Scholarlink

<p>Variable geometry turbochargers (VGTs) are increasingly applied to gasoline engines to improve efficiency and performance, but packaging and durability constraints make direct measurement of all turbocharger states difficult in practice. This study develops a mean value model (MVM) for a 1.6&#xa0;L TGDI engine equipped with a development VGT, using only a minimal set of measurable inputs. Engine tests were conducted to obtain performance data, which were divided into modeling and validation sets. The compressor was modeled in a forward structure and the turbine in a backward structure to reflect sensor availability. Curve and surface fitting of experimental data were used to identify key model coefficients. The resulting sub-models were validated against independent data, with compressor outlet pressure and temperature achieving coefficients of determination above 85%, and turbine inlet pressure, inlet temperature, and shaft speed exceeding 95%. The results demonstrate that the proposed model can reproduce unmeasured turbocharger states with sufficient accuracy for system-level simulation and engine analysis. Future work will extend the model to transient operation and control applications for turbocharger-equipped gasoline engines.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Mean Value Modeling of Turbocharger for A 1.6 L TGDI Gasoline Engine: Experimental Identification and Validation

  • 期刊论文

  • 发表日期：

    2026年5月28日

查看全文
  • Hyunjun Kim,
  • Changhee Lee,
  • Jeongho Kang

摘要

Variable geometry turbochargers (VGTs) are increasingly applied to gasoline engines to improve efficiency and performance, but packaging and durability constraints make direct measurement of all turbocharger states difficult in practice. This study develops a mean value model (MVM) for a 1.6 L TGDI engine equipped with a development VGT, using only a minimal set of measurable inputs. Engine tests were conducted to obtain performance data, which were divided into modeling and validation sets. The compressor was modeled in a forward structure and the turbine in a backward structure to reflect sensor availability. Curve and surface fitting of experimental data were used to identify key model coefficients. The resulting sub-models were validated against independent data, with compressor outlet pressure and temperature achieving coefficients of determination above 85%, and turbine inlet pressure, inlet temperature, and shaft speed exceeding 95%. The results demonstrate that the proposed model can reproduce unmeasured turbocharger states with sufficient accuracy for system-level simulation and engine analysis. Future work will extend the model to transient operation and control applications for turbocharger-equipped gasoline engines.