Implementation of the use of artificial neural networks for the predictive calculation of concrete resistance, Trujillo

Cóndor-Palomino J.B.; Huamán-Sandoval C.S.; Noriega-Vidal E.M.; Martell-Ortiz J.C.; Valdiviezo-Velarde A.Y.; Herrera-Viloche Á.A. · Proceedings of the LACCEI international Multi-conference for Engineering, Education and Technology (2024) · DOI: 10.18687/laccei2024.1.1.488

Resumen

This research aimed to implement artificial neural networks for predicting concrete strength. Specific objectives included assessing the accuracy of these networks, determining the progress of their training, and evaluating the mean square error. The applied methodology employed a non-experimental descriptive cross-sectional design. The overall conclusion highlighted the successful implementation of artificial neural networks in predicting concrete strength. The network architecture was detailed, comprising input, hidden, and output layers with 10, 19, and 1 neuron, respectively, using Matlab. Specific findings indicated a high level of accuracy at 99.997%, confirming the effectiveness of concrete strength prediction. Furthermore, a training progress of 100.00% was achieved, demonstrating the neural network's ability to adjust internal parameters and learn to predict concrete strength with provided data. Regarding the neural network's mean square error, a coefficient of MSE = 1.5949 was obtained. Ultimately, it was concluded that the use of artificial neural networks is a valid approach for estimating concrete compressive strength. This research opens promising perspectives for their future application in monitoring concrete quality.

Instituciones: Universidad César Vallejo