Artificial Neural Network Model Research Articles

Weak nonlinear thermo bioconvection: Heat transfer via artificial neural network

The present study undertakes an analytical exploration of the onset of regular and chaotic thermo-bioconvection in a suspension of gravitactic microorganisms. Additionally, it employs a machine learning approach for numerical computation and prediction of heat transfer rates. The two-dimensional flow governing dynamics are modeled using the Hamilton-Crosser model for microorganisms. The amplitude of convection is determined by solving the cubic Ginzburg-Landau equation derived by applying the Lorenz technique. Stationary curves are plotted with thermal Rayleigh number and wave number, while Nusselt numbers are depicted over a range of time. The onset of chaotic motion is briefly discussed through Lyapunov plots and a bifurcation diagram. Further, to predict the heat transfer rate with multiple interconnected parameters, an artificial neural network is trained with the Levenberg-Marquardt algorithm to understand the underlying patterns of simulated data. The trained neural network is then employed to estimate the Nusselt number for various values of bioconvection Rayleigh number, bioconvection Lewis number, and bioconvection Péclet number. The values obtained from the artificial neural network models are compared with numerical data for validation and are found to be in good agreement. The findings of weak non-linear stability indicate that chaotic motion emerges in the system at the Hopf-Rayleigh number of 24.635, and fast-swimming microorganisms significantly increase the heat transfer rate. The coefficient of correlation is higher than 0.99, supporting the accuracy of developed ANN models to predict the Nusselt number with high accuracy. This observation supports the potential utilization of microbes in heat transfer applications.

Oscillation-static cycle cultivation enhances the synthesis of triterpenes and mycelium activity in Ganoderma lucidum

Ganoderma lucidum is a precious fungus with both edible and medicinal values and has a long history of medical use. Triterpenes as the main active components endow G. lucidum with anti-tumor, antioxidant, and other pharmacological activities. The present study endeavors to establish a proficient liquid-state fermentation technology for the enhanced production of triterpenes. In view of the limitations inherent in conventional submerged fermentation and oscillation-static two-stage cultivation, this study established an oscillation-static cycle cultivation process and optimized the cultivation conditions by building an artificial neural network model based on genetic algorithms. The cultivation conditions for the high-yield production of triterpenes were optimized as follows: 2.8 days of oscillation, 7.3 days of static cultivation, 0.2 day of oscillation, and 0.3 day of static cultivation. Under these conditions, the content of triterpenes reached 20.82 mg/g. The yield of triterpenes reached 129.09 mg/L, showing a remarkable increase of 324.78% compared with that of the Z10J0 method. Moreover, the established method shortened the cultivation cycle by 10.6 days. The mycelia cultivated under this regimen exhibited commendable anti-tumor and antioxidant activities. This study not only presents an economical liquid-state fermentation approach but also streamlines the fermentation flow, reduces fermentation duration, and effectively ameliorates drawbacks associated with conventional cultivation methods. In addition, this study gives valuable insights into the scaled application of liquid-state fermentation in the high-yield production of triterpenes, which showcases broad prospects.

Utilisation of Potassium Chloride in the Production of White Brined Cheese: Artificial Neural Network Modeling and Kinetic Models for Predicting Brine and Cheese Properties during Storage

Excessive sodium consumption is a worldwide problem, prompting the industry to develop sodium-reduced products and substitute salts. High sodium consumption is a significant risk factor for high blood pressure, cardiovascular disease and kidney disease. Excessive sodium intake also impairs the immune system in the gastrointestinal tract. Potassium chloride (KCl) is the most commonly used mineral salt due to its similarity to sodium chloride (NaCl), and its consumption has been shown to lower blood pressure when consumed in adequate amounts. The aim of this study was to partially replace NaCl with KCl at levels of 25%, 50% and 75% in the brine used to make white brined cheese. Parameters such as acidity, total dissolved solids, salinity, conductivity, colour, texture and sensory properties were evaluated during a 28-day refrigerated storage period. KCl can replace NaCl by 50%, and no significant differences in physicochemical and sensory parameters were observed during cheese storage compared to the control sample. In addition, the study investigates the use of Artificial Neural Network (ANN) models to predict certain brine and cheese properties. The study successfully developed four different ANN models to accurately predict various properties such as brine pH, cheese colour and hardness over a 28-day storage period.

