There is a lack of standardized rules on how the fluid in the turbine should be included in rotor models of hydraulic machinery. This thesis is an attempt to shed some light on this issue. We approach the problem from two viewpoints, situated at place at a hydropower plant and by mathematical analysis.One goal of the thesis is to develop a measurement system that monitors the instantaneous pressure at several locations of a runner blade on a 10 MW Kaplan prototype in Porjus along Lule river. Paper A outlines the development of the measurement system and the instrumentation of the runner blade. Miniature piezo-resisitive pressure transducers were mounted flush to the surface. If instrumentation is successful, the pressure field of the runner blade could be measured simultaneously as the loads and displacements of the guide bearings and the generator.The second objective is concerned with how the motion-induced fluid force affects the dynamic behaviour of the rotor. Inertia and angular momentum of the fluid and shrouding are expected to influence the dynamic behaviour of the turbine. Paper B scrutinizes this assumption by presenting a simple fluid-rotor model that captures the effects of inertia and angular momentum of the fluid on the motion of a confined cylinder. The simplicity of the model allows for powerful analytical solution methods. The results show that fluid inertia, angular momentum and shrouding of hydraulic turbines could have substantial effects on lateral rotor vibrations. This calls for further investigation with a more complex fluid-rotor model that accounts for flexural bending modes.

Lightweight floor structure is widely used in building industries and to have better sound insulation builders come up with different ways of construction. Depending on the construction the floor structure could either be coupled (floor and ceiling coupled by beams) or decoupled (no mechanical connection between floor and ceiling). Although there are many models on coupled structure but for decoupled structure the number is not too many. Keeping that in mind the present thesis talks about lightweight floors: the construction, properties, behaviour etc with a focus on developing a model for decoupled floor structure where the core contribution being the decoupling and adding the moment effect at plate beam joints. The advantage of decoupled structure is that it disconnects the sound bridge through the beams. One consequence on the other hand is that cavity resonance dominates the low frequency region. A comparative analysis is also done with the coupled model. While developing the model this talks about different mathematical tools such as Fourier transform, Floquet principle, Poisson's sum formula etc This also gives an overview of different types of modelling technique available such as analytical, Numerical, energy based approach, empirical method etc. A parametric study is also done here to find out the relative influence of different elements on sound pressure level.

Complete vehicle, system and component target setting for noise, vibration and harshness (NVH) are important activities within automotive product development. New challenges arise when electric motors are introduced into cars traditionally powered by internal combustion engines (ICE). Although the electric powertrain is generally quieter than ICEs, the sound character however is completely different and necessarily not more preferable. The noise signature from an electric traction motor is characterized by speed-dependent high frequency tonal components from the dominating electro-magnetic harmonics, covering a wide rpm-range. With relatively low levels of broad band masking noise from tires and wind, the tonal components can be accented in a large frequency range and contribute to perceived annoyance for the occupants inside the car. Methods for good practice of NVH development work for traditional vehicles have been established and refined during the last decades. With the rapid increase of various types of electric powertrain-equipped low emission vehicles, new methodology is required in order to understand and fulfill expectations on these unaccustomed vehicle-acoustic phenomena. The objective with this thesis was to investigate the audible perception of the electric powertrain and develop and validate methods for assessment of noise transmission. The findings aim to support the complete vehicle, system and component target setting processes to secure a desired sound quality in future electric vehicle launches. The thesis is based upon four papers. The first two papers address the human perception of the sound from the electric powertrain during driving. In Paper 1, the evaluation methodology of electric powertrain sound quality was investigated. It was concluded that the mean ratings of a specific acceleration sound stimulus was judged similarly in two laboratory test environments (listening room and “sound car” respectively) as during on-road driving. Further, with lower level of high frequency (>1 kHz) tonal components, the overall satisfaction and annoyance ratings were improved. The annoyance due to tonal components, stationary as well as speed-dependent, was further studied in Paper 2. The psycho-acoustic metric prominence ratio (PR) was used to quantify the tone levels relative to the adjacent background noise. It was found that for PR-levels below 3 dB, the perceived annoyance was similar independent of frequency range. Increased PR-level provided higher probability of detecting the tones and the perceived annoyance was significantly increased compared to PR<3 dB for tones above 2.5 kHz. The findings from Paper 1 and 2 provide foundation for complete vehicle interior acoustic target specification. The following two papers (Paper 3 and 4) focus on the main mechanical system properties that are responsible for the airborne radiation and transfer of electric powertrain induced noise. Paper 3 has emphasis on the experimental determination of acoustic transfer functions (ATFs) between the car’s interior to multiple positions located on the stator shell surface of an electric rear axle drive. It addresses the ATFs’ spatial resolution with respect to estimated interior noise for theoretically derived and measured surface vibrations. A reciprocal relation was assumed; therefore the measured ATFs could be used to inversely reproduce the sound transmission from the stator shell to the interior. ATFs and surface velocities are also central in Paper 4, where a method for in-situ determination of radiated sound power from a stator shell was presented and validated. By acquiring the operational deflection shape due to an electro-magnetic radial force wave, a simplified expression for the radiation efficiency was estimated. Paper 3 and 4 highlights the advantages of the rotational symmetric force excitation acting on the stator housing which enables sound power as an acoustic source characterization which perhaps is the most adequate measure for system target setting directed towards external suppliers.

