博士 / 國立臺灣大學 / 工程科學與海洋工程學系 / 92 / Abstract
The prediction of damage for submerged structures subjected to an underwater explosion (UNDEX) is an important topic for naval ship designers. Most studies on the dynamic responses and failure modes of ship structures subjected to UNDEX are experimental. An UNDEX experiment is generally very expensive, and the available information is frequently limited by budget or the performance of instruments used. The application of a simplified formula to complex structures may be difficult. Recently, owing to the rapid developments of hardware and software, numerical methods have provided accurate and effective predicting ability to improve the strength and reduce the scantling of structures of both surface vessels and submarines.
This study begins in collecting references involving UNDEX phenomenon and the non-linear properties of materials. By means of verifying experimental data and numerical results at the same time, this study hope to finish two goals: at first, to establish the measurement technique to gain useful data of submerged structures to UNDEX loading; secondly, to verify an accurate and effective process to predict the dynamic response of structures to UNDEX loading.
In this study, all experiments were conducted in a water tank. A highly sensitive combined charge, which consists of a DP60 detonator and a Detasheet, was selected to enable a small quantity of charge to react fully when it explodes in the small water tank. At first this work focus on the verification of the intensity of peak pressure of the primary shock wave. The equivalent quantity of this charge to TNT and the pressure regressed formula was determined. The pressure regressed formula is very useful in the subsequent numerical analysis when the standoff distance is very small.
A plate panel is a common structural component of ships. The experiments involving an UNDEX attack on the air-backed 6061-T6 aluminum square plate with size with different standoff distances were performed. A small quantity of explosive was used, so that the response of the plate panel remained within the elastic range, and no permanent deformation occurred. The time histories of pressure, acceleration, and strain near plate center were measured. A numerical analysis of dynamical responses for the same plate panel to UNDEX loadings using USA/DYNA codes was performed. Due to the mechanical property of the pressure transducer, if the measured pressure time-history was applied as the shock loading in numerical analysis, the peak pressure has to be modified properly instead of original measured data. The tendency and amplitude of numerical results and the experimental data are in good agreement. The peak value of dynamic responses predicated by simplified formula is acceptable compared with both measured data and numerical results.
In this paper, the elasto-plastic dynamic response of the plate panel subjected to underwater shock loading has been numerically examined. The dynamic responses of the target plate are examined under different level of shock factors, which cover elastic and plastic deformation ranges. The effect of strain rate on the dynamic response during the material nonlinear phase is taken into consideration. The numerical results are compared with experimental data accomplished by the specified reference. The advanced study of the reasonable finite element models is prompted and indicates that the permanent deformation was result from both the primary shock and the subsequent bubble impulse when the plastic strain has occurred after the pulsation of the primary shock. It existed some significant errors in the setting of the boundary conditions and the interactions between fluid and structures of the numerical analysis accomplished by the reference paper.
This study examined the dynamic response of a 6061-T6 aluminum cylinder subjected to a top-on UNDEX shock wave with two different standoff distances. The cylinder model is not internally stiffened and has two thick endplates. The quantity of high sensitive explosive was selected to be sufficiently small, so the cylinder response stays within the elastic range and there is no permanent deformation if the larger standoff distance was selected. When a smaller standoff distance was used, two types of damage appeared in the test cylinder model, namely plastic deformation on the upper side of the cylinder, and obvious permanent damage at the weld line. The second type of damage should not occur in the perfect model. That is, manufactured structures generally have more or less imperfections. The main imperfections of the cylinder model discussed here result from the welding process, which causes changes in material properties and geometry discontinuity. The numerical analysis of the experiment model was performed using the USA-DYNA software, and the calculated results were compared to the experimental data. High-speed photography also was used to observe the deformation time-history of the cylinder model during the experiment; it indicated that the bubble impulse would expand permanent deformation at the locations existing plastic strains resulted from the primary shock.
At last, this study chose a 500 tons ship as a target to analyze the dynamic responses to at different keel shock factors conditions. The amplitude and decay tendency of dynamic responses of local structures on this ship were determined. In addition, this study prompted an effective method to predict and reduce the vibration level of the important equipments in this ship.
| Identifer | oai:union.ndltd.org:TW/092NTU00345019 |
| Date | January 2004 |
| Creators | Hsu Pei-Yu, 徐培譽 |
| Contributors | Hung Chen-Far, Kong Chin-Hwa, 洪振發, 孔慶華 |
| Source Sets | National Digital Library of Theses and Dissertations in Taiwan |
| Language | zh-TW |
| Detected Language | English |
| Type | 學位論文 ; thesis |
| Format | 160 |
Page generated in 0.0018 seconds