论文部分内容阅读
为研究麻纤维化学成分对其增强复合材料界面性能的影响,选取麻纤维纤维素、半纤维素、果胶、木质素、水溶物、脂蜡质成分含量及回潮率作为影响因素,以麻纤维/不饱和聚酯树脂(UP)复合材料界面性能作为影响结果,构建Back Propagation(BP)神经网络的训练样本。首先,利用灰关联分析法对影响麻纤维/UP复合材料界面性能的因素进行关联度计算;其次,按照影响程度的大小进行排序,建立3层BP神经网络模型进行迭代训练;最后,预测麻纤维化学成分含量对麻纤维/UP复合材料界面性能的影响。预测结果表明:学习结束后模型的输出比较接近实测值,说明BP神经网络具有很强的学习能力,同时也证明了将BP神经网络用于麻纤维/UP复合材料界面剪切力预测的可行性;灰关联与BP神经网络联用后预测精度得到大大提高,预测误差最大可减小83.28%。
In order to study the influence of the chemical composition of hemp fiber on the interfacial properties of the composites, the content of hemp fiber, hemicellulose, pectin, lignin, water soluble matter, fat wax content and moisture regain were selected as the influencing factors. / Unsaturated Polyester Resin (UP) Interfacial Properties As a consequence, a training sample of Back Propagation (BP) neural network was constructed. Firstly, the gray relational analysis method is used to calculate the correlative degree of the factors that affect the interface properties of hemp fiber / UP composite material. Secondly, the three-layer BP neural network model is established for iterative training according to the degree of influence. Finally, Effect of Chemical Compositions on Interfacial Properties of Hemp Fiber / UP Composites. The results show that the output of the model is close to the measured value at the end of the study, which shows that BP neural network has a strong learning ability and also proved the feasibility of using BP neural network to predict the interfacial shear stress of hemp fiber / UP composites The prediction accuracy of gray association and BP neural network has been greatly improved, the prediction error can be reduced by 83.28%.