CFD modelling and validation of a large 2-stroke marine diesel engine

  • Wei Jin

Student thesis: Doctoral Thesis

Abstract

With the increased understanding of the physical processes of the large 2-stroke marine engine and the corresponding numerical models as turbulence, injection, evaporation and chemical combustion, it has become realisable to investigate the whole working processes using Computational Fluid Dynamics (CFD). However, CFD itself needs to be validated to produce reliable results. The focus of this thesis is on the CFD validation of the large 2-stroke marine diesel engine working processes. The turbulence models and the wall functions in the CFD solver were firstly tested using two basic 2D cases to obtain performance results in the modelling of turbulence flow and heat transferring near the wall. It was concluded that the RNG and the realisable Ҡ-Ɛ turbulence models predict the same best flow field and non-equilibrium wall function produces the best heat transfer results. Following this, the spray breakup models were evaluated through the constant volume spray chamber test. The Stochastic Secondary Droplet (SSD) model was verified as the best one for its generality and accuracy. During this process, the appropriate mesh for droplet calculation using Euler-Lagrange approach was determined. Based on the ontained conclusions, the computation model of a large 2-stroke marine diesel engine MAN B&W S60MC-C6 was generated, with the proven best models, the RNG Ҡ-Ɛ turbulence model, the Non-Equilibrium wall function and the SSD breakup models. In addition, different combustion models including the Finite Rate / Eddy Dissipation (FRED), the Non-Premixed Equilibrium, the Non-Premixed Steady Flamelet and the Non-Premixed Unsteady Flamelet Diesel were investigated at four engine loads (25%, 50%, 75% and 100% Maximum Continuous Revolution (MCR)). The obtained in-cylinder pressure traces were compared with the test shop data. It was proven that the Non-Premixed Equilibrium combustion model presented the best prediction performance at four loads.The derived conclusions can be used as guidelines for CFD simulations of the large 2-stroke marine diesel engines working processes. It also provides the starting point for engine optimisation to increase engine efficiency and reduce pollutant emissions.
Date of Award1 Nov 2012
LanguageEnglish
Awarding Institution
  • University Of Strathclyde
SupervisorDracos Vassalos (Supervisor) & (Supervisor)

Cite this