1,721,085 research outputs found
Dynamic shakedown analysis and bounds in non-associative poroplasticity with saturation hardening
No full text
Stima dell'errore per l'integrazione temporale di tipo 'backward-difference' in plasticità computazionale, Una
No full text
Adaptive backward-difference time-stepping procedure for finite element hardening elastoplasticity
No full textStatistical approach to damage diagnosis of concrete dams by radar monitoring: formulation and a pseudo-experimental test
No full textThis paper deals with the investigation of a diagnostic procedure especially suited for large concrete dams. The main features of such methodology are as follows: static excitation of the dam; displacement monitoring by radar device; identification of the unknown Young moduli in pre-defined homogeneous zones through a batch least square method. The uncertainty of the identified parameters is assessed by means of a thorough statistical processing. The numerical validation of the proposed method is carried out on the basis of pseudo-experimental data. The most important results, summarized in the paper together with some computational remarks, are critically examined
Finite-friction least-thickness self-standing domains of symmetric circular masonry arches. Part II: Milankovitch-like self-weight distribution
Full text linkThe present paper constitutes a sort of compendium to a recent analysis considering the role of finite (Coulomb) friction in the mechanics of (symmetric circular) masonry arches, towards addressing least-thickness self-standing states and collapse modes possibly including sliding. Thereby, a classical Heyman-like uniform self-weight distribution along geometrical centreline was considered, and all characteristic features delivered, by a comprehensive analytical approach, corroborated by consistent outcomes from a self-made numerical Complementarity Problem/Mathematical Programming implementation. The same is now updated, for the real Milankovitch-like uniform self-weight distribution, considering the true positions of the centres of gravity of the ideal wedged-shaped chunks of the arch, with radial stereotomy. The main achieved result is that the specific conceived distribution does not alter, conceptually, the salient recorded characteristic features, in terms of types of self-standing/collapse states, though some differences are displayed, in the details of the final analytical-numerical findings, with few physical implications. However, the main implant, put to light from the previous, simpler to be mathematically handled, classical self-weight distribution, is confirmed, showing definite reference, in methodological terms, discovered features and first-order amount of technical results. In general sense, the present true Milankovitch-like self-weight distribution leads to more precise results, which support further bearing capacity for the real arch, thus anyway with conservative estimates arising from the previous approximate Heyman-like self-weight distribution, quite simpler in the underlying mechanical and mathematical treatment
A rigorous bound on error in backward-difference elastoplastic time-integration
No full textThe finite element analysis of elastoplastic structures requires in general a time-stepping procedure and, in most cases, the integration of the constitutive law within each time-step has to be carried out by numerical integration. The error associated to this numerical integration depends on the degree of non-linearity of the structural response and can be used as an indicator for the adaptive definition of the time-step size. Based on Martin’s and Ortiz theorem on minimum total work, a simple estimate of the integration error associated to a backward-difference scheme for elastoplastic models is derived. It is shown that the proposed estimate is a rigorous upper bound on the error in the case of assigned constant strain rate. Finally, a simple strategy for the automatic definition of the time-step size is proposed. The estimator and the adaptive strategy are validated by application to problems with a perfectly plastic material model
Elastic-plastic and limit-state analyses of frames with softening plastic-hinge models by mathematical programming
No full textFrames (and more general beam systems) subjected to monotonic loading are modelled by conventional finite elements with the traditional assumption of possible plastic deformations concentrated in pre-selected “critical sections”. The inelastic behaviour of these beam sections, i.e. the development of “plastic hinges”, is described by piece-wise-linear constitutive models allowing for hardening and/or softening, in terms of generalized stresses and conjugate kinematic variables.
The following topics are discussed: step-by-step analysis methods, both “exact” and stepwise holonomic; path bifurcations and overall stability; limit and deformation analyses combined, as an optimization problem under complementarity constraints apt to compute the safety factor (with respect to global or local failures); numerical tests of non-conventional algorithms by means of simple representative applications.
The objective of the paper is to provide a unified methodology and to propose novel procedures for inelastic analyses of frames up to failure, in the light of recent results in mathematical programming, particularly on complementarity theory
- …
