Modeling of CVI process in fabrication of carbon/carbon composites by an artificial neural network  

作　　者：李爱军 李贺军 李克智 顾正彬 

机构地区：[1]Superhigh Temperature Composites Key Laboratory, Northwestern Polytechnical University

出　　处：《Science China(Technological Sciences)》2003年第2期173-181,共9页中国科学（技术科学英文版）

基　　金：supported by the National Natural Science Foundation of China(Grant No.50072019);the Aeronautical Foundation of China under Grant No.99G53092

摘　　要：The chemical vapor infiltration(CVI) process in fabrication of carbon-carbon composites is very complex and highly inefficient, which adds considerably to the cost of fabrication and limits the application of the material. This paper tries to use a supervised artificial neural network(ANN) to model the nonlinear relationship between parameters of isothermal CVI(ICVI) processes and physical properties of C/C composites. A model for preprocessing dataset and selecting its topology is developed using the Levenberg-Marquardt training algorithm and trained with comprehensive dataset of tubal C/C components collected from experimental data and abundant simulated data obtained by the finite element method. A basic repository on the domain knowledge of CVI processes is established via sufficient data mining by the network. With the help of the repository stored in the trained network, not only the time-dependent effects of parameters in CVI processes but also their coupling effects can be analyzed and predicted. The results show that the ANN system is effective and successful for optimizing CVI processes in fabrication of C/C composites.The chemical vapor infiltration(CVI) process in fabrication of carbon-carboncomposites is very complex and highly inefficient, which adds considerably to the cost offabrication and limits the application of the material. This paper tries to use a supervisedartificial neural network(ANN) to model the nonlinear relationship between parameters of isothermalCVI(ICVI) processes and physical properties of C/C composites. A model for preprocessing dataset andselecting its topology is developed using the Levenberg-Marquardt training algorithm and trainedwith comprehensive dataset of tubal C/C components collected from experimental data and abundantsimulated data obtained by the finite element method. A basic repository on the domain knowledge ofCVI processes is established via sufficient data mining by the network. With the help of therepository stored in the trained network, not only the time-dependent effects of parameters in CVIprocesses but also their coupling effects can be analyzed and predicted. The results show that theANN system is effective and successful for optimizing CVI processes in fabrication of C/Ccomposites.

关 键 词：C/C composites  ICVI process  artificial neural network  Levenberg-Marquard algorithm  FINITE element method. 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程]