Thesis. 1979. Ph.D.--Massachusetts Institute of Technology. Alfred P. Sloan School of Management. / MICROFICHE COPY AVAILABLE IN ARCHIVES AND DEWEY. / Bibliography: leaves 141-150. / Ph.D.

Structural identification is the inverse problem of estimating the physical parameters, e.g. element masses and stiffnesses, of a model representing a structural system, using response measurements obtained from the actual structure subjected to operational or well-defined experimental excitations. It is one of the principal focal areas of modal testing and structural health monitoring, with the identified model finding a wide variety of applications, from obtaining reliable response predictions to timely detection of structural damage (location and severity) and consequent planning and validating of maintenance/retrofitting operations. However, incomplete instrumentation of the monitored system and ambient vibration testing generally result in spatially incomplete and arbitrarily normalized measured modal information, often making the inverse problem ill-conditioned and resulting in non-unique identification results. The problem of parameter identifiability addresses the question of whether or not a parameter set of interest can be identified from the available information. The identifiability of any parameter set of interest depends on the number and location of sensors on the monitored system. In this dissertation we study the identifiability of the mass and stiffness parameters of shear-type systems, including 3-dimensional laterally-torsionally coupled rigid floor systems, with incomplete instrumentation, simultaneous to the development of algorithms to identify the complete mass and stiffness matrices of such systems. Both input-output and output-only situations are considered, and mode shape expansion and mass normalization approaches are developed to obtain the complete mass normalized mode shape matrix, starting from the incomplete modal parameters identified using any suitable experimental or operational modal analysis technique. Methods are discussed to decide actuator/sensor locations on the structure which will ensure identifiability of the mass and stiffness parameters. Several possible minimal and near-minimal instrumentation set-ups are also identified. The minimal a priori information necessary in output-only situations is determined, and different scenario of available a priori information are considered. Additionally, tests for identifiability are discussed for both pre- and post-experiment applications. The different theoretical discussions are illustrated using numerical simulations and experimental data. It is shown that the proposed identification algorithms are able to obtain reliably accurate physical parameter estimates even under the constraints of minimal instrumentation, minimal a priori information, and unmeasured input. The different actuator/sensor placement rules and identifiability tests are useful for both experiment design purposes, to determine the necessary number and location of sensors, as well as in identifying possibilities of multiple solutions post-experiment. The parameter identification methods are applied for structural health monitoring using experimental data, and an approach is discussed for probabilistic characterization of structural damage location and severity. A perturbation based uncertainty propagation approach is also discussed for the identification of the distributions of mass and stiffness parameters, reflecting the variability in the test structure, using very limited measured and a priori information.

The reliability of the main cable of a suspension bridge is crucial to the reliability of the entire bridge. Throughout the life of a suspension bridge, its main cables are subject to corrosion due to various factors, and the deterioration of strength is a slowly evolving and dynamic process. The goal of this research is to find the pattern of how the strength of steel wires inside a suspension bridge cable changes with time. Two methodologies are proposed based on the analysis of five data sets which were collected by testing pristine wires, artificially corroded wires, and wires taken from three suspension bridges: Severn Bridge, Forth Road Bridge and Williamsburg Bridge. The first methodology is to model wire strength as a random process in space whose marginal probability distribution and power spectral density evolve with time. Both the marginal distribution and the power spectral density are parameterized with time-dependent parameters. This enables the use of Monte Carlo methods to estimate the failure probability of wires at any given time. An often encountered problem -- the incompatibility between the non-Gaussian marginal probability distribution and prescribed power spectral density -- which arises when simulating non-Gaussian random processes using translational field theory, is also studied. It is shown by copula theory that the selected marginal distribution imposes restrictions on the selection of power spectral density function. The second methodology is to model the deterioration rate of wire strength as a stochastic process in time, under Ito's stochastic calculus framework. The deterioration rate process is identified as a mean-reversion stochastic process taking non-negative values. It is proposed that the actual deterioration of wire strength depends on the deterioration rate, and may also depend on the state of the wire strength itself. The probability distribution of wire strength at any given time can be obtained by integrating the deterioration rate process. The model parameters are calibrated from the available data sets by matching moments or minimizing differences between probability distributions.

