Informations et ressources scientifiques
sur le développement des zones arides et semi-arides

Accueil du site → Master → Turquie → Estimation of groundwater level fluctuation using adaptive neural fuzzy and support vector machines in arid and semi-arid regions

İskenderun Teknik Üniversitesi (2019)

Estimation of groundwater level fluctuation using adaptive neural fuzzy and support vector machines in arid and semi-arid regions

KÖRLÜ Selçuk

Titre : Estimation of groundwater level fluctuation using adaptive neural fuzzy and support vector machines in arid and semi-arid regions

Yeraltı suyu seviye değişiminin bulanık mantık ve destek vektör makineleri yöntemleri ile tahmini

Auteur : KÖRLÜ Selçuk

Université de soutenance : İskenderun Teknik Üniversitesi

Grade : Master of Science (MS) 2019

Résumé
Groundwater is one of the most frequently used water resources for domestic, urban, agricultural and industrial purposes, especially in arid and semi-arid regions. Sustainable management of groundwater resources along with surface water has become an urgent need to secure future water. Accurate and reliable estimation of groundwater levels is a very important component in the water basins in arid and semi-arid regions, which are more susceptible to hydrological extreme events, especially in drought form. In this study, the groundwater level of the region was estimated using the monthly average rainfall and temperature values of the area between 2000 and 2015 and the monthly average values of the groundwater level of Turkish General Directorate of State Hydraulic Works (DSİ), located in Hatay Reyhanlı region. With this data, Multiple Linear Regression (MLR), Anfis (Adaptive Neural Fuzzy Inference System), support vector machines with radial basis functions (SVM-RBF) and support vector machines with poly kernels (SVM- PK) methods were estimated. The performance of the groundwater level estimation was evaluated. In the estimation of the groundwater level of the region, it is seen that SVM models have better results than the MLR and ANFIS model.

Présentation

Version intégrale

Page publiée le 6 janvier 2023