Effect of Different Refractory Structures on the Flammability of Fiber Reinforced Plastics

Yusuke Ishihara･Asami Nakai･Masayuki Okoshi (Gifu University)･Atsushi Yuki･Hiroto Shigeta･Haruhiko Nakamura (DaikyoNishikawa)･Shuhei Yasuda･Junichi Ogawa･Yuki Yamada (Mazda)･Masahiko Shigetsu (Hiroshima University)

Installation of electric components for electrification of automobiles has led to a demand for improved fire resistance of FRP. The objective of this study was to improve the fire resistance of hybrid molded products in which the resin base material is covered with a continuous fiber base material. Hybrid structures were fabricated by selecting materials and surface treatments to provide fire resistance, and their effects on flammability were evaluated by combustion tests.

Strength Improvement of CFRTP by Controlling Fiber Orientation (Effect of Product Shape on Fiber Orientation)

Mutsuki Hamada･Souichiro Nishino (Ibaraki University)･Hidemaru Sootome･Kenta Iwasawa (Industrial Technology Innovation Center of Ibaraki Prefecture)

CFRTP, with its light weight and high rigidity, is finding an increasing number of applications as a structural material. On the other hand, it has been pointed out that disruption of the orientation of carbon fibers during injection molding is a factor in strength reduction. In this study, we report the results of our investigation of the effects of fiber orientation inside injection-molded products and product shape, based on strength evaluations and internal observations using CT scans.

Material Data Enrichment with Machine Learning for the Orthotropic Material Property of the FRP

Masakazu Takeuchi (Celanese)

When designing parts using fiber reinforced plastics, it is necessary to take into account the anisotropic properties caused by fiber orientation, but the burden of measuring the properties required for material modeling is not small, and it is difficult to prepare for all candidate material grades. Therefore, we report the results of our investigation into the use of machine learning to predict anisotropic property values ​​from uniaxial tensile test results.

Optimization for thickness design of polymer parts using machine learning

Hiroshi Asayama･Hirofumi Kishi･Kazuyoshi Baba (DAIHATSU MOTOR)

We propose multiobjective optimization for thickness design of polymer parts in vehicles to achieve both lightweight and thermostability. This method is based on CAE data of polymer parts with controlled gradient of thickness to balance variation and moldability. Surrogate modeling according to this method generate Pareto front to obtain optimal solution regarding weight of polymer parts and heat-threshold temperature to deform them. We have achieved thickness design with weight saving and improved thermostability by using proposed method.