Each year storms impact coastal areas, sometimes causing significant morphologic change. Cold fronts are associated with increased wave energy and frequently occur during the winter months along many coasts, such as the Atlantic and Gulf of Mexico. The higher wave energy can be responsible for a large quantity of the sediment transport resulting in rapid morphologic change. Using streamer traps, the vertical distribution of onshore-directed sediment transport during two different cold fronts on two low-wave energy beaches (i.e., along the northern Yucatan and southeast Florida) were compared with the resulting morphologic change. The objectives of this study are to: 1) analyze the grain size distribution (statistics) of sediment transported during a cold front, 2) compare the vertical sediment distribution throughout the water column, and 3) compare characteristics of bed sediment to the sediment within the water column. Understanding the changing grain size distribution of bottom sediments in comparison to directional transport (throughout the water column) should help determine the sediment fraction(s) being eroded or deposited, which could greatly improve predictions of storm-induced morphology change. / Includes bibliography. / Thesis (M.S.)--Florida Atlantic University, 2017. / FAU Electronic Theses and Dissertations Collection

A hypoid gear optimization procedure using the method of feasible directions has been developed. The objective is to reduce the gear set weight with bending strength, contact strength and facewidth-diametral pitch ratio as constraints. The objective function weight, is calculated from the geometric approximation of the volume of the gear and pinion. The design variables selected are number of gear teeth, diametral pitch, and facewidth. The input parameters for starting the initial design phase are power to be transmitted, speed, gear ratio, type of application, mounting condition, type of loading, and the material to be used. In the initial design phase, design parameters are selected or calculated using the standard available procedures. These selected values of design parameters are passed on to the optimization routine as starting points.

In numerical simulations of geomechanics problems, a grand challenge consists of overcoming the difficulties in making accurate and robust predictions by revealing the true mechanisms in particle interactions, fluid flow inside pore spaces, and hydromechanical coupling effect between the solid and fluid constituents, from microscale to mesoscale, and to macroscale. While simulation tools incorporating subscale physics can provide detailed insights and accurate material properties to macroscale simulations via computational homogenizations, these numerical simulations are often too computational demanding to be directly used across multiple scales. Recent breakthroughs of Artificial Intelligence (AI) via machine learning have great potential to overcome these barriers, as evidenced by their great success in many applications such as image recognition, natural language processing, and strategy exploration in games. The AI can achieve super-human performance level in a large number of applications, and accomplish tasks that were thought to be not feasible due to the limitations of human and previous computer algorithms. Yet, machine learning approaches can also suffer from overfitting, lack of interpretability, and lack of reliability. Thus the application of machine learning into generation of accurate and reliable surrogate constitutive models for geomaterials with multiscale and multiphysics is not trivial. For this purpose, we propose to establish an integrated modeling process for automatic designing, training, validating, and falsifying of constitutive models, or "metamodeling". This dissertation focuses on our efforts in laying down step-by-step the necessary theoretical and technical foundations for the multiscale metamodeling framework. The first step is to develop multiscale hydromechanical homogenization frameworks for both bulk granular materials and granular interfaces, with their behaviors homogenized from subscale microstructural simulations. For efficient simulations of field-scale geomechanics problems across more than two scales, we develop a hybrid data-driven method designed to capture the multiscale hydro-mechanical coupling effect of porous media with pores of various different sizes. By using sub-scale simulations to generate database to train material models, an offline homogenization procedure is used to replace the up-scaling procedure to generate path-dependent cohesive laws for localized physical discontinuities at both grain and specimen scales. To enable AI in taking over the trial-and-error tasks in the constitutive modeling process, we introduce a novel “metamodeling” framework that employs both graph theory and deep reinforcement learning (DRL) to generate accurate, physics compatible and interpretable surrogate machine learning models. The process of writing constitutive models is simplified as a sequence of forming graph edges with the goal of maximizing the model score (a function of accuracy, robustness and forward prediction quality). By using neural networks to estimate policies and state values, the computer agent is able to efficiently self-improve the constitutive models generated through self-playing. To overcome the obstacle of limited information in geomechanics, we improve the efficiency in utilization of experimental data by a multi-agent cooperative metamodeling framework to provide guidance on database generation and constitutive modeling at the same time. The modeler agent in the framework focuses on evaluating all modeling options (from domain experts’ knowledge or machine learning) in a directed multigraph of elasto-plasticity theory, and finding the optimal path that links the source of the directed graph (e.g., strain history) to the target (e.g., stress). Meanwhile, the data agent focuses on collecting data from real or virtual experiments, interacts with the modeler agent sequentially and generates the database for model calibration to optimize the prediction accuracy. Finally, we design a non-cooperative meta-modeling framework that focuses on automatically developing strategies that simultaneously generate experimental data to calibrate model parameters and explore weakness of a known constitutive model until the strengths and weaknesses of the constitutive law on the application range can be identified through competition. These tasks are enabled by a zero-sum reward system of the metamodeling game and robust adversarial reinforcement learning techniques.

