Prediction of compressive and flexural strengths of a modified zeolite additive mortar using artificial neural network

Abstract Artificial neural network (ANN) has been used to predict the compressive and flexural strength of a mortar made with a modified zeolite additive (MZA). The ANN had three layers, which included the input, hidden and output layer. The input layer had six parameters: cement quantity, silica sand quantity, modified zeolite additive (MZA) quantity, water quantity, curing period, and load weights. The output layer consisted of either the compressive or the flexural strength. While developing the ANN model, 30 samples were used for training and testing. Two assessments were carried out, first to determine the effective number of neurons in the hidden layer in predicting the compressive strength. The second assessment evaluated the accuracy with which the neural network would predict the compressive or flexural strength under different load weights. In general, the ANN learnt from the training data and was able to give excellent results. For the compressive strength, the testing data was estimated with a correlation coefficient (R2) of 0.9967, 0.9938, 0.9924 and 0.6676 for four load weights, while the flexural strength was predicted with R2 of 0.9984 and 0.9990 for two load weights. The ANN can be used to augment or in lieu of experimental work to determine both the compressive and the flexural strength of mortar.

[1]  T. Kohonen.  Self-organized formation of topographically correct feature maps , 1982 .

[2]  Rafat Siddique,et al.  Prediction of compressive strength of self-compacting concrete containing bottom ash using artificial neural networks , 2011, Adv. Eng. Softw..

[3]  Iskender Akkurt,et al.  Prediction of compressive strength of heavyweight concrete by ANN and FL models , 2010, Neural Computing and Applications.

[4]  J. Beaudoin,et al.  Zeolite-based additives for high alumina cement products , 1996 .

[5]  Baoju Liu,et al.  Effect of curing conditions on the permeability of concrete with high volume mineral admixtures , 2018 .

[6]  Harun Tanyildizi,et al.  Predicting the core compressive strength of self-compacting concrete (SCC) mixtures with mineral additives using artificial neural network , 2011 .

[7]  H. Marzouk,et al.  Crack width in concrete using artificial neural networks , 2013 .

[8]  C. Goodier,et al.  Durability performance of sustainable structural concrete: effect of coarse crushed concrete aggregate on microstructure and water ingress , 2017 .

[9]  E. Kadri,et al.  Effect of cement and admixture on the utilization of recycled aggregates in concrete , 2017 .

[10]  B. Ikotun,et al.  Strength and durability effect of modified zeolite additive on concrete properties , 2010 .

[11]  Yves Chauvin,et al.  Backpropagation: theory, architectures, and applications , 1995 .

[12]  Amir Hossein Rafiean,et al.  Compressive strength prediction of environmentally friendly concrete using artificial neural networks , 2018 .

[13]  Geoffrey E. Hinton,et al.  Learning internal representations by error propagation , 1986 .

[14]  Teuvo Kohonen,et al.  Self-organized formation of topologically correct feature maps , 2004, Biological Cybernetics.

[15]  R. Hecht-Nielsen.  Counterpropagation networks. , 1987, Applied optics.

[16]  M. J. D. Powell,et al.  Radial basis functions for multivariable interpolation: a review , 1987 .

[17]  J J Hopfield,et al.  Learning algorithms and probability distributions in feed-forward and feed-back networks. , 1987, Proceedings of the National Academy of Sciences of the United States of America.

[18]  Kurt Hornik,et al.  Multilayer feedforward networks are universal approximators , 1989, Neural Networks.