No description available.

No description available.

No description available.

No description available.

[pt] Neste trabalho são apresentados dois reflectômetros ópticos por contagem de fótons no domínio do tempo para o monitoramento de fibras ópticas. O primeiro foi projetado para obter faixas dinâmicas altas. Demonstrou-se a sua capacidade de sintonização no monitoramento de redes passivas WDM-PON durante o tráfego de dados. 32 dB de faixa dinâmica com 6 m de resolução foram atingidos. O segundo reflectômetro foi projetado para atingir resoluções ultra altas. As aplicações neste caso, além do monitoramento de uma rede TDM-PON de curto alcance, foram na caracterização de redes de Bragg dispersivas e na descrição e modelagem de um fenômeno não reportado antes na literatura, chamado nesta tese de reflexão por curvatura. Foi demonstrada uma resolução menor que 3 cm com faixa dinâmica maior que 14.0 dB. / [en] This thesis presents the development of two photon counting optical time domain reflectometers for fiber optic links monitoring. The first one was focused on high dynamic range. It is demonstrated its tunable capability for a WDM-PON in-service monitoring. 32 dB on dynamic range and a two-point resolution of 6 m is achieved. The second reflectometer was design to accomplish an ultra high resolution. The monitoring of a short TDM-PON is performed. Moreover, due to its high resolution, a chirped fiber Bragg grating is characterized and a non previous reported phenomena, the bend reflection, is shaped and described. It is demonstrated 3 cm two-point resolution and more than 14 dB on dynamic range.

As the light source, semiconductor laser diodes play an important role in the fiber-optic communication systems. The main function of a laser diode is to convert signals from the electrical domain to the optical carriers so that they can be transmitted through an optical fiber. Modeling and simulation of directly modulated laser diodes are necessary for understanding and prediction of their performance in fiber-optical communication links. The alternatives based on a comprehensive experimental evaluation are normally costly and time consuming. This is particularly true for systems running at high bit-rate such as the 10Gb/s transmission systems that are used in tele and data communication applications. This thesis presents a modeling and simulation study for directly modulated laser diodes for high-speed fiber-optical communication systems. The work is based on the conventional rate equation model used as the governing equation for the simulation of the behavior of semiconductor lasers. In modeling of the system performance, each device is treated as a symbolic node that takes input signal and generates output signal all in time domain. For the semiconductor lasers, the original signals in electrical domain are taken as the input while the modulated lights in optical domain are as the output. The rate equations then link the output to the input. For any given time domain signal input, the modulated light (power and wavelength) as the output is calculated through the solutions of the rate equations. In seeking for the solution to the rate equations, we utilized a numerical approach to solve the rate equations which are a system of coupled nonlinear ordinary differential equations where analytical solution does not generally exist. In this work, a comprehensive study on the behavior of semiconductor lasers has been performed through static and dynamic analyses of the rate equations. The noise characteristic is also examined as it may become a major concern in some applications for the noise of the directly modulated laser transmitter may cause degradation to the signals and therefore lead to system penalty. Further, the numerical models and simulators developed for semiconductor lasers are incorporated into a general simulation platform on which similar models and simulators for other optoelectronic and optical components are connected to form a system-level simulator for point-to-point multiple channel fiber-optical communication links. This platform is capable of handling different system configurations with different component selection options. It simulates the time domain waveform in any point along the signal transmission path following a strict data-flow approach; i.e., the simulation is performed sample-by-sample on “real time” rather than frame-by-frame at “flush” mode. Finally, the simulation results, both on the device level and on the system level, have been compared with the experimental data and the results from other models in literature and found qualitative agreement. / Thesis / Master of Applied Science (MASc)

A procedure for computational dosimetry to verify safety standards compliance of mobile communications base stations is presented. Compared with the traditional power density method, a procedure based on more rigorous physics was devised, requiring computation or measurement of the specific absorption rate (SAR) within the biological tissue of a person at an arbitrary distance. This uses a hybrid methd of moments/finite difference time domain (MoM/FDTD) numerical method in order to determine the field or SAR distribution in complex penetrable media, without the computational penalties that would result from a wholly FDTD simulation. It is shown that the transmitted power allowed by the more precise SAR method is, in many cases, between two and five times greater than that allowed by standards implementing the power flux density method.

