Abstract
This paper presents a data driven case study of two outer-race spalling faults in double-row spherical roller main-bearings, with the purpose of identifying key features and relevant measurements associated with this failure mode in wind turbine main-bearings. Supervisory data is analysed for one fault case and vibration data for the other. The aim of this work is to inform practitioners and assist in improving fault detection systems for this component.
Highlights
Wind energy is playing a key role in the decarbonisation of the power sector
This paper presents a data driven case study of two outer-race spalling faults in double-row spherical roller main-bearings, with the purpose of identifying key features and relevant measurements associated with this failure mode in wind turbine main-bearings
Some subtleties associated with the data were investigated, with findings indicating that thermal inertia effects should be considered when developing fault detection models
Summary
Wind energy is playing a key role in the decarbonisation of the power sector. In 2019 electricity production from wind power totaled 417 Terra Watt Hours, accounting for 15% of EU electricity consumption [1]. Existing work which does consider fault detection and prognosis for main-bearings is often based on simulated and idealised data [8, 9, 10] or, in some cases, is based on scaled rotors whose inertias are orders of magnitude less than that of operational wind turbines [11]. Even in studies reporting promising main-bearing fault detection performance [13, 14], it is not clear whether employed data was ascertained to be the most appropriate for the task at hand, or what was available.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
