In recent years advances in technology have allowed the transition of composite structures from secondary to primary structural components. Consequently, a lot of applications demand development of thicker composite structures to sustain heavier loads. Typical sandwich panels consist of two thin metallic or composite face sheets separated by a honeycomb or foam core. This configuration gives the sandwich panel high stiffness and strength and enables excellent energy absorption capabilities with little resultant weight penalty. This makes sandwich structures a preferred design for a lot of applications including aerospace, naval, wind turbines and civil industries. Most aerospace structures can be analyzed using shell and plate models and many such structures are modeled as composite sandwich plates and shells. Accurate theoretical formulations that minimize the CPU time without penalties on the quality of the results are thus of fundamental importance. The classical plate theory (CPT) and the first order shear deformation theory (FSDT) are the simplest equivalent single-layer models, and they adequately describe the kinematic behavior of most laminates where the difference between the stiffnesses of the respective phases is not huge. However, in the case of sandwich structures where the core is a much more compliant and softer material as compared to the face sheets the results from CPT and FSDT becomes highly inaccurate. Higher order theories in such cases can represent the kinematics better, may not require shear correction factors, and can yield much more accurate results. An advanced Extended Higher-order Sandwich Panel Theory (EHSAPT) which is a two-dimensional extension of the EHSAPT beam model that Phan presented is developed. Phan had extended the HSAPT theory for beams that allows for the transverse shear distribution in the core to acquire the proper distribution as the core stiffness increases as a result of non-negligible in-plane stresses. The HSAPT model is incapable of capturing the in-plane stresses and assumes negligible in-plane rigidity. The current research extends that concept and applies it to two-dimensional plate structures with variable aspect ratios. The theory assumes a transverse displacement in the core that varies as a second order equation in z and the in-plane displacements that are of third order in z, the transverse coordinate. This approach allows for five generalized coordinates in the core (the in-plane and transverse displacements and two rotations about the x and y-axes respectively). The major assumptions of the theory are as follows: 1) The face sheets satisfy the Euler-Bernoulli assumptions, and their thicknesses are small compared to the overall thickness of the sandwich section; they undergo small strains with moderate rotations. 2) The core is compressible in the transverse and axial directions; it has in-plane, transverse and shear rigidities. 3) The bonding between the face sheets and the core is assumed to be perfect. The kinematic model is developed by assuming a displacement field for the soft core and then enforcing continuity of the displacement field across the interface between the core and facesheets. The constitutive relations are then defined, and variational and energy techniques are employed to develop the governing equations and associated boundary conditions. A static loading case for a simply supported sandwich plate is first considered, and the results are compared to existing solutions from Elasticity theory, Classical Plate Theory (CPT) and First-Order Shear Deformation Plate Theory (FSDT). Subsequently, the governing equations for a dynamic analysis are developed for a laminated sandwich plate. A free vibration problem is analyzed for a simply supported laminated sandwich plate, and the results for the fundamental natural frequency are compared to benchmark elasticity solutions provided by Noor. After validation of the new Extended Higher Order Sandwich Panel Theory (EHSAPT), a parametric study is carried out to analyze the effect of variation of various geometric and material properties on the fundamental natural frequency of the structure. After the necessary verification and validation of the theory by comparing static and free vibration results to elasticity solutions, a nonlinear static analysis for square and rectangular plates is carried out under various sets of boundary conditions. The analysis was carried out using variational techniques, and the Ritz method was used to find an approximate solution. The kinematics were developed for a sandwich plate undergoing small strain and moderate rotations and nonlinear strain displacement relations were evaluated. Approximate and assumed solutions satisfying the geometric boundary conditions were developed and substituted in the total potential energy relations. After carrying out the spatial integrations, the total potential energy was then minimized with respect to the unknown coefficients in the assumed solution resulting in nonlinear simultaneous algebraic equations for the unknown coefficients. The simultaneous nonlinear equations were then solved using the Newton-Raphson method. A convergence study was carried out to study the effect of varying the number of terms in the approximate solution on the overall result and rapid convergence was observed. The rapid convergence can be attributed to the fact that the assumed approximate solution not only satisfies the geometric boundary conditions of the problem but also the natural boundary conditions. During calculations four cases of boundary conditions were considered 1) Simply Supported with moveable edges. 2) Simply Supported with fixed edges. 3) Clamped with moveable edges. 