A GA-based Comparative Study of DI Diesel Engine Emission and Performance Using a Neural Network Model

In diesel engines, applying design techniques such as computer simulations has become a necessity in view of the fact that these methods can result in small amounts of NOx and SOOT and a reasonable fuel economy. To achieve such a target, multi-objective optimization methodology is a good choice In this paper, this technique is implemented on a closed cycle two-zone combustion model of a DI diesel engine. The combustion model is developed by Matlab programming and validated by a single cylinder Ricardo data obtained from the engine. The main outputs of this model are NOx, SOOT and engine performance. The optimization goal is to minimize NOx and SOOT at the same time while maximizing engine performance. Injection timing, injection duration and AFR (Air-fuel ratio) are selected from engine inputs as design variables. A neural network model of the engine is developed based on model data as an alternative for the complicated and time-consuming combustion model in a wide range of engine operation. Design variables are optimized using GA (Genetic Algorithm). Here, three common algorithms for multi-objective optimization, MOGA, NSGA-II, and SPEA2+ are applied and the results are compared.

[1]  Carlos A. Coello Coello,et al.  Handling preferences in evolutionary multiobjective optimization: a survey , 2000, Proceedings of the 2000 Congress on Evolutionary Computation. CEC00 (Cat. No.00TH8512).

[2]  Lothar Thiele,et al.  Multiobjective Optimization Using Evolutionary Algorithms - A Comparative Case Study , 1998, PPSN.

[3]  J. Ringuest.  Multiobjective Optimization: Behavioral and Computational Considerations , 1992 .

[4]  R. K. Ursem.  Multi-objective Optimization using Evolutionary Algorithms , 2009 .

[5]  Sang Woo Kim,et al.  Detection of line defects in steel billets using undecimated wavelet transform , 2008, 2008 International Conference on Control, Automation and Systems.

[6]  Dimitrios C. Kyritsis,et al.  Development and validation of a comprehensive two‐zone model for combustion and emissions formation in a DI diesel engine , 2003 .

[7]  Tomoyuki Hiroyasu,et al.  SPEA2+: Improving the Performance of the Strength Pareto Evolutionary Algorithm 2 , 2004, PPSN.

[8]  Tomoyuki Hiroyasu,et al.  Comparison study of SPEA2+, SPEA2, and NSGA-II in diesel engine emissions and fuel economy problem , 2005, 2005 IEEE Congress on Evolutionary Computation.

[9]  Tomoyuki Hiroyasu,et al.  Reduction of Heavy Duty Diesel Engine Emission and Fuel Economy with Multi-Objective Genetic Algorithm and Phenomenological Model , 2004 .

[10]  Zbigniew Michalewicz,et al.  Evolutionary Computation 2 , 2000 .

[11]  Youngjean Jung,et al.  Robust optical measurement of water vapor in steel-making process , 2008, 2008 International Conference on Control, Automation and Systems.

[12]  Kalyanmoy Deb,et al.  Muiltiobjective Optimization Using Nondominated Sorting in Genetic Algorithms , 1994, Evolutionary Computation.

[13]  H. Hiroyasu,et al.  Development and Use of a Spray Combustion Modeling to Predict Diesel Engine Efficiency and Pollutant Emissions : Part 1 Combustion Modeling , 1983 .

[14]  Aravind Seshadri,et al.  A FAST ELITIST MULTIOBJECTIVE GENETIC ALGORITHM: NSGA-II , 2000 .

[15]  Erick Cantú-Paz,et al.  A Survey of Parallel Genetic Algorithms , 2000 .

[16]  J. Heywood,et al.  Experimental and Theoretical Study of Nitric Oxide Formation in Internal Combustion Engines , 1970 .

[17]  Kevin N. Gurney,et al.  An introduction to neural networks , 2018 .

[18]  William H. Lipkea,et al.  Direct injection diesel engine soot modeling: Formulation and results , 1994 .

[19]  村田 忠彦,et al.  Genetic algorithms for multi-objective optimization , 1997 .

[20]  Kalyanmoy Deb,et al.  A fast and elitist multiobjective genetic algorithm: NSGA-II , 2002, IEEE Trans. Evol. Comput..