Data Decomposition Method Combining Permutation Entropy and Spectral Substitution with Ensemble Empirical Mode Decomposition

Shengxiang Huang,Xinpeng Wang,Chenfeng Li,Chao Kang

DOI: https://doi.org/10.1016/j.measurement.2019.01.026

IF: 5.6

2019-01-01

Measurement

Abstract:As themajority of real signals (e.g. the dynamic information obtained during bridgemonitoring) are nonstationary and nonlinear, empirical mode decomposition (EMD) and its derivatives have become popular research topics in recent yearsworldwide. Existing methods, such asEMD, ensembleempirica lmode decomposition (EEMD), complementary ensemble empirical mode decomposition (CEEMD), complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN) and partly ensemble empirical mode decomposition (PEEMD), may generate modal aliasing. This condition exerts an adverse effect on the denoising and loss of high-frequency components during signal decomposition. A new method, which combines permutation entropy and spectral substitution with ensemble EMD, is proposed to solve this problem. Results of our analysis indicate that themethod can extract multi-frequencies and effectively denoise signals. Advantages emerge with regard to practical signals collected via the global navigation satellite systems (GNSS) monitoring of the dynamic characteristics of a bridge in China. In conclusion, the natural vibration frequency in the x-andy-directions at each section is 0.2267 Hz, whichdemonstrates the health of the bridge. (C) 2019 Published by Elsevier Ltd.