Using Energy Fluxes to Analyze the Hydrodynamic Performance of Marine Propulsion Systems

Sammanfattning: The strive towards more fuel efficient ships is a continuously ongoing process, motivated by both economic and regulatory reasons. An important aspects to consider for the final fuel consumption is the propulsion system performance in relevant operating conditions. The propulsion system performance is most commonly described using a well-established terminology, including thrust deduction, wake fraction, and propulsive efficiency, a decomposition with its primary origin in the experimental procedures used to establish ship scale performance. Since this decomposition does not really provide us with any details about the flow, it can imply limitations in design and optimization of the propulsion system, as the interaction thus may not be correctly represented and fully understood. Numerical methods, such as Computational Fluid Dynamics (CFD) based on the Reynolds-Averaged Navier-Stokes (RANS) equations, can be used to extract detailed data of the flow around marine propulsion systems. It is proposed in this thesis to conduct control volume analysis of energy based on CFD results to describe the performance of the propulsion system. Control volume analyses of energy is actually a power balance, since it is expressed in terms of energy fluxes and can be directly coupled to the delivered power. Through a decomposition of the energy fluxes over the control volume surface the system performance can be described in terms of kinetic energy in axial direction, rate of pressure work, kinetic energy in transverse directions, internal energy, and turbulent kinetic energy, the two last representing the viscous losses. In general there are no restrictions of how to construct the control volume, it rather depends on the analysis objectives, for which it needs to enclose the entire flow domain of interest. However, it is shown that the downstream surface preferably is located in the vicinity of the studied object, to obtain more details of the flow before it has dissipated into internal energy. Further, from conducted studies it is clear that the control volume for flexibility preferably is constructed in the post-processing phase. It is evident that the possibility to characterize the flow is entirely dependent on the underlying CFD solution, which needs a sufficiently refined grid and suitable models to accurately capture the flow field around the propulsion system. Important aspects to consider for the propulsion system modelling discussed within the thesis are: representation of hull boundary layers, interaction with the free surface, the possible influence from laminar boundary layers on the propeller (model scale), and surface roughness of both hull and propeller (ship scale). The control volume analysis of energy has within this project successfully been applied to describe the performance differences for a vessel operating with an open and ducted propeller respectively, and for describing the reasons behind the established optimal propeller diameter reduction in behind conditions relative a homogeneous inflow.