Vibration-based fault diagnostics of planetary gearbox using time synchronous averaging with multiple window functions

Abstract

Recently, numerous number of studies have been made to advance TSA for vibration-based diagnostics of planet gears of a planetary gearbox. To increase signal-to-noise ratio of the vibration signals, various narrow-range window functions centered on sensor's position have been developed to capture the instances when the planet gears are adjacent to the sensor. However, TSA with such narrow-range window functions is unable to detect the abnormal signal if it is amplified at outside of the sensor's position due to an unexpected vibration modulation characteristics of the gearbox. This paper proposes a TSA which is robust toward the unexpected vibration modulation characteristics of the gearbox. Multiple narrow-range window functions were employed to perform multiple TSAs rather than employing the sole window function centered on the sensor's position. Condition indicators with regard to every ring gear's teeth were derived, and accumulated for the purpose of condition monitoring. Then, optimal position of the window function was determined to maximize capability to detect signals from the faulty gear. For demonstration of the proposed TSA, test was performed with a 2kW testbed having one-stage planetary gearbox within which a planet gear with artificial fault was assembled.