International audienceIn this paper, the specific effect of additional constraints on the stability of undamped non-conservative elastic systems is studied. The stability of constrained elastic system is compared to the stability of the unconstrained system, through the incorporation of Lagrange multipliers. It is theoretically shown that the second-order work criterion, dealing with the symmetric part of the stiffness matrix corresponds to an optimization criterion with respect to the kinematics constraints. More specifically, the vanishing of the second-order work criterion corresponds to the critical kinematics constraint, which can be interpreted as an instability direction when the material stability analysis is considered (typically in the field of soil mechanics). The approach is illustrated for a two-degrees-of-freedom generalised Ziegler's column subjected to different constraints. We show that a particular kinematics constraint can stabilize or destabilize a non-conservative system. However, for all kinematics constraints, there necessarily exists a constraint which destabilizes the non-conservative system. The constraint associated to the lowest critical load is associated with the second-order criterion. Excluding flutter instabilities, the second-order work criterion is not only a lower bound of the stability boundary of the free system, but also the boundary of the stability domain, for all mixed perturbations based on proportional kinematics conditions
International audienceThis paper deals with the buckling of a column which ismodeled by some finite rigid segments and elastic rotational springs and relating its solution to continuum nonlocal elasticity. This problem, which can be referred to Hencky's chain, can serve as a basic model to rigorously investigate the effect of the microstructure on the buckling behaviour of a simple equivalent continuum structural model. The buckling problem of the pinned-pinned discretized column is analytically investigated by introducing a Lagrange multiplier. Such a buckling problem is mathematically treated as an iterative eigenvalue problem. It is shown that the buckling load of this finite degree-of-freedom system is exactly obtained by a recursive formula involving Chebyschev polynomials. Euler's buckling load is asymptotically obtained at larger scales. However, at smaller scales, the buckling model highlights some scale effect that can be only captured by nonlocal elasticity for the equivalent continuum. We show that Eringen's nonlocal continuum is well suited to capture this scale effect. The small scale coefficient of the equivalent nonlocal continuum is then identified from the specific microstructure features, namely the length of each cell. It is shown that the small length scale coefficient valid for this buckling problem is very close to the one already identified from a comparison with the Born-Karman model of lattice dynamics using dispersive wave properties
International audienceAlthough the concept of the second-order work criterion dates back to the middle of the past century, its physical meaning often continues to be debated. Recent papers have established that a certain class of instabilities, related to the occurrence of an outburst in kinetic energy, could be properly detected by the vanishing of the second-order work. This manuscript attempts to extend the second-order work formalism to boundary value problems. For this purpose, the role of the boundary stiffness tensor (relating external forces and displacement components) is put forward in the occurrence of instability by divergence. Omitting body forces, a global method is then given to compute the second-order work terms directly. The capability of this formalism is finally demonstrated in the context of engineering issues
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.