Accurate prediction of the real voyaging speed for ocean‐going vessels in the actual weather condition is important for the expanding shipping industry. Ship owners are always aiming at achieving the highest profit, so delivering the goods to the destination within the designed schedule time is of great concern. At the same time, the increasing attention on environmental issues as well as increasing fuel price have put more pressure on optimizing the ship design with respect to energy efficiency. Also, accurate prediction of the real attainable speed will greatly improve the sea margin prediction. From these respects, the research work on speed loss study is important and necessary. Speed loss is usually categorized as voluntary or involuntary. Voluntary speed loss is when the ship master actively reduces the ship speed to avoid slamming, propeller racing, excessive ship motion or other effects that might cause danger or severe discomfort. Voluntary speed loss is subjective and relies on ship master's experience. Different ship masters can have quite different actions to the same circumstance based on their ability to estimate the potential danger. Involuntary speed loss is due to added resistance from waves, wind, current and reduced thrust and efficiency due to waves and other changes in operational conditions. Involuntary speed loss is in focus in this thesis work. In order to investigate the nature of speed reduction in the real environment, integrated knowledge about resistance, propulsion, ship machinery, seakeeping, automatic control and maneuvering are required. This thesis takes advantage of model tests and a numerical simulation tool based on simplified modular concept to investigate the nature of speed reduction in seakeeping and maneuvering conditions.
A series of experiments were carried out on a model of 8000 DWT tanker in the large towing tank and ocean basin at Marine Technology Center in Trondheim, Norway. The model was selfpropelled and mainly running in moderate long wave conditions. For speed loss in head sea conditions, three different bow shapes were tested in order to investigate the influence of added wave resistance. Due to the practical difficulty of applying towrope force to balance the frictional resistance coefficient difference between model scale and full scale ship, a method is proposed in this thesis to make compensation for the frictional resistance difference and predict speed loss of the ships when propeller is working at 'ship propulsion point'. Test results show that a conventional bow with bulb and flare gave least calm water resistance, while a bow without bulb and flare gave the least total resistance in the tested wave conditions. That means bulb can have positive influence on calm water resistance and negative effect on added wave resistance. Reduction of speed loss in waves up to order of 10% can be gained by relatively minor changes to the bow shape of a vessel. Another two tests for speed loss in zigzag maneuvering in head waves and speed loss in oblique waves were carried out in the towing tank and ocean basin respectively. Speed loss when doing zigzag maneuvering in waves is due to yawing, added wave resistance and loss of thrust due to steering; while for speed loss in oblique waves is due to added wave resistance and loss of thrust due to steering. For the test of speed loss in oblique waves, towrope force is added by an air fan. Due to limitation of basin length, converged speed is not always achieved during the tests due to the large mass of the model. A converged speed prediction method is proposed which can be used to correct the nonconverged tests results. This method is carefully verified and can give good prediction of attainable speed in waves. However, this method is sensitive to the selection of thrust deduction factor in waves. Also how much speed loss due to added wave resistance and due to steering is pointed out. Speed loss due to added wave resistance and steering is of the same magnitude in head sea and bow sea conditions. While for beam sea, speed loss due to steering is dominating.
Numerical simulation work was carried out in order to make comparisons with experimental results. Comparable calculations were performed in the frequency domain tool ShipX and in the time‐domain tool Vessel Simulator. Ship motion is calculated by linear strip theory and added wave resistance in head sea is calculated by the method developed by Gerritsma and Beukelman and the Direct Pressure Integration method. Wave resistance in other than head sea condition in surge, sway and yaw directions is calculated by the method proposed by Loukakis and Sclavounos, which is an extension of the Gerritsma and Beukelman method to other than head sea. For the speed loss in head sea condition, both frequency domain calculations using ShipX and time domain calculations using Vessel Simulator were carried out. It is concluded that both methods can give a good prediction of speed loss in moderate sea conditions. For the cases that steering effect has to be taken into account, time domain simulation is a preferable method. Generally speaking, numerical results can give a good correlation with model test data. Based on the good validation of the numerical tool, further numerical investigations were carried out to compare the speed loss characteristics between conventional propeller‐rudder system and azimuth thrusters. The results showed that conventional propeller‐rudder system has better speed keeping ability than azimuth thrusters. That is because in the conventional propeller‐rudder system propeller force is not decomposed by the rudder angle and also conventional propeller‐rudder system has higher ratio of longitudinal thrust force/transverse thrust force than azimuth thrusters within the working steering angle range.
In the end, recommendations to ITTC procedures to predict speed loss and power increase in irregular waves and sea margin were proposed. A method to predict 'ship propulsion point' from the self‐propulsion test carried out at 'model propulsion point' is specified in this thesis. This method will greatly improve the procedures to predict power increase in irregular waves from model tests in regular waves. Also another suggestion is proposed that when evaluating power increase in irregular waves and sea margin prediction, thrust deduction due to ventilation and steering effect should be taken into account in addition to thrust diminution factor.