In this thesis the response variables (RAOs) of a semi submersible unit are inspected. Both operational and survival condition as well as a shallow draft are inspected. The survival condition is inspected with respect to an element analysis. And both operational- and shallow draft condition are case studies, where the operational condition is inspected for two different damage cases. The unit in question is a four column semi submersible, based on the GG5000 design. This is a relatively new design, and the first vessel to get this design is in its final engineering stage. Construction start is planned to be in August this year (2012). This unit will get the name COSLProspector and will be built in CIMC Yantai Raffles shipyard in China.The unit is symmetrical about the centre line and close to symmetrical about the vertical transverse plane, only pontoon tips are different. Because of this, no significant simplifications have been necessary in order to simplify the calculation due to computational time. Another reason for not doing any simplifications to the geometry is due to the fact that the results are desired to be the most realistic. However, to reduce computational time, only half the unit has been modelled due to symmetry about centre line.To find the appropriate element size for the mesh, an element analysis has been carried out. The results from this analysis resulted in a chosen element size of 2.5m. This element size both gives accurate results, and requires a relatively short computational time. The units’ resonance periods has been investigated, and verified by help of hand calculations and comparison with RAOs done by Global Maritime (2011). However not all the values were identical to each other, but many factors can influence on that result. The GM value was not changed in this thesis, but was in Global Maritime (2011), in addition the additional damping was in this thesis taken as 3% of critical damping, while in Global Maritime (2011), Morrison elements were taken into account. These factors, and perhaps a few shortenings are assumed to be the reason for the small difference in the responses, they are however small differences for most of the periods.Two damage cases have been modelled by flooding two different water ballast tanks. These damages will give an angle of list for the unit. Damage case 1 gives an angle of list of 13.18o with a rotation of heel axis of 7o forward. Damage case 2 gives a list angle of 11.68o with a rotation of the heel axis of 39o forward. An earlier study like this one is done by Henriksen (2011), found in Grenland Groups internal archive. AQWA does not give out the tilt angles in damage cases as this is not the main purpose of this program. Therefore, the list angles for the different cases have been obtained from the report done by Henriksen (2011). However, AQWA will be used to obtain the RAOs for both cases, as well as confirm floating equilibrium in such conditions.It is assumed that the tanks are completely emptied for air, and that seawater is filling the entire volume. A table showing the different tanks flooded and its weight with seawater is shown in table 1.UnitVolume [m3]Sea water weight [MT]BWT ST-2692.51709.82BWT ST-8616.83632.25Table 1: Weight of water for flooded tanksWhen it comes to the RAOs in the damage cases, they are very hard to read. This is mainly due to the fact that the motions in these cases are highly dependent on each other due to coupled motions. Due to an angle of list, the unit is no longer symmetric. As a consequence of this a RAO for a specific degree of freedom can no longer be read like it is only this degree of freedom which is affecting the responses, but one or more of the other degrees of freedom are strongly influencing. This makes some of the peaks appear where not normally expected.It is also noticeable that the highest motions are encountered for damage case 1, which is natural because this case has the highest list angle. The resonance periods are lower in the damage cases compared to the normal operational condition, however not to a degree which is dangerously low. The lowest resonance period is still in heave.From the RAOs in the shallow draft case, it can be noticed that the highest responses in heave are encountered for the shallow condition (14.5m) compared to the survival condition (15.5m) and operational condition (17.5m), however only up to about 18s, where after that it has the smallest response, and the operational condition has the highest.In roll and pitch the trends are fairly similar. The graphs are a little uneven until the first cancellation period, and then the shallow draft gives a higher response until reaching the resonance period. In the resonance region, the operational condition has the highest response for both roll and pitch, same as for the heave.As a conclusion, the optimal approach in a situation where the unit is heavily tilted is to try to ballast the unit to an even keel. But of course risks of doing this are a possibility, such as slamming problems and the fact that the resonance periods will be shorter.

Increasing demand for clean and effective energy production turns interests of the world on floating offshore wind technology. To establish floating wind farms, a wind turbine have to be mound on a floating structure. The floating structure has to be carefully design according to sea environmental condition and kept in precise position. Different types of floating structure and stationkeeping systems have been proposed for floating wind turbines. This project deals with design a spar floater with tension leg mooring system, where the vertical fairlead position located between center of buoyancy and center of gravity. In this project a details study about floater design and tension leg concept was presented. Further, a model was established in computer program RIFLEX and static and dynamic analysis was carried out for two different environmental conditions, one for an operation condition to understand the model behaviour in normal sea state. The second one was for extreme condition to estimate the extreme tether forces and find out slacking possibility.

