基于最小二乘法和SSA算法的行星齿轮箱微弱振动故障探测

Weak vibration fault detection of planetary gearbox based on least squares method and SSA algorithm

摘要:

行星齿轮箱是一种常用于工业机械和车辆传动系统中的重要装置,其长期运转和负荷导致行星齿轮箱存在着微弱振动故障的风险,通过研究行星齿轮箱微弱振动故障的特征和变化趋势,可以提前发现潜在故障迹象,提高齿轮箱的安全性。然而,行星齿轮箱微弱振动信号具有复杂的时域和频域特征,且其振动信号非常微弱,并富含大量噪声,为行星齿轮箱的故障探测工作带来较大难度。为此,提出基于最小二乘法和SSA算法的行星齿轮箱微弱振动故障探测方法。模拟行星齿轮箱工作与振动流程,根据不同故障下齿轮箱的工作特征,设置微弱振动故障探测标准。采集行星齿轮箱微弱振动信号,利用最小二乘法降低行星齿轮箱微弱振动信号中的噪声含量。利用SSA算法提取行星齿轮箱微弱振动信号特征,根据提取特征与设置标准的匹配,得出故障类型和故障参数的探测结果。通过性能测试实验得出结论:所提方法探测行星齿轮箱微弱振动信号的信噪比明显降低,故障探测范围扩大了353.94mm2,具有较好的探测能力。

Abstract:

planetary gearbox is a kind of commonly used in industrial machinery and vehicle transmission system, its long-term operation and load lead to the risk of weak vibration failure, by studying the characteristics of the planetary gearbox weak vibration fault and trend, can find potential signs of failure in advance, improve the safety of the gear box. However, the weak vibration signal of planetary gearbox has complex time domain and frequency domain characteristics, and its vibration signal is very weak, and rich in a large amount of noise, which brings great difficulty to the fault detection work of planetary gearbox. Therefore, the method of detecting weak vibration fault based on least squares and SSA algorithm is proposed. Simate the working and vibration process of planetary gearbox, and set the weak vibration fault detection standard according to the operating characteristics of the gearbox under different faults. Collect the weak vibration signal of the planetary gearbox and reduce the noise content of the planetary gearbox by least squares. The SSA algorithm is used to extract the weak vibration signal features of the planetary gearbox, and to obtain the detection results of fault types and fault parameters according to the matching between the extracted features and the setting standard. Through the performance test experiment, the conclusion is that the signal-to-noise ratio of the weak vibration signal of the planetary gearbox is significantly reduced, the fault detection range is expanded by 353.94mm2, and it has a good detection ability.