<p>The numerical simulation of physical systems plays a key role in different fields of science and engineering. The popularity of numerical methods stems from their ability to simulate complex physical phenomena for which analytical solutions are only possible for limited combinations of geometry, boundary, and initial conditions. Despite their flexibility, the computational demand of classical numerical methods quickly escalates as the size and complexity of the model increase. To address this limitation, and motivated by the unprecedented success of Deep Learning (DL) in computer vision, researchers started exploring the possibility of developing computationally efficient DL-based algorithms to simulate the response of complex systems. To date, DL techniques have been shown to be effective in simulating certain physical systems. However, their practical application faces an important common constraint: trained DL models are limited to a predefined set of configurations. Any change to the system configuration (e.g., changes to the domain size or boundary conditions) entails updating the underlying architecture and retraining the model. It follows that existing DL-based simulation approaches lack the flexibility offered by classical numerical methods. An important constraint that severely hinders the widespread application of these approaches to the simulation of physical systems.</p> <p><br></p> <p>In an effort to address this limitation, this dissertation explores DL models capable of combining the conceptual flexibility typical of a numerical approach for structural analysis, the finite element method, with the remarkable computational efficiency of trained neural networks. Specifically, this dissertation introduces the novel concept of <em>“Finite Element Network Analysis”</em> (FENA), a physics-informed, DL-based computational framework for the simulation of physical systems. FENA leverages the unique transfer knowledge property of bidirectional recurrent neural networks to provide a uniquely powerful and flexible computing platform. In FENA, each class of physical systems (for example, structural elements such as beams and plates) is represented by a set of surrogate DL-based models. All classes of surrogate models are pre-trained and available in a library, analogous to the finite element method, alleviating the need for repeated retraining. Another remarkable characteristic of FENA is the ability to simulate assemblies built by combining pre-trained networks that serve as surrogate models of different components of physical systems, a functionality that is key to modeling multicomponent physical systems. The ability to assemble pre-trained network models, dubbed <em>network concatenation</em>, places FENA in a new category of DL-based computational platforms because, unlike existing DL-based techniques, FENA does not require <em>ad hoc</em> training for problem-specific conditions.</p> <p><br></p> <p>While FENA is highly general in nature, this work focuses primarily on the development of linear and nonlinear static simulation capabilities of a variety of fundamental structural elements as a benchmark to demonstrate FENA's capabilities. Specifically, FENA is applied to linear elastic rods, slender beams, and thin plates. Then, the concept of concatenation is utilized to simulate multicomponent structures composed of beams and plate assemblies (stiffened panels). The capacity of FENA to model nonlinear systems is also shown by further applying it to nonlinear problems consisting in the simulation of geometrically nonlinear elastic beams and plastic deformation of aluminum beams, an extension that became possible thanks to the flexibility of FENA and the intrinsic nonlinearity of neural networks. The application of FENA to time-transient simulations is also presented, providing the foundation for linear time-transient simulations of homogeneous and inhomogeneous systems. Specifically, the concepts of Super Finite Network Element (SFNE) and network concatenation in time are introduced. The proposed concepts enable training SFNEs based on data available in a limited time frame and then using the trained SFNEs to simulate the system evolution beyond the initial time window characteristic of the training dataset. To showcase the effectiveness and versatility of the introduced concepts, they are applied to the transient simulation of homogeneous rods and inhomogeneous beams. In each case, the framework is validated by direct comparison against the solutions available from analytical methods or traditional finite element analysis. Results indicate that FENA can provide highly accurate solutions, with relative errors below 2 % for the cases presented in this work and a clear computational advantage over traditional numerical solution methods. </p> <p><br></p> <p>The consistency of the performance across diverse problem settings substantiates the adaptability and versatility of FENA. It is expected that, although the framework is illustrated and numerically validated only for selected classes of structures, the framework could potentially be extended to a broad spectrum of structural and multiphysics applications relevant to computational science.</p>

