Machine Learning-Based Short-Term Forecasting of Wind Speed, Turbulence Intensity, and Sea Breeze for Offshore Wind Farm Applications

Clicks: 25
ID: 328019
2026
Free PDF
Article Quality & Performance Metrics
Overall Quality
Not rated
Combines reader engagement with the AI quality analysis. This article has not been analysed, so there is no overall score — reader engagement is measured and shown alongside.
AI Quality Assessment
Not analyzed
Readership in this journal
Emerging

Ranked #34 of 45 articles by views in journal of modern power systems and clean energy

Most read Least read

Bar heights use a square-root scale.

Mint this article as an NFT
Not yet minted

Create a permanent, verifiable on-chain record of this article on the Scimatic Network. The NFT is held in your Journament account, and you can withdraw it to your own wallet at any time.

5 SUSD one-off · no wallet required
Sign in to mint
Abstract
Abstract Accurate estimation of wind energy resources significantly improves wind farm performance. However, predicting wind parameters such as wind speed, turbulence intensity, and sea breezes is complicated by interactions among these factors. The suggested methodology involves a multi-parametric workflow-based machine learning approach whereby independent models will be constructed for wind speed forecast, turbulence intensity forecast, and sea breeze classification by utilizing a shared meteorological data set obtained from a light detection and ranging sensor. The output of these task-based models is combined under one analytical process for examining offshore wind properties and quantifying the effects of sea breeze on power production by wind turbines. In wind speed prediction, random forest performed best, with root mean square error = 0.172 m/s and R2 = 0.998. For predicting turbulence intensity, long short-term memory was more efficient than the other models, with root mean square error = 0.0196, mean absolute error = 0.0139, and R2 = 0.922. For sea breeze classification, extreme gradient boosting was more successful, with area under curve = 0.93, precision = 0.87, recall = 0.98, and F1-score = 0.92. There is also evidence of decreasing forecast accuracy with longer prediction periods. In particular, root mean square error values for wind speed forecasts were 0.172 m/s and 1.249 m/s, while root mean square error values for turbulence intensity were 0.0196 and 0.0575 as the forecasting period increased from 10 to 60 min. In addition, it is noted that the power generation is 14% higher during sea-breeze in the National Renewable Energy Laboratory's offshore wind turbines with rated power of 8-18 MW. Overall, the proposed approach will be beneficial for short-term forecasts, wind turbine loads, and power production.
Reference Key
openalex_W7211902287 Use this key to autocite in the manuscript while using SciMatic Manuscript Manager or Thesis Manager
Authors L Selvarajan, Krishnamoorthy Amarnath, S. Senthil Kumar, N Satish, A Palanisamy, S Umamaheswari, N Kasthuri
Journal journal of modern power systems and clean energy
Year 2026
DOI
10.1093/ce/zkag059
URL
https://doi.org/10.1093/ce/zkag059
Keywords Keywords not found

Citations

No citations found. To add a citation, contact the admin at info@scimatic.org

Comments

No comments yet. Be the first to comment on this article.