A NEW SINGLE-ZONE MULTI-STAGE SCAVENGING MODEL FOR REAL-TIME EMISSIONS CONTROL IN TWO-STROKE ENGINES

被引:0
|
作者
Bajwa, Abdullah U. [1 ]
Patterson, Mark [2 ]
Linker, Taylor [1 ]
Jacobs, Timothy J. [1 ]
机构
[1] Texas A&M Univ, College Stn, TX 77843 USA
[2] Baker Hughes GE Co, Houston, TX USA
关键词
D O I
暂无
中图分类号
TH [机械、仪表工业];
学科分类号
0802 ;
摘要
Gas exchange processes in two-stroke internal combustion engines, i.e. scavenging, remove exhaust gases from the combustion chamber and prepare the fuel-oxidizer mixture that undergoes combustion. A non-negligible fraction of the mixture trapped in the cylinder at the conclusion of scavenging is composed of residual gases from the previous cycle. This can cause significant changes to the combustion characteristics of the mixture by changing its composition and temperature, i.e. its thermodynamic state. Thus, it is vital to have accurate knowledge of the thermodynamic state of the post-scavenging mixture to be able to reliably predict and control engine performance, efficiency and emissions. Several simple-scavenging models can be found in the literature that - based on a variety of idealized interaction modes between incoming and cylinder gases - calculate the state of the trapped mixture. In this study, boundary conditions extracted from a validated 1-D predictive model of a single-cylinder two-stroke engine are used to gauge the performance of four simple scavenging models. It is discovered that the assumption of thermal homogeneity of the incoming and exiting gases is a major source of inaccuracy. A new non-isothermal multi-stage single-zone scavenging model is thus, proposed to address some of the shortcomings of the four models. The proposed model assumes that gas-exchange in cross-scavenged two-stroke engines takes place in three stages; an isentropic blowdown stage, followed by perfect-displacement and perfect-mixing stages. Significant improvements in the trapped mixture state estimates were observed as a result.
引用
收藏
页数:17
相关论文
共 23 条
  • [21] A real-time two-input stream multi-column multi-stage convolution neural network (TIS-MCMS-CNN) for efficient crowd congestion-level analysis
    Santosh Kumar Tripathy
    Rajeev Srivastava
    [J]. Multimedia Systems, 2020, 26 : 585 - 605
  • [22] A real-time two-input stream multi-column multi-stage convolution neural network (TIS-MCMS-CNN) for efficient crowd congestion-level analysis
    Tripathy, Santosh Kumar
    Srivastava, Rajeev
    [J]. MULTIMEDIA SYSTEMS, 2020, 26 (05) : 585 - 605
  • [23] Implementation and Validation of an Event-Based Real-Time Nonlinear Model Predictive Control Framework with ROS Interface for Single and Multi-robot Systems
    Dentler, Jan
    Kannan, Somasundar
    Olivares-Mendez, Miguel A.
    Voos, Holger
    [J]. 2017 IEEE CONFERENCE ON CONTROL TECHNOLOGY AND APPLICATIONS (CCTA 2017), 2017, : 1000 - 1006