Background

Helicopter usage and monitoring system (HUMS) is one of the critical systems for helicopter’s safety and reliability. Whilst HUMS has proven to be effective in detecting gears’ defects, bearing failures are not adequately detected using current monitoring indicators. Detection of bearing faults in helicopter gearboxes is made challenging by the presence of the complicated signal transmission path attenuating the monitored signal to the receiving sensor.

Purpose

To ensure safe operation of helicopters, this research proposes a novel condition indicator to detect bearing faults in helicopter gearboxes.

Methods

For this purpose, vibration measurements captured from a CS-29 category ‘A’ helicopters test rig were utilized for detecting a bearing defect that has occurred in the epicyclic module of the main gearbox. Signals of rolling element bearings under various fault conditions were collected, and an adaptive filter algorithm was utilized to separate the random component of the signal. The resultant signatures were then further processed using wavelet analysis to extract the bearing signal of interest.

Results

Results showed that this new indicator successfully detect bearing faults. Besides, the impulse energy indicator responds consistently to the fault severity compared to the traditional indicators such as RMS and kurtosis.

Conclusion

A technique to extract frequency band corresponding to the bearing fault impulses has been developed and tested. The technique employs the adaptive filter signal separation, wavelet packet decomposition and the combination of RMS and kurtosis to select the optimum filter band.

中文翻译：

---

用于直升机传动系统轴承故障检测的新型状态指示器

背景

直升机使用和监视系统（HUMS）是提高直升机安全性和可靠性的关键系统之一。虽然HUMS已被证明可以有效地检测齿轮的缺陷，但使用当前的监控指示器并不能充分检测出轴承故障。由于复杂的信号传输路径会衰减监测到接收传感器的信号，因此在直升机齿轮箱中检测轴承故障变得非常困难。

目的

为了确保直升机的安全运行，本研究提出了一种新颖的状态指示器来检测直升机齿轮箱中的轴承故障。

方法

为此，从CS-29类别“ A”型直升机测试台获得的振动测量结果被用于检测主变速箱的周转组件中出现的轴承缺陷。收集各种故障条件下滚动轴承的信号，并采用自适应滤波算法分离信号的随机分量。然后，使用小波分析对所得签名进行进一步处理，以提取感兴趣的方位信号。

结果

结果表明，该新指示器可以成功检测轴承故障。此外，与传统指标（如RMS和峰度）相比，脉冲能量指标对故障的严重程度具有一致的响应。

结论

已经开发并测试了提取与轴承故障脉冲相对应的频带的技术。该技术采用自适应滤波器信号分离，小波包分解以及RMS和峰度的组合来选择最佳滤波器频带。