Reappraisal of methods for calculating flexural capacity of reinforced concrete members

Abstract

A comparison of the predicted and experimentally-established behaviour of over 150 reinforced concrete beam specimens (selected from 465 test results considered) revealed that around 20% of the specimens exhibited shear failure rather than the expected flexural failure. The work presented in this paper investigated the possibility that the causes of shear failure reflected shortcomings of the code methods adopted for calculating flexural capacity. It was found that the predicted values of flexural capacity tended to underestimate their experimentally-established counterparts by up to 17% on average. It was shown that by accounting for the triaxial stress conditions invariably developing in the compressive zone through a simple modification of code-proposed stress blocks, the correlation between predicted and experimental values was similar to the best possible one resulting from the development and use of an artificial neural network model.

Neural Network-Based Prediction: The Case of Reinforced Concrete Members under Simple and Complex Loading

“…However, this is not true in all cases, as the literature shows that the prediction of these CDCs are mostly underestimated compared to the experimental values. Therefore, another theory named the compressive force path (CFP) [14] based on the new, modified design approach or opposite assumptions of the TA, predict RC member responses very close to their experimental values [9][10][11][12]. The CFP based on the member response included the critical region of the member between points of the counter flexural, unlike the sectional approach of the TA, as shown in Figure 1.…”

“…The function of artificial neural networks (ANNs) is inspired by the human biological nervous system [9][10][11][12][13]. ANNs can experience, learn, classify, recognize and predict the assigned problems, based on the information [46,47] fed to them during a training process.…”

“…ANNs contain a different combination of layers, i.e., input layer, hidden layer(s) and output layer. These layers are interconnected through weighted links, summarized by the activation function, with the free forward and backward calculation phase during the training of the ANN, as illustrated in Figure 2 [9][10][11][12][13]. Equation (1) describe the simplified mathematical function of an artificial neuron:…”

“…The results of the activation task generate the values of the input process of the neurons of the next layer, as shown in Figure 2. The final selected weight values are obtained from the available data training process after giving the first random weights [9][10][11][12][13]. In the present study, ANNs were used to predict the load carrying capacity of the RC members at ULS.…”

“…Based on these theories, different Current Design Codes (CDC) [1][2][3][4][5][6] were established and used around the globe. However, the predictions from these CDC mostly underestimate the member response at ULS [9][10][11][12][13]. All these CDC are based on the truss analogy theory (TA), which is a sectional approach.…”