摘要
齿轮箱早期的故障信号往往十分微弱 ,信噪比低 ,这大大限制了已有诊断方法在早期诊断中的应用 ,因此如何获取真实的振动信号是提高齿轮箱早期故障诊断质量的关键 ,独立分量分析 (ICA)为此提供了一种新的思路。文中研究了ICA在齿轮箱故障早期诊断中的应用 ,首先分析了齿轮箱的混合振动信号模型 ,然后针对具体的轴承故障进行了实验 ,并使用快速ICA算法分离出轴承的振动信号 ,再将其功率谱与原始振动信号的谱相比较 ,结果表明ICA更易于实现故障的早期诊断 ;最后提出了进一步的研究建议。
Early gearbox fault signal is often very weak and its SNR(signal to noise ratio) is low, which greatly constrains the use of existing diagnosis methods. Thus how to determine the true vibration signals is the key to improve the diagnostic performance. Independent component analysis (ICA) provides a way for it. This paper proposes the application of ICA to early diagnosis of helicopter gearboxes. First the composite model of gearbox vibration signal is built; then one experiment is done on one actual faulty bearing and the bearing vibration signal is separated by Fast ICA algorithm. Different PSDs(Power Spectrum Densities) between separated signal and original signal are compared and the results demonstrate that ICA can be used for early fault diagnosis easily indeed.
出处
《机械科学与技术》
CSCD
北大核心
2004年第4期481-483,500,共4页
Mechanical Science and Technology for Aerospace Engineering
基金
国防预研项目资助