4) Clamped with fixed edges. For movable boundary conditions, in-plane displacements along the normal direction to the supported edges are allowed whereas the out-of-plane displacement is fixed. For the immovable boundary condition cases, the plate is prevented from both in-plane and out-of-plane displacements along the edges. For the simply supported cases rotations about the tangential direction are allowed, and for the clamped cases no rotations are allowed.In recent years advances in technology have allowed the transition of composite structures from secondary to primary structural components. Consequently, a lot of applications demand development of thicker composite structures to sustain heavier loads. Typical sandwich panels consist of two thin metallic or composite face sheets separated by a honeycomb or foam core. This configuration gives the sandwich panel high stiffness and strength and enables excellent energy absorption capabilities with little resultant weight penalty. This makes sandwich structures a preferred design for a lot of applications including aerospace, naval, wind turbines and civil industries. Most aerospace structures can be analyzed using shell and plate models and many such structures are modeled as composite sandwich plates and shells. Accurate theoretical formulations that minimize the CPU time without penalties on the quality of the results are thus of fundamental importance. The classical plate theory (CPT) and the first order shear deformation theory (FSDT) are the simplest equivalent single-layer models, and they adequately describe the kinematic behavior of most laminates where the difference between the stiffnesses of the respective phases is not huge. However, in the case of sandwich structures where the core is a much more compliant and softer material as compared to the face sheets the results from CPT and FSDT becomes highly inaccurate. Higher order theories in such cases can represent the kinematics better, may not require shear correction factors, and can yield much more accurate results. An advanced Extended Higher-order Sandwich Panel Theory (EHSAPT) which is a two-dimensional extension of the EHSAPT beam model that Phan presented is developed. Phan had extended the HSAPT theory for beams that allows for the transverse shear distribution in the core to acquire the proper distribution as the core stiffness increases as a result of non-negligible in-plane stresses. The HSAPT model is incapable of capturing the in-plane stresses and assumes negligible in-plane rigidity. The current research extends that concept and applies it to two-dimensional plate structures with variable aspect ratios. The theory assumes a transverse displacement in the core that varies as a second order equation in z and the in-plane displacements that are of third order in z, the transverse coordinate. This approach allows for five generalized coordinates in the core (the in-plane and transverse displacements and two rotations about the x and y-axes respectively). The major assumptions of the theory are as follows: 1) The face sheets satisfy the Euler-Bernoulli assumptions, and their thicknesses are small compared to the overall thickness of the sandwich section; they undergo small strains with moderate rotations. 2) The core is compressible in the transverse and axial directions; it has in-plane, transverse and shear rigidities. 3) The bonding between the face sheets and the core is assumed to be perfect. The kinematic model is developed by assuming a displacement field for the soft core and then enforcing continuity of the displacement field across the interface between the core and facesheets. The constitutive relations are then defined, and variational and energy techniques are employed to develop the governing equations and associated boundary conditions. A static loading case for a simply supported sandwich plate is first considered, and the results are compared to existing solutions from Elasticity theory, Classical Plate Theory (CPT) and First-Order Shear Deformation Plate Theory (FSDT). Subsequently, the governing equations for a dynamic analysis are developed for a laminated sandwich plate. A free vibration problem is analyzed for a simply supported laminated sandwich plate, and the results for the fundamental natural frequency are compared to benchmark elasticity solutions provided by Noor. After validation of the new Extended Higher Order Sandwich Panel Theory (EHSAPT), a parametric study is carried out to analyze the effect of variation of various geometric and material properties on the fundamental natural frequency of the structure. After the necessary verification and validation of the theory by comparing static and free vibration results to elasticity solutions, a nonlinear static analysis for square and rectangular plates is carried out under various sets of boundary conditions. The analysis was carried out using variational techniques, and the Ritz method was used to find an approximate solution. The kinematics were developed for a sandwich plate undergoing small strain and moderate rotations and nonlinear strain displacement relations were evaluated. Approximate and assumed solutions satisfying the geometric boundary conditions were developed and substituted in the total potential energy relations. After carrying out the spatial integrations, the total potential energy was then minimized with respect to the unknown coefficients in the assumed solution resulting in nonlinear simultaneous algebraic equations for the unknown coefficients. The simultaneous nonlinear equations were then solved using the Newton-Raphson method. A convergence study was carried out to study the effect of varying the number of terms in the approximate solution on the overall result and rapid convergence was observed. The rapid convergence can be attributed to the fact that the assumed approximate solution not only satisfies the geometric boundary conditions of the problem but also the natural boundary conditions. During calculations four cases of boundary conditions were considered 1) Simply Supported with moveable edges. 2) Simply Supported with fixed edges. 3) Clamped with moveable edges. 4) Clamped with fixed edges. For movable boundary conditions, in-plane displacements along the normal direction to the supported edges are allowed whereas the out-of-plane displacement is fixed. For the immovable boundary condition cases, the plate is prevented from both in-plane and out-of-plane displacements along the edges. For the simply supported cases rotations about the tangential direction are allowed, and for the clamped cases no rotations are allowed.

