Citas bibligráficas
Marín, D., Pineda, I. (2019). Modelo predictivo machine learning aplicado a análisis de datos hidrometeorológicos para un SAT en represas [Tesis, Universidad Tecnológica del Perú]. https://hdl.handle.net/20.500.12867/3300
Marín, D., Pineda, I. Modelo predictivo machine learning aplicado a análisis de datos hidrometeorológicos para un SAT en represas [Tesis]. PE: Universidad Tecnológica del Perú; 2019. https://hdl.handle.net/20.500.12867/3300
@misc{renati/807932,
title = "Modelo predictivo machine learning aplicado a análisis de datos hidrometeorológicos para un SAT en represas",
author = "Pineda Torres, Ian Augusto",
publisher = "Universidad Tecnológica del Perú",
year = "2019"
}
The present research work focuses on the efforts being made to predict the effluent discharge (discharge) in the Aguada Blanca Dam, belonging to the Autonomous Authority of Majes (AUTODEMA), where the use of wastewater networks is introduced. short and long term memory (LSTM, by English Long Short Term Memory). This prediction constitutes an important stage for the good agreement and decision making of the competent area of the Early Warning System (SAT). For this reason, it is proposed to develop a methodology for the implementation of the LSTM network to evaluate the effluent flow and serve for prevention and good preventive decision making in the different levels of early warnings. The neural network LSTM, is trained with different sensor registers, whose characteristics are effluent water volume (discharge flow), evaporation, minimum temperature, maximum temperature and precipitation. The obtained metric is 1.30 Mean Quadratic Error (RMSE) obtained in the training of the neural network, making it acceptable for the prediction of the next discharge discharge.
Este ítem está sujeto a una licencia Creative Commons Licencia Creative Commons