<p> The logit format for a modal split model, which has previously been used for only binary cases, is used to build a new set of behavioural, probabilistic, multi-mode models. The models and the testing were carried out on a CDC 6400 Computer. </p> <p> A program developed at Chicago was used to construct the models while a separate program was developed to analyze the results. The type and number of variables to be used in the different sections of the model were investigated and an attempt was made to find the best method of aggregation. An inferred 'value of time' was also calculated and statistical testing of the individual and aggregate models was made. </p> <p> It is shown that this method of modelling is indeed feasible in terms of the significance of the models and the accuracy of the predictions on a separate data set. </p> / Thesis / Master of Engineering (ME)

No description available.

Optical waveguide structures and devices are the fundamental basic building blocks of photonic cireuits which play important roles in modern telecommunication and sensing systems. With the fast development of fabrication technologies and in response to the needs of miniaturization and fast increased functionality in future integrated photonic chips, various structures based on high-index contrast waveguides, surface plasmonic polaritons structures, etc., have been widely proposed and investigated. Modeling and simulation methods, as efficient and excellent cost performance tools comparing to costly facilities and time-consuming fabrication procedures, are demanded to explore and design the devices and circuits before their finalization. This thesis covers a series of techniques for modeling, simulation and design of photonic devices and circuits with the emphasis of handling of radiation wave and the related power couplings. The fundamental issue in optical waveguide analysis is to obtain the complete mode spectrum. In principle, we need the radiation modes to expand the arbitrary fields of an open waveguide. In practice, however, the continuum nature of the radiation modes makes them hard to use. The discrete leaky modes may approximately represent a cluster of radiation modes under some circumstance and can be utilized in mode expansion together with guided modes to significantly simplify the analysis of mode coupling problems in optical waveguides. However, the leaky modes are unbounded by nature and hence lack the usual characteristics of normal guided modes in terms of normalization and orthogonality. Recently a novel scheme for handling of radiation optical fields was proposed and demonstrated by applying perfectly matching layers (PML) terminated with a perfectly reflecting boundary (PRB) condition. In this scheme, the radiation fields are represented in terms of a set of complex modes, some of which resemble the conventional leaky modes and others associated with the interaction between the PML media and the reflecting numerical boundaries. The mode spectrum is therefore split into the guided modes and complex modes which possess the normal mode features such as normalization and modal orthogonality. The seemingly paradoxical application of both the PML and PRB in the new method has in fact overcome one of the main challenges assoiated with this traditional method, i.e., the desire for discrete, orthogonal, and normalized modes to represent radiation fields and the need for elimination and reduction of spurious reflections from the edges of the finite computation window. With the understanding of mode spectrum, a full vector mode matching method and a complex coupled mode method for analyzing the wave propagation in optical waveguides under the framework of PRL and PRB computation model have been proposed. The methods have been validated through various structures such as waveguide facet, polarization rotators, long/short period gratings etc. Then the proposed techniques have been utilized to design a series of waveguide structures based on surface plasmonic polaritons, slot waveguides etc. / Thesis / Doctor of Philosophy (PhD)

The objective of the study was to investigate multi-modality effects (visual and auditory) on the perception of urgency in elderly individuals addressing issues of safety and comfort. Twenty individuals (10 young, 10 elderly) from Montgomery County of Virginia participated in the laboratory study. In the pre-experimental audiometric tests, as expected elderly individuals had higher hearing thresholds (dBHL) than young across all the frequencies (250, 1000, 2000 and 4000 Hz) of pure tones. The difference was more pronounced at 4000 Hz. In visual acuity tests administered, elderly individuals had a lower Snellen VA than young at both near and far distances. The participants also were administered Bausch and Lomb color test; all participants were able to differentiate red color. Method of adjustment was used to collect empirical data. Rating test was administered after every treatment to objectively assess the participants' feeling of perceived urgency associated with the auditory warning signal. All the participants rated the experimental sound as urgent. Repeated measures was conducted to remove variations due to time and treatment conditions. A mixed-factorial design was used to investigate the main and interaction effects. The age effect was found to be marginally statistically significant. Modality and trial effects were not found to be significant. The reanalysis conducted after removal of outliers resulted in marginal statistical significance for age and modality main effects. Although not significant, a trend was seen in the pulse intensity levels (dBL) across modalities in the elderly individuals. Future research should be conducted to investigate modality effects in elderly individuals in a naturalistic driving environment which would give a better insight of the effects of dual modalities. / Master of Science

