Rrecommendations for Full Long-Term Analysis of Long-Span Bridges

Abstract

Investigations of extreme buffeting responses on long-span bridges suggest revisiting the current design guidelines as these might be non-conservative. For such purposes, the full long-term analysis is the most accurate methodology. Despite its well-known advantages, the method is not the standard practice in bridge design, mainly due its high computational demand. Recent studies have contributed to make the analysis feasible by employing enhancing strate-gies such as machine learning and efficient simulation techniques. The results are promising, and the computational effort can be reduced to a fraction of the original formulation. It’s necessary that the more accurate long-term methods are included in the design guidelines for long-span bridges. In the Norwegian Handbook for bridge design, only a simplified methodology is recommended, which exposes a gap between research and engineering practice that must be fulfilled. Therefore, this paper presents a set of recommendations for implementing the full long-term analysis based on our experience with the calculations from the Sulafjord bridge, a 2800m single span suspension bridge currently in design and feasibility stage. The objective is to provide recommendations to back-up the abovementioned regulating effort and contribute to close the gap between academia and engineering practice.