Search

The Flexible Sandtraps (FlekS) project has been a four-year research project on upgrading of existing pressurized sandtraps in operational hydropower plants. Upgrading of existing sandtraps have several challenges compared with construction of new sandtraps; (i) limitations imposed by the existing infrastructure and (ii) the need to limit downtime to avoid production losses. The purpose of FlekS has thus been to develop cost-efficient solutions that can be installed with limited downtime to retrofit existing sandtraps in operational hydropower plants. The project has been divided in three main activities; (A1) literature review and mapping of challenges in the hydropower industry, (A2) numerical simulations of preliminary concepts, and (A3) physical modelling of the most promising concepts. The literature review and mapping of challenges in the industry reveals that hydropower plants comprising over 8000 MW equaling about 25% of the total installed capacity in Norway have experienced challenges with silt, sand or gravel transported during operation. The challenges range from slightly faster than expected wear on the turbine to severe damages on the turbines and operational restrictions. Almost all these hydropower plants have existing pressurized sandtraps at the downstream end of the unlined headrace tunnel. Common for all is that a well-functioning sandtrap may be able to mitigate the challenges. To find potential concepts for upgrading of existing sandtraps, numerical simulations and physical scale modelling have been applied. The sandtrap no. III at the 960 MW Tonstad hydropower plant has been used as a case study. Several field trips have been conducted and detailed information from the case-study has been made available, including a full 3D scan, velocity profile measurements from three ADCPs, pictures and videos taken during operation from a camera installed inside, and samples of the trapped material inside. The field data has been used to calibrate the numerical simulations and physical models. These tools have thereafter been used to test several different concepts for upgrading of existing sandtraps. In conclusion, it is found possible to upgrade existing sandtraps with limited downtime and costs. The main conclusions from the FlekS research project are presented below. 1. Hydropower plants comprising over 25% of the total installed capacity in Norway have experience challenges related to transport of silt, sand and gravel during operation. Many existing sandtraps do not function satisfactory. 2. The FlekS-project confirms previous research that the closed type of sandtraps is superior compared with the open type of sandtrap for trapping of bed load. The main difference is that the closed type successfully separates the deposited material from the main water flow, preventing further transport as bed load or resuspension. 3. An efficient solution for upgrading of open sandtraps to closed sandtraps has been developed. The solution entails limited construction works in the downstream end of the sandtrap and does not require expanding of the sandtrap volume. This solution has been recommended for the upgrading of the case-study sandtrap at Tonstad power plant. Physical model tests at TU Graz indicate that this solution can increase the trap efficiency from 0% to 90% for particle sizes in the range 0.3 to 1 mm. Physical model tests at NTNU for particle sizes with d50 equal to 3 mm do not show any significant different in the trap efficiency for tests with and without ribs (about 87% for both situations). 4. The proposed solution includes the use of a flushing system, which is necessary to enable the minimum reconstruction works inside the sandtrap. Flushing systems are in general found to be a promising measure when upgrading existing sandtraps. Most of the Norwegian sandtraps reviewed in this work do not have a flushing system and require dewatering and manual removal of the deposited material. Flushing systems allow emptying of the deposited material without dewatering the sandtrap and will hence reduce both outage and the stress on the tunnel stability resulting from a dewatering. 5. The results concerning the effect of flow calming structures on the trap efficiency of sandtraps are inconclusive. Some tests show a positive effect, and some show a negative effect. A concept with a passable flow calming structure was developed, allowing personnel and machines to pass the structure without dismantling during dewatering. 6. Heightening of the downstream weir was not found to have any effect on the trap efficiency. However, it can have a positive effect by preventing free surface flow with high velocities in the sandtrap. 7. Geometrical improvements are usually possible in existing sandtraps. Sudden expansions, contractions and abrupt changes in the geometry will cause turbulence and can decrease the trap efficiency of sandtraps. 8. For power plants where the sandtrap is constructed in conjunction with a surge tank, a new innovative solution for combined upgrading of the surge tank is developed. The “semi-air cushion surge tank” is able to increase the capacity of both the upper and lower surge chamber of two-chamber surge tanks with only expanding the volume of the lower chamber. Upgrading of the sandtrap may enable upgrading of the installed capacity in hydropower plants. 9. Physical model tests were not able to confirm any significant difference in trap efficiency for stable operation and transient operation of hydropower plants. However, only one test to investigate this effect was conducted and further research is necessary to conclude. The results presented above are primarily based on the results from the physical scale models. However, it is experienced that numerical simulations and physical modelling and variations of the two methods gave significantly different results. Three different numerical simulation software were applied by different users and with different setup, and two different scaling methods for the physical modelling were applied. By comparing with the sand deposition in the field measurements, the results from the physical scale model from TU Graz seems to be most representative. Developing scaling methods has not been an objective in the FlekS-project, and further research is necessary to investigate the accuracy of the two different scaling methods. Tests in the TU Graz model with the equivalent same sand size as applied in the NTNU model are planned to check if the difference in results is still significant. Tests are also planned to be conducted according to the Euler scaling method with lightweight material (natural sand is not possible at scale 1:37 owing to cohesion). Also, field data from the future case-study Tonstad sandtrap with installation of the ramp and a short section with ribs, will provide more information to validate the physical scale models. Further investigation of the numerical simulations methods is also required. The turbulence models and necessary resolution, in addition to the particle simulation and two-way coupling between water and particles needs further testing to give credibility as a design method for pressurized sandtraps. A PhD project in HydroCen will continue parts of the work described in this report. The physical models, the field data, and the 3D geometry are still available for future research, teaching and master thesis work. © NTNU 2021 The publication may freely be cited with source acknowledgement.

AlternaFuture has been a multidisciplinary research project in Hydrocen to investigate the potential of extreme upgrading of existing hydropower system with a positive effect on the environmental conditions. The project is a desk study and is carried out through developing future scenarios of an extreme upgrading of an existing hydropower system, to create potential for new innovations and solutions from the multidisciplinary scientists within the project. The existing hydropower system in the Mandal river has been applied as a case-study, where the current situation has been the baseline for developing alterative future scenarios and evaluating the economic and ecological results. The project has been divided in three main activities; (A1) mapping the current situation and defining environmental restrictions for reconstruction, (A2) developing physical scenarios consisting of hydropower projects, environmental projects and use of new innovations, and (A3) economic and environmental evaluation of the scenarios. Three main scenarios were developed for the Mandal river; (1) triple installed capacity, (2) maximum flexibility, and (3) flood protection. The hydropower optimization program ProdRisk was used to compare the hydropower operation and water management in present situation with the three scenarios. Energy price forecasts from HydroCen where used to assess the economic income from hydropower production, which in turn where compared with the estimated construction costs of each scenario to consider the economic feasibility. Thereafter, the impacts on the environmental status and recreational value in different parts of the watercourse, including reservoirs, lakes and river reaches are evaluated for three hydropower scenarios and compared with the present situation. In conclusion, it is found possible to realize extreme upgrading of existing hydropower systems, and at the same time in sum have a positive effect on the environmental conditions. It is noted that the positive effects require a significant effort in mapping and planning the environmental measures. For such upgrading consisting of multiple projects, single projects that have severe negative ecological impacts must be cancelled and cannot be included in the final scheme. Planning of such upgrading projects therefore must include environmental experts from the very beginning. The main conclusions from the AlternaFuture research project are presented below. 1. Extreme upgrading of existing hydropower systems can be done while also in sum improving the environmental conditions. 2. Extreme upgrading of hydropower systems that include pumped storage plants are economically feasible if the energy price variability increases sufficiently. For the Mandal river, the necessary increase is in between the 2030 price forecast and the 2030-scaled forecast as described in Memo 1. 3. It is the pumped storage plants that generate the main increase of revenue in the upgrading scenarios. Extreme upgrading without pumped storage plants has not been found economically feasible for any price forecast. 4. The pumped storage plants result in a reduction of the total energy production for the hydropower system, but a higher income. In the current tax regime, the pumped storage plants result in reduced taxation to the local municipalities because some of the Norwegian hydropower taxes are related to produced energy. The taxation to the central government increases depending on the economic profit from the pumped storage plants. 5. There is potential to find new hydropower projects in existing hydropower systems. 6. It is possible to construct a flood power plant to mitigate the flood challenges in the Mandal river. The flood power plant is not found to be economically feasible only from hydropower production, and the remaining costs have to be financed by other stakeholders such as insurance companies or local municipalities. The flood power plant can reduce a 200-year flood to a 20-year flood. 7. The extreme upgrading scenarios have a positive impact on flood mitigation owing to new reservoirs and pumped storage plants. This positive impact has not been quantified in this project. 8. Recommended future work includes developing a best-practice guideline for environmentally friendly upgrading of existing hydropower systems based on the methodology developed and applied in this project. In addition, 18 new research projects have been proposed and are described in Memo 5. © NTNU 2020. The publication may freely be cited with source acknowledgement.

Killingtveit, Å. (red.). 2017. Utfordringer og muligheter for norsk vannkraft ved integrasjon med vind- og solkraft i Europa. En oppsummering fra HydroPEAK-prosjektet. – NINA Temahefte 71. 91 s.

