Downloads & Free Reading Options - Results

Assumed-displacement versions of the finite-element method are developed to predict large-deformation elastic-plastic transient deformations of structures. Both the conventional and a new improved finite-element variational formulation are derived. These formulations are then developed in detail for straight-beam and curved-beam elements undergoing (1) Bernoulli-Euler-Kirchhoff or (2) Timoshenko deformation behavior, in one plane. For each of these categories, several types of assumed-displacement finite elements are developed, and transient response predictions are compared with available exact solutions for small-deflection, linear-elastic transient responses. The present finite-element predictions for large-deflection elastic-plastic transient responses are evaluated via several beam and ring examples for which experimental measurements of transient strains and large transient deformations and independent finite-difference predictions are available.

The main goal of this project is to estimate theoretically the overall or effective constitutive properties of nonlinear composite materials undergoing large deformations. Two types of large deformations are of interest; large elastic deformations, and large viscous deformations. The proposed method is to apply variational principles that are under development to characterize the range of the effective properties given partial statistical information about the microstructure (such as the volume fractions of the phases). For some particular microstructures of interest exact estimates may be given. Significant progress was made over the first year with the development of a new variational principle allowing the estimation of the effective properties of a given nonlinear composite in terms of the effective properties of linear composites (which are assumed to be known). The potential significance of this work derives from its simplicity allowing the application of a large body of prior research on linear composites to nonlinear composites. The method has been applied to the case of large viscous deformations, and some results for particular materials systems have already been reported in the pertinent literature.

The main goal of this project is to estimate theoretically the overall or effective constitutive properties of nonlinear composite materials undergoing large deformations. Two types of large deformations are of interest: large elastic deformations, and large viscous deformations. The proposed method is to apply variational principles that are under development to characterize the range of the effective properties given partial statistical information about the microstructure (such as the volume fractions of the phases). For some particular microstructures of interest exact estimates may be given.

A circular twist disclination is a nontrivial example of a defect in an elastic continuum that causes large deformations. The minimal potential energy and the corresponding displacement field is calculated by solving the Euler-Lagrange-equations. The nonlinear incompressibility constraint is rigorously taken into account. By using an appropriate curvilinear coordinate system a finer resolution in the regions of large deformations is obtained and the dimension of the arising nonlinear PDE's is reduced to two. The extensive algebraic calculations that arise are done by a computer algebra system (CAS). The PDE's are then solved by a difference scheme using the Newton-Raphson algorithm of successive approximations for multidimensional equations. Additional features for global convergence are implemented. To obtain basic states that are sufficiently close to the solution, a one dimensional linearized version of the equation is solved with a numerical computation that reproduces the analytical results of Huang and Mura (1970). With this method, rigorous solutions of the nonlinear equations without any additional simplifications can be obtained. The numerical results show a contraction of the singularity line which corresponds to the well-known Poynting effect in nonlinear elasticity. This combination of analytical and numerical computations proves to be a versatile method to solve nonlinear boundary value problems in complicated geometries.

The development of a general mathematical model and solutions of test problems to analyze large nonisothermal elasto-visco-plastic deformatisms of structures is discussed. Geometric and material type nonlinearities of higher order are present in the development of the mathematical model and in the developed solution methodology.

The development of a general mathematical model and solutions of test problems to analyze large nonisothermal elasto-visco-plastic deformatisms of structures is discussed. Geometric and material type nonlinearities of higher order are present in the development of the mathematical model and in the developed solution methodology.

A user-oriented FORTRAN 4 computer program, called JET 3, is presented. The JET 3 program, which employs the spatial finite-element and timewise finite-difference method, can be used to predict the large two-dimensional elastic-plastic transient Kirchhoff-type deformations of a complete or partial structural ring, with various support conditions and restraints, subjected to a variety of initial velocity distributions and externally-applied transient forcing functions. The geometric shapes of the structural ring can be circular or arbitrarily curved and with variable thickness. Strain-hardening and strain-rate effects of the material are taken into account.

These structural ring deflections lie essentially in one plane and, hence, are called two-dimensional (2-d). The structural rings may be complete or partial; the former may be regarded as representing a fragment containment ring while the latter may be viewed as a 2-d fragment-deflector structure. These two types of rings may be either free or supported in various ways (pinned-fixed, locally clamped, elastic-foundation supported, mounting-bracket supported, etc.). The initial geometry of each ring may be circular or arbitrarily curved; uniform-thickness or variable-thickness rings may be analyzed. Strain-hardening and strain-rate effects of initially-isotropic material are taken into account. An approximate analysis utilizing kinetic energy and momentum conservation relations is used to predict the after-impact velocities of each fragment and of the impact-affected region of the ring; this procedure is termed the collision-imparted velocity method (CIVM) and is used in the CIVM-JET 5 B program. This imparted-velocity information is used in conjunction with a finite-element structural response computation code to predict the transient, large-deflection, elastic-plastic responses of the ring. Similarly, the equations of motion of each fragment are solved in small steps in time. Provisions are made in the CIVM-JET 5B code to analyze structural ring response to impact attack by from 1 to 3 fragments, each with its own size, mass, translational velocity components, and rotational velocity. The effects of friction between each fragment and the impacted ring are included.