Designing a decomposition-based multi-phase pre-processing strategy coupled with EDBi-LSTM deep learning approach for sediment load forecasting

Ecological Indicators, Год журнала: 2023, Номер 153, С. 110478 - 110478

Опубликована: Июнь 13, 2023

Forecasting accurately suspended sediment load (SSL) in the basin is one of most critical issues for river engineering, environment, and water resources management which effectively reduces flood damages. In this study, a new multi-criteria hybrid expert system comprised empirical wavelet decomposition (EWT) integrated with Encoder-Decoder Bidirectional long short-term memory (EDBi-LSTM), supported by five feature selection (FS) methods was developed first time to forecast daily SSL at two study sites (Bamini Ashti) Godavari basin, India. The employed FS schemes are including Boruta-Random forest (BRF), simulated annealing (SA), Relief algorithm, Ridge regression (RR), Mutual information (MI) where BRF coupled EWT EDBi-LSTM (i.e., EWT-EDBi-LSTM-Boruta) identified as main forecasting paradigm. Here original signals monsoon season (2001–2015) only input were considered events scale both zones. decomposed using technique considering significant antecedent time-lagged inputs based on partial auto-correlation function (PACF). next stage, strategies addressed specify sub-sequences reduce computational cost enhance accuracy. Besides, extreme gradient boosting (XGB) approach implemented compare potential standalone counterpart models sites. According several goodness-of-fit indices validation tools, outcomes Bamini Ashti demonstrated that EWT-EDBi-LSTM-Boruta model, achieved best accuracy, followed EWT-XGB-Boruta, EWT-EDBi-LSTM-SA, EWT-XGB-SA, respectively. Comparing all showed BRF, SA, RR performed better integration machine learning (ML) models.

Язык: Английский

---

A novel global solar exposure forecasting model based on air temperature: Designing a new multi-processing ensemble deep learning paradigm

Expert Systems with Applications, Год журнала: 2023, Номер 222, С. 119811 - 119811

Опубликована: Март 8, 2023

Язык: Английский

---

Computation of energy across the type-C piano key weir using gene expression programming and extreme gradient boosting (XGBoost) algorithm

Energy Reports, Год журнала: 2023, Номер 9, С. 310 - 321

Опубликована: Апрель 15, 2023

An accurate assessment of energy loss across dams and weir systems is a critical technical monetary remedy for understanding the hydraulic system's downstream morphology during flood. Using standard empirical formulas, accurately estimating different devices tedious challenging procedure. Consequently, new concise techniques remain highly sought after. This paper presents two models based on Gene Expression Programming (GEP) Extreme Gradient Boosting (XGBoost) to examine losses throughout type-C PKW. The have been developed consider five non-dimensional parameters, viz L/W, Ht/P, Wi/Wo, N, Si/So, that influence over significantly. were created using experimental data from wide range residual release capacities. Additionally, adequacy constructed GEP XGBoost was assessed RMSE (root mean square error) statistical variables coefficient (R2). As per outcomes, model beats with determination (R2) = 0.999, MAE 0.0062, MAPE 1.4% 0.0012 in training stage R2=0.998, 0.001, 2.1, 0.001 testing data. These findings show algorithm more than algorithms this study prediction

Язык: Английский

---

Boruta extra tree-bidirectional long short-term memory model development for Pan evaporation forecasting: Investigation of arid climate condition

Alexandria Engineering Journal, Год журнала: 2023, Номер 86, С. 425 - 442

Опубликована: Дек. 7, 2023

In this study, two deep learning approaches, bidirectional long short-term memory (BiLSTM) and (LSTM), were used along with adaptive boosting general regression neural network to forecast multi-step-ahead pan evaporation in arid climate stations Iran (Ahvaz Yazd). Lagged time series of meteorological data input the machine models. Two feature selection methods, i.e., Boruta extra tree XGBoost, select significant inputs reduce number model complexity. Different statistical metrics investigate performance. The results demonstrated that Boruta-extra-tree-based models more accurate than XGBoost-based Compared techniques, combination BiLSTM enabled one-day-ahead forecasting for both sites (Root Mean Square Error (RMSE) = 1.6857, Ahvaz station, RMSE 1.3996 Yazd station). proposed was up 30 days ahead stations. showed Boruta-BiLSTM could accurately

