Identification of influential weather parameters and seasonal drought prediction in Bangladesh using machine learning algorithm

Scientific Reports, Journal Year: 2024, Volume and Issue: 14(1)

Published: Jan. 4, 2024

Abstract Droughts pose a severe environmental risk in countries that rely heavily on agriculture, resulting heightened levels of concern regarding food security and livelihood enhancement. Bangladesh is highly susceptible to hazards, with droughts further exacerbating the precarious situation for its 170 million inhabitants. Therefore, we are endeavouring highlight identification relative importance climatic attributes estimation seasonal intensity frequency Bangladesh. With period forty years (1981–2020) weather data, sophisticated machine learning (ML) methods were employed classify 35 agroclimatic regions into dry or wet conditions using nine parameters, as determined by Standardized Precipitation Evapotranspiration Index (SPEI). Out 24 ML algorithms, four best methods, ranger, bagEarth, support vector machine, random forest (RF) have been identified prediction multi-scale drought indices. The RF classifier Boruta algorithms shows water balance, precipitation, maximum minimum temperature higher influence occurrence across trend spatio-temporal analysis indicates, has decreased over time, but return time increased. There was significant variation changing spatial nature intensity. Spatially, shifted from northern central southern zones Bangladesh, which had an adverse impact crop production rural urban households. So, this precise study important implications understanding how mitigate impacts. Additionally, emphasizes need better collaboration between relevant stakeholders, such policymakers, researchers, communities, local actors, develop effective adaptation strategies increase monitoring meticulous management

Language: Английский

---

Scour depth prediction around bridge abutments: A comprehensive review of artificial intelligence and hybrid models

Physics of Fluids, Journal Year: 2025, Volume and Issue: 37(2)

Published: Feb. 1, 2025

Scouring around the bridge structure is a major concern of globe. Therefore, precise estimation scour depth essential to minimize failure and provide preventive measures. This review paper aims analyze critical various artificial intelligence (AI) techniques utilized in literature estimate abutment including neural networks (ANN), adaptive neuro-fuzzy inference systems (ANFIS), gene expression programming (GEP), support vector machines (SVM), extreme learning (ELM). The predictive power each technique was assessed terms different performance indicators, such as correlation coefficient (R), mean square error (MSE), predicted values, Taylor's diagram, sensitivity analysis, violin plot. highlights that by comparing AI techniques, ELM GEP have superior performance, especially predicting dealing with complex large datasets. However, limitations proposed solutions been reported for ANN, ANFIS, SVM, group method data handling (GMDH). main challenges GMDH were overfitting hyperparameter tuning. Based on technique, current found satisfactory because its computation speed capability. Moreover, would be helpful researchers working field hydraulics engineering, particularly scouring abutment.

Language: Английский

---

Machine learning prediction of pyrolytic products of lignocellulosic biomass based on physicochemical characteristics and pyrolysis conditions

Bioresource Technology, Journal Year: 2022, Volume and Issue: 367, P. 128182 - 128182

Published: Oct. 25, 2022

Language: Английский

---

Multi-step daily forecasting of reference evapotranspiration for different climates of India: A modern multivariate complementary technique reinforced with ridge regression feature selection

Agricultural Water Management, Journal Year: 2022, Volume and Issue: 272, P. 107812 - 107812

Published: July 30, 2022

Language: Английский

---

A Comprehensive Experimental and Computational Investigation on Estimation of Scour Depth at Bridge Abutment: Emerging Ensemble Intelligent Systems

Water Resources Management, Journal Year: 2023, Volume and Issue: 37(9), P. 3745 - 3767

Published: May 17, 2023

Language: Английский

---

Prediction of maximum scour depth at clear water conditions: Multivariate and robust comparative analysis between empirical equations and machine learning approaches using extensive reference metadata

Journal of Environmental Management, Journal Year: 2024, Volume and Issue: 354, P. 120349 - 120349

Published: Feb. 24, 2024

Language: Английский

---

Air quality monitoring based on chemical and meteorological drivers: Application of a novel data filtering-based hybridized deep learning model

Journal of Cleaner Production, Journal Year: 2022, Volume and Issue: 374, P. 134011 - 134011

Published: Sept. 8, 2022

Language: Английский

---

Development of an enhanced bidirectional recurrent neural network combined with time-varying filter-based empirical mode decomposition to forecast weekly reference evapotranspiration

Agricultural Water Management, Journal Year: 2023, Volume and Issue: 290, P. 108604 - 108604

Published: Nov. 24, 2023

Evapotranspiration is one of agricultural water management's most significant and impactful hydrologic processes. A new multi-decomposition deep learning-based technique proposed in this study to forecast weekly reference evapotranspiration (ETo) western coastal regions Australia (Redcliffe Gold Coast). The time-varying filter-based empirical mode decomposition (TVF-EMD) was used first break down the original meteorological variables/signals into intrinsic functions (IMFs), which included maximum minimum temperature, relative humidity, wind speed, solar radiation. Using a partial autocorrelation function (PACF), lagged values were then calculated from decomposed sub-sequences (i.e., IMFs). novel Extra Tree- Boruta feature selection algorithm extract important features IMFs. Four machine learning approaches, including bidirectional recurrent neural network (Bi-RNN), multi-layer perceptron (MLP), random forest (RF), extreme gradient boosting (XGBoost), using TVF-EMD-based data. Different statistical metrics applied evaluate model performances. results showed that input data by TVF-EMD significantly improved accuracy compared with non-decomposed inputs (single models without decomposition). findings indicate TVF-BiRNN model, as presented, achieved highest level simulating ET0 at both Redcliffe Coast stations (Redcliffe: R=0.9281, RMSE=3.8793 mm/week, MAPE = 9.2010%; Coast: R=0.8717, RMSE=4.1169 11.5408%). hybrid modeling can potentially improve management through its ability generate more accurate ETo estimates weekly. methodology exhibits potential applicability various other environmental hydrological issues.

Language: Английский

---

Machine Learning Guided Strategies to Develop High Efficiency Indoor Perovskite Solar Cells

Advanced Theory and Simulations, Journal Year: 2024, Volume and Issue: 7(5)

Published: March 20, 2024

Abstract Indoor Perovskite Solar Cells (IPSCs) have recently gathered massive research attention, driven by their promising role in powering the continuously expanding Internet of Things (IoT) devices and simultaneous advancements solar field. To further accelerate development IPSCs, a machine learning (ML) approach to assist advancement IPSCs is proposed current study. Here, ML model predict most important performance parameters such as short circuit ( J SC ), open voltage V OC fill factor (FF), power conversion efficiency (PCE) under various light sources intensities presented. This developed can effectively performances (PSCs) operated indoor illumination close true/experimental values. The factors affecting IPSC Correlation matrix SHAPley analysis are also analyzed. These findings demonstrate that provides accurate predictions , FF, PCE ultimately contributing optimization cell environments renewable energy technology.

Language: Английский

---

Forecasting Daily Flood Water Level Using Hybrid Advanced Machine Learning Based Time-Varying Filtered Empirical Mode Decomposition Approach

Water Resources Management, Journal Year: 2022, Volume and Issue: 36(12), P. 4637 - 4676

Published: Aug. 1, 2022

Language: Английский

---