The overarching aim of this work is to study the freezing process of a single water droplet freezing on a cold surface, which is an interesting and important phenomenon with possible applications in many areas. Understanding the freezing process of a single water droplet is for example an important step when preventing unwanted icing, e.g. in the case of airplane wings and propellers, wind turbine rotor blades, and road surfaces.As a step in understanding the freezing process, the study specifically focuses on the internal flow in the droplet during the freezing process. To do this, the study combines the use of Computational Fluid Dynamics (CFD) to build a model of the freezing process and experimental methods, i.e. Particle Image Velocimetry (PIV) to validate the numerical results. Focus is to start with the heat- and mass transfer inside the droplet using simple geometries with a rigid boundary, not modelling the outside environment as the air and the cooling plate. These components will be incorporated in the model further on.Three papers will be included in the study. In Paper A the CFD model is created and tested on a simple 2D-geometry of a droplet. The numerical result is partially compared to experimental work found in literature. In Paper B the numerical model is developed even further and a more realistic geometry of a real droplet, although with rigid boundaries, is used. The numerical results are as for Paper A validated with experimental results found in literature. In Paper C the internal flow inside the droplet has been investigated experimentally to estimate the velocities in the water, so that in the future the results can be used to validate the numerical work.The results show that is possible to work with a very simple CFD model and still capture the main flow features and freezing characteristics in a freezing water droplet. In line with previous research, this study confirms that the natural convection induced by gravity is significant for the internal flow, as compared to conduction and effects of ice creation. If studying the freezing time the internal flow has little effect. However, when estimating the velocities in the water it is crucial. It can be seen that the gravity effects are most pronounced around the density maximum for water (at T = 4◦C). The experiments show that the method used to study the flow inside the droplet is a working method, and the velocities in the water has been estimated. The next step is to further develop the CFD model and validate the numerical work with the experimental results. An interesting next step is to incorporate a moving interface to capture the volume expansion during the phase change.

