CNN-Based Virtual Sensing of Rate of Heat Release for Fuel-Robust Diesel Engine Control (First Report) Rate of Heat Release Estimation and Improvement in Estimation Accuracy
畳み込みニューラルネットワークモデルを用いたディーゼルエンジンの熱発生制御 （第1報）熱発生率推定と精度向上
- Delivery
- Available on this site
- Format
- Price
- Non-members (tax incl.)：￥1,100 Members (tax incl.)：￥880
- Publication code
- 20266320
- Paper/Info type
- Proceedings (Autumn)
No.159-26
- Pages
- 1-2(Total 2 p)
- Date of publication
- Oct 2026
- Publisher
- JSAE
- Language
- Japanese
- Event
- 2026 JSAE Annual Congress (Autumn)
Detailed Information
| Author(J) | 1) 関口 亮, 2) 岡本 毅, 3) 小澤 恒, 4) 栁内 滉洋, 5) Ratnak Sok, 6) 草鹿 仁 |
|---|---|
| Author(E) | 1) Ryo Sekiguchi, 2) Takeshi Okamoto, 3) Hisashi Ozawa, 4) Akihiro Yanai, 5) Ratnak Sok, 6) Jin Kusaka |
| Affiliation(J) | 1) いすゞ中央研究所, 2) いすゞ中央研究所, 3) いすゞ中央研究所, 4) 早稲田大学, 5) 早稲田大学, 6) 早稲田大学 |
| Affiliation(E) | 1) Isuzu Advanced Engineering Center, 2) Isuzu Advanced Engineering Center, 3) Isuzu Advanced Engineering Center, 4) Waseda University, 5) Waseda University, 6) Waseda University |
| Abstract(J) | カーボンニュートラル燃料の多様化に対応するため，燃料特性によらず目標性能を実現する燃焼制御が求められる．本研究では，CNNを用いた熱発生率推定に基づく制御ロジックを構築し，セタン価の異なる軽油を用い，筒内圧力とクランク角加速度をそれぞれモデル入力として，推定精度への影響を評価した． Translation |
| Abstract(E) | To enable the future use of diverse carbon-neutral fuels, robust combustion control is required to achieve target engine performance regardless of fuel properties, without fuel-specific engine calibration. In this study, RHR-based combustion control logic for feedback correction of engine control variables was developed using a convolutional neural network (CNN) -based rate of heat release (RHR) estimation method. The accuracy of RHR estimation was evaluated for diesel fuels with different cetane numbers using in-cylinder pressure and rotary-encoder-derived crank-angle acceleration. The results clarify the influence of training conditions and image resolution on estimation accuracy and demonstrate the feasibility of fuel-robust combustion control. |
