Marine Propulsion System Performance Beyond the Propulsive Factors

Doktorsavhandling, 2021

The marine propulsion system often operates in the wake of the ship it is mounted on. This implies that the propulsion system affects the flow around the hull, and the inflow to the propulsion system is dependent on the hull shape. Due to the complexity of the flow, the performance of marine propulsion systems has historically been assessed through model-scale testing in basins, alternative numerical tools were for a long time not available. However, the limited possible measurements does not provide any detailed descriptions of the propulsion system performance. A second factor complicating the use of model-scale testing is the inevitable low Reynolds number in relation to the real ship. The need for a more detailed description of the propulsion system performance motivates the use of Computational Fluid Dynamics (CFD); there are no limitations on what to extract in terms of flow details or forces on surfaces and it can be applied for the ship-scale system. However, this requires a representative model of the flow in ship-scale which is not yet well established. In this thesis, using CFD, two alternative approaches to the propulsive factors originating from the model-test scaling procedure are proposed to describe the propulsion system interaction effects: A detailed evaluation of forces on the propulsion system and hull surfaces and a control volume approach based on energy fluxes describing the power required by the propulsion system in terms of various energy fluxes. For the first approach a powerful tool is the thrust over torque coefficient for a blade around a revolution, or studies of its radial distribution at specific positions. A clear advantage of the control volume approach is its possibilities to describe the viscous losses. As a step towards ship-scale CFD a review and comparison of different methods to model hull roughness is conducted; it shows no convergence towards specific roughness functions or methods to obtain the roughness length scales and there is neither a strong correlation between the additional resistance predicted by various hull roughness models and the Average Hull Roughness (AHR). Applying the proposed approaches, a few generic interaction effects could be explained. For instance: The old rule of thumb regarding optimal propeller diameter in-behind based on model-scale tests from the 1950s is shown to be mainly due to that operation together with a rudder favours smaller propeller diameters since the rudder can make use of the stronger slipstream rotation. However, the results indicate that this only holds within the same scale factor, and the optimal propeller diameter for the ship is most probably larger than what is indicated by propeller series data. Other generic interaction effects explained is how tip-unloading deteriorates propeller performance to a larger extent in-behind since the wake distribution further decreases the load on the blade tip, and how blunter leading edges has a superior performance at low load, since they are less sensitive to poor performance at negative angles of attack.