The need for water conservation continues to increase as global freshwater resources dwindle. In response, many golf course superintendents are implementing new methods and tools to become more frugal with their water applications. For example, scheduling irrigation using time-domain reflectometer (TDR) soil moisture sensors can decrease water usage. Still, TDR measurements are time-consuming and only cover small scales, leading to many locations being unsampled. Remotely sensed data such as the normalized difference vegetation index (NDVI) offer the potential of estimating moisture stress across larger scales; however, NDVI measurements are influenced by numerous stressors beyond moisture availability, thus limiting its reliability for irrigation decisions. An alternative vegetation index, the water band index (WBI), is primarily influenced by water absorption within a narrow spectral range of near-infrared light. Previous research has established strong relationships between moisture stress of creeping bentgrass (CBG) grown on sand-based root zones, a typical scenario for golf course putting greens. However, this relationship characterizes only a small portion of total acreage across golf courses, which limits widespread adoption. In our research, '007' CBG and 'Latitude 36'hybrid bermudagrass (HBG) were grown on three soil textures, USGA 90:10 sand (S), sand loam (SL) and clay (C), arranged in a 2 x 3 factorial design, randomized within six individual dry-down cycles serving as replications. Canopy reflectance and volumetric water content (VWC) data were collected hourly between 0700 and 1900 hr using a hyperspectral radiometer and an embedded soil moisture sensor, until complete turf necrosis. The WBI had the strongest relationship to VWC (r = 0.62) and visual estimations of wilt (r = -0.91) compared to the green-to-red ratio index (GRI) or NDVI. Parameters associated with non-linear regression were analyzed to compare grasses, soils, indices, and their interactions. The WBI and GRI compared favorably with each other and indicated significant moisture stress approximately 28 hr earlier than NDVI (P = 0.0010). WBI and GRI respectively predicted moisture stress 12 to 9 hr before visual estimation of 50% wilt, whereas NDVI provided 2 hr of prediction time (P = 0.0317). When considering the time to significant moisture stress, the HBG lasted 28 hr longer than CBG, while S lasted 42 hr longer than either SL and C (P ≤ 0.0011). Nonlinear regression analysis showed that WBI and GRI can be useful for predicting moisture stress of CBG and HBG grown on three diverse soils in a highly controlled environment. Our results provide substantial evidence and direction for future research investigating how WBI and GRI can expedite moisture stress assessment and prediction on a large-acreage basis. / Master of Science in Life Sciences / Managed turfgrasses provide several benefits including filtering pollutants, cooling their surroundings, generating oxygen, preventing erosion, serving as recreational surfaces, and increasing landscape aesthetics. Intensively managed turfgrass systems, such as on golf courses and sports fields, require more inputs to maintain acceptable conditions. Freshwater use is often excessive on intensively managed turfgrasses to maintain proper plant growth. Drought conditions often limit water availability, especially in regions with limited rainfall. Turf managers tend to over-apply water across large acreage when few localized areas begin to show symptoms of drought. Additionally, turf managers sometimes wrongly identify stressed areas from other factors as ones being moisture-deprived. Advancements such as the use of soil moisture meters have simplified irrigation decisions as an aid to visual inspections for drought stress. While this method enhances detection accuracy, it still provides no solution to increase efficiency. Expanding our current knowledge of turfgrass canopy light reflectance for rapid moisture stress identification can potentially save both time and water resources. The objective of this research was to enhance our ability to identify and predict moisture stress of creeping bentgrass (CBG) and hybrid bermudagrass (HBG) canopies integrated into varying soil textures (USGA 90:10 sand (S), sand loam (SL) and Clay (C)) using light reflectance measurements. Dry-down cycles were conducted under greenhouses conditions collecting soil moisture and light reflectance data every hour from 7 am to 7 pm after saturating and withholding water from established plugs. Moisture stress was most accurately estimated over time using two vegetation indices, the water band index (WBI) and green-to-red ratio index (GRI), with approximately ninety percent accuracy to visible wilt stress. The WBI and GRI predicted moisture stress of CBG in all soil types and HBG in SL and C approximately 14 hours before the grasses reached 50% wilt. While light reflectance varies on exposed soils, our research shows that underlying soils do not interfere with measurements across typical turfgrass stands. This research provides a foundation for future research implementing rapid, aerial measurements of moisture stressed turfgrasses on a broad application of CBG and HBG on constructed or native soils.