Buckling Performance Evaluation of Double-Double Laminates with Cutouts Using Artificial Neural Network and Genetic Algorithm

Although the double-double (DD) laminates proposed by Tsai provide a promising option for achieving better structural performance with lower manufacturing and maintenance costs, the buckling performance of perforated DD laminates still remains clear. In this study, optimal ply angles, rotation angles, and the corresponding maximum buckling loads are determined for DD laminates with various cutouts, which are used for comparisons to evaluate the effects of cutout size and shape on the buckling behaviour of perforated DD laminates. Apart from conventional circular and elliptical cutouts, the use of a combined-shape cutout for DD laminates is also investigated. As a large number of optimisations are required to obtain the maximum buckling loads for different cases in this study, an efficient optimisation method for perforated DD laminates is proposed based on an artificial neural network (ANN) and a genetic algorithm (GA). Unlike conventional quadaxial (QUAD) laminates, the repetition of a four-ply sublaminate in DD laminates makes their layup to be represented by only two ply angles; hence, the application of ANN models for predicting the buckling behaviour of various perforated DD laminates is studied in this paper. The superior performance of the ANN models is demonstrated by comparisons with other machine learning models. Instead of using the time-consuming FEA, the developed ANN model is utilised within a GA to obtain the maximum buckling load of perforated DD laminates. Compared to the circular cutout, the use of elliptical and combined-shape cutouts leads to more noticeable changes in the optimal ply angles as the cutout size increases. Based on the obtained results, the use of the combined-shape cutout is recommended for DD laminates.

Machine learning approach for predicting the machining characteristics of additive manufactured ABS material

An Additive manufacturing process offers advantage for the production of intricate designs with increased efficiency and reduced material waste. In this research, an experimental and analysis work has been conducted to identify the significant 3-D printed input parameters effecting the output response characteristics as kerf width. The correlation between experimental and predicted values has been established through the application of Artificial Neural Network (ANN), Support Vector Machine Regression (SVMR), and Response Surface Methodology (RSM) techniques. The validation of predicted models conducted with the help of mean prediction error, maximum error and minimum error. The performance of ANN model was analyzed by feedforward-propagation method using Levenberg- Marquardt algorithm. The maximum prediction error investigated by ANN, SVM, and RSM were noted as 4.28, 5.18, and 5.84 respectively. The optical micrographs have been investigated to understand the hole and crater formation phenomenon.

Modeling and Forecasting High Frequency Currency Exchange Rates. A Comparative Study of ARIMA, ANN, and Hybrid ARIMA-ANN Models

Aims/ Objectives: The study develops comparative results on the modeling and prediction performance of the autoregressive integrated moving average (ARIMA), artificial neural network (ANN) and the hybrid ARIMA-ANN time series models for high frequency data. Methodology: The study made use of ARIMA, ANN, and hybrid ARIMA-ANN models to forecasts the East Africa Community countries' daily currency exchange rates data which were obtained from the Central Bank of Kenya website and covered the period from January 2017 to December 2023. Stationarity of the time series data was established using the ADF test. The Ljung Box test and ACF plots were used to establish and compare the goodness-of-fit of the resultant models while the Mean Absolute Percentage Error (MAPE) and Root Mean Squared Error(RMSE) values were used to compare the prediction performance. Results: The study established that the hybrid ARIMA-ANN methodolgy provided better-fitting models for the currency exchange rates data compared to ARIMA and ANN modeling strategies since all the Lyung Box test statistics had p values greater than 5%. Comparatively, the hybrid methodology registered lower MAPE and RMSE values hence had better prediction accuracy compared to ARIMA and ANN methods. Conclusion: The Hybid methodology improves the modeling and forecasting accuracy over the ARIMA and ANN models for high frequency time series data due to its ability to captures both the linear and nonlinear patterns in the time series data.

