I løpet av 2005-2006 ble det installert en ny kombinert varmepumpe/kjølemaskin i Kongsberg Teknologipark. Den kombinerte varmepumpen/kjølemaskinen er et totrinns ammoniakkanlegg som har som formål å levere prosesskjøling til Volvo Aero Norge (VAN) og FMC Technologies (FMC), som har produksjons- og testlokaler inne i teknologiparken. I tillegg leverer varmepumpen varme til Kongsberg Teknologiparks interne fjernvarmeanlegg. Dimensjonerende kjøleytelse og varmeeffekt er henholdsvis 1600 kW og 2200 kW. Anlegget er bygget med et akkumuleringsbasseng for å akkumulere kaldt vann med hensikt å utjevne de tidvis store effektuttakene hos VAN. Temperatursjiktning i akkumuleringsbassenget er ønskelig, slik at varmt vann til fordamperen kan hentes fra toppen av bassenget, og kaldt vann til prosesskjølingen kan hentes fra bunnen av bassenget. Det har imidlertid vist seg at sjiktningen i bassenget er dårlig.Formålet med masteroppgaven har blant annet vært å kartlegge energi- og væskestrømmer inn og ut av akkumuleringstanken, for å undersøke om sjiktningsproblemet skyldes stor væskeomrøring på grunn av store væskemengder inn og ut av bassenget. Det har også blitt utarbeidet et måleopplegg og en målerigg for temperaturmålinger i bassenget, og målinger har blitt gjennomført. I tillegg har det blitt gjennomført en analyse av både prosess og konsekvenser av effektvariasjoner i anlegget, og en analyse av og beregninger på foreslåtte tiltak for optimalisering av akkumuleringsbassenget med tanke på energigjenvinning. Hovedfokus er satt på den kalde siden av anlegget og akkumuleringsbassenget.Både Kongsbergs og egne målinger verifiserer sjiktningsproblemet ved at det er observert liten temperaturdifferanse mellom topp og bunn i akkumuleringsbassenget. En sammenligning av temperaturmålinger og målinger av væskestrømmen i bassenget har vist at sjiktning ikke oppnås, selv ved den minste målte væskestrøm på 169 m3/h. Om sjiktning skal forekomme må følgelig væskestrømmen senkes under dette nivået.Analysen av prosessen over måleperioden viser at det fra et prosesskjølings-perspektiv bør tilstrebes å holde kjølekapasiteten høyere enn effektuttaket til prosesskjølingen til enhver tid, og dermed opprettholde en jevn turtemperatur til prosesskjølingen. Dette for å unngå unødvendig store væskestrømmer som blir resultatet av en høy turtemperatur. Fra et energioptimerings-perspektiv ønskes på sin side så varmt vann som mulig inn på fordamperen, og så lav temperatur som mulig i fjernvarmekretsen, ettersom dette gir den beste energifaktoren, grunnet lavt trykkforhold.Det bør derfor gjøres modifikasjoner på akkumuleringsbassenget for å optimalisere temperaturene til fordamperen og prosesskjølingen, enten ved å fysisk skille kaldt og varmt vann, eller gjennom å muliggjøre sjiktning ved omkobling av rør i akkumuleringsbassenget slik at de store væskestrømmene i bassenget vil reduseres. En løsning der energifaktoren forbedres med 10,5 % er foreslått.

