Massive infrastructure networks such as electric power, natural gas, or water systems play a pivotal role in everyday human lives. Development and operation of these networks is extremely capital-intensive. Moreover, security and reliability of these networks is critical. This work identifies and addresses a diverse class of computationally challenging and time-critical problems pertaining to these networks. This dissertation extends the state of the art on three fronts. First, general proofs of uniqueness for network flow problems are presented, thus addressing open problems. Efficient network flow solvers based on energy function minimizations, convex relaxations, and mixed-integer programming are proposed with performance guarantees. Second, a novel approach is developed for sample-efficient training of deep neural networks (DNN) aimed at solving optimal network dispatch problems. The novel feature here is that the DNNs are trained to match not only the minimizers, but also their sensitivities with respect to the optimization problem parameters. Third, control mechanisms are designed that ensure resilient and stable network operation. These novel solutions are bolstered by mathematical guarantees and extensive simulations on benchmark power, water, and natural gas networks. / Doctor of Philosophy / Massive infrastructure networks play a pivotal role in everyday human lives. A minor service disruption occurring locally in electric power, natural gas, or water networks is considered a significant loss. Uncertain demands, equipment failures, regulatory stipulations, and most importantly complicated physical laws render managing these networks an arduous task. Oftentimes, the first principle mathematical models for these networks are well known. Nevertheless, the computations needed in real-time to make spontaneous decisions frequently surpass the available resources. Explicitly identifying such problems, this dissertation extends the state of the art on three fronts: First, efficient models enabling the operators to tractably solve some routinely encountered problems are developed using fundamental and diverse mathematical tools; Second, quickly trainable machine learning based solutions are developed that enable spontaneous decision making while learning offline from sophisticated mathematical programs; and Third, control mechanisms are designed that ensure a safe and autonomous network operation without human intervention. These novel solutions are bolstered by mathematical guarantees and extensive simulations on benchmark power, water, and natural gas networks.

Because of the inherent complexity of fiber-reinforced laminated composites, it can be challenging to manufacture composite structures according to their exact design specifications, resulting in unwanted material and geometric uncertainties. Thus the understanding of the effect of uncertainties in laminated structures on their static and dynamic responses is highly important for a reliable design of such structures. In this research, we focus on the deterministic and probabilistic stability analysis of laminated structures subject to subtangential loading, a combination of conservative and nonconservative tangential loads, using the dynamic criterion. Thus a shear-deformable laminated beam element, including warping effects, is derived to study the deterministic and probabilistic response of laminated beams. This twenty-one degrees of freedom element can be used for solving both static and dynamic problems. In the first-order shear deformable model used here we have employed a more accurate method to obtain the transverse shear correction factor. The dynamic version of the principle of virtual work for laminated composites is expressed in its nondimensional form and the element tangent stiffness and mass matrices are obtained using analytical integration. The stability is studied by giving the structure a small disturbance about an equilibrium configuration, and observing if the resulting response remains small. In order to study the dynamic behavior by including uncertainties into the problem, three models were developed: Exact Monte Carlo Simulation, Sensitivity-Based Monte Carlo Simulation, and Probabilistic FEA. These methods were integrated into the developed finite element analysis. Also, perturbation and sensitivity analysis have been used to study nonconservative problems, as well as to study the stability analysis using the dynamic criterion. / Ph. D.

The precise and detailed characterization of the dynamic response of structures has become increasingly important in recent years. As a consequence, the accuracy of experimental data, which is often used to validate and update finite element models, has become extremely important. However, as researchers have attempted to identify and quantify sources of error in the experimental modal analysis (EMA) process, an important potential error source has been largely overlooked. Instabilities in the dynamic response of structures due to small variations in test environmental conditions may result in significant errors in experimental and analytical results, leading to erroneous and/or misleading conclusions. This thesis presents an experimental and analytical investigation of the effects of ambient temperature variations on the dynamic characteristics of a thin, square steel plate. The modal properties of the plate with two different boundary conditions and at temperatures above and below standard room temperature are examined. In addition, an analytical model is developed accounting for the effects of temperature-dependent material properties. Results indicate that natural frequencies and damping are significantly affected by changes in temperature. In the case of the natural frequency variations, the temperature-dependence of Young's modulus is the dominant factor, but boundary condition effects may also be important. Also, FRF magnitudes at spectral lines close to the resonances are very sensitive to temperature. Finally, only minor variations in the plate response shapes are observed, although significant changes in the imaginary component of the velocity field are evident. / Master of Science

