This report presents opportunities for use of CO2 as refrigerant in the air conditioning system in public trains. The CO2 system shall provide cooling in the summer and heating in the winter. CO2 is a natural fluid which means that it exists naturally in the biosphere. Today 75% of the air conditioning systems in trains use R134a as refrigerant. The GWP of R134a is 1410 while CO2 used as refrigerant is 0. A replacement from R134a to CO2 gives possibilities of large environmental savings. Three different technical system solutions of the heat pump are presented, each with its own method of provide cooling and heating. Solution I changes between cooling and heating by change the direction of the refrigerant flow through the system. Solution II changes between cooling and heating by change the configuration of the air streams through the heat exchangers. In Solution III the refrigerant flow direction and the configurations of the air streams is always the same. The whole heat pump is placed on a rotatable unit and the change between cooling and heating is done by rotating the whole heat pump 180°. In all the three technical solutions there are separated heat exchangers for fresh and exhaust air. This gives an energy efficient system which recover heat from the exhaust air. Computer simulation shows that a system solution with one evaporation pressure and one stage compression is problematic for low ambient temperatures; the system must stand temperatures to 40 °C. A system solution with two levels on the evaporation pressure and a two stage compression showed to improves the COP from 1,7 to 3,2 when the ambient temperature at 40 °C. A railway coach need cooling when the ambient temperature is above 20 °C and heating below 15 °C. Norway is a country with cold climate. Weather statistic show that a train which drives in Oslo every day from 0600 to 1800 throughout a year will need cooling 3% of the time and heating 83% the time. This heating should be done by a heat pump and not with electrical heating as today. Results of the computer simulation shows that the annual energy consumption of heating the train will be reduced by 78 % if the designed CO2 heat pump is used in stead of electrical heating.

Vane pack demisters in the industry operate with natural gas at pressures up to 100 bara. A new vane pack has been compared with the traditional one used by the industry. The vane packs have been investigated through experiments and Computational Fluid Dynamics, CFD. The fluid flow inside a vane pack consist of turbulence and two fluid phases. The simulations were carried out with a Large Eddy Simulation model and a Scale-Adaptive Simulation model. Phenomena observed in the experiments were confirmed by CFD. A transient Discrete Phase Model,DPM, that should be capable of modeling the generation of a liquid film together with droplets was used. The DPM simulation gave a mist flow pattern that agreed with the one observed in the laboratory. Separation efficiency measurements of the two vane packs using Exxsol D60 as liquid and SF6 as gas were performed at different pressures, ranging from 1 to 8 barg. This corresponds to natural gas density ranging from 8 to 65 barg. Non of the experiments achieved the specification given by the oil and gas industry. The efficiency measurements did show that the pressure had a great influence on the performance. The low pressure measurements were the only experiments which had a efficiency above 97% at a k-value between 0.2 - 0.25 m/s.

A multiregional input-output model representing the world in the year 2000 was constructed based on statistical data, and combined with process specific data on a primary aluminium supply chain, to create a model of the global primary aluminium industry. Using input-output methodology, total emissions of eight substances due to primary aluminium production, their size and origins, were estimated and expressed in terms of global warming potential (GWP) and acidification potential (AP). Simulations from 2000 to 2030 were run based on final demand estimates from external GDP projections and three assumed development scenarios. The baseline, scenario 0, assumed no changes in technologies or relative production and trade patterns - only the model's response to the expected change in final demand was analyzed. By contrast, both scenarios 1 and 2 assumed that the additional aluminium production predicted by the baseline would be produced exclusively in China. Scenario 2 employed the added assumption that the Norwegian aluminium production would experience a steady decline from its 2000 level to zero by 2030. The baseline scenario showed rapidly increasing aluminium output towards 2030, following the expected GDP developments. Emissions followed the same trend, increasing about 3.3 times over the three decades. As for total cradle-to-gate impacts of primary aluminium production, the model showed large variations from one region to another. Emissions per ton of Chinese primary aluminium were high relative to most other regions, hence the total global GWP and AP from primary aluminium production rose more rapidly in scenarios 1 and 2 than in scenario 0. By 2030, the GWP in scenarios 1 and 2 were 11.4% and 12.5 higher than in the baseline, while AP were 50.0 and 51.9 higher.

Subsea processing can make production from otherwise unprofitable fields profitable. In subsea processing controlled cooling of the process fluid will often be required. Robust and simple solutions are desirable in subsea processing. Coolers that rely on natural convection from the surrounding seawater are therefore interesting, but control of the process fluid outlet temperature is hard to obtain in such coolers. In this study a calculation model for subsea coolers has been developed. The commercial software MATLAB has been used for developing a program. Heat transfer and frictional pressure drop correlations have been studied and recommendations are made for the model. The model is based on tubes in parallel, and the tubes can be oriented vertically or horizontally. The program allows for open, semi-open and closed arrangements on the waterside, and both natural and forced convection is implemented. The program has been tested through simulations of two test cases and found to be performing as desired.

