Formålet med masteravhandlingen er å gjennomføre en systematisk evaluering av ulike end-flash løsninger i forhold til kraft- og brenselbalanse, produksjonskapasitet, LNG-produktspesifikasjoner, spesifikt kraftforbruk og samlet kraftforbruk i forhold til LNG-produksjon. For å begrense analysen fokuserer studiet seg om følgende typer end-flash konfigurasjoner; i) ingen spesifikt end-flash system (avdampning inn i LNG-lagringstank), ii) enkelt end-flash system med flashgasseparator, iii) systemer med mer separasjon mellom C1 og N2 og økt metangjenvinning (strippekolonne), og iv) resirkulering av end-flash gass til oppstrøms kondenseringsprosessen. Bruk av flashgass for brensel og gjenvinning av kald energi fra flashgassen vurderes når slike alternativer er mulige. For system i) og ii) vurderes kaldgasskompresjon av EFG og BOG mot kuldegjenvinning og varmere kompresjon av EFG og BOG til fyrgass.For å oppnå en metodisk sammenligning og analyse implementeres de ulike end-flash systemene inn i en C3MR-kjøleprosess. Sammenligning for systemene baseres på tilgjengelig kraft fra to Frame-7 gassturbiner med tilhørende hjelpemotorer. Det elektriske kraftgenereringssystemet drives av gassturbingeneratorer. Fyrgass fås fra NG oppstrøms kjøleprosessen, fra EFG og BOG, eller fra en kombinasjon av disse. Simuleringene gjennomføres med to ulike gassammensetninger med henholdsvis høyt og lavt N2-innhold. Lav-N2 har et N2-innhold på 0,5 mol%, som er lavere enn LNG-krav. Alle end-flash konfigurasjonene vurderes derfor med lav-N2. Høy-N2 innholder 3 mol% nitrogen. Evaluering av høy-N2 og system i) gjennomføres ikke grunnet uoppnåelig tilfredsstillelse av LNG-krav. System med lav-N2 og avdampning inn i LNG-lagringstank oppnår et høyt spesifikt kraftforbruk og dårligst parameter for samlet kraftforbruk i forhold til LNG-produksjon, sammenlignet med de andre systemene. Systemløsning med kolonne oppnår et LNG-produkt med et veldig lavt N2-innhold, som kan bety større gevinst ved salg. Med tanke på produktrate og kraftforbruk, er system med separator en optimal løsning med lav-N2. End-flash system med høy-N2 og separator medfører at noe av EFG og BOG må resirkuleres oppstrøms kjøleprosessen for at systemet ikke skal levere for mye fyrgass og samtidig tilfredsstille LNG-krav. Med strippekolonne og høy-N2 kan EFG og BOG dekke nødvendig fyrgassmengde og tilfredsstille LNG-krav uten resirkulering. Samlet krever prosessen et mindre kraftforbruk enn system med separator. Resirkulering oppstrøms kjøleprosessen bør unngås dersom mulig. Dersom N2-innholdet i EFG og BOG er for høyt og ikke kan benyttes til fyrgass, må nitrogenet tas ut av systemet med en dobbeltkolonne, som medfører mer kompleks drift. Et mer komplekst system gir ikke nødvendigvis økt LNG-produksjon for et bestemt kraftforbruk. Termodynamisk sett kan et LNG-anlegg tjene på å benytte kaldgasskomprimering framfor kuldegjenvinning og varmere kompresjon av EFG og BOG.

Aeroelastic design and fatigue analysis of large utility-scale wind turbine blades are performed. The applied fatigue model is based on established methods and is incorporated in an iterative numerical design tool for realistic wind turbine blades. All aerodynamic and structural design properties are available in literature. The software tool FAST is used for advanced aero-servo-elastic load calculations and stress-histories are calculated with elementary beam theory.According to wind energy design standards, a turbulent wind load case is implemented. Fatigue loads are estimated based on 100% availability and a site-specific annual wind distribution. Rainflow cycle counting and Miner’s sum for cumulative damage prediction is used together with constant life diagrams tailored to actual material S-N data. Material properties are based on 95% survival probability, 95% confidence level, and additional material safety factors to maintain conservative results. Fatigue performance is first evaluated for the baseline blade design of the 10MW NOWITECH reference wind turbine. Results show that blade damage is dominated by tensile stresses due to poorer tensile fatigue characteristics of the shell glass fiber material. The interaction between turbulent wind and gravitational fluctuations is demonstrated to greatly influence the damage. The need for relevant S-N data to closely predict such blade stress cycle events is investigated to avoid non-conservative conclusions. State-of-art wind turbine blade trends are discussed and different designs of the NOWITECH baseline blade are analyzed in a parametric study focusing on fatigue performance and material costs.

