ANN-based modeling and reducing dual-fuel engine's challenging emissions by multi-objective evolutionary algorithm NSGA-II. (1st August 2016)

• Record Type: Journal Article
• Authors: Lotfan, S.; Ghiasi, R. Akbarpour; Fallah, M.; Sadeghi, M.H.
• Is Part Of: Applied energy. Volume 175(2016)
• Page Start: 91
• Page End: 99
• Publication Date: 2016-08-01
• Subjects: Artificial neural networks -- Non-dominated sorting genetic algorithm -- Dual-fuel engine -- Emission reduction; Power (Mechanics) -- Periodicals; Energy conservation -- Periodicals; Energy conversion -- Periodicals; 621.042
• DOI: 10.1016/j.apenergy.2016.04.099
• Languages: English
• ISSNs: 0306-2619

Abstract

Highlights: A turbocharged heavy duty 4 stroke direct injection dual-fuel engine is considered. CO and NO x are predicted with training correlation factors of 0.9969 and 0.9953. The Pareto-optimal CO and NO x reduction shows their negatively correlated nature. The optimum percentage of the gaseous fuel reduces the CO and NO x, simultaneously. Abstract: In this study, the combination of artificial neural network (ANN) and non-dominated sorting genetic algorithm II (NSGA-II) has been implemented for modeling and reducing CO and NO x emissions from a direct injection dual-fuel engine. A multi-layer perceptron (MLP) network is developed to predict the values of the emissions based on experimental data. The controllable variables such as engine speed, output power, intake temperature, mass flow rate of diesel fuel, and mass flow rate of the gaseous fuel are considered as input parameters. In order to identify the uncertainties due to the experiments and the ANN-based model, uncertainty analysis is carried out. Finally, optimum values of intake temperature, mass flow rate of diesel and gaseous fuels are obtained for a desired output power and engine speed via NSGA-II. The use of the developed evolutionary optimization algorithm allows the calculation of the Pareto-optimal set of designs under any combination of engine speed and output power, defined in the range of the experiments.