A new method for identifying flow pattern of spouted fluidized bed by coupling Hilbert–Huang transform characteristics of differential pressure signals in different zones--颗粒学报PARTICUOLOGY

Mei, Y., Wang, Y., Zhang, L., & Duan, F. (2024). A new method for identifying flow pattern of spouted fluidized bed by coupling Hilbert–Huang transform characteristics of differential pressure signals in different zones. Particuology, 89, 67-78. https://doi.org/10.1016/j.partic.2023.10.010

A new method for identifying flow pattern of spouted fluidized bed by coupling Hilbert–Huang transform characteristics of differential pressure signals in different zones

Yongzhi Mei^{a b}, Yang Wang^c, Lihui Zhang^{a b}, Feng Duan^{a b *}

a School of Energy and Environment, Anhui University of Technology, Maanshan, 243002, China
b Wuhu Technology and Innovation Research Institute, Wuhu, 241000, China
c Wuhu Cigarette Factory of China Tobacco Anhui Industrial Co., Ldt., Wuhu, 241002, China

10.1016/j.partic.2023.10.010

Volume 89, June 2024, Pages 67-78

Received 5 June 2023, Revised 5 October 2023, Accepted 20 October 2023, Available online 2 November 2023, Version of Record 6 December 2023.

E-mail: ddffeng@126.com

Highlights

• Hilbert-Huang transform is used to identify the flow pattern of spouted fluidized bed.

• Hilbert-Huang transform characteristics of different flow patterns are different.

• Middle-frequency range energy of Intrinsic modal functions is related to flow pattern.

• Flow pattern recognition based on Hilbert-Huang transform has high accuracy.

Abstract

On a cold spouted fluidized bed, this study compares the characteristic differences in intrinsic mode function (IMF) energy and Hilbert–Huang spectrum between the spout zone and annulus zone under different combinations of spouted gas and fluidized gas flow rates for five typical flow patterns. The energy distribution characteristics under different flow patterns are also analyzed. The Hilbert–Huang spectrum and IMF energy of pressure difference signals exhibit distinct variations in different zones as the flow pattern changes. Moreover, there exists a correlation between the energy in the middle-frequency range and the flow pattern. Leveraging the K-means algorithm, the middle-frequency range energy of IMFs in the spout zone and annulus zone is subjected to clustering analysis, leading to the identification of partition boundaries for each flow pattern. Based on this, a flow pattern identification method of spouted fluidized bed coupled with middle-frequency range energy in spout zone and annulus zone is proposed, which has very high identification accuracy.

Graphical abstract

Keywords

Spouted fluidized bed; Flow pattern; Zoning signal; Flow pattern transition; Hilbert–Huang transform

文章导读