The report is written to show as to what extent test-driven development (TDD) and continuous integration (CI) can be used on KBE models and how a unit testing framework for KBE models can be developed.Test-driven development (TDD) and continuous integration (CI) has changed the way software is tested. Software testing was often a separate process at the end of a project. It is now being worked on during the entire development period. TDD and CI relies on unit tests. Unit tests are done by dividing the code into the smallest possible units and testing each of them independently. This master’s thesis asks how these practices can be used for testing knowledge based engineering (KBE) models.A unit testing framework for the Adaptive Modeling Language (AML), AUnit, has been developed. It is explained in detail and an introductory guide to using AUnit for testing KBE models in AML is included. AUnit was used to perform TDD and CI on different KBE models, both creating new models and testing existing ones. Testing KBE models differ to a large degree from testing regular object-oriented software. Different approaches for unit testing and TDD has been performed on several KBE models. It was concluded that the basic attributes in KBE models cannot be unit tested in a sensible way. This includes adding any superclasses and simple parameters like height and width. Without including these attributes, unit testing cannot fully be performed on KBE models using AUnit. However, the models can highly benefit from having unit tests for the logic in the model, which is where the most severe bugs will be. When the attributes are implemented in the model, test-driven development (TDD) can be performed on the models.Automatic continuous integration (CI) has been performed on a KBE model andthe basic principles of CI have been accounted for. CI for KBE models does notdiffer much from other software projects, so its focus is reduced.

Related to the technology of CO2 capture, transport, and storage (CCS), an accurate transport model which predicts the behaviour of carbon-dioxide mixtures during steady-state and transient situations, is needed. A correct estimation of the frictional pressure-drop is an important part of such a model.A homogenous friction-model, the Friedel model, and the Cheng et al. model have been compared with six steady-state experiments using pure CO2. The experiments were nearly adiabatic and within the following range: mass velocities from 1058 to 1663 kg/m2s, saturated temperatures from 3.8 to 17 C (reduced pressures from 0.52 to 0.72), vapor fractions from 0.099 to 0.742, and pipe diameter of 10 mm.The Friedel model was found to be the most accurate model with a standard deviation of 9.7 % versus 55.74 % for the Cheng et al. model and 29.18 % for the homogenous model.The selected friction models were implemented into a numerical model for pipe flow of multiphase CO2, and one of the mentioned experiments was reproduced. The result illustrates how the accuracy of the friction model is even more important when used as a part of the complete transport-model. This is mainly because the friction model and other sub-models, such as the equation of state, are coupled. During the implementation of the Cheng et al. model, certain errors in the original paper were found and corrected.In the case of a transient flow, the influence of the friction model and the associated slip relation, were explored. It was shown that wave speeds strongly depends on the slip relation used. The friction model itself will indirectly affect the wave speed. This is mainly because of the reducedfluid velocity arising when the driving force across the wave is reduced. However, the main effect of the friction model is the pressure gradient arising in regions where the velocity is non-zero.

Resonator AS is developing a hammer drilling system suitable for oil and geothermal drilling. The nature of the device involves high speed linear motion at high frequencies. As a result of this, the gas springs used in their device are subject to high pressures and temperatures. To prevent pressure leakage from the device, seals are installed.The seals used in the device are fit for use in a limited range of temperatures and pressures. This pose a challenge since a high amount of heat is generated as a result of the friction between the seal and and the gas spring walls.The main objective of this thesis has been to develop a model to investigate the effects of the frictional heat. Control volume techniques have been used to model the effects on the gas pressure and temperature. A detailed heat transfer formulation comprising convection and conduction has been developed. To model the conduction it was necessary to derive two-dimensional heat transfer equations for cylindrical coordinates. A MATLAB-code has been written in order to simulate the behavior of the system.Test results have been compared with the simulations to validate the model. The comparison showed that the model described the thermal inertia of the system appropriately. It also showed that the present friction model is not able to describe the friction in a satisfactory manner.