ANN modeling and optimization of friction stir welding performance for AA6061 and AA5083 alloy joints

This research focused on the characteristics of the AA6061/AA5083 alloy joints which are fabricated by the friction stir welding (FSW) technique. The impact of FSW parameters such as tool tilt angle, pin depth, and tool geometry on the mechanical properties and fractographic characteristics were studied. Ultimate tensile strength (UTS), hardness (HBN), and specific wear rate (SWR) were examined for characterization while fractography analyses were done using scanning electron microscopy. Taguchi-Grey relational analysis (GRA) was employed to execute multicriteria optimization and recognize the best grouping of parameters. The optimized values attained from this analysis were 259.2 MPa for UTS, 94.9 HV for HBN, and 0.0819 m³/Nm for SWR. The analysis of variance (ANOVA) results derived from the GRA revealed that tool geometry exerted the most significant influence (39.79%) on output factors, followed by pin depth (24.90%) and tilt angle (24.26%). To enhance the predictive accuracy of the output responses, an artificial neural network (ANN) model was developed utilizing the Levenberg-Marquardt (LM) algorithm. The optimized ANN architecture, configured as (3–10–3), exhibited robust regression analysis outcomes, showcasing correlation coefficients ( R) of 0.99999, 0.99967, and 0.99994 for the training, validation, and test datasets. The overall R-value, computed at 0.99958, affirmed high conformity between experimental and predicted values.

Bridge to Professional Practice: Evaluating Internship Effectiveness and Coordination for Civil and Hydropower Engineering Graduates, A Study of Mid-West University of Nepal using ANN

The transition from academia to professional engineering practice is a pivotal and formidable challenge for graduating engineers. This research explores the effectiveness of internship programs for final semester undergraduates’ students in Civil and Hydropower Engineering of Mid-West University, based on data collected from 77 intern students by evaluating the acquired knowledge and skills during internships, while also exploring the associated benefits and the level of co-ordination between university and the internship organizations. Questionnaires assessed self-evaluated knowledge and skills gained, perceived benefits of the internship, co-ordination by the university and the department, and challenges encountered. Data analysis utilized descriptive statistics and an Artificial Neural Network (ANN) model to identify factors influencing internship effectiveness. Results indicated that the effective coordination between universities and industry significantly enhances internship outcomes, impacting students’ practical skill development. The study underscores the importance of collaborative efforts in fostering industry-relevant competencies among future construction engineers. These findings suggest practical implications for universities in refining internship initiatives, improving curricula, and establishing robust internship guidelines to better prepare students for professional careers in engineering disciplines.

Green synthesis, characterization, and multifunctional applications of Ag@CeO2 and Ag@CeO2-pullulan nanocomposites for dye degradation, antioxidant, and antifungal activities

The synthesis and characterization of novel nanocomposites with unique properties have garnered significant interest. Ag@CeO2 nanocomposite and its pullulan counterparts were prepared using a green approach involving rosemary extract. Characterization techniques, including Fourier Transform Infrared Spectroscopy, UV–visible spectroscopy, zeta potential, Dynamic Light Scattering, High-Resolution Transmission Electron Microscopy, Energy-Dispersive X-ray Spectroscopy, Scanning Electron Microscopy, and X-ray Diffraction, confirmed the formation of Ag@CeO2 nanoparticles (NPs). Pullulan led to increased particle size and improved homogeneity. Employing the Artificial Neural Networks (ANN) model to optimize methylene blue removal by Ag@CeO2 NPs and Ag@CeO2-pullulan NPs demonstrated predictive capabilities up to 97.53 % of MB removal (R2 = 0.9991). The antioxidant test demonstrated that rosemary extract exhibited the highest activity (IC50 = 0.011 mg/mL), then Ag@CeO2 NPs (IC50 = 0.039 mg/mL), and Ag@CeO2-pullulan NPs (IC50 = 0.041 mg/mL). Both Ag@CeO2 NPs and Ag@CeO2-pullulan NPs inhibited Candida albicans growth, with the latter exhibiting enhanced efficacy (MIC = 468.27, MFC = 936.53, and IC50 = 129.60 μg/mL). The study successfully synthesized novel Ag@CeO2-based nanocomposites coupled with pullulan with promising applications in dye removal, and antimicrobial therapy. The incorporation of pullulan improved the properties of the nanocomposites, enhancing their potential for practical use in environmental and biomedical applications.

