期刊
POLYMER COMPOSITES
卷 43, 期 8, 页码 5235-5249出版社
WILEY
DOI: 10.1002/pc.26816
关键词
3D printing; biocomposites; continuous ramie fiber; dynamic mechanical properties; machine learning
资金
- Hu-Xiang Youth Talent Program [2020RC3009]
- InnovationDriven Project of Central South University [2019CX017]
- National Natural Science Foundation of China [51905555]
3D-printed continuous natural fiber reinforced biocomposites have promising prospects due to their environmental friendliness and suitable mechanical properties. In this study, the dynamic mechanical properties of 3D-printed biocomposites were investigated. Different printing parameters were found to have nonlinear and interactive influences on the dynamic strength of the biocomposites. Machine learning models were employed to predict the dynamic strength based on the printing parameters, and good agreement was found between the predictions and experimental results. The relationships between parameters, microstructural characteristics, and dynamic strength of the printed biocomposites were quantitatively analyzed.
3D-printed continuous natural fiber reinforced biocomposites have promising prospects due to their environmental friendliness and suitable mechanical properties. Understanding the dynamic mechanical properties of 3D-printed biocomposites is essential to expand their application. In this study, the continuous ramie fiber reinforced biocomposites (CRFRC) with different layer thicknesses and hatch spacings were fabricated via 3D printing technique with microstructure characterized. In addition, the dynamic strengths of 3D-printed CRFRC at four strain rates were investigated. The experimental results exhibited that the printing parameters presented nonlinear and interactive influences on the dynamic strength of CRFRC. Given this circumstance, machine learning methods were employed to link the dynamic strength of 3D-printed CRFRC with different printing parameters. The experimental data were used to train, calibrate, and validate the machine learning models. The trained models were then utilized to predict the dynamic strength of CRFRC printed using different conditions. Behaviors under multiple strain rates were investigated over the whole parameter space. A good agreement was found between experimental results and predictions. Based on the prediction results, the relationships between parameters, microstructural characteristics and dynamic strength of printed CRFRC were quantitatively analyzed.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据