Wind turbine generator bearing Condition Monitoring with NEST method

Condition Monitoring (CM) can greatly reduce the maintenance cost for a wind turbine. In this paper, history data of Supervisory Control and Data Acquisition (SCADA) system is analyzed to detect the incipient failure of wind turbine generator bearing. A new condition monitoring method based on the Nonlinear State Estimate Technique (NSET) is proposed. NSET is used to construct the normal behavior model of the generator bearing temperature. Detail of NSET is introduced. When the generator bearing has an incipient failure, the residuals between NSET model estimates and the measured generator bearing temperature will become significant. When the residual exceeds the predefined thresholds, an incipient failure is flagged. Analysis of a manual drift added on the historical SCADA data for a wind turbine generator bearing demonstrates the effectiveness of this new condition monitoring method.