The Dynamic Effect of Downburst Winds on the Longitudinal Forces Applied to Transmission Towers

Abstract

Electric power transmission line systems are a key component in the development of our modern societies. The uninterrupted availability of electric power mandates building transmission line structures that can withstand severe natural hazards. The current study focuses on the performance of conductor line systems subjected to downburst loading, which along with other types of wind events are responsible for 80% of weather related failures of transmission lines. Due to the special configuration of downburst wind fields, a unique failure mode of transmission line structures can be found to damage the cross-arms of the supporting towers as a result of the development of longitudinal forces in the conductor lines. The main concern of the current study is to explore the dynamic effect of downburst loading on these longitudinal forces, and to examine the validity of using quasi-static analysis. This is done through two levels of analysis, model-scale and full-scale. The model-scale analysis is first used to propose a simulation technique that can predict the dynamic response of conductor systems due to downburst loading overcoming the complexity induced by time varying aerodynamic damping coefficients. The proposed model is validated using a unique experiment conducted at WindEEE dome testing facility at The University of Western Ontario in Canada. The validated model is used to conduct analysis at the full-scale level using a wind field generated by a Computational Fluid Dynamics (CFD) model. Results show that the longitudinal forces within the conductor system are dynamically insensitive and can therefore be treated quasi-statically. Lastly, to judge the applicability of the wind field used, a comparison was made between the CFD wind field and full-scale records. The comparison was done through two parameters that would affect the quasi-static response; the first relating the peak velocity to the post peak velocity, and the second defining the ramping period. It was concluded from the comparison that varying the defined parameters would result in a maximum difference of 5% in the computed reactions.