Machine Learning Application to Predict Turbocharger Performance under Steady-State and Transient Conditions

Kanto Kobayashi*, Arravind Jeyamoorthy, Iku Tanabe, Rio Asakawa, Kyohei Yamaguchi, Jin Kusaka


研究成果: Conference article査読


Performance predictions of advanced turbocharged engines are becoming difficult because conventional engine models are built using performance map data of turbochargers with a proportional integral derivative (PID) controller. Improving prediction capabilities under transient test cycles or real driving conditions is a challenging task. This study applies a machine learning technique to predict turbocharger performances with high accuracy under steady-state and transient conditions. The manipulated signals of engine speed and torque created based on Compressed High-Intensity Radiated Pulse (Chirp signal) and Amplitude-modulated Pseudo-Random Binary Signal (APRBS) are used as inputs to the engine testbed. Data from the engine experiments are used as training data for the AI-based turbocharger model. High prediction accuracy of the AI turbocharger model is achieved with the co-efficient of determination in the model, and cross-validation results are higher than 0.8. Further, an integrated engine model coupled with the AI turbocharger model is developed and simulated to compare its performance result against the conventional turbocharged engine model (performance map + PID controller). The results show that the diesel engine system's performance predictions can be achieved with reasonable accuracy using the integrated engine-AI turbocharger model. The results also confirm the suitability of the proposed method to develop the advanced turbocharger model.

ジャーナルSAE Technical Papers
出版ステータスPublished - 2021 9月 5
イベントSAE 15th International Conference on Engines and Vehicles, ICE 2021 - Capri, Italy
継続期間: 2021 9月 122021 9月 16

ASJC Scopus subject areas

  • 自動車工学
  • 安全性、リスク、信頼性、品質管理
  • 汚染
  • 産業および生産工学


「Machine Learning Application to Predict Turbocharger Performance under Steady-State and Transient Conditions」の研究トピックを掘り下げます。これらがまとまってユニークなフィンガープリントを構成します。