Natural gas is a vital component of the worlds’ supply of energy. In Norway all the gas can be found offshore where it needs to processed before transport. A vital part of the processing is gas and liquid separation. Current gas-liquid separators are big and expensive, and the offshore industry is looking for more compact separators. One of the more prominent technologies is the NNNGLseparator developed at NTNU. To further identify its possibilities, a thorough research program has been startedIn this thesis one-dimensional models describing pressure drop and separation performance of the NTNU Natural Gas Liquid Separator (NNGLseparator) for dispersed gas-liquid flows has been studied. Here modeling of separation performance was divided into cyclonic separation and droplet capture by the meshpad, and then combined in sequence. The droplet capture is assumed to occur before cyclonic separation.To analyze the impact of centrifugal force on droplet capture, the force was included in a previous proposed model describing droplet capture by a single fiber in the meshpad. With this as basis, modeling of total droplet capture for the meshpad was analyzed. Through this analysis we proposed an extension on the existing model for the case of the flow not following the rotation of the meshpad. The droplets that are not captured are then separated through cyclonic separation. To model this separation mechanism, a modified time of flight model was developed. The modification includes the mesh porosity, and a - factor describing the droplet’s reduced radial velocity due to the obstructing meshpad.Existing models for pressure drop across fixed porous media were compared to experimental data to identify which model best applies to the meshpad used in the NNGLseparator. How this model performs in describing a rotating porous media was then analyzed for the two scenarios; fully developed flow before mesh entry and developing flow inside a rotating meshpad. Through this analysis an extension to the pressure drop model was proposed, which includes the tangential velocity difference between rotating mesh and gas flow. A previous proposed model for pressure drop across wet mesh pad was reviewed. This led to a discussion on how liquid hold up differs in the NNGLseparator from conventional fixed meshpads. The proposed one-dimensional models were then analyzed through a parametric study of the separator performance in terms of pressure drop and efficiency of droplet separation for different flow conditions and geometries.

An investigation of the turbulent fluctuating kinetic energy dissipation in low Reynolds number channel flow is made, both analytically and numerically with means of Direct Numerical Simulation (DNS). The unsteady Navier-Stokes equations are solved at a Reynolds number of 360, based on the shear velocity and channel height, for four grid resolutions 48^3, 88^3, 128^3 and 192^3. The results are compared with data from Kim et al. (1987) [9], and good agreement is found for the 192^3 grid resolution.The viscous term in the kinetic energy equation is derived and described, from there the “isotropic” dissipation equation is shown to be the homogeneous dissipation equation which is compared with the thermodynamically correct dissipation. The results are in agreement with the findings of Bradshaw and Perot (1993) [2], with a difference of maximum ≈2.5% from the correct dissipation.The isotropic dissipation, often used as approximation in experiments, is also calculated and compared with the homogeneous dissipation. The results are un- surprisingly poor, and are only in agreement about the centerline. A comparison with an equation from the k-ε-model is also made, most as a curiosity, and also shows poor agreement.The Kolmogorov length scale is calculated from the dissipation, and it shows clear grid dependency even though the grid is smaller than the Kolmogorov length scale in the z-direction with max(∆z+/η+) = 0.8. The dissipation of the Reynolds stress components are used to create Kolmogorov length scales in x, y and z-direction. They are also grid dependent, even though max(∆z+/η+ ) ≈ 0.7. A length scale tensor analogous to the Kolmogorov length scale is proposed. It is based on the connection between the Reynolds stress equation and the turbulent fluctuating kinetic energy equation. It relaxes the grid restrictions compared to the Kolmogorov length scale, but investigation of its validity requires simulations with a super computer and is therefore not performed.