This work has studied the hydraulic and thermodynamics of closed surge tanks for hydropower plants. Surge tanks are applied in hydropower plants to control and reduce the acceleration forces of the water during change of turbine flow. Surge tanks reduce the design pressure of the tunnel and pipes that convey water from the reservoir to the turbines, and enable automatic frequency and power control in the hydropower plant. Closed surge tanks are constructed as underground caverns filled with pressurized air located close to the turbines. The work comprises four different studies. In the first study (1), the benefits and challenges of two different surge tank types, the open and the closed surge tanks, were compared. The state-of-the-art design for these two solutions were described. In the second study (2), a hydraulic scale model of an existing hydropower plant with a closed surge tank was constructed in the scale 1:65 in the Hydraulic laboratory at NTNU. A new method for scaling of the absolute pressure in the closed surge tanks was developed and tested. The results of the hydraulic scale model was compared with field measurements from the prototype, and the accuracy was quantified. In the third study (3) the thermodynamics of the air in the closed surge tank was investigated. The impact of heat transfer from the air to the surrounding media was assessed, and different theoretical models for simulation of the thermodynamics where compared with field measurements from an existing hydropower plant. A new improved theoretical model for the thermodynamics was developed and programmed and tested in the freeware hydropower simulation program LVTrans. In the fourth study (4), the effect of installing a surge tank throttle on governor stability, power control and hydraulic transients was investigated. A new method for evaluation of the throttle effect has been developed, and a quantification of the effects for an example hydropower plant with a closed surge tank was conducted.

The Duge pumped storage power plant in Norway currently experiences a problem with rotor-lifting. One of the two vertical 100 MW units lifts up off its bearings when operated above a certain load. The vertical forces on the units have been known to be marginally balanced, but the problem increased when the turbines, commissioned in 1979, were refurbished in 2017. One of the reasons is the significant submergence of the units combined with a 12 km long tailrace tunnel, which produces significant friction losses and therefore high downstream pressure. A one-dimensional (1D) numerical model of the power plant's waterway is established with the software LVTrans and calibrated with field measurements to accurately simulate the hydraulic transients. Three different measures to solve the problem are considered. An economic analysis is conducted to assess the economic feasibility of the three solutions.

For å nå nullutslippsmålet innen 2050 blir det bygd ut fornybar kraftproduksjon for å erstatte fossile energikilder som olje, gass og kullkraft. Den største økningen i fornybar kraftutbygging skjer innenfor solkraft- og vindkraftsektoren. Både sol og vind er uregulerbare kraftkilder som styres av været, og kan ikke reguleres etter etterspørsel og forbruk. Energiproduksjonen som overgår forbruket, må balanseres av lagringstjenester for å unngå overbelastning på strømnettet. Pumpekraftverk kan spille en viktig rolle i fornybaromstillingen ved å levere reguleringstjenester til strømnettet, og å drive storskala lagring av energi. Formålet med masteroppgaven har vært å undersøke muligheten for ombygging av tradisjonelle kraftverk til pumpekraftverk ved å gjenbruke så mye som mulig av eksisterende vannei og kraftstasjon. Utfordringene knyttet til ombygging er kartlagt gjennom litteraturstudie av gjennomføre ombyggingsprosjekt, gjennomgang av vannveien til kraftverkene i Røldal-Suldal kraft, RSK, og numerisk modellering av tunnelsystemet Suldal I dersom turbinen byttes ut med en reversibel pumpeturbin. Arbeidet med oppgaven har kartlagt flere utfordringer knyttet til ombygging til pumpekraftverk, hvor de mest inngripende hydrauliske utfordringene er manglende dykking av løpehjulet, at kraftverksutløpet er plassert for høyt i nedre magasin, og luftinnsug i svingekammer og bekkeinntak. I tillegg vil høy brukstid og underdimensjonert tunnelsystem påvirke lønnsomheten i ombyggingen. I oppgaven presenteres ulike alternativ for ombygging, hvor kun å skifte av turbinløpehjul til et pumpeløpehjul er det rimeligste alternativet for ombygging til pumpekraftverk. Løpehjulsskifte og installasjon av oppstartsutrustning er kostnadsestimert til 20 $\%$ av kostnaden av en full oppgradering, som omfatter ny vannvei og kraftstasjon i parallell. Oppgaven har vurdert løsninger som boosterpumpe i avløpstunnel som en løsning på manglende dykking av løpehjulet. Denne løsningen er kostnadsestimert til 31 $\%$ av kostnaden av ny vannvei og kraftstasjon. Av kraftverkene i RSK er det kun Novle kraftverk som kan bygges om med løpehjulsskifte og oppstartsutrustning i form av frekvensomformer eller ponnimotor. I Røldal, Suldal I og Svandalsflona kraftverk er turbinsenteret plassert for høyt i forhold til nedre magasin, og pumpeturbinen vil derfor være utsatt for kavitasjon. To achieve the net zero emission by 2050, renewable power production will be developed to replace fossil energy sources such as oil, gas and coal. The largest increase in renewable power development is in solar power and wind power business. Both solar and wind are unregulated power sources that are controlled by the weather, and can not be regulated according to demand and consumption. Energy production that exceeds consumption must be balanced by storage services to avoid congestion on the power grid. Pumped power plants can play an important role in the renewable energy transition by providing regulation services to the electricity grid, and operating large-scale energy storage. The purpose of the master's thesis has been to investigate the possibility of converting traditional hydropower plants to pumped storage power plants by reusing as much as possible of existing waterways and power stations. The challenges associated with conversion have been mapped through a literature study of carrying out conversion projects, a review of the waterway to the power plants in Røldal-Suldal kraft, RSK, and numerical modeling of the tunnel system Suldal I if the turbine is replaced with a reversible pump turbine. The work on the master's thesis has identified several challenges related to the reconstruction of power plants to pumped storage power plants. The most intrusive hydraulic challenges are the lack of submergence of the runner, that the power plant outlet is located too high in the lower reservoir and air entrainment in the surge tank and brook intakes. In addition, high working time and undersized tunnel system will affect the profitability of the conversion. The thesis presents various alternatives for reconstruction, where changing the impeller to a pump impeller is the most affordable alternative for reconstruction. Replacing the impeller and installing start-up equipment is estimated to cost 20 $\% $ of the cost of a full upgrade, which includes a new waterway and power station in parallel. The thesis has presented solutions such as a booster pump in a drainage tunnel as a solution to the lack of diving of the impeller. This solution is cost estimated at $ 31 \% $ of the cost of a new waterway and power station. Of the power plants in RSK, only Novle power plants can be rebuilt with impeller replacement and start-up equipment in the form of a frequency converter or pony motor. In Røldal, Suldal I and Svandalsflona power plants, the turbine center is located too high in relation to the lower reservoir, and the pump turbine will therefore be exposed to cavitation.

