直升机动力传动系统振动信号分析及故障诊断

发布时间：2018-07-03 19:22

本文选题：直升机动力传动系统 + 行星齿轮　；参考：《南昌航空大学》2012年硕士论文

【摘要】：为了降低飞行事故，保证飞行安全，对直升机动力传动系统进行故障诊断研究。行星齿轮传动系是直升机动力传动系统的关键部分，动力系统事故多出于行星传动系；振动诊断是动力系统诊断的基本方法，所以直升机动力传动系统故障诊断的关键就是行星齿轮的振动诊断。如果能把齿轮箱振动信号进行分解，就可以通过检查各个齿轮振动信号的特点，发展齿轮状态估计与故障诊断的各种方法。行星齿轮振动诊断首先必须分离出每个行星轮的振动信号，然后对分离出的信号采用适当的方法做进一步的分析处理，从而诊断出行星轮的故障所在，以达到直升机动力传动系统故障诊断的目的。本文针对行星齿轮失效中最常见的轮齿故障进行诊断和分析。首先结合信号加窗等相关理论，提出基于时域同步平均的行星齿轮振动信号分离方法，，该方法采用多传感器测量的方式获取齿轮振动信号，并根据行星齿轮振动信号波形的整体形状来确定分离窗函数。然后设计并制作直升机动力传动系统故障诊断试验台，建立基于LABVIEW虚拟仪器技术的振动信号采集系统。再对行星齿轮做人为毁坏失效试验，使行星齿轮分别在无故障、一个轮齿破损以及连续五个轮齿破损的情况下进行啮合试验，采集每种情况下齿轮的振动信号。最后利用基于时域同步平均的行星齿轮振动信号分离方法直接分离齿轮箱振动采样原始信号，并通过时域统计指标、频谱、倒频谱、细化谱、连续小波变换以及提升小波变换等分析方法对分离后的信号进行综合对比分析。结果表明：基于时域同步平均的行星齿轮振动信号分离方法可实现对行星齿轮振动信号的充分分离，实现在信号平均时不丢失信息，而且不再履行传统信号平均的那些要求，无需较长采样周期；小波变换方法能够有效诊断和定位行星轮的故障信息，反映齿轮失效的演变，诊断准确率高，其中连续小波变换可以很容易判断出齿轮的故障类型，提升小波变换可以判断出齿轮中具体有几个轮齿产生了破损现象。充分验证了基于时域同步平均的行星齿轮振动信号分离方法的有效性，以及小波分析方法在行星齿轮故障诊断中的优越性和可行性。
[Abstract]:In order to reduce flight accidents and ensure flight safety, the fault diagnosis of helicopter power transmission system is studied. The planetary gear transmission system is the key part of the helicopter power transmission system, and the accidents of the power system are mostly caused by the planetary transmission system, and vibration diagnosis is the basic method of the power system diagnosis. So the key to the fault diagnosis of helicopter power transmission system is the vibration diagnosis of planetary gear. If the vibration signal of the gear box can be decomposed, various methods of gear state estimation and fault diagnosis can be developed by checking the characteristics of each gear vibration signal. The vibration diagnosis of planetary gear must first separate out the vibration signal of each planetary gear, and then make further analysis and processing of the separated signal, so as to diagnose the fault of the planetary gear. In order to achieve the helicopter power transmission system fault diagnosis purpose. In this paper, the most common gear failure in planetary gear failure is diagnosed and analyzed. Based on the theory of signal windowing and so on, a method of separating planetary gear vibration signal based on time-domain synchronous averaging is proposed. The method uses multi-sensor measurement to obtain gear vibration signal. The separation window function is determined according to the whole shape of the vibration signal waveform of planetary gear. Then the fault diagnosis test-bed of helicopter power transmission system is designed and made, and the vibration signal acquisition system based on LabVIEW virtual instrument technology is established. Then the planetary gear is damaged and failure test, which makes the planetary gear meshing test under the condition of no fault, one gear tooth breakage and five consecutive gear teeth breakage, and collects the vibration signal of each kind of gear. Finally, the planetary gear vibration signal separation method based on time-domain synchronous averaging is used to directly separate the original signal of gear box vibration, and through time domain statistical index, frequency spectrum, cepstrum, fine spectrum. Continuous wavelet transform and lifting wavelet transform are used to analyze the separated signals. The results show that the method of separation of planetary gear vibration signals based on time-domain synchronous averaging can realize the full separation of planetary gear vibration signals, and realize that there is no loss of information when the signals are averaged, and that the requirements of traditional signal averaging are no longer fulfilled. The wavelet transform method can effectively diagnose and locate the fault information of planetary gear, reflect the evolution of gear failure, and have high diagnostic accuracy. The continuous wavelet transform can easily determine the fault type of gear. Lifting wavelet transform can be used to determine the damage of several gear teeth in the gear. The validity of the method based on time-domain synchronous averaging for separating planetary gear vibration signals and the superiority and feasibility of wavelet analysis in fault diagnosis of planetary gears are fully verified.
【学位授予单位】：南昌航空大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：TH132.41;TH165.3

【参考文献】