This report considers different energy solutions for a floating LNG production vessel. The two alternatives analyzed are gas turbine and steam turbine. In addition to being stand-alone alternatives they are the basis for other alternatives, such as electric drive and combined cycle. Light side studies have been performed on the two latter ones. A LNG process model has been built in Aspen HYSYS, and from this several cases has been run. There are 3 levels of different parameters that have been run: 1) Energy system, 2) Feed Gas Composition, and 3) Liquefaction process. As mentioned the two energy systems are GT and ST. Three different feed gas compositions have been analyzed: 1) Low content of both CO2 and N2 (0.5 % and 1 % respectively), 2) High content of CO2 (9.5 %), and 3) High content of both CO2 and N2 (9.5 % and 3 %). The liquefaction processes analyzed are two of the most promising for a floating LNG application: Dual mixed refrigerant and dual N2 expander.The feed gas compositions was chosen to give a wide area of applications for the results, and to give illustration on how the two different energy solution would respond to changing feed gas composition. The DMR liquefaction process was chosen mainly because this is the one being implemented in Shell Prelude FLNG. Being the most proposed solution for offshore application the dual N2 expander was a natural alternative to the DMR.The analysis show a clear advantage for gas turbine and DMR process, when exclusively looking at efficiency. However; as the objective of the study states, important factors such as safety, vessel motion sensitivity, reliability, availability is also to be considered. The results show 245 kWh/ton LNG energy consumption with the DMR liquefaction process, whereas the dual N2 expander requires 424 kWh/ton LNG; over 70 % increase. However; the side- cases run in this report show advantages to the N2 dual expander in safety, weight/space requirements and ease of start-up and shut down.The ST/N2 has fuel gas consumption 4% higher than the GT/N2. On the basis of the results in this report and other studies performed on FLNG a selection of the ST/N2 setup will be favorable as long as there is a high CO2 content in the feed. With low CO2 content, hence heat demand, the advantage of the ST is smaller thanks to lower heat recovery demand.

This objective of this thesis has been to development and analysis a measurement apparatus designed to determine thermal conductivity of porous materials. A literature survey concerning available experimental techniques for thermal conductivity measurements was conducted. A steady state radial heat transfer method with cylindrical geometry and a centered heating element was found to be most suited technique for achieving accurate and reliable results. A side wall cooling arrangement was used to achieve desired cooling temperatures. To restrict the extent of the work, it was decided to only investigate heat transfer behavior at cryogenic temperatures. Test specimen with a thermal conductivity of 0.05 W/(m*K), (assumed to be the thermal conductivity of the materials to be tested in the apparatus) and a thermal conductivity of 0.01 W/(m*K) for the insulation components, were the ones chosen for investigations. The design process of the new apparatus, using the software COMSOL Multiphysics 4.2, was initiated by evaluating heat transfer behavior in a simple cylinder, containing a hollow heating element and the test specimen. Radial heat transfer was verified, hence, the design process proceeded. Extensive, step-wise analyses were conducted to evaluate heat transfer behavior as the complexity of the apparatus increased. Implemented elements such as insulation blocks, a heater support and three thermocouples proved to cause heat losses in the test section, which resulted in errors in the calculated thermal conductivities. Furthermore, an electric wire, supplying the heating element with current, was included in the model. In addition, the hollow heater was replaced by an aluminum oxide heater since such an element is to be used when building the apparatus. Unexpected results revealed critical heat transfer into the test section from the wire. This led to an investigation of the wire length to reduce such effects. Lastly, as a result of the analyses carried out, the overall error of the thermal conductivity measurements due to heat losses was determined. Dimensional drawings of the characteristic dimensions, as well as practical solutions for the final compilation of the apparatus, were suggested as the last step of the design process. It was of interest to estimate the overall uncertainty of the apparatus when all parameters effecting the measurements, were included. For this, a comprehensive uncertainty analysis was conducted and compared to previous work. Results showed that temperature recordings from the thermocouples placed in the mid-section of the test cylinder would provide the most reliable results for the determination of thermal conductivity in the test apparatus.

