Dynamic response of RC chimney under seismic excitations

Abstract

The industrial reinforced concrete (RC) chimneys almost are tall and slender structures with circular cross-sections. In dynamic analyses and design of such structures, the seismic loads, wind-induced pressure and dead loads are taken into account. Exact geometry of the reinforced concrete chimney, it plays an important role in structural behaviour under the dynamic loads such as winds and earthquakes. The geometric dimensions and types of the reinforced concrete chimneys effect its stiffness parameters. Therefore, it should be careful in the modelling of the structures. Basic design parameters such as the height above the ground and the diameter at the top, etc., of the chimney frequently are determined due to the own national environmental requirements for where the structure is to be constructed. The aim of the study is to obtain and compare the behaviour of RC chimneys excited by an earthquake for different design codes. The standards considered in this study are Turkish Seismic Code 2018 and Eurocode 8 and their elastic response spectrum will be considered. In this study, the RC chimney it was built in 1992 to serve as a power generation company in Jiangsu, China. The selected reinforced chimney was designed according to 1989 China seismic design regulations. The chimney is modelled using the same design details. The focus in this work is on the comparison of seismic analyses results in accordance with Eurocode 8 and Turkish Seismic Code 2018. In 3-D modelling of the reinforced concrete chimney, Abaqus finite element program is utilized. The concrete damage plasticity model (CDP) is used to demonstrate the inelastic behaviour of concrete, defining the behaviour of concrete under periodic or dynamic loads and displaying residual concrete damage. To demonstrate the plastic behaviour of steel rebars, an isotropic hardening model is used. Nonlinear dynamic time history analyses are applied to study the seismic response of RC chimney according to Eurocode 8 and Turkish Seismic Code 2018. The analyses are performed by simultaneously affecting the model's seismic acceleration records for one horizontal directions. The vertical component of the earthquake record is not taken into account. The model is released along the seismic acceleration record and fixed along the perpendicular direction. Seismic loads are applied horizontally, in the X-axis direction. Seven seismic records are selected according to Eurocode 8 and Turkish seismic code. Using SeismoMatch software, it is matched to the elastic response spectrum of the viscous damping 5 percent by the spectral matching method. To apply time history analyses in Abaqus, the implicit method of time integration is followed. Abaqus's dynamic implicit steps use the Hilber-Hughes-Taylor method, which it is an extension of the Newmark β- method, and it is associated with three parameters, α, β, and γ. And their values, -1/3≤ α ≤ 0, β >0, γ ≥ 1/2. In this study, the value of (α) is assumed to be (-0.05), the recommended value by Abaqus. A modified iterative Newton Raphson procedure is performed during the dynamic analyses. A modal analysis is done by changing the element mesh size of the model to obtain the most appropriate finite element mesh size (the number of elements changes accordingly). In the analysis due to the mesh size of the element, the effect of it on frequency is investigated. When the obtained results are almost fixed related to the mesh number, it is acceptable as the most suitable mesh size. The lateral displacement, stress distribution, and compression and tension damages of the chimney under seismic loading are studied and compared with Eurocode 8 and Turkey Seismic Code 2018 for fixed base. Then a foundation base will be added to the model for one case, and compared with the fixed base. The results of the seismic analyses showed that maximum stress was formed in the lower area of the chimney. At the top, the stress is significantly reduced. The lateral displacement increased linearly towards the top of the chimney and remained relatively constant around the circumference. The model has not been damaged by compression. A tension damage occurred in the upper part of a chimney in a particular area. When the raft foundation is included in the analyses, the lateral displacement has slightly increased as well as the vertical stress. And the vertical stress occurs in the connection region between chimney and raft foundation. Also, in this case no compression damage occurs, some increase in the tension damage area.