Tjørhom og Tonstad kraftverk er to kraftverk eigd av Sira-Kvina. %Tjørhom kraftverk er leverer vatn til Tonstad. Låg installert effekt i Tjørhom, gjer at kraftverket produserar samanhengande vinterstid for å levere nok vatn til Tonstad. Låg magasinkapasitet, kombinert med stort lokaltilsig, gjer at Tonstad tvangskøyrast mykje på lokaltilsiget. Ved å utvide Tjørhom kraftverk med eit reversibelt aggregat, kan Tjørhom produsere meir prisoptimalt, og delar av tvangskøyringa i Tonstad kan fjernast. Målet for oppgåva er å utvikle ein metodikk for å modellere drifta av Tonstad og Tjørhom kraftverk. Modelleringa skal kunne sjå på endringa ei utviding av Tjørhom kraftverk til eit pumpekraftverk på 240 MW, gjer med drifta av begge kraftverka. Det vart utvikla ein produksjonsimuleringsmodell for produksjonen i Tjørhom og Tonstad kraftverk. % vart simulert i ein eigenutvikla modell. Både eksisterande Tjørhom kraftverk og tre alternative utbyggingar av Tjørhom pumpekraftverk, vart modellert på historiske data, for tiårsperioden 2010-2019. Den teoretiske maksimale inntektsauka i dei to kraftverka, er rundt 190 millionar kroner årleg, basert på historiske prisar og produksjon. Utvidinga av Tjørhom kraftverk ga auka inntekt og redusert produksjon. Modellen kunne berre optimalisere innanfor ei veke, som avgrensa inntekta i modellen. Manglande langsiktig planlegging gjorde, saman med inndata som ikkje var nøyaktige nok, at den modellerte produksjonen og inntekta i Tonstad vart lågare enn historisk. Den største potensielle inntektsauka frå bygging av Tjørhom pumpekraftverk, er optimalisert drift i Tonstad. Etter ei utbygging vil enkelte av avgrensingane til kraftverka betrast, særleg knytt til tvangskøyring av Tonstad. Andre avgrensingar, særleg kor mykje vatn som kjem til Tonstad frå Sira-strengen, vil framleis kunne hindre optimal drift i både Tjørhom og Tonstad. Forenklingar av avgrensingane i systemet kan overvurdere inntektspotensialet til ei utbygging. Manglande vurdering av dei eksisterande avgrensingane, kan òg overvurdere fleksibiliteten i dagens system. Auka i fleksibilitet som Tjørhom pumpekraftverk gir, vert då undervurdert, som kan gi for låg inntektsauke. På bakgrunn av modelleringa, tilråast det at avløpstunnelen til det nye aggregatet i Tjørhom, drivast heilt til Ousdalsvatn. Det gir størst fleksibilitet, både i produksjon og pumping. Tjørhom and Tonstad are two hydropower plants owned by Sira-Kvina. Low installed capacity in Tjørhom forces the power plant to produce almost constantly during winter, in order to supply Tonstad with water. In Tonstad, low reservoir capacity, combined with large local discharge, forces the power plant to produce on low prices to prevent loss of water. By expanding Tjørhom with a reversible turbine, Tjørhom can produce more energy on the highest prices. It will also allow for pumping of the local discharge in Tonstad, thus reducing the forced production on low prizes. The assignment is to develop a methodology for modelling production in Tjørhom and Tonstad power plants. The modelling should be able to assess how an expansion of Tjørhom hydropower plant to a 240 MW pumped storage plant will change the production in both power plants. A production simulation model was developed for Tjørhom and Tonstad power plants. Both the existing 120 MW Tjørhom hydropower plant, and three alternatives for the 240 MW expanded Tjørhom pumped storage plant, was modelled. The model was based on historical data for the period 2010-2019. The maximal theoretical increase in income, was approximately 190 MNOK per year, based historical prices and production. The expansion of Tjørhom increased the income in Tjørhom, but reduced the production. The model could only optimise week by week, which limited the modelled income. Together with inadequate input data, the lack of long term planning in the model lead to lower production and income in Tonstad than historically. The largest potential increase in income from expanding Tjørhom, is more optimized production i Tonstad. Some of the limitations in the existing system, especially in regards of forced production in Tonstad, will be reduced by an expansion. Other limitations, especially how much water Tonstad gets from each of its two reservoirs, can still limit the production in both power plants. Simplifications of the systems limitations can overestimate the potential income. On one hand, the increased flexibility from expanding Tjørhom can be exaggerated, giving to high potential income from an expansion. On the other, underestimating the limitations in the existing system will exaggerate the flexibility and modelled income of the existing system. Based on the modelling, it is recommended that the tailrace tunnel of Tjørhom pumped storage plant should be built directly to Ousdalsvatn. This gives the largest flexibility, both during production and pumping.

For å forlenge livsløpet til hydrauliske turbiner er det nødvendig å minimere skader forårsaket av sedimentslitasje. En mulig løsning er å forbedre sandfanget for å redusere mengden sedimenter som når turbinene. I dette arbeidet brukes numeriske simuleringer til å undersøke hvordan installering av v-formede raker ved sandfangets innløp og ribber ved utløpet påvirker sandfangseffektiviteten. Det har blitt laget tre-dimensjonelle modeller av sandfanget i Tonstad kraftverk. Arbeidet viser at inkludering av ribber reduserer den totale sedimentmassen som forlater sandfanget med 24.5 \% fra basisverdien, noe som leder til økt sandfangseffektivitet. Som en konsekvens øker sandfangets falltap med 1.8 \%. Inkludering av de v-formede rakene i tillegg øker heller den totale sedimentmassen som forlater sandfanget med 48.5 \% fra basisverdien. Dermed synker sandfangseffektiviteten. Dette øker også falltapet med 12.7 \%. Resultatene viser også at turbulente strømninger som oppstår i diffuseren gjør at sedimenter med diameter mindre enn 1 mm forhindres i å slå seg til ro i sandfanget. Den hydrauliske representasjonen av den numeriske modellen valideres ved å sammenligne med PIV-målinger av strømningsfeltet i skalaekperimenter og med ADCP-målinger fra det faktiske sandfanget. In order to increase the life span of hydraulic turbines in hydropower plants, it is necessary to minimize damages caused by sediment erosion. One solution is to reduce the amount of sediments by improving the sand trap. In the present work, the effects on sand trap efficiency by installing v-shaped rake structures for flow distribution and rib structures for sediment trapping is investigated numerically using the SAS-SST turbulence model. Three-dimensional models of the sand trap in Tonstad hydropower plant are created. The v-shaped rake structures are located in the diffuser near the inlet of the sand trap, while the ribs cover a section of the bed in the downstream end. The present study showed that when including ribs in the model, the total weight of sediments escaping the sand trap is reduced by 24.5 \% from the base value. This leads to an improved sand trap efficiency. Consequently, the head losses in the sand trap are increased by 1.8 \%. By including the v-shaped rakes in addition, the total weight of sediments escaping the sand trap is instead increased by 48.5 \% from the base value, thus worsening the sand trap efficiency. This increases head losses by 12.7 \%. The results also show that turbulent flow commencing at the sand trap diffuser prevents the downstream settling of sediments with a diameter of < 1 mm. The hydraulic representation of the numerical model is validated by comparing to PIV measurements of the flow field from scale experiments and ADCP measurements from the prototype.

Energy storage is needed to enable the transition from fossil to renewable electrical energy sources. As wind and solar power are unregulated and volatile, energy storage is necessary. Pumped hydro can deliver both short- and long-term electrical energy storage. The motivation of this work is to enable cost-efficient and more environmentally friendly construction of pumped storage plants, by finding solutions to the technical challenges. This thesis presents research on hydropower tunnels for pumped storage plants with multiple surge tanks that resulted in three journal papers. This thesis is organized to answer the following four research objectives: (1) Review of existing Norwegian pumped hydro: design challenges, technological solutions, and operational experience. (2) Verify hydraulic scale modelling for investigations of reconstruction of HPPs to PSPs. Demonstrate on a case-study. (3) Identify main challenges associated with the upgrade in terms of tunnel system design and provide solution alternatives. (4) Assess the effect of brook intakes on mass oscillations stability, and its implications for upgrading of hydropower plants with brook intakes. The three research methods applied to answer the research objectives are: field measurements, 1D numerical simulations, and hydraulic scale modelling. Much of the work is conducted based on a case-study, namely the 50 MW Roskrepp hydropower plant located in southern Norway. A feasibility study for reconstructing this HPP to a PSP is currently undertaken by the power plant owner Sira-Kvina kraftselskap, making it an ideal case-study for the work. The power plant owner granted access to conduct field measurement during operation, a 3D scanning of the tunnel system, and available documentation and reports. The work results in three journal papers, presented in this thesis. In addition, four secondary papers are published. The main contributions from this work are: 1. A technical review of currently existing Norwegian PSPs. This review provides a foundation for future development of PSPs. 2. A new method for determining the distribution of head loss factors in hydropower tunnel systems with multiple surge shafts. 3. Identification of the main limitations for upgrading HPP to PSP and provided solution alternatives. 4. An investigation of the effect of brook intakes on the stability of mass oscillations in existing hydropower plants. It is concluded that it is possible to upgrade existing hydropower plants to pumped storage plants by using the existing tunnel system infrastructure with minor modifications. Suggestions for future work are included at the end of this thesis.

