New method for condition monitoring of low-speed bearings using peak frequency of acoustic emission hits

The continuous advancement of wind power technology highlights the critical importance of condition monitoring (CM) for maintaining operational performance. Among the vital components in wind turbine (WT) systems, main shaft bearings have emerged as the most critical, especially with the ongoing trend towards larger and heavier rotors. This paper introduces a new approach to monitoring the condition of the main shaft bearings in a WT with a two-point suspension drivetrain, using acoustic emission (AE). The measurements take place in a real-life setting where AE signals are detected, transmitted to an AE analyser and translated to a PC for initial analysis and storage. This system is fixed securely inside the WT nacelle, while wirelessly connected to the internet, and is thus accessible at any time from any device. Further analysis of the recorded AE signals takes place in the laboratory, where the analysis focuses on the frequency domain of the AE hits. The Welch method is used for transforming the signal to the frequency domain because of its superior spectral accuracy and computational efficiency. It enables fast identification of the peak frequency (PF) for each AE hit and these frequencies are mapped over specific time segments, generating a 'PF distribution plot'. These plots are combined into an overlay for a comprehensive assessment of the bearing condition. Results reveal distinct PF bands in the distribution plots, correlating with bearing condition. The proposed approach offers simplicity and the potential for automation, facilitating seamless integration into a future predictive maintenance plan for WTs.