Neural Network Model to Predict Shear Strength of RC Beam

EasyChair Preprint 8951

Abstract

In recent years, application of Artificial Neural Networks (ANN) in civil engineering has drawn lot of attention. The potential of ANN as an analytical substitute for conventional methodologies, which are usually bound by inflexible assumptions, is recognized and accepted widely. Artificial neurons, which are a set of interconnected units or nodes that loosely resemble the neurons in a biological brain, are the foundation of ANN. Like the synapses in a human brain, each link has the ability to send a signal to neighboring neurons. After receiving inputs, an artificial neuron processes them and the output of each neuron is computed by a function of the sum of its inputs. The present work focuses on the use of ANN to predict the shear strength of reinforced concrete beams without shear reinforcement. The conventional stress analysis criteria are neither adequate to anticipate the shear strength of reinforced concrete beams nor competent to characterize the failure mechanism in beams. The neural network is trained using google colaboratory platform considering experimental data gathered from previous research studies. The results are compared with experimentally measured shear strength as well as those computed from various codes of practice. It is inferred that with adequate training ANN can predict the shear strength of RC beam satisfactorily.

Keyphrases: Artificial Neural Network (ANN), Reinforced Concrete beam (RC beam), shear strength

@booklet{EasyChair:8951,
  author    = {S Raviraj and G N Anil},
  title     = {Neural Network Model to Predict Shear Strength of RC Beam},
  howpublished = {EasyChair Preprint 8951},
  year      = {EasyChair, 2022}}