This thesis presents the results of three independent but related studies of aspects of the Mw 7.8 2012 Haida Gwaii earthquake, which was the second largest Canadian earthquake in recorded history. This event ruptured an area of roughly 150 by 40 km on a gently northeast-dipping thrust fault off the west coast of Moresby Island, British Columbia. This event was felt over 1600 km away from the epicentre, and produced tens of thousands of aftershocks. Adjacent to the mainshock fault plane is the Queen Charlotte Fault, the site of the largest event recorded in Canada: the 1949 Ms 8.1 strike-slip earthquake whose rupture extended as far south as this 2012 event and roughly as far north as an Mw 7.5 strike slip event which occurred on 5 January 2013. The 2012 thrust event was a surprise to some members of the seismological community as it ruptured a slab offshore of a major strike slip boundary. This earthquake therefore presents an excellent opportunity to constrain the tectonics and seismic hazard off the northwest coast of British Columbia. Herein a Coulomb stress transfer analysis is performed using finite fault models which incorporate both seismological and geodetic data. Static stress changes are projected onto optimally-oriented fault planes, determined using regional tectonic stresses in addition to mainshock stress; nodal planes, determined by aftershock centroid moment tensors; and onto the Queen Charlotte Fault. I find that aftershocks are generally consistent with Coulomb stress changes using optimal planes and known nodal planes, although the latter have slightly higher percentages of events consistent with triggering. I find that the Queen Charlotte Fault experienced stress changes greater than the empirically-determined threshold for triggering. This is particularly important as the southern extent of this fault is believed to lie in a seismic gap going back at least 116 years. With added stress from the mainshock and a lack of post-mainshock seismicity occurring in this seismic gap, it is a likely location for future earthquakes on this portion of the plate boundary. To obtain estimates of rupture parameters, an empirical Green's function technique and directivity analysis is performed. This method constrains rupture kinematics of the mainshock using systematic azimuthal variations in relative source time functions. My results indicate a rupture that progressed mainly to the northwest and updip. Subevent analysis confirms the existence of at least two subevents, with the first being roughly twice as large as the second. The results herein are similar to those found using finite fault inversion, but are better able to explain observed surface wave amplification at Alaskan seismic stations. My findings help support the idea that strong surface wave shaking may have resulted in delayed-onset dynamic triggering of the 2013 Craig event, through an unknown but intermediate mechanism that accounts for the two-month hiatus. Finally, an attempt was made to relocate all offshore aftershocks for this sequence by improving locations for events during a two-week ocean bottom seismometer deployment. This dataset includes a wider range of source-station azimuths and decreases the minimum source-receiver distance, relative to locations that only use land stations. My locations therefore represent the best-constrained depths for M greater than or equal to 3 offshore aftershocks occurring during the two-week deployment, and help constrain reasonable depth estimates from other relocation techniques. For events located using only ocean bottom seismometers I determine the time residual between observed and predicted phase arrivals at land stations to be used as a correction for all aftershocks recorded at land stations through the entire aftershock sequence. Although I was not able to find consistent time residuals I present suggestions for future implementations of this technique to this dataset, and discuss challenges associated with location of offshore earthquakes in regions with sparse regional seismic networks. All of these findings contribute to a more thorough understanding of this 2012 earthquake, as well as the tectonics of southern Haida Gwaii. I pay particular attention to identification of hazard within a seismic gap south of Moresby Island, and the northwest rupture directivity of the 2012 mainshock. / Graduate / 0373 / tiegan.hobbs@gmail.com

>Magister Scientiae - MSc / This research project had two different parts: The first was about the mapping of a section of the Warm Zand Structure and the collection of strain data. The second part focused on the analysis of the finite strain and strain pattern in the Warm Zand Structure. The Warm Zand Structure consists of strongly deformed calc-silicates of the Puntsit formation and feldpathic quarzites of the Goede Hoop formation which gradually change into pure quartzites in some locations. The second phase of folding F2 in the calc-silicates gave open folds. Strain markers are very scarce in the Warm Zand Structure; on the Emmanuel Farm pebbles were not found. However on the Compion Farm, pebbles are scattered and occur in few number. In the study area boudins were found in loose rocks hence they could not be used to estimate orientation of the XY plane of the strain ellipsoid (Ramsay, 1967). Folds on the other hand was common and was used to investigate the bulk shortening of the rocks and to understand the kinematical folding mechanisms involved in the folding process of the rock in the Warm Zand Structure. The strain contour map and the Sherwin and Chapple graph were used to investigate the strain and viscosity contrast respectively. The viscosity contrast of the folds collected in the Puntsit falls between 100-53 whereas the bulk shortening on the other hand is between 60-53%. The values of the shortening found agree with those found by van Bever Donker (1980). The numerical modelling on the other demonstrated that in addition to layer parallel shortening and the flattening mention by van Bever Donker (1980) Tangential Longitudinal Strain and Flexural Flow are also involved in the folding process of the fold collected in the Puntsit Formation.