De gjeldende retningslinjene for design av lukkede sandfang er basert på fysiske modellforsøk gjennomført i vassdragslaboratoriet på 1960-tallet, der resultatene av modellforsøkene ikke kan oppdrives, bare konklusjonene. Denne masteroppgaven ble gjennomført for å verifisere de tidligere funnene. I denne masteroppgaven ble 3D Computational Fluid Dynamics (CFD) brukt til å undersøke to forskjellige variabler i ribbedesign for lukkede sandfang, og hvordan disse påvirker fangsteffektiviteten. Geometrien av CFD-modellen er basert på sandfang no. 3 i Tonstad vannkraftverk, 960 MW, og resultater fra et fysisk modellforsøk har blitt brukt til å kalibrere og sammenligne den numeriske simuleringsmodellen. Flere endringer av prototypgeometrien er gjennomført for å unngå case-spesifikke effekter på løsningen, og for å gjøre resultatene mer generelt gyldige. Teori om sediment-transport og design av lukkede sandfang med ribber er presentert. Teori om CFD er presentert sammen i metodekapittelet der den numeriske tilnærmingen er beskrevet. For kalibreringen og valideringen, ble stasjonære Reynolds-Averaged Navier-Stokes (RANS) -simuleringer gjennomført med to forskjellige vannføringer for å teste validiteten til CFD-modellen. Hastigheten rundt ribbene i CFD-modellen ble sammenlignet med hastigheten målt ved hjelp av Particle Image Velocimetry (PIV) -eksperimenter fra den fysiske skalamodellen, henholdsvis i prototyp- og modellskala. Sammenligningen viste god overensstemmelse mellom CFD-modellen og den fysiske modellen. Ikke-stasjonære RANS-simuleringer med sedimenter ble gjennomført for å undersøke hvordan ribbedesignet påvirker fangsteffektiviteten for to ulike variabler. Den første variabelen er ribbebredden, der bredden av ribben er lik åpningen mellom ribbene. Den andre variabelen er åpningen mellom ribbene, der ribbebredden holdes lik på 1.0 m. Totalt 20 simuleringer med injeksjon av partikler med 0.1 mm diameter ble kjørt. Tre forskjellige hypoteser ble testet: 1) Det er mulig å forbedre ribbedesignet for lukkede sandfang sammenlignet med nåværende designanbefalinger, 2) Den optimale avstanden mellom ribbene er 0.5 m, 3) Den optimale bredden på hver ribbe er 1.0 m. Simuleringene viste at det ikke var mulig å forbedre designet med variablene som ble testet i denne oppgaven. For hypotese no. 2 viste resultatene fra simuleringene at den optimale avstanden mellom ribbene var 1 m. Simuleringene avslørte at fangsteffektiviteten var høyere for simuleringene der ribbebredden var fastsatt til 1 m, sammenlignet med simuleringene der ribbebredden varierte. Det er ikke mulig å konkludere med en optimal ribbebredde ettersom at det ble gjennomført få simuleringer i området nært 1.0 m. Resultatene viser at det optimale designet er 1 m brede ribber med 1 m åpning mellom ribbene, som bekrefter tidligere litteratur på lukkede sandfang. The current state-of-the-art guidelines to closed sandtraps are based on physical model studies performed in the hydraulic laboratory in the 1960s, where the model test results cannot be found, only the conclusions. This master thesis was conducted to verify the previous findings. In this master thesis, 3D Computational Fluid Dynamics (CFD) was used to investigate two different variables in rib design for closed sandtraps, and how these affect the trap efficiency. The geometry of the CFD model is based on sandtrap no. 3 in the 960 MW Tonstad hydropower plant, and results from a physical model test have been used to calibrate and compare the numerical simulation model. Several changes to the prototype geometry are done to avoid case-specific effects on the solution, to make the results more generally valid. Theory on sediment transport and design of closed sandtraps with ribs is presented. Theory for CFD is presented together in the methodology chapter where the numerical approach is described. For the calibration and validation, steady-state Reynolds-Averaged Navier-Stokes (RANS) simulations were conducted with two different discharges to test the validity of the CFD model. The velocity around the ribs in the CFD model was compared to the velocity measured from Particle Image Velocimetry (PIV) experiments from the physical scale model, in prototype and model scale, respectively. The comparison showed good correlation between the CFD model and the physical model. Transient RANS simulations with sediments were carried out to investigate how the rib design affect the trap efficiency for two different variables. The first variable is the rib width, where the width is equal to the opening between the ribs. The second variable is the opening between ribs, where the rib width is fixed at 1.0 m. A total of 20 simulations were run with 0.1 mm particles injected. Three different hypotheses were tested: 1) It is possible to further improve the rib design for closed sandtraps compared to current design recommendations, 2) The optimum distance between the ribs is 0.5 m, and 3) The optimum width of each rib is 1.0 m. The simulations showed that with the variables tested in this thesis, the rib design could not be improved. For hypothesis 2, the results from the simulations showed that the optimal distance between the ribs was 1 m. The simulations revealed that the trap efficiency was higher for simulations where the rib width was fixed at 1 m, compared to the simulations where the width varied. However, owing to the limited number of simulations run in proximity to 1.0 m, the optimal width could not be concluded. The results showed that the optimal design is 1 m wide ribs with 1 m opening between the ribs, confirming previous literature on closed sandtraps.

The concept of unlined pressure tunnel design is well-tested and has a history of more than 100 years. In Norway, more than 95% of hydropower pressure tunnels are unlined and most of it was built between 1950 and 1990. It is also popular outside of Norway because of its cost-effectiveness and shorter construction time. The main design principle is to prevent hydraulic jacking, which is obtained by suitably aligning the tunnel such that the in-situ stresses are sufficient to withstand the internal water pressure, without the use of extensive rock support and lining. Minor rockfalls are accepted during operation as long as they do not develop significantly and increase the frictional loss or cause blockage in the tunnel. It is seen that the operational regime of power plants in Norway has changed after the power market de-regulation in 1991. In the demand driven market, the power prices can vary on an hourly basis and the power plants can experience multiple load changes per day to benefit from the variable power prices, causing frequent pressure transients in the waterway. Further, an increasing share of unregulated energy from solar and wind power in the energy system as seen in the recent years will demand more operational changes from regulated hydropower systems which are used to maintain the balance between supply and demand. Such an operation will lead to frequent pressure pulsations and cyclic loading on the rock mass around unlined tunnels, and may contribute to increased instances of block falls as a result of rock mass fatigue. This research is focused on understanding the effects of frequent pressure pulsations in the long-term stability of unlined water tunnels. The work is based on cases from Norway and includes observations from inspection of four dewatered tunnels, instrumentation, and monitoring of one tunnel, operational data of 10 hydropower plants and numerical modelling using the distinct element code 3DEC. Results indicate that pressure transients can have significant influence on the pore pressure variation and joint displacement in the rock mass around unlined pressure tunnels as a result of the time-lag between the pressure transient in the tunnel and the rock mass pore pressure. It is the source of hydraulic stresses in the rock mass and is dependent on their hydro-mechanical properties. Results confirm the previous knowledge that mass oscillations cause larger hydraulic stresses in the rock mass as compared to water hammer. However, exceptions are known and the effect of water hammer may not be completely ignored. It is seen that 200-400 start/stops and more than 1000 load changes of varying magnitudes occur every year per generating unit in Norwegian power plants, causing frequent pressure transients. It is envisaged that this trend will further increase in the future due to addition of larger share of unregulated power from wind and solar energy. This implies that rock mass fatigue in unlined pressure tunnels may occur at an accelerated rate. The results indicate that an increased conservatism may be needed in rock support decisions in critical areas where the rock mass permeability permits significant pore pressure changes in the rock mass during pressure transient, especially for tunnels excavated in schistose rock mass, and power plants with multiple load changes within a day. Power plant operation is seen to have a significant influence on the amount of hydraulic stress acting on the rock mass during pressure transients. The shutdown/opening duration is usually dependent on the individual operator due to lack of standard guidelines for speed of load changes. Especially for large load changes, the power is usually changed in smaller steps, where the size and number of these steps are decided by the individual power plant operator. Results show that the shutdown/opening duration during load changes directly affects the time-lag between pressure in the tunnel water and in the rock mass. It is seen that shorter shutdown/opening duration i.e., faster speed, can cause significantly high hydraulic stresses on the rock mass. Thus, slowing down the load change operation can provide significant benefit in slowing down the fatigue process. Hence, it is recommended that more emphasis should be given towards keeping the speed of load changes consistently slow. A new term called “Hydraulic impact” is proposed to quantify the hydraulic stress on the rock mass caused by pressure transients in unlined hydropower tunnels. It can also be used to define a suitable shutdown speed of the power plants in order to help slow down the fatigue process. It is recommended to instrument and monitor more tunnels in order to validate and expand the results.

Denne masteroppgaven handler om numerisk modellering (“Computational Fluid Dynamics”, CFD) av Sandfang 3 ved Tonstad kraftverk med dagens utforming og rekonstruert med horisontale ribber. Det ble observert sedimentproblemer i sandfanget etter kapasiteten økte da en femte turbin ble installert. Simuleringene er basert på geometrien for dette sandfanget, og de numeriske simuleringene er gjennomført for å studere effekten av å oppgradere sandfanget ved å installere horisontale ribber. Det er gitt en innføring i teori om sandfang, sedimenttransport og CFD. Teorikapitlene etterfølges av fremgangsmåten for numerisk modellering med Ansys Fluent, inkludert tillaging av geometrien, generering av grid og oppsett av modellene. To ulike siutasjoner er simulert for å studere effekten av å installere horisontale ribber i sandfanget. Stasjonære “Reynolds-Averaged Navier-Stokes” (RANS) simuleringer ble kjørt for å sammenlikne vannstrømningen med og uten ribber i sandfanget. Transiente RANS simuleringer ble gjennomført for å analysere effekten av ribbene når sedimenter slippes inn i sandfanget. Resultatene fra de stasjonære simuleringene viste at hastigheten langs bunnen synker når sandfanget rekonstrueres med horisontale ribber. Dette skal i teorien bedre forholdene for sedimentering og redusere medrivningskraften på avsatte sedimenter. De transiente simuleringene med innslipp av sedimenter bekreftet at redusert hastighet langs bunnen gir bedre forhold for sedimentering. Sandfangets fangsteffektivitet øker når horisontale ribber konstrueres i sandfanget. Det er vist at Ansys Fluent er egnet som CFD-program for modellering av partikler i et trykksatt sandfang. In this master thesis, Computational Fluid Dynamics (CFD) was used to simulate Sand trap 3 at Tonstad hydropower plant in the current configuration and reconstructed with horizontal shear plates. The operators of the power plant observed sediment problems when the capacity was increased along with the installation of a fifth turbine. The simulations are based on the geometry of this sand trap, and the numerical simulations are carried out to investigate the effect of upgrading the sand trap by installing horizontal shear plates. It is given an introduction to sand trap, sediment transport and CFD theory. The theory chapters are followed by the procedure for numerical modelling with Ansys Fluent, including creating the geometry, mesh generation and setting up the models. Two different cases are simulated to study the effect of installing horizontal shear plates in the sand trap. Steady state Reynolds-Averaged Navier-Stokes (RANS) simulations were run to compare the flow behaviour with and without shear plates installed in the sand trap. Transient RANS simulations were carried out to investigate the effect of the shear plates when particles are injected into the sand trap. The results of the steady state simulations showed that the velocity decreases close to the bed when the sand trap is reconstructed with horizontal shear plates, allowing particles to settle and reducing the entrainment force on the deposits. The transient simulations with injection of particles confirmed that the reduced velocity close to the bed gives better conditions for particle settling. The trap efficiency increases when horizontal shear plates are installed in the sand trap. It is shown that Ansys Fluent is a suitable CFD software for the modelling of particles in a pressurized sand trap.

