Approach to quality improvement of simulation model for MBD applying AI method - Introduction of time-domain data analysis using Deep Learning method -
AIによるMBDシミュレーションモデルの品質改善への取り組み －ディープラーニングによる時系列データ分析の紹介－
- Delivery
- Available on this site
- Format
- Price
- Non-members (tax incl.)：￥1,100 Members (tax incl.)：￥880
- Publication code
- 20214221
- Paper/Info type
- Journal of Society of Automotive Engineers of Japan
Vol.75 No.4
- Pages
- 64-69(Total 6 p)
- Date of publication
- Apr 2021
- Publisher
- JSAE
- Language
- Japanese
Detailed Information
| Category(J) | 試作と実験をしない仮想自動車開発へ Translation |
|---|---|
| Category(E) | Hot Topics |
| Author(J) | 1) 黒川 和彦, 2) 塚原 賢祐, 3) 矢島 祐二, 4) Hoang Pham Truc Phuong, 5) Dinh Xuan Tue, 6) Nguyen Dong Dung, 7) Dao Huu Hung |
| Author(E) | 1) Kazuhiko Kurokawa, 2) Kensuke Tsukahara, 3) Yuji Yajima, 4) Hoang Pham Truc Phuong, 5) Dinh Xuan Tu, 6) Nguyen Dong Dun, 7) Dao Huu Hung |
| Affiliation(J) | 1) MCOR, 2) MCOR, 3) MCOR, 4) MCOR, 5) MCOR, 6) MCOR, 7) MCOR |
| Abstract(J) | MBDによる制御開発において、1Dシミュレーションは応用性が高い事から、その需要増による人材不足が懸念される。そこで、この問題の解決にAIの適用を試みた。その結果、初級技術者の計算結果をAIに評価させることにより、非現実的な挙動を見つける事ができ、1Dシミュレーションモデルの精度向上の可能性が確認された。 Translation |
| Abstract(E) | MBD has become a method of major process management in software development. Furthermore, in recent advanced development, cooperating with 1D CAE is important solution for improving development efficiency. On the other hand, the many demands of related human resource is higher worldwide due to the application expended. In order to solve the issue, the availability of artificial intelligence (AI) is studied to increase 1D CAE labor efficiency. Deep learning method is convenience to validate the results simulated by little experience person and some unrealistic behaviors can be found out. The possibility of accuracy improvement of the simulation model was confirmed. |
