A First Principles–Neural Networks Approach to Model a Vibrated Fluidized Bed Dryer: Simulations and Experimental Results

2005 ◽  
Vol 23 (1-2) ◽  
pp. 187-203 ◽  
Author(s):  
P.I. Alvarez ◽  
R. Blasco ◽  
J. Gomez ◽  
F.A. Cubillos
2011 ◽  
Vol 29 (3) ◽  
pp. 295-307 ◽  
Author(s):  
Tayyeb Nazghelichi ◽  
Mortaza Aghbashlo ◽  
Mohammad Hossein Kianmehr ◽  
Mahmoud Omid

2019 ◽  
Vol 7 (2) ◽  
Author(s):  
S. Wongsiriwan ◽  
Thongchai Rohitatisha Srinophakun ◽  
Pakon Laopreecha

The particle motion, temperature behavior, and drying rate of particle inside a vibrated fluidized bed dryer were numerically investigated in this work. In the simulation, the Distinct Element Method (DEM) based on the Newton’s second law of motion was used to solve the particle motion. The physical aspects of fluid motion and heat transfer were obtained by applying Computational Fluid Dynamics (CFD) technique. For the drying of particle, only the constant rate period was considered in order to save the computational time. Programming was developed in Standard-C language and using MATLAB to visualize the results. In the simulation, 2,000 particles with stiffness 800 N m-1 were simulated in a rectangular bed. The developed model was validated with an experimental result of Gupta et al. [1]. The program was then used to study the effect of superficial gas velocity (U0), frequency of vibration (f) and amplitude of vibration (a) in fluidized bed dryer. At low velocities and no vibration of bed,  articles in the bed were not fluidized but smoothly circulated. Thus, the heat transfer occurred only near the orifice. When superficial gas velocity increased, the fluidization of the particles was observed. The fluidization and drying rate improved with increased in superficial velocity for both vibrated fluidized bed and stationary bed. With introducing of vibration, the fluidization behavior of the particle was improved. The particles in the bed were well mixed and also increased the drying rate. From the simulation results, increasing of frequency and amplitude could not significantly improve rate of drying.


1995 ◽  
Vol 60 (12) ◽  
pp. 2074-2084
Author(s):  
Petr Mikulášek

The microfiltration of a model fluid on an α-alumina microfiltration tubular membrane in the presence of a fluidized bed has been examined. Following the description of the basic characteristic of alumina tubular membranes, model dispersion and spherical particles used, some comments on the experimental system and experimental results for different microfiltration systems are presented. From the analysis of experimental results it may be concluded that the use of turbulence-promoting agents resulted in a significant increase of permeate flux through the membrane. It was found out that the optimum porosity of fluidized bed for which the maximum values of permeate flux were reached is approximately 0.8.


2021 ◽  
Vol 40 (1) ◽  
pp. 551-563
Author(s):  
Liqiong Lu ◽  
Dong Wu ◽  
Ziwei Tang ◽  
Yaohua Yi ◽  
Faliang Huang

This paper focuses on script identification in natural scene images. Traditional CNNs (Convolution Neural Networks) cannot solve this problem perfectly for two reasons: one is the arbitrary aspect ratios of scene images which bring much difficulty to traditional CNNs with a fixed size image as the input. And the other is that some scripts with minor differences are easily confused because they share a subset of characters with the same shapes. We propose a novel approach combing Score CNN, Attention CNN and patches. Attention CNN is utilized to determine whether a patch is a discriminative patch and calculate the contribution weight of the discriminative patch to script identification of the whole image. Score CNN uses a discriminative patch as input and predict the score of each script type. Firstly patches with the same size are extracted from the scene images. Secondly these patches are used as inputs to Score CNN and Attention CNN to train two patch-level classifiers. Finally, the results of multiple discriminative patches extracted from the same image via the above two classifiers are fused to obtain the script type of this image. Using patches with the same size as inputs to CNN can avoid the problems caused by arbitrary aspect ratios of scene images. The trained classifiers can mine discriminative patches to accurately identify some confusing scripts. The experimental results show the good performance of our approach on four public datasets.


Author(s):  
Sebastian Alexander Pérez Cortés ◽  
Yerko Rafael Aguilera Carvajal ◽  
Juan Pablo Vargas Norambuena ◽  
Javier Antonio Norambuena Vásquez ◽  
Juan Andrés Jarufe Troncoso ◽  
...  

Sign in / Sign up

Export Citation Format

Share Document