A three layer stacked multimodel transfer learning approach for deep feature extraction from Chest Radiographic images for the classification of COVID-19

Engineering Applications of Artificial Intelligence, Journal Year: 2025, Volume and Issue: 147, P. 110241 - 110241

Published: Feb. 25, 2025

Language: Английский

---

Peristaltic flow of electromagnetic tri-hybrid Carreau nanofluid using backpropagated Levenberg–Marquardt technique: an entropy generation analysis in blood cells

Electromagnetic Biology and Medicine, Journal Year: 2025, Volume and Issue: unknown, P. 1 - 19

Published: Feb. 27, 2025

The present research concentrates on examining entropy generation during the flow phenomenon of a three-dimensional peristaltic motion magnetized tri-hybrid nanofluid within curved rectangular duct using machine learning technique called backpropagated Levenberg-Marquardt (BLMT). Carreau constitutive model is used for base liquid (blood). To obtain most accurate solutions governing equations, an analytical tool Homotopy Perturbation Method (HPM) utilized along with methodology ANN-BLMT method MatLab. data HPM and are also compared to assess how framework partial differential equations (PDEs) occurring in problem can be improved. It shows highest correlations between output prediction method. convergence analysis reveals that two scenarios, velocity exhibits best validation performance values around 7.3117×10-11 1.0082×10-10. A detailed comparison blood has been presented graphically enhance benefits ternary hybrid nanoparticles simple fluid. found slowed by curvature increase because increment pure blood. noted rate heat transfer nanofluids greater than Research findings have obvious implications comprehending enhancing dynamics biological processes such as intestinal tract.

Language: Английский

---

Problem optimization of ray tracing through the crystalline lens of the eye with an artificial neural network and Grey Wolf optimizer

Communications in Nonlinear Science and Numerical Simulation, Journal Year: 2025, Volume and Issue: unknown, P. 108733 - 108733

Published: March 1, 2025

Language: Английский

---

Applications to medical and failure time data: Using a new extension of the extended exponential model

Journal of Radiation Research and Applied Sciences, Journal Year: 2025, Volume and Issue: 18(2), P. 101373 - 101373

Published: March 6, 2025

Language: Английский

---

An updated software reliability model using the shanker model and failure data

Quality and Reliability Engineering International, Journal Year: 2024, Volume and Issue: 40(4), P. 2078 - 2095

Published: Feb. 26, 2024

Abstract Software developers' goal is to develop reliable and superior software. Due the fact that software errors frequently generate large societal or financial losses, reliability essential. growth models are a widely used technique for assessment. This study examines various nonhomogeneous Poisson process with newly developed distribution evaluates unknown model parameters based on frequentist Bayesian methods of estimation. Finally, we conduct evaluations real datasets using variety evaluation criteria compare results previous show how proposed may be applied under both approaches in practical setting. According this study, innovative model's mean square error, R 2 , bias, predicted relative variation, Theil statistic, error prediction values lowest approach data sets II IV, perform well set I. These implementation findings demonstrate effectiveness our specific examination failure data.

Language: Английский

---

Validation Challenges in Data for Different Diesel Engine Performance Regimes Utilising HVO Fuel: A Study on the Application of Artificial Neural Networks for Emissions Prediction

Machines, Journal Year: 2024, Volume and Issue: 12(4), P. 279 - 279

Published: April 21, 2024

Artificial neural networks (ANNs) provide supervised learning via input pattern assessment and effective resource management, thereby improving energy efficiency predicting environmental fluctuations. The advanced technique of ANNs forecasts diesel engine emissions by collecting measurements during trial sessions. This study included experimental sessions to establish technical ecological indicators for a across several operational scenarios. VALLUM01, novel tool, has been created with user-friendly interface data input/output, intended the purposes testing prediction. There was comprehensive collection 12 parameters 10 output that were identified as relevant sufficient objectives training, validation, proper value ranges transforming into fuzzy sets input/output an ANN found. Given ANN’s training session comprises 1,000,000 epochs 1000 perceptrons within single-hidden layer, its effectiveness can be considered high. Many statistical distributions, including Pearson, Spearman, Kendall, validate prediction accuracy. accuracy from 96% on average, in some instances, it may go up 99%.

