Comparison of LSTM and SVM methods through wavelet decomposition in drought forecasting

Earth Science Informatics, Год журнала: 2025, Номер 18(1)

Опубликована: Янв. 1, 2025

Язык: Английский

---

Coupling SWAT and Transformer Models for Enhanced Monthly Streamflow Prediction

Sustainability, Год журнала: 2024, Номер 16(19), С. 8699 - 8699

Опубликована: Окт. 9, 2024

The establishment of an accurate and reliable predictive model is essential for water resources planning management. Standalone models, such as physics-based hydrological models or data-driven have their specific applications, strengths, limitations. In this study, a hybrid (namely SWAT-Transformer) was developed by coupling the Soil Water Assessment Tool (SWAT) with Transformer to enhance monthly streamflow prediction accuracy. SWAT first constructed calibrated, then its outputs are used part inputs Transformer. By correcting errors using Transformer, two effectively coupled. Monthly runoff data at Yan’an Ganguyi stations on Yan River, first-order tributary Yellow River Basin, were evaluate proposed model’s performance. results indicated that performed well in predicting high flows but poorly low flows. contrast, able capture low-flow period information more accurately outperformed overall. SWAT-Transformer could correct predictions overcome limitations single model. integrating SWAT’s detailed physical process portrayal Transformer’s powerful time-series analysis, coupled significantly improved offer optimal resource management, which crucial sustainable economic societal development.

Язык: Английский

---

Multimodal deep learning models incorporating the adsorption characteristics of the adsorbent for estimating the permeate flux in dynamic membranes

Journal of Membrane Science, Год журнала: 2024, Номер 709, С. 123105 - 123105

Опубликована: Июль 18, 2024

Язык: Английский

---

Spatial prediction of groundwater salinity in multiple aquifers of the Mekong Delta region using explainable machine learning models

Water Research, Год журнала: 2024, Номер 266, С. 122404 - 122404

Опубликована: Сен. 6, 2024

Язык: Английский

---

Evaluating Performances of LSTM, SVM, GPR, and RF for Drought Prediction in Norway: A Wavelet Decomposition Approach on Regional Forecasting

Water, Год журнала: 2024, Номер 16(23), С. 3465 - 3465

Опубликована: Дек. 2, 2024

A serious natural disaster that poses a threat to people and their living spaces is drought, which difficult notice at first can quickly spread wide areas through subtle progression. Numerous methods are being explored identify, prevent, mitigate distinct metrics have been developed. In order contribute the research on measures be taken against Standard Precipitation Evaporation Index (SPEI), one of drought indices has developed accepted in recent years includes more comprehensive definition, was chosen this study. Machine learning deep algorithms, including support vector machine (SVM), random forest (RF), long short-term memory (LSTM), Gaussian process regression (GPR), were used model droughts six regions Norway: Bodø, Karasjok, Oslo, Tromsø, Trondheim, Vadsø. Four architectures employed for goal, as novel approach, models’ output enhanced by using discrete wavelet decomposition/transformation (WT). The outputs evaluated correlation coefficient (r), Nash–Sutcliffe efficiency (NSE), root mean square error (RMSE) performance evaluation criteria. When findings analyzed, GPR (W-GPR), acquired after WT, typically produced best results. Furthermore, it discovered that, out all recognized models, M04 had most effective structure. Consequently, successful outcomes obtained with W-SVM-M04 Bodø W-GPR-M04 Oslo region results across (r: 0.9983, NSE: 0.9966 RMSE:0.0539).

Язык: Английский

---

Temporal fusion transformer model for predicting differential pressure in reverse osmosis process

Journal of Water Process Engineering, Год журнала: 2025, Номер 70, С. 106914 - 106914

Опубликована: Янв. 9, 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

---

Modeling the First Flush Effect of Urban Micropollutants with Sensitivity Analysis and Uncertainty Analysis

Опубликована: Янв. 1, 2025

Язык: Английский

---

Feasibility study of real-time virtual sensing for water quality parameters in river systems using synthetic data and deep learning models

Journal of Environmental Management, Год журнала: 2025, Номер 380, С. 125191 - 125191

Опубликована: Апрель 1, 2025

Язык: Английский

---

A hybrid approach to improvement of watershed water quality modeling by coupling process–based and deep learning models

Water Environment Research, Год журнала: 2024, Номер 96(8)

Опубликована: Авг. 1, 2024

Watershed water quality modeling to predict changing is an essential tool for devising effective management strategies within watersheds. Process-based models (PBMs) are typically used simulate modeling. In watershed utilizing PBMs, it crucial effectively reflect the actual conditions by appropriately setting model parameters. However, parameter calibration and validation time-consuming processes with inherent uncertainties. Addressing these challenges, this research aims address various challenges encountered in of PBMs. To achieve this, development a hybrid model, combining uncalibrated PBMs data-driven (DDMs) such as deep learning algorithms proposed. This intended enhance integrating strengths both DDMs. The constructed coupling Soil Water Assessment Tool (SWAT) Long Short-Term Memory (LSTM). SWAT, representative PBM, using geographical information 5-year observed data from Yeongsan River Watershed. output variables streamflow, suspended solids (SS), total nitrogen (TN), phosphorus (TP), well precipitation day previous day, training TP load. For comparison, conventional SWAT calibrated validated results revealed that load simulated predicted better than model. Also, reflects seasonal variations load, including peak events. Remarkably, when applied other sub-basins without specific training, consistently outperformed conclusion, application SWAT-LSTM could be useful decreasing uncertainties improving overall predictive performance PRACTITIONER POINTS: We aimed process-based water-quality Tool-Long model's (TP) matched TP. It exhibited superior sub-basins. will overcome constraints also enable more efficient

Язык: Английский

---