Hydrogenlagring er en nøkkelteknologi for etablering av en hydrogenøkonomi. Adsorpsjon er en av hydrogenlagringsteknologiene som for tiden utforskes sammen med hydrogen lagret som komprimert gass, væske, og i metallhydrider. Omfanget av denne oppgaven har vært eksergianalyse og utvikling av eksergilikninger for hydrogenlagring ved adsorpsjon ved kryogen temperatur. Eksergianalyse utført på kryo-adsorpsjonssystem er i hovedsak basert på det eksperimentelle oppsettet til Aleksic (2010) utført på NTNU.Fundamental informasjon om adsorpsjonssystemene er representert av adsorpsjonsisotermer. Disse viser lagringskapasitet til et adsorbent ved konstant temperatur som funksjon av trykk. I dette arbeidet er adsorpsjonsisotermer modellert med Langmuir og Sips likning med tilhørende parametere tilpasset etter eksperimentell data gitt av Aleksic (2010) og Paggiaro (2008) for adsorpsjonslikevekt ved utvalgte temperaturer og trykk. Modellene med Sips likning viser god overenstemmelse med eksperimentell data, og med prinsippene fra løsningstermodynamikk har adsorpsjonssystemet blitt beskrevet termodynamisk, med massebalanser, energibalanser og eksergibalanser. Oppgaven kan deles opp i tre deler. I den første delen er det introdusert relevant informasjon om utvalgte hydrogenlagringsteknologier og termodynamikk tilhørende adsorpsjon, inkludert adsorpsjonsisotermer for relevante adsorbenter. I den neste delen er det utledet termodynamiske likninger for masse-, energi- og eksergibalanser for kryo-adsorpsjonslagringssystem og kryo-kompresjonslagringssystem. Oppgaven avsluttes med et kapittel som inneholder fire eksergianalyser. I den første analysen studeres ekserginivå i adsorpsjonssystem mot utvalgte teknologier, og i den neste analysen studeres eksergibidrag i oppsettet til Aleksic (2010) i forhold til et kryo-kompresjonssystem ved tilsvarende trykk og temperatur. Den tredje analysen er utført av oppsettet til Aleksic (2010) ved fylling og tapping, og sammenlignet med kryo-kompresjonslagringssystem operert ved tilsvarende tilstander. Kapittelet avsluttes med analyse av lagring av hydrogen i kryo-adsorpsjonsbeholder, analysert med utgangspunkt i temperatur-, trykk- og masse-utvikling ved lagring hentet fra arbeidet til Paggiaro (2008). Analysen er beriket med sammenligning av andre lagringsmetoder ved samme operasjonsforhold: en lavtrykkstank med flytende hydrogen og en høytrykkstank fylt med enten flytende hydrogen eller komprimert hydrogengass ved 80 K.Teknologien for adsorpsjonslagring av hydrogen er på et eksperimentelt stadium, og har vist seg gjennomførbar. Teknologien har fordel av operasjon ved lavt trykk (typisk 20-30 bar) i forhold til kompresjonssystemer (300 – 700 bar), og oppnår likevel sammenlignbar volumetrisk og gravimetrisk lagringskapasitet. Teknologien er likevel langt fra å oppnå hydrogenlagringskravene fra U.S DoE for 2015. Dette er også tilfellet for de andre hydrogenlagringsmetodene som er under utvikling: komprimert gass, hydrogen som væske og hydrogen lagret i metallhydrider. Enkle forbedringer i disse teknologiene ser ikke ut til å føre til at kravene nåes. Adsorpsjonssystemet har i tillegg andre karakteristikker som kan hindre kommersialisering av teknologien (Paggiaro 2008), og som vil bli diskutert i oppgaven. Eksergianalyseresultatene viste at i adsorpsjonslagringsystemer består den totale eksergimengde i hovedsak av kjemiske eksergi og omtrent 7–10 % termomekanisk eksergi. Den termomekaniske eksergien kan delvis utvinnes ved installasjon av turbin for utnyttelse av trykket i beholderen. Resterende termomekanisk eksergi kan teoretisk utvinnes ved en varmekraftmaskin som benytter beholderen som kuldereservoar og omgivelsene som varmereservoar. Ved implementering av slike systemer er det av viktighet å minimere økningen av systemets kompleksitet som helhet, som kan vise seg vanskelig. Derfor mener forfatteren at det er ønskelig å minimere termomekanisk eksergi i lagringssystemet og maksimere den kjemiske eksergien. Kjemisk eksergi er direkte avhengig av mengde hydrogengass lagret i systemet, og derfor er teknologiens lagringskapasitet viktigste parameter for å maksimere det kjemiske eksergiinnholdet. Lagringskapasitet i adsorpsjonssystem viste seg å være sterkt temperaturavhengig. Ved en temperaturøkning på 20 grader (fra 77-97 K) faller lagringsevnen i oppsettet til Aleksic med adsorbent Cu-BTC og NORIT R0.8 med omtrent 21 % ved 2 MPa. Det ble utført eksergianalyser av fylle- og tappeprosesser med utgangspunkt i eksperimentell data fra oppsettet til Aleksic (2010) med adsorbent NORIT R0.8. Ved fylling viser utregningene et tap i lagringskapasitet på 15,1 % grunnet generasjon av varme i prosessen. Ved endt tappeprosess er 59,0 % av den opprinnelige hydrogenmengden igjen i lagringstanken, grunnet mangel på intern varmekilde. Forslag til forbedringer er foreslått i oppgavens avsluttende kapittel.Eksergianalyse er utført av hydrogenlagring i kryo-adsorpsjonssystem over tid og sammenlignet med lagring i utvalgte alternative lagringsbeholdere. Resultatene viser at varmelekkasje inn i tankene som er undersøkt fører til trykkøkning i samtlige tanker og ventilering av hydrogengass. Adsorpsjonsbeholderen viste tilsvarende, og noe høyere ventileringsrate av hydrogen enn høytrykksbeholderen med komprimert hydrogen. Beholderen som viste best lagringsevne var høytrykksbeholderen fylt med flytende hydrogen.

