Thermal shock resistance of nanocomposites reinforced concrete pier shape structures: Presenting hybrid deep neural networks to obtain properties of construction and building materials under high temperature

Within a structural system, the load-bearing columns hold significant importance as well as being susceptible to damage. The destruction of piers can result in partial failure or possibly catastrophic gradual collapse of the structure, leading to considerable casualties and substantial economic damage, as demonstrated by the unintentional detonation of 2750 tons of Ammonium Nitrate in the city of Beirut, Lebanon. The ability of columns to withstand severe loads and maintain little deflection is crucial for the survival of the bridge construction and the safety of its occupants. So, this work introduces graphene nanoplatelets (GPLs) as a nanocomposite reinforcement to improve the critical positions of the concrete piers subjected to thermal shock loading as the main part of bridge construction. Using mathematical modeling of the bridge pier subjected to external thermal shock, the governing equations of this kind of applicable structure are obtained. For extracting the results, the differential quadrature method (DQM) and the Laplace transform approach are used to solve the governing equations of the problem. In this work, at first, using presented hybrid deep neural networks, the material properties of the current concrete bridge pier subjected to external thermal shock via a dataset in the open literature is presented. After obtaining the thermo-mechanical properties of the current concrete bridge pier, the results are presented. The results show that relaxation time, thermal shock loading, and residual stresses have important roles in the bending properties of the bridge pier structures.