Development of High-speed Prediction System using Surrogate Models for Magnet Wire Forming
コイル線成形CAEのサロゲートモデルを用いた高速化技術開発
- Delivery
- Available on this site
- Format
- Price
- Non-members (tax incl.)：￥1,100 Members (tax incl.)：￥880
- Publication code
- 20246326
- Paper/Info type
- Proceedings (Autumn)
No.154-24
- Pages
- 1-4(Total 4 p)
- Date of publication
- Oct 2024
- Publisher
- JSAE
- Language
- Japanese
- Event
- 2024 JSAE Annual Congress (Autumn)
Detailed Information
| Author(J) | 1) 中野 慎太郎, 2) 寺部 俊紀, 3) Le Dinh Thanh |
|---|---|
| Author(E) | 1) Shintaro Nakano, 2) Toshiki Terabe, 3) Le Dinh Thanh |
| Affiliation(J) | 1) トヨタ自動車, 2) トヨタ自動車, 3) トヨタ自動車 |
| Affiliation(E) | 1) Toyota Motor, 2) Toyota Motor, 3) Toyota Motor |
| Abstract(J) | 電動車用モータのコイル線成形時の加工による膜厚減少を予測するため，基礎実験とCAEを組み合わせて予測技術を開発．更にCAEの高速化としてCNNを用いてサロゲートモデルを構築し計算時間短縮を実現．性能CAEと同期したモータ開発による初期設計素性の向上により，開発期間短縮に貢献できる． Translation |
| Abstract(E) | This study presents a high-speed prediction system using surrogate models for coiled wire forming. First, we confirmed that CAE results coincide with the real. Next, we constructed the surrogate model using a Convolutional Neural Network (CNN) and CAE data taken from CAE results. The surrogate models can provide prediction on strain distribution within 10% error, while reduce computation time from days to seconds. It is highly expected that using such surrogate models, motor development period can be significantly shortened. |
