The complexities of the Prairie watersheds, including potholes, drainage interconnectivities, changing land-use patterns, dynamic watershed boundaries and hydro-meteorological factors, have made hydrological modelling on Prairie watersheds one of the most complex task for hydrologists and operational hydrological forecasters. In this study, four hydrological models (WATFLOOD, HBV-EC, HSPF and HEC-HMS) were developed, calibrated and tested for their efficiency and accuracy to be used as operational flood forecasting tools. The Upper Assiniboine River, which flows into the Shellmouth Reservoir, Canada, was selected for the analysis. The performance of the models was evaluated by the standard statistical methods: the Nash-Sutcliffe efficiency coefficient, correlation coefficient, root mean squared error, mean absolute relative error and deviation of runoff volumes. The models were evaluated on their accuracy in simulating the observed runoff for calibration and verification periods (2005-2015 and 1994-2004, respectively) and also their use in operational forecasting of the 2016 and 2017 runoff.
The application of a data-driven adaptive neuro-fuzzy modelling technique for predicting bed load and total bed-material load for the River Rhine is summarized. Four main parameters affecting sediment transport are used to construct the model, using 560 and 510 measured bed load and total bed-material load data, respectively. Two-thirds of the available data sets are used for training and one third for testing. The initial fuzzy model is obtained by grid partitioning of the input variables. The optimization of the model is performed by data-driven tuning of the fuzzy model parameters using the adaptive neuro-fuzzy inference system, so that the model output is able to reproduce the measured value. A sensitivity analysis for the combination of input parameters, as well as the number and type of membership functions, is also performed. The model results show that the data-driven adaptive neuro-fuzzy modelling approach can be a powerful alternative technique for estimating both bed load and total bed-material load. Keywords data-driven; neuro-fuzzy model; optimization; prediction of sediment transport; sensitivity analysis Une approche de modélisation neuro-floue pour le calcul du transport de sédiments Résumé On résume ici l'application d'une technique de modélisation neuro-floue adaptative pilotée par les données pour estimer la charge de fond et la charge totale en matériaux du Rhin. Quatre paramètres principaux affectant le transport des sédiments ont été utilisés pour construire le modèle, en utilisant 560 et 510 données mesurant respectivement la charge de fond et la charge totale en matériaux. Les deux tiers des jeux de données disponibles ont été utilisés pour l'apprentissage et un tiers pour les tests. Le modèle flou initial est obtenu par séparation sur un maillage des variables d'entrée. L'optimisation du modèle est effectuée par un ajustement des paramètres du modèle flou piloté par les données, en utilisant le système d'inférence neuro-flou adaptatif, de telle sorte que la sortie du modèle soit capable de reproduire la valeur mesurée. On a également réalisé une analyse de sensibilité portant sur la combinaison des paramètres d'entrée, ainsi que sur le nombre et le type des fonctions d'appartenance. Les résultats du modèle montrent que l'approche de modélisation neuro-floue adaptative pilotée par les données peut être une technique alternative puissante pour estimer à la fois la charge de fond et la charge totale en matériaux. Mots clefs pilotage par les données; modèle neuro-flou; optimisation; estimation du transport des sédiments; analyse de sensibilité
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.