Skip to main navigation Skip to search Skip to main content

CFD analysis of spray and combustion for alternative fuels in a two-stroke marine engine

    Activity: Examinations and supervisionSupervisor activities

    Description

    Master thesis by Jonathan Thilo

    Abstract:

    In the recent years stricter regulations are being implemented to the marine engine emissions. To comply with the new regulations alternative fuels are combined with the conservative and robust two-stroke diesel marine engine in dual-fuel engines. This thesis investigates the two sustainable alternative fuels methanol and ammonia in a dual-fuel setup with diesel. MAN Energy Solutions has provided experimental data for a diesel-methanol dual-fuel research engine, 4T50ME-X, located in Sydhavn, Copenhagen, Denmark. CFD simulations in Simcenter Star CCM+ are setup to match the experimental setup. A multiphase Eulerian-Lagrangian approach is applied to the model with the Reitz-Diwakar secondary breakup model. The K-Omega SST model are utilized to model the turbulence, while the real gas model, Redlich-Kwong, determines the equation of state. The liquid fuels properties are defined as temperature-dependent to reflect the actual properties when heated. Adaptive mesh refinement is setup to refine the mesh in regions of interest. The reduced chemical mechanism by Müller-Elna is used to reduce the computational cost compared to the LU mechanism. The model is validated in terms of motion, initial conditions and injected fuel mass. A convergence analysis is performed on the mesh and the timestep, while the sensitivity of the turbulence model, chemical mechanism and water content is investigated. The main focal point for the study is the effect of replacing the solid cone nozzles and Rosin-Rammler particle size distribution with the nozzle injector and Huh atomization model. Another crucial part of the study is the effect on the particle breakup of adjusting the Reitz-Diwakar Cs2 coefficient value, which affects fuel particles and their resistance to breakup. An increase of the Cs2 coefficient increases the breakup time, evaporation time, liquid penetration length and thus decreases the resulting in-cylinder pressure. The effect of changing to the Huh atomization model was significant and showed a pressure decrease which corrected the in-cylinder pressure results closer to the experimental data. Defining the specific heat and surface tension of the liquid fuel showed to have crucial influence on the timing and magnitude of the pressure peak. Comparing methanol and ammonia in a non-reacting simulation shows that methanol leaves the region cooler and less fuel rich as the evaporation process is slower, also shown in terms of the liquid penetration length which is longer for methanol.
    Period5 Jun 2021