Many studies have attempted to detect structural damage by examining differences in the frequency response functions of a structure before and after damage. In an experimental setting, this variation can not be attributed solely to the addition of damage. Other sources of variation include testing and structure variation. Examples of testing variation include the error introduced by modal parameter extraction, measurement noise, and the mass loading of the accelerometer. Structure variability is due to slight differences in the supposedly identical structures. Dimensional tolerancing is one example. This study began with six "identical" undamaged aluminum cylinders, of which three were later damaged to varying extents. The frequency response functions of the undamaged and damaged cylinders were measured. Also, the frequency response function of the same undamaged cylinder was measured multiple times to investigate testing variation. The contributions of testing, cylinder, and damage variation to the differences between cylinder responses was elucidated by specifically examining their frequency response functions in two ways: comparing the natural frequencies and directly investigating the entire frequency response function. The curvature of the frequency response functions was then used to determined the presence, location, and severity of the imparted damage. / Master of Science

<p dir="ltr">This study encompasses the design and dynamic analysis of a previously used compact, portable vibrating plate machine. Utilizing Siemens NX 2021 for the precise modeling of the machine's components, the design prioritized simplicity and functionality, resulting in a 450 mm x 450 mm aluminum alloy structure, suitable for a wide range of research applications. A detailed modal analysis, conducted to ascertain the system's natural frequencies, revealed six predominant modes, ensuring operational frequencies of 110 Hz to 130 Hz were strategically avoided to mitigate resonance risks. Complementing this, harmonic response analysis evaluated the system's behavior under an applied cyclic load, confirming the suitability of the chosen actuator, model VL181206-160H, which provides optimal vibrational force without over stressing the machine. The findings affirm the machine's capability to perform efficiently within the target frequency range, with the design and selected actuator offering a robust solution for consistent and safe vibrational analysis, essential for field and laboratory applications.</p>

No description available.

The constituent parts of a magnetorheological (MR) sandwich beam combine to create a powerful damping device. The apparent viscosity of an MR fluid can change with the application of an adjacent magnetic field. Typically, this apparent viscosity rises as the field gets stronger. The resulting resistance to motion makes MR fluids a great choice for damping applications. The sandwich beam is a composite structure with two elastic outer layers and a damping material in between that forms the core layer. Using MR fluid as the damping agent produces good and tunable damping performance while possessing the benefits of a sandwich beam, such as its ease of construction and its form factor. This dissertation pursues two goals. The first goal is related to modeling a beam's dynamics. In addition to creating a new beam model, this research tests the limitations of the DiTaranto model, a popular model in the literature. The second goal is to apply time-varying magnetic fields to the beam to balance damping performance and electromagnet power consumption when the beam undergoes free-decay vibration. The model findings showed that while the DiTaranto model works better than the proposed models, this beam model tends to work best for stiffer outer layers and for electromagnets (assuming they do not span the beam's entire length) placed at the beam's middle or free end, while softer outer layers lead to improved damping performance. With respect to the free-decay analysis, time-varying magnetic fields can possess comparable or faster decay when compared to a constant magnetic field; turning the field on or off during free decay leads to slightly decreased damping performance but with less power consumption for stiffer beams, while softer beams can enjoy both reduced consumption and increased damping when using a damping method like synchronous switch damping.

