Development of an ANN based system identification tool to estimate the performance-emission characteristics of a CRDI assisted CNG dual fuel diesel engine

Sumit Roy; Rahul Banerjee; Ajoy Kumar Das; Probir Kumar Bose

doi:10.1016/j.jngse.2014.08.002

Development of an ANN based system identification tool to estimate the performance-emission characteristics of a CRDI assisted CNG dual fuel diesel engine

Sumit Roy, Rahul Banerjee, Ajoy Kumar Das, Probir Kumar Bose

NIT Agartala

Research output: Contribution to journal › Article › peer-review

78 Citations (Scopus)

Abstract

In the present study the performance and emission parameters of a single cylinder four-stroke CRDI engine under CNG-diesel dual-fuel mode have been modeled by Artificial Neural Network. An ANN model was developed to predict BSFC, BTE, NO_x, PM and HC based on the experimental data, with load, fuel injection pressure and CNG energy share as input parameters for the network. The developed ANN model was capable of predicting the performance and emission parameters with commendable accuracy as observed from correlation coefficients within the range of 0.99833-0.99999, mean absolute percentage error in the range of 0.045-1.66% along with noticeably low root mean square errors provided an acceptable index of the robustness of the predicted accuracy.

Original language	English
Pages (from-to)	147-158
Number of pages	12
Journal	Journal of Natural Gas Science and Engineering
Volume	21
Early online date	24 Aug 2014
DOIs	https://doi.org/10.1016/j.jngse.2014.08.002
Publication status	Published - Nov 2014
Externally published	Yes

Keywords

Artificial neural network
CNG
CRDI
Dual-fuel
Engine performance
Exhaust emissions

ASJC Scopus subject areas

Fuel Technology
Geotechnical Engineering and Engineering Geology
Energy Engineering and Power Technology

Access to Document

10.1016/j.jngse.2014.08.002

Cite this

@article{e857233c41fa4f9683080b4049915416,

title = "Development of an ANN based system identification tool to estimate the performance-emission characteristics of a CRDI assisted CNG dual fuel diesel engine",

abstract = "In the present study the performance and emission parameters of a single cylinder four-stroke CRDI engine under CNG-diesel dual-fuel mode have been modeled by Artificial Neural Network. An ANN model was developed to predict BSFC, BTE, NOx, PM and HC based on the experimental data, with load, fuel injection pressure and CNG energy share as input parameters for the network. The developed ANN model was capable of predicting the performance and emission parameters with commendable accuracy as observed from correlation coefficients within the range of 0.99833-0.99999, mean absolute percentage error in the range of 0.045-1.66\% along with noticeably low root mean square errors provided an acceptable index of the robustness of the predicted accuracy.",

keywords = "Artificial neural network, CNG, CRDI, Dual-fuel, Engine performance, Exhaust emissions",

author = "Sumit Roy and Rahul Banerjee and Das, \{Ajoy Kumar\} and Bose, \{Probir Kumar\}",

year = "2014",

month = nov,

doi = "10.1016/j.jngse.2014.08.002",

language = "English",

volume = "21",

pages = "147--158",

journal = "Journal of Natural Gas Science and Engineering",

issn = "1875-5100",

publisher = "Elsevier B.V.",

}

TY - JOUR

T1 - Development of an ANN based system identification tool to estimate the performance-emission characteristics of a CRDI assisted CNG dual fuel diesel engine

AU - Roy, Sumit

AU - Banerjee, Rahul

AU - Das, Ajoy Kumar

AU - Bose, Probir Kumar

PY - 2014/11

Y1 - 2014/11

N2 - In the present study the performance and emission parameters of a single cylinder four-stroke CRDI engine under CNG-diesel dual-fuel mode have been modeled by Artificial Neural Network. An ANN model was developed to predict BSFC, BTE, NOx, PM and HC based on the experimental data, with load, fuel injection pressure and CNG energy share as input parameters for the network. The developed ANN model was capable of predicting the performance and emission parameters with commendable accuracy as observed from correlation coefficients within the range of 0.99833-0.99999, mean absolute percentage error in the range of 0.045-1.66% along with noticeably low root mean square errors provided an acceptable index of the robustness of the predicted accuracy.

AB - In the present study the performance and emission parameters of a single cylinder four-stroke CRDI engine under CNG-diesel dual-fuel mode have been modeled by Artificial Neural Network. An ANN model was developed to predict BSFC, BTE, NOx, PM and HC based on the experimental data, with load, fuel injection pressure and CNG energy share as input parameters for the network. The developed ANN model was capable of predicting the performance and emission parameters with commendable accuracy as observed from correlation coefficients within the range of 0.99833-0.99999, mean absolute percentage error in the range of 0.045-1.66% along with noticeably low root mean square errors provided an acceptable index of the robustness of the predicted accuracy.

KW - Artificial neural network

KW - CNG

KW - CRDI

KW - Dual-fuel

KW - Engine performance

KW - Exhaust emissions

UR - https://www.scopus.com/pages/publications/84906482966

U2 - 10.1016/j.jngse.2014.08.002

DO - 10.1016/j.jngse.2014.08.002

M3 - Article

AN - SCOPUS:84906482966

SN - 1875-5100

VL - 21

SP - 147

EP - 158

JO - Journal of Natural Gas Science and Engineering

JF - Journal of Natural Gas Science and Engineering

ER -

Development of an ANN based system identification tool to estimate the performance-emission characteristics of a CRDI assisted CNG dual fuel diesel engine

Abstract

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this