TY  - JOUR
AU  - Hassanian, Reza
AU  - Shahinfar, Abdollah
AU  - Helgadóttir, Ásdís
AU  - Riedel, Morris
TI  - Optimizing Wind Energy Production: Leveraging Deep Learning Models Informed with On-Site Data and Assessing Scalability through HPC
JO  - Acta polytechnica hungarica
VL  - 21
IS  - 9
SN  - 1785-8860
CY  - Budapest
M1  - FZJ-2025-00812
SP  - 45 - 56
PY  - 2024
AB  - This study suggests employing a deep learning model trained on on-site windspeed measurements to enhance predictions for future wind speeds. The model uses a gatedrecurrent unit (GRU) derived from the long short-term memory (LSTM) variant, and istrained using actual measured wind velocity data collected at both 10-minute and hourlyintervals. The approach relies on using same-season data for predicting wind velocity,necessitating regular updates to the model with recent measurements to ensure accuratepredictions in a timely manner.The results from the prediction model, particularly at a 10-minute interval, demonstrate asignificant alignment with the actual data during validation. Comparative analysis of theemployed model over a two-year span, with a 24-year distinction, indicates its efficiencyacross different time periods and seasonal conditions, contingent upon frequent updateswith recent on-site wind velocity data.Given the reliance of sequential deep learning models on extensive data for enhancedaccuracy, this study emphasizes the importance of employing high-performance computing(HPC). As a recommendation, the study proposes equipping the wind farm or wind farmcluster with an HPC machine powered by the wind farm itself, thereby transforming it intoa sustainable green energy resource for the HPC application. The recommended approachin this work is enforcing the smart power grid to respond to the power demand that isconnected to predictable wind farm production.
LB  - PUB:(DE-HGF)16
DO  - DOI:10.12700/APH.21.9.2024.9.4
UR  - https://juser.fz-juelich.de/record/1037647
ER  -