liu, yang (China University of Petroleum (East China)), Sheng, Li (China University of Petroleum (East China)), Li, Chunyu (China University of Petroleum (East China)), XI, XIAOPENG (Universidad Técnica Federico Santa María), Orchard, Marcos (Faculty of Physical and Mathematical Sciences, Universidad de Ch)

PhysGAIN: A Physical-Constraint-Based Imputation Model for Missing SCADA Data in Wind Turbines

Scheduled for presentation during the Invited Session "Advanced Prognostics and Health Management of Mechatronic Systems: Methods and Applications" (WeAT1), Wednesday, July 16, 2025, 11:40−12:00, Room 105

Joint 10th IFAC Symposium on Mechatronic Systems and 14th Symposium on Robotics, July 15-18, 2025, Paris, France

This information is tentative and subject to change. Compiled on July 16, 2025

Abstract

SCADA data from wind turbines is crucial for the monitoring and optimization of wind power systems; however, missing data is a significant challenge. While generative models, particularly the GAIN model, have been shown to yield promising results in imputing missing data, the physical relationships inherent in SCADA data are generally overlooked. To address this limitation, a novel data imputation model, PhysGAIN, is proposed. The physical characteristics of wind turbines, such as the power curve, are integrated into the GAIN framework by incorporating a physical loss function. This integration ensures that the imputed data is consistent with the operational constraints of wind turbine systems. Experimental evaluations on a SCADA dataset from wind turbines demonstrate that PhysGAIN outperforms the baseline GAIN model in terms of imputation accuracy, physical consistency, and convergence speed. PhysGAIN provides a robust solution for missing data imputation in energy systems, with significant potential for broader applications in fields that demand physical consistency.