Artificial Neural Network Modeling Approach Research Articles

Estimation of cutting forces in CNC slot-milling of low-cost clay reinforced syntactic metal foams by artificial neural network modeling

PurposeThis study aims to elucidate the machining properties of low-cost expanded clay-reinforced syntactic foams by using different neural network models for the first time in the literature. The main goal of this endeavor is to create a casting machining-neural network modeling flow-line for real-time foam manufacturing in the industry.Design/methodology/approachSamples were manufactured via an industry-based die-casting technology. For the slot milling tests performed with different cutting speeds, depth of cut and lubrication conditions, a 3-axis computer numerical control (CNC) machine was used and the force data were collected through a digital dynamometer. These signals were used as input parameters in neural network modelings.FindingsAmong the algorithms, the scaled-conjugated-gradient (SCG) methodology was the weakest average results, whereas the Levenberg–Marquard (LM) approach was highly successful in foreseeing the cutting forces. As for the input variables, an increase in the depth of cut entailed the cutting forces, and this circumstance was more obvious at the higher cutting speeds.Research limitations/implicationsThe effect of milling parameters on the cutting forces of low-cost clay-filled metallic syntactics was examined, and the correct detection of these impacts is considerably prominent in this paper. On the other side, tool life and wear analyses can be studied in future investigations.Practical implicationsIt was indicated that the milling forces of the clay-added AA7075 syntactic foams, depending on the cutting parameters, can be anticipated through artificial neural network modeling.Social implicationsIt is hoped that analyzing the influence of the cutting parameters using neural network models on the slot milling forces of metallic syntactic foams (MSFs) will be notably useful for research and development (R&D) researchers and design engineers.Originality/valueThis work is the first investigation that focuses on the estimation of slot milling forces of the expanded clay-added AA7075 syntactic foams by using different artificial neural network modeling approaches.

Enhanced defluoridation by nano-crystalline alum-doped hydroxyapatite and artificial intelligence (AI) modeling approach

The study aimed to investigate the defluoridation capacity of nano-hydroxyapatite (HAP) [Ca10(PO4)6(OH)2] and alum-doped hydroxyapatite (AHAP) [Ca8Al(PO4)6.(OH)2] as an environmental friendly adsorbent. The physicochemical characteristics of both the material (HAP and AHAP) were examined using XRD, FE-SEM-EDS, and BET techniques. The batch adsorption study revealed a fluoride removal efficiency of up to 83% (AHAP) and 74% (HAP) under acidic conditions (pH-2). The doping of alum alters the surface chemistry and enhances the affinity of AHAP for fluoride adsorption. The pseudo-second-order kinetic (R2–0.9941) and Langmuir isotherm (R2–0.9425) models best describe the adsorption mechanism and behavior. The thermodynamic analysis indicated the spontaneous and endothermic nature of the adsorption process. The study also tested the applicability of the artificial neural network (ANN) modeling approach using MATLAB R2013a to simulate the simulated absorptive efficiency of AHAP. This study suggests that AHAP proved an effective adsorbent for defluoridation.

Artificial neural network modeling of solar photovoltaic panel energy output

Solar panel energy output is an essential parameter for the design and operation of renewable energy systems. Previously, little was known about the precise relationship between the energy outputs of solar panels with various meteorological, radiometric, and weather conditions in the southern California region. Without precise modeling or prediction systems, solar energy can potentially be wasted due to grid energy fluctuation. Thus, it is intended to use an artificial neural network (ANN) to develop solar panel energy output prediction model with a high degree of accuracy. A self-developed feedforward ANN model utilizing the Rectified linear unit (ReLu) activation function was used in the present study. Meteorological, weather, and sun irradiation data collected throughout the last year from a residential location have been used to train the models. The model’s performance was identified based on the minimum mean absolute error (MAE) and root mean square error (RMSE) and maximum linear correlation coefficient (R2). Further, the present self-developed ANN model was consistent with other solar energy experimental results and theoretical analysis. The developed ANN model using the Python programming language achieved a high R2 of more than 85% which ascertains the accuracy and suitability of the model to predict the daily solar energy output in local southern California area. This ANN modeling approach can be extended to many other applications such as SCORE, commercial, and residential building design.

Adsorption of lead ions from wastewater using electrospun zeolite/MWCNT nanofibers: kinetics, thermodynamics and modeling study.

