Dhanak, Manhar R.

A computational investigation of the hydrodynamic and seakeeping performance of a catamaran in calm, and in the presence of transforming head and following seas in waters of constant and varying depths is described. Parametric studies were conducted for a selected WAM-V 16 catamaran geometry using OpenFOAM® to uncover the physical phenomena. In the process a methodology has been developed for simulating the interactions between the vehicle and the shallow water environment akin to that in the coastal environment. The multiphase flow around the catamaran, including the six degrees-of-freedom motion of the vehicle, was modeled using a Volume of Fluid (VoF) method and solved using a dynamic mesh. The numerical approach was validated through computing benchmark cases and comparing the results with previous work. It is found that in a calm shallow water environment the total resistance, dynamic trim and sinkage of a catamaran in motion can be significantly impacted by the local water depth. The variations of the impact with depth and length-based Froude numbers are characterized. The impact varies as the vehicle moves from shallow waters to deep water or vice versa. In the presence of head and following small-amplitude seas, interesting interactions between incident waves and those generated by the vehicle are observed and are characterized for their variation with Froude number and water depth.

Modeling, system identification and controller design for a 16’ catamaran is
described with the objective of enhanced operation in the presence of environmental
disturbances including wind, waves and current. The vehicle is fully-actuated in surge,
sway and yaw degrees of freedom. Analytical and experimental system identification is
carried out to create a numerical model of the vehicle. A composite system of a Multiinput
multi-output Proportional-Derivative (PD) controller and a nonlinear disturbance
observer is used for station-keeping and transiting modes of operation. A waypoint
transiting algorithm is developed to output heading and cross-track error from vehicle
position and waypoints. A control allocation method is designed to lower azimuthing
frequency and incorporate angle saturation and rate limits. Validation is achieved with
improvement in simulation with the addition of the nonlinear observer.

Automatic target recognition capabilities in autonomous underwater vehicles has
been a daunting task, largely due to the noisy nature of sonar imagery and due to the lack
of publicly available sonar data. Machine learning techniques have made great strides in
tackling this feat, although not much research has been done regarding deep learning
techniques for side-scan sonar imagery. Here, a state-of-the-art deep learning object
detection method is adapted for side-scan sonar imagery, with results supporting a simple
yet robust method to detect objects/anomalies along the seabed. A systematic procedure
was employed in transfer learning a pre-trained convolutional neural network in order to
learn the pixel-intensity based features of seafloor anomalies in sonar images. Using this
process, newly trained convolutional neural network models were produced using
relatively small training datasets and tested to show reasonably accurate anomaly
detection and classification with little to no false alarms.

Boundary layer control on a circular cylindrical body through oscillating Lorentz
forcing is studied by means of numerical simulation of the vorticity-stream
function formulation of the Navier-Stokes equations. The model problem
considers axisymmetric seawater flow along an infinite cylinder controlled by an
idealized radially directed Lorentz force oscillating spatially and temporally.
Under optimum forcing parameters, it is shown that sustainable Lorentz induced
vortex rings can travel along the cylinder at a speed equivalent to the phase
speed of forcing . Wall stress is shown to locally change sign in the region
adjacent to the vortex, considerably decreasing net viscous drag . Adverse flow
behaviors are revealed as a result of studying the effects of the Reynolds
numbers, strength of the Lorentz force, and phase speed of forcing for boundary
layer control. Adverse flow behaviors inc I ude complex vortex configurations
found for suboptimal forcing resulting in a considerable increase in wall stress.

A microbubble generation system has been designed, constructed, and tested in a
circulating water tunnel. A 1.0 m long flat plate was subjected to a flow where the
Reynolds number ranged from ReL = 7.23x 10^5 - 1.04 x 10^6. Bubble diameters and skin
friction measurements were studied at various airflow rates and water velocities.
Bubbles were produced by forcing air through porous plates that were mounted
flush with the bottom of the test plate. Once emitted through the plates, the bubbles
traveled downstream in the boundary layer. The airflow rate and water velocity were
found to have the most significant impact on the size of the bubbles created.
Skin friction drag measurements were recorded in detail in the velocity and
airflow rate ranges. The coefficient of skin friction was determined and relationships
were then established between this coefficient and the void ratio.

The primary objective of this research is to investigate the viability of magnetic
anomaly localization with an autonomous underwater vehicle, using a genetic algorithm
(GA). The localization method, first proposed by Sheinker. et al. 2008, is optimized here
for the case of a moving platform. Extensive magnetic field modeling and algorithm
simulation has been conducted and yields promising results. Field testing of the method is
conducted with the use of the Ocean Floor Geophysics Self-Compensating Magnetometer
(SCM). Extensive out-of-water field testing is conducted to validate the ability to
measure a target signal in a uniform NED frame as well as to validate the effectiveness of
the GA. The outcome of the simulation closely matches the results of the conducted field
tests. Additionally, the SCM is fully integrated with FAU’s Remus 100 AUV and
preliminary in-water testing of the system has been conducted.

