Hybrid Deep Learning Models With Attention Mechanisms For Short-Term Wind Speed Forecasting: A Comprehensive Review And Benchmarking Of CNN–Bilstm–Attention Using Synthetic SCADA-Style Data For Robust Preliminary Validation. International Journal of Multidisciplinary Engineering In Current Research, [S. l.], v. 11, n. 5, p. 1–7, 2026. DOI: 10.63665/IJMEC.1105.01. Disponível em: https://ijmec.com/index.php/multidisciplinary/article/view/1075. Acesso em: 23 may. 2026.