Estimation of ground-based atmospheric turbulence strength (C n2) by neural network architecture

Estimating the atmospheric turbulence strength (C n 2) becomes significant in the research field of electromagnetic radiation transmission through the atmosphere, particularly optical waves. As turbulence strength increases, the quality and strength of these optical waves may decrease and cause much trouble as they pass through the atmosphere. Throughout the years, C n 2 has been formulated by different research groups for various geographical locations and seasons using macro-meteorological variables empirically and theoretically. However, since these models are based on the data collected from numerous places and conditions, such as deserts or coastal areas, they do not provide accurate C n 2 predictions for our experimental site, as demonstrated for three well-known models in the paper. In this study, a novel, to our knowledge, artificial neural network (ANN) model named as quadratic Fourier neural network (QFNN) is trained to estimate C n 2 from experimentally measured ground-based atmospheric turbulence strength and macro-meteorological variables during the winter season in a rural area. The trained model gives reliable estimations, achieving a value of R2=0.92 for experimental C n 2 values.

Optimizing locally delivered periodontitis therapy: Development of chitosan‐hydroxyapatite‐encapsulated drug via electrospraying

AbstractThis study presents the development of an optimized drug delivery system for periodontitis treatment utilizing chitosan‐hydroxyapatite nanoparticles. Employing the electrospraying technique, researchers encapsulated the antibiotic amoxicillin within chitosan‐hydroxyapatite composite matrices while varying the concentrations of chitosan and hydroxyapatite nanoparticles. The resulting microparticles displayed sizes ranging from 276 to 386 μm. Characterization of the produced drug encapsulations indicated that higher concentrations of chitosan and hydroxyapatite resulted in the formation of larger particles with rougher surfaces, increased mechanical strength, and enhanced thermal stability. Notably, nanoindentation analysis revealed an increase in hardness from 0.21 to 0.35 GPa, and in elastic modulus from 5.32 to 8.72 GPa with escalating chitosan and hydroxyapatite content, meeting or surpassing requirements for load‐bearing regenerative biomaterials. In vitro drug release assessments exhibited sustained amoxicillin release over 24 h, predominantly through diffusion and dissolution mechanisms. Formulations with lower polymer and mineral content showcased elevated release rates, with the F1 encapsulation (0.5% chitosan, 0.2% HAP) achieving a cumulative release of 78.3 ± 3.2% in contrast to 65.1 ± 2.7% for the F5 formulation (2.5% chitosan, 1% HAP). These encapsulations selectively modulated the overproduction of proinflammatory cytokines, including IL‐1β, IL‐6, and TNF‐α, in gingival fibroblasts and osteoblasts without compromising cell viability. An artificial neural network (ANN) model accurately forecasted the material properties and biological performance of the formulations based on their compositional variations, yielding correlation coefficients exceeding 0.97. This computational platform offers a virtual screening tool for refining regenerative and drug delivery therapies. The developed chitosan‐hydroxyapatite microparticulate system demonstrates promising potential for prolonged treatment of periodontitis through controlled antibiotic delivery, improved mechanical properties, and targeted regulation of the inflammatory response.

Predictive analysis of tensile strength ratios in laminated bamboo composites: Unraveling the stochastic impact of ply angle variations through machine learning model

