The Application of Artificial Neural Network in Predicting and Optimizing Power and Emissions in a Compressed Natural Gas Direct Injection Engine

2007 ◽  
Author(s):  
Wendy H. Kurniawan ◽  
Shahrir Abdullah ◽  
Zulkifli M. Nopiah ◽  
Kamaruzzaman Sopian
Keyword(s):  
Neural Network ◽  
Artificial Neural Network ◽  
Natural Gas ◽  
Direct Injection ◽  
Compressed Natural Gas ◽  
Direct Injection Engine ◽  
Artificial Neural

Artificial Neural Network-Based Prediction of Performances-Exhaust Emissions of Diesohol Piloted Dual Fuel Diesel Engine Under Varying Compressed Natural Gas Flowrates

2018 ◽  
Vol 140 (11) ◽  
Author(s):  
Abhishek Paul ◽  
Subrata Bhowmik ◽  
Rajsekhar Panua ◽  
Durbadal Debroy
Keyword(s):  
Neural Network ◽  
Artificial Neural Network ◽  
Diesel Engine ◽  
Natural Gas ◽  
Goodness Of Fit ◽  
Specific Energy Consumption ◽  
Dual Fuel ◽  
Mean Square ◽  
Compressed Natural Gas ◽  
Artificial Neural

The present study surveys the effects on performance and emission parameters of a partially modified single cylinder direct injection (DI) diesel engine fueled with diesohol blends under varying compressed natural gas (CNG) flowrates in dual fuel mode. Based on experimental data, an artificial intelligence (AI) specialized artificial neural network (ANN) model have been developed for predicting the output parameters, viz. brake thermal efficiency (Bth), brake-specific energy consumption (BSEC) along with emission characteristics such as oxides of nitrogen (NOx), unburned hydrocarbon (UBHC), carbon dioxide (CO2), and carbon monoxide (CO) emissions. Engine load, Ethanol share, and CNG strategies have been used as input parameters for the model. Among the tested models, the Levenberg–Marquardt feed-forward back propagation with three input neurons or nodes, two hidden layers with ten neurons in each layer and six output neurons, and tansig-purelin activation function have been found to the optimal model topology for the diesohol–CNG platforms. The statistical results acquired from the optimal network topology such as correlation coefficient (0.992–0.999), mean square error (MSE) (0.0001–0.0009), and mean absolute percentage error (MAPE) (0.09–2.41%) along with Nash–Sutcliffe coefficient of efficiency (NSE), Kling–Gupta efficiency (KGE), mean square relative error, and model uncertainty established itself as a real-time robust type machine learning tool under diesohol–CNG paradigms. The study also incorporated a special type of measure, namely Pearson's Chi-square test or goodness of fit, which brings up the model validation to a higher level.

Computational optimization of reactivity controlled compression ignition combustion to achieve high efficiency and clean combustion

2020 ◽  
pp. 146808742093173 ◽  
Author(s):  
Avilash Jain ◽  
Anand Krishnasamy ◽  
Pradeep V
Keyword(s):  
Neural Network ◽  
Artificial Neural Network ◽  
Fuel Injection ◽  
Direct Injection ◽  
Network Models ◽  
Compression Ignition ◽  
Neural Network Models ◽  
Reactivity Controlled Compression Ignition ◽  
Artificial Neural ◽  
Artificial Neural Network Models

One of the major limitations of reactivity controlled compression ignition is higher unburned hydrocarbon and carbon monoxide emissions and lower thermal efficiency at part load operating conditions. In the present study, a combined numerical approach using a commercial three-dimensional computational fluid dynamics code CONVERGE along with artificial neural network and genetic algorithm is presented to address the above limitation. A production light-duty diesel engine is modified to run in reactivity controlled compression ignition by replacing an existing mechanical fuel injection system with a flexible electronic port fuel injection and common rail direct injection systems. The injection schedules of port fuel injection and direct injection injectors are controlled using National Instruments port and direct injection driver modules. Upon validation of combustion and emission parameters, parametric investigations are carried out to establish the effects of direct-injected diesel fuel timing start of injection (SOI), premixed fuel ratio and intake charge temperature on the engine performance and emissions in reactivity controlled compression ignition. The results obtained show that the start of injection timing and intake charge temperature significantly influence combustion phasing, while the premixed fuel ratio controls mixture reactivity and combustion quality. By utilizing the data generated with the validated computational fluid dynamics models, the artificial neural network models are trained to predict the engine exhaust emissions and efficiency. The artificial neural network models for gross indicated efficiency and oxides of nitrogen (NOx) are then coupled with genetic algorithm to maximize gross indicated efficiency while keeping the NOx and soot emissions within Euro VI emission limits. By optimizing the start of injection timing, premixed fuel ratio and intake charge temperature simultaneously using the artificial neural network models coupled with genetic algorithm, 19% improvement in gross indicated efficiency, 60% and 64% reduction in hydrocarbon and carbon monoxide emissions, respectively, are obtained in reactivity controlled compression ignition compared to the baseline case.

Sign in / Sign up

Export Citation Format

Share Document