<p dir="ltr">The CFD-CAA method combines computational fluid dynamics (CFD) and computational aeroacoustics (CAA) techniques to analyze the interaction between fluid flow and the generation and propagation of sound. CFD is primarily concerned with simulating fluid flow patterns, while CAA focuses on predicting noise generation and its propagation in fluids. The CFD-CAA method provides a powerful tool for understanding and predicting the acoustic behavior of turbulent flows. By combining the strengths of CFD and CAA, this approach provides more precise and comprehensive analyses across various fields, thereby contributing to enhanced designs and noise control strategies.</p><p dir="ltr">Within industrial applications, a primary concern is noise source identification. This process enables engineers to locate and quantify the strength of noise sources within a system, facilitating the implementation of more effective strategies during the design process. A novel methodology, computational statistically optimized near-field acoustic holography (C-SONAH), is proposed to virtually identify aeroacoustic sources. Initially, sound pressure is obtained using the CFD-CAA method, followed by the application of the SONAH algorithm to locate acoustic sources and predict the sound field. C-SONAH offers computational advantages over direct CAA methods for simulating sound produced by systems with rotating elements, as CAA analyzes sources on the moving elements, making sound field calculation computationally expensive. The SONAH procedure converts these rotating sources into a series of equivalent stationary planar or cylindrical waves, reducing the number of sources and the time required to compute the sound field from each source. This methodology was demonstrated by characterizing the aerodynamic noise produced by a bladeless fan. The sound pressure level obtained by C-SONAH method was validated by the data predicted by the direct CFD-CAA method. Acoustic maps were reconstructed at different locations and frequencies, revealing that the C-SONAH method can predict noise sources generated by airflow and rotating components within the fan. Thus, it serves as an effective tool for understanding the aeroacoustic noise generation mechanism and guiding the design optimization of similar products.</p><p dir="ltr">The CFD-CAA method is also a powerful tool for design optimization. Computational simulations are typically less expensive and time-consuming than building and maintaining experimental setups, particularly for large or complex projects. Additionally, simulations reduce the need for multiple physical prototypes, which can shorten the development cycle. CFD-CAA simulations provide detailed flow and acoustic field data, including variables that may be difficult or impossible to measure experimentally, such as pressure distributions, velocity fields, and turbulent structures. In this dissertation, aeroacoustic characteristics and flow field information of vortex whistles were investigated using the CFD-CAA method. The simulation results clearly illustrate the swirling motion created in the vortex whistle cylinder and also demonstrate the linear frequency versus flow rate relationship characteristic of the whistle. The design of the vortex whistle was optimized based on the acoustic response and flow resistance by both simulations and experiments. The results suggest that the whistle with a thin inlet exhibits the best performance at high flow rates, while the whistle with a scale of 0.5 is the most sensitive to low flow rates, making it suitable for pediatric applications.</p><p dir="ltr">In CFD-CAA simulations, the time step typically cannot be too small due to limited computational resources. This constraint results in an aliasing error in spectral analysis. Consequently, an anti-aliasing operation prior to sampling is necessary to eliminate such errors from the acoustic source terms. In the present study, an anti-aliasing filter based on the compact finite difference formulation was designed within a time-domain, compact filter scheme. This filter was directly applied to the Navier-Stokes solver prior to sampling for CAA analysis. A cavity flow case was simulated to validate this mitigation strategy. The results indicate that the artificial spectral peak induced by aliasing error is removed without affecting other signature peaks. The anti-aliasing filter was also applied to more complex cases for predicting the acoustic field of a vortex whistle. The acoustic field around the vortex whistle, with both constant and variable inlet flow rates, was simulated, and the aliasing peak was successfully removed. Although the peak magnitudes decreased slightly due to the filter, the signature frequencies remained unchanged. Thus, the simulation with anti-aliasing operation can predict acoustic features without introducing aliasing errors, even if the time step is not sufficiently small, thereby significantly reducing simulation time.</p><p dir="ltr">In engineering applications, once noise sources are identified, the subsequent concern is noise reduction. An effective strategy for noise reduction involves acoustical absorbing materials to minimize noise emissions from components. Traditionally, experiments in engineering applications have focused on surface treatments to explore noise control techniques. However, the CFD-CAA method commonly assumes smooth and purely reflective wall surfaces. Consequently, there is growing interest in incorporating impedance boundary conditions into the CFD-CAA method. Since impedance boundary conditions are defined in the frequency domain, while CFD-CAA simulations operate in the time domain, direct implementation is not feasible. To address this issue, several methods have been proposed to define time-domain impedance boundary conditions in simulations. In the present study, a wall softness model was implemented in the CFD-CAA method and to examine a vortex whistle featuring an acoustically permeable surface. In simulations, an impedance boundary condition representing the properties of melamine foam was defined over the surface of a cylindrical cavity. The simulation results were validated against experimental data obtained from a vortex whistle with melamine foam. The findings revealed that the impedance of the melamine foam contributed to noise reduction at high frequencies. Additionally, at low airflow rates, the impedance boundary condition enhanced the signal-to-noise ratio for the low-frequency peak, which is advantageous in clinical applications.</p>