Designing and Building an Autonomous Surface Vehicle (ASV)
Our research group is currently building an Autonomous Surface Vehicle (ASV) from scratch and is looking for students to help with the design and fabrication of the craft. Currently a KCS hull form has been chosen for the ASV with a 1:75.5 scale ratio to result in a 3 meter model. This vehicle will be a self propelled vehicle with an onboard computer and sensor suite that will allow the vehicle to localize itself and navigate in an environment. The vehicle will have multiple uses including investigation of reinforcement learning based control for trajectory tracking, cooperative control with other unmanned vehicles, testing of various underactuated control algorithms and investigating the effect of maneuvering in waves.
SIMDYN - Time Domain Hydrodynamic Simulation Tool
SIMDYN is a time domain simulation program written in Fortran to simulate the nonlinear motions of a vessel at sea. It can handle the large amplitude of motions of vessel by using an Euler angle formulation and also computes the nonlinear force vector to calculate the nonlinear motions of a vessel in an irregular seas. In particular it includes both nonlinear Froude Krylov and nonlinear hydrostatic forces and moments and also includes a viscous roll damping model for ship shaped structures to help simulate nonlinear phenomena such as parametric roll.
While including the nonlinear Froude Krylov and nonlinear hydrostatics allows modelling strong nonlinear motions, the program is also capable of handling the traditional consistent second order analysis where the second order drift forces and moments are included to evaluate the performance of the mooring system. A mooring model based on the catenary equation is also included to allow for the analysis of moored offshore platforms.
A newer version of the SIMDYN is currently being developed as a Python web application.
Parametric Roll of Container Ships (Finished)
Parametric response is an instability which results in large responses even when the excitation is close to zero. This nonlinear phenomenon is primarily caused by the time variation of a system parameter such as stiffness (in waves) or inertia. Typically, for a long time parametric roll motion was believed to be a problem for fishing vessels in following waves. However, in the recent years it has come to light that even fast container ships (which have a slender form to achieve higher operational speeds) are highly susceptible to parametric roll in head seas. This instability manifests when the incident regular wave has an encounter frequency which is twice the roll natural frequency. The video below demonstrates this instability of C11 (modified APL China hull form) simulated using SIMDYN.
The vulnerability of a vessel to parametric excitation is currently determined either through time domain simulations (as seen in the above video) or through experimental model tests. Both these approaches are time consuming and expensive. For a designer to consider vulnerability to parametric response during the design process an efficient but reliable appraoch is needed to assess the vulnerability. My research explores the development of simplified approaches (analytical or semi-analytical methods) to assess vulnerability of vessels to parametric response in an efficient manner. These approaches come under the Level 2 criteria for dynamic stability set forth by International Maritime Organization (IMO). My research has led to the development of two assessment criteria:
Rate of phase space flux that stems from a nonlinear dynamical systems appraoch
Mean first passage time to capsize that stems from a stochastic dynamics approach
More information regarding these criteria can be obtained from the relevant publications listed below.
Somayajula, A., & Falzarano, J. (2018). Volterra GZ approach – a new method to accurately calculate the non-linear and time-varying roll restoring arm of ships in irregular longitudinal seas. Ships and Offshore Structures, 13(4), 423–431.
Somayajula, A., Falzarano, J., & Lutes, L. (2019). An efficient assessment of vulnerability of a ship to parametric roll in irregular seas using first passage statistics. Probabilistic Engineering Mechanics, 58, 102998.
System Identification of Roll Damping (Finished)
While we can model most of the dynamics governing the motions of a vessel in a seaway, there are cetain phenomena that cannot be modelled easily. Ship roll damping is one such phenomenon that has been studied for long by researchers. Although simplified empirical models have been proposed, the actual damping differs significantly from the reality. One way to overcome this problem is to determine the damping coefficients from expeirmental roll motion time series. This approach of identifying parameters from input-output time histories is known as system identification. In this research a specific appraoch known as Reverse-Multiple Input Single Output (R-MISO) is used to predict the linear and nonlinear damping coefficients from time series data. The advantage of R-MISO is that it not only helps determine the unknown coefficients but also calculates the relative importance of various inputs in governing the dynamics of the system.
More information regarding this method and its application to roll damping problem can be obtained from the relevant publications listed below.
Somayajula, A., & Falzarano, J. (2017) Critical assessment of reverse-MISO techniques for system identification of coupled roll motion of ships. Journal of Marine Science and Technology, 22(2), 231-244
Somayajula, A., & Falzarano, J. (2017) Application of advanced system identification technique to extract roll damping from model tests in order to accurately predict roll motions. Applied Ocean Research, 67, 125-135.
Somayajula, A., & Falzarano, J. (2016) Estimation of roll motion parameters using R-MISO system identification technique. 26th International Offshore and Polar Engineering (ISOPE 2016) Conference, 568-574.
Wave Energy Converter (Finished)
In this project that was undertaken in collaboration with Texas A&M University, the motions of a wave energy converter were simulated using SIMDYN that was integrated with an external mooring module MAP++. Since a wave energy converter is designed to experience large motions to increase the power take off, the nonlinear effects become important. Similarly, large amplitude motions also means that there will be unmodelled dynamics that needs to be identified from the experiments to improve the numerical models. R-MISO based system identification is performed for this system to improve the pitch modeling of the simulation.
More information on this can be obtained from the relevant publications listed below.
Wang, H., Somayajula, A., Falzarano, J., & Xie, Z. (2020). Development of a blended time-domain program for predicting the motions of a wave energy structure. Journal of Marine Science and Engineering, 8(1), 1.