Bearing temperature monitoring of a Wind Turbine using physics-based model

The purpose of this paper is to propose a method to monitor a Wind Turbine’s (WT) main bearing, based on the difference between the temperature as measured by the Supervisory Control and Data Acquisition system (SCADA).

The monitoring of the main bearing is based on the difference between the measured temperature and the estimated temperature obtained from a dynamic model. The model used is based on the law of energy conservation. Several validation metrics have suggested that this model is accurate.

The Exponentially Weighted Moving Average control chart for two cases studies is used for the monitoring for the main bearing; this method has shown great potential for industrial applications. A failure was detected three weeks before the current actual alarm settings used by SCADA were able to identify the issue.

The proposed method is a monitoring method that can be used on most industrial wind farms and provide important information on the condition of the WTs’ main bearing.