P-56 Neural networks modeling of carbon layer thickness of cross driver carbonized in fluidized bed

2007 ◽  
Vol 2007.6 (0) ◽  
pp. _P-56-1_-_P-56-5_
Author(s):  
M. Szota ◽  
J. Jasinski ◽  
T. Mrozinski ◽  
W. Napadlek
Keyword(s):  
Neural Networks ◽  
Fluidized Bed ◽  
Layer Thickness ◽  
Carbon Layer

Uniform coating of nano-carbon layer on SiOx in aggregated fluidized bed as high-performance anode material

Carbon ◽  
2019 ◽  
Vol 149 ◽  
pp. 462-470 ◽  
Author(s):  
Zhexi Xiao ◽  
Chunhui Yu ◽  
Xianqing Lin ◽  
Xiao Chen ◽  
Chenxi Zhang ◽  
...  
Keyword(s):  
Fluidized Bed ◽  
Anode Material ◽  
High Performance ◽  
Carbon Layer ◽  
Uniform Coating ◽  
Nano Carbon

Application of neural networks to mass-transfer predictions in a fast fluidized bed of fine solids

AIChE Journal ◽  
1997 ◽  
Vol 43 (7) ◽  
pp. 1684-1690 ◽  
Author(s):  
Piroz Zamankhan ◽  
Pekka Malinen ◽  
Hannu Lepomäki
Keyword(s):  
Neural Networks ◽  
Mass Transfer ◽  
Fluidized Bed

Experimental Study and Artificial Neural Networks Simulation of Activated Carbon Synthesis in Fluidized Bed Reactor

2009 ◽  
Vol 7 (1) ◽  
Author(s):  
Hassan Hashemipour ◽  
Saeid Baroutian ◽  
Esmail Jamshidi ◽  
Alireza Abazari
Keyword(s):  
Neural Networks ◽  
Artificial Neural Networks ◽  
Fluidized Bed ◽  
Fluidized Bed Reactor ◽  
Operating Conditions ◽  
Synthesis Process ◽  
Back Propagation Algorithm ◽  
Feed Forward Neural Network ◽  
Activation Temperature ◽  
Artificial Neural

In this work, thermal activation of pistachio shell was studied in a fluidized bed reactor. The effects of operating conditions on the pore development within the char particles were studied experimentally. The results showed that activation temperature, residence time, oxidizing gas type and concentration have main effects on the Iodine adsorption capacity of the product. The highest surface area was obtained using steam activation at temperature 850°C for 45 min. The synthesis process was also simulated using artificial neural networks (ANNs) to estimate the Iodine number of the product. The present work applied a Tan-sigmoid transfer function in three layers in the feed forward neural network with back propagation algorithm. The results of the network are in good agreement with the experimental data with a maximum relative deviation of 0.015%.

Sign in / Sign up

Export Citation Format

Share Document