Bibliographic citations
Castro, K., Brousset, J. (2023). Propuesta de Red Neuronal para Caracterizar el Macizo Rocoso en una Mina Subterránea en el Centro del Perú [Trabajo de suficiencia profesional, Universidad Peruana de Ciencias Aplicadas (UPC)]. http://hdl.handle.net/10757/670755
Castro, K., Brousset, J. Propuesta de Red Neuronal para Caracterizar el Macizo Rocoso en una Mina Subterránea en el Centro del Perú [Trabajo de suficiencia profesional]. PE: Universidad Peruana de Ciencias Aplicadas (UPC); 2023. http://hdl.handle.net/10757/670755
@misc{renati/410862,
title = "Propuesta de Red Neuronal para Caracterizar el Macizo Rocoso en una Mina Subterránea en el Centro del Perú",
author = "Brousset Cornejo, Julyans Giuseppe",
publisher = "Universidad Peruana de Ciencias Aplicadas (UPC)",
year = "2023"
}
In underground mines, challenges arise in the characterization of the rock mass using current empirical methods. This is because these characterization methods are designed based on specific geologies, and when applied in different and heterogeneous geological environments, uncertainties in rock characterization emerge. Furthermore, in underground mines, highly fractured rocks are often encountered, or situations where empirical classification methods are not applicable. In recent years, artificial intelligence, particularly Neural Networks, has been used to estimate geomechanical parameters such as Young's Modulus and uniaxial compressive strength. Therefore, it is possible to use Artificial Neural Networks to determine the Rock Mass Rating (RMR) of the rock. Artificial Neural Networks are based on learning, which means that by providing a large amount of data to train the neural network, it generates results with the highest possible accuracy. In the rock mass characterization process of this research, the following parameters were considered as input for the neural network: joint condition, spacing, Rock Quality Designation (RQD), presence of water, and uniaxial compressive strength. Additionally, the RMR index of the rock was determined as the output variable, as this value is used for rock mass characterization. The research results showed that the application of Artificial Neural Networks achieved an estimation accuracy of 99% when determining the RMR index. This represents a significant advancement in the characterization of rock masses in underground mining environments and contributes to more informed and precise decision-making in geotechnical engineering and mine safety.
This item is licensed under a Creative Commons License