The Gulf Stream current in the Straits is typically dominated by a strong northerly current,
associated shear, and eddies. The water column also includes a prominent thermocline and
periodically features internal waves centered on the upper or lower edges of the thermocline.
Despite numerous previous related studies, there is limited available field data on internal waves
in the Straits of Florida. Here, study and analysis of velocity, temperature and conductivity data
acquired in the Straits over a period of time are described, in support of identifying presence of
internal waves in the flow. A systematic procedure is employed in modifying the universal Garrett-
Munk spectrum for internal waves in the open ocean for application to flow in the Straits of
Florida. Using this process, identified internal waves are characterized and related velocity
fluctuations in the time series are isolated to facilitate consideration of their correlations with
simultaneously observed magnetic fields.

In near-shore transforming seas, as waves approach the shoreline, wave shoaling
and sometimes wave breaking take place due to the decreasing water depth. When
a ship advances through the transforming seas, the ship body and waves interact with
each other substantially and can lead to unknown motions of the ship hull. The physical
process of how the wave transforms in the surf zone and how the vehicle actually
behaves when it passes through the transforming seas is a complicated issue that
triggers considerable research interest.
The goal of my research is to characterize the dynamics of a high-speed surface
ship model in transforming seas through a parametric numerical study of the shipwave
interactions. In this study, the vehicle of interest is a surface effect ship (SES)
and we aim to contribute to developing a methodology for simulating the transforming
wave environment, including wave breaking, and its interactions with the SES.
The thesis work uses a commercial software package ANSYS Fluent to generate
numerical waves and model the interface between water and air using the volume
of fluid (VoF) method. A ship motion solver and the dynamic mesh are used to
enable the modeled ship to perform three degree-of-freedom (DoF) motion and the
near-region of the ship hull to deform as well as re-mesh. Non-conformal meshes with hybrid compositions of different cell types and various grid sizes are used in the
simulations for different purposes. Five user-defined functions (UDFs) are dynamically
linked with the flow solver to incorporates ship/grid motions, wave damping
and output of the numerical results. A series of steps were taken sequentially: 1)
validation for ship motions including simulation of a static Wigley hull under steady
flows to compare against previous experimental results by other researchers, and the
comparison between the static SES model under steady flows and the moving SES
model advancing in the calm water; 2) study of the ship with 3 DoF advancing in
calm water of both constant depth and varying depth; 3) validation for numerical
waves, including predictions of numerically progressive waves in both a regular tank
and a tank with a sloped fringing reef to compare with theoretical and experimental
results, respectively; 4) investigation of the transforming characteristics of the wave
traveling over the sloped fringing reef, which mimics the near-shore wave environment
and a study of the dynamics of the SES through transforming waves.
We find that the flow solver used in this study reliably models the wave profiles
along the ship hull. The comparison between a static SES in a current and a moving
SES in calm water at the same Froude number shows that although the velocity fields
around the vehicle are significantly different, the wave profiles inside and outside the
rigid cushion of the vehicle are similar and the resistance force experienced by the
vehicle in the two scenarios agree well over time. We conducted five numerical simulations
of the vehicle traveling from shallow water to deep water across the transition
zone for different Froude numbers. From the results, we find that as the Froude number
increases, the wave resistance force on the vehicle becomes larger in both shallow
water and deep water. In addition, the overall mean resistance force experienced by
the vehicle over the whole trip increases with the Froude number. Statistical analysis
of the wave motions suggests that the energy flux decreases dramatically in the
onshore direction as the waves break. The more severe the wave-breaking process, the greater the decrease in energy flux. Both the increase of Froude number and the
wave steepness apparently increase the resistance force on the vehicle in the shallow
water.
This thesis work captures the impact of the transforming characteristics of
the waves and closely replicates the behavior of how waves interact with a ship in
transforming seas through numerical modeling and simulation.

An Ocean Current Turbine (OCT) numerical simulation for creating, testing and
tuning flight and power takeoff controllers, as well as for farm layout optimization is
presented. This simulation utilizes a novel approach for analytically describing oceanic
turbulence. This approach has been integrated into a previously developed turbine
simulation that uses unsteady Blade Element Momentum theory. Using this, the
dynamical response and power production of a single OCT operating in ambient
turbulence is quantified.
An approach for integrating wake effects into this single device numerical
simulation is presented for predicting OCT performance within a farm. To accomplish
this, far wake characteristics behind a turbine are numerically described using analytic
expressions derived from wind turbine wake models. These expressions are tuned to
match OCT wake characteristics calculated from CFD analyses and experimental data. Turbine wake is characterized in terms of increased turbulence intensities and decreased
mean wake velocities. These parameters are calculated based on the performance of the
upstream OCT and integrated into the environmental models used by downstream OCT.
Simulation results are presented that quantify the effects of wakes on downstream turbine
performance over a wide range of relative downstream and cross stream locations for
both moored and bottom mounted turbine systems. This is done to enable the
development and testing of flight and power takeoff controllers designed for maximizing
energy production and reduce turbine loadings.