Transverse Connectivity and Durability Evaluation of Hollow Slab Bridges Using Surface Damage and Neural Networks: Field Test Investigation

Prefabricated concrete hollow slab bridges are widely used in short- and medium-span highway China due to the advantages of high production quality, installation convenience, low construction cost. Field investigation shows that severe hinge joint damage occurred during service life, mechanical performance also deteriorated with weakened joints. It is important accurately evaluate ensure safety system. In this paper, transverse connectivity durability investigated a field test using surface neural networks. Hollow Wu-He system were taken as background bridge. Surface was visually checked statistically analyzed. Static load conducted joints based on girder responses. The then demolished, total 75 segments cut off. Durability girders evaluated conditions rebars, analytic hierarchy process along fuzzy comprehensive evaluation method employed. Results showed there two main types defects bridges, i.e., cracks bottom plates longitudinal distribution deflection each non-uniform weakening connectivity, within moderate deterioration condition after 25 years’ life. An networks verified by data. maximum crack width at different locations input layer network, or considered output results. prediction error set 15.0% for 40% result girder, indicating feasibility method.