Open AccessEvaluation method of spindle performance degradation based on VMD and random forests
Author(s) -
Wei Pangshi,
Wang Hongjun
Publication year - 2019
Publication title -
the journal of engineering
Language(s) - English
Resource type - Journals
ISSN - 2051-3305
DOI - 10.1049/joe.2018.9127
Subject(s) - degradation (telecommunications) , reliability (semiconductor) , modal , eigenvalues and eigenvectors , computer science , signal (programming language) , domain (mathematical analysis) , frequency domain , algorithm , time domain , modal analysis , decomposition , pattern recognition (psychology) , biological system , artificial intelligence , mathematics , acoustics , materials science , vibration , physics , computer vision , chemistry , mathematical analysis , telecommunications , power (physics) , organic chemistry , quantum mechanics , polymer chemistry , biology , programming language
As one of the core components of the machine tool, the reliability of spindle is very important for improving machine tool quality. To effectively evaluate the performance degradation degree of the spindle, a method of degradation evaluation of the spindle performance based on variational mode decomposition (VMD) and random forests (RFs) was proposed. Firstly, VMD is used to process the current signal to obtain several modal components. Then, the time domain and frequency domain features of each modes component are calculated as eigenvalues. Finally, the RFs algorithm is used to classify the eigenvalues. The experimental results show that VMD can decompose the signal better and avoid the phenomenon of the modal mixture. The combination of VMD and RFs can accurately and effectively evaluate the performance degradation of the spindle.