Journal of Zhejiang University SCIENCE
(ISSN 1009-3095, Monthly)
2002   Vol. 3   No. 5   p.543-548

A method for predicting in-cylinder compound combustion emissions

SU Shi-chuan(苏石川)(The Institute Power Machinery and Vehicular Engineering, Zhejiang University, Hangzhou 310027, China)  
YAN Zhao-da(严兆大)(The Institute Power Machinery and Vehicular Engineering, Zhejiang University, Hangzhou 310027, China)  
YUAN Guang-jie(元广杰)(The Institute Power Machinery and Vehicular Engineering, Zhejiang University, Hangzhou 310027, China)  
CAO Yun-hua(曹韵华)(The Institute Power Machinery and Vehicular Engineering, Zhejiang University, Hangzhou 310027, China)  
ZHOU Chong-guang(周重光)(The Institute Power Machinery and Vehicular Engineering, Zhejiang University, Hangzhou 310027, China)  

Abstract:This paper presents a method using a large steady-state engine operation data matrix to provide necessary information for successfully training a predictive network, while at the same time eliminating errors produced by the dispersive effects of the emissions measurement system. The steady-state training conditions of compound fuel allow for the correlation of time-averaged in-cylinder combustion variables to the engine-out NOx and HC emissions. The error back-propagation neural network (EBP) is then capable of learning the relationships between these variables and the measured gaseous emissions, and then interpolating between steady-state points in the matrix. This method for NOx and HC has been proved highly successful.
Keywords:Back-propagation neural network (EBP), Compound fuel, Emissions, Prediction

CLC Number:TK421+.5  Document ID:A

Author Resume:SU Shi-chuan(苏石川)E-mail: ssczju@hotmail.com

References:

[1]Atkinson, C., Long, T., Hanzevack, E. 1998. Virtual sensing: an emissions prediction system for on-board diagnostics and engine control. SAE Paper 980516.
[2], Hafner, M., SchuK ler, M., Nelles, O., Isermann, R., 2000. Fast neural networks for diesel engine control design. Control Engineering Practice 8:1211-1221.
[3]Haykin, S., 1994. Neural Networks: AComprehensive Foundation.Macmillan College Publishing Company, Inc., Englewood Cliffs, New Jercy.
[4]Hornik, K., Stinchcombe, M., White, H., 1989, Neural networks. Int. J. Control, 2:359.
[5]Nelles, O., 1997. Orthonormal basis functions for nonlinear system identification with local linear model trees (LOLIMOT). Proceedings of the 11th IFAC Symposium on System Identification (SYSID). Fukuoka, Japan, p.667-672.
[6]Raina, P., 1994. Comparison of learning and generalization capabilities of the Kak and the backpropagation algorithms. Information Sciences, 81:261-274.
[7]Watanabe, S., Machida, K., Iijima, K., Tomisawa, N., 1996. A sophisticated engine control system using combustion pressure detection. SAE Paper: 960042.
[8]Widrow, B., Lehr, M.A.,1990. 30 years of adaptive neural networks: Perception, Madaline, and Back propagation, Proceedings of IEEE, 78(9):1415-1442.
[9]Yuan, G.J., 2000. Test Study the Performance of Emissions of Compound fuel Combustion in Diesel Engine. Thesis Requirement for the Degree of Master of Engineering, Zhejiang University.


Manuscript Received:2001 Oct. 8

Manuscript Revised:2002 Jan. 20

Published:2002 Dec. 1