High performance concrete, prediction of its resistance to compression through artificial neuronal networks

• Autores: Acuña P., Luis, Espinoza H., Pedro C., Moromi N., Isabel, Torre C., Ana V., García F., Francisco
• Formato: artículo
• Fecha de Publicación: 2017
• Institución: Universidad Nacional de Ingeniería
• Repositorio: Revistas - Universidad Nacional de Ingeniería
• Lenguaje: español
• OAI Identifier: oai:oai:revistas.uni.edu.pe:article/125
• Enlace del recurso: https://revistas.uni.edu.pe/index.php/tecnia/article/view/125
• Nivel de acceso: acceso abierto
• Materia: Red Neuronal Artificial; probeta; compresión axial; aditivos; Artificial Neural Network; test tube; axial compression; additives

The building of modern housing concrete is a fundamental element that intervenes. On the other hand, in the construction of bridges, dams, tunnels, this is in the construction of non‐standard civil engineering structures, the concrete that is used is the high performance (CAR) that apart from the basic components such as water, Cement, fine and coarse aggregates, contain other cementing additives, such as microsílices. The problem is to get a technological resource that helps predict the resistance of CAR from its manufacturing data, but this is impossible. However, we have artificial neural networks that fulfill this role, which after being transformed into true mathematical functions that approximate the expected values ??of the resistance of concrete specimens. The approximation level is estimated by the correlation between the response and the expected value of the network. It is then very useful to have a neural network that simulates numerically the resistance of the concrete, even before its manufacture. In this investigation, several artificial neural networks have been obtained that predict the resistance to compression of the CAR with correlations that vary between 0.86 and 0.91.