment installed at the site. The meteorological instruments measured temperature, humidity, dew point, and wind characteristics at different heights, while power generation data was recorded from the wind turbines' output. This dataset is a unique compilation of field-based meteorological observations and wind power generation data, collected directly from one of our company's operational sites. The dataset represents a detailed hourly record, starting from January 2, 2017. This rich dataset provides real-world insights into the interplay between various weather conditions and wind energy production.

Potential Uses: This dataset is ideal for industry experts, researchers, and data scientists exploring renewable energy, especially wind power. It can aid in developing predictive models for power generation, studying environmental impacts on renewable energy sources, and enhancing operational efficiency in wind farms.

Wind resource data for North America was produced using the Weather Research and Forecasting Model (WRF). The WRF model was initialized with the European Centre for Medium Range Weather Forecasts Interim Reanalysis (ERA-Interm) data set with an initial grid spacing of 54 km. Three internal nested domains were used to refine the spatial resolution to 18, 6, and finally 2 km. The WRF model was run for years 2007 to 2014. While outputs were extracted from WRF at 5 minute time-steps, due to storage limitations instantaneous hourly time-step are provided for all variables while full 5 min resolution data is provided for wind speed and wind direction only. The following variables were extracted from the WRF model data: - Wind Speed at 10, 40, 60, 80, 100, 120, 140, 160, 200 m - Wind Direction at 10, 40, 60, 80, 100, 120, 140, 160, 200 m - Temperature at 2, 10, 40, 60, 80, 100, 120, 140, 160, 200 m - Pressure at 0, 100, 200 m - Surface Precipitation Rate - Surface Relative Humidity - Inverse Monin Obukhov Length

Context

High precision and reliable wind speed forecasting is a challenge for meteorologists. Severe wind due to convective storms, causes considerable damages (large scale forest damage, outage, buildings/houses damage, etc.). Convective events such as thunderstorms, tornadoes as well as large hail, strong winds, are natural hazards that have the potential to disrupt daily life, especially over complex terrain favoring the initiation of convection. Even ordinary convective events produce severe winds which causes fatal and costly damages. Therefore, wind speed prediction is an important task to get advanced severe weather warning. This dataset contains the responses of a weather sensor that collected different weather variables such as temperatures and precipitation.

Content

The dataset contains 6574 instances of daily averaged responses from an array of 5 weather variables sensors embedded in a meteorological station. The device was located on the field in a significantly empty area, at 21M. Data were recorded from January 1961 to December 1978 (17 years). Ground Truth daily averaged precipitations, maximum and minimum temperatures, and grass minimum temperature were provided.

Attribute Information

1. DATE (YYYY-MM-DD)
2. WIND: Average wind speed [knots]
3. IND: First indicator value
4. RAIN: Precipitation Amount (mm)
5. IND.1: Second indicator value
6. T.MAX: Maximum Temperature (°C)
7. IND.2: Third indicator value
8. T.MIN: Minimum Temperature (°C)
9. T.MIN.G: 09utc Grass Minimum Temperature (°C)

Citation Request

If you want to cite this data:

fedesoriano. (April 2022). Wind Speed Prediction Dataset. Retrieved [Date Retrieved] from https://www.kaggle.com/datasets/fedesoriano/wind-speed-prediction-dataset

