人工神经网络用于炭/炭复合材料CVDSiC涂层的研究  

STUDY ON APPLICATION OF ARTIFICIAL NEURAL NETWORK TO THE CVD-SiC PROCESS FOR COATING ON CARBON-CARBON COMPOSITES

作　　者：李贺军[1] 徐志淮[1] 

机构地区：[1]西北工业大学炭/炭复合材料研究所

出　　处：《炭素技术》1999年第2期17-20,共4页Carbon Techniques

基　　金：国防基金;航空基金

摘　　要：炭／炭复合材料抗氧化涂层是该材料可否用于高温结构件的关键。将人工神经网络用于ＣＶＤＳｉＣ抗氧化涂层的建模，试图解决该过程影响因素复杂、数学模型难于建立、无法对整个过程进行有效预测和控制的问题。初步探讨表明，采用人工神经网络建模，可以比较确切与全面地反映ＣＶＤＳｉＣ涂层过程的影响因素和内在规律，预测的工艺参数与沉积速率的关系同实验结果吻合较好。Anti-oxidation coatings are essential for carbon-carbon composites utilized as thermal structure materialsAmong many kinds of coatings,the CVD-SiC coating is one really utilizedHowever the coating methods are generally influenced by many complicated factors,thus its mathematical model cant be deduced exactly,and the whole process cant be predicted and controlled effectivelyThe artificial neural network(ANN) is used for establishing the process model of the CVD-SiC coating for the first timeThe preliminary research result shows that the method is effective and feasibleThe model established by ANN reveals the objective factors and the inner regularity to process of CVD-SiC coating,and the predicted relation between the technical parameters and deposition rate is identified well with the experimental data

关 键 词：化学气相沉积 神经网络 涂层 炭/炭复合材料 CVD 

分 类 号：TB332.04[一般工业技术—材料科学与工程]