From the beginning of 20th century, huge tensions rising about global warming and environmental pollution from utilizing fossil fuels enormously. Implementing pump hydro storage profoundly can reduce the extensive use of fossil fuels and ameliorate the green energy production. Regarding this, Kuli PHSP (Kuli pumped hydro storage project) has been studied in this thesis. This study focuses on planning and designing of possible tunnel layout alternatives for Kuli PHSP. Tunnel layouts are selected with minimum geological conflicts and maximum utilization of natural resources. 1D numerical simulation of waterway systems have carried out to determine the system stability during pump-turbine operation. Threshold price is imposed based on roundtrip efficiency losses to regulate the plant operation. From this regulation strategy net production energy is calculated and a financial analysis is conducted by employing total project cost for each alternative. Finally, an economic and sensitivity analysis is performed to estimate the projects economic feasibility and robustness. Regarding this outcomes best alternative is selected for further investigation.

Several large hydropower plants in Norway have been upgraded with a higher installed capacity, but unfortunately, this upgrade has also brought with it operational problems because of sediments entering penstock and causing damage to the turbine. This was due to a sand trap which was not functioning as projected after the upgrade. To improve sand trap efficiency, several proposals were considered. One of them was to calm and homogenize the flow by using a flow calming structure. The main objective of this thesis was to propose the optimum design of a sand trap flow calming structure for Tonstad hydropower plant, which was one of the hydropower plants facing operational problems after an upgrade. To find the optimum designed of such a structure, three different flow calming structures were designed, constructed in the model scale, tested in the hydraulic scale model and compared to each other. The results showed that the implementation of a flow calming structure improved sand trap efficiency. All three flow calming structures provided very similar, satisfactory, results. Sand trap efficiency increased from 95.67 % to more than 99 %, sedimentation began already in the diffuser and occurred mainly in the upstream half of the sand trap. In terms of headloss were the results various. While flow calming structure No. 3, layout 3, increased sand trap efficiency to 99.96 % with a small headloss of 3.83 mm, flow calming structure No. 2 reached sand trap efficiency of 99.19 % with the headloss 5.2 times higher equal to 19,89 mm. This made flow calming structure No. 3 the one to be recommended for further testing and possibly installation in sand trap 3 in Tonstad HPP.

Snømåling med drone ved bruk av fotogrammetri for å lage 3D modellar av eit område er ein ny teknologi som kan vere veldig nyttig for vasskraftselskap for å berekne snømagasin. Med det kan vasskraftselskapa finne forventa avrenning frå snømagasinet til vassmagasina. Eit forbetra anslag for forventa avrenning frå snømagasinet vil gje eit betre grunnlag for å planlegge bruk av vassmagasinet og auke inntekt frå produksjon. Snøvolum var førespurt til å bli funnen på tre forskjellege plassar i Sira-Kvina kraftselskap sitt nedbørsfelt. Dei tre valde plassane var Flatstøl, Tjørhom og Nesjen som er lokalisert i ulike områder av nedbørsfeltet. Ved desse stadane var det utført fotogrammetri ved hjelp av drone med og utan snø på bakken. 3D modellar frå fotogrammetrien var laga i programmet Pix4Dmapper og volumkalkuleringar vart utført i programmet CloudCompare. Før flygning med fotogrammetri på vinterstid vart fleire prøveflygningar utført i forskjellege vêrforhold for å validere kva slags vêrforhold som fungerte og ikkje. Resultatet frå desse prøveflygningane slo fast at det var kun vêrforhold med sol frå klar himmel som var tilstrekkeleg på grunn av for mykje lysrefleksjon i overskya vêr. Frå vinterflygningane ved Tjørhom var ikkje GPS-en i drona korrekt kalibrert og 3D modellen som vart laga var dermed ubrukeleg. Frå 3D modellane ved Nesjen var modellen frå bar grunn og snødekt grunn skeive i forhold til kvarandre, noko som gjorde dei ubrukelege for volumkalkulering. 3D modellane frå Flatstøl var utan problem utanom ein liten forskyving i x og y retning mellom modellane. Frå Flatstøl vart det funne ei gjennomsnittleg snødjupn på 0,48 meter frå fotogrammetri og ei snødjupn på 0,35 meter frå manuelle målingar. Det vart funne ein auke i snøvolum på 38% i 3D modellane i forhold til manuelle målingar. Resultata viser at med eit større areal dekt ved hjelp av ei drone så finn ein meir forskjellar i snødjupne enn ved manuelle målingar. Det auka kalkulerte snøvolumet gjev moglegheiter for betre planlegging av bruk av vassmagasin sidan den kalkulerte mengda avrenning er høgare. Ved å anta at kalkulert auka avrenning er gyldig for halve nedbørsfeltet til Sira-Kvina kraftselskap, så er det funne ei auka avrenning på 59,67 millionar m3. Ved å produsere meir kraft i periodar med høg etterspørsel av straum, er ei ekstra inntekt på 1,496 millionar kroner forventa. Snow measurement with drones by use of photogrammetry to create 3D models of an area is a new technology that can be very useful for hydropower companies to estimate snow storage. With that, the hydropower companies can find the expected runoff from the snow storage into the reservoirs. An improved estimation of expected runoff from snow storage will give better foundation to plan the operation of the reservoir and increase the income from production. Snow volume was requested to be measured at three different sites in Sira-Kvina Power Company’s catchment area. The three chosen sites were Flatstøl, Tjørhom and Nesjen which is located in different areas of the catchment area. At these sites, photogrammetry with drone were performed with and without snow on the ground. 3D models from photogrammetry was made in the Pix4Dmapper software and volume calculations were performed in the software CloudCompare. Several test flight were performed in different weather conditions to validate which conditions were working or not, before flights in wintertime to take photogrammetry of the snow. The results from the test flights revealed that only conditions with sun from clear sky was appropriate due to much light reflection from the snow in overcast weather. From the winter flight at Tjørhom, the GPS in the drone was not correctly calibrated and the obtained 3D model was not usable. At Nesjen, 3D models from bare ground and snow covered ground was twisted according to each other and not usable for volume calculation. The 3D model at Flatstøl was without issues except from a small offset between the two models in x and y direction. From Flatstøl, a mean snow depth obtained from photogrammetry was 0.48 meters and from manual measurement, a snow depth of 0.35 meter was found. An increase in snow volume of 38% was found from the 3D models according to the manual measurements. According to this, the results showed that with a larger area covered by the drone, more differences in snow accumulation is obtained than with manual measurements. The increased calculated snow volume gives possibility for better planning of operation of reservoirs since the calculated runoff would be higher. By assuming that the calculated increased expected runoff is valid for half of Sira-Kvina Power Company’s catchment area, an increase in calculated runoff of 59.67 million m3 of water is found. By producing more power in high demand periods, an extra income of 1.496 million NOK is expected.