This thesis presents the study of laminar to turbulent transition of particle laden flows. When a flow becomes turbulent, the drag increases one order of magnitude compared to a laminar flow, therefore, much research is devoted to understand and influence the transition. Previous research at the Linne Flow Centre at KTH has concentrated on the understanding of the bypass transition process of single-phase fluids. Though there are still questions, the principles of this process are now, more or less, known. However, little is known of the influence of particles on transition. While experiments in the 1960s already showed that particles can reduce the friction in turbulent channel flows significantly. The question explored in this thesis is whether this can be attributed to their influence on transition. The initial onset of transition has been investigated with both modal and non-modal linear stability analysis in a Poiseuille flow between two parallel plates. Particles are introduced as a second fluid and they are considered to be solid, spherical and homogeneously distributed. When the fluid density is much smaller than the particle density, ξ (≡ ρf/ρp) &lt;&lt; 1, an increase of the critical Reynolds number is observed. However, transient growth of streamwise vortices resulting in streaks is not affected by inclusion of particles. Particles with ξ ∼ 1 hardly seem to have an effect on stability. Although linear analysis shows that particles hardly influence the transient growth of disturbances, they might affect other (non-linear) stages of transition. To investigate such effects, the full Navier-Stokes equations for 3D Poiseuille flow between two parallel plates are numerically solved and particles are introduced as points with two-way coupling. For particles in a channel flow with ξ&lt;&lt;1, results show that the transition to turbulence is delayed for mass fractions ƒ (=mp N / ρf) larger than 0.1. For a mass fraction of ƒ=0.4 the initial disturbance energy needed to get a turbulent flow increases with a factor of four. Even if lower particle mass fractions ƒ are used, locally there could be large particle mass fractions. Therefore, the next step is to investigate the generation of local large particle mass fractions ƒ. Such particle clusters can be as large as the typical flow structures in the flow, like streak width and vortex size. Then they might change the flow field and (in)stability mechanisms. Numerical simulations of bypass transition in a boundary layer flow are used to determine whether particles cluster and where they tend to cluster. It is found that point particles with ξ&lt;&lt;1 and a large particle relaxation time tend to move in the low speed regions of the flow. In case of streaks, the low speed streaks are most favourable. For smaller particle relaxation times, particles act as tracers and do not have a preferential position and are homogeneously distributed. For particles with ξ∼1 the linear stability analysis showed no transition effect at any ƒ. However, one effect neglected until now is that of particle size. For particles with dimensions of the same order of magnitude of the flow disturbance, particles might influence the flow field. To investigate whether such particles migrate towards positions where they can affect transition some exploratory numerical simulations and experiments are performed. Numerically, the lateral migration of large particles (H/d=5) with ξ=1 in a 3D Poiseuille flow between two parallel plates is investigated. In laminar channel flow, large particles tend to move laterally due to shear to an equilibrium position. For a single large particle some key parameters for migration are identified: the size of the particle and the velocity of the fluid. When multiple particles are present, they tend to form particle trains. If particles are close, they influence each other and the equilibrium position shifts towards the wall, where the final position is dependent on the inter particle spacing. Also, not one steady equilibrium position is present, but particles move around an equilibrium position. Experimentally, migration of particles in bypass transition with ξ=1 is investigated to find out whether neutrally buoyant particles have a preferential position within streaks. The first results with tracer particles (d∼50μm) and few large particles (d∼200μm) do not show detectable preferential positioning. / <p>QC 20131030</p>

[pt] Confiança é um aspecto fundamental em sistemas distribuídos abertos de larga-escala. Ela está no núcleo de todas as interações entre as entidades que precisam operar em ambientes com muita incerteza e que se modificam constantemente. Dada essa complexidade, esses componentes, e o sistema resultante, são cada vez mais contextualizados, desenhados e construídos usando técnicas baseadas em agentes. Portanto, confiança é fundamental em um sistema multi-agentes (MAS) aberto. Logo, este trabalho investiga como se ter um modelo de confiança explicitamente em um agente inteligente, que possui crenças (Beliefs), desejos (Desires) e intenções (Intentions), chamado de agente BDI. Ou seja, o agente passa a ter um quarto componente chamado confiança (Trust). Dessa forma, é necessário uma lógica para englobar o conceito de confiança em um MAS BDI aberto. Isso é feito usando uma lógica multi-modal indexada, onde os mundos possíveis que modelam um sistema multi-agentes representam quais agentes estão presentes em um dado instante de tempo. E, para cada uma três componentes originais de um agente BDI, há também uma representação de mundos possíveis, pois as mesmas são tratadas como modalidades. Já a confiança é modelada como sendo um predicado, e não uma modalidade. / [en] Trust is a fundamental concern in large-escale open distributed sytems. It lies at the core of all interactios between the entities that have to operate in such uncertain and constantly changing environmonts. Given the complexity of the interactions, these components, and the ensuing system, are increasingly being conceptualised, desined, and built using agent-based techiques. Therefore, the presence of trust is imperative in a multi-agent system (MAS). Consequently, this work studies how to have a explicit trust model in intelligent agent, which has beliefs, desires and intentions (BDI agent). Thas is, the agent now has a fourth component called Trust. This way, a logic to include the concept of trust in an open BDI MAS is interesting, so that the different aspects of a trust model can be expressed formally and accuratelly. This is achieved by using an indexed multi-modal logic, where the possible worlds which model a multi-agent system represent which agents are in the system in a given moment. Moreover, for each one of the three original components of a BDI agent, where the components represent beliefs, desires and intentions, there is a representation of possible worlds, because these are treated as modalities. However, trust is modelled as predicate, not as a modality.