摘要
针对行星变速箱齿轮故障信号特征易被噪声湮没且不同齿轮故障信号较难区分的特点,提出了一种基于离散小波变换(Discrete Wavelet Transform,DWT)和局部保持映射(Local Preserving Projection,LPP)的行星变速箱故障信号特征增强方法。首先,利用DWT对信号进行多频段的重构扩充信号维度;然后,通过LPP对多维度信号进行降维,减弱噪声影响并增强信号的稳定性;最后,以排列熵(Permutation Entropy,PE)、样本熵(Sample Entropy,SE)和功率谱熵(Power Spectral Entropy,PSE) 3种信息熵表征信号特征。对台架试验采集不同故障状态的振动信号进行分析,结果表明:该方法对故障信号特征增强明显,依据3种信息熵值的三维坐标有效实现了行星变速箱齿轮故障的分类识别。
The planetary gearbox fault signal characteristics are easy to be annihilated by noise and different gear fault signals are difficult to distinguish.To solve this problem,a planetary gearbox fault signal feature enhancement method based on the Discrete Wavelet Transform(DWT)and the Local Preserving Projection(LPP)is proposed.Firstly,DWT is used to reconstruct the multi-band signal to expand the signal dimension.Then,the multi-dimensional signal is reduced by LPP,which reduces the influence of noise and enhances the stability of the signal.Finally,the signal characteristics are characterized by three kinds of information entropy:Permutation Entropy(PE),Sample Entropy(SE)and Power Spectral Entropy(PSE).The vibration signals of different fault states are collected by bench test.The results show that the method enhances the characteristics of fault signals obviously.The classification and identification of planetary gearbox faults are realized based on the three-dimensional coordinates of three kinds of information entropy values.
作者
江鹏程
丛华
吴春志
冯辅周
JIANG Peng-cheng;CONG Hua;WU Chun-zhi;FENG Fu-zhou(Vehicle Engineering Department,Army Academy of Armored Forces,Beijing 100072,China)
出处
《装甲兵工程学院学报》
2019年第2期75-80,共6页
Journal of Academy of Armored Force Engineering
基金
军队科研计划项目