In aerospace applications, use of laminates made of composite materials as face sheets in sandwich panels are on the rise. These composite laminates have low transverse shear and transverse normal moduli compared to the in-plane moduli. It is also seen that the corresponding transverse strength values are very low compared to the in-plane strength leading to delaminations. Further, in sandwich structures, the core is subjected to significant transverse shear stresses. Therefore the interlaminar stresses (i.e., transverse shear and normal) can govern the design of sandwich structures. As a consequence, the first step in achieving efficient designs is to develop the ability to reliably estimate interlaminar stresses. Stress analysis of the composite sandwich structures can be carried out using 3-D finite elements for each layer. Owing to the enormous computational time and resource requirements for such a model, this process of analysis is rendered inefficient. On the other hand, existing plate/shell finite elements, when appropriately chosen, can help quickly predict the 2-D displacements with reasonable accuracy. However, their ability to calculate the thickness-wise distributions of interlaminar shear and normal stresses and 3-D displacements remains as a research goal. Frequently, incremental refinements are offered over existing solutions. In this scenario, an asymptotically correct dimensional reduction from 3-D to 2-D, if possible, would serve to benchmark any ongoing research. The employment of a mathematical technique called the Variational Asymptotic Method (VAM) ensures the asymptotical correctness for this purpose. In plates and sandwich structures, it is typically possible to identify (purely from the defined material distributions and geometry) certain parameters as small compared to others. These characteristics are invoked by VAM to derive an asymptotically correct theory. Hence, the 3-D problem of plates is automatically decomposed into two separate problems (namely 1-D+2-D), which then exchange relevant information between each other in both ways. The through-the-thickness analysis of the plate, which is a 1-D analysis, provides asymptotic closed form solutions for the 2-D stiffness as well as the recovery relations (3-D warping field and displacements in terms of standard plate variables). This is followed by a 2-D plate analysis using the results of the 1-D analysis. Finally, the recovery relations regenerate all the required 3-D results. Thus, this method of developing reduced models involves neither ad hoc kinematic assumptions nor any need for shear correction factors as post-processing or curve-fitting measures. The results are most general and can be made as accurate as desired, while the procedure is computationally efficient. In the present work, an asymptotically correct plate theory is formulated for composite sandwich structures. In developing this theory, in addition to the small parameters (such as small strains, small thickness-to-wavelength ratios etc.,) pertaining to the general plate theory, additional small parameters characterizing (and specific to) sandwich structures (viz., smallness of the thickness of facial layers com-pared to that of the core and smallness of elastic material stiffness of the core in relation to that of the facesheets) are used in the formulation. The present approach also satisfies the interlaminar displacement continuity and transverse equilibrium requirements as demanded by the exact 3-D formulation. Based on the derived theory, numerical codes are developed in-house. The results are obtained for a typical sandwich panel subjected to mechanical loading. The 3-D displacements, inter-laminar normal and shear stress distributions are obtained. The results are compared with 3-D elasticity solutions as well as with the results obtained using 3-D finite elements in MSC NASTRAN®. The results show good agreement in spite of the major reduction in computational effort. The formulation is then extended for thermo-elastic deformations of a sandwich panel. This thesis is organized chronologically in terms of the objectives accomplished during the current research. The thesis is organized into six chapters. A brief organization of the thesis is presented below. Chapter-1 briefly reviews the motivation for the stress analysis of sandwich structures with composite facesheets. It provides a literature survey on the stress analysis of composite laminates and sandwich plate structures. The drawbacks of the existing anlaytical approaches as opposed to that of the VAM are brought out. Finally, it concludes by listing the main contributions of this research. Chapter-2 is dedicated to an overview of the 3-D elasticity formulation of composite sandwich structures. It starts with the 3-D description of a material point on a structural plate in the undeformed and deformed configurations. Further, the development of the associated 3-D strain field is also described. It ends with the formulation of the potential energy of the sandwich plate structure. Chapter-3 develops the asymptotically correct theory for composite sandwich plate structure. The mathematical