Analysis of PPCI mode and multi-objective comprehensive optimization for a dual-fuel engine

Abstract

The aim of this research is to find a favorable combination of the complex control parameters of a diesel/natural gas dual-fuel engine, so that polluting emissions can meet regulatory requirements and achieve relatively low fuel consumption. This research covers the effects of control parameters on the performance and emissions of a dual-fuel engine for different combustion modes as well as discussions on the optimization of the entire operating conditions in the partial premixed compression combustion ignition (PPCI) mode combined with long short-term memory networks (LSTM) and non-dominated sorting genetic algorithms (NSGA-II). LSTM is used for the first time to establish a predictive regression model for the calibration parameters and output response of a dual-fuel engine, achieving the optimized mapping of input and output parameters. All of the optimization scenarios can achieve a trade-off between fuel consumption and emissions. The engine test results show that for the entire working conditions for the optimized dual-fuel engine, the nitrogen oxides (NOx) emissions are reduced by 77%, and the fuel consumption is roughly equal to that of the original diesel engine. The test results of the four condition test cycle of a marine engine for propulsion show that the NOx emissions of the optimized dual-fuel engine are reduced by 26.6% compared with the Tier-III legal requirements.