Loading...

A fast online bandwidth empirical mode decomposition scheme for avoidance of the mode mixing problem

Momeni Massouleh, S. H ; Sharif University of Technology | 2018

682 Viewed
  1. Type of Document: Article
  2. DOI: 10.1177/0954406217741515
  3. Publisher: SAGE Publications Ltd , 2018
  4. Abstract:
  5. The main objective of this work is to propose a scheme to extract intrinsic mode functions of online data with an acceptable speed as well as accuracy. For this purpose, an individual block framework method is firstly employed to extract the intrinsic mode functions. In this method, buffers are selected such that they overlap with their neighbors to prevent the end effect errors with no need for the averaging process. And in order to avoid the mode mixing problem, a bandwidth empirical mode decomposition scheme is developed to effectively improve the results. Through this scheme, an auxiliary function made of both high- and low-frequency components corresponding to noise and dominant frequency is added to data for the strengthening of the components for the better extraction of intrinsic mode functions during sifting process. An index criterion as well as a threshold limit is also introduced to separate high- and low-frequency parts of data at desired frequency range. Advantages of the proposed scheme are assessed and comparisons with the available methods are presented. Solution of different types of examples and experimentally generated data for two faulty ball bearings reveals that the present easily implemented scheme achieves results with lower computational efforts and accuracy. © IMechE 2017
  6. Keywords:
  7. Bandwidth empirical mode decomposition method ; Denoising ; Mode mixing problem ; Ball bearings ; Functions ; Mixing ; Signal processing ; Computational effort ; De-noising ; Empirical Mode Decomposition ; Empirical mode decomposition method ; Intrinsic Mode functions ; Low-frequency components ; Mode mixing ; Online data ; Bandwidth
  8. Source: Proceedings of the Institution of Mechanical Engineers, Part C: Journal of Mechanical Engineering Science ; Volume 232, Issue 20 , 2018 , Pages 3652-3674 ; 09544062 (ISSN)
  9. URL: https://journals.sagepub.com/doi/abs/10.1177/0954406217741515