Millions of people sustain a mild traumatic brain injury (concussion) each year. While most clinical signs and symptoms resolve within 7-10 days for the majority of typical concussions, some gait and balance tasks have shown abnormalities lasting beyond the resolution of clinical symptoms. These abnormalities can persist after athletes have been medically cleared for competition, yet the implications of such changes are unclear. Most prior research has examined straight gait and standard measures of balance, yet there is a lack of knowledge regarding potential persistent effects on non-straight maneuvers or on indicators of motor control variability or complexity. To expand the knowledge of post-concussion locomotor and postural changes, this investigation examined the recovery of recently concussed athletes longitudinally, over the course of one year, in three domains: 1) path selection and body kinematics during turning gait, 2) non-linear local dynamic stability during straight gait, and 3) postural control complexity during quiet standing. Compared to matched health controls, concussed athletes exhibited significant and persistent differences in turning kinematics, local dynamic stability, and postural complexity over the initial six weeks following injury. These motor differences may increase the risk of injury to concussed athletes who are cleared to return to play. Given the persistent nature of these effects, future clinical tests may benefit from incorporating gait assessments before returning athletes to competition. Future research should prospectively and longitudinally monitor locomotor and postural control in conjunction with structural and functional changes within the brain to better understand the pathophysiology of concussions and potential rehabilitation strategies. / Ph. D.

abstract: Injuries and death associated with fall incidences pose a significant burden to society, both in terms of human suffering and economic losses. The main aim of this dissertation is to study approaches that can reduce the risk of falls. One major subset of falls is falls due to neurodegenerative disorders such as Parkinson’s disease (PD). Freezing of gait (FOG) is a major cause of falls in this population. Therefore, a new FOG detection method using wavelet transform technique employing optimal sampling window size, update time, and sensor placements for identification of FOG events is created and validated in this dissertation. Another approach to reduce the risk of falls in PD patients is to correctly diagnose PD motor subtypes. PD can be further divided into two subtypes based on clinical features: tremor dominant (TD), and postural instability and gait difficulty (PIGD). PIGD subtype can place PD patients at a higher risk for falls compared to TD patients and, they have worse postural control in comparison to TD patients. Accordingly, correctly diagnosing subtypes can help caregivers to initiate early amenable interventions to reduce the risk of falls in PIGD patients. As such, a method using the standing center-of-pressure time series data has been developed to identify PD motor subtypes in this dissertation. Finally, an intervention method to improve dynamic stability was tested and validated. Unexpected perturbation-based training (PBT) is an intervention method which has shown promising results in regard to improving balance and reducing falls. Although PBT has shown promising results, the efficacy of such interventions is not well understood and evaluated. In other words, there is paucity of data revealing the effects of PBT on improving dynamic stability of walking and flexible gait adaptability. Therefore, the effects of three types of perturbation methods on improving dynamics stability was assessed. Treadmill delivered translational perturbations training improved dynamic stability, and adaptability of locomotor system in resisting perturbations while walking. / Dissertation/Thesis / Doctoral Dissertation Biomedical Engineering 2019