Heavy metal contamination in water is a serious environmental issue due to the toxicity of metals like lead. This study developed zeolite and multi-walled carbon nanotube (MWCNT) incorporated polyacrylonitrile (PAN) nanofibers via needleless electrospinning and examined their potential for lead ion adsorption from aqueous solutions. The adsorption process was optimized using response surface methodology (RSM) and artificial neural network (ANN) modeling approaches. The adsorbent displayed efficient lead removal of 84.75% under optimum conditions (adsorbent dose (2.21 g), adsorption time (207 min), temperature (48 °C), and initial concentration (62 ppm)). Kinetic studies revealed that the adsorption followed pseudo-first-order kinetics governed by interparticle diffusion. Isotherm analysis indicated Langmuir monolayer adsorption with improved 5.90 mg g-1 capacity compared to pristine PAN nanofibers. Thermodynamic parameters suggested the adsorption was spontaneous and endothermic. This work demonstrates the promise of electrospun zeolite/MWCNT nanofibers as adsorbents for removing lead from wastewater.

Groundwater quality index development using the ANN model of Delhi Metropolitan City, India.

Groundwater is widely recognized as a vital source of fresh drinking water worldwide. However, the rapid, unregulated population growth and increased industrialization, coupled with a rise in human activities, have significantly harmed the quality of groundwater. Changes in the local topography and drainage systems in an area have negative impacts on both the quality and quantity of groundwater. This underscores the critical need to assess the susceptibility of groundwater to pollution and implement measures to mitigate these risks. The water quality index (WQI) is an approach that simulates the water quality at peculiar locations for a particular period of time. The artificial neural network (ANN) model approach is such an idealistic methodology that can be utilized for WQI development and provides better results for specific locations in optimum time. Therefore, the goal of the current study is to provide a unique way for using artificial neural networks (ANN) to characterize the groundwater quality of Delhi Metropolitan City, India. In order to make the water fit for residential and drinking use, the research also pinpoints the geographical variability and spots where the contaminated region has to be sufficiently cleaned. A minimum WQI of 41.51 was obtained at the Jagatpur location while a maximum value of 779.01 was at the Peeragarhi location. During the training phase, the results obtained using the ANN model were highly favorable, demonstrating a strong association with an R-value of 98.10%, thus highlighting the program's exceptional efficiency. However, in accordance with the correlation regression findings, the prediction outcomes of the ANN model in testing are observed to be an R-value of 99.99-100%. This study confirms the promise and advantages of employing advanced artificial intelligence in managing groundwater quality in the studied area.

Obtaining an accurate prediction model for viscosity of a new nano-lubricant containing multi-walled carbon nanotube-titanium dioxide nanoparticles with oil SAE50

This study aims to investigate the viscosity behavior of multi-walled carbon nanotube (MWCNT) - titanium dioxide (TiO2) (40−60) - SAE50 oil nanofluid using an Artificial Neural Network (ANN) modeling approach. The main objective is to develop a highly accurate predictive model for viscosity by considering three input parameters: temperature, solid volume fraction (SVF), and shear rate (SR). Rheological measurements provide experimental data used to train and validate the ANN model. The ANN model's architecture, activation functions, and training algorithms are carefully chosen. Data are divided to three subsets including train, validation and test. ANN is trained using trainlm algorithm for 50 times to vanish the effect of random nature of ANN weight initialization. The trained ANN model is then utilized to predict the viscosity of the nanofluid under varying conditions. The results demonstrate the efficacy of the proposed ANN model in capturing the complex relationship between viscosity and the input parameters, providing accurate viscosity predictions for the MWCNT-TiO2-oil SAE50 nanofluid. Furthermore, the influence of temperature, SVF, and SR on viscosity is analyzed, offering valuable insights into the flow behavior of the nanofluid. According to the obtained results, the developed ANN model presents a reliable and efficient approach to estimate the viscosity of the MWCNT-TiO2-SAE50 oil nanofluid, eliminating the need for costly and extensive experimental measurements within the analyzed range. ANN could model the nanofluid viscosity with R2 = 0.9998 and MSE= 0.000189 that is quite acceptable. Also, the experimental data revealed that for the investigated nanofluid, temperature and shear rate have impressive effect on the viscosity (changing viscosity more than 100% for the analyzed margin), on the other hand, the nanoparticle volume fraction effect is much lower, to be more precise, increasing the nanoparticle percentage will increase the viscosity mean value around 30%.

