Structural investigation of masonry stone arch bridges

Turkey, with its ancient civilization history, has an invaluable inventory of structural heritages inherited from the past eras. Existing structural reserves include historic mosques, bridges, churches, viaducts, palaces, traditional baths, schools and residential houses etc. Historical bridges have an important part within these heritages in Turkey, and the structural system type is mostly stone arch masonry. Protection of historic masonry bridges and their transition to next generations are priority national and international goals. Historical masonry bridges, which are mostly located far from urban areas, typically do not undergo sufficient maintenance and repair. Recently, greater emphasis has been placed on maintaining the integrity of existing stone masonry bridges as important heritage structures. Numerous restoration projects have been undertaken by the General Directorate of Turkish Highways (KGM) to preserve and protect historic bridges. There are many studies in literature about modeling and structural evaluation of stone arch bridges. Analysis and performance evaluation of historic stone arch bridges can be performed using simplified analytical approaches as well as numerical approaches using computer technology. In the latter, important aspects of modeling are the determination of the materials properties and selection of a suitable modeling approach that is sufficiently accurate yet computationally inexpensive. There are two ways to assess these type of bridges; (a) simple and rapid assessment method which use linear elastic and isotropic material model basis, (b) complex and detailed methods which use nonlinear models that take into account the behavior of the material. In this study, four historic stone arch bridges Dicle (On Gozlu), Malabadi, Papaz and Sinanlı-Alpullu, which were constructed in different periods and have different geometric properties such as span, rise, and length, were selected for evaluation of structural behavior. The followed approach included performing analyses using two different modeling approaches. 3D FE models of the bridges were created using FX+ DIANA software using two different mesh sizes for each bridge. Linear static analysis, modal analysis and response spectrum analysis were carried out for both models involving different mesh sizes. The Specification for Buildings to be Built in Seismic Zones (Turkish seismic code, 2007) was used to perform structural response spectrum analysis under various load combinations. The obtained results for both models with two mesh sizes, which are referred to as fine and coarse meshes, were compared and discussed for selected bridges. Considering computation time and obtained minor differences between results of fine and coarse meshes models, coarse mesh was used to perform complex and detailed nonlinear static and dynamic analyses. Nonlinear static analyses of the studied bridges were performed using the incremental lateral force method for coarse mesh models. According to the first mode shape, which had the maximum mass participating ratio, gradually increased lateral force was applied to the bridges until capacity is reached. The obtained results, in the form of base shear force/weight (V/W) ratio corresponding to the drift ratio (d/h), were discussed in relation to the geometric characteristics. It was concluded that, span, rise and thickness of the arch is directly related with the ultimate drift capacity of the studied bridges. In the nonlinear dynamic analyses, ten earthquake ground motion records for site class A were used in the analysis of the studied bridges. Maximum base shear corresponding to the maximum displacement under ten ground motion records were assessed for each selected bridge. The performance of the bridges was discussed according to the base shear force/weight (V/W) ratio corresponding to the drift (d/h) ratio. Duzce, 1999 (Turkey) earthquake motion with a peak ground acceleration in the selected 10 ground motion records, cause the maximum displacement for each bridge. It was observed that for Duzce, 1999 (Turkey) ground motion, the capacity of Sinanlı-Alpullu Bridge was exceeded. In the case of nonsymmetrical geometry, the pushover analysis results.