The 2023 National Offshore Wind data set (NOW-23) is the latest wind resource data set for offshore regions in the United States, which supersedes, for its offshore component, the Wind Integration National Dataset (WIND) Toolkit, which was published about a decade ago and is currently one of the primary resources for stakeholders conducting wind resource assessments in the continental United States. The NOW-23 data set was produced using the Weather Research and Forecasting Model (WRF) version 4.2.1. A regional approach was used: for each offshore region, the WRF setup was selected based on validation against available observations. The WRF model was initialized with the European Centre for Medium Range Weather Forecasts 5 Reanalysis (ERA-5) data set, using a 6-hour refresh rate. The model is configured with an initial horizontal grid spacing of 6 km and an internal nested domain that refined the spatial resolution to 2 km. The model is run with 61 vertical levels, with 12 levels in the lower 300m of the atmosphere, stretching from 5 m to 45 m in height. The MYNN planetary boundary layer and surface layer schemes were used the North Atlantic, Mid Atlantic, Great Lakes, Hawaii, and North Pacific regions. On the other hand, using the YSU planetary boundary layer and MM5 surface layer schemes resulted in a better skill in the South Atlantic, Gulf of Mexico, and South Pacific regions. A more detailed description of the WRF model setup can be found in the WRF namelist files linked at the bottom of this page. For all regions, the NOW-23 data set coverage starts on January 1, 2000. For Hawaii and the North Pacific regions, NOW-23 goes until December 31, 2019. For the South Pacific region, the model goes until 31 December, 2022. For all other regions, the model covers until December 31, 2020. Outputs are available at 5 minute resolution, and for all regions we have also included output files at hourly resolution. The NOW-23 data are provided here as HDF5 files. Examples of how to use the HSDS Service to Access the NOW-23 files are linked below. A list of the variables included in the NOW-23 files is also linked below. No filters have been applied to the raw WRF output.

In this dataset the anther's analysis is based on data from NREL about Solar & Wind energy generation by operation areas.

NASA Prediction of Worldwide Energy Resources

Solar

Monthly averages for global horizontal radiation over 22-year period (Jul 1983

• Jun 2013)

Wind

Monthly average wind speed at 50m above the surface of earth over a 30-year period (Jan 1984 - Dec 2013)Year: Averaged Over 10 to 15 years

COA = central operating area.

EOA = eastern operating area.

SOA = southern operating area.

WOA = western operating area. Source: NRELSource Link

This dataset comes from a single turbine on an inland wind farm. The dataset covers the duration of one year, but data at some of the time instances are missing. Two time resolutions are included in the dataset: the 10-min data and the hourly data; the latter is the further average of the former. For each temporal resolution, the data is arranged in three columns. The first column is the time stamp, the second column is the wind speed, and the third column is the wind power. This dataset is used in Chapter 2 of the Data Science for Wind Energy book.

Data repository for measurements from 12 wind masts in Pakistan. Data transmits daily reports for wind speed, wind direction, air pressure, relative humidity and temperature. Please refer to the country project page for additional outputs and reports, including the installation reports: http://esmap.org/node/3058. For access to maps and GIS layers, please visit the Global Wind Atlas: https://globalwindatlas.info/ Please cite as: [Data/information/map obtained from the] “World Bank via ENERGYDATA.info, under a project funded by the Energy Sector Management Assistance Program (ESMAP).

WIND Toolkit Long-term Ensemble Dataset (WTK-LED), an updated version of the meteorological WIND Toolkit, is a meteorological dataset providing high-resolution time series, including interannual variability and model uncertainty of wind speed at every modeling grid point to indicate ranges of possible wind speeds. The data were produced using the Weather Research and Forecasting Model (WRF). The vertical grid used in WTK-LED includes many vertical layers in the atmospheric boundary layer to provide information of atmospheric quantities across the rotor layer of utility scale and distributed wind turbines. The WTK-LED includes: (1) Numerical simulations of wind speed and other meteorological variables covering the contiguous United States (CONUS) and Alaska, with high-resolution (5-minute [min], 2-kilometer [km]) data for 3 years (2018-2020): WTK-LED CONUS, WTK-LED Alaska. (2) Climate simulations from Argonne National Laboratory covering North America, including Alaska, Canada, and most of Mexico and the Caribbean islands. These simulations complement the new WTK-LED to offer a 4-km, hourly dataset covering 20 years (2001-2020): WTK-LED Climate. (3) Specific long-term, high-resolution offshore simulations have been conducted separately for the U.S. coasts, Hawaii, and the Great Lakes, leading to the 2023 National Offshore Wind dataset: NOW-23. The data for Hawaii include land-based data and are part of WTK-LED Hawaii. Because the accuracy of simulations from a mesoscale model, such as WRF, varies depending on the location and weather situation, and can reach up to several m/s for wind speed, we provide simulated wind speed uncertainty estimates to the community to be used in conjunction with the deterministic model simulations. This dataset was developed to satisfy a wide group of stakeholders across various wind energy disciplines, including but not limited to stakeholders in the distributed and utility scale wind industry, the new emerging airborne wind energy field, grid integration, power systems modeling, environmental modeling, and researchers in academia, and to close some of the gaps that current public datasets have. Based on our validation results to date, we suggest use cases and applications for each dataset of the WTK-LED as shown in "WTK-LED Use Cases" resource below.

