View non-flash version
July 2013 www.sname.org/sname/mt ( abstracts )SNAME Paper Abstracts Physics-Based Learning Models for Ship Hydrodynamics BY GABRIEL D. WEYMOUTH AND DICK K.P. YUE PUBLISHED IN THE MARCH JOURNAL OF SHIP RESEARCHWe present the concepts of physics-based learning models (PBLM) and their relevance and application to the eld of ship hydrody- namics. e utility of physics-based learning is motivated by contrasting generic learning models for regression predictions, which do not presume any knowledge of the system other than the training data provided with methods such as semi-empirical models, which incor- porate physical insights along with data-tting. PBLM provides a framework wherein interme- diate models, which capture some physical aspects of the problem, are incorporated into modern generic learning tools to substantially improve the predictions of the latter, mini- mizing the reliance on costly experimental measurements or high-resolution high-delity numerical solutions. To illustrate the versatility and ecacy of PBLM, we present three wave- ship interaction problems: at speed waterline proles; ship motions in head seas; and three- dimensional breaking bow waves. PBLM is shown to be robust and produce error rates at or below the uncertainty in the generated data at a small fraction of the expense of high-reso- lution numerical predictions. Uncertainty Analysis of Load Combination Factors for Global Longitudinal Bending Moments of Double-hull Tankers BY ANGELO P. TEIXEIRA, C. GUEDES SOARES, NIAN-ZHONG CHEN, AND GE WANG PUBLISHED IN THE MARCH JOURNAL OF SHIP RESEARCHis article aims at assessing the probabilistic characteristics of the load combination fac- tors for global longitudinal bending moments of double-hull tankers. e calculations are performed based on a sample of oil tankers representative of the range of application of the Association of Classication Societies (IACS)-Common Structural Rules (CSR) design rules. e article starts by reviewing the probabilistic models that have been pro- posed to model still water and wave-induced loads and their characteristic extreme val- ues. Different load combination methods are also reviewed, including an analytical method that provides the combined charac- teristic value of still water and wave-induced bending moments based on the Poisson assumption for upcrossing events and using the rst-order reliability method in combi- nation with the point-crossing method. e predictions of the dierent load combina- tion methods are assessed on the basis of a sample of ve oil tankers adopted during the IACS-CSR design rules development process. A parametric and an uncertainty propaga- tion study are then performed to identify the range of variation and the probabilistic mod- els of the load combination factors that are applicable to double-hull tankers. Comparisons Between Prediction and Experiment for Lift Force and Heel Moment for a Planing Hull BY CAROLYN Q. JUDGE PUBLISHED IN THE FEBRUARY JOURNAL OF SHIP PRODUCTION AND DESIGN Even in calm water, high-speed vessels can display unstable behaviors such as chine walking, sudden large heel, and porpoising. Large heel results from the loss of transverse stability at high forward speed. When a plan- ing craft begins to plane, the hydrodynamic lift forces raise the hull out of the water. e available righting moment resulting from the hydrostatic buoyancy is, therefore, reduced. As the righting moment resulting from hydrostatic buoyancy is reduced, the righting moment resulting from dynamic effects becomes important. These hydro- dynamic righting effects are related to the hydrodynamic lift. is article explores the relationship between the hydrostatic lift and righting moment, the hydrodynamic lift and righting moment, and the total lift and heel- restoring moment of a planing craft operating at planing speeds. A series of tow tests using a prismatic hull with a constant deadrise of 20° measured the lift force and righting moment at various angles of heel and at various model velocities. e model was completely con- strained in surge, sway, heave, roll, pitch, and yaw. e underwater volume is deter- mined from the known hull conguration and the underwater photography of the keel and chine wetted lengths. e results presented include the total lift and righting moment with the hydrostatic and hydrodynamic con- tributions for various model speeds at two model displacements. Ride Severity Index: A Simpli?ed Approach for Comparing Peak Acceleration Responses of High- Speed Craft BY MICHAEL R. RILEY, TIM COATS, KELLY HAUPT, AND DONALD JACOBSON PUBLISHED IN THE FEBRUARY JOURNAL OF SHIP PRODUCTION AND DESIGN is article presents a simplied approach to quantifying the comparison of accelera- tion responses of high-speed craft in rough seas. Statistical acceleration values, used to characterize craft seakeeping responses, including average of the highest one-third, one-tenth, and 1/100th peak accelerations and the root mean square acceleration, are used to dene the relative ride sever- ity index (RSI). e article rst summarizes an unambiguous computational procedure for multiple investigators to calculate simi- lar acceleration values. It then explains the theory and rationale for relating statisti- cal acceleration ratios to an indication of potential damage, whether resulting from cumulative wave impacts or single severe slam events, that can be used in compar- ative assessments of structural integrity, equipment susceptibility to malfunction, or personnel comfort and safety. Example ride Editors note: The following are abstracts of papers recently presented at SNAME events and/or published in SNAME publications. The papers can be found at www. sname.org/sname/mt/featuredabstracts View current and previous issues of Journal of Ship Production and Design and Journal of Ship Research at http://www.sname.org/SNAME/Pubs/Journals1/