基于CEEMDAN和SE算法的打夯机负荷振动信号识别研究  

Research on load vibration signal identification of crusher based on CEEMDAN and SE algorithm

作　　者：刘剑 王强[2] LIU Jian;WANG Qiang(Gansu Province Transportation Planning,Survey&Design Institute Co.,Ltd.,Lanzhou 730030,Gansu,China;Department of Material Forming and Control Engineering,Lanzhou University of Technology,Lanzhou 730050,Gansu,China)

机构地区：[1]甘肃省交通规划勘察设计院股份有限公司,甘肃兰州730030 [2]兰州理工大学材料成型与控制工程系,甘肃兰州730050

出　　处：《中国工程机械学报》2024年第2期220-224,共5页Chinese Journal of Construction Machinery

基　　金：国家自然科学基金资助项目(51775157)。

摘　　要：为了提高打夯机在复杂工作环境中筒体容易产生多种非线性振动信号扛干扰能力,设计了一种基于自适应噪声完备经验模态分解(CEEMDAN)算法和标准误差(SE)算法的打夯机负荷振动信号识别方法。采用CEEMDAN算法分解信号数据,以极限学习机为打夯机负荷建立模型,完成打夯机负荷的精确判断。研究结果表明:CEEMDAN对打夯机振动信号起到良好的预处理作用,各内涵模态(IMF)分量SE值均未出现相互重叠,有效IMF分量SE总体表现为欠负荷>常负荷>过负荷的特征。该模型对过负荷达到最高识别率,形成了98.85%的过负荷识别率,比EMD-SE与MEEMD-SE的过负荷识别率依次增大15.61%、12.14%。该研究可以有效识别负荷情况,为下一步驱动打夯机做出相应动作奠定基础,有效地提高节能效果。In order to improve the ability of the cylinder to generate nonlinear vibration signals in the complex working environment of the tamper,a load vibration signal recognition method based on CEEMDAN and SE algorithms is designed.The signal data is decomposed by CEEMDAN algorithm,and the load model of the rammer is established by using the extreme learning machine to determine the load of the rammer accurately.The results show that CEEMDAN has a good preprocessing effect on the vibration signals of the rammer,and the SE values of the IMF components do not overlap with each other,and the effective IMF components SE generally show the characteristics of underload>normal load>overload;Characteristics of overload.This model achieves the highest overload recognition rate,with an overload recognition rate of 98.85%,which is 15.61%and 12.14%higher than that of EMD-SE and MEEMD-SE respectively.The research can effectively identify the load situation,lay the foundation for the next step to drive the tamper to make the corresponding action,and effectively improve the energy-saving effect.

关 键 词：CEEMDAN SE 相关系数 ELM 负荷状态识别 

分 类 号：TD921[矿业工程—选矿]