Dynamic post-elastic response of transmission towers

Collapse of transmission towers can occur due to accidental loads such as conductor breakages, failures of insulators or other components, either under every day conditions (components with marginal strengths) or under extreme conditions such as ice storms, thunderstorms, tornadoes, fires, explosions, heavy mass impacts, etc.Furthermore, the trigger of one tower collapse may cause a catastrophic cascading failure of the whole transmission line section as was observed in the 1998 ice storm in Quebec, Canada.Knowledge of the post-elastic capacity of towers is necessary to mitigate the risk of cascading failures in overhead lines.The thesis presents a detailed study of the post-elastic response of latticed towers combining advanced (highly nonlinear) finite element analysis and full-scale dynamic testing of four tower section prototypes.USFOS, commercial software developed for post-elastic analysis of offshore platform structures, was selected as the numerical analysis tool to perform the nonlinear static and transient dynamic analysis of transmission towers.USFOS good performance was demonstrated on several case studies.The lattice towers are modeled with special three-node beam elements that include nonlinear material constitutive models for post-elastic response and the geometric stiffness matrices for elements are progressively updated to account for the second order effects.The numerical models also include the effects of connection eccentricities between diagonal members and the main leg members.The numerical models have been used to plan the physical tests and for re-analysis of the models with the experimental loads as measured during the physical tests.Four full-scale transmission tower model sections were built and tested under different load scenarios to verify the results from the ii numerical analysis.The prototypes were loaded at the tip of the cross arm, in the tower transverse direction (along the cross arm length) for the bending test and in the perpendicular direction, for the flexure-torsion test.Each of these loading cases was applied first in the quasi-static regime, in a pushover test, and then in the dynamic regime, under a strong dynamic pulse.These dynamic pulses were realized by dropping a large mass weighing 12.6kN from different heights (2 m for flexure-torsion and 6 m for bending).The salient conclusions of the research as follows: The numerical models and the physical test results are in excellent agreement, both in terms of predicting the collapse loads and the sequence of element failures until collapse. The research demonstrates that it is possible to use post-elastic analysis to accurately predict the reserve strength of bolted lattice towers provided connection eccentricities are properly modeled at peaks or cross arms loading points and in diagonals connected only on one leg. Both the numerical model and experimental results indicate significant postelastic reserve strength of the tower section.In the tower prototypes tested in this research, the post-elastic reserve strength was 1.22 for flexure-torsion (i.e.tower under longitudinal loading) governed by diagonals, and 1.37 for bending (i.e.tower in transverse loading) governed by inelastic buckling of the main legs. Diagonal members affect the failure modes of transmission towers and their connection design may be a weak link in the development of their post-elastic capacity.Diagonal members connected on one leg only are subject to biaxial iii bending, they cannot develop the full strength of their cross section since the unconnected leg takes much less stress on its entire length. Consideration of the tower post-elastic capacity is necessary for realistic assessment of tower vulnerability to extreme loads. Accurate pushover post-elastic analysis is an essential design tool to ensure that the tower capacity is adequate and that failure modes are safe, i.e. not leading to progressive collapse.With appropriate training, such analysis is feasible in a design office. Observations from physical tests, confirmed by numerical simulations, suggest that failure modes under pushover static and dynamic pulse loading are similar. The ultimate loads sustained by the prototypes in the dynamic tests are higher than their static counterparts (162kN vs. 126kN in bending, and 57kN vs. 51.2kN in flexure-torsion); this can be explained by the strain rate effects, which were particularly large in the bending test due to a mass dropping height of 6 m. The numerical models could not accurately predict the displacements of the prototypes due to foundation movement and splice connection slippage, which were not modelled.However, such movements do not have a significant effect on the tower capacity and post-elastic response.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.