XXXI International Mineral Processing Congress 2024 Proceedings/Washington, DC/Sep 29–Oct 3 585
bed. These images were processed using Matlab through
binarization techniques. After processing, the bubble con-
tours were marked with spots to calculate their equivalent
diameters. Subsequently, the evolving patterns of adjacent
bubble images were analyzed by comparing and studying
them. The process is illustrated in Figure 2.
Traditionally, research on bubble movement in gas-solid
fluidized beds has been primarily conducted using two-
dimensional fluidized bed imaging tests. However, there
has been limited investigation on the real-time correlation
between bubble movement and bed pressure drop fluctua-
tions. To address this gap, Dong Liang et al. [5,6] proposed
an innovative approach to identify bubbles by combining
high-speed dynamic imaging technology with pressure
drop sensors. They utilized a high-speed dynamic camera to
capture images of bubbles’ dynamic behavior during their
ascent, coalescence, and rupture processes. Simultaneously,
they measured the response pressure drop signals of bub-
ble evolution using pressure sensors. By employing digi-
tal image processing and signal analysis techniques, they
conducted a comparative analysis of various bubble move-
ment behavior images and their corresponding pressure
drop signals. This analysis enabled the accurate identifi-
cation and extraction of pressure drop signals induced by
bubble movement, coalescence, and rupture. Furthermore,
they quantitatively studied the evolution of characteristic
parameters such as bubble shape, size, and quantity distri-
bution in the fluidized bed.
ECT Identification
We used electrical capacitance tomography technology to
identify bubbles in a three-dimensional fluidized bed. The
working principle of electrical capacitance tomography
(ECT) is as follows: The theoretical foundation of ECT is
the Poisson equation
,yh@ 0
0 $$d d{ f f^x,y =^h 6 (1)
where ∇· is the divergence operator,
0 f is the absolute per-
mittivity in vacuum, ,yh f^ is the permittivity distribution
Figure 2. Processing flow for image acquisition of non-uniform mesoscale structures[4]
bed. These images were processed using Matlab through
binarization techniques. After processing, the bubble con-
tours were marked with spots to calculate their equivalent
diameters. Subsequently, the evolving patterns of adjacent
bubble images were analyzed by comparing and studying
them. The process is illustrated in Figure 2.
Traditionally, research on bubble movement in gas-solid
fluidized beds has been primarily conducted using two-
dimensional fluidized bed imaging tests. However, there
has been limited investigation on the real-time correlation
between bubble movement and bed pressure drop fluctua-
tions. To address this gap, Dong Liang et al. [5,6] proposed
an innovative approach to identify bubbles by combining
high-speed dynamic imaging technology with pressure
drop sensors. They utilized a high-speed dynamic camera to
capture images of bubbles’ dynamic behavior during their
ascent, coalescence, and rupture processes. Simultaneously,
they measured the response pressure drop signals of bub-
ble evolution using pressure sensors. By employing digi-
tal image processing and signal analysis techniques, they
conducted a comparative analysis of various bubble move-
ment behavior images and their corresponding pressure
drop signals. This analysis enabled the accurate identifi-
cation and extraction of pressure drop signals induced by
bubble movement, coalescence, and rupture. Furthermore,
they quantitatively studied the evolution of characteristic
parameters such as bubble shape, size, and quantity distri-
bution in the fluidized bed.
ECT Identification
We used electrical capacitance tomography technology to
identify bubbles in a three-dimensional fluidized bed. The
working principle of electrical capacitance tomography
(ECT) is as follows: The theoretical foundation of ECT is
the Poisson equation
,yh@ 0
0 $$d d{ f f^x,y =^h 6 (1)
where ∇· is the divergence operator,
0 f is the absolute per-
mittivity in vacuum, ,yh f^ is the permittivity distribution
Figure 2. Processing flow for image acquisition of non-uniform mesoscale structures[4]