The pelletizing process where the crude ore from the mine is upgraded to pellets is a process which includes several stages involving complex fluid dynamics. In this thesis, focus is on the grate-kiln pelletizing process and especially on the rotary kiln, with the objective to get a deeper understanding of the aerodynamics and its influence on the combustion process. The aim is to discover flow features taking place in the kiln, and the kiln hood, by using Computational Fluid Dynamics (CFD) on simplified models of the real kiln, and to validate the set-ups of the numerical model with physical experiments using Particle Image Velocimetry (PIV) and Laser Doppler Velocimetry (LDV). By starting as simple as possible, studying only the cold flow field without combustion and validating the simulations, a foundation for future geometrical optimizations can be achieved. Later on more realistic geometries may be studied with the validated simulations as a base. In Paper A the initial down-scaled, simplified model of the real kiln is studied, and both numerical and experimental analyses of the flow field are performed. Paper B focuses on the turbulent secondary flow that arises in ducts with non-circular cross-section. One of the inlet ducts to the kiln of interest here is close to semi-circular in cross section, hence the focus of this work. Numerical and experimental results are reported. Paper C is a development of the model, where instead of parallel inlet ducts as in Paper A, the top one has an inclination angle to the kiln axis. A thorough experimental analysis of the flow field is performed in this case. Conclusions are that steady state simulations can be used to get an overview over the main features of the flow field. Precautions should though be taken when analyzing the recirculation zone which is important for the flame stabilization. A stable flame is safe and crucial for efficient combustion. Steady state simulations do not capture the transient, oscillating behavior of the flow seen in the physical experiment. These oscillations will under certain conditions considerably affect the size of the recirculation zone. Another parameter affecting the size of the recirculation zone is the inclination of the upper inlet duct, where a decrease in recirculation length is seen although the actual inclination of the incoming jet is only about 3-4º. The choice of turbulence model affects the prediction of turbulent secondary flow. If this flow feature needs to be revealed, a more advanced turbulence model should be used. / Fastelaboratoriet - VINNEXC

In an iron ore pelletizing plant, crude ore is upgraded to pellets to be used as feedstock in steel-making plants. As part of a grate-kiln pelletizing plant, the rotary kiln is an indurating furnace in which the pellets are sintered. The rotary kiln involves complex flow of large amounts of gas and the process is strongly coupled to the fluid dynamics, which is not well understood. The present work focuses on increasing the understanding of the aerodynamics of the rotary kiln.Though the kiln geometry is relatively simple a rather complex flow arises, which is known to occur for turbulent flows in similar geometries. In order to isolate the underlying flow mechanisms, simplified models of the kiln are studied both numerically using Computational Fluid Dynamics (CFD) and experimentally using Particle Image Velocimetry (PIV). The understanding of the flow phenomena that arises for the simplified models is essential for maintaining a solid comprehension of the fluid dynamics when increasing the complexity of the models. Computations are validated against available experimental data to evaluate the capability of the numerical procedure in capturing the underlying physics of the flow. In this way, the reliability of the predictions is improved when increasing the complexity of the model.In Paper A the unsteady non-reacting flow is computed and a preliminary coal combustion model is proposed, which is in need of further development to yield reliable predictions of the reacting flow. Paper B is an experimental investigation of a down-scaled model of the kiln and also an extension to previous experimental work by introducing an inclination of the upper inlet duct to the kiln and carrying out a more thorough analysis of the fluid dynamics. In Paper C, the periodic flow observed in Paper A is investigated further using a more sophisticated turbulence closure and carefully validating the predictions against available experimental data.For the simplified models under investigation, it is concluded that the flow is dominated by the periodic shedding and downstream convection of von-Karman-like vortices originating in the free shear layers enclosing the recirculation zone formed in the inlet end of the kiln. Both numerical and experimental investigations show a strong dependence of momentum flux ratio between the two inlet ducts on the flow field. The large-scale periodic fluctuations, which are resolved in an unsteady computation but completely neglected in a steady computation, are seen to contribute significantly to the turbulent transport in the recirculation zone. This indicates the need for unsteady computations to accurately predict the transport processes. The recirculation zone is important for flame stabilization as it feeds back hot gas to the near-burner region. Hence, a challenging requirement of the numerical model is to accurately capture the physics of the recirculating flow. The use of a second-moment turbulence closure is shown to significantly improve the predictions over the use of an eddyviscosity turbulence model and give promising results for further work on more complex models of the kiln.

