Daily Streamflow Forecasting Based on the Hybrid Particle Swarm Optimization and Long Short-Term Memory Model in the Orontes Basin

Basit Öğe Kaydı
dc.authoridKilinc, Huseyin Cagan/0000-0003-1848-2856
dc.contributor.authorKılınç, Hüseyin Çağan
dc.date.accessioned2025-03-26T17:34:45Z
dc.date.available2025-03-26T17:34:45Z
dc.date.issued2022
dc.departmentİstanbul Esenyurt Üniversitesi, Fakülteler, Mühendislik ve Mimarlık Fakültesi, İnşaat Mühendisliği Bölümü
dc.description.abstractWater, a renewable but limited resource, is vital for all living creatures. Increasing demand makes the sustainability of water resources crucial. River flow management, one of the key drivers of sustainability, will be vital to protect communities from the worst impacts on the environment. Modelling and estimating river flow in the hydrological process is crucial in terms of effective planning, management, and sustainable use of water resources. Therefore, in this study, a hybrid approach integrating long short-term memory networks (LSTM) and particle swarm algorithm (PSO) was proposed. For this purpose, three hydrological stations were utilized in the study along the Orontes River basin, Karasu, Demirkopru, and Samandag, respectively. The timespan of Demirkopru and Karasu stations in the study was between 2010 and 2019. Samandag station data were from 2009-2018. The datasets consisted of daily flow values. In order to validate the performance of the model, the first 80% of the data were used for training, and the remaining 20% were used for the testing of the three FMSs. Statistical methods such as linear regression and the more classical model autoregressive integrated moving average (ARIMA) were used during the comparison process to assess the proposed method's performance and demonstrate its superior predictive ability. The estimation results of the models were evaluated with RMSE, MAE, MAPE, SD, and R-2 statistical metrics. The comparison of daily streamflow predictions results revealed that the PSO-LSTM model provided promising accuracy results and presented higher performance compared with the benchmark and linear regression models.
dc.identifier.doi10.3390/w14030490
dc.identifier.issn2073-4441
dc.identifier.issue3
dc.identifier.scopus2-s2.0-85124337496
dc.identifier.scopusqualityN/A
dc.identifier.urihttps://doi.org/10.3390/w14030490
dc.identifier.urihttps://hdl.handle.net/20.500.14704/877
dc.identifier.volume14
dc.identifier.wosWOS:000754574800001
dc.identifier.wosqualityQ2
dc.indekslendigikaynakWeb of Science
dc.indekslendigikaynakScopus
dc.institutionauthorKilinc, Huseyin Cagan
dc.language.isoen
dc.publisherMDPI
dc.relation.ispartofWater
dc.relation.publicationcategoryMakale - Uluslararası Hakemli Dergi - Kurum Öğretim Elemanı
dc.rightsinfo:eu-repo/semantics/openAccess
dc.snmzKA_WOS_20250326
dc.subjectwater resources; streamflow; particle swarm optimization; long short-term memory; time series
dc.titleDaily Streamflow Forecasting Based on the Hybrid Particle Swarm Optimization and Long Short-Term Memory Model in the Orontes Basin
dc.typeArticle

Dosyalar

Koleksiyon

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