Duge kraftverk har utfordringer knyttet til aksiell last ved ett av to aggregat, som gjør at rotor har løftet seg ved et par anledninger. For dette prosjektet blir de aksielle kreftene undersøkt for å bygge opp en bedre forståelse av kreftene som virker. Utfordringene med aksiell last begrenser mulighetene til å kjøre på bestemte driftspunkter og reduserer fleksibiliteten til kraftverket. Ved å løse de nevnte utfordringene, kan Duge kraftverk operere med større fleksibilitet og høyere driftspunkter enn i dag, og dermed oppnå stor økonomisk gevinst. Det ble gjennomført forsøk på en Francismodell i Vannkraftlaboratoriet ved NTNU for å undersøke de aksielle kreftene i første del av oppgaven. Feltmålinger fra Duge ble analysert i andre del. Det ble utviklet en analytisk modell for å estimere og simulere aksiallasten. Aksiallastmodellen baseres på et sett av ligninger som bruker ulike parametre og geometrier for å beregne aksielle krefter som virker over og under løpehjulet. For å evaluere aksiallastmodellen, ble det gjennomført forsøk i Vannkraftlaboratoriet for å sammenligne målt aksiallast med resultatene fra simuleringene. Francisløpehjul og løpehjul i reversible Francis pumpeturbiner, er begge designet slik at de hydrauliske kreftene over og under løpehjulet aktivt brukes til å balansere de aksielle kreftene fra de store massene som roterer. Ved å balansere de aksielle kreftene, letter det lasten på bærelageret. Det er et mål om å ha en aksiallast på rotor som gir en margin mot at rotor løfter seg på 10% i forhold til roterende masse ved største negative hydrauliske lastbidraget ved drift på stasjonære driftspunkter. Resultatet fra målt og beregnet aksiallast i Vannkraftlaboratoriet ble sammenlignet for å undersøke om aksiallastmodellen treffer på dette designmålet. Resultatene viste at det for driftspunktene på bestpunkt og overlast blir estimert en last som er under 10% forskjell fra målt aksiallast for de fleste forsøkene. På dellast var det en forskjell på rett i overkant av 15%. Felles for alle simuleringene var at de overestimerte aksiallasten i forhold til den målte lasten. Resultatene fra laboratorieforsøkene viste at det er en stor variasjon i aksiallast mellom de ulike driftspunktene. I andre del av prosjektet, ble feltmålinger fra Duge analysert. Med målinger fra hendelsene med løft av rotor og andre målinger utført ved Duge, ble det gjort undersøkelser rundt hva som kan påvirke de aksielle kreftene. Aksiallastmodellen ble brukt for å se på hvilke krefter som er årsaken til løft, og om det er noe forvarsel som kan observeres fra simuleringen. Resultatene fra simuleringen viste at det er lite som indikerer at rotor skal løfte seg. Simuleringene med data fra siste løftehendelse, viste en margin på 30 tonn mot løft samtidig som løftet skjer. Flere analyser av målingene ved Duge viser at trykkene ved innløp og avløp til løpehjul kan føre til at marginen mot at rotor løfter seg blir liten. Ved begge løftehendelsene er trykkdifferansen mellom innløpstrykket til spiraltrommen og sugerørskonus 200 mVs. Videre undersøkelser viser at en tenkt situasjon med økt innløpstrykk, økt sugerørstrykk og stor volumstrøm er kombinasjoner som gir en større negativ hydraulisk last som virker oppover og fører til mindre margin mot løft ved stasjonær tilstand. Et grovt estimat gjort ut fra resultater i Vannkraftlaboratoriet og målinger fra Duge viser at 52 tonn mer last på rotor, kan gi en sikker drift med god margin mot løft. En løsning for å oppnå dette kan være å øke trykket i spaltevannshulrommet. Med en trykkøkning i øvre spaltevannshulrom på 18 mVs, kan dette gi effekten som trengs i form av økt trykk over flaten i spaltevannshulrommet nedstrøms siste øvre labyrint. Tester i laboratoriet viser positive resultater i form av økt last ved struping av ventil koblet til spaltevannsarrangementet i øvre spaltevann. Struping av ventil på øvre spaltevann er også mulig å gjøre ved Duge. Hvorvidt dette kan løse situasjonen på Duge må undersøkes videre og testes på prototype i felt. Duge power plant struggles with one of the turbines lifts at certain areas of operation. Due to the lift problems, the powerplant cannot operate at some desired operation points. In order to investigate the axial forces and what affect the forces, experiments were conducted in the Waterpower Laboratory at NTNU in the first part of this thesis, and analyses of field measurements from Duge in the second part. A model was developed in order to calculate and simulate the axial forces acting on the runner. In the Waterpower Laboratory it is possible to measure the axial forces on the turbine. The measurements were used for validation of the simulated axial forces. A Francis runner and a reversible Francis pump turbine runner are both designed to balance the axial forces in order to ease the load on the thrust bearing. A design goal is a hydraulic axial force that leads to an axial resultant force of 10% of the total rotating mass. Results from measured axial force in the laboratory and calculated axial force from the analytical model were compared. The results from simulations of the axial forces showed some variations on how accurate the model is calculating the forces compared to the measured axial forces. The results from the simulations showed less than 10% difference between simulated and measures axial force at best efficiency point and at over load. At part load the difference between simulated and measured axial force was 15%. All of the simulations showed an overestimated axial force compared to the measured force. The results from the measured axial forces also showed that there are significant differences in total axial force at different operation points. The field measurements from the lift incidents and a couple of other operations from Duge power plant were analysed with the aim to investigate what cause the lift. The axial force simulation model was used to see if it is possible to find what triggers the lift of the rotor by use of the measured parameters from the incidents. Analyses of the measured parameters were conducted to investigate the cause of the lift. The results from the simulations show no sign of the lift that happens. From the last lift incident, the results from the simulation show a force acting downwards, corresponding to 30 tons, the moment the rotor lifts. The analyses of the measurements from Duge power plant show that the pressure at inlet and outlet of the runner may be one of the reasons for the lift. At both lift incidents the differential pressure between the inlet of the spiral casing and the draft tube cone is 200 mWc. A situation with high inlet pressure, large volumetric flow and high outlet pressure yield a large negative hydraulic axial force at steady state condition. This leads to a low resultant force acting downwards and risk of lift of the rotor if the resultant force becomes negative. A rough estimate, based on the results from the analyses at theWaterpower laboratory and Duge combined with the theory, shows that with additional 52 tons acting downwards, could give a safe operation when it comes to lift incidents. A solution in order to provide the additional axial force is by increasing the pressure in the leakage water from the upper clearance downstream the labyrinth seal. With an increase of 18 mWc of pressure in the clearance, this may be enough to provide the additional axial force. Tests in the laboratory show positive results from increasing the pressure from the leakage water when it comes to increased total axial force. At Duge the leakage water can be throttled by an existing valve. Further work is needed to validate if this solution can provide the wanted effect and result in a safe operation at all operation points.

I senere år har flere norske vannkraftverk opplevd driftsproblemer på grunn av sedimenter i turbinvannet. Dette kan føre til redusert virkningsgrad på turbin og dermed økonomiske tap. Et forskningsprosjekt på sandfang har blitt startet for å løse dette problemet. I Norge er sandfang normalt plassert oppstrøms trykksjakta med det formål å la sedimenter bunnfelle, og dermed redusere middelpartikkelstørrelsen og den totale sedimentmengden i turbinvannet. Alle sandfang fungerer imidlertid ikke optimalt og denne oppgaven er en del av et større forsknings-prosjekt som går på å optimalisere utformingen til norske sandfang. I denne oppgaven var målet å foreslå hvilke modellsedimenter som skulle brukes til fremtidig fysisk modellering av trykksatte sandfang. For å kunne bestemme dette, ble det gjennomført et litteraturstudie som beskriver tidligere forskning rundt temaene hydraulisk modellering, skalering av sedimenttransport og -erosjon, og trykksatte sandfang. Dette ble etterfulgt av generell teori om hydraulikk, likhet og dimensjonsanalyse. Den teoretiske delen av avhandlingen legger grunnlaget for en bedre forståelse av skaleringen og mekanikken bak sedimenttransport i tunneler og sandfang. En felttur ble organisert til Sandfang 3, som er en del av Tonstad kraftverk (960 MW) i Norge og det sandfanget som er i fokus. Denne turen ble lagt til den årlige tømmingen av sandfanget. En rekke sedimentprøver ble tatt fra sandfanget og analysert på vassdragslaboratoriet ved NTNU. De resulterende kornfordelingskurvene la noe av grunnlaget for å bestemme hvilke modellsedimenter som skulle testes i en eksperimentell renne før ferdigstillelsen av sandfang-modellen. Tre forskjellige modellsedimenter som var tilgjengelige i laboratoriet ble valgt til testing. Dette var (i) sand, (ii) karbonatstein og (iii) PMMA. Modellsedimentene hadde forskjellige egenskaper og oppførte seg derfor forskjellig fra hverandre i modellen. En spørreundersøkelse utført viste at sand ofte er det foretrukne modellsedimentet for hydrauliske modeller på grunn av dets like egenskaper til naturlige sedimenter. Ved testing av sanden i modellen med en hastighet som tilsvarer prototypen, oppsto bølgeslagslinjer på bunnen langs hele lengden av modellen. Karbonatsteinen hadde samme tetthet som sanden, men større partikler og deponerte derfor kun i den oppstrøms delen av sandfanget. PMMA var det eneste lettvekts modellerings-materiale og representerte de minste fraksjonene av prototype sedimenter som ble gjenskapt i modellen. Dette materialet gav det deponeringsmønsteret i modellen mest lik prototypen og er sammen med sand, anbefalt for videre bruk i modellering av trykksatte sandfang. In recent years, several Norwegian hydropower plants have experienced operational problems due to entrainment of sediments in the turbine water. This may lead to reduced turbine efficiency and hence financial losses. To address this problem, new research on sand traps has been initiated. The sand traps are located upstream or inside the headrace tunnel with the purpose of allowing sediments to settle, reducing the mean particle size and overall sediment load in the turbine water. All sand traps do unfortunately not function optimally, and this thesis is part of an ongoing research project aiming to improve Norwegian sand traps. In this thesis, the aim was to propose which modeling materials to use for future physical modeling of pressurized sand traps. In order to properly address this, a literature review outlining the previous contributions on the topics hydraulic scale modeling, scaling of sediment transport and erosion, and pressurized sand traps was conducted. This was followed by general theory regarding hydraulics, similitude and dimensional analysis. The theoretical part of the thesis lays the foundation for a better understanding of the scaling and the mechanics behind sediment transport in tunnels and sand traps. A field trip was organized to Sand trap 3 in the 960 MW Tonstad hydropower plant in southern Norway. This trip was set for the annual draining and cleaning of the sand trap. A series of sediment samples were taken from the sand trap and analyzed back at the Hydraulic Laboratory at NTNU. The resulting grain size distribution curves became the basis for deciding on which modeling sediments to test in an experimental flume prior to the completion of the sand trap model. Three different modeling materials available in the laboratory were selected for testing. Those were (i) sand, (ii) carbonate rock and (iii) PMMA. The modeling materials had different properties and hence behaved differently in the model. A survey showed that sand often is the preferred modeling material for hydraulic scale models due to its similar properties to natural sediments. When testing the sand in the model at a velocity equivalent to the prototype, the sand formed ripple marks throughout the length of the model upon settling. The carbonate rock had the same density as the sand, but larger particles and settled therefore in the upstream part of the sand trap. PMMA was the only lightweight modeling material and represented the smallest fraction of prototype sediments recreated in the model. The PMMA also gave the deposition pattern in the model most similar to the prototype and is together with sand, recommended for further use in the modeling of pressurized sand traps.

