Fault detection enhancement in wind turbine planetary gearbox via stationary vibration waveform data

Abstract

Planetary gearbox fault diagnosis is very important for reducing the downtime, maintenance cost, and for improving the safety, reliability, and lifespan of wind turbines. The present work reports the results concluded by long-term experiments to a defected planetary gearbox system, with a transverse cut with a depth of 1.0 mm and thickness of 0.2 mm to simulate the planetary gearbox component crack. For each defect, recordings every 60.0 min were acquired and a total of 7 recordings (∼ 6.0 h of test duration) were resulted until the termination of the test. Fault is assured by increasing the test period to the point of where the remaining metal in the tooth area has enough stress to be in the plastic deformation region. An experimental procedure is developed to assess the severity of the gearbox component fault. Gearbox components faults of cracked planet gear tooth, cracked planet gears carrier, and cracked main bearing inner race were tested under accelerated fault conditions, where a comparative analysis of condition monitoring indicators for various crack detection has been done. The experimental localized fault signals (vibration acceleration signals) were subjected to the same diagnostic techniques such as spectrum comparisons, spectral kurtosis analysis, skewness analysis, and crest factor analysis. The method is validated on a set of seeded localized faults on all gears and components: sun gear, ring gear, etc. The results look promising, where the root mean square value analysis could be a good indicator when compared with the other indicators in terms of early detection and characterization of faults.