description of VAM and the procedure involved in developing the dimensionally reduciable structural models from 3-D elasticity functional is first described. The 1-D through-the-thickness analysis procedure followed in developing the 2-D plate model of the composite sandwich structure is then presented. Finally, the recovery relations (which are one of the important results from 1-D through-the-thickness analysis) to extract 3-D responses of the structure are obtained. The developed formulation is applied to various problems listed in chapter 4. The first section of this chapter presents the validation study of the present formulation with available 3-D elasticity solutions. Here, composite sandwich plates for various length to depth ratios are correlated with available 3-D elasticity solutions as given in [23]. Lastly, the distributions of 3-D strains, stresses and displacements along the thickness for various loadings of a typical sandwich plate structure are correlated with corresponding solutions using well established 3-D finite elements of MSC NASTRAN® commerical FE software. The developed and validated formulation of composite sandwich structure for mechanical loading is extended for thermo-elastic deformations. The first sections of this chapter describes the seamless inclusion of thermo-elastic strains into the 3-D elasticity formulation. This is followed by the 1-D through-the-thickness analysis in developing the 2-D plate model. Finally, it concludes with the validation of the present formulation for a very general thermal loading (having variation in all the three co-ordinate axes) by correlating the results from the present theory with that of the corresponding solutions of 3-D finite elements of MSC NASTRAN® FE commercial software. Chapter-6 summarises the conclusions of this thesis and recommendations for future work.

The deterioration of the nation's civil infrastructure has prompted the investigation of numerous solutions to offset the problem. Some of these solutions have come in the form of innovative materials for new construction, whereas others have considered rehabilitation techniques for repairing existing infrastructure. A relatively new system that appears capable of encompassing both of these solution methodologies is the Sandwich Plate System (SPS), a composite bridge deck system that can be used in both new construction or for rehabilitation applications. SPS consists of steel face plates bonded to a rigid polyurethane core; a typical bridge application utilizes SPS primarily as a bridge deck acting compositely with conventional support girders. As a result of this technology being relatively new to the bridge market, design methods have yet to be established. This research aims to close this gap by investigating some of the key design issues considered to be limiting factors in implementation of SPS. The key issues that will be studied include lateral load distribution, dynamic load allowance and deck design methodologies. With SPS being new to the market, there has only been a single bridge application, limiting the investigations of in-service behavior. The Shenley Bridge was tested under live load conditions to determine in-service behavior with an emphasis on lateral load distribution and dynamic load allowance. Both static and dynamic testing were conducted. Results from the testing allowed for the determination of lateral load distribution factors and dynamic load allowance of an in-service SPS bridge. These results also provided a means to validate a finite element modeling approach which would could as the foundation for the remaining investigations on lateral load distribution and dynamic load allowance. The limited population of SPS bridges required the use of analytical methods of analysis for this study. These analytical models included finite element models and a stiffened plate model. The models were intended to be simple, but capable of predicting global response such as lateral load distribution and dynamic load allowance. The finite element models are shown to provide accurate predictions of the global response, but the stiffened plate approach was not as accurate. A parametric investigation, using the finite element models, was initiated to determine if the lateral load distribution characteristics and vibration response of SPS varied significantly from conventional systems. Results from this study suggest that the behavior of SPS does differ somewhat from conventional systems, but the response can be accommodated with current AASHTO LRFD bridge design provisions as a result of their conservativeness. In addition to characterizing global response, a deck design approach was developed. In this approach the SPS deck was represented as a plate structure, which allowed for the consideration of the key design limit states within the AASHTO LRFD specification. Based on the plate analyses, it was concluded that the design of SPS decks is stiffness-controlled as limited by the AASHTO LRFD specification deflection limits for lightweight metal decks. These limits allowed for the development of a method for sizing SPS decks to satisfy stiffness requirements. / Ph. D.

