Analysis of chemical reactive nanofluid flow on stretching surface using numerical soft computing approach for thermal enhancement

Engineering Applications of Computational Fluid Mechanics, Journal Year: 2024, Volume and Issue: 18(1)

Published: April 17, 2024

In this work, computational intelligence methodologies are used to investigate the trihybrid nanofluid, a new theoretical model with remarkable thermal transmission properties enhance liquid performance. The nanostructure Cu, Al2O3, and TiO2 were immersed in base (C2H6O2) produce (Cu + Al2O3 TiO2/C2H6O2) nanofluid. Darcy-Forchheimer porous medium over stretching Riga sheet, study examines electromagnetic ternary hybrid nanofluid flow under various slip situations. takes into account complex interactions between number of variables, including as viscous dissipation, radiation, heat sources, chemical reactions. Similarity transformations convert partial differential equations for flow, energy, concentration nonlinear ordinary equations. highly problem solved numerically use techniques from bvp4c approach. results method reference dataset required Levenberg-Marquardt backpropagation neural networks (LMBNN). network performance is validated using regression analysis, mean square errors, error histogram data. problem's consistency precision evaluated absolute error, which given each instance at around 10−06–10−08, 10−05–10−10 10−06 05–10−09. order reduce fluid dynamics system's numerical solutions have been taken consideration. Using comparative configurations MSE, histograms, state transitions, correlation, regression, reliability competence stochastic technique verified.

Language: Английский

---

Validation of four optimization evolutionary algorithms combined with artificial neural network (ANN) for landslide susceptibility mapping: A case study of Gilan, Iran

Ecological Engineering, Journal Year: 2024, Volume and Issue: 201, P. 107214 - 107214

Published: Feb. 29, 2024

Language: Английский

---

Evaluation of student failure in higher education by an innovative strategy of fuzzy system combined optimization algorithms and AI

Heliyon, Journal Year: 2024, Volume and Issue: 10(7), P. e29182 - e29182

Published: April 1, 2024

This research suggests two novel metaheuristic algorithms to enhance student performance: Harris Hawk's Optimizer (HHO) and the Earthworm Optimization Algorithm (EWA). In this sense, a series of adaptive neuro-fuzzy inference system (ANFIS) proposed models were trained using these methods. The selection best-fit model depends on finding an excellent connection between inputs output(s) layers in training testing datasets (e.g., combination expert knowledge, experimentation, validation techniques). study's primary result is division participants into performance-based groups (failed non-failed). experimental data used build measured fourteen process variables: relocation, gender, age at enrollment, debtor, nationality, educational special needs, current tuition fees, scholarship holder, unemployment, inflation, GDP, application order, day/evening attendance, admission grade. During evaluation, scoring was created addition mean absolute error (MAE), square (MSE), area under curve (AUC) assess efficacy utilized approaches. Further revealed that HHO-ANFIS superior EWA-ANFIS. With AUC = 0.8004 0.7886, MSE 0.62689 0.65598, MAE 0.64105 0.65746, failure pupils assessed with most significant degree accuracy. MSE, MAE, precision indicators showed EWA-ANFIS less accurate, having amounts 0.71543 0.71776, 0.70819 0.71518, 0.7565 0.758. It found optimization have high ability increase accuracy performance conventional ANFIS predicting students' performance, which can cause changes management improve quality academic programs.

Language: Английский

---

Utilization of a micro-squeezed channel in drug delivery through hybrid nanofluids under Hall current, nanoparticle radius, and interparticle spacing effects

International Journal of Numerical Methods for Heat &amp Fluid Flow, Journal Year: 2025, Volume and Issue: unknown

Published: May 19, 2025

Purpose The purpose of this study is to focus on the laminar flow blood-based (Ag-TiO 2 ) Darcy−Forchheimer hybrid nanofluids (DF-HNF) in an artificial micro-squeezed channel confined by two porous walls. Design/methodology/approach This model could be advantageous for medical diagnostics using intelligence (AI), particularly drug delivery and biomedical mechanisms. nanofluid can infiltrate escape both sides with similar widths while dilating expanding at a constant speed due medium flow. control volume finite element method (CVFEM) neural network (ANN) are used solve governing equations. Findings Investigating effects nanoparticle radius interparticle spacing, allowing uniform distribution. CVFEM examined microscopic view DF-HNF based blood, achieving best validation performance 250 epochs mean squared error 4.356 × 10 −9 Nusselt number prediction, confirming strong correlation between key heat transfer parameters output. Parameters examined, reasoning physical explanations beyond them also shown. current results compared existing literature. Research limitations/implications simulation helpful experimental analysis. Other than Ag TiO nanoparticles, idea further research. Practical implications In field, compression extension essential transporting blood delivering targeted drugs. intends concentrate gradual movement within that surrounded Social Targeted important phenomenon field engineering. advanced AI-based analysis achieve targets. These phenomena requirements each human. Originality/value particles ionic, addition Hall innovative make more natural, which has been study. space diameter flexibility novel approach analyze ANN new contributions

Language: Английский

---

A review of the satellite remote sensing techniques for assessment of runoff and sediment in soil erosion

Journal of Hydrology and Hydromechanics, Journal Year: 2024, Volume and Issue: 72(2), P. 252 - 267

Published: May 9, 2024

Abstract Soil erosion monitoring is essential for the ecological evaluation and dynamic of land resources via remote sensing technology. In this paper, we provide new insights into existing problems development directions traditional models, which are supported by technologies. An important role played information acquisition technology in qualitative quantitative soil erosion, data technical support provided systematically reviewed. We a detailed overview research progress associated with empirical statistical models physically driven process limitations their application also summarized. The preliminary integration sources high spatial temporal resolution technologies enables high-precision estimation sediment transport trajectories, watershed river network density, terrain slope, enhancing accuracy factor identification, such as spectral feature recognition from information, gully extraction, vegetation coverage estimation. However, current algorithms not comprehensive enough, particularly terms extraction there applicability accurate models.