Laminated bamboo composites (LBC), made by sandwiching bamboo strips, offer promising alternatives to traditional construction materials, especially for housing. However, subjection to the continuous static loading makes these materials initiate cracks inside their various ply. This study uses classical laminate theory (CLT) to determine the strength ratio (SR) of the LBC at different ply orientations by applying Tsai-Wu and Tsai-Hill failure criteria using MATLAB. The study aims to calculate the SRs for LBCs using CLT, employing an ANN model and stochastic finite element (FE) modeling to investigate SRs of five-layered LBCs with varying ply orientations. Applying CLT, the highest SR was determined to be 1.5375 × 107 N/m for the [0°/0°/0°/0°/0°] laminate, as per both failure theories. The study reveals substantial SR variations depending on ply orientation, consistently showing higher SR predictions with the Tsai-Wu theory compared to the Tsai-Hill theory. Next, the study emphasizes the deterministic methodologies to account for the stochastic effects of ply angle on the obtained SR of LBCs. Monte Carlo simulation (MCS) was utilized to model 10,000 randomly generated ply angle inputs and their associated SRs. By incorporating MCS to introduce ±1% variations in ply angles and utilizing normally distributed data, this research effectively captures uncertainties associated with ply orientation. Finally, to forecast the random SR of LBC with various ply orientations, an artificial neural network (ANN) surrogate model is employed. The stochastic analysis confirms the need to quantify the associated uncertainties. The findings of this study are crucial for advancing the application of LBC in sustainable construction, providing valuable insights into their mechanical behavior under different ply orientations, considering the stochastic effect in properties.

Optimizing High-Performance Predictive Modeling of the Medium-Speed WEDM Processing of Inconel 718

The purpose of this research was to create a predictive model for a medium-speed wire electrical discharge machine (WEDM) utilizing an artificial neural network (ANN). Medium-speed WEDM experiments were developed based on the I-optimal mixture design for machining, the Inconel 718 superalloy. During the experiment, the input parameters were the spark ontime, spark offtime, wire feed, and current, with the material removal rate (MRR) and surface roughness (Ra) selected as performance indicators. The ANN model was trained on experimental data and built using a feed-forward backpropagation neural network with a (4-8-2) structure and the Bayesian regularization (BR) learning approach. The model correctly predicted the relationship between the medium-speed WEDM’s primary process parameters and machining performance. An integrated ANN model and the Non-Dominated Sorting Genetic Algorithm-II (NSGA-II) were used to determine the ideal parameters for the MRR and Ra, resulting in a set of Pareto-optimal solutions. The confirmation experiment revealed that the mean prediction error between the experimental and ideal solutions had a maximum error percentage of 1% for the MRR and 2% for the Ra, which are within acceptable ranges. This showed that the best process–parameter combinations were better for the MRR and Ra.

Predicting MHD mixed convection in a semicircular cavity with hybrid nanofluids using AI

This study presents a numerical analysis of magnetohydrodynamic (MHD) mixed convection in a semicircular enclosure containing a rotating inner cylinder and filled with nanofluids and hybrid nanofluids. The investigation explores the effects of Al2O3-TiO2-SWCNT-water hybrid nanofluids with varying nanoparticle compositions, as well as Al2O3-water, TiO2-water, and SWCNT-water nanofluids. The analysis includes the development of an artificial neural network (ANN) model to predict outcomes, achieving 97.34 % accuracy in training and 97.41 % in testing for the average Nusselt number. The study examines the impact of Reynolds number (Re), Richardson number (Ri), Hartmann number (Ha), cylinder rotation speed (Ω), cylinder size, and nanoparticle volume fraction (φ) on heat transfer and fluid flow. Key findings include a 6.98 % increase in heat transfer for SWCNT-water nanofluid from Ri = 1 to Ri = 10, a reduction in heat transfer with higher Hartmann numbers, and a significant 21.12 % enhancement when cylinder speed increases to Ω = 10 compared to a stationary cylinder. Larger cylinder sizes also improve convective heat transfer, with a 66.14 % increase for SWCNT-water nanofluid. Additionally, higher concentrations of SWCNT and Al2O3 in hybrid nanofluids enhance heat transfer performance.

