Experimental and numerical investigation on the performance of GFRP-confined expansive concrete-filled unplasticized polyvinyl chloride tubes

Abstract

This paper describes both an experimental study and a numerical investigation on the compressive behavior of new types of solid as well as hollow composite columns, which consist of Unplasticized Polyvinyl Chloride (UPVC) tubes filled with expansive concrete. 32 specimens were categorized into four groups and cast with expansive concrete. Half of the specimens were confined additionally with two layers of GFRP wraps. Furthermore, two specimens were cast with ordinary concrete to be compared with those expansive counterparts. All the composite columns were tested under the monotonic compressive loading. The effects of GFRP confinement, aspect ratio of composite columns ( L/ D), and hollowness of concrete core on compressive behavior were investigated. It was demonstrated that adding expansive agents improved the load-carrying capacity and ductility of columns by enhancing the tri-axial state of stress in the concrete core. Moreover, an increase in aspect ratio decreased both peak load and its corresponding axial strain. Approximately 40% increase was observed in the values of peak load for specimens confined with GFRP wraps. Generally, concrete core removal in the hollow specimens resulted in slightly higher average compressive stress. Finally, a finite element (FE) simulation was performed using ABAQUS software, and the numerical results were validated with experimental tests of the present study.

Ensemble Learning Models for Prediction of Punching Shear Strength in RC Slab-Column Connections

“…Conventional experimental tests are the established method for evaluating various characteristics of concrete structures [10,11]. Recently, machine learning (ML) techniques have found extensive applications across diverse domains within the field of structural engineering, aiming to enhance prediction performance (i.e., accuracy) [12][13][14].…”