A Comparative Study of Three Fault Diagnosis Schemes for Wind Turbines

In wind turbine systems, early diagnosis and accommodation of faults are crucial for the reliable and cost effective operation of wind turbines and their success as viable renewable energy conversion solutions. This paper proposes and compares three different diagnostic schemes that address the issue of fault detection and isolation for the drivetrain and generator-converter subsystems of a wind turbine. The first diagnostic scheme is based on a cascade of two Kalman filters intended to alleviate the effect of the nonlinear aerodynamic torque generation in the drivetrain dynamics. The second scheme uses a bank of dedicated observers, each of which exploits Thau´s argument for systems featuring nonlinear static feedback. The third scheme is a secondary H_∞ filtering mechanism constructed from parity equations by treating the nonlinearity as bounded uncertainty. The performance of each scheme is demonstrated using simulations of the wind turbine system. Robustness of the schemes has been analyzed in terms of parametric uncertainties and different operating conditions. A detailed comparison is also presented pointing to the positive and negative aspects of each scheme.