Dynamic behavior analysis method for the normal and faulty drivetrain of wind turbine under multi‐working conditions

Abstract

Due to the random fluctuation of wind energy, the working conditions of the drivetrain have complex time‐varying characteristics. To deeply understand time‐varying dynamic behavior, a dynamic behavior method for normal and faulty drivetrain under multiple operating status is proposed. First, the working condition is divided into multiple working conditions by combining the control principle of the wind turbine with the K‐means clustering algorithm. Second, the torsional dynamics model of the drivetrain is established, solved, and verified. Finally, the dynamic behavior of normal and faulty drivetrain under multiple working conditions is analyzed. The dynamic behavior of drivetrain in normal and faulty states under different working conditions can be determined by the method proposed in this paper, which is affected by the system itself, external load, and the control strategy of the wind turbine. At the same time, the fault‐sensitive working condition range and the fault feature index under each working condition are determined. This research can provide an important theoretical basis for variable condition fault diagnosis.