Development of a model for a walking robot made of Desai mechanism using ANN and regression approach

The study focuses on computing the optimized foot profile for a walking leg mechanism using artificial neural network (ANN), genetic algorithm, and regression approaches. The technique adopted in this work is the benchmark approach and acts as a tool for complex problems. A mathematical model using regression and ANN is developed for the 8-link coplanar mechanism. Optimum link lengths are obtained to minimize the objective function (error). The output response is the foot length with a minimum foot height of 124 mm for obstacle clearance. A neural network is designed with seven neurons (one neuron/link) in the input layer. Optimum neurons in the hidden layer are determined based on the output obtained through simulation. A single neuron is used to represent the foot profile length at the output layer. The foot lengths obtained from the regression model and ANN are compared and validated with a genetic algorithm for the data sets of 100, 200, 300, 400, and 500. Simulation studies of the walking leg mechanism revealed a difference of 19%, 22.4%, and 5.23% in the foot profile by ANN and mathematical, ANN and regression model, and mathematical and regression approach respectively. This paper reveals that different approaches viz., ANN, mathematical and regression models generate dissimilar foot profiles.

Optimization of sequential ultrasound-microwave assisted extraction of polyphenols-rich concrete from tuberose flowers through modelling

The polyphenols-rich concrete from tuberose flowers can be extracted by sequential ultrasound-microwave assisted extraction (UMAE). However, additional study is required to optimise the processing parameters of sequential UMAE of tuberose concrete. Therefore, in this work, we have focused on establishing the optimal conditions for sequential UMAE of concrete from tuberose flowers using the Response Surface Methodology (RSM) and Artificial Neural Network (ANN) model approach. The impact of different extraction conditions (microwave power (160–800 W), irradiation time (60–300 s), sonication intensity (200–600 W) and sonication time (360−840)) of sequential UMAE on the percent concrete yield from the fresh chopped tuberose flowers were investigated. The percent concrete yield (%), total phenols (mg/g) and antioxidant activity of concrete obtained were 92.28 ± 0.52%, 232.50 ± 0.46 mg/g and 54.48 ± 0.46%, respectively under optimized conditions of sonication power-300 W, microwave power-480 W, irradiation time-2.2 min and sonication time-10 min. The findings suggested that the RSM and ANN models may adequately describe the experimental data. However, ANN prediction is more accurate than the RSM model with a higher coefficient of determination (R2) (0.987 vs. 0.904) and lower mean squared error (1.385 vs. 1.65) and root mean squared error (1.177 vs. 1.284). Gas chromatography-mass spectrometry analyzed that the tuberose concrete was rich in benzyl benzoate (14.5%), trans-Farnesol (11.02%), indole (8.52%), eugenol (5.35%), isoeugenol (5.16%), methyl benzoate (2.17%), methyl anthranilate (2%).

Quantification of contamination, ecological risk index, and health risk assessment of groundwater using artificial neural network and multi-linear regression modeling approaches within Egbema, Nigeria

Quantification of contamination, ecological risk index, and health risk assessment of groundwater using artificial neural network and multi-linear regression modeling approaches within Egbema, Nigeria

Prediction of Refractive Index of Petroleum Fluids by Empirical Correlations and ANN