This report documents the adaptation of a two dimensional two phaseNavier-Stokes solver to axisymmetric problems. The changes fromCartesian to cylindrical coordinates are thoroughly described with finitedifference methods for the heat equation, Poisson equation, singleand two phase Navier-Stokes equations.The jump conditions at interfacesare modified to accommodate these changes for the two phaseNavier-Stokes equations.The changes to the solver are done step by step, and every changeis verified through intermediate test cases with analytical solutions tolimit the possible sources of errors.Finally all stepwise changes are joined together to form an axisymmetrictwo phase Navier-Stokes solver. Results are presented for aresting bubble, and for both viscous and inviscid oscillating ellipticbubbles.

Simulations on liquid loads and flow rates which the Taitel-Dukler model predicts to have multiple solutions have been performed with ANSYS Fluent and LedaFlow. Both steady state and transient results in one, two and three dimensional flows are reported in this work. The hypothesis that the holdup of a pipe operated in the multiple solution region will be determined by the downstream holdup is investigated. Some results indicate that the hypothesized interface level gradients effects are correct.The Fluent steady state simulations had mass imbalance issues in addition to being both grid and geometry dependent, but produced results consistent with the independent Fluent transient simulations. The one dimension LedaFlow solver illustrated the effect shear stress modeling have on the multivalued solution region. The solver chose the intermediate solution for some flow rates, which by physical arguments can be excluded. The novel solver LedaFlow Q3D produced transient results displaying the wavy surface of the high holdup solutions. The results from the different models are deviating, but it is hard to predict which results are most the accurate since no comparison with experimental results have been conducted.

Sediment erosion in Francis turbines is a big problem in hydropower plants in and around the Himalayas. The sediment composition in the rivers contains high levels of the hard mineral quarts. When the sediments enter the turbine they cause erosive damage to exposed parts such as covers, guide vanes and runner. The sediment concentration is at its highest during the monsoon period. During this period some turbines are stopped when the sediment consentration reaches certain levels to reduce the damage. Jhimruk power plant in the mid-western part of Nepal is a good example of how the sediment erosion affects the operation of a power plant. During the monsoon period the turbines can be eroded to an almost unrepairable state. The turbines have to go through substantial annually maintenance. A result of this is reduced power output and high maintenance costs. It is therefore of interest to design a new Francis turbine that can better withstand the sediment erosion. A cooperation project between Kathmandu University and The Norwegian University of Science and Technology was started as a part of the RenewableNepal project which aims to develop and start manufacturing of erosion resistant Francis turbines.A parameter study of different blade designs have been performed to find a more erosion resistant design. In this thesis FSI analyses have been performed on three different designs to verify their structural integrity. The designs transfers the hydraulic energy from the water to the blade in different sections. The results showed a stress distribution which coincided with the energy transfer along the blade. The reference design was analyzed with two different blade thickness. For all the designs the stress was relatively low compared to the criteria for hydraulic turbines.