Extrapolating low-occurrence strong wind speeds at pedestrian levels using artificial neural networks trained by a single turbulent dataset

Predicting low-occurrence strong wind speeds in urban areas is crucial for enhancing the comfort and ensuring the safety of residents. Artificial neural network (ANN) models outperform traditional statistical methods in accurately estimating low-occurrence wind distributions. However, generalizing trained models to be applicable to various datasets remains a significant challenge. Therefore, three ANN models with different input variables, trained by a single turbulent dataset of a building array case, were applied to four types of datasets to evaluate the models’ generalization ability for predicting low-occurrence strong wind speeds in urban areas. It found that the ANN model, which uses the mean, standard deviation, and skewness of the turbulent flow, shows the best generalization ability among the three ANN models for all types of datasets. In addition, for the four datasets, the results calculated by this model show that the majority of points have an error of less than 20% compared to the true values. It is advisable to construct models using more extensive training data that encompasses a broad range of mean and skewness of wind speeds. This study contributes to the development of more effective and adaptable ANN models for urban wind safety in building engineering.

Impact of Soybean Biodiesel Blends with Mixed Graphene Nanoparticles on Compression Ignition Engine Performance and Emission: An Experimental and ANN Analysis

The extensive use of fuels in power generation plants, industries, and transportation has led to a scarcity of fossil fuels and has contributed to global warming. This has prompted researchers to focus on improving internal combustion engine performance, as the transportation system accounts for 50% of global fuel consumption. Soybean biodiesel plays a vital role in reducing emissions and fuel consumption in engines. In the current work, a blend of soybean oil is used as a biodiesel (B20, B40, and B60), with and without the addition of graphene nanoplatelets (GNPs) examined on a constant-speed compression ignition (CI) engine with respect to pure diesel. The blends are denoted as B20GNP10, B20GNP20, B20GNP40, B20GNP60, B20GNP80, and B20GNP100, according to their proposition of biodiesel and graphene nanoplatelets. Similar blends were prepared for B40 and B60 combinations with similar GNPs concentrations. The prepared blend properties were measured, and good thermophysical properties were found. The trial includes testing the engine emissions and performance at altering loads ranging from 0 to 12 kg for all blends. The artificial neural network (ANN) tool is used to forecast the accuracy of experimental results. Compared to pure diesel, the B60GNP100 blend at 12 kg load condition showed the lowest brake specific fuel consumption at 12.58 % and the highest brake thermal efficiency at 27.13%. Emissions were estimated using a gas analyzer, and the outcomes indicated that the biodiesel blends have controlled levels of CO, CO2, NOx, and HC compared to pure diesel. The ANN model with 99.99% accuracy was developed using experimental data, confirming the accuracy of the experimental results with lower simulation time and cost. Additionally, the B60GNP100 blend yielded better results compared to previous studies.

Application of artificial intelligence to predict energy consumption and thermal efficiency of hybrid convection-radiation dryer for garlic slices

This study aims to examine the energy-related aspects of a hybrid convection-radiation drying technique employed in the drying process of garlic slices. Several factors, including infrared radiation intensity (1500, 2000, and 3000 W/m2), air temperature (40, 50, and 60 °C), and airflow rates of the drying chamber (0.7, 1.0, and 1.5 m/s) were evaluated to optimize the responses of drying time (min), energy consumption (kJ), thermal efficiency (%), and specific energy consumption (MJ/kg). Furthermore, an Artificial Neural Network model was employed to predict the effects of the parameters above on the drying system's specific energy consumption, energy consumption, drying time, and thermal efficiency. The obtained data were analyzed using a self-organizing map and principal component analysis. Interestingly, the findings showed that increasing the air temperature and infrared radiation intensity led to shorter drying times, while higher airflow rates resulted in longer drying durations. Additionally, it was found that higher infrared intensity and air temperature, combined with lower airflow rates, improved the energy indices. Increasing the air velocity to 1.0 and 1.5 m/s at the same air temperature and radiation intensity resulted in a considerable increase in drying time by 9 and 22.7%, respectively. The lowest efficiency value of 10.1% was observed at 40 °C temperature and an air velocity of 1.5 m/s. The Artificial Neural Network model demonstrated high accuracy with a Root Mean Square Error of 0.05 and an overall accuracy of (95%) in predicting the drying parameters. Notably, the self-organizing map visualization revealed that higher air temperatures and radiation corresponded to lower drying times, energy consumption, and specific energy consumption. Conversely, higher airflow rates resulted in increased drying time and energy consumption. The present study suggested that 60 °C, 3000 W/m2 and 0.7 m/s are optimum conditions for efficiently drying garlic slices under the hybrid dryer. This investigation addresses the limitations and deficiencies of previous work on energy optimization and efficiency analysis of garlic under a hybrid dryer with infrared heating systems. Further studies on the economic feasibility of applying such a system for sample drying and the impact of drying on the thermal behaviors of the products need additional investigation.