The main objective of this thesis is to describe the transient thermodynamics during physisorption of hydrogen gas using a commercial numerical software.Simulations of thermal effects during adsorption are valuable tools for the efficient design of hydrogen adsorption storage systems. Transient mass and energy equations are used for describing the adsorption process. For this purpose, experimental adsorption data has to be presented analytically. Several models have been developed for this objective.The thesis consists of two parts. In the first, a literature study on adsorption theories and thermodynamic assumptions for development of transient mass and energy balances is conducted. The models are discussed, and from this, the Langmuir approach is selected to be used for numerical calculations. The model is implemented into a lumped-parameter analysis describing an infinitesimal element within an adsorbent bed, allowing for neglecting heat leaks into the system as well as the structural steel mass. The second part describes the simulations conducted in the study. The numerical software COMSOL Multiphysics 4.2.a is used for numerical calculations. Modules for implementation of the transient mass and energy balances are considered, before Heat Transfer in Porous Media and Brinkman Equations are applied, for heat transfer, pressure- and velocity calculations, respectively. The simulations are run for different initial and boundary conditions. The porous material is defined with Fe-btc properties. The simulation model is built step by step, and problems encountered are analyzed continuously in the process towards a complete model. After completion, the model geometry is adjusted and the porous material is changed to MOF-5 properties, to resemble a selected published paper. Numerical results are compared and discussed. Modeling restrictions for the present study is accounted for, and all choices made when considering the assigned task are justified. The report is completed by listing the conclusions drawn from the present study, and concrete suggestions for further work are given.Simulation results found in the present study differs slightly from the published research work. Instabilities in the solver results in a temperature dip in the simulated domain. This leads to an increased adsorption rate. Furthermore, it appears that mass is not conserved, which means that the inlet velocity of the feed gas does not change as expected when the adsorption is disabled from the model.

A Pelton design software is currently being developed at the Waterpower laboratory at NTNU. The motivation behind this software is to streamline the parametric design process for Pelton turbines. A numerical flow model is a cornerstone in this application, but the lack of a bucket geometry and model runner has prevented the development of such a model. DynaVec, a turbine producer who specializes on sediment erosion and corrosion problems, offered to help by providing a bucket geometry and a model runner.The objective of this Master's thesis was to develop and validate a CFD model that predicts the torque applied to a non-stationary Pelton bucket, subject to a high-speed water jet. The numerical model was based on a method proposed by DynaVec, and the bucket geometry used in the simulations was identical (1:1) to the model runner.Numerous simulations were conducted, testing mesh dependency and different operational points (e.g. head). Mesh independence occurred at approximately 4.5 million elements. Furthermore, simulations of varying heads showed that the model may be independent of the head (40-80m), but this was not verified properly.Experiments showed that the numerical prediction was fairly accurate. A comparison of the numerical and experimental measurements showed that the CFD model over-predicts the torque by approximately 1.5%. This prediction was validated for the specific geometry used in the simulations, and a head of 75m.Overall, the results suggest that the numerical model is promising as a parametric design tool, but further development is required to obtain a true validation of the model.Task three and four were changed in agreement with Ole Gunnar Dahlhaug, because Solemslie's design program was delayed. In essence, the parametric study proceeded in favor of the development of a CFD model. To ensure that this work would benefit future research, especially students at the Waterpower laboratory, a detailed procedure for the CAD modeling, meshing and physical setup was included in the Appendix.

The incentive to reduce energy consumption in the industry is big, especiallyin high temperature systems. Heat pipes are of great interest for this purposedue to their favorable thermomechanical properties. This master thesis is apart of the ongoing study of Ph.D. candidate Geir Hansen, who is currentlydeveloping a rectangular heat pipe with potassium as working fluid at NTNU.The rectangular heat pipe is intended to be implemented in the walls ofelectrolysis cells as a part of a heat recovery system.The present work reports results of theoretical calculations of two importantheat transfer limitations, the incipience of boiling and the capillarylimitation for two types of nickel foam wicks. Results of experimental testscarried out on the cooling circuit for the proposed rectangular heat pipe arealso reported. The foam porosity, permeability and effective pore radius forwick 1 is 0.797, 31·10−12m2 and 62·10−6m, respectively, and for wick 2; 0.886,205·10−12m2 and 126·10−6m.A literature survey showed that porous coated surfaces improves the heattransfer and requires less superheat for boiling to commence. Calculationsperformed showed no danger of homogeneous nucleation in the proposed heatpipe. Boiling inside the nickel foam wick(s) were found to only be of concernfor wick 2 at high heat fluxes and a operating temperature of 600C.Calculations of the capillary limit showed that wick 2 is the best choicefor sustaining high heat fluxes. Increasing the wick length to 20cm madewick 2 not suitable for usage, and wick 1 was the best choice for increasedwick length. Combination of the two wick types showed to be very effectiveand significantly (factor of almost 4) improved the performance. An unevenheat flux distribution where a lower heat flux is at the bottom region of theevaporator is found to lower the performance, while a higher heat flux at thebottom region increases the performance.Early tests revealed that the PID controller was marginally stable, so thecontroller was tuned and stable operating conditions were achieved. Experimentsshowed that in order to get an accurate heat balance for the test rig,knowledge about the exact position of the thermocouples is needed.