Lean premixed combustion has potential for reducing emissions from combustion devices without compromising fuel efficiency, but it is prone to instabilities which presents design difficulties. From emissions point of view reliable predictions of species formation rates in the flame zone are required while from the point of view of thermo-acoustics the prediction of spatial variation of heat release rate is crucial; both tasks are challenging but imperative in CFD based design of combustion systems. In this thesis a computational model for turbulent premixed combustion is proposed in the RANS framework and its predictive ability is studied. The model is based on the flamelet concept and employs strained laminar flamelets in reactant-to-product opposed flow configuration. The flamelets are parametrised by scalar dissipation rate of progress variable which is a suitable quantity to describe the flamelet structure since it is governed by convection-diffusion-reaction balance and represents the flame front dynamics. This paramaterisation is new. The mean reaction rate and mean species concentrations are obtained by integrating the corresponding flamelets quantity weighted by the joint pdf of the progress variable and its dissipation rate. The marginal pdf of the progress variable is obtained using β-pdf and the pdf of the conditional dissipation rate is presumed to be log-normal. The conditional mean dissipation rate is obtained from unconditional mean dissipation rate which is a modelling parameter. An algebraic model for the unconditional mean scalar dissipation rate is proposed based on the relevant physics of reactive scalar mixing in turbulent premixed flames. This algebraic model is validated directly using DNS data. An indirect validation is performed by deriving a turbulent flame speed expression using the Kolmogorov-Petrovskii-Piskunov analysis and comparing its predictions with experimental data from a wide range of flame and flow conditions. The mean reaction rate closure of the strained flamelets model is assessed using RANS calculations of statistically planar one-dimensional flames in corrugated flamelets and thin reaction zones regimes. The flame speeds predicted by this closure were close to experimental data in both the regimes. On the other hand, an unstrained flamelets closure predicts flame speed close to the experimental data in the corrugated flamelets regime, but over predicts in the thin reaction zones regime indicating an over prediction of the mean reaction rate. The overall predictive ability of the strained flamelets model is assessed via calculations of laboratory flames of two different configurations: a rod stabilised V-flame and pilot stabilised Bunsen flames. For the V-flame, whose conditions correspond to the corrugated flamelets regime, the strained and unstrained flamelets models yield similar predictions which are in good agreement with experimental measurements. For the Bunsen flames which are in the thin reaction zones regime, the unstrained flamelet model predicts a smaller flame brush while the predictions of the strained flamelets model are in good agreement with the experimental data. The major and minor species concentrations are also reasonably well predicted by the strained flamelets model, although the minor species predictions seem sensitive to the product stream composition of the laminar flamelets. The fluid dynamics induced attenuation of the reaction rate is captured by the strained flamelets model enabling it to give better predictions than the unstrained flamelets model in the thin reaction zones regime. The planar flames and laboratory flames calculations illustrate the importance of appropriately accounting for fluid dynamic effects on flamelet structure and the scalar dissipation rate based strained flamelet model seems promising in this respect. Furthermore, this model seems to have a wide range of applicability with a fixed set of model parameters.

Rock-ice avalanches are three-phase flows composed of rock, ice, and a liquid. As their occurrences might increase in the future due to climate change, constructing mathematical and numerical models able to simulate these flows could be necessary for good hazard assessment and management in cold mountainous regions. With this aim, in this work, a framework of simplified rock-ice avalanche models is derived from a complete three-phase approach by applying two assumptions. Thanks to these two hypotheses, we obtain five classes of simplified mathematical models that simplify the rock- ice avalanche physics with different levels of approximation. Among these simplified approaches, the mathematical model, which simplifies the flow dynamics to a lesser extent, is a new mathematical model for rock-ice avalanches. For numerical purposes, a detailed analysis of the eigenvalues is performed for the one-dimensional depth-integrated version of the proposed model. Results show that the proposed approach loses hyperbolicity for specific ranges of the flow variables. Due to this feature, numerical modelling is performed by maintaining the numerical solutions in the hyperbolic domain of the flow variables. In this way, we consider the uniformly accelerated flow and the small perturbation of the flow depth and ice concentration as test cases. Additionally, we implement three numerical methods to identify the numerical scheme that can solve the proposed model accurately and to compare the obtained numerical results with those associated with the other simplified rock-ice avalanche approaches. Finally, we apply the linear stability theory to the proposed model to investigate its potential ill-posedness in the ranges of the flow variables where hyperbolicity is lost. Since the proposed approach turns out to be ill-posed, the model regularization is performed by trying to recover its hyperbolicity through a strategy suggested for a two- phase gas-liquid model. Although this strategy can enlarge the hyperbolic flow-variable range, hyperbolicity is still lost for specific flow conditions.

