Matsuzaki Ryosuke, Tachikawa Takeshi, Ishizuka Junya
Department of Mechanical Engineering, Tokyo University of Science, 2641 Yamazaki, Noda, Chiba 278-8510, Japan.
Heliyon. 2018 Mar 1;4(3):e00554. doi: 10.1016/j.heliyon.2018.e00554. eCollection 2018 Mar.
Accurate simulations of carbon fiber-reinforced plastic (CFRP) molding are vital for the development of high-quality products. However, such simulations are challenging and previous attempts to improve the accuracy of simulations by incorporating the data acquired from mold monitoring have not been completely successful. Therefore, in the present study, we developed a method to accurately predict various CFRP thermoset molding characteristics based on data assimilation, a process that combines theoretical and experimental values. The degree of cure as well as temperature and thermal conductivity distributions during the molding process were estimated using both temperature data and numerical simulations. An initial numerical experiment demonstrated that the internal mold state could be determined solely from the surface temperature values. A subsequent numerical experiment to validate this method showed that estimations based on surface temperatures were highly accurate in the case of degree of cure and internal temperature, although predictions of thermal conductivity were more difficult.
准确模拟碳纤维增强塑料（CFRP）成型对于开发高质量产品至关重要。然而，此类模拟具有挑战性，并且之前通过纳入从模具监测获取的数据来提高模拟准确性的尝试并未完全成功。因此，在本研究中，我们开发了一种基于数据同化的方法来准确预测各种CFRP热固性成型特性，数据同化是一种将理论值与实验值相结合的过程。利用温度数据和数值模拟估算了成型过程中的固化程度以及温度和热导率分布。初步数值实验表明，仅根据表面温度值就可以确定模具内部状态。随后验证该方法的数值实验表明，尽管热导率的预测更困难，但基于表面温度的固化程度和内部温度估算非常准确。