Artificial neural network based forecasting of diesel engine performance and emissions utilizing waste cooking biodiesel

Ecological and environmental problems resulting from fossil fuels are due to the harmful emissions released into the atmosphere. The rising interest in searching for alternative fuels like biodiesel is growing to solve these problems. Waste cooking oil (WCO) is transformed into methyl ester and combined with biodiesel in percentages of 25, 50, 75, and 100%. Research is done on the impacts of methyl ester blends on engine performance and emissions. Compared to diesel, the methyl ester combination showed 25% lower brake power and 24% loss in thermal efficiency at maximum load and 1500 rpm. However, diesel fuel showed 23% lower specific fuel consumption increase than biodiesel. Compared to diesel, methyl ester exhibits 15% lower air-fuel ratio and 4% volumetric efficiency. Biodiesel lowers CO, HC, and smoke concentrations by 12, 44, and 48%, respectively, compared to diesel. Biodiesel emits 23% higher NOx at 1500 rpm and 100% engine load. To predict the emissions and performance of different percentages of biodiesel at engine speed variation, an artificial neural network (ANN) model is presented. ANN modeling minimizes labor, time, and finances and uses nonlinear data. Predictions were produced about the brake output power, specific fuel consumption, thermal efficiency, air-fuel ratio, volumetric efficiency, and emissions of smoke, CO, HC, and NOx as a function of engine speed and blend ratio. All correlation coefficients (r) over 0.99 and R2\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$R^{2}$$\\end{document} values were beyond 0.98 for all variables. There were low values of MSE, MAPE, and MSLE with significant predictive ability. WCO’s biodiesel is a viable diesel engine replacement fuel.

Influence of core–shell intermetallic on dynamic recrystallization and processing maps of novel Al-Cu-Ni composite through Arrhenius and artificial neural network (ANN) models

Influence of core–shell intermetallic on dynamic recrystallization and processing maps of novel Al-Cu-Ni composite through Arrhenius and artificial neural network (ANN) models

Predicting extended hospital stay following revision total hip arthroplasty: a machine learning model analysis based on the ACS-NSQIP database.

Prolonged length of stay (LOS) following revision total hip arthroplasty (THA) can lead to increased healthcare costs, higher rates of readmission, and lower patient satisfaction. In this study, we investigated the predictive power of machine learning (ML) models for prolonged LOS after revision THA using patient data from a national-scale patient repository. We identified 11,737 revision THA cases from the American College of Surgeons National Surgical Quality Improvement Program database from 2013 to 2020. Prolonged LOS was defined as exceeding the 75th value of all LOSs in the study cohort. We developed four ML models: artificial neural network (ANN), random forest, histogram-based gradient boosting, and k-nearest neighbor, to predict prolonged LOS after revision THA. Each model's performance was assessed during training and testing sessions in terms of discrimination, calibration, and clinical utility. The ANN model was the most accurate with an AUC of 0.82, calibration slope of 0.90, calibration intercept of 0.02, and Brier score of 0.140 during testing, indicating the model's competency in distinguishing patients subject to prolonged LOS with minimal prediction error. All models showed clinical utility by producing net benefits in the decision curve analyses. The most significant predictors of prolonged LOS were preoperative blood tests (hematocrit, platelet count, and leukocyte count), preoperative transfusion, operation time, indications for revision THA (infection), and age. Our study demonstrated that the ML model accurately predicted prolonged LOS after revision THA. The results highlighted the importance of the indications for revision surgery in determining the risk of prolonged LOS. With the model's aid, clinicians can stratify individual patients based on key factors, improve care coordination and discharge planning for those at risk of prolonged LOS, and increase cost efficiency.