The research conducted in this study sought to appraise and validate the IEC-TC-114 technical specification for wave energy resource assessment in terms of its methodologies, requirements, and sources of error. This assessment was conducted through the pilot application of the IEC-TS to a project site located on the west coast of Vancouver Island in British Columbia. The objectives of this research were to (1) establish a firm scientific rational for the technical specification, (2) suggest improvements for future iterations of the technical specification, (3) conduct a sensitivity analysis to determine sources or error and uncertainty within the specification, (4) develop a user guidance , and (5) to complete an assessment of the wave energy resource for all three classes of assessment at the pilot project site. At the time of this research there has been limited assessment and validation of the technical specification and its applicability to real world studies. This study is also limited in its scope to a singular site within Canada and so specific observations made may not be applicable to all locations. It was observed over the course of this study that the IEC-TS could be applied to the assessment of wave energy with reasonable computational effort. Additionally, the IEC-TS is overall relatively easy to understand and apply, with the only exception to this being the validation procedure which lacked clarity. However, it was noted that in many cases the IEC-TS would not provide defined requirements, but only recommendations which would allow researchers to potentially disregard the recommendations provided within. While this is likely due to the fact that different sites may have different requirements, it does call into question the need for a standard if it is not going to have strict requirements to ensure that all resource assessments are completed using state of the art knowledge and procedures.

<p dir="ltr">Inverse problems in infinite dimensions are ubiquitously encountered across the scien- tific disciplines. These problems are defined by the need to reconstruct continuous fields from incomplete, noisy measurements, which oftentimes leads to ill-posed problems. Almost universally, the solutions to these problems are constructed in a Bayesian framework. How- ever, in the infinite-dimensional setting, the theory is largely restricted to the Gaussian case, and the treatment of prior physical knowledge is lacking. We develop a new framework for Bayesian reconstruction of infinite-dimensional fields which encodes our physical knowledge directly into the prior, while remaining in the continuous setting. We then prove various characteristics of the method, including situations in which the problems we study have unique solutions under our framework. Finally, we develop numerical sampling schemes to characterize the various objects involved.</p>

Yes / This study focuses on numerical modeling of condensation of water vapor in a Laval nozzle, using the liquid drop nucleation theory. Influence of nozzle geometry, pressure, and temperature on the average drop size is reported. A computer program written in MATLAB was used used to calculate the nucleation and condensation of water vapor in the nozzle. The simulation results are validated with the available experimental data in the literature for steam condensation. The model reveals that the average drop size is reduced by increasing the divergent angle of the nozzle. The results also confirm that increasing the inlet pressure has a direct effect on the average drop size while temperature rise has an inverse effect on the drop size.

The large proportion of energy used in the built environment has made improving energy efficiency in buildings, in particular their heating, ventilation, and air conditioning (HVAC) systems, a policy objective for reducing energy consumption and CO2 emissions nationally and internationally. Ground source heat pump (GSHP) systems, due to their high coefficient of performance (COP) and low CO2 emissions are consequently, receiving increasing attention. This work is concerned with the modelling of borehole heat exchangers (BHEs), the commonest form of ground heat exchangers found in GSHP systems. Their careful design is critical to both the short timescale and long timescale performance of geothermal heat pump systems. Unlike conventional components of HVAC systems, BHEs cannot be designed on the basis of peak load data but require 3 application of dynamic thermal models that are able to take account of the heat transfer inside the borehole as well as the surrounding ground. The finite volume method has been applied to develop a dynamic three-dimensional (3D) model for a single BHE and BHE arrays. The multi-block boundary fitted structured mesh used in this model allows the complex geometries around the pipes in BHEs and the surrounding ground around the borehole to be represented exactly. The transport of the fluid circulating along the pipe loop has been simulated explicitly in this model. The ground underneath the borehole has also been represented in this model. Validation of the 3D model has been carried out by reference to analytical models of borehole thermal resistance and fluid transport in pipes, as well as experimental data. In this work, the 3D numerical model has been applied to investigate the three-dimensional characteristics of heat transfer in and around a BHE at both short and long timescales. By implementing a two-dimensional (2D) model using the same numerical method and comparing the simulation results from the 3D and 2D models, the most significant three-dimensional effects have been identified and quantified. The findings have highlighted some of the limitations of 2D models, and based on the findings, methods to improve the accuracy of a 2D model have been suggested and validated. Furthermore, the 3D and 2D finite volume models have been applied to simulate an integrated GSHP system and their effects on overall system performance predictions have been investigated. The 3D numerical model has also been applied to examine thermal interactions within BHE arrays and to evaluate the assumptions of the line source model and their implications in the analysis of thermal response test data.

