Comparative Analysis of the Prediction of the Mechanical Properties of Lightweight Aggregate Concrete via Artificial Neural Network and Finite Elements Method

Abstract

Lightweight Aggregate Concrete (LWAC) is a composite comprising cement-based mortar and Lightweight Aggregates (LWA) widely employed around the world. Due to the particularities of the LWA properties, which are difficult to measure experimentally, the design of LWAC mixtures is a rather complicated task. This fact justifies the search for analytical and/or numerical methods evaluate the LWAC’s properties. Thus, the present work aims to compare the performances of two strategies to predict the compressive strength of LWAC’s samples: Finite Element models and Artificial Neural Network. To this end, both strategies use the Young’s modulus and compressive strength of the mortar and the LWA obtained from an experimental program in the literature. The results for both methods show good agreement with the validation data, and encourage further studies towards the development of a numerical tool, which may assist engineers for practical purposes.