This thesis presents an off-design simulation of offshore combined cycles. Offshore installations have a substantial power demand to facilitate the oil and gas production. To cover this need of power almost all the platforms use one or several gas turbines, often described as a simple cycle. However, because of high taxes on emissions, and increasing gas prices, more efficient technologies have been reviewed. One solution has been installing combined cycles (CC) offshore. Between 1999 and 2000 three combined cycles were installed on the Norwegian continental shelf and are still in operation. A combined plant may operate for prolonged time at off-design conditions, depending on power demand, ambient condition offshore. First, this thesis gives a description of combined cycles from a thermodynamic and technical point of view. A study of existing offshore combined cycles is performed, and some of the implications of using combined cycles offshore are discussed. In the study, also off-design performance regarding the gas turbine and steam cycle is presented. Further, the simulation tool GTPRO is used to model two CC plants, one designed for offshore installations, and one designed to achieve high efficiency. As part of the design process a sensitivity analysis is performed to find a good trade-off between efficiency and weight for the offshore plant. The model showed good agreements compared with the existing offshore plants, with a power output of 50.3MW, plant efficiency of 50.3%, and similar weight of the skids. The high efficient plant, based on the same gas turbine, and the same assumptions produced 53.1MW. This model gained 2.4MW more in power output, however with a penalty of 209 ton in extra weight.To review the plants performance and operability, off-design simulations were performed in GTMASTER. Both part load and changing ambient temperature were investigated. The results showed that both plants had similar behavior in performance at off-design, and that the GT strongly dictates the behavior of the steam cycle. At part load the relative SC efficiency increases, resulting in general high plant efficiency. At 60% GT load, the relative gas turbine efficiency is 81% compared to the relative plant efficiencies of about 90%. The difference in efficiency between the high efficient plant and the offshore plant remains constant at part load. The result from the simulations of ambient temperature is that none of plants will achieve higher plant gross efficiency at changing ambient temperature. The best plant efficiency occurs at design point. However, both plants have a long interval with approximately 100 % plant efficiency. From 15 to 0°C, the relative SC gross efficiency drops with 5 %, and the relative GT efficiency increase with 2%. However, the power output changes for both the GT and ST. From 28°C to about 0°C the power output increase almost linearly for the SC and GT.

Rock Energy is a Norwegian company with a patented solution for drilling deep geothermal wells, for exploitation of deep geothermal energy from Hot Dry Rocks. The concept involves a drilled sub-surface heat exchanger, referred to as cross wells. The concept is well suited for production of heat for direct heat applications. In this thesis an analysis of the existing district heating plant at Oslo Airport Gardermoen has been conducted, together with examining possibilities of implementing geothermal energy as base load at the plant. A geothermal design that could meet the needs of the district heating plant has been established, and for evaluating the geothermal system in an environmental perspective an analysis based on LCA methodology has been conducted. Hafslund operates two district heating centrals at Gardermoen (Gardermoen heating central and a smaller mobile central) for which both have been analyzed to determine the potential for implementing deep geothermal energy as base load for the systems. Gardermoen heating central is connected to the airport and to the area close to the airport. This central is again connected to the mobile heating central, which is situated near the industrial estate south-east of the airport. Based on Hafslund’s production data from February 2011 to January 2012, a heat load duration curve for the two existing centrals have been established. When adding the two curves together the duration curve show a maximum load of 25,7 MW at present, and a yearly energy production of 74 GWh. The mobile central accounts for only 7,2% of the total load and heat production at present.Future heat demand in the Gardermoen area is expected to increase beyond existing capacity. Hafslund is therefore considering to increase the capacity of both their district heating centrals. The enlargement plans involves that the heating central will be expanded to a design load of 37,4 MW (24 MW at present), while the mobile central need to be increased to a design load of 15,2 MW (1,7 MW at present). Assessment of the geothermal installation showed that it is preferable to include the geothermal system in the base load of the mobile central. The additional geothermal capacity will cover 10 MW, and thus deliver 65% of the required heat load and 90% of the energy production from the mobile central. The geothermal installation