The random wetlay process is used to make fiber-reinforced thermoplastic sheets that can be compression molded into composite panels at little cost. By utilizing these composite panels as the facesheets of honeycomb sandwich structures, it is possible to greatly increase the bending stiffness of the composite without adding significant weight. The random wetlay composite facesheets used in this research consisted of 25% E-glass fibers and 75% PET by weight. The thickness uniformity of the facesheets was difficult to control. The core of the sandwich structure was HexWeb&174; EM. Three low-cost adhesives were examined for secondarily bonding the facesheets to the core: polyurethane glue; epoxy paste; and 3M Scotch-Grip&174; plastic adhesive. The polyurethane glue mixed with Cab-O-Sil filler was easiest to apply and provided the largest flatwise tensile strength. Mathematical models were developed to predict the static behavior of sandwich beams and plates in bending. Three-point bend tests were performed on a sandwich beam in accordance with ASTM C 393. A sandwich plate simply supported along two opposite edges and free along the other two edges was subjected to a line-load using weights and a wiffle tree arrangement. An effective facesheet modulus and Poisson's ratio were found by comparing the measured displacements to the sandwich plate theory. The shadow moiré technique was used to visualize the displacement of the line-loaded sandwich plate. The overall shape of the displacement was very similar to the shape predicted by the sandwich plate theory. / Master of Science

The Sandwich Plate System (SPS) is created by bonding two steel plates together with an elastomer core that is injected into a cavity formed by the steel plates and perimeter bars. The result is a stiffer and lighter panel that can be used for plate-like structures such as bridge decks, stadium risers or ship decks. For more versatility, the effects of welding post-injection to the SPS panels were investigated. Three post-injection welded joints were tested to determine fatigue resistance and the effects of cyclic loading on the localized debonding of the heat affected zone at the post-injection welded joint of a SPS bridge deck. Seven panels containing one of three post-injection weld configurations were investigated. Each panel was fatigue tested to ten million cycles or until failure, by applying remote bending to the post-injection welded joint. Experimental deflections and strains were compared to finite element analyses. Fatigue-life predictions were made using code based S-N curves, and a relatively new mesh-insensitive structural stress method with a master S-N curve approach. The post-injection welded joint demonstrated good fatigue resistance to recommended AASHTO loading when shims were used under the middle support to offset the camber in the SPS panels. It was also found that stresses caused by draw down of the camber had an adverse affect on the post-injection welded joint and greatly reduced its fatigue resistance. / Master of Science

An innovative approach to possible construction or rehabilitation of bridge decks can be found in a bridge construction system called the Sandwich Plate System (SPS). The technology developed and patented by Intelligent Engineering Canada Limited in conjunction with an industry partner, Elastogran GmbH, a member of BASF, may be an effective alternative to traditional bridge rehabilitation techniques. Although the system's behavior has been studied the connection of the SPS deck to the supporting girders has not been investigated. Two types of connection are presented in this research. The use of a bent plate welded to the SPS deck and subsequently bolted to the supporting girder utilizing slip-critical connections has been utilized in the construction of a SPS bridge. A proposed SPS bridge system utilizes the top flange of the supporting girder welded directly to the SPS deck as the deck-to-girder connection. The fatigue performance of a deck-to-girder connection utilizing a bent plate welded to the deck and bolted to the supporting girder using slip-critical connections was tested in the Virginia Tech Materials and Structures Laboratory. The testing concluded that the fatigue performance of the welded and bolted bent plate connection was limited by the weld details and no slip occurred in the slip-critical connections. Finite element modeling of the two types of deck-to-girder connections was also used to determine influence of the connections on the local and global behavior of a SPS bridge system. A comparison of the different connection details showed that the connection utilizing the flange welded directly to the SPS deck significantly reduces the stresses at location of the welds in the connections, but the connection type has a limited influence on the global behavior of a SPS bridge. / Master of Science

Three series of tests were conducted on a sandwich plate bridge deck, which consisted of two steel plates and an elastomer core. The first series of testing was conducted by applying a static load on a full scale sandwich plate bridge deck panel. Local strains and deflections were measured to determine the panel's behavior under two loading conditions. Next, fatigue tests were performed on the longitudinal weld between two sandwich plate panels. Two connections were tested to 10 million cycles, one connection was tested to 5 million cycles, and one connection was tested to 100,000 cycles. The fatigue class of the weld was determined and an S-N curve was created for the longitudinal weld group. Finally, a series of experiments was performed on a half scale continuous bridge deck specimen. The maximum positive and negative flexural bending moments were calculated and the torsional properties were examined. Finite element models were created for every load case in a given test series to predict local strains and deflections. All finite element analyses were preformed by Intelligent Engineering, Ltd. A comparison of measured values and analytical values was preformed for each test series. Most measured values were within five to ten percent of the predicted values. Shear lag in the half scale bridge was studied, and an effective width to be used for design purposes was determined. The effective width of the half scale simple span sandwich plate bridge deck was determined to be the physical width. Finally, supplemental research is recommended and conclusions are drawn. / Master of Science