Language: Английский

---

Fluvial bedload transport modelling: advanced ensemble tree-based models or optimized deep learning algorithms?

Engineering Applications of Computational Fluid Mechanics, Journal Year: 2024, Volume and Issue: 18(1)

Published: May 10, 2024

The potential of advanced tree-based models and optimized deep learning algorithms to predict fluvial bedload transport was explored, identifying the most flexible accurate algorithm, optimum set readily available reliable inputs. Using 926 datasets for 20 rivers, performance three groups tested: (1) standalone Alternating Model Tree (AMT) Dual Perturb Combine (DPCT); (2) ensemble Iterative Absolute Error Regression (IAER), ensembled with AMT DPCT; (3) Long Short-Term Memory (LSTM) Recurrent Neural Network (RNN) Grey Wolf Optimizer. Comparison predictive that commonly used empirical equations sensitivity analysis driving variables revealed that: (i) coarse grain-size percentile D90 effective variable in prediction (where Dx is xth bed surface grain size distribution), followed by D84, D50, flow discharge, D16, channel slope width; (ii) all displayed 'very good' or 'good' performance, outperforming equations; (iii) performed best when input parameters were used. Thus, a range different combinations must be considered optimization these models. Overall, provided more predictions than their counterpart. In particular, model IAER-AMT strongly, displaying great produce robust coarse-grained rivers based on few variables.

Language: Английский

---

Novel embedding model predicting the credit card's default using neural network optimized by harmony search algorithm and vortex search algorithm

Heliyon, Journal Year: 2024, Volume and Issue: 10(9), P. e30134 - e30134

Published: April 23, 2024

In today's banking and financial system, using a credit card has become indispensable. The industry existed due to shift in consumer preferences rise national economic growth. number of issuing banks, issuers, transaction volumes increased significantly. Nevertheless, owing the growth transactions made with cards, both total amount rate defaults on loans have issues that cannot be neglected. This issue must resolved ensure continued prosperous years come. Currently, few optimization algorithms—Whale algorithm (WOA), Harmony Search (HS), Multi-verse (MVO), Vortex (VS)—have been used achieve this purpose. However, because default data is volatile unequal, it challenging for typical algorithms offer steady approaches optimal performance. Studies indicated optimizing suitable properties can significantly improve To performance, some tuning was applied ANN. study will assess twenty-three parameters, efficacy all four compared ROC AUC evaluations. suggested model's performance contrasted scenario where classifiers were trained original data. contrast, values VS-MLP 0.7407 0.7271, while those HS-MLP 0.7074 0.6997. training testing phases, 0.7469 0.7329 from MVO-MLP 0.72 0.7185 WOA-MLP, respectively. results show accuracy HS, VSA, MVO, WOA are similar; MVO highest accuracy. benefit methodology, which may help resolve probabilities.

Language: Английский

---

Optimized phased planning for dynamic rehabilitation of integrated municipal infrastructure

AQUA - Water Infrastructure Ecosystems and Society, Journal Year: 2024, Volume and Issue: 73(7), P. 1389 - 1405

Published: July 1, 2024

ABSTRACT Phased planning for municipal infrastructure is based on the time-dependent status of multiple networks, which in contrast to traditional approach, where one-phase construction and a single are considered system activities. This study integrates optimizes corridor-wise intervention water, sewer, road networks number equally long phases decisions among decision variables showing extent phase optimization can impact cost coordination interventions interdependent systems. Optimizing within an evolutionary algorithm challenging task due recombination between numerous solutions with different variable lengths. A multi-phase design approach developed rehabilitation real case Montreal, Canada. The involves 20 corridors street section co-located water sewer pipes. metaheuristic single-objective engine employed minimize total net present value plan costs whole integrated system. results show that phased could bring about 25% saving master coordinated multi-systems

Language: Английский

---

Modeling the total hardness (TH) of groundwater in aquifers using novel hybrid soft computing optimizer models

Environmental Earth Sciences, Journal Year: 2024, Volume and Issue: 83(13)

Published: June 12, 2024

Language: Английский

---

Impact of fuzzy volume fraction on unsteady stagnation-point flow and heat transfer of a third-grade fuzzy hybrid nanofluid over a permeable shrinking/stretching sheet

Engineering Applications of Computational Fluid Mechanics, Journal Year: 2024, Volume and Issue: 18(1)

Published: July 26, 2024

In current work, the unsteady stagnation point's flow on a special third-grade fuzzy hybrid (Al2O3 + Cu/SA) nanofluid (HNF) through permeable convective shrinking/stretching sheet has been scrutinized. addition, adverse consequences of heat source, viscous dissipation, nonlinear thermal radiation, and nanoparticle volume fraction are likewise taken into consideration. Non-linear coupled partial differential equations (PDEs) get transformed ordinary (ODEs) using an effective similarity transformation. After that, ODEs numerically solved bvp4c algorithm. Regarding validation, present results align with earlier published research. The effects distribution, rate, Nusselt number, skin friction coefficient dynamics explored graphical tabular forms. is considered triangular number (TFN) [0, 5%, 10%]. With use TFNs, (FDEs). TFNs controlled widely used ζ - cut technique cut∈[0,1], which requires minimal computational effort to examine their dynamical performance. Also, comparison Al2O3/SA, Cu/SA Al2O3 membership functions (MFs). MFs show that in terms rate transfer better than both Al2O3/SA nanofluids.

Language: Английский

---