Язык: Английский

---

Hybrid Machine Learning Models for Soil Saturated Conductivity Prediction

Water, Год журнала: 2022, Номер 14(11), С. 1729 - 1729

Опубликована: Май 27, 2022

The hydraulic conductivity of saturated soil is a crucial parameter in the study any engineering problem concerning groundwater. Hydraulic mainly depends on particle size distribution, compaction, and properties that influence aggregation water retention. Generally, finding simple accurate analytical equations between characteristics which it very hard task. Machine learning algorithms can provide excellent tools for tackling highly nonlinear regression problems. Additionally, hybrid models resulting from combination multiple machine further improve accuracy predictions. Five different were built to predict using dataset extracted Soil Water Infiltration Global database. based predictors. Seven variants each model compared, replacing implemented algorithm. Three individual models, while four models. employed Multilayer Perceptron, Random Forest, Support Vector Regression. largest number predictors led most In addition, across all three hybridized Forest Regression proved be (R2 values up 0.829). However, showed tendency overestimate soils where low.

Язык: Английский

---

Developing ensemble models for estimating sediment loads for different times scales

Environment Development and Sustainability, Год журнала: 2023, Номер 26(6), С. 15557 - 15575

Опубликована: Апрель 21, 2023

Язык: Английский

---

River water temperature prediction using hybrid machine learning coupled signal decomposition: EWT versus MODWT

Ecological Informatics, Год журнала: 2023, Номер 78, С. 102376 - 102376

Опубликована: Ноя. 14, 2023

Язык: Английский

---

Discharge Estimation over Piano Key Weirs: A Review of Recent Developments

Water, Год журнала: 2022, Номер 14(19), С. 3029 - 3029

Опубликована: Сен. 26, 2022

The piano key (PK) weir has advanced over the labyrinth to increase discharge capacity. Piano weirs exhibit nonlinear flow behavior and are easy place on existing spillway or newly constructed dam with less base area. Various investigators given equations calculate coefficient for free submerged conditions. study focuses reviewing impacts of PK geometry coefficient, including length height, upstream downstream widths, apex overhangs. In this study, all possible aspects were briefly reviewed. From sensitivity analysis, it is observed that more sensitive L/W dimensionless ratio followed by B/P ratio. L total crest, W width weir, B P height weir. This review paper intended serve as an accessible resource hydraulic structures researchers engineering professionals alike interested in hydraulics weirs.

Язык: Английский

---

Surface water electrical conductivity and bicarbonate ion determination using a smart hybridization of optimal Boruta package with Elman recurrent neural network

Process Safety and Environmental Protection, Год журнала: 2023, Номер 174, С. 115 - 134

Опубликована: Март 29, 2023

Язык: Английский

---

Enhancing Meteorological Drought Modeling Accuracy Using Hybrid Boost Regression Models: A Case Study from the Aegean Region, Türkiye

Sustainability, Год журнала: 2023, Номер 15(15), С. 11568 - 11568

Опубликована: Июль 26, 2023

The impact of climate change has led to significant changes in hydroclimatic patterns and continuous stress on water resources through frequent wet dry spells. Hence, understanding effectively addressing the escalating patterns, especially context meteorological drought, necessitates precise modeling these phenomena. This study focuses assessing accuracy drought using well-established Standard Precipitation Index (SPI) Aegean region Türkiye. utilizes monthly precipitation data from six stations Cesme, Kusadasi, Manisa, Seferihisar, Selcuk Izmir at Kucuk Menderes Basin covering period 1973 2020. dataset is divided into three sets, training (60%), validation (20%), testing (20%) sets. aims determine SPI-3, SPI-6 SPI-12 a multi-station prediction technique. Three boosting regression models (BRMs), namely Extreme Gradient Boosting (XgBoost), Adaptive (AdaBoost), (GradBoost), were employed optimized with help Weighted Mean Vectors (INFO) Model performances then evaluated Root Square Error (RMSE), Absolute (MAE), Percentage (MAPE), Coefficient Determination (R2) Willmott (WI). Results demonstrated distinct superiority XgBoost model over AdaBoost GradBoost terms accuracy. During test phase, achieved RMSEs 0.496, 0.429 0.389 for SPI-12, respectively. WIs 0.899, 0.901 0.825 These are considerably lower than corresponding values obtained by other models. Yet, comparative statistical analysis further underscores effectiveness extended periods

Язык: Английский

---