Title :
Simulation on the Performance of Ceramic-Lined Steel Pipe Prepared by SHS Process Based on Artificial Neural Network
Author :
Yu Zhu ; Yuxi Ge ; Feng Huang ; Hongjun Ni
Author_Institution :
Sch. of Mech. Eng., Nantong Univ., Nantong, China
Abstract :
In order to study the relationship between reaction recipe and the performance of ceramic-lined steel pipe prepared by SHS process, 21 groups data obtained in the experiment were used. the different reaction recipes were taken as input data. Besides, the crushing strength and the density of ceramic layer were taken as output data. the BP neural network model was established to simulate the performance of ceramic lined composite steel pipe under different reaction recipes. Simulation results show that: the use of BP neural network simulation of ceramic lined composite tube crushing strength and the density of the steel pipe ceramic layer maximum error of 2.6742% and 4.8445%.It meets the needs in the engineering.
Keywords :
backpropagation; ceramics; combustion synthesis; composite materials; compressive strength; high-temperature techniques; mechanical engineering computing; neural nets; pipes; steel; BP neural network model; artificial neural network-based SHS process; ceramic layer density; ceramic lined composite steel pipe; ceramic lined composite tube crushing strength; reaction recipe; self-propagating high temperature synthesis; Biological neural networks; Ceramics; Data models; Neurons; Steel; Training; composite pipe; performance simulating; self-propagating high-temperature synthesis; the BP neural network;
Conference_Titel :
Computational Intelligence and Design (ISCID), 2012 Fifth International Symposium on
Conference_Location :
Hangzhou
Print_ISBN :
978-1-4673-2646-9
DOI :
10.1109/ISCID.2012.19