A model has been developed that can be used to predict the transport of non-spherical particles in the nano- and micro scale range for different applications. This may be the flow in the lungs or flow taking place during composites manufacturing, but the model can be applied to many applications where the particle Stokes and Reynolds numbers are small. The model can, for instance, be used to simulate an evenly random distribution of particles and then follow them through a laminar flow in a straight circular tube, either to study the statistical congregation of multiple particles or to follow the path of an individual particle. Both gravitational settling and Brownian motions are included in the model and their influence was also examined. To increase the understanding of the influence of the breathing pattern on the deposition of inhaled nano- and micro-fibres simulations were done in a straight model airway. Maximum deposition rate was found when particles were released in the beginning of therespiratory cycle while a minimum when the release came at the peak of inhalation. A comparison was done of a cyclic flow field and a quasi-steady one to see if the latter could accurately be used to replace the former. A quasi-steady solution generally provides a relatively good approximation to cyclic flow if an average velocity over one residence time of the particles moving with the mean fluid velocity is used. A statistical study was done to compare the deposition rates of oblate and prolate particles of different size and aspect ratio as they travel down narrowing bronchi in a steady, fully developed parabolic flow field. The model shows a clear correlation between increased particle size and increased deposition, it also consistently yielded a higher deposition rate for oblate particles compared to prolate particles with a similar geometricdiameter. A study of the motion and orientation of single oblate and prolate particles with large aspect ratio and the same geometric diameter has also been done. To see the effect the different forces have on the particle it was first studied with only the force of the flow field acting on it. Clear Jeffery orbits were visible in the simulations, although the periods of the orbits were shorter for the oblate particles than the prolate ones. When Brownian motion was introduced the motion of the particles became less periodic. For prolate particles Jeffery orbits could still be distinguished, unlike for the oblate particles whose movements mostly resembled random tumbling. In some methods to produce fibre reinforced polymer composites a fabric is impregnatedwith a fluid that may contain particles on the micro- and nano scales. Theseparticles are aimed to give the final product additional properties. It is therefore interesting to be able to reveal how the distribution and orientation of such particles are affected by the processing condition. During the manufacturing of the fabric and during the subsequent lay-up in a mold relatively large channels are formed between bundles of fibres where the impregnating fluid flows; there is also micro channels within the bundle that are also impregnated by the fluid and the capillary action there may be modelled as a suctioning force on the walls of the channels. Therefore in this study the channel between the bundles are represented as a tube with a circular diameter and a flow fieldthat are being sucked to the sides as it travels down the tube. A random distribution of particles is introduced at the inlet of the channel and the deposition is studied and the results are compared to a case when the flow is purely driven by an applied pressure gradient without any suction on the walls.

