Runoff prediction using a multi-scale two-phase processing hybrid model

Stochastic Environmental Research and Risk Assessment, Год журнала: 2025, Номер unknown

Опубликована: Янв. 19, 2025

Язык: Английский

---

A runoff prediction approach based on machine learning, ensemble forecasting and error correction: A case study of source area of Yellow River

Journal of Hydrology, Год журнала: 2025, Номер unknown, С. 133190 - 133190

Опубликована: Март 1, 2025

Язык: Английский

---

Mountain Flood Level Forecasting in Small Watersheds Based on Recurrent Neural Networks and Multi-Dimensional Data

IEEE Access, Год журнала: 2024, Номер 12, С. 82706 - 82719

Опубликована: Янв. 1, 2024

Язык: Английский

---

Monthly rainfall prediction model based on VMD-PSO-BiLSTM-case study: Handan City, China

Theoretical and Applied Climatology, Год журнала: 2025, Номер 156(2)

Опубликована: Фев. 1, 2025

Язык: Английский

---

Research on monthly runoff prediction model considering secondary decomposition of multiple fitness functions and deep learning

Geomatics Natural Hazards and Risk, Год журнала: 2025, Номер 16(1)

Опубликована: Фев. 27, 2025

Язык: Английский

---

A monthly runoff prediction model based on ICEEMD-L-SHADE-SRU

Journal of Freshwater Ecology, Год журнала: 2025, Номер 40(1)

Опубликована: Апрель 8, 2025

Язык: Английский

---

Using a seasonal and trend decomposition algorithm to improve machine learning prediction of inflow from the Yellow River, China, into the sea

Frontiers in Marine Science, Год журнала: 2025, Номер 12

Опубликована: Май 9, 2025

The Yellow River is the largest inflow into Bohai Sea, and its changes directly affect ecological environment marine health of Sea. Therefore, accurate prediction crucial for maintaining balance Sea protecting resources. Time decomposition algorithms, combined with machine learning, are effective tools to enhance capabilities models. However future data leakage from items was ignored in many studies. It necessary develop right method operate time avoid leakage. In this study, sea predicted based on a learning model (light gradient boosting machine, LightGBM) algorithm (seasonal trend using loess, STL), different ways STL were evaluated. results showed that overall performance STL–LightGBM better than LightGBM model. took historical 8 days as input, average NSE next 1–7 would reach 0.720. Even when forecast period 7 days, (NSE: 0.549 7-day lead time) 0.105 higher 0.444 time). We found pretreatment entire test set overestimated true STL–LightGBM. recommended preprocesses each sample study can provide help water resources management offshore environmental management.

Язык: Английский

---

Comprehensive objective function- guided decomposition-prediction co-optimization framework: Enhanced Transformer model for high-accuracy forecasting of non-stationary runoff

Journal of Hydrology Regional Studies, Год журнала: 2025, Номер 60, С. 102482 - 102482

Опубликована: Май 21, 2025

Язык: Английский

---

Enhanced successive variational mode decomposition with improved fitness function and optimizer for noise analysis of electric vibrators

Journal of Mechanical Science and Technology, Год журнала: 2025, Номер unknown

Опубликована: Май 30, 2025

Язык: Английский

---

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

Scientific Reports, Год журнала: 2025, Номер 15(1)

Опубликована: Июнь 3, 2025

Streamflow contemplates a fundamental criterion to evaluate the impact of human activities and climate changes on hydrological cycle. In this study, novel innovative deep neural network (DNN) structure by integrating double Gated Recurrent Units (GRU) model with multiplication layer meta-heuristic whale optimization algorithm (WOA) (i.e., hybrid 2GRU×-WOA model) is developed improve prediction accuracy performance mean monthly Chehel-Chai River's streamflow (CCRSFm) in Iran. The Pearson's correlation coefficient (PCC) Cosine Amplitude Sensitivity (CAS) as feature (input) selection process determine only precipitation (Pm) most effective input variable among list on-site potential time series parameters recorded study area. Thanks well-proportioned structural framework suggested model, it leads an appropriate total learnable parameter (TLP) compared standard individual GRU Bi-GRU benchmark models comparable meta-parameters. This under optimal meant meta-parameters tuned i.e., coupling state activation functions (SAF) tanh-softsign, dropout rate (P-rate) 0.5, numbers hidden neurons (NHN) 70, outperforms R2 0.79, NSE 0.76, MAE 0.21 (m3/s), MBE -0.11(m3/s), RMSE 0.36 (m3/s). Hybridizing 2GRU× WOA causes increase value 6.8% reduce 20.4%. Comparatively, result 0.59 0.66, 0.55 0.6, 0.91 0.53 0.047 - 0.06 1.29 0.83 respectively.

Язык: Английский

---