This thesis deals with the issue of surge tank stability in hydropower plants, investigating methods used to design the surge tank area required to guarantee stability. Currently, Thoma equation (1910) is widely accepted as the standard approach for surge tank dimensioning. However, this approach has proven excessively conservative, often leading to over-dimensioned surge tanks. Recent developments in computational power enable us to consider alternative methods, with the direct simulation, consisting in the numerical simulation of the entire hydropower plant, as one of the most promising. This study draws a comparison between Thoma equation approach and the direct simulation. A 1D numerical model of Roskrepp power plant (Vest-Agder County, Norway) is developed and calibrated in the LVtrans freeware. Transients originating from different scenarios are analysed. Simulation results show that the direct simulation method proves superior, with a more accurate dimensioning of the surge tank. However, Thoma equation, albeit less accurate, still proves relevant for the early stages of design, when limited design data are available. Keywords Hydropower, Stability, Transients, Surge tanks, Direct simulation, 1D numerical model, LVtrans, Thoma method

Rebuilding a hydropower plant into a pumped-storage plant in an underground tunnel, means that the water will rapidly change directions, instead of streaming evenly in one direction. This will cause extra water pressure in the headrace tunnel and can cause stability problems that can damage the turbines. The hydropower plant, Roskrepp, is being considered to become a pumped-storage plant. During the construction of the hydropower plant, an asphalt layer was put at the floor. With rapidly change of water pressure, the asphalt can tear up and cause damage to the turbines. Investigation of the rock condition along the tunnel alignment and an assessment of the asphalt lining in conjunction with the possibilities of pumped-storage plant for Roskrepp has been done. To evaluate the problem, literature research, field investigation, laboratory testing on rock samples, numerical analysis and physical model test has been carried out. The stability assessment of the rock mass included literature study, field investigation, laboratory testing and numerical analysis. One of the six possible weakness zones crossing the tunnel area, appears to be more crucial regarding the stability of the tunnel. This could be of crushed rock material. If extra pressure occurs, a caving situation can happen in the ceiling. Erosion under asphalt from the crushed rock materials can also cause instabilities. With water streaming under asphalt lining, the crushed rock can erode and tear up the asphalt lining. Eventually the materials can stream down to the turbines and cause destruction. The assessment of the asphalt lining included literature study, numerical analysis, and physical model test. If cracks are developed in the asphalt, or the contact between the asphalt layer and the rock walls are not fully sealed, water can easily stream under the lining and disturb the aggregate under it. Literature study and physical model test results shows that pressure under asphalt lining is delayed comparing with pressure over asphalt lining when mass oscillation is present. If a fine combination between trapped air and water under the lining are present, there will be a possibility of lifting the asphalt when mass oscillation is on its way down. This can cause tearing up the asphalt and destroy the turbines.

Formålet med masteroppgaven har vært å studere luftmedrivning i modellen av det lukkede flomløpet ved dam Follsjøen i Rindal kommune. Modellen er opprinnelig bygget for et modellforsøk av flomavledningskapasiteten, men er nå tilgjengelig for undersøkelse av luftmedrivningen. Modellen er bygget i en skala på 1/40, ut fra Froudes modellov. Modellen er ikke lenger geometrisk lik prototypen, da utvidelser i modellen har vært nødvendig for å oppnå ønsket flomavledningskapasitet. Det har i forbindelse med denne masteroppgaven blitt bygget en lufttett kasse over overløpet, for å oppnå et fullstendig lukket system, med kontroll av all luftmedrivning. Kassens innvirkning på systemet har blitt kontrollert ved hjelp av differensialtrykk og vannstandsmålinger i kasse og basseng. Kassens innvirkning er funnet til å være lav, maksimalt 23,82 mm differanse i vannstand og 0,56 mm i differensialtrykk. Luftmedrivningen har blitt målt med en varmetrådsprobe i tre lufterør påmontert kasse, sjakt og overføringstunnel. Luftmedrivningen er undersøkt og sammenlignet for ulike vannhastigheter, ruheter og ulike punkt for lufttilgang. Vannhastigheten har blitt endret ved å variere vannføringen, og ruheten har blitt endret ved å variere antall innlimte lister i tunnel og sjakt. Målinger viser at luftmedrivningen i modellen øker for økende vannføring frem til 74,1 l/s, deretter avtar luftmedrivningen for en ytterligere økning i vannføring. Maksimal vannføring i forsøket er 138,8 l/s. Luftmedrivningen i modellen er betydelig større for en lavere ruhet ved samtlige vannføringer, og den største økningen i luftmedrivning for lavere ruhet er 4,7 l/s. Ved å variere hvor lufttilgangen gis i systemet ved å stenge eller åpne lufterørene på sjakt og overføringstunnel, kommer det frem at ulike luftepunkt gir ulik luftmedrivning i systemet, både i menge og i område. Maksimal luftmedrivning for systemet og luftmedrivning ved lukking av innløp har blitt undersøkt. Målingene viste at luftmedrivningen i modellen er størst mellom en vannføring på 74,1 og 79,1 l/s, og at innløpet lukkes ved 133,4 l/s. Et tydelig fall i luftmedrivning observeres ikke når innløpet dykkes. De målte verdiene i modellen lar seg ikke overføre direkte til prototype grunnet manglende skaleringslover for luft.

This thesis discusses two different criteria for the stability of surge tanks in hydropower plants. In 1910, the first stability criterion was derived by Dieter Thoma. This criterion sets a minimum cross-sectional area for a surge tank to give stable conditions for the power plant. The criterion was put into use, but was found to give inaccurate results, and later modified. In 1970, Hallbjørn Roald Svee derived a second stability criterion, improving on the inaccuracy by including several factors that Thoma had left out. Svee s criterion was written as a reply to bad results in using the Thoma criterion for the stability of a surge tank in Brazil. However, after its use by Svee himself, the criterion has not been put into widespread use. The derivation of the two criteria are presented side by side, where both criteria have been rewritten to a modern form and non-explained steps from the original sources are derived. To test the criterion, three studies were carried out. A parameter sensitivity analysis, a case study of Kvinen power plant and a one-dimensional numerical simulation on the same power plant, comparing the Thoma and Svee criteria. For the parameter sensitivity study, most of the results show the Svee criterion giving a smaller surge tank area than the Thoma criterion. A smaller surge tank is easier and less expensive to construct. The only parameter giving an opposite result is the turbine efficiency. For a system operating on a load higher than the best efficiency point, the Svee criterion gives a larger area. The area of the tunnels in the system is found to be the parameter most influential on the area of a stable surge tank. The case study shows similar results, smaller surge tanks for Svee except in high load cases. The numerical simulation shows slow dampening of the mass oscillations for a system dimensioned after the Thoma criterion, and unstable oscillations for the Svee criterion. This should imply that the Svee criterion gives a surge tank design that is unstable, but the limitations of a numerical simulation suggests a physical model test should be performed.

The objective of this master's thesis is to describe the existing surge tank solutions and additional improvements, and to simulate the hydraulic behavior of the surge tanks and how it affects the hydropower plant. The high head hydropower plant Torpa and the low head hydropower plant Åna-Sira are used for the case study, thus the surge tanks can be simulated under different hydraulic conditions. Turbine pressures and mass oscillations after a complete turbine shutdown, and turbine regulation parameters from operating the hydropower plant, are simulated in the LVTrans software. The result is a \emph{Surge Tank Atlas}, where the surge tank solutions are described with conceptual drawings, advantages and disadvantages, and documentation from the simulations. This Surge Tank Atlas provides a useful tool to compare different surge tank solutions for both a high head and low head hydropower plant. Some of the conclusions are valid for both hydropower plants and all the surge tank solutions. Increasing the distance from surge tank to turbine, decreasing the surge tank volume or a strong throttling of the surge tank will increase the water hammer significantly, and to some extent reduce the mass oscillations. An optimal throttling of the surge tank will stabilize the running of the hydropower plant, reduce the surge tank mass oscillations and hence the necessary height and volume of the surge tank. Other conclusions vary between the open and the closed surge tank, and the high and low head hydropower plant. The closed surge tank shows a better dampening of the water hammer than the open surge tank for both hydropower plant, the closed surge tank can be throttled stronger than the open surge tanks, and the surge tanks at the low head hydropower plant requires more volume than for the high head plant, even though the hydraulic head is a tenth of the high head hydropower plant. Hence there are some significant differences in the hydraulic behavior between the surge tank solutions and the high head and the low head hydropower plant, that will be presented and discussed in the Surge Tank Atlas.

