Reduced Order Modeling of CFD Model Using Machine Learning and an Application for Heat Damage Evaluation
機械学習を用いたCFDモデル低次元化の熱害検討への適用
- Delivery
- Available on the other site
- Click here to order.
- Publication code
- 20234221
- Paper/Info type
- JSAE Transaction
Vol.54 No.3
- Pages
- 658-663(Total 6 p)
- Date of publication
- May 2023
- Publisher
- JSAE
- Language
- Japanese
Detailed Information
Category(J) | 論文 Translation |
---|---|
Category(E) | Paper |
Author(J) | 1) 河合 悠奈, 2) 新谷 浩平, 3) 菅井 友駿, 4) 笹川 崇 |
Author(E) | 1) Haruna Kawai, 2) Kohei Shintani, 3) Tomotaka Sugai, 4) Takashi Sasagawa |
Affiliation(J) | 1) トヨタ自動車, 2) トヨタ自動車, 3) トヨタ自動車, 4) 豊田中央研究所 |
Abstract(J) | 本研究では,CFD計算コスト削減のためCFDサロゲートモデルの構築手法を提案した.提案手法では,CFD結果からテンソルを構成しタッカー分解を用いて特徴量を抽出した.次に,ガウス過程回帰を用いて設計変数を特徴量との間の回帰モデルを作成した.提案手法を熱害のためのCFDモデルに適用し,技術検証を行った. Translation |
Abstract(E) | The purpose of this paper is to propose a method to construct a surrogate model which can predict flow filed of velocity and temperature aiming at decrease the computational cost of CFD. In the proposed method, training data are corrected from CFD simulation based on a Design of Experiments (DOE). Then, after performing missing value interpolation on the training data, Tucker decomposition is applied to training data to extract features from the tensor type training data. For regression model, Gaussian process is introduced to construct surrogate models. The feasibility of the proposed method is illustrated by an application for CFD model using for the heat damage design problem. |