The data in this repository consists of 4 files. This includes a readme file [readme.txt], a file summarizing the wind speed [All_Windspeed_Data.csv], a file for the resulting power outputs [All_Power_Data.csv],and a zip-file including detailed data for each wind farm [Data_Per_Wind_Farm.zip]. Each file can be downloaded seperatly or colectivly by clicking the "Download all"-Button.The structure of this repository is as follows:├── readme.txt (this file)├── All_Power_Data.csv (Power time series of wind farms)├── All_Windspeed_Data.csv (Windspeed time series of wind farms)├── Data_Per_Wind_Farm (folder including csv-files for each wind farm) ├── Baie_de_Saint_Brieuc ├── Baltic_Eagle ├── Beatrice ├── Borkum_Riffgrund ├── Borssele_(Phase_1,2) ├── Borssele_(Phase_3,4) ├── Dieppe_et_Le_Treport ├── Dogger_Bank_(Phase_A,B) ├── East_Anglia_One ├── Gemini ├── Gode_Wind ├── Greater_Gabbard ├── Gwynt_y_Mor ├── Hautes_Falaises ├── Hohe_See ├── Hollandse_Kust_Noord ├── Hollandse_Kust_Zuid ├── Horns_Rev ├── Hornsea_(Project_1) ├── Hornsea_(Project_2) ├── Iles_dYeu_et_de_Noirmoutir ├── Kriegers_Flak ├── London_Array ├── Moray_Firth ├── Race_Bank ├── Seagreen ├── Seamade ├── Triton_Knoll ├── WalneyIn the 29 files included in the zip-file [Data_Per_Wind_Farm.zip], we report detailed data for each wind farm. Therein, each column includs one variable while each row represents one point in time. Namely, the columns contain:- time- u-component of wind 100m above ground- v-component of wind 100m above ground- forecasted surface roughness (fsr)- scaled windspeed at hub heigts (heigt given in parentheses - multiple time series possible)- Wind direction in degrees- Power of wind turbines (type given in parentheses - multiple time series possible)- Turn_off (0: turbine turned off because of strong winds, 1: turbines active)- Power (resulting power output of wind farm over all turbine types).Starting from January 1, 1980, 00:00 am UTC in the first row, the data set ranges up to December 31, 2019, 11:00 pm in the last of 350640 rows.Similar to the detailed files per wind farm, each row in the two csv files [All_Power_Data.csv , All_Windspeed_Data.csv] reporting wind speed at hub height and total power represent one point in time for the same period.In the [All_Power_Data.csv] each row gives the sythetic resulting power outout in MW of one wind farm. I.e., the dataset includes 29 columns one for each wind farm. In the [All_Windspeed_Data.csv] each row gives the calculated windspeed im 100m above ground in m/s at the position of each wind farm. I.e., the dataset includes 29 columns one for each wind farm. Data generated using Copernicus Climate Change Service information [1980-2019] and containing modified Copernicus Climate Change Service information [1980-2019].

This dataset is published together with the paper "CARE to Compare: A real-world dataset for anomaly detection in wind turbine data" which explains the dataset in detail and defines the CARE score that can be used to evaluate anomaly detection algorithms on this dataset. When referring to this dataset, please cite the paper mentioned in the related work section.

The data consists of 95 datasets, containing 89 years of SCADA time series distributed across 36 different wind turbinesfrom the three wind farms A, B and C. The number of features depends on the wind farm; Wind farm A has 86 features, wind farm B has 257 features and wind farm C has 957 features.

