基于BP神经网络的高温快速固化环氧树脂胶膜制备-性能模型预测  

Model Prediction of Preparation-Property of Expediting Curable Epoxy Film at High Temperature Based on Artificial Neural Network

作　　者：邵康宸[1] 史洪源[1] 陈文静 张咏军[1] SHAO Kang-chen;SHI Hong-yuan;CHEN Wen-jing;ZHANG Yong-jun(College of Aeronautical Materials Engineering,Xi’an Aeronautical Polytechnic Institute,Xi’an 710089,Shaanxi,China)

机构地区：[1]西安航空职业技术学院航空材料工程学院,陕西西安710089

出　　处：《合成材料老化与应用》2021年第5期32-34,40,共4页Synthetic Materials Aging and Application

基　　金：西安航空职业技术学院2020年度科研项目(20XHZK-12);陕西高校青年创新团队资助项目(2019-73)。

摘　　要：根据正交试验,采用BP神经网络优化工艺制备高温快速固化的环氧树脂胶膜;并对环氧树脂胶膜的结构进行了表征和分析。研究结果表明,运用BP神经网络优化环氧树脂胶膜的最佳工艺为:增韧剂聚氨酯(PU)含量为30%,固化剂双氰胺含量为10%,固化温度为230℃,固化时间为120s;模型的预测值与实验值曲线具有较高的契合度,相对误差小,可以控制最高的误差值范围不超过3%,相关系数是0.99875。采用该优化参数制备的环氧树脂胶膜固化速度快、粘结强度高。On the basis of the orthogonal test,BP neural network was used to optimize the process of preparing high temperature fast curing epoxy film,the structure of epoxy film were characterized.The results showed that the optimum technological conditions were as follows:the content of toughener PU was 30%,the content of curing agent dicyandiamide was 10%,the curing temperature was 230℃,and the curing time was 120 s.The predicted value of the model has a high agreement with the experimental value curve,the relative error is small,the maximum error range can be controlled not more than 3%,the correlation coefficient is 0.99875.The epoxy film prepared by the optimized process has the advantages of high bonding strength and fast curing speed.

关 键 词：环氧树脂 粘结强度 BP神经网络 优化 

分 类 号：Q433.437[生物学—生理学]