was designed using the spreadsheet “Geocalc”. The outputs from Geocalc are used in an analysis of the environmental performance of the designed system through a Life Cycle Assessment (LCA). LCA introduces a technique to assess environmental impacts associated with all stages of a product’s life from “cradle to grave”. The report aims at giving normative results for the environmental impacts of a geothermal installation at Gardermoen. The method provides the ability to quantitatively compare results to other sources of heat provision processes for district heating. It is important to emphasize that the analysis has provided an overview of the potential environmental impact, and not necessarily the actual results of environmental consequences. The system analyzed has a thermal output of 10 MW, lifetime of 30 years, 5000 annual operating hours. The functional unit of district heating produced is kWh. The analysis is based on the main contributing processes to construction, operation and demolition of Rock Energy’s geothermal system. The district heating grid is not included in the analysis, as it is already in place at the site. Each contributing process has been systematically validated. It is however uncertainties associated with the data collection mainly due to contradictory information gathered. The information considered to be mostly uncertain is the energy consumption used for drilling purposes.Possible scenarios for the energy supply to drilling were established. These scenarios were simulated in a system model in Excel. The model is based on data and information gathered from existing literature, the database Ecoinvent, published reports and personal communication with drilling experts and specialists within the relevant fields of study. The results are assessed for the following impact categories: Climate change, metal depletion, fossil depletion, terrestrial acidification and freshwater eutrophication. The evaluated potential energy sources for the drilling operation are electricity from the Norwegian grid, electricity from the European grid, and diesel. The climate change category has especially been in focus when conducting the simulations and this category shows large spread in the results, from 0,9993 g CO2-eq/kWh for the best scenario to 23,6 g CO2-eq/kWh for the worst scenario. As expected, the analysis concludes that electricity from the Norwegian grid for the drilling is preferable. For a geothermal system in Europe, the results show that it would be advantageous to use diesel as energy supply for the drilling operation instead of European electricity mix, for which the emissions are doubled.For the metal depletion impact category, the variation of energy supply to drilling cause the least fluctuation. This is also the only impact category where the Norwegian electricity mix has higher impacts than for the diesel consumption. This can be explained by the infrastructure related to electricity transmission. The results of the study have been compared to other heat sources for district heating (waste incineration, biofuel and solar thermal). The comparison shows that from an LCA perspective geothermal energy based on Rock Energy’s concept is an environmentally friendly energy supplier for district heating. The studies compared are however based on varying assumptions, and thus a generalized conclusion cannot be drawn from this.

The goal with this master thesis was to establish Hill diagrams and improve a Kaplan turbine intended for use in Afghanistan. The turbine efficiency has been tested in setting 1 and 2. Turbine efficiency in setting 3 and 4 could not be tested because the runner blades interfere with the housing making it impossible to rotate the turbine. The efficiency was tested with an effective pressure head ranging from 2 to 8 meters. Best efficiency point was not reached because of limitations in the test rig making it impossible to reach a lower effective head. The best efficiencies tested in the two different settings are presented in the table below together with the uncertainty in the actual test point. All tests are done according to the IEC standard for model testing of hydraulic turbines. The computational fluid dynamics (CFD) simulations done on the inlet bend indicates that the bend should be rounded and flow controllers should be extended over the entire bend. This should be considered to get a more even velocity distribution at the inlet of the guide vane. An alternative placement of the lower bearing was designed but is discarded because of the disadvantages the modification leads to. High wear due to sand erosion on the seals causing high maintenance and costly stops makes the solution not optimal for use in water with high sand content. The runner blade design is checked against the design procedure presented by Professor Hermod Brekke in Pumper og Turbiner and found to be satisfying. It is concluded that time should rather be spent on optimizing the inlet of the turbine. Fluctuations in the measurements make it necessary to change the measuring equipment or search for error in the existing equipment before further tests can be carried out. In order to be able to test in setting 3 and 4 the runner needs to be placed while the blades are fixed in setting 4.