USE OF POM AND ARTIFICIAL NEURAL NETWORKS IN THE THREEDIMENSIONAL MODELING OF LAKES : GOKPINAR DAM RESERVOIR AS A CASE STUDY

• Main Authors: Mahmut FIRAT, Fatih DİKBAŞ
• Format: Article
• Language: English
• Published: Pamukkale University 2006-01-01
• Series: Pamukkale University Journal of Engineering Sciences
• Subjects: Princeton okyanus modeli, Yapay sinir ağları, Hidrodinamik modelleme
• Online Access: http://dergipark.ulakbim.gov.tr/pajes/article/view/5000089193

The circulation pattern in lakes and reservoirs varies according to many external factors. In situ measurement of the occuring flow pattern in every point of the lake is a very costly and hard task. For this reason, models determining the velocities and surface fluctuations are developed by using computers. The use of these models enables the generation of the foundation for the prediction of possible environmental problems and water pollution concentrations. Today, three dimensional models are widely used in the modelling of lakes and reservoirs. In this study, the velocity profiles and surface fluctuation values generated under various wind speed and directions at some sections in Gokpinar Lake in Denizli are obtained by applying artificial neural networks (ANN) on the results of three dimensional hydrodynamic model of the lake made with Princeton Ocean Model (POM). The developed ANN model is applied to the same sections for different wind conditions and it is found that the results are in accordance with the results of POM. As a result of the comparisons of the models, the superiorities of the models on each other at the model generation and solution phases are determined and mentioned.