Akademik Veri Yönetim Sistemi
ENGINEERING GEOLOGY, cilt.209, ss.44-55, 2016 (SCI-Expanded)
Turkey has been faced with an escalating energy demand and recurring droughts within the last few decades. The construction of the BAGISTAS 1 Hydroelectric Power Plant Dam, one of the dams constructed in order to solve these problems, resulted in the partial submersion of a number of pre-existing railway bridges and retaining walls of the Divrigi-Ilic-Erzincan Railway System (NE Turkey). Before the accumulation of dam water, the structural safety of these 86-year-old infrastructures, which were constructed using carbonate rocks, were investigated under saturated conditions. The maximum uniaxial compressive strength (UCS) losses under saturated conditions, after the application of freezing-thawing, and after wetting-drying cycles, were determined. For the mortar samples obtained from a drill core, the wet-to-dry UCS ratio was determined to be 0.82, suggesting a high durability performance. The natural filling material, which was used behind the retaining structures and as the railway embankment, was classified as the selective filling material, representing the best conditions for a filling material. The samples representing the retaining wall and filling materials had very high slake durability indexes, showing that they are very durable under the effect of water. The closed-form analysis for partially submerged retaining walls indicated that the structures are safe against overturning and have permissible internal wall stresses under operational conditions. In addition, the structural safety assessment of a masonry bridge was investigated using 3D Finite Element Modeling (FEM) under the designed train and expected earthquake loads, in both dry and partly immersed conditions. The results of the study showed that the strength reduction of masonry in saturated conditions, under the raised waters of the newly constructed dam, has an insignificant effect on the submerged sections and does not pose any danger to the overall structural performance. (C) 2016 Elsevier B.V. All rights reserved.