The scope of this master thesis is to design a surge tank throttle based on literature values, and compare the expected head loss factor with the results from a 3D numerical simulation, and a physical model test. A literature review on throttle effects and the state-of-the-art throttle design will also be carried out as a part of the thesis. As a supplement to the thesis scope, the hydraulic effects of installing the throttle in the physical model will be discussed. The literature review reveal that a throttle has a high influence on the dampening of mass oscillations in a hydropower plant. In addition, a properly designed throttle lowers the volume demand of the surge tank, improves the governing stability for the machines, and reduce the maximum pressure in front of the turbine. The 1D numerical program LVTrans is used to calculate the hydraulic transients in the whole system and the 3D numerical program STAR-CCM+ is used to calculate the throttle head loss. The physical model used in this work is a 1:65 scale model of Torpa hydropower plant constructed at the Norwegian University of Science and Technology (NTNU). The load case for the experiments is a multiple closing and opening case to get resonance and bring the system to the maximum possible state of oscillations, which is a method used in Austria since the 1960s. The throttle design tested in this work is an asymmetric throttle with a relatively simple design, to reduce construction time and cost. The main parameter for the throttle design was found to be the smallest diameter. Two surge tank throttles with different diameters were made and used to compare the different methods. The literature values underestimate the head loss coefficient in both directions compared to the physical model test. The literature values show acceptable correlation to the physical model test in the upward direction, which is the direction with the simplest hydraulic conditions. The conditions in the downward direction are more complicated, and this is also reflected in the results, where the difference is larger. The 3D numerical simulations give results that are lower than the physical model test for downward flow, and higher for upward flow. The numerical simulations give the highest ratio between head loss in the upward and downward directions. The findings also shows a significant impact on the pressure development and dampening of mass oscillations by inserting a surge tank throttle, as was found in the literature review. The two throttles tested in the physical model reduced the maximum pressure by 5.2% and 3.9% for the smallest and largest throttle diameter respectively. This was with the use of the scaled maximum discharge from the prototype.

The objective of this thesis has been to evaluate the effect of throttling the surge tanks at Tonstad Hydropower Plant, by the means of one-dimensional numerical modelling in the program LVTrans. The background of the thesis is problems with the amplitude of mass oscillations in the surge tanks at Tonstad, causing restrictions on operation, due to the fear of drawing the surge tank water level down to a level where air enters the sand trap and initiates free surface flow. The numerical model of Tonstad hydropower plant, used for simulations, is currently running as a superset regulator at the plant. The calibration and validation shows good representation of steady state operation and period of mass oscillations. The amplitude of mass oscillations, does however show high deviations that are attributed to the inaccurate representation of transient friction in the numerical method. The simulations are interpreted relatively to minimise the error from the numerical model to the prototype, meaning that throttle effect is evaluated on the basis of improvement of mass oscillation amplitude from the restricted surge tank steady state water level. The critical situation for drawdown at this restriction level has been found to be with an output effect of 660 MW, reservoir levels at 482 m.a.s.l. in Homstøl and Ousdal, 49.5 m.a.s.l. in Sirdalsvann and with no inflow of water to the creek intakes. An optimization of throttle losses was performed by comparing a simulation of the current situation with simulations with varying throttle losses. The throttles asymmetric geometry was calculated from tabular values. The optimization finds that an asymmetric throttle, with loss ratio 1:1.5 from upwards to downwards flow respectively, may reduce downswing of the water level by 9.6 meters. A simulation where the restriction level in the surge tanks is reduced by 8 meters, show that the surge tank water level downswing is further reduced by 5.3 meters. It is concluded that the optimized throttle allows for a reduction of the restricted water level in the surge tank from 470 to 462 m.a.s.l., provided that all reservoir gates are fully open and water level at Ousdal is equal or higher than the water level at Homstøl. Some uncertainties connected with the numerical model are high, but these are outweighed by several conservative assumptions made in the simulations. The annual economic loss due to restricted operation is estimated to 2.5 million NOK, resulting in an allowed throttle cost of 33.3 million NOK to ensure profitability. The evaluation of surge tank throttling at Tonstad Hydropower Plant exemplifies benefits that may be achieved by detailed surge tank throttle design at other high head hydropower plants.

Sammendrag

Denne oppgaven omhandler trykkstøt og massesvingninger i vannveien for store vannkraftverk. Svingekammer er den mest brukte løsningen for å håndtere trykkstøt, som oppstår i vannveien når vannføringen endres. Luftputekammer er den nyeste teknologien innen svingekammer, og kan i mange tilfeller være et godt alternativ til åpen svingesjakt. Målet med denne oppgaven er å skape et bedre vurderingsgrunnlag for bruk av luftputekammer som løsning for svingekammer.Det har blitt utført feltmålinger på Oksla kraftverk i Tyssedal, og det samme kraftverket er modellert i beregningsprogrammet LVTrans. Til slutt har måleverdier og simuleringer blitt sammenlignet for å se om LVTrans klarer å gjengi de målte verdiene. Trykket i tilløpstunnelen ble målt like oppstrøms turbinen med en kvartscelle. For måling av vannivået i kammeret ble displayet på et gammelt måleinstrument filmet, og verdiene ble i etterkant registrert manuelt.Resultatet viser at simuleringene for tilløpstunnelen er sammenlignbare med målingene. Grafene har samme form og omtrent samme størrelse, men størrelsen på svingeamplituden avviker noe. Størst avvik er det for maksimal trykkøkning ved fullt avslag, der LVTrans gir omtrent 10 m større trykkøkning enn i målingene. For de øvrige lastendringene er avviket 1-3 m. Prosentvis ligger alle avvikene mellom 18 og 22 prosent. LVTrans gir derfor konservative beregninger med tanke på maksimalt trykk. Perioden i LVTrans er omtrent 60 s, det vil si 27 prosent lavere enn i målingene. Målingene for vannivået i luftputekammeret ble forstyrret av overflatebølger i kammeret, og det var derfor vanskelig å sammenligne målinger og simuleringer. For senere målinger anbefales det derfor å måle lufttrykket i kammeret i tillegg til vannivået, siden lufttrykket trolig ikke vil påvirkes av overflatebølgene.Lastrampen i regulatoren avgjør hvor raskt en lastendringene skal gjøres. Simulering med ulike verdier for lastrampen viser at den har stor betydning for trykksvingningenes amplitude. Jo raskere en lastendring gjøres, jo større blir trykksvingningenes amplitude. Det vil derfor være konservativt å velge en kort lastendringstid ved dimensjonering av vannvei og luftputekammer.Det er ikke funnet noe store driftsmessige problemer ved bruk av luftputekammer, men det bør nevnes at tapping og fylling av kammeret er en tidkrevende prosess. Dette vil imidlertid ikke være til hinder for den daglige driften. Lufttap fra kammeret kan være et problem dersom fjellet har høy permeabilitet og trykket i kammeret er høyere enn det omgivende grunnvannstrykket. Vanngardin har vist seg å være en god løsning for å redusere lufttapet, både som forebyggende og utbedrende tiltak.

Duge kraftverk er eit pumpekraftverk med effekt på 100 MW. I denne avhandlinga skulle det undersøkjast om kraftverket kan utvidast med større vassføring, og om det kan nyttast med hyppige avslag og pådrag. Duge kraftverk er bygd med ein 12 km lang utløpstunnel med tri svingesjakter. Det kan føra til store svingingar på eit så stort anlegg, og det kan gjeva utfordringar og setja avgrensingar for kva utvidingar som er praktiske å byggja. For å undersøkja kva som er mogleg å gjera, vart det nytta eit dataprogram som heiter LVTrans til å modellera kraftverket. LVTrans finn vasstrykk og vassføring i alle punkt som funksjon av tid. Det simulerer alle komponentar i kraftverket på ein realistisk måte, og kunne difor nyttast til å testa kapasiteten og til å prøva ut endringar i kraftverket.For at modellen skulle gjeva brukande resultat, måtte han matast med parametrar. Det vart gjort trykkmålingar for å gjera det mogleg å validera og kalibrera modellen. Det er svært mange parametrar som kan endrast på eit kraftverk, og sjølv med berre fokus på utløpstunnelen var det vanskeleg å få svingingane til å oppføra seg heilt likt.Modellen vart deretter brukt til forsøk med turbindrift og pumpedrift med avslag og pådrag på minst gunstige tidspunkt. Utan å byggja ut svingekammer, kan ikkje turbinene nyttast med vassføring over 141m3/s utan raske avslag og pådrag, og over 130m3/s med raske avslag og pådrag. Pumpedrift kan ikkje nyttast med vassføring over 120m3/s utan raske avslag og pådrag og over 94m3/s med raske avslag og pådrag. Det minst gunstige tidsrommet mellom to pådrag er 5--6\,minutt.