Artificial neural network modeling to predict the flexural behavior of RC beams retrofitted with CFRP modified with carbon nanotubes

Hashem K. Almashaqbeh, Mohammad R. Irshidat, Yacoub Najjar and Weam Elmahmoud

Abstract

In this paper, the artificial neural network (ANN) is employed to predict the flexural behavior of reinforced concrete (RC) beams retrofitted with carbon fiber/epoxy composites modified by carbon nanotubes (CNTs). Multiple techniques are used to improve the accuracy of the ANN prediction, as the data represents a multivalued function. These techniques include static ANN modeling, ANN modeling with load history, and ANN modeling with double load history. The developed ANN models are used to predict the load-displacement profiles of beams retrofitted with either CFRP or CNTs modified CFRP, flexural capacity, and maximum displacement of the beams. The results demonstrate that the ANN is able to predict the flexural behavior of the retrofitted RC beams as well as the effect of each parameter including the type of the used epoxy and the presence of the CNTs.

Key Words

artificial neural network, carbon nanotubes, CFRP, composites, flexural, RC beams

Address

Hashem K. Almashaqbeh: Department of Civil Engineering, Isra University, Amman, Jordan Mohammad R. Irshidat: Center for Advanced Materials (CAM), Qatar University, Doha, Qatar Yacoub Najjar: Department of Civil Engineering, The University of Mississippi, MS 38677, USA Weam Elmahmoud: Department of Civil Engineering, The University of Mississippi, MS 38677, USA

← Back to Volume 30, No. 3