An oscillating water column (OWC) wave energy converter (WEC) is a device designed to extract energy from waves at sea by using the water to move trapped air and thus drive an air turbine. Because the incident waves and the force caused by the power take-off (PTO) interact, control of the power take off (PTO) system can increase the total energy converted. A numerical model was developed to study the interaction of an OWC with the water and other structures around it. ANSYS AQWA is used here to find the effects on the water surface in and around the central column of a five-column, breakwater-mounted OWC. For open OWC structures, coupled modes were seen which lead to sensitivity to incident wave period and direction. The frequency-domain displacements of the internal water surface of the central column were turned into a force-displacement, time-domain model in MATLAB Simulink using a state space approximation. The model of the hydrodynamics was then combined with the thermodynamic and turbine equations for a Wells turbine. A baseline situation was tested for fixed turbine speed operation using a wave climate for a region off the north coast of Devon. A linear feedforward controller and a controller based on maximising turbine efficiency were tested for the system. The linear controller was optimised to find the combination of turbine speed offset and proportional constant that gave maximum energy in the most energy abundant sea state. This increased the converted energy by 31% in comparison to the fixed speed case. For the turbine efficiency control method, the increase was 36%. Energy conversion increases are therefore clearly possible using simple controllers. If increased converted energy is the only criterion for controller choice, then the turbine efficiency control is the best method, however the control action involves using very slow turbine speeds which may not be physically desirable.

Thesis (MEng)--Stellenbosch University, 2015. / ENGLISH ABSTRACT: Rubber is the main component of the pneumatic tire. When rubber is put under cyclic loading, like when a tire is rolled, heat is generated and stored in the rubber, due to hysteresis. Heat stored in the tire is increased by factors like under-inflation, overloading, speeding and defects in the tire. The heat causes high temperatures in the tire due to the poor thermal conductivity of rubber. When the temperature in the rubber increases to 185 °C, pyrolysis and thermo-oxidation starts and can cause the tire to eventually explode. A numerical model of a rolling passenger vehicle tire was developed to calculate the temperature distribution inside the tire and analyse the effect of different operating conditions on the temperature. Operating conditions include loading, inflation pressure, rolling velocity and ambient temperature. The tire was modelled by a single rubber type, using the Mooney-Rivlin material model. The bead wire was modelled using an isotropic material model, while the body and steel cord plies were modelled as rebars. The cavity, used to inflate the tire, included the pressure increase due to the volume change, when the tire is loaded. The numerical model was validated using experimental data from tests done on an actual tire. These tests included deformation and contact stress analysis, as well as surface temperature measurements. Numerical results showed an increase in temperature when the load, rolling velocity and the ambient temperature were increased, as well as when the inflation pressure was decreased. The trends of the numerical data matched the trends of the experimental data. However, the values of the numerical model were not consistent with the experimental data due to material properties from literature being used to model the tire. / AFRIKAANSE OPSOMMING: Rubber is die hoofkomponent in die pneumatiese band. As rubber onder ’n sikliese las geplaas word, soos wanneer ’n band rol, word hitte gegenereer en in die rubber gestoor as gevolg van histerese. Die hitte wat in die band gestoor word, word verhoog deur faktore soos lae inflasiedruk, hoë las, hoë rol snelhede en gebreke in die band. Die hitte veroorsaak hoë temperature in die band weens die swak termiese geleiding van rubber. As die temperatuur in die band hoër as 185 °C raak, vind piroliese en termo-oksidasie plaas en die band kan uiteindelik ontplof. ’n Numeriese model van ’n passasiersmotorband is ontwikkel om die temperatuurverspreiding te bepaal, asook om die effek van verskillende werkstoestande op die temperatuur te analiseer. Die band is gemodelleer met een tipe rubber en die Mooney-Rivlin materiaal-model is gebruik om die rubber te beskryf. Die spanrand van die band is deur ’n isotropiese materiaalmodel gemodelleer, terwyl die hoof- en staalkoordlae as bewapening gemodelleer is. Die holte wat gebruik word om die band op te blaas, neem die druk toename as gevolg van die verandering in volume in ag wanneer die band belas word. Die numeriese model was bekragtig met eksperimentele data wat deur toetse op ’n werklike band onttrek is. Die toetse sluit die volgende in: vervormingen kontakspanninganalises, asook temperature wat op die oppervlak van die band gemeet is. Die numeriese resultate toon ’n toename in temperatuur wanneer die las, rolsnelheid en omgewingstemperatuur verhoog word, asook waneer die inflasiedruk verlaag word. Die numeriese model se tendense stem ooreen met die eksperimentele data, maar die waardes van die numeriese model is nie in ooreenstemmig met die eksperimentele data nie. Die verskil is as gevolg van die materiaaleienskappe wat uit die literatuur geneem is.