| 作成者 |
|
|
|
|
|
| 本文言語 |
|
| 出版者 |
|
| 発行日 |
|
| 収録物名 |
|
| 巻 |
|
| 開始ページ |
|
| 終了ページ |
|
| 会議情報 |
|
| 出版タイプ |
|
| アクセス権 |
|
| 権利関係 |
|
| 権利関係 |
|
| 概要 |
This study investigates the optimization of natural fiber-reinforced polymer composite (NFRPCs) creation for bulletproof applications by integrating computational simulation and machine learning (ML).... We incorporate abaca (Musa textilis) or pineapple leaf fibers (Piñatex), along with aramid and carbon fibers, into layered composite plates. Ballistic performance was modeled and predicted using simulated data from ANSYS Explicit Dynamics and validated through live bullet testing. ML models, such as Support Vector Machine (SVM) and Random Forest (RF) with optimized hyperparameters, achieved up to 80% prediction accuracy and an F1-score of 82% for abaca-reinforced composites, closely aligning with experimental results. However, lower prediction accuracy was observed for Piñatex-based composites, due to fiber variability and other factors identified in the study. This hybrid methodology highlights the potential of combining simulation and ML to reduce reliance on extensive live bullet testing, providing a data-driven pathway for the efficient development of high-performance bulletproof composite materials.続きを見る
|
Mendeley出力
2025_p0472