Vibration based fault diagnostics in a wind turbine planetary gearbox using machine learning

Abstract

To reduce wind turbine operations and maintenance costs, we present a machine learning framework for early damage detection in gearboxes based on the cyclostationary and kurtogram analysis of sensor data. The application focus is fault diagnostics in gearboxes under varying load conditions, particularly turbulent wind. Faults in the gearbox rotating components can leave their signatures in vibrations signals measured by accelerometers. We analyze data stemming from a simulated vibration response of a 5 MW multibody wind turbine model in a healthy and damaged scenarios and under different wind conditions. With cyclostationary and kurtogram analysis applied on acquired sensor data, we generate two types of 2D maps that highlight signatures related to the fault damage. Using these maps, convolutional neural networks are trained to identify faults, including those of small magnitude, in test data with a high accuracy. Benchmark test cases inspired by an NREL study are tested and faults successfully detected.