摘要
针对球磨机振动信号非线性、非平稳性导致的负荷状态难以识别问题,提出一种基于经验小波变换(EWT)和奇异值熵相结合的磨机负荷识别方法。首先,采用经验小波变换对磨机筒体的原始振动信号进行分解得到本征模态分量,初步提取球磨机负荷状态特征值;其次,通过相关系数法选出能表征原始信号状态的敏感模态分量进行信号重构;最后,利用奇异值分解技术对重构信号进行处理并计算出奇异值熵,利用奇异值熵的大小判断球磨机的负荷状态。球磨机在3种工况下的磨矿实验结果表明:相较于其他2种识别方法,EWT-奇异值熵识别时,球磨机3种负荷状态之间的熵值差异很大,识别区间分别为欠负荷(48,61)、正常负荷(98,111)、过负荷(34,45),能够较好地区分磨机负荷状态。
In order to solve the problem that the load state of ball mill was difficult to identify due to the non-linear and nonstationary vibration signals,a load identification method based on the combination of empirical wavelet transform(EWT)and singular value entropy was proposed.Firstly,the original vibration signals of the mill barrel were decomposed by the empirical wavelet transform to obtain the eigenmode component,and the characteristic value of the load state of the ball mill was extracted preliminarily.Secondly,the sensitive mode component which can represent the original signal state was selected by the correlation coefficient method for signal reconstruction.Finally,the reconstructed signal was processed by the singular value decomposition technology to calculate the singular value entropy.The load state of ball mill can be judged by singular value entropy.The results of grinding experiments under three working conditions of ball mill showed that compared with other two identification methods,the entropy values of three load states of ball mill were very different as using EWT singular value entropy to identify.The identification ranges were under load(48,61),normal load(98,111),and overload(34,45),which can better distinguish the load states of ball mill.
作者
高纯生
刘鑫
谢文涓
蔡改贫
GAO Chunsheng;LIU Xin;XIE Wenjuan;CAI Gaipin(Chinalco Mining Co.,Ltd,Zhengzhou,Henan 450041,China;School of Mechanical and Electrical Engineering,Jiangxi Universityof Science and Technology,Ganzhou,Jiangxi 341000,China)
出处
《矿业研究与开发》
CAS
北大核心
2019年第11期130-136,共7页
Mining Research and Development
基金
国家自然科学基金项目（51464017）
江西省教育厅科技重点项目（GJJ150618）
