基于OSGMD-Hilbert包络对数分析的齿轮箱齿面磨损早期故障诊断  

Early fault diagnosis of gearbox teeth surface wear based on OSGMD-Hilbert envelope logarithmic analysis

作　　者：俞香熔 王友仁[1] 王胤博 YU Xiangrong;WANG Youren;WANG Yinbo(College of Automation Engineering,Nanjing University of Aeronautics and Astronautics,Nanjing 211106,China;State Key Lab of Mechanical System and Vibration,School of Mechanical Engineering,Shanghai Jiao Tong University,Shanghai 200240,China)

机构地区：[1]南京航空航天大学自动化学院,南京211106 [2]上海交通大学机械与动力工程学院机械系统与振动国家重点实验室,上海200240

出　　处：《振动与冲击》2025年第7期225-231,274,共8页Journal of Vibration and Shock

基　　金：航空发动机及燃气轮机重大专项基础研究项目(J2019-ΙV-0018-0086);国家自然科学基金项目(52372430);航空科学基金项目(20230033052002);南京航空航天大学研究生科研与实践创新计划资助项目(xcxjh20230327)。

摘　　要：针对环境噪声下齿轮箱齿面磨损早期微弱故障特征难以提取的问题,提出了一种基于优化型辛几何模态分解-Hilbert包络对数分析的齿轮磨损故障诊断方法。新方法中引入Cao算法和功率谱密度,提出最近邻波动偏差实现嵌入维数的自适应确定,利用奇异值分解进行降噪,采用Pearson-功率谱熵差和闵氏距离作为重构准则以获取特征模态分量,通过Hilbert包络对数分析法突出故障频率成分,并进行故障诊断。该新方法克服了辛几何模态分解嵌入维数依赖经验公式、重构准则单一和噪声鲁棒性欠佳的缺陷。仿真与试验结果分析表明,与辛几何模态分解(symplectic geometric mode decomposition,SGMD)、迭代SGMD、变分模态分解和经验模态分解相比,该新方法能够有效提取早期齿面磨损故障特征信息,表现出更好的鲁棒性。Here,aiming at the problem of difficulty in extracting early weak fault features of gearbox teeth surface wear under environmental noise,a gear wear fault diagnosis method based on optimized symplectic geometry mode decomposition-Hilbert envelope logarithmic analysis was proposed.In the new method,Cao algorithm and power spectral density were introduced,and the nearest neighbor fluctuation deviation was proposed to adaptively determine embedding dimension number.Singular value decomposition was used for denoising.Pearson power spectral entropy difference and Minkowski distance were taken as reconstruction criteria to obtain characteristic modal components.Hilbert envelope logarithmic analysis method was used to highlight fault frequency components and perform fault diagnosis.This new method could overcome shortcomings of dependency on empirical formulas for embedding dimension number in symplectic geometry modal decomposition,singularity of reconstruction criteria and poor noise robustness.Simulation and experimental results showed that compared with symplectic geometry modal decomposition(SGMD),iterative SGMD,variational mode decomposition and empirical mode decomposition,this new method can effectively extract early gear teeth surface wear fault feature information,and exhibit better robustness.

关 键 词：早期故障诊断 振动信号特征信息提取 优化型辛几何模态分解(OSGMD) 齿轮磨损 Hilbert包络对数分析法 辛几何模态分解(SGMD) 

分 类 号：TH113[机械工程—机械设计及理论]