The performance of a hot wire probe with three wires is investigated for two different flow cases. The wires are made of a platinum/rhodium alloy, and has a diameter of 5 micrometer. The three wires make a probe volume with acrossection of approximately 5 mm. A cosinus fit using the effective angle method gives a deviation of plus/minus 1 degree for a variation of yaw angle equal to plus/minus 20 degrees. First the probe was tested in a fully developed turbulent pipe flow, for Re_D = 10^5. Good results were obtained for |y/R|<0.8, both for mean velocities and turbulent stresses. Closer to the wall the mean flow gradient was to large relative to the probe resolution, giving large errors. The second flow case was a cylinder wake. A traverse of the flow at x/D = 10 was performed at Re_D = 3*10^3. The mean velocities and turbulent stresses was partly found to be in qualitative agreement with results found in litterature. The shear stresses uw and vw were however found to be unphysically large, this is belived to be due to the velocity gradient in the wake. Conditional averaging of the wake results with respect to shedding frequency was also conducted.

The process used by the Norwegian company Hydro for making Vinyl Chloride Monomer (VCM) from natural gas and sodium chloride has been studied. A three dimensional CFD model representing the firebox of the EDC cracker has been developed using the commercial CFD tool Fluent. Heat to the cracker is delivered by means of combustion of a fuel gas consisting of methane and hydrogen. In the developed CFD model used in this work, the combustion reaction itself is omitted, and heat is delivered by hot flue gas. With the combustion reaction left out, the only means of tuning the CFD model is through the flue gas inlet temperature. With the flue gas inlet temperature near the adiabatic flame temperature, the general temperature level of the EDC cracker was reported to be too high. The outer surface temperature of the coil was reported to be 3-400 K higher than what was expected. By increasing the mass flow of flue gas and decreasing the temperature, the net delivered heat to the firebox was maintained at the same level as the first case, but the temperature on the coil was reduced by 100-150 K. Further reductions in the flue gas inlet temperature and modifications in the mass flow of flue gas at the different burner rows, eventually gave temperature distributions along the reaction coil, and flue gas and refractory temperatures, that resemble those in the actual cracker. The one-dimensional reactor model for the cracking reaction represents the actual cracker in a satsifactorily manner. The cracking reaction was simulated using a simple, global reaction mechanism, thus only the main components of the process fluid, EDC, VCM and HCl, can be studied. The model is written in a way suitable for implementation of more detailed chemical reaction mechanisms. The largest deviation in temperature between measured and simulated data are about 5%. At the outlet the temperature of the process fluid is equal to the measured data. The conversion of EDC out of the firebox is assumed to be 50 wt-%, this value is met exactly by the model.

This master thesis describes a quadratic programming model used to calculate the spot prices in an efficient multi-area power market. The model has been adapted to Northern Europe, with focus on Denmark West and the integration of large quantities of wind power. In the model, demand and supply of electricity are equated, at an hourly time resolution, to find the spot price in each area. Historical load values are used to represent demand which is assumed to be completely inelastic. Supply is modeled according to the type of generation: Thermal generators are represented by piecewise linear, upward sloping, marginal cost curves. Historical wind generation data is used to model the fluctuating wind power output, and wind power is considered to have zero marginal cost. Hydro power is modeled by one aggregate reservoir for Norway and one for Sweden; the marginal cost of hydro power is set as a function of the difference between the reservoir level and the historical median reservoir level. Additionally, decentral combined heat and power plants in Denmark are considered to operate irrespective of the market. Six separate price areas constitute the model: Denmark West, Denmark East, Norway, Sweden/Finland, Germany, and Central Europe. The areas are modeled as having no internal bottlenecks and are connected by tie-lines constrained by active power limits. This report quantifies the impact the installed wind power capacity has on the power price in Denmark West by scaling up the wind power output in the model. Because wind power has a marginal cost close to zero, it will force prices down. The effect will be most prominent during high wind speed hours in a power system with substantial amounts of wind power. Results show that the impact is modest; average power prices fall by only 10% if the installed wind power capacity is doubled, and thermal generation will set the power price in all hours until wind energy exceeds 50% of domestic demand in Denmark. Since prices fall the most during hours with high wind power output, income to wind turbine owners will decline quickly as the installed capacity becomes large. The effect is most pronounced at wind energy shares above 40%, thereafter the income -- per MWh sold -- falls rapidly. In absence of government subventions, this effect will limit the economically viable level of installed wind power capacity. Expansion of the cross-border transmission capacity and higher thermal generation costs can both help offset the income reduction to wind turbine owners from higher wind power penetration. Alone, a 30% increase in thermal generation costs can allow 50% of wind energy and still retain todays income to wind turbine owners. Use of the Norwegian hydro reservoirs to balance out fluctuations in wind power output is found to stabilize and reduce the price. This benefits both consumers and wind turbine owners in Denmark. Expansion of transmission capacity to Norway will further stabilize the price; a new 1000MW cable lets the Danish market easily accomodate 50% wind energy. With lower and more volatile prices as a result of high wind power penetration, a load can profit by being flexible. Water electrolysis is one such load; it uses electricity to produce hydrogen, and production can quickly be ramped up and down in accordance with the power price. Presently, steam methane reforming is the least expensive method of producing hydrogen, but with higher wind power penetration, electrolysis might become competitive. Using a previously developed model to assess the cost of electrolysis, in combination with the power market model developed here, this report finds that wind energy must exceed 85% of domestic demand in Denmark, combined with higher natural gas prices, for electrolysis to break even with steam methane reforming.

