Using acoustic emission for condition monitoring of the main shaft bearings in 4-point suspension wind turbine drivetrains

Abstract

The continuous growth of wind power technology makes condition monitoring of wind turbine components crucially important for their operational efficiency. The main shaft bearings in wind turbines have been identified as one of the most critical components in the system, especially with the ongoing increase in rotor size and weight. This increase made the 4-point suspension drivetrain more preferable. In this study, we present a novel approach for condition monitoring of the main shaft bearings in a 2 Megawatt wind turbine with 4-point suspension drivetrain using primarily acoustic emission (AE). The focus was on the analysis of time and frequency domains of the AE signal, where the dominant frequency of each AE hit was identified and plotted back in the time domain to create the so-called dominant frequency map in specific time intervals for each bearing. A comparison between the two dominant frequency maps of the two bearings gives valuable insights into the condition of the two bearings. The distinctive nature of the dominant frequency bands in the dominant frequency maps presented promising potential for this method. The presented method is straightforward and can be automated and then integrated into a planned predictive maintenance programme for this wind turbine.The continuous growth of wind power technology makes condition monitoring of wind turbine components crucially important for their operational efficiency. The main shaft bearings in wind turbines have been identified as one of the most critical components in the system, especially with the ongoing increase in rotor size and weight. This increase made the 4-point suspension drivetrain more preferable. In this study, we present a novel approach for condition monitoring of the main shaft bearings in a 2 Megawatt wind turbine with 4-point suspension drivetrain using primarily acoustic emission (AE). The focus was on the analysis of time and frequency domains of the AE signal, where the dominant frequency of each AE hit was identified and plotted back in the time domain to create the so-called dominant frequency map in specific time intervals for each bearing. A comparison between the two dominant frequency maps of the two bearings gives valuable insights into the condition of the two bearings. The distinctive nature of the dominant frequency bands in the dominant frequency maps presented promising potential for this method. The presented method is straightforward and can be automated and then integrated into a planned predictive maintenance programme for this wind turbine.