Integrating Oil Debris and Vibration Gear Damage Detection Technologies Using Fuzzy Logic

A diagnostic tool for detecting damage to spur gears was developed. Two different measurement technologies, wear debris analysis and vibration, were integrated into a health monitoring system for detecting surface fatigue pitting damage on gears. This integrated system showed improved detection and decision-making capabilities as compared to using individual measurement technologies. This diagnostic tool was developed and evaluated experimentally by collecting vibration and oil debris data from fatigue tests performed in the NASA Glenn Spur Gear Fatigue Test Rig. Experimental data were collected during experiments performed in this test rig with and without pitting. Results show combining the two measurement technologies improves the detection of pitting damage on spur gears. Introduction One technology area recommended for helicopter accident reduction is the design of helicopter Health Usage Monitoring Systems (HUMS) capable of predicting imminent equipment failure for on-condition maintenance and a more advanced system capable of warning pilots of impending equipment failure. Today’s helicopter health monitoring systems (HUMS) are not at this level. Data collected by HUMS is often processed after the flight and plagued with high false alarm rates and undetected faults. The current fault detection rate of commercially available HUMS through vibration analysis is 70 percent [1]. False warning rates average 1 per hundred flight hours [2]. Often these systems are complex and require extensive interpretation by trained diagnosticians [3]. Transmission diagnostics are an important part of a helicopter HUMS because helicopters depend on the power train for propulsion, lift, and flight maneuvering. In order to predict transmission failures, the diagnostic tools used in the health monitoring system must provide real-time performance monitoring of aircraft operating parameters and must demonstrate a high level of reliability to minimize false alarms. Various techniques exist for diagnosing damage in helicopter transmissions. The method most widely used involves vibration. Algorithms are developed, using vibration data collected from gearbox accelerometers, to detect when gear damage has occurred. Oil debris is also used to identify abnormal wear related conditions at an early stage. Oil debris monitoring for gearboxes consists mainly of off-line oil analysis, or plug type chip detectors. Although not commonly used for gear damage detection, many engines have on-line oil debris sensors for detecting the failure of rolling element bearings. These on-line, inductance type sensors count the number of particles, their approximate size, then calculate an accumulated mass. Integrating the sensors into one system can potentially improve the detection capabilities and the probability that damage is detected. Recent investigations have shown the benefits of using an oil debris monitor with vibration data to improve current HUMS, but have not fully demonstrated a system with improved detection and decisionmaking capability when integrating the two measurement systems [4], [5]. The objective of the work reported herein is to improve the detection capability of vibration and oil based damage detection features by applying fuzzy logic analysis techniques to gear failure data collected from the NASA Glenn Spur Gear Fatigue Rig. A simple model was defined by the fuzzy rules and the membership functions for the experiments when pitting damage occurred. The ability to define valid ranges and limits for each membership function was found to be critical to the success of the model at predicting damage. Vibration data were collected from accelerometers and used in previously validated gear vibration diagnostic algorithms. Oil debris data were collected using a commercially available in-line oil debris sensor. Oil debris and vibration data will be integrated using fuzzy logic analysis techniques. The goal of this research is to provide the end user with a simple tool to determine reliably the health of this geared system.