Strength Prediction Model of Aluminum Alloy Friction Stir Spot Welding Using Machine Learning -Part I-
機械学習によるアルミ合金摩擦撹拌点接合の強度予測モデル（第一報）
- Delivery
- Available on this site
- Format
- Price
- Non-members (tax incl.)：￥1,100 Members (tax incl.)：￥880
- Publication code
- 20265243
- Paper/Info type
- Proceedings (Spring)
No.57-26
- 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) 福原 俊昭, 2) 杉本 幸弘, 3) 深堀 貢, 4) 田中 耕二郎, 5) 島田 聡子, 6) 藤田 一輝 |
|---|---|
| Author(E) | 1) Toshiaki Fukuhara, 2) Yukihiro Sugimoto, 3) Mitsugi Fukahori, 4) Kojiro Tanaka, 5) Satoko Shimada, 6) Itsuki Fujita |
| Affiliation(J) | 1) マツダ, 2) 広島大学, 3) マツダ, 4) マツダ, 5) マツダ, 6) マツダ |
| Affiliation(E) | 1) Mazda, 2) Hiroshima University, 3) Mazda, 4) Mazda, 5) Mazda, 6) Mazda |
| Abstract(J) | アルミ合金接合における重要な技術として摩擦撹拌点接合（FSSW) に着目し，強度予測技術構築を目指している．FSSWの接合強度は，接合条件により変化する接合部断面形状と相関があるが，従来この関係の定量化が困難だった．そこで粒子法解析と機械学習モデルを直列的に組合わせることで，強度予測モデルを構築した． Translation |
| Abstract(E) | We focus on friction stir spot welding (FSSW) as a critical technique for joining aluminum alloys and aim to establish a methodology for predicting joint strength. The strength of FSSW joints is closely correlated with the cross-sectional geometry of the joint, which varies according to welding conditions; however, quantifying this relationship has traditionally been challenging. In this study, we developed a strength prediction model by sequentially integrating flow analysis based on a particle method with a machine learning approach. |
