Developing a novel hybrid model based on GRU deep neural network and Whale optimization algorithm for precise forecasting of river’s streamflow

Abstract Streamflow contemplates a fundamental criterion to evaluate the impact of human activities and climate changes on the hydrological cycle. In this study, a novel innovative deep neural network (DNN) structure by integrating a double Gated Recurrent Units (GRU) neural network model with a mul...

Full description

Saved in:
Bibliographic Details
Main Authors: Amin Gharehbaghi, Redvan Ghasemlounia, Farshad Ahmadi, Rasoul Mirabbasi, Ali Torabi Haghighi
Format: Article
Language:English
Published: Nature Portfolio 2025-06-01
Series:Scientific Reports
Subjects:
Online Access:https://doi.org/10.1038/s41598-025-03185-3
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849332996311613440
author Amin Gharehbaghi
Redvan Ghasemlounia
Farshad Ahmadi
Rasoul Mirabbasi
Ali Torabi Haghighi
author_facet Amin Gharehbaghi
Redvan Ghasemlounia
Farshad Ahmadi
Rasoul Mirabbasi
Ali Torabi Haghighi
author_sort Amin Gharehbaghi
collection DOAJ
description Abstract Streamflow contemplates a fundamental criterion to evaluate the impact of human activities and climate changes on the hydrological cycle. In this study, a novel innovative deep neural network (DNN) structure by integrating a double Gated Recurrent Units (GRU) neural network model with a multiplication layer and meta-heuristic whale optimization algorithm (WOA) (i.e., hybrid 2GRU×–WOA model) is developed to improve the prediction accuracy and performance of mean monthly Chehel-Chai River’s streamflow (CCRSF m ) in Iran. The Pearson’s correlation coefficient (PCC) and Cosine Amplitude Sensitivity (CAS) as feature (input) selection process determine the only precipitation (P m ) as the most effective input variable among a list of on-site potential climate time series parameters recorded in the study area. Thanks to a well-proportioned layer network structural framework in the suggested hybrid 2GRU×–WOA model, it leads to an appropriate total learnable parameter (TLP) compared to standard individual GRU and Bi-GRU as the benchmark models developed in the comparable meta-parameters. This hybrid model under the optimal meant meta-parameters tuned i.e., coupling a state activation functions (SAF) of tanh-softsign, dropout rate (P-rate) of 0.5, numbers of hidden neurons (NHN) of 70, outperforms with an R 2 of 0.79, NSE of 0.76, MAE of 0.21 (m3/s), MBE of -0.11(m3/s), and RMSE of 0.36 (m3/s). Hybridizing the 2GRU× model with WOA algorithm causes to increase in the value of R 2 by 6.8% and reduce in the value of RMSE by 20.4%. Comparatively, standard individual GRU and Bi-GRU models result in an R 2 of 0.59 and 0.66, NSE of 0.55 and 0.6, MAE of 0.91 and 0.53 (m3/s), MBE of 0.047 and − 0.06 (m3/s), RMSE of 1.29 and 0.83 (m3/s), respectively.
format Article
id doaj-art-d2cd9566e04f4e95a04398ffa347316c
institution Kabale University
issn 2045-2322
language English
publishDate 2025-06-01
publisher Nature Portfolio
record_format Article
series Scientific Reports
spelling doaj-art-d2cd9566e04f4e95a04398ffa347316c2025-08-20T03:46:01ZengNature PortfolioScientific Reports2045-23222025-06-0115112010.1038/s41598-025-03185-3Developing a novel hybrid model based on GRU deep neural network and Whale optimization algorithm for precise forecasting of river’s streamflowAmin Gharehbaghi0Redvan Ghasemlounia1Farshad Ahmadi2Rasoul Mirabbasi3Ali Torabi Haghighi4Department of Civil Engineering, Faculty of Engineering, Hasan Kalyoncu UniversityDepartment of Civil Engineering, Faculty of Engineering, Istanbul Gedik UniversityDepartment of Hydrology & Water Resources Engineering, Shahid Chamran University of AhvazDepartment of Water Engineering, Shahrekord UniversityWater, Energy, and Environmental Engineering Research Unit, University of OuluAbstract Streamflow contemplates a fundamental criterion to evaluate the impact of human activities and climate changes on the hydrological cycle. In this study, a novel innovative