University of Qom Civil Infrastructure Researches 2783-140X 4 2 2019 02 20 Comparative study of the performance of finite element method and evolutionary model in seepage discharge predicting from the body of an earth dam Comparative study of the performance of finite element method and evolutionary model in seepage discharge predicting from the body of an earth dam 1 15 1255 10.22091/cer.2018.2349.1094 FA Hassan Sani M.Sc, Department of Civil Engineering, University of Tabriz, Tabriz, Iran Kiyoumars Roushangar Associate Professor, Department of Civil Engineering, University of Tabriz Roghayeh Ghasempour PhD student, Department of Civil Engineering, University of Tabriz. Journal Article 2017 08 07 In this study, the capability of gene expression programming (GEP) and a finite element method (SEEP/W) was assessed for predicting water seepage from body and foundation of zonouz earthen dam located in East Azerbaijan Province. Therefore, different models using observational data of Zonouz dam (water level in the reservoir, pizometer levels implanted in the body and injected tunnel dam and seepage discharge) were developed in three states: single-mode of seepage discharge related to pizometer level, single-mode of seepage discharge related to reservoir level and combined state of seepage discharge related to reservoir and pizometer level and the results were analyzed. From the sensitivity analysis the most effective parameter for any states were determined. Then the best model of GEP was reanalyzed using SEEP/W software. The results showed that the single-mode which the seepage discharge depends to only reservoir level leads to more accurate results and in this regard, the gene expression programming approach is more successful than the applied finite element method. The sensitivity analysis showed that the water level related to one day ago in pizometer which located at level of 1850m (PL105(t-1)) and reservoir level related to two days ago (RL(t-2)) are the most effective parameters in modeling. In this study, the capability of gene expression programming (GEP) and a finite element method (SEEP/W) was assessed for predicting water seepage from body and foundation of zonouz earthen dam located in East Azerbaijan Province. Therefore, different models using observational data of Zonouz dam (water level in the reservoir, pizometer levels implanted in the body and injected tunnel dam and seepage discharge) were developed in three states: single-mode of seepage discharge related to pizometer level, single-mode of seepage discharge related to reservoir level and combined state of seepage discharge related to reservoir and pizometer level and the results were analyzed. From the sensitivity analysis the most effective parameter for any states were determined. Then the best model of GEP was reanalyzed using SEEP/W software. The results showed that the single-mode which the seepage discharge depends to only reservoir level leads to more accurate results and in this regard, the gene expression programming approach is more successful than the applied finite element method. The sensitivity analysis showed that the water level related to one day ago in pizometer which located at level of 1850m (PL105(t-1)) and reservoir level related to two days ago (RL(t-2)) are the most effective parameters in modeling. https://cer.qom.ac.ir/article_1255_8216cf143d29576f87f5c1d0227c2cc1.pdf