A Hybrid ANFIS-GA Approach for Estimation of Hydrological Time Series

Basit Öğe Kaydı
dc.authoridHaznedar, Bulent/0000-0003-0692-9921
dc.contributor.authorHaznedar, Bulent
dc.contributor.authorKılınç, Hüseyin Çağan
dc.date.accessioned2025-03-26T17:35:02Z
dc.date.available2025-03-26T17:35:02Z
dc.date.issued2022
dc.departmentİstanbul Esenyurt Üniversitesi, Fakülteler, Mühendislik ve Mimarlık Fakültesi, İnşaat Mühendisliği Bölümü
dc.description.abstractRiver flow modeling plays a leading role in the management of water resources and ensuring sustainability. The complex nature of hydrological systems and the difficulty in the application process have led researchers to seek more instantaneous methods for flow predictions. Therefore, with the development of artificial intelligence-based techniques, hybrid modeling has become popular among hydrologists in recent years. For that reason, this study seeks to develop a hybrid model that integrates an adaptive neuro-fuzzy inference system (ANFIS) with a genetic algorithm (GA) to predict river flow. Fundamentally, the performance of an ANFIS model depends on the optimum model parameters. Thus, it is aimed to increase the prediction performance by optimizing the ANFIS parameters with the population-based GA algorithm, which is a powerful algorithm. In this respect, the data gathered from Zamanti and Korkun Flow Measurement Stations (FMS) of Seyhan River, one of Turkey's significant rivers, were employed. Besides, the proposed hybrid ANFIS-GA approach was compared to classical ANFIS model to demonstrate the improvement of its performance. Also, within the scope of simulation studies, the traditional artificial neural networks (ANN) and the long-short term memory (LSTM) method which is a quite popular in recent years were used to predict streamflow data. The estimation results of the models were evaluated with RMSE, MAE, MAPE, SD, and R-2 statistical metrics. In a nutshell, the outcomes indicated that the proposed ANFIS-GA method was the most successful model by achieving the highest values of R-2 (approximate to 0.9409) and R-2 (approximate to 0.9263) for the Zamanti and Korkun FMS data.
dc.identifier.doi10.1007/s11269-022-03280-4
dc.identifier.endpage4842
dc.identifier.issn0920-4741
dc.identifier.issn1573-1650
dc.identifier.issue12
dc.identifier.scopus2-s2.0-85135615272
dc.identifier.scopusqualityQ1
dc.identifier.startpage4819
dc.identifier.urihttps://doi.org/10.1007/s11269-022-03280-4
dc.identifier.urihttps://hdl.handle.net/20.500.14704/1013
dc.identifier.volume36
dc.identifier.wosWOS:000836827600001
dc.identifier.wosqualityQ1
dc.indekslendigikaynakWeb of Science
dc.indekslendigikaynakScopus
dc.language.isoen
dc.publisherSpringer
dc.relation.ispartofWater Resources Management
dc.relation.publicationcategoryMakale - Uluslararası Hakemli Dergi - Kurum Öğretim Elemanı
dc.rightsinfo:eu-repo/semantics/closedAccess
dc.snmzKA_WOS_20250326
dc.subjectANFIS; Genetic algorithm; Soft computing; Streamflow; Forecasting; River flow
dc.titleA Hybrid ANFIS-GA Approach for Estimation of Hydrological Time Series
dc.typeArticle

Dosyalar

Koleksiyon

WoS İndeksli Yayınlar Koleksiyonu
Mühendislik ve Mimarlık Fakültesi Koleksiyonu
Scopus İndeksli Yayınlar Koleksiyonu