The lattice Boltzmann method has become a valuable tool in computational fluid dynamics, one of the reasons is due to the simplicity of its coding. In order to maximize the performance potential of today’s computers, code has to be optimized for parallel execution. In order to achieve parallel execution of the lattice Boltzmann method, the data dependency has to be solved. And to get good performance, the memory has to be organized for unit stride access. Here we investigate the most known algorithms for lattice Boltzmann, and implement a code which runs on a parallel graphics processor, using a library for parallelization called C++ AMP. Furthermore, we show how the code compares to known solutions of fluid flows to verify the numerical results. The optimized parallel code achieves a speed up of 650 times the un-optimized code, on a current generation high-end graphics card.

As the worlds demand for energy is increasing mostly due to the increase in population, and coal, oil and gas deposits are limited, it is desirable to gather energy from renewable energy sources. Wind energy is a form of renewable energy. Wind turbines have been common on land and near shores for some time, but now one wants to take advantage of the wind resources further away from the coast. As the length from the coast increases, so does the water depth making it necessary to use other foundations than the well-known monopile. In this thesis the “Federal Maritime and Hydrographic Agency” in Germany, also known as Bundesamt für Schifffarth und Hydrographie requires in the standard “Design of Offshore Wind Turbines” that an offshore wind turbine needs to be risk evaluated against a ship tanker of more than 500 MJ in collision energy in order to be classified as collision energy. This is to prevent environmental pollution in form of oil spill. Whether the offshore wind turbine models used in this thesis are collision friendly or not relies on different factors. When the given soil properties are used the analyses show in all cases except one that the offshore wind turbine models can be called collision friendly. The case were it could not be called collision friendly was a case with the small jacket at a water depth of 27 m getting hit by a loaded ship at a column of the jacket, but installing a horizontal brace on this jacket would make it also collision friendly. The effects of soil, water depth and a horizontal x-brace are looked further into in this thesis. If the soil had been stronger it is not certain that the outcome would be the same. When the jackets were fixed to the sea bed, several of the models could collapse in the dangerous direction over the ship. A horizontal brace was seen to have a positive effect when installed on the different jacket models. It seems also that it is more favorable to use jackets at deeper water.

Det hydro-pneumatiske strekksystemet til stigerørene virker som en passiv hivkompensator med en olje/gass-akkumulator for a utligne de relative bevegelsene mellom semien og stigerørene. Semiresponsen fører direkte til at stempelet beveger seg i sylinderen, noe som gir kompressjon eller ekspansjon av gassen basert på ideell gasslov. Dette gir en ekstra ikkelineær stivhet, som også er avhengig av gassvolumet i systemet. Effekten av semiresponsen for et strekksystem med variabelt gassvolum har blitt evaluert.Strømmen av hydraulisk væske mellom sylinderen og akkumulatoren i kombinasjon med strømmen av gass mellom akkumulatoren og gassreservoaret, gir viskøs dempning i strekksystemet. Disse friskjonskreftene er bestemt av lengde og diameter i rørsystemet, ventiler og andre komponenter i det hydraulisk-pneumatiske systemet. I denne masteroppgaven er det brukt forenklede beregningsmodeller som gir gode, men ikke eksakte resultat. En numerisk metode har blitt skrevet for å beregne trykktap basert på Bernoullis ligning og konservering av masse og moment i systemet.En numerisk metode har også blitt skrevet for å løse svingeligningen i tidsdomenet med ikkelineær stivhet og dempning fra det hydro-pneumatiske strekksystemet. En sensitivitetsanalyse basert på forskjellige perioder og bølgehøyder ble gjort for å separere de forskjellige bidragene fra stigerør-strekksystemet, og lettere finne trender og sammenhenger som definerer dette. I forprosjektet ble transferfunksjoner regnet ut for noen gitte senarioer. Disse gir viktig og mye brukt informasjon i tidlig designfase for prosjekter. Nye transferfunksjoner er tatt ut ved å plotte maksimum respons i stabil fase, basert på tidsserier fra de forskjellige bølgehøydene og periodene.Det finnes friskjonskrefter i ulike kontaktpunkter mellom bevegelige deler på semien, noe som er spesielt viktig for små sjøtilstander. Semiresponsen blir påvirket av friksjonskreftene mellom stigerørene og kjølguiden, og i pakninger i strekksystem-sylinderene. Når friskjonskreftene blir større enn eksitasjonskreftene fra bølgene, vil semien bli «forankret» til stigerørene. Dette fører til at bølgekreftene blir overført direkte til stigerørene, og denne belastningen kan redusere forventet levetid. Analyser har blitt gjort for hvilke regulærbølger dette vil skje. Det er også sett på hvordan den økte stivheten ved en «forankret» tilstand reduserer egenperioden til systemet.