The refractive index is an important physical property that is used to estimate the structural characteristics, thermodynamic, and transport properties of petroleum fluids, and to determine the onset of asphaltene flocculation. Unfortunately, the refractive index of opaque petroleum fluids cannot be measured unless special experimental techniques or dilution is used. For that reason, empirical correlations, and metaheuristic models were developed to predict the refractive index of petroleum fluids based on density, boiling point, and SARA fraction composition. The capability of these methods to accurately predict refractive index is discussed in this research with the aim of contrasting the empirical correlations with the artificial neural network modelling approach. Three data sets consisting of specific gravity and boiling point of 254 petroleum fractions, individual hydrocarbons, and hetero-compounds (Set 1); specific gravity and molecular weight of 136 crude oils (Set 2); and specific gravity, molecular weight, and SARA composition data of 102 crude oils (Set 3) were used to test eight empirical correlations available in the literature to predict the refractive index. Additionally, three new empirical correlations and three artificial neural network (ANN) models were developed for the three data sets using computer algebra system Maple, NLPSolve with Modified Newton Iterative Method, and Matlab. For Set 1, the most accurate refractive index prediction was achieved by the ANN model, with %AAD of 0.26% followed by the new developed correlation for Set 1 with %AAD of 0.37%. The best literature empirical correlation found for Set 1 was that of Riazi and Daubert (1987), which had %AAD of 0.40%. For Set 2, the best performers were the models of ANN, and the new developed correlation of Set 2 with %AAD of refractive index prediction was 0.21%, and 0.22%, respectively. For Set 3, the ANN model exhibited %AAD of refractive index prediction of 0.156% followed by the newly developed correlation for Set 3 with %AAD of 0.163%, while the empirical correlations of Fan et al. (2002) and Chamkalani (2012) displayed %AAD of 0.584 and 0.552%, respectively.

Enhancing EDM Machining Precision through Deep Cryogenically Treated Electrodes and ANN Modelling Approach.

The critical applications of difficult-to-machine Inconel 617 (IN617) compel the process to be accurate enough that the requirement of tight tolerances can be met. Electric discharge machining (EDM) is commonly engaged in its machining. However, the intrinsic issue of over/undercut in EDM complicates the achievement of accurately machined profiles. Therefore, the proficiency of deep cryogenically treated (DCT) copper (Cu) and brass electrodes under modified dielectrics has been thoroughly investigated to address the issue. A complete factorial design was implemented to machine a 300 μm deep impression on IN617. The machining ability of DCT electrodes averagely gave better dimensional accuracy as compared to non-DCT electrodes by 13.5% in various modified dielectric mediums. The performance of DCT brass is 29.7% better overall compared to the average value of overcut (OC) given by DCT electrodes. Among the non-treated (NT) electrodes, the performance of Cu stands out when employing a Kerosene-Span-20 modified dielectric. In comparison to Kerosene-Tween-80, the value of OC is 33.3% less if Kerosene-Span-20 is used as a dielectric against the aforementioned NT electrode. Finally, OC's nonlinear and complex phenomena are effectively modeled by an artificial neural network (ANN) with good prediction accuracy, thereby eliminating the need for experiments.

A Hybrid Analysis of Consumer Preference for Domestic Products: Combining PLS-SEM and ANN Approaches

This study aims to investigate the factors influencing Algerian consumers’ willingness to purchase domestic cosmetics products over French imported products. The study employs a hybrid methodology, combining Partial Least Squares Structural Equation Modeling (PLS-SEM) and Artificial Neural Network (ANN) modeling approaches. Data were collected from 237 Algerian consumers through snowball sampling and an online survey. The results indicate that consumer ethnocentrism, consumer animosity, product quality judgment, and perceived price significantly influence willingness to buy domestic cosmetic products. The ANN results support these findings, revealing that consumer ethnocentrism is the key predictor factor with the greatest normalized importance ratio. Furthermore, a significant mediating role of product quality judgment was found in the relationship between consumer ethnocentrism, perceived price, and willingness to buy domestic cosmetic products. The study provides insights for marketers interested in the Algerian market, helping them to better understand consumer behavior, particularly in the context of cosmetics products. This study is unique in its focus on the assessment of global vs. domestic cosmetic product usage in developing nations such as Algeria.

Prediksi kekasaran permukaan baja S45C terhadap parameter pemesinan dan getaran pada proses bubut menggunakan metode artificial neural network

Based on machining characteristics, this study gives surface roughness modeling for machine parts. The artificial models used the Artificial Neural Network (ANN) modeling approach and multivariable regression analysis were used to create the prediction model. S45C steel was one of the materials utilized in this research. With a depth of cut 0.5 mm, the parameters are spindle (n) of 165, 330, 585, and 1170 rpm and feed (f) of 0.2 mm/rev. Utilizing TIBCO software, surface roughness values will be predicted. Equations derived from multivariable linear regression serve as the study's findings. At 1170 rpm spindle rotation and 0.5 mm of cut depth, the lowest surface roughness measurement of 1.114 (μm) was recorded. At spindle speed 585 and a cut depth of 2.0 mm, a roughness value of 2.999 (μm) was recorded as the maximum value. Roughness rises at spindle speeds between 585 and 900 rpm when cutting at shallower depths. The third modeling had the smallest error value, which was 11.21%, and surface roughness value using an artificial neural network with five simple multi-layer models.