Falling during walking leads to millions of emergency room visits every year for all age groups and is a significant medical concern. While gait training has shown some promise for fall prevention, we know relatively little about how humans maintain stability, how we can quantify it and how we can use this knowledge to increase the success of fall prevention training. In this dissertation, I studied how human stability responds to continuous, small magnitude perturbations and to voluntary changes in gait characteristics by examining movement variability and long-term and instantaneous dynamic stability. In the first set of experiments, participants were exposed to continuous, pseudo-random external perturbations of the visual field and support surface in a Computer Assisted Rehabilitation ENvironment (CAREN). Participants exhibited increased step widths, shorter step lengths and increased step variability, orbital and short-term local instability. Despite this, mean instantaneous lateral stability remained approximately constant. In the second set of experiments, participants voluntarily adopted changes in their step widths and step lengths. Wider steps were associated with increased step width variability, decreased nonlinear stability, decreased anterior-posterior margins of stability and increased instantaneous lateral stability. Shorter steps were associated with decreased short-term and orbital stability but did not affect mean instantaneous stability. When instantaneous stability was examined between steps, as opposed to as an average over many steps, results from both studies indicated a relationship between each step’s stability and the stability of the immediately preceding step. From these studies, we now know that unpredictable, continuous perturbations during human walking applied in a given direction can be used to elicit predictable responses in motion variability and stability in that same direction. We know that the type of stability examined can influence the conclusions drawn about an individual’s stability during perturbed walking. For example, an individual’s variability may indicate increased risk of falling while he or she simultaneously demonstrates increased orbital stability and instantaneous lateral stability. A challenge faced in this area of research will be to understand how quantitative measures of stability relate to how we perceive our stability. / text

While the active dynamic stability of automobiles has increased over the past 17 years there have been very few similar advances made with electrically powered wheelchairs. This lack of improvement has led to a long standing acceptance of less-than-optimal stability and control of these wheelchairs. Accidents due to loss of stability are well documented. Hence, the healthcare industry has made several efforts for improved control of electric powered wheelchairs (EPWs) to provide enhanced comfort, safety and manoeuvrability at a lower cost. In response, an area of stability control was identified that could benefit from a feedback control system using solid state sensors. To design an effective closed–loop feedback controller with optimal performance to overcome instabilities, an accurate model of wheelchair dynamics needed to be created. Such a model can be employed to test various controllers quickly and repeatedly, without the difficulties of physically setting a wheelchair up for each test. This task was one central goal of this research. A wireless test-bed of a front wheel drive (FWD) wheelchair was also developed to validate a dynamic wheelchair model. It integrates sensors, a data control system, an embedded controller, and the motorised mechanical system. The wireless communication ensures the integrity of sensor data collected and control signals sent. The test-bed developed not only facilitates the development of feedback controllers of motorised wheelchairs, but the collected data can also be used to confirm theories of causes of dynamic instabilities. The prototype test-bed performed the required tasks to satisfaction as defined by the sponsor. Data collected from live tests in which the test-bed followed set patterns, was processed and analysed. The patterns were designed to induce instability. The analysis revealed that an occupied wheelchair is more stable than an unoccupied wheelchair, disproving an initial instability theory proposed in this research. However, a proximal theory explaining over-steer is confirmed. Two models of the FWD test-bed were created. First, a dynamic model inherited from prior research, based on equations of motion was tested and enhanced based on measured data. However, even with alterations to correct parameter values and variables in the equations, a complete model validation was not possible. Second, a kinematic model was created with a factor to compensate for dynamics not normally accounted in kinematic models. The kinematic model was partially validated versus the measured data. Although, still highly accurate, there is room for improvement in this model. Both models contained a sub-system drive motor model, to account for input forces to the FWD wheelchair system model, which is fully validated.

