Tenza-Abril, Antonio José, Villacampa, Yolanda, Solak, Afonso M., Baeza Brotons, Francisco
Prediction and sensitivity analysis of compressive strength in segregated lightweight concrete based on artificial neural network using ultrasonic pulse velocity
Construction and Building Materials. 2018, 189: 1173-1183. doi:10.1016/j.conbuildmat.2018.09.096
URI: http://hdl.handle.net/10045/80949
DOI: 10.1016/j.conbuildmat.2018.09.096
ISSN: 0950-0618 (Print)
Abstract: 
Due to the low density of the aggregates used, lightweight aggregate concrete (LWAC) is susceptible to segregation because of the differences between the densities of their components. The segregation in LWAC causes a great variability in the concrete properties causing negative effects in its mechanical properties and durability. Ultrasonic velocity and artificial neural network (ANN) were applied by diagnosis and prediction in the impact of the compressive strength in LWAC specimens. 640 experimental observations were used to select the best ANN model. A sensitivity analysis was performed to observe the response of the model to perturbations in longitudinal wave velocity up to ±10% of the value observed experimentally. ANN was found to be suitable to predict the compressive strength through ultrasonic pulse velocity. This study leads to future research in non-destructive measurements to describe the segregation phenomenon in LWAC.
Keywords:ANN, Compaction, Concrete, Lightweight, Prediction, Segregation, Vibration
Elsevier
info:eu-repo/semantics/article