摘要
针对工业工程各领域广泛应用的螺旋锥齿轮振动信号受环境噪声干扰大,出现故障时信号呈现非线性、非平稳特性,故障特征信息微弱,故障特征提取难、诊断效率低的难题,提出一种基于多尺度排列熵(multi-scale permutation entropy,MPE)-局部保持投影(locality preserving projections,LPP)与极限学习机(extreme learning machine,ELM)的螺旋锥齿轮状态识别方法。首先,构造MPE作为原始高维特征矢量;然后使用LPP对原始高维特征矢量降维,获得最优低维敏感特征矢量,挖掘并保留高维特征矢量的非线性结构特点;最后将所得敏感特征量输入ELM进行识别诊断。该方法应用于3种转速下4种故障状态螺旋锥齿轮的诊断中,并与基于MPE-PCA-ELM与MPE-ELM进行对比识别,结果有效地证明了方法的准确性和优越性。
For spiral bevel gears widely used in various fields of industrial engineering, the vibration signal is greatly disturbed by environmental noise. When the fault occurs, the signal exhibits nonlinear, non-stationary characteristics, the fault feature information is weak, the fault feature extraction is difficult, and the diagnostic efficiency is low. Therefore, a spiral bevel gear state recognition method based on MPE-LPP and ELM is proposed. Firstly, construct multi-scale entropy values as the original high-dimensional feature vectors, then use LPP to obtain the optimal low-dimensional sensitive feature vectors by reducing the original high-dimensional feature vectors, which can mine and preserve the nonlinear structural features of the original high-dimensional features. The obtained sensitive feature quantity is input into the ELM for recognition diagnosis. The method is applied to the diagnosis of four kinds of fault state spiral bevel gears under three kinds of speeds, and compared with MPE-PCA-ELM and MPE-ELM. The results prove the accuracy and superiority of the proposed method.
作者
谭鸿创
杨大炼
蒋玲莉
李学军
Tan Hongchuang;Yang Dalian;Jiang Lingli;Li Xuejun(Hunan Provincial Key Laboratory of Health Maintenance for Mechanical Equipment,Hunan University of Science and Technology,Xiangtan 411201,China;Foshan University,Mechanical and Electrical Engineering,Foshan 528225,China)
出处
《电子测量与仪器学报》
CSCD
北大核心
2020年第2期44-52,共9页
Journal of Electronic Measurement and Instrumentation
基金
国家自然科学基金(11872022,51575177)
湖南省科技厅“科技人才专项-湖湘青年英才”项目(2017RS3049)
湖南省自然科学基金(11702091)资助项目。