Language: Английский

---

A neuro‐computational study of viscous dissipation and nonlinear Arrhenius chemical kinetics during the hypodicarbonous acid‐based hybrid nanofluid flow past a Riga plate

ZAMM ‐ Journal of Applied Mathematics and Mechanics / Zeitschrift für Angewandte Mathematik und Mechanik, Journal Year: 2024, Volume and Issue: 104(9)

Published: July 17, 2024

Abstract We examine the flow of Casson hybrid nanofluid (Cu+/) through a Riga plate sensor with perforations that act as an electromagnetic actuator. The hypodicarbonous acid is considered base fluid. impact Arrhenius chemical kinetics and viscous dissipation are taken into account during dynamics. problem formulated by considering heat mass transfer. An appropriate scaling used to reduce complexity problem, further transform it system ordinary differential equations (ODEs). reduced set for first‐order analyzed Artificial Neural Network (ANN) which trained Levenberg–Marquardt algorithm. results state variables displayed graphs tables performing 1000 independent iterations tolerance . Hartman, Casson, Richardson numbers their increasing values enhance velocity profile. reaction parameter Prandtl number decline thermal concentration profiles, respectively. Statistical analysis in form regression histograms also carried out each case. absolute error (AE) ranges up validations range presented varying parameter. A comparative (NF) (HNF) performed case study. skin friction Nusselt numerically compared available literature, where accuracy performance ANN proved.

Language: Английский

---

An Efficient Deep Learning Prognostic Model for Remaining Useful Life Estimation of High Speed CNC Milling Machine Cutters

Results in Engineering, Journal Year: 2024, Volume and Issue: 24, P. 103420 - 103420

Published: Nov. 16, 2024

Language: Английский

---

Artificial Neural Network Modeling for Predicting Thermal Conductivity of EG/Water-Based CNC Nanofluid for Engine Cooling Using Different Activation Functions

Frontiers in Heat and Mass Transfer, Journal Year: 2024, Volume and Issue: 22(2), P. 537 - 556

Published: Jan. 1, 2024

A vehicle engine cooling system is of utmost importance to ensure that the operates in a safe temperature range.In most radiators are used cool an engine, water serves as fluid.The performance radiator terms heat transmission significantly influenced by incorporation nanoparticles into water.Concentration and uniformity nanoparticle distribution two major factors for practical use nanofluids.The shape size also have great impact on transfer.Many researchers investigating transfer.This study aims develop artificial neural network (ANN) model predicting thermal conductivity ethylene glycol (EG)/waterbased crystalline nanocellulose (CNC) nanofluid internal combustion engine.The implementation considering different activation functions hidden layer made find best using nanofluid.Accuracies with networks analyzed concentrations temperatures.In networks, Levenberg-Marquardt optimization approach functions, including Tansig Logsig training phase.The findings each training, testing, validation phase presented demonstrate provides highest level accuracy.The result was obtained Tansig, which has correlation 0.99903 error 3.7959 ×10 -8 .It been noticed function can be good due its 0.99890 4.9218 .Thus our ANN demonstrates high between actual output predicted output.

Language: Английский

---

Buongiorno’s radiative Prandtl-Eyring nanofluid flow through porous medium over a stretching surface with cross-diffusion effects

Proceedings of the Institution of Mechanical Engineers Part N Journal of Nanomaterials Nanoengineering and Nanosystems, Journal Year: 2024, Volume and Issue: unknown

Published: Aug. 31, 2024

The main focus of the current analysis is to describe thermo-magnetic flow, heat and mass transport features two-dimensional dissipative Prandtl-Eyring nanofluid (PE-NF) flow over a stretching sheet under influence porous medium magnetic Ohmic dissipation numerically. An innovative Buongiorno’s model in terms Brownian motion thermophoresis deployed accurately simulate nano behavior within boundary layer regime. thermal radiation source/sink effects are also included process. In addition this, thermo-diffusion diffusion-thermo effect demonstrate temperature concentration diffusion mechanism presence chemical reaction emerged coupled nonlinear time-independent partial differential equations rendered their dimensionless form through appropriate similarity transformations solved by deploying Matlab-based BVP4C technique. graphical visualization showed that, increasing number diminished field enhanced profiles Rising parameter decayed velocity. Increasing Eckert numbers enhance Magnifying suppressed velocity diffusion. parameters increases profile. Amplifying Soret upsurges Skin-friction coefficient amplified with numbers. Heat transfer rate acts as function parameters. objective this study dissipation, radiation, medium, source/sink, Dufour Buongiorno consideration generalizes former studies gives new re-defined mathematical formulation sheet. Finally, numerical accuracy results validated available solutions noticed good agreement.

Language: Английский

---