Review of transformers applied to wind turbine data: Exploring the potential of transformers in revolutionizing wind turbine data analysis

Population growth in recent decades has significantly increased global electricity demand, underscoring the need for renewable energy sources to diversify the energy mix and mitigate environmental impacts. Wind energy has experienced exponential growth, but this expansion poses major challenges, particularly in accurately forecasting wind power potential and detecting anomalies in critical components to prevent unexpected outages. Recently, Artificial Intelligence techniques—especially deep learning models based on Transformer architectures—have emerged as promising tools to address these issues. This study presents the first systematic review of Transformer-based models for wind turbine data analysis, offering direct comparisons with established benchmark models. A combined quantitative and qualitative approach was adopted. A bibliometric analysis identified research trends, while an in-depth content analysis examined model architectures, datasets, and evaluation procedures. The search resulted in 9994 articles; after applying inclusion and exclusion criteria, 37 were analyzed in detail. The results reveal increasing adoption of Transformer-based models across three main applications: (I) wind speed forecasting, (II) wind power forecasting, and (III) anomaly detection. In these domains, Transformer models and their derivatives consistently outperform benchmark models, demonstrating superior accuracy, robustness, and the ability to model long-term dependencies. Transformer-based models offer substantial benefits for wind farm operations, enabling improved generation forecasting, proactive maintenance, and reduced downtime. As Transformer architectures become increasingly specialized and efficient for wind energy applications, alongside richer datasets and enhanced hardware, these models are becoming increasingly viable for real-world deployment, representing a transformative advance in wind energy analytics.