Modelación hidrológica de largo plazo en la cuenca baja del Río Magdalena

A long term forecast of river levels is beneficial for entities in charge of preparing control measures, flood risk prevention, as well as for other entities interested in these long term level forecasts. The objective of this research project is to implement a long term hydrological model, these models will be made through the use of the ARIMA function in its "AR" or auto-regression form, also the model based on an artificial neural network (RNA) of multilayer perception type will be implemented, when implementing these two models it is sought to measure their efficiency by means of the coefficient of efficiency NSE, its quadratic error RMSE, finally, the Mann-Kendall analysis will identify sets of trends in the input data obtained from the IDEAM database, each of these parameters will help to define which forecast model has a better behavior in relation to its real data. Discussions and conclusions will be drawn from these forecasts and the parameters used to measure their efficiency and error.

• Autores: De La Parra Tovar, Ernesto Ricardo; Turizo Martínez, Jesús Isaac
• Tipo de recurso: Trabajo de grado de pregrado
• Fecha de publicación: 2020
• Institución: Corporación Universidad de la Costa
• Repositorio: REDICUC - Repositorio CUC
• Idioma: spa
• OAI Identifier: oai:repositorio.cuc.edu.co:11323/7804
• Acceso en línea: https://hdl.handle.net/11323/7804; https://repositorio.cuc.edu.co/
• Palabra clave: Forecast; Level; Arima; RNA; Model; Hydrological; Data; Trend; Pronóstico; Nivel; Modelo; Hidrológico; Datos; Tendencia
• Rights: openAccess
• License: Attribution-NonCommercial-ShareAlike 4.0 International