Excessive release of phosphorus from the effluent of on-site sanitation facilities to the environment is one source of aquatic pollution. The primary effect of excess phosphorus in the aquatic environment is eutrophication. In Sweden, there are about one million private on-site wastewater treatment facilities, half of which are considered unacceptable by the Swedish Environmental Protection Agency (SEPA). Recent guidelines issued by SEPA recommended total phosphorus reduction of 70-90 % in wastewater in small scale treatment facilities. Using reactive filter materials as compact filter systems following a biological pretreatment can be a convenient option for upgrading the present on-site sewage treatment facilities. Research on reactive filter materials has been undertaken in many different ways as there is no standard method for such investigations. Therefore, there is a need to develop methods and models to evaluate the reduction capacity and predict the longevity of potential reactive filter materials.In the first part of this thesis a filter material Filtra P (Nordkalk, Finland) is investigated. The phosphorus sorption characteristics of Filtra P were investigated in batch and column experiments. The batch test method was evaluated with respect to the effects of liquid-to-solid ratio and particle size on phosphorus sorption. Column experiments were carried out at different flow rates and phosphate (PO4-P) feed concentrations. The experimental data obtained from batch and column experiments were used to investigate the feasibility of modeling PO4-P breakthrough curves in fixed-bed columns by means of the convection-dispersion equation and a Langmuir isotherm. The model successfully predicted the PO4-P breakthrough curves for the columns with low influent concentration (11 mg dm-3), which is within the range normally found in wastewater (5-15 mg dm-3). Experimental observations however yield that the possibility to combine the convection-dispersion equation with a model accounting for precipitation of calcium-phosphate compounds in the bulk should be examined.The second part of the thesis focuses on dispersion in laminar flow through a randomly packed porous-medium model. The study of dispersion in flow through porous media is an important issue in many branches of science and engineering and has direct applications in several industries. For example, dispersion occurs in the disposal of sewage waste into aquifers and flow through reactive compact-bed filters that are used for on-site sanitation systems. In order to shed light on this phenomenon, a detailed longitudinal and transverse dispersion numerical experiments were performed in two-dimensional (2D) randomly packed beds of thousands of circular particles in a laminar flow regime. The calculated longitudinal and transverse dispersion coefficients were presented as Peclet numbers based on longitudinal and transverse dispersion coefficients versus Peclet number based on molecular diffusion coefficient, and were compared to three-dimensional (3D) experimental data available in the literature. It was found that at very low superficial liquid velocities both longitudinal and transverse dispersions were governed by molecular diffusion. Based on the comparisons between numerical results obtained from the 2D model to the 3D experimental data, it was found that the present 2D approach to study the longitudinal dispersion in flow through porous media can be considered to be acceptable.The present numerical data may be used to estimate the longitudinal and transverse dispersion coefficients in a 2D process by which one fluid is displaced from a porous medium by a second fluid which is miscible with the first in a laminar flow regime.

This licentiate thesis deals with the topic of variations and uncertainties in building acoustic parameters for lightweight timber constructions. A higher safety margin to the legal requirements is needed to compensate for acoustical uncertainties, which leads to higher costs. Building costs can be reduced if the variations can be identified and controlled. The project was limited to industrially prefabricated timber frame based volumes and massive timber based plate elements. This thesis is based on the work reported in three papers (A, B and C). In paper A, the variations in impact and airborne sound insulation were assessed and quantified in 30 nominally identical volume built apartments in a four-storey building. Large variations were found and the underlying causes were investigated. A statistically significant difference between floor numbers was found as the highest floor achieved better sound insulation. This difference was assumed to be caused by the higher static load on lower floors affecting the elastic layer used to structurally connect the apartments. In paper B, three room volumes were followed and measured at different stages of completion throughout the construction process. The objective was to test if acoustical deviations in the field can be identified at earlier construction stages. An ISO tapping machine was used to excite the floors and the response was measured at 20 positions. The airborne and impact sound insulation were measured in the finished building. Deviations were found, but these could not be traced to earlier stages of completion. In Paper C, the variations in sound insulation of a cross-laminated timber (CLT) building system was investigated. The construction was based on prefabricated wall and floor plate elements which were mounted at the building site. A number of acoustical uncertainties related to the measurement procedure were also investigated. The measurement uncertainty was small in comparison to the total variations. The degree of prefabrication for the CLT system was lower compared to the volume system, which indicated a greater scope for poor workmanship. All papers indicate a higher sound insulation on the upper floors in a building. It is therefore important to carefully design the elastic layer between floor numbers. The measurement uncertainty has been continuously considered in this thesis. In order to properly identify and quantify variations, the measurement uncertainty should be minimised. Advantages and drawbacks with different measurement methods and directions for future research are discussed in the concluding chapters.

Open channel flow near hydropower stations is of interest for both engineering and environmental applications. In this research project Computational Fluid Dynamics simulations of free surface flow in regulated rivers were applied with both fish migration and validation of numerical simulations in focus. In the first paper, numerical simulations has been used to evaluate the flow downstream a hydropower plant with regards to upstream migrating fish. Field measurements with an Acoustic Doppler Current Profiler were performed and the measurements were used to validate the simulations. In the second paper ,simulations on the spilling from a dam were performed and compared to experimental results from a physical scale model. The third paper deals with more in depth analysis of field measurements where the variations in the flow downstream a hydropower plant was examined.