Three-dimensional scour propagation along offshore pipelines is a major reason to pipeline failures in an offshore environment. Although the research on scour in both numerical and experimental aspect has been extensive over the last three decades, the focus of the investigation has been limited to the two-dimensional aspect. The knowledge on three-dimensional scour is still limited. This dissertation presents the results of an experimental investigation on threedimensional scour along offshore pipelines in (1) steady currents (2) waves only and (3) combined waves and current. The major emphasis of the investigation is to investigate the propagation of the scour hole along the pipeline after the initiation of scour. Physical experiments conducted were used to quantify the effects of various parameters on scour propagation velocities along the pipeline. The problem of monitoring real time scour below a pipeline was solved by using specifically developed conductivity scour probes. Effects of various parameters such as pipeline embedment depth, incoming flow Shields parameter, Keuglegan- Carpenter (KC) number and flow incident angle to the pipeline on scour propagation velocities along the pipeline were investigated. The investigations clearly reveal that scour propagation velocities generally increase with the increase of flow but decrease with the increase of the pipeline embedment depth. A general predictive formula for scour propagation velocities is proposed and validated against the experimental results. There are still some common issues related to pipeline scour that is lacking in the literature to date. One of these issues is the effects of Reynolds number on two-dimensional scour beneath pipelines. A numerical approach was adopted to investigate the Reynolds-number dependence of two-dimensional scour beneath offshore pipelines in steady currents. A novel wall function is proposed in calculating the suspended sediment transport rate in the model. The effects of Reynolds number were investigated by simulating the same undisturbed Shields parameters in both model and prototype but with different values of Reynolds number in two separate calculations. The results revealed that scour depths for prototype pipelines are about 10~15% smaller than those for model pipelines. The normalized time scales was found to be approximately the same, and the simulated scour profiles for the model pipelines agree well with the experimental results from an independent study. The backfilling of pipeline trenches is also an important issue to the design and management of offshore pipelines. A numerical model is developed to simulate the self-burial of a pipeline trench. Morphological evolutions of a pipeline trench under steady-current or oscillatory-flow conditions are simulated with/without a pipeline inside the trench. The two-dimensional Reynolds-averaged continuity and Navier-Stokes equations with the standard k-e turbulence closure, as well as the sediment transport equations, are solved using finite difference method in a curvilinear coordinate system. Different time-marching schemes are employed for the morphological computation under unidirectional and oscillatory conditions. It is found that vortex motions within the trench play an important role in the trench development.

Abstract not available

Design of commercial aircraft structure, composed of composite material, requires the prediction of failure loads given large scale damage. In particular, a fuselage of graphite/epoxy lamination was analyzed for damage tolerance given a standard large crack that severed both skin and internal structure. Upon loading, a zone of damage is known to develop in front of a crack-tip in composite laminates; and, its material behavior within the damage zone is characterized as strain softening. This investigation sought to develop a computational model that simulates progressive damage growth and predicts failure of complex laminated shell structures subject to combined tensile and flexural load conditions. This was accomplished by assuming a macroscopic definition of orthotropic damage that is allowed to vary linearly through the shell thickness. It was further proposed that nonlocal plate strain and curvature act to force damage growth according to a set of uniaxial criteria. Damage induced strain softening is exhibited by degradation of laminate stiffness. An expression for the damage reduced laminated plate stiffness was derived which assumed the familiar laminated plate [AM] stiffness matrix format. The model was implemented in a finite element shell program for simulation of fracture and evaluation of damage tolerance. Laminates were characterized for damage resistance according to material parameters defining nonlocal strain and the damage growth criteria. These parameters were selected using an inverse method to correlate simulation with uniaxial strength and fracture test results. A novel combined tension-plus-flexure fracture test was developed to facilitate this effort. Analysis was performed on a section of pressurized composite fuselage containing a large crack. Good agreement was found between calculations and test results. / Graduation date: 1998

by H.-N. Tan and S.B. Gershwin. / Bibliography: leaves 13-15. / "February, 1979." Caption title. / Supported by the U.S. Dept. of Transportation under Contract DOT-TSC-1456