Applying thermophysics for wind turbine drivetrain fault diagnosis using SCADA data

Abstract

Cost-effective wind turbine (WT) diagnosis using supervisory control and data acquisition (SCADA) data is a promising technology for future intelligent wind farm operation and management. This study presents a thermophysics-based method for WT drivetrain fault diagnosis. A synthesised thermal model is formed by incorporating thermal mechanisms of the drivetrain into a WT system model. Applications of the model are demonstrated in case studies of the gearbox and generator comparing simulation results and SCADA data analysis. The results show non-linearity of the gearbox oil temperature rise with wind speed/output power that can effectively indicate gearbox efficiency degradation, which may be attributed to gear transmission problems such as gear teeth wear. Electrical generator faults such as ventilation failure and winding voltage unbalance will cause changes to heat transfer and result in temperature changes that can be used for diagnosis purposes. This is shown by different patterns of stator winding temperature associated with power generation, while the simulation reveals the thermal mechanism. The method can be applied to diagnose some failure modes which are hard to identify from vibration analysis. The developed thermal model can play a central role for the purpose of fault diagnosis, by deriving relationships between various SCADA signals and revealing changes in the thermophysics of WT operation.