The overall dataset is balanced, as 45 out the 95 datasets contain a labeled anomaly event that leads up to a turbine fault and the other 50 datasets represent normal behavior. Additionally, the quality of training data is ensured by turbine-status-based labels for each data point and further information about some of the given turbine faults are included.

The data for Wind farm A is based on data from the EDP open data platform (https://www.edp.com/en/innovation/open-data/data), and consists of 5 wind turbines of an onshore wind farm in Portugal. It contains SCADA data and information derived by a given fault logbook which defines start timestamps for specified faults. From this data 22 datasets were selected to be included in this data collection. The other two wind farms are offshore wind farms located in Germany. All three datasets were anonymized due to confidentiality reasons for the wind farms B and C.Each dataset is provided in form of a csv-file with columns defining the features and rows representing the data points of the time series. Files

More detailed information can be found in the included README-file.

Version Changes:

Version 4->5:

Corrections to labels were made:

Previously missing status_type_id 4 labels were added to datasets in Wind Farm A.

Event 51 from Wind Farm A was wrongly labeled as a normal event. With the newly added status_type_id 4 occurences, it is to be considered an anomaly event due to a gearbox bearing damage within the prediction data.

Wind Farm A no longer contains status_type_id 5. All occurences of status_type_id 5 have been changed to 0 and are considered normal time stamps. This change is done, because status_type_id 5 was set as a result of a wind speed and power analysis, flagging potential anomalous data. This is not based on a fixed ground truth, so status_type_id 5 was removed. For Wind Farms B and C status_type_id 5 is still valid since it is based on real SCADA-status codes.

The event_info.csv files now contain an additional column 'asset_id'.

Version 3->4:

The change of the timestamp anonymization lead to duplicate timestamps when transitioning from a leap year to 2022. This is now fixed in Version 4.

Version 2->3:

In version 2 timestamp changes were not consistent with the timestamps in the event-info-files. Version 3 fixes this.

Version 1->2:

Version 2 contains one deviation from version 1 regarding the anonymization procedure. Instead of shifting the timestamps of each sub-dataset by a random number of years, the size of the time shift is now determined to be the number of years so that each sub-dataset starts in 2022. This change is made to make the timestamp anonymization more consistent and to avoid future timestamps being present within the data.

Data repository for measurements from 8 wind masts in Zambia. Data transmits daily reports for wind speed, wind direction, air pressure, relative humidity and temperature. Please refer to the country project page for additional outputs and reports, including installation reports: https://www.esmap.org/re-mapping/zambia. For access to maps and GIS layers, please visit the Global Wind Atlas: http://globalwindatlas.info/ Please cite as: [Data/information/map obtained from the] “World Bank via ENERGYDATA.info, under a project funded by the Energy Sector Management Assistance Program (ESMAP).

The Global Wind Atlas version 3 data-sets contain microscale wind information at approximately 250m grid point spacing.The data is created by first dynamically down-scaling ERA5 reanalysis data from 2008-2017 to 3km resolution using the WRF mesoscale model.The WRF results are then generalized using DTU's generalization methodology, and then down-scaled using the WAsP model to the final 250m resolution.The data in this directory consist of the entire global tiff at the full 0.0025 degree resolution on the WGS84 map projection. These data also include four sets of overview pyramids to improve the viewing of the data at low resolution.Most of the data are named as follows: gwa_{variable}_{height}.tif, where variable is one of* wind-speed - The mean wind speed at the location for the 10 year period* power-density - The mean power density of the wind, which is related to the cube of the wind speed, and can provide additional information about the strength of the wind not found in the mean wind speed alone.* combined-Weibull-A and combined-Weibull-k - These are the all sector combined Weibull distribution parameters for the wind speed. They can be used to get an estimate of the wind speed and power density at a site. However, caution should be applied when using these in areas with wind speeds that come from multiple directions as the shapes of those individual distributions may be quite different than this combined distribution.* air-density - The air density is found by interpolating the air density from the CFSR reanalysis to the elevation used in the global wind atlas following the approach described in WAsP 12.* RIX - The RIX (Ruggedness IndeX) is a measure of how complex the terrain is. It provides the percent of the area within 10 km of the position that have slopes over 30-degrees. A RIX value greater than 5 suggests that you should use caution when interpreting the results.The files which do not follow the naming convention above are the capacity-factor layers. The capacity factor layers were calculated for 3 distinct wind turbines, with 100m hub height and rotor diameters of 112, 126, and 136m, which fall into three IEC Classes (IEC1, IEC2, and IEC3). Capacity factors can be used to calculate a preliminary estimate of the energy yield of a wind turbine (in the MW range), when placed at a location. This can be done by multiplying the rated power of the wind turbine by the capacity factor for the location (and the number of hours in a year): AEP = Prated*CF*8760 hr/year, where AEP is annual energy production, Prated is rated power, and CF is capacity factor.

