摘要
文章针对轴承早期故障特征的提取,提出了基于改进EWT-SVD的算法。首先,改进的经验小波(EWT)提出了模态分解数量确定的思路,自适应地将预处理信号分解到合适数量的模态分量,通过相关度系数验证了分解模态的信号有效性;其次,通过计算各分量的峭度值确定最优的特征提取模态分量,并通过变阈值奇异值分解(SVD)对模态信号进行去噪;最后,通过对重构特征信号进行Hilbert变换包络处理提取振动信号频率特征。实验证明了文章算法的可行性,同时,算法还具有计算速度快、以数据为主要驱动的特点。
An improved EWT-SVD algorithm is proposed to extract the early fault features of bearings in the thesis. Firstly, the IEWT algorithm proposed the method of calculating the number of decomposition modal, which adaptively decomposes the preprocessed signal to the appropriate components number, and verified the validity by the relevancy coefficient. Secondly, the optimal component is determined by the kurtosis, and the modal signal is denoised by SVD with variable threshold method. After that, the frequency feature of the reconstructed signal is extracted by the Hilbert transform envelope processing. Finally, the feasibility of the algorithm is proved by experiments. And, the algorithm has the characteristics of fast computing speed and mainly data-driven.
作者
车守全
包从望
江伟
陈俊
肖钦兰
CHE Shou-quan;BAO Cong-wang;JIANG Wei;CHEN Jun;XIAO Qin-lan(School of Mines and Civil Engineering,Liupanshui Normal University,Liupanshui Guizhou 553000,China)
出处
《组合机床与自动化加工技术》
北大核心
2020年第9期80-83,共4页
Modular Machine Tool & Automatic Manufacturing Technique
基金
贵州省矿山装备数字化技术工程研究中心(黔教合KY字[2017]026号)
六盘水市科研创新平台和人才团队建设(52020-2019-5-12)。