A three-dimensional, parallel, finite volume solver which uses Roe&#039 / s upwind flux differencing scheme for spatial and Runge-Kutta explicit multistage time stepping scheme for temporal discretization on unstructured meshes is developed for the unsteady solution of external viscous flow around rotating bodies. The main aim of this study is to evaluate the aerodynamic dynamic stability derivative coefficients for rotating missile configurations. Arbitrary Lagrangian Eulerian (ALE) formulation is adapted to the solver for the simulation of the rotation of the body. Eigenvalues of the Euler equations in ALE form has been derived. Body rotation is simply performed by rotating the entire computational domain including the body of the projectile by means of rotation matrices. Spalart-Allmaras one-euqation turbulence model is implemented to the solver. The solver developed is first verified in 3-D for inviscid flow over two missile configurations. Then inviscid flow over a rotating missile is tested. Viscous flux computation algorithms and Spalarat-Allmaras turbulence model implementation are validated in 2-D by performing calculations for viscous flow over flat plate, NACA0012 airfoil and NLR 7301 airfoil with trailing edge flap. The ALE formulation is validated in 2-D on a rapidly pitching NACA0012 airfoil. Afterwards three-dimensional validation studies for viscous, laminar and turbulent flow calculations are performed on 3-D flat plate problem. At last, as a validation test case, unsteady laminar and turbulent viscous flow calculations over a spinning M910 projectile configuration are performed. Results are qualitatively in agreement with the analytical solutions, experimental measurements and previous studies for steady and unsteady flow calculations.

abstract: In this work, the hydrodynamics of Suction Stabilization is studied. Suction stabilization was found to stabilize floating platforms/floats in a much better way as compared to the conventional methods. This was achieved by an effective increment in the metacentric height due to the Inverse Slack Tank (IST) effect. The study involves the analysis of the existing designs and optimizing its performance. This research investigates the stability of such floats and the hydrodynamic forces acting on the same for offshore applications, such as wind turbines. A simple mathematical model for the condition of parametric resonance is developed and the results are verified, both analytically and experimentally. / Dissertation/Thesis / Real time working of SSF in heavy wind and rain conditions / Animation explaining working of SSF / Experiment with patio umbrella mounted on SSF / Masters Thesis Mechanical Engineering 2014

L’introduction de fuselages et de pales de plus en plus légers durant le développement des nouveaux hélicoptères, combinée à une puissance disponible augmentée peut donner lieu à des couplages rotor/structure d’un nouveau genre. Ces instabilités complexes apparaissent à des fréquences plus élevées que les couplages connus et étudiés tels que les résonances sol et air, et impliquent des modes de pale souple, des modes de structure, et des phénomènes aérodynamiques. Des codes de calcul multi-corps aéromécaniques tels que HOST sont capables de déterminer la stabilité de l’hélicoptère, mais sont difficilement modifiables et manipulables. Des modèles analytiques existent également pour les instabilités maîtrisées citées précédemment, mais n’ont pas les capacités de modélisation nécessaires à la prédiction de ces couplages haute fréquence. Ce travail de thèse se concentre sur le développement d’un modèle semi-analytique, capable de prédire la stabilité de l’hélicoptère vis-à-vis de ces phénomènes. Cette approche est différente de l’approche multi-corps et a un double avantage car elle permet des études paramétriques rapides et une analyse terme à terme des équations de la dynamique de l’hélicoptère. Ce modèle a été validé à l’aide de HOST et le mécanisme de l'instabilité a été détaillé. Enfin, l’influence des paramètres de rotor, de structure, et de vol a été évaluée et les considérations architecturales pour éviter l'apparition de tels phénomènes sont présentées. / The introduction of lightweight fuselages and blades during new developments, combined with an increased available power, may lead to the triggering of a new kind of rotor/structure coupling. These complex instabilities appear at higher frequencies than known and studied couplings, such as ground and air resonance, and involve elastic blade modes, structure modes, and aerodynamic phenomena. Comprehensive analysis codes, like HOST, are able to determine the helicopter stability but can hardly be tweaked and handled. Rotor/structure coupling analytical models also exist for ground and air resonance, but do not have the modeling capabilities required to predict these high frequency couplings. This research work focuses on the development of a semi-analytical model, able to predict the helicopter stability with respect to these phenomena. This approach has a two-fold advantage since fast parametric studies can be carried out and a term-by-term analysis of the helicopter stability equations can be performed. This model has been validated with HOST and the triggering mechanism has been detailed. Finally, the influence of rotor, structure, and flight parameters has been evaluated and architectural considerations to avoid the appearance of such couplings are presented.