This thesis investigates the characteristics of a close proximity underwater explosion and its effect on a ship-like structure. Finite element model tests are conducted to verify and validate the propagation of a pressure wave generated by an underwater explosion through a fluid medium, and the transmission of the pressure wave in the fluid to a structure using the Multi-Material Arbitrary Lagrangian/Eulerian method. A one dimensional case modeling the detonation of a spherical TNT charge underwater is investigated. Three dimensional cases modeling the detonation of an underwater spherical TNT charge, and US Navy Blast Test cases modeling a shape charge and a circular steel plate, and a shape charge and a Sandwich Plate System (SPS) are also investigated. This thesis provides evidence that existing tools and methodologies have some capability for predicting early-time/close proximity underwater explosion effects, but are insufficient for analyses beyond the arrival of the initial shock wave. This thesis shows that a true infinite boundary condition, a modified Gruneisen equation of state near the charge, and the ability to capture shock without a very small element size is needed in order to provide a sufficient means for predicting early-time/close proximity underwater explosion effects beyond the arrival of the initial shock wave. / Master of Science

Shear links are used as fuse elements in lateral load resisting systems to provide ductility and dissipate seismic energy. These links have traditionally been employed in eccentrically braced frames, but have more recently been suggested for use in the innovative linked column frame system (LCF). Current design specifications for shear links require intermediate web stiffeners to provide out-of-plane web stability so ductility requirements can be achieved. This research focused on moving from discrete transverse web stiffening to continuously stiffened webs in built up shear links. Built up links were designed to yield in shear when subjected to severe cyclic loading, however the webs of the links were designed using two metal sheets joined by an elastic core. These composite "sandwich" webs allowed for an increase in web thickness (and inherent flexural rigidity) without increasing the shear strength of the links. Numerical and experimental investigations were conducted to assess the performance of composite sandwich links subjected to severe loading. Numerical results showed improved web behavior in sandwich links in which the core material was assigned an elastic modulus greater than 5000psi. Due to fabrication limitations, experimental specimens were fabricated with a core material elastic modulus of 1000psi. These specimens did not perform as well as unstiffened base case links in terms global hysteretic behavior or ductility.

The response of submerged structures subject to loading by underwater blast waves is governed by complex interactions between the moving or deforming structure and the surrounding fluid and these phenomena need to be thoroughly understood in order to design structural components against underwater blast. This thesis has addressed the response of simple structural systems to blast loading in shallow or deep water environment. Analytical models have been developed to examine the one-dimensional response of both water-backed and air-backed submerged rigid plates, supported by linear springs and loaded by underwater shock waves. Cavitation phenomena as well as the effect of initial static fluid pressure are explicitly included in the models and their predictions were found in excellent agreement with detailed FE simulations. Then, a novel experimental apparatus has been developed, to reproduce controlled blast loading in initially pressurised liquids. It consists of a transparent water shock tube and allows observing the structural response as well as the propagation of cavitation fronts initiated by fluid-structure interaction in a blast event. This experimental technique was then employed to explore the one-dimensional response of monolithic plates, sandwich panels and double-walled structures subject to loading by underwater shock waves. The performed experiments provide great visual insight into the cavitation process and the experimental measurements were found to be in good agreement with analytical predictions and dynamic FE results. Finally, underwater blast loading of circular elastic plates has been investigated by theoretically modelling the main phenomena of dynamic plate deformation and fluid-structure interaction. In addition, underwater shock experiments have been performed on circular composite plates and the obtained measurements were found in good correlation with the corresponding analytical predictions. The validated analytical models were then used to determine the optimal designs of circular elastic plates which maximise the resistance to underwater blast.