Enhancement of ISO Score Accuracy in Rear-Impact Simulations Using Multi-Objective Optimization
多目的最適化を用いた後突シミュレーションにおけるISOスコアの精度向上
- Delivery
- Available on this site
- Format
- Price
- Non-members (tax incl.):¥1,100 Members (tax incl.):¥880
- Publication code
- 20265184
- Paper/Info type
- Proceedings (Spring)
No.44-26
- Pages
- 1-4(Total 4 p)
- Date of publication
- May 2026
- Publisher
- JSAE
- Language
- Japanese
- Event
- 2026 JSAE Annual Congress (Spring)
Detailed Information
| Author(J) | 1) 植栗 友洋 |
|---|---|
| Author(E) | 1) Tomohiro Ueguri |
| Affiliation(J) | 1) SUBARU |
| Affiliation(E) | 1) SUBARU |
| Abstract(J) | 衝突安全のアセスメント評価でVirtual Testの導入が進んでおり,CAE精度の重要性が高まっている.最適化ソフトを用いて鞭打ち解析結果を活用し,精度相関評価を対象とした多目的最適化のワークフローを構築した.設計変数に対し機械学習を用いた最適化を実施し,実機試験との相関性を最大化する組み合わせを探索したので発表する. Translation |
| Abstract(E) | The introduction of Virtual Testing in crash safety assessments is accelerating, increasing the importance of CAE simulation accuracy. In this study, we developed a multi-objective optimization workflow that leverages whiplash analysis results to evaluate and improve correlation accuracy. Using an optimization platform with machine-learning–based techniques, we explored design-variable combinations that maximize correlation with physical test results. This presentation summarizes the workflow and its effectiveness. |