Development of a Digital Twin Framework for Condition Monitoring and Performance Assessment of a Wind Turbine

The paper presents a comprehensive quasi-static digital twin of a utility-scale wind turbine developed in Python. The proposed digital twin framework combines dynamically acquired turbine measurements with static system parameters to reconstruct the steady-state operational behaviour of the wind turbine under real-world conditions. The digital twin integrates historical and real-time operational data to provide dynamic visualization of key performance indicators, enabling continuous insight into turbine behaviour and operational conditions. The model is trained using comprehensive SCADA datasets, including turbine operational data, grid measurements, meteorological observations, and alarm logs, with target variables encompassing the turbine’s active power output and efficiency-related performance metrics. Through systematic data processing, filtering, and aggregation, the digital twin provides a representation of expected turbine behaviour across a wide range of operating conditions. The developed digital twin supports performance benchmarking, operational condition assessment, and early detection of deviations that may indicate aerodynamic inefficiencies, control limitations, or emerging component degradation. Its quasi-static formulation ensures computational efficiency and numerical robustness, making it suitable for real time deployment within wind farm monitoring environments. The presented approach is fully transparent, reproducible, and modular, allowing straightforward extension toward advanced digital twin functionalities, including condition-based maintenance, performance degradation analysis, and integration with forecasting or control-oriented applications. Key words. Digital twin, Wind turbine, Power production, Prediction