A measurement of the hydraulic turbine efficiency at the Vessingfoss hydro power station by the thermodynamic method has been attempted, but has not given the desired results. Two problems have been encountered. The high pressure side temperature measurements show an abnormal scatter resulting in standard deviations of sy=0.05ºC. The reason for the scatter may be temperature layers in the reservoir lake Nesjø. This theory has been investigated, but needs further work. The other problem has been the mechanical strength of the low pressure side collector probes. Two different collectors have been tried, and both have broken down. The second attempt was made with a collector design based on wire rope, which failed because the turnbuckles were under-dimensioned. With proper dimensions, this solution is interesting in the future, as it was easy to install and may contribute to lose collector weight. The relative turbine efficiency has been calculated based on pressures and levels measured during the thermodynamic test. An uncertainty analysis of the result has been carried out. The head loss has been calculated based on technical drawings of the penstock and loss coefficients from the literature.

A multi-reactor, multi-temperature, waste-heat driven biomass thermochemical converter is proposed and simulated in the process simulation tool Aspen Plus™. The thermochemical converter is in Aspen Plus™ integrated with a gas turbine power cycle and a combined fuel cell/gas turbine power cycle. Both power cycles are recuperative, and supply the thermochemical converter with waste heat. For result comparison, the power cycles are also integrated with a reference conventional single-reactor thermochemical converter, utilizing partial oxidation to drive the conversion process. Exergy analysis is used for assessment of the simulation results. In stand-alone simulation, the proposed thermochemical shows high performance. Cold gas efficiency is 108.0% and syngas HHV is 14.5 MJ/kg on dry basis. When integrated with the gas turbine power cycle, the proposed converter fails to improve thermal efficiency of the integrated cycle significantly, compared to reference converter. Thermal efficiency is 41.8% and 40.7%, on a biomass HHV basis, with the proposed and the reference converter respectively. This is despite superior cold gas efficiency for the proposed converter, and the gas turbine cycle is found not to be able to properly take advantage of the high chemical energy in the syngas of the proposed converter. When integrated with the combined fuel cell/gas turbine power cycle, the proposed converter significantly improves the thermal efficiency of the integrated cycle, compared to the reference converter. Thermal efficiency is 56.0% and 51.2%, on a biomass HHV basis, with the proposed and the reference converter respectively. The fuel cell is found to be able to take advantage of the high chemical energy in the syngas of the proposed converter, which is the main cause of increase in thermal efficiency. Operation of the proposed thermochemical converter is found to be feasible at a wide range of operating conditions, although low operating temperatures in the converter may cause problems at very high carbon conversion ratios.

In this work, a 23-1 designed experiment has been performed to evaluate the effect of selected operating conditions on PEMFC performance and durability. Relative humidity, clamping pressure and back pressure were studied at two levels for Gore MEAs and GDLs. Two replicated experiments were performed. An ON/OFF test cycle was used to accelerate degradation. Total duration of the tests, after a break in procedure suggested by Gore, was ten days. In addition to sampling of voltage and current response and ohmic resistance, effluents were manually sampled from both electrodes every 24 hours and analyzed. Experiments with low humidification levels showed inferior durability. The combination of high relative humidity (100 %), high clamping pressure (10 barg) and high back pressure (1.5barg) result in the best performance and the lowest degradation rate. Results indicate that relative humidity is important both for performance and durability. Generally, fluoride emission rates (FER) showed an increasing trend with time. Higher rates were observed at the cathode. For the experiment with low relative humidity (25 %), low clamping pressure (5 barg) and high back pressure (1.5 barg) FER was significantly higher compared to the other experiments. For all tests the sulfur emission rates (SER) are initial high. Rates are higher at the anode. For the experiment with high relative humidity, low clamping pressure and no back pressure, the SER was significantly higher than for the other experiments. The sustained high levels of sulfur are probably a result of sulfuric acid residue from production of the MEA and/or GDL. High humidification of gases appears to more effectively wash out the sulfur.