Performance analysis of WEDM during the machining of Inconel 690 miniature gear using RSM and ANN modeling approaches

Abstract The present work aims to carry out a feasible study of wire electro-discharge machining (WEDM) during the machining of Inconel 690 superalloy gears. Processing conditions of power-on time, power-off time, current, and spark-gap voltage are varied to evaluate the process performance in terms of material removal rate (MRR), surface roughness (SR), and wire consumption. Parametric optimization has been carried out using combined approach of response surface methodology (RSM) and artificial neural network (ANN). Results revealed that ANN predicted values are 99% in agreement with the experimental results which validates its effectiveness as compared to RSM predicted values. A viability study of noise characteristics of the processed gear is also done using a noise testing setup. Additionally, FE-SEM has been used to analyze the machined surface’s topography. Greater discharge energy brought by a longer pulse length raises the values of MRR, SR, and recast layer thickness (RLT). This study explores the capability of WEDM to produce a more precise and smooth gear profile as compared to other conventional machining methods. Additionally, RLT and microhardness of the machined surface have been critically studied to comprehend the better understanding of the process mechanism.

Land use and land cover changes along the Boteti-Thamalakane River system in Ngamiland District, Botswana

Analysis of the dynamics of land use and land cover change (LULCC) along rivers that supply water to settlements provides crucial information to decision-makers; it offers information for planners and water supply authorities on water management issues in general. The paper analyzed LULCC along Boteti and Thamalakane rivers and predicted the probable LULCC for the year 2050 using the artificial neural network (ANN) of the module for land use evaluation (MOLUCE) within the QGIS 2.9 software. Adopting a supervised classification method, the Maximum Likelihood Classification (MLC) algorithm was used to classify the study area into five LULC categories. Findings show an increase in a built-up area (46.5%), barren land (90.1%) and a decrease in shrubs/trees (67.4%), water/waterlogged area (48%) and hydrophytes (39.3%) between 1990 and 2020. Using 2020 as the baseline, the ANN modelling approach predicted an increase in the built-up area and a continued shrinking of water and shrubs/trees spatial coverage in the next 30 years. Findings also reveal the presence of building constructions within less than 500 m from the river channel and/or flood plains. While a strict monitoring of the construction works along the riverbank needs to be constantly carried out, the paper also recommends the creation of a buffer zone of 500 m along the rivers where construction activities are prohibited.

Neural network modelling of mechanical joints for the application in large-scale crash analyses

This paper presents an artificial neural network (NN) modelling approach to represent a connector model in large-scale finite element explicit crash simulations. The NN model was established to describe the local force–deformation response of point connectors in automotive applications, namely self-piercing rivets and flow-drill-screws. The study is limited to two-sheet connections and to the use of feedforward NNs. Successive loading and unloading of the joints is not studied. Various architectures and complexities of the feedforward NNs were evaluated and trained based on data generated from a constraint model found in the literature. This forms a proof of concept for implementing a modelling technique not based on physics-motivated constitutive equations. The impact of the network complexity and training data diversity was investigated. The NN model was implemented as a cohesive zone model for incremental force prediction in an explicit finite element code. In order to have a wide selection of joint types, five different joint configurations including self-piercing rivets (SPR) and flow-drill screws (FDS) were investigated. Numerical results from the NN model were compared to physical tests from all joint configurations. It was shown that a rather basic machine learning technique like a feedforward NN was able to reproduce path-dependent force–deformation behaviour for the application in explicit FE solvers.

