Automated Structural Assessment of existing reinforced concrete underpasses

Abstract

Much of the Dutch infrastructure was built in the 1960s and 1970s and includes an estimated 55.000 bridges and viaducts. These aging structures have reached the end of their designed service life or will reach it in the short term. It is estimated that 15% of these structures have a box-shaped cross-section as being reinforced concrete underpasses and culverts. Replacing all these structures in the short term is not practical and financially feasible. In order to preserve these structures, it must be demonstrated based on a structural assessment that the structural safety is guaranteed. So many assessments will be performed by different engineers which is a time-consuming process and results in a non-consistent assessment where unnecessary disapproval occurs which in turn leads to unnecessary costs and loss of time. The aim of this thesis is to set up a structured consistent automated method with which unnecessary disapproval and unjustified approval is avoided, a necessary efficiency improvement is achieved and structures where structural safety is at risk can be identified. Quantitative research has been conducted in which numerical data from parameter studies is collected and analysed to investigate relationships between the studied variables, find patterns and generalize results. First a data analysis of existing underpasses and culverts in the Netherlands was carried out, after which the modeling of the load effect and the capacity of existing reinforced concrete structures was investigated. An automated analytical model was created for the first step to determine the load effect, including the method of Guyon-Massonnet for traffic load distribution for the determination of bending moments and an application of method 7.3-7 from CEB-fib model code 2010 to determine the shear force as a result of the local tandem system loads. The determination of the capacity of existing concrete structures was investigated arithmetically. As a second step to determine the load effect, a model refinement was performed by automating a 2,5D FEM plate model. Structures from the database were assessed with both the analytical model and the 2,5D FEM model and a sensitivity analysis was performed based on theoretical structures within the boundaries derived from the data analysis. It is concluded that it is the most efficient method to first assess the large batches with the analytical model and that the 2,5D FEM plate model for structures with an intersection angle smaller than 80gon and structures that are disapproved with the analytical model offers added value in terms of a lower UC. From the comparison with the 2,5D FEM plate model it is concluded that the method as an extension of method 7.3-7 from (CEB-fib, 2010) results in too optimistic values for the occurring shear force as a result of the tandem system load due to an overestimation of the vertical load distribution. This research shows that an automated structural assessment can be of added value on the classic approach of assessing existing concrete underpasses and culverts by significantly improving efficiency and therefore to realize a cost-efficient method.