deep neural network (DNN) structure by integrating a double Gated Recurrent Units (GRU) neural network model with a multiplication layer and meta-heuristic whale optimization algorithm (WOA) (i.e., hybrid 2GRU×–WOA model) is developed to improve the prediction accuracy and performance of mean monthly Chehel-Chai River’s streamflow (CCRSF m ) in Iran. The Pearson’s correlation coefficient (PCC) and Cosine Amplitude Sensitivity (CAS) as feature (input) selection process determine the only precipitation (P m ) as the most effective input variable among a list of on-site potential climate time series parameters recorded in the study area. Thanks to a well-proportioned layer network structural framework in the suggested hybrid 2GRU×–WOA model, it leads to an appropriate total learnable parameter (TLP) compared to standard individual GRU and Bi-GRU as the benchmark models developed in the comparable meta-parameters. This hybrid model under the optimal meant meta-parameters tuned i.e., coupling a state activation functions (SAF) of tanh-softsign, dropout rate (P-rate) of 0.5, numbers of hidden neurons (NHN) of 70, outperforms with an R 2 of 0.79, NSE of 0.76, MAE of 0.21 (m3/s), MBE of -0.11(m3/s), and RMSE of 0.36 (m3/s). Hybridizing the 2GRU× model with WOA algorithm causes to increase in the value of R 2 by 6.8% and reduce in the value of RMSE by 20.4%. Comparatively, standard individual GRU and Bi-GRU models result in an R 2 of 0.59 and 0.66, NSE of 0.55 and 0.6, MAE of 0.91 and 0.53 (m3/s), MBE of 0.047 and − 0.06 (m3/s), RMSE of 1.29 and 0.83 (m3/s), respectively.https://doi.org/10.1038/s41598-025-03185-3GRU and Bi-GRU modelsMeta-heuristic Whale optimization algorithmNovel hybrid 2GRU×–WOA modelTLP parameterChehel-Chai river’s streamflow
spellingShingle Amin Gharehbaghi
Redvan Ghasemlounia
Farshad Ahmadi
Rasoul Mirabbasi
Ali Torabi Haghighi
Developing a novel hybrid model based on GRU deep neural network and Whale optimization algorithm for precise forecasting of river’s streamflow
Scientific Reports
GRU and Bi-GRU models
Meta-heuristic Whale optimization algorithm
Novel hybrid 2GRU×–WOA model
TLP parameter
Chehel-Chai river’s streamflow
title Developing a novel hybrid model based on GRU deep neural network and Whale optimization algorithm for precise forecasting of river’s streamflow
title_full Developing a novel hybrid model based on GRU deep neural network and Whale optimization algorithm for precise forecasting of river’s streamflow
title_fullStr Developing a novel hybrid model based on GRU deep neural network and Whale optimization algorithm for precise forecasting of river’s streamflow
title_full_unstemmed Developing a novel hybrid model based on GRU deep neural network and Whale optimization algorithm for precise forecasting of river’s streamflow
title_short Developing a novel hybrid model based on GRU deep neural network and Whale optimization algorithm for precise forecasting of river’s streamflow
title_sort developing a novel hybrid model based on gru deep neural network and whale optimization algorithm for precise forecasting of river s streamflow
topic GRU and Bi-GRU models
Meta-heuristic Whale optimization algorithm
Novel hybrid 2GRU×–WOA model
TLP parameter
Chehel-Chai river’s streamflow
url https://doi.org/10.1038/s41598-025-03185-3
work_keys_str_mv AT amingharehbaghi developinganovelhybridmodelbasedongrudeepneuralnetworkandwhaleoptimizationalgorithmforpreciseforecastingofriversstreamflow
AT redvanghasemlounia developinganovelhybridmodelbasedongrudeepneuralnetworkandwhaleoptimizationalgorithmforpreciseforecastingofriversstreamflow
AT farshadahmadi developinganovelhybridmodelbasedongrudeepneuralnetworkandwhaleoptimizationalgorithmforpreciseforecastingofriversstreamflow
AT rasoulmirabbasi developinganovelhybridmodelbasedongrudeepneuralnetworkandwhaleoptimizationalgorithmforpreciseforecastingofriversstreamflow
AT alitorabihaghighi developinganovelhybridmodelbasedongrudeepneuralnetworkandwhaleoptimizationalgorithmforpreciseforecastingofriversstreamflow