Posters

Abstract

This study will provide important insight for stakeholders who want to invest in the wind energy in Türkiye or for consultants conducting feasibility studies for wind farm projects by quantifying icing events and their occurrences in the regions of interest, which will help in identifying potential risks to turbine operations and optimizing the planning process for wind energy projects. In this study, the anemometers icing events occurrences are analysed and extracted for 100 meteorological masts located in different regions in Türkiye. Consequently, a statistical regression technique is applied between the icing events (icing days/year) with snow depth EMD WRF+ ERA5 data in order to generate a relation between both data and generalise it for whole regions of Türkiye averaged to a 20 years period snow depth data (2003-2023). Furthermore, in order to ensure the robustness of the relation between different regions and different sets, a K-fold cross-validation is applied.  The occurrence where snow depth data are more than zero are taken and multiplied with the average of these available occurrences to produce (Snow depth/year), and after filtering out anomaly data of the masts studied which are due to the usage of heated anemometers, strong polynomial correlation with an R² value of 0.95 was found. In order to assess the model’s predictive accuracy, the equation generated is applied to predict the (icing days/year) from (snow depth/year) to be able to compare it with the actual data of the icing events using RMSE (11.7), and MAE (2.6). Furthermore, regions with moderate icing events showed a strong linear relationship with snow depth data, with a Pearson coefficient of (0.95). Previous studies that found a link between icing events detected by anemometers and turbine icing losses in wind turbines. These studies also found a relationship between snow depth and elevation. Using this information, snow depth of 3 km resolution is spatially interpolated to a 50 m resolution and converted into a map showing turbine icing losses.1,2,3,4. The results emphasize the strong relationship between icing events and snow depth, and as the snow depth data will be average for 20 years period, this will enable more accurate decisions for the development of wind energy projects in icing-prone regions. Additionally, a factor was derived from this polynomial relationship to directly translate snow depth into anemometer and turbine icing occurrences, providing a practical tool for estimating icing risk from basic snow depth predictions.   References [1] Ribeiro, C., & Beckford, T. (2016). Icing losses: What can we learn from production and meteorological data? DNV GL Energy. [2] Switanek, M., Resch, G., Gobiet, A., Günther, D., Marty, C., & Schöner, W. (2024). Snow depth sensitivity to mean temperature, precipitation, and elevation in the Austrian and Swiss Alps. The Cryosphere, 18(1. [3] Grünewald, T., Bühler, Y., & Lehning, M. (2014). Elevation dependency of mountain snow depth. The Cryosphere, 8(1), 2381–2394. [4] Davis, N., Hahmann, A. N., Clausen, N.-E., & Zagar, M. (2014). Modeling Power Loss Due to Wind Turbine Icing.