Artificial neural network modelling approach for the prediction of turbidity removal efficiency of PACl and Moringa Oleifera in water treatment plants

Artificial neural network modelling approach for the prediction of turbidity removal efficiency of PACl and Moringa Oleifera in water treatment plants

Modelling of Activity Concentrations Using Machine Learning Technique

This study develops a modelling framework for activity concentration prediction following a Machine Learning (ML) approach. The study aims to understand the spatial and temporal evolution of activity considering the effects of the built environment, time of day, and past activities. This study utilizes the Nova Scotia Travel Activity (NovaTRAC) survey and Info Canada Business Establishment dataset to extract activity and built environment information. An Artificial Neural Network (ANN) modelling approach is adopted for prediction modelling as it provides advantages over other modelling approaches in dealing with assumptions, non-linearity, and model accuracy. The ANN model is trained and validated through a trial-and-error process using various model architectures and evaluation metrices, such as mean square error (MSE) and mean absolute error (MAE). Results show that model fitness is satisfactory with significantly smaller values of MSE (0.37) and MAE (0.23). Furthermore, results indicate high prediction accuracy. The outcomes of this study provide insights into factors affecting activity density which will further aid in designing public spaces, transportation infrastructures and safety protocols.

Regression Analysis and Artificial Neural Network Approach to Predict of Surface Roughness in Milling Process

Surface roughness (Ra) has a significant influence on the fatigue strength, corrosion resistance, and aesthetic appeal of machine components. Ra is hence a crucial manufacturing process parameter. This study predicts Ra of aluminum alloy Al-7024 after milling. Regression analysis and artificial neural network (ANN) modeling approaches are suggested for predicting Ra values. For better surface roughness, the cutting parameter must be set properly. Spindle speed, feed rate, and depth of cut have been chosen as predictors. Through 31 study cases, regression and ANN were used to examine how these parameters affected Ra. The measurement of surface roughness, together with comprehensive Ra analysis and regression analysis. The findings of this investigation indicate that Ra was predicted by both the regression and ANN models. convergent results from model predictions are obtained. This convergence highlights the promising methodology used in this work to forecast Ra in the milling of Al-7024. The findings demonstrated that, in comparison to the regression model, which had an average variation from the actual values of roughly 1%, The surface roughness was accurately predicted by the ANN model.

In-situ selective epoxidation of Colocynthis Vulgaris shrad seed oil for the synthesis of a methacrylated biobased resin; An artificial neural network (ANN) modelling approach

Colocynthis Vulgaris shrad seed oil (CVSSO) contains unsaturated fatty triglycerides predominantly (oleic, linoleic and linolenic acid), indicating its suitability for epoxide development. The challenges associated with petroleum-based epoxides, such as price escalation, non-renewability and environmental pollution, have led the research for its substitution with vegetable-based epoxide; artificial intelligence techniques have been adopted to model and predict complex processes. This study predicted CVSSO epoxidation parameters using an Artificial Neural Network (ANN). First, pure CVSSO was epoxidized via a conventional in-situ method with peroxyacetic acid. Then, methacrylated biobased resin was synthesized by further modifying the epoxidized oil with methacrylic acid in the presence of hydroquinone. The input parameters of the neural networks are time, temperature, composition and stirring speed, while the output parameters are iodine value, oxirane value, conversion, yield and selectivity. ANN was evaluated with 11 backpropagation (BP) algorithms; each algorithm was evaluated with four neurons in the input layer, a hidden layer with ten neurons and an output layer with five neurons. Coefficient of determination (R 2 ) and mean square error (MSE) have been implemented and correlated to test the adequacy and predictive ability of the model. Results showed that the Levenberg Marquardt algorithm gave a good prediction for iodine and oxirane value with mean square error (MSE) of 2.2397E-11 and 1.67718E-12; at the same time, Bayesian regularization predicted conversion, yield and selectivity with MSE of 4.3925E-12, 1.86759E-12 and 8.53013E-13 respectively; nevertheless, Levenberg Marquardt and Bayesian regularization gave the best prediction than other different BP algorithms. Other statistical indicators such as mean absolute error (MAE) showed 0.0232,0.0781,8.14,12.13 and 19.73), while mean absolute percentage error (MAPE) gave (0.00028%, 0.000864%, 0.000376%, 0.00019% and 0.00017%) for iodine value, oxirane value, conversion, yield and selectivity respectively; the results indicate a high forecast on the ANN simulated process parameters and suggest a good agreement between the experimental ad predicted values. The characterization results from physicochemical and FTIR analysis indicated that CVSSO was suitably epoxidized, and further modification to Methacrylated epoxidized Colocynthis Vulgaris shrad oil (MECVSSO) was actualized. These results show that ANN can potentially estimate epoxidation parameters of CVSSO. This development may have significant potential to improve the product quality of epoxides and reduce time and cost by minimizing epoxidation experiments.