The Canadian Wind Turbine Database contains the geographic location and key technology details for wind turbines installed in Canada. This dataset was jointly compiled by researchers at CanmetENERGY-Ottawa and by the Centre for Applied Business Research in Energy and the Environment at the University of Alberta, under contract from Natural Resources Canada. Additional contributions were made by the Department of Civil & Mineral Engineering at the University of Toronto. Note that total project capacity was sourced from publicly available information, and may not match the sum of individual turbine rated capacity due to de-rating and other factors. The turbine numbering scheme adopted for this database is not intended to match the developer’s asset numbering. This database will be updated in the future. If you are aware of any errors, and would like to provide additional information, or for general inquiries, please use the contact email address listed on this page.

This is the data set behind the Wind Generation Interactive Query Tool created by the CEC. The visualization tool interactively displays wind generation over different time intervals in three-dimensional space. The viewer can look across the state to understand generation patterns of regions with concentrations of wind power plants. The tool aids in understanding high and low periods of generation. Operation of the electric grid requires that generation and demand are balanced in each period.

which corresponding to an hours

Within the project WEA-Acceptance¹, extensive measurement campaigns were carried out, which included the recording of acoustic, meteorological and turbine-specific data. Acoustic quantities were measured at several distances to the wind turbine and under various atmospheric and turbine conditions. In the project WEA-Acceptance-Data², the acquired measurements are stored in a structured and anonymized form and provided for research purposes. Besides the data and its documentation, first evaluations as well as reference data sets for chosen scenarios are published.

In this version of the data platform, a specification 2.0, an anonymized data set and three use cases are published. The specification contains the concept of the data platform, which is primarily based on the FAIR (Findable, Accessible, Interoperable, and Reusable) principle. The data set consists of turbine-specific, meteorological and acoustic data recorded over one month. Herein, the data were corrected, conditioned and anonymized so that relevant outliers are marked and erroneous data are removed in the data set. The acoustic data includes anonymized sound pressure levels and one-third octave spectra averaged over ten minutes as well as audio data. In addition, the metadata and an overview of data availability are uploaded. As examples for the application of the data, three use cases are also published. Important information such as the approach for data anonymization is briefly described in the ReadMe file.

For further information about the measurements, it is referred to "Martens, S., Bohne, T., and Rolfes, R.: An evaluation method for extensive wind turbine sound measurement data and its application, Proceedings of Meetings on Acoustics, Acoustical Society of America, 41, 040001, https://doi.org/10.1121/2.0001326, 2020.

¹The project WEA-Acceptance (FKZ 0324134A) was funded by the German Federal Ministry for Economic Affairs and Energy (BMWi).

²The project WEA-Acceptance-Data (FKZ 03EE3062) was funded by the German Federal Ministry for Economic Affairs and Energy (BMWi).

The UK mean wind data contain the mean wind speed and direction, and the direction, speed and time of the maximum gust, all during 1 or more hours, ending at the stated time and date. The data were collected by observation stations operated by the Met Office across the UK and transmitted within the following message types: SYNOP, HCM, AWSHRLY, DLY3208, HWNDAUTO and HWND6910. The data spans from 1949 to 2023.

