数学形态学和LMD算法下滚动轴承全生命周期故障检测研究  

Research on Full Life Cycle Fault Detection of Rolling Bearings under Mathematical Morphology and LMD Algorithm

作　　者：严峰军[1] YAN Fengjun(Foundation Department,Xi’an Siyuan University,Xi’an 710038,China)

机构地区：[1]西安思源学院基础部,西安710038

出　　处：《计算机测量与控制》2024年第12期50-56,66,共8页Computer Measurement &Control

摘　　要：当滚动轴承在高速旋转时,会产生振动和摩擦,容易引起轴承表面的细微磨损和损伤,且在恶劣的工作环境中,会加剧轴承的磨损和腐蚀,难以区分表面缺陷;为此,对滚动轴承全生命周期故障检测方法进行了研究;根据滚动轴承的故障机理及特征,设置滚动轴承故障检测标准,模拟滚动轴承全生命周期工作过程;采集并预处理滚动轴承的表面图像数据和内部振动数据,利用数学形态学基于形状特征提取滚动轴承表面图像的微小特征,通过LMD算法分解复杂信号为多个单一调频和窄带调频分量,提取峭度、频率等关键特征;采用特征匹配的方式,得出滚动轴承故障类型、位置以及故障量的检测结果;通过实验得出结论:优化设计方法的故障类型误检率明显降低,具有良好的故障检测能力。When the rolling bearings rotate at high speed,the vibration and friction generated can easily cause minor wear and damage in the bearing surface.In harsh working environments,it can exacerbate bearing wear and corrosion,making it difficult to distinguish surface defects.For this purpose,a fault detection method for the entire life cycle of rolling bearings is studied.Based on the fault mechanism and characteristics of rolling bearings,set fault detection standards for rolling bearings,and simulate the entire life cycle working process of rolling bearings.Collect and preprocess surface image data and internal vibration data of rolling bearings,extract small features of rolling bearing surface images based on shape features using mathematical morphology,decompose complex signals into multiple single frequency and narrowband frequency components using the LMD algorithm,and extract key features such as kurtosis and frequency.The feature matching is used to obtain the detection results of the type,location,and quantity of rolling bearing faults.Experimental results show that the fault type false detection rate of the optimized design method is significantly reduced,with a good fault detection ability.

关 键 词：数学形态学 LMD算法 滚动轴承 全生命周期 故障检测 

分 类 号：TP133[自动化与计算机技术—控制理论与控制工程]