Hydrodynamic Aspects of Sail-Assisted Merchant Vessels
Kramer, J.V. (2022). Hydrodynamic Aspects of Sail-Assisted Merchant Vessels. PhD Thesis. Norwegian University of Science and Technology (NTNU): Trondheim. e-ISBN 978-82-326-5848-0. 171 pp. |
| Authors | | Top | - Kramer, J.V.
- Eloot, K., revisor, more
| | |
| Abstract | Modern sail technology is a potential solution for reducing the fuel consumption for merchant ships. This can be used to either reduce the emission of greenhouse gasses with conventional fuels, or the cost related to zero-emission fuels. This thesis explores wind-power for commercial shipping, with particular focus on the additional hydrodynamic challenges related to such vessels. These challenges primarily arise because a sail does not only push a ship forward, but often also strongly sideways. For a wind-powered vessel to move with a steady speed and direction, this aerodynamic side force must be balanced by opposing hydrodynamic forces and moments. A conventional ship will do this by moving with steady drift, rudder, and heel angles. As a direct consequence, the resistance of the ship will increase, and therefore remove some of the positive effect of the sails. The added resistance on a ship due the sails are labeled the sail-induced resistance in this report.
The research work was divided into three main questions related to the negative hydrodynamic effects: how can we accurately and practically model a wind-powered ship when the negative hydrodynamic effects are a concern? How important are these negative effects for merchant ships?
What can be done to reduce the negative effects to a minimum?
The project started by exploring the drift-induced forces on simple foil-like ship geometries with different aspect-ratios and bottom edge shapes based on towing tank experiments. This was done to explore the accuracy of simplified methods for low aspect-ratio lifting surfaces and to see how much the lift and lift-induced drag are sensitive to design details. It was found that simplified theories where not able to capture the physics with acceptable accuracy, and that relatively small changes to the bottom edge shape could have a large impact on the results. The cross-flow drag experienced by the ship was found to not only affect the non-linear lift, but also the linear lift and lift-induced drag.
Because of these complicated physical effects, the details in the geometry are important when the goal is to create an accurate hydrodynamic model. An automated and scriptable framework for setting up CFD simulations using the open-source library OpenFOAM was therefore developed. This was used to simulate the forces acting on ship hulls with both rudders and keels, at various speeds, drift angles, rudder angles, and propeller loadings. Different simplifications were tested for the simulation strategy, including neglecting the free surface and performing the simulations in model scale. These simplifications were found to give acceptable accuracy, although with some important limits on both model size and Froude number. The assumptions in established maneuvering theory for predicting the forces due to drift angle, rudder angle, and propeller thrust were evaluated with the purpose of reducing the necessary test matrix for a given ship design to a minimum. Most simplifications in the MMG maneuvering model were found to be acceptable, including the models for the rudder-hull interaction. However, an update to the rudder force model was suggested to better estimate the liftinduced resistance. In addition, heel was found to affect the drift-induced forces significantly when the drift angle was large, but much less for small drift angles. The most likely explanation is that heel is mainly affecting the cross-flow drag on a merchant ship geometry, while the circulatory lift is less affected.
Although the focus of this thesis is hydrodynamic effects, some attention was also given to the aerodynamics of wind-power devices. A custom discrete lifting line method was developed specially to model wingsails. This method includes the interaction effects between multiple sails and viscous effect on both lift and drag. The lifting line model was compared against CFD simulations for both single wings and for multiple wings as a function of wind-direction. The accuracy was found to be good, especially considering the computational speed and simplicity of the method. The importance of the negative hydrodynamic effects due to the sails was evaluated by performing route simulations of two different case study ships representing a 5 000 DWT general cargo ship. The necessary software for performing such simulations was developed as part of this project. The exact magnitude of the sail-induced resistance depends on several factors, such as total sail area, ship speed, hydrodynamic design, and control strategy for the sails. In general, the relative importance of the sailinduced resistance is highest for low ship speeds and for cases with large fuel savings. As an example, at 8 knots ship speed, and three 56 m tall sails, the fuel savings without hydrodynamic effects were estimated to approach 70% based on route simulations with weather data from the north Atlantic. When the hydrodynamic effects where included, the fuel savings were reduced to right below 60%. In other words, around 10 percentage points – which corresponds to 14% of the sail thrust – was lost due to the sail-induced resistance. The source of the added resistance was found to be a combination of drift-induced forces on the hull and lift-induced resistance on the rudder. The rudder was found to often be the largest source of resistance, depending on the sail-placement. One way to reduce the liftinduced resistance on the rudder was to balance more of the side force with the hull. This could be achieved by adding either a fixed high-aspect ratio keel to the hull or bilge keels. A surprising result was that low-aspect ratio bilge keels essentially gave the same improvement as a high-aspect ratio keel. The reason was that the lift coefficient and the resulting lift-induced drag on the hull was small, despite the small aspect ratio of the geometry. The best design solution tested in this project was a dynamic high-aspect ratio keel that was both rotatable and retractable. This allowed the lift on the keel to be adjusted independently of the drift angle on the hull. The loss of fuel savings due to hydrodynamic effects were reduced to right above 6 percentage points at 8 knots with this design solution.
Another important solution for managing the negative hydrodynamic effects was to alter the control strategy for the sails. The case studies showed that the side force from the sails can be significantly limited in unfavorable weather conditions without significant loss of fuel savings. This allowed the ship to operate with smaller rudder angles and heel angles. This is therefore a simple way to ensure safe and comfortable operation of wind-powered ships without any physical design changes to the system. Although the thrust from the sails is reduced along with the side force, the reduction in sail-induced resistance was almost equal in magnitude, depending on the exact ship design. A simple control algorithm designed to maximize thrust but with strict hydrodynamic limits was found to be comparable to the more advanced control algorithms that optimized the operation of the sails including hydrodynamic effects.
In short, the work presented in this report shows that the negative hydrodynamic effects on windpowered merchant vessels are important to consider if accurate fuel savings are the goal of the analysis. Practical and efficient computational methods for analyzing these effects are suggested. However, the magnitude of sail-induced resistance is not so large that it should be considered as a major problem for the concept of wind-propulsion for merchant ships. The negative hydrodynamic effects can be managed quite easily with either simple design changes, modifications to the control algorithm of the sails, or both. |
|
