An Evaluation of Empirical Approach for Gearbox Diagnosis in the Construction Equipment

Abstract

A planetary gearbox is one of the most important components in rotating machinery. In construction equipment, it is mainly used for swing and traveling devices due to huge torque capability and good torque transmission. Unexpected failure of the planetary gearbox can cause increased idle time, unnecessary cost due to inability of the construction site and safety hazards such as an operator isolation in a remote area. For these reasons, the need for timely prediction of unexpected failure becomes increasingly important in the field of construction machinery. In general, studies on fault diagnosis and condition monitoring of planetary gearboxes have been carried out in fields such as aircraft, wind-turbines and power plants. Researches in the above-mentioned fields usually require high performance computing power, and the burden of cost for diagnosis is relatively small. In addition, construction machinery also faces difficulties due to various uncertainties such as uncertain operating conditions that affect the vibration characteristics of gearboxes.

This study focuses on an approach to the vibration-based fault diagnosis methodology that can distinguish gear faults in a planetary gearbox in an excavator using vehicle-based test. First, we analyze the types of gear faults and the parts where failures occur mainly through the database of field failures. From this result, several fault types to be used in the experiment are selected, and an arbitrary fault is applied to gears. Secondly, since the vibration data is acquired directly from the excavator, the signal processing method that can remove the noise as much as possible and distinguish the fault is selected. Finally, optimized features are selected to minimize the uncertainty impacts that cannot be eliminated or unknown. Through this study, we confirmed the effect of the signal processing method which can be used in the planetary gearbox of the excavator as follows: 1) Several kinds of fault can be distinguished. 2) Faults and methodologies that can be distinguished in the constant speed condition 3) Faults and methodologies that can be distinguished in the transition speed condition.