Performance research and optimization of marine dual-fuel engine based on RSM and NSGA-II

Abstract

ABSTRACT In order to improve the economic efficiency of the ship when sailing and reduce the engine optimization cost. We propose the Response Surface Methodology (RSM) combined with NSGA-II to optimize the engine parameters. First, a simulation model of a marine four-stroke dual-fuel engine is established in AVL-BOOST software. Then, control parameters such as engine speed, exhaust valve opening (EVO) and compression ratio (CR) are planned by design of experiments. The response surface model was established in Design-Expert software. The significant influence of control parameters on performance parameters was studied by analysis of variance (ANOVA). Finally, with the output power, indicated fuel consumption rate and nitrogen oxide emissions as the optimization objectives. Non-dominated Sequential Genetic Algorithm (NSGA-II) is used to optimize the parameters to improve engine performance and reduce emissions. The results show that the established response surface model has good prediction accuracy. The response surface model visualizes the mathematical relationship between the control parameters and the optimization targets. The ANOVA results show that engine speed, EVO and CR have significant effects on engine performance and emissions. The optimization results show that the engine speed is 793 rpm, the EVO is 145°CA, and the CR is 12.3. Compared to standard settings, the optimized data shows a 3.4% increase in power, a 0.3% reduction in ISFC, and a 6.2% reduction in nitrogen oxide (NOx) emissions. The combination of response surface analysis and NSGA-II algorithm to optimize engine performance and emissions is thereby a feasible method.