This version supersedes the previous version of this dataset and a change log is available in the archive, and in the linked documentation for this record, detailing the differences between this version and the previous version. The change logs detail new, replaced and removed data. These include the addition of data for calendar year 2023.

For further details on observing practice, including measurement accuracies for the message types, see relevant sections of the MIDAS User Guide linked from this record (e.g. section 3.3 details the wind network in the UK, section 5.5 covers wind measurements in general and section 4 details message type information).

This dataset is part of the Midas-open dataset collection made available by the Met Office under the UK Open Government Licence, containing only UK mainland land surface observations owned or operated by the Met Office. It is a subset of the fuller, restricted Met Office Integrated Data Archive System (MIDAS) Land and Marine Surface Stations dataset, also available through the Centre for Environmental Data Analysis - see the related dataset section on this record.

HRD wind analysis requires input of all available surface weather observations (e.g., ships, buoys, coastal platforms, surface aviation reports, reconnaissance aircraft data adjusted to surface, etc.). These data are downloaded on a regular schedule and processed to fit the analysis framework. This framwork is based on height (10 m), exposure (marine or open terrain over land, and averaging period (maximum sustained 1 minute wind speed). This framework is consistent with that used by the National Hurricane Center (NHC). This dataset contains data for 2005 hurricanes Dennis, Emily, Gert, Katrina, Ophelia, Rita and Wilma.

This dataset provides locations and technical specifications of wind turbines in the United States, almost all of which are utility-scale. Utility-scale turbines are ones that generate power and feed it into the grid, supplying a utility with energy. They are usually much larger than turbines that would feed a house or business. The regularly updated database contains wind turbine records that have been collected, digitized, and locationally verified. Turbine data were gathered from the Federal Aviation Administration's (FAA) Digital Obstacle File (DOF) and Obstruction Evaluation Airport Airspace Analysis (OE-AAA), American Clean Power (ACP) Association (formerly American Wind Energy Association (AWEA)), Lawrence Berkeley National Laboratory (LBNL), and the United States Geological Survey (USGS), and were merged and collapsed into a single dataset. Verification of the turbine positions was done by visual interpretation using high-resolution aerial imagery in ESRI ArcGIS Desktop. A locational error of plus or minus 10 meters for turbine locations was tolerated. Technical specifications for turbines were assigned based on the wind turbine make and models as provided by manufacturers and project developers directly, and via FAA datasets, information on the wind project developer or turbine manufacturer websites, or other online sources. Some facility and turbine information on make and model did not exist or was difficult to obtain. Thus, uncertainty may exist for certain turbine specifications. Similarly, some turbines were not yet built, not built at all, or for other reasons cannot be verified visually. Location and turbine specifications data quality are rated, and confidence is recorded for both. None of the data are field verified.

This data (csv file) provides simulated hourly time series of existing offshore wind generation for the regions shown in the attached map. Only regions with existing (by the time of modeling) offshore wind power plants are simulated (otherwise the data are NaN). The map shows the resulting capacity factors (annual mean). Wake losses are modeled, with additional 5 % of other losses and unavailability considered. The time stamps are in GMT; the variable (column) names relate to the region names shown in the maps. The data include also country-level aggregations, e.g., UK00_OFF is the aggregated offshore wind generation of all the UK regions (weighted by regional installed capacities). The data are part of the variable renewable energy generation time series created for ENTSO-E in the 2021 update of the Pan-European Climate Database (PECD) dataset. ENTSO-E has used the data in ERAA 2021 and Winter Outlook 2021-2022 assessments, and they are used in TYNDP 2022. The simulations are carried out by DTU Wind Energy, with the future technology selection and data validation discussed and agreed with ENTSO-E and its members. The linked journal paper (1st link) describes the ERA5-based simulation methodology. It is requested that the paper is cited when the data are used. The linked related journal paper (2nd link) describes the modeling of wake losses and storm shutdown behaviour for the offshore wind power plants. This item is part of a larger collection of wind and solar data: https://doi.org/10.11583/DTU.c.5939581