Hydrogeological model based on the numerical deepening method, applying the Back Propagation Neural Network technique for the evaluation of large water seeps inside an underground mine in Peru

The objective of this research is to propose a conceptual model based on neural networks (ANN), which lies in its ability to approximate any measurable Borel function, with the desired degree of precision, as indicated by Hornik et al. to the. (1989). ANNs became very useful in predictions, such as time series; since its ability to learn, within a large amount of data; it is potentially noisy. Starting from the collection of field information through diamond drilling (for the characterization of the rocky massif) and the use of geological maps, aerial photographs and satellite images as a means of studying groundwater. All this, in order to determine the main study parameters that will be used for the design of the hydrogeological model to be proposed. Likewise, a quality control tool is used to carry out the respective analysis of the main sources of mistakes detected. However, it requires the establishment of some standards to evaluate the quality of the data, including the successive manipulations carried out on the raw data acquired by the sensors. Finally, the hydrogeological model is constituted whose purpose is to contribute together with the geomechanical model for the mining design and to determine possible cases of risks that are frequent inside an underground mine due to the subsoil water.