Levenberg-Marquardt Artificial Neural Network Research Articles

Simulation of hybrid boiling nano fluid flow with convective boundary conditions through a porous stretching sheet through Levenberg Marquardt artificial neural networks approach

In the domain of artificial neural networks, the Levenberg-Marquardt technique stands out for its innovative approach and convergent stability. It generates a numerical approach for the progression of a hybrid Cu–Al2CO3/water nanofluid on a porous stretched sheet (HNF-PSS), employing regression plots, state transition measures, histogram representations, and mean squared errors. This work explores the analysis of the fluid flow problem associated with HNF-PSS, introducing a novel application of an intelligent computing system that utilizes the effectiveness of neural networks trained by the Levenberg-Marquardt algorithm (NN-TLMA). The initial mathematical expression, expressed in terms of partial differential equations (PDEs) for HNF-PSS, undergoes transformation into dimensionless nonlinear ordinary differential equations (ODEs). The collection of data for the envisioned NN-TLMA involves parameters influencing the velocity of the HNF-PSS system model, generated using the Lobatto IIIa formula. The efficacy of the suggested NN-TLMA for solving the HNF-PSS is robustly verified by examinations of state transition dynamics, regression analyses, mean square error, and error histogram studies. The consistent alignment of the obtained results with corresponding solutions attests to the legitimacy of the structure, achieving a high level of accuracy within the range of 10–6 to 10–8. Insightful findings demonstrate that the thermal Biot number contributes significantly to an upsurge in fluid temperature. Furthermore, an augmentation in nanofluid concentration and thermal profile is found to increase the porosity strength of the medium. Nevertheless, an attenuation in the velocity profile is noted under these conditions. The application of this simulation work can be found in optimizing the designs of heat exchanger, in biomedical application and in renewable energy processes.

Agarwood oil quality identification using artificial neural network modelling for five grades

Agarwood (Aquilaria Malaccensis) oil stands out as one of the most valuable and highly sought-after oils with a hefty price tag due to its widespread use of fragrances, incense, perfumes, ceremonial practices, medicinal applications and as a symbol of luxury. However, nowadays the conventional method that rely on color alone has its limitations as it yields varying results depending on individual panelists' experiences. Hence, the quality identification system of Agarwood oil using its chemical compounds had been proposed in this study to enhance the precision of the Agarwood oil grades thus addressing the shortcomings of traditional methods. This study indicates that the primary chemical compounds of Agarwood oil encompass ɤ-Eudesmol, ar-curcumene, β-dihydroagarofuran, ϒ-cadinene, α-agarofuran, allo-aromadendrene epoxide, valerianol, α-guaiene, 10-epi-ɤ-eudesmol, β-agarofuran and dihydrocollumellarin. This study employed artificial neural network analysis with the implementation of Levenberg-Marquardt algorithm to identify the Agarwood oil grades. The study's findings revealed that this modeling system of five grades got 100% accuracies with mean square error of 0.14338×10-08. Notably, this lowest mean square error (MSE) value falls within the best hidden neuron 3. These study outcomes play a pivotal role in highlighting the Levenberg Marquardt- artificial neural network (LM-ANN) modeling that contribute to the successful of Agarwood oil quality identification using its chemical compounds.

Computing neural network to analyze heat and mass transfer in the flow of nanofluid between two disks

The model of copper nanoparticles which are suspended in the engine oil (EO) and rotated between two stretchable disks is analyzed. The flow, heat, and mass transmission phenomena of nanofluid with magnetohydrodynamics (MHD) have a vital role in many industries. A magnetic field in the vertical direction is imposed in the flow of the nanofluid and Dufour and Soret (DS) effects are discussed in the equations of energy and concentration. The main equations of motion and energy are converted into a set of nonlinear ordinary differential equations (ODEs) after applying the similarity conversions. A popular semi-analytical approach, namely the differential transform method (DTM) is used to get the solution of velocity, temperature, and concentration profiles. The effect of the various parameters on all profiles is graphically presented and explained. The present data of shear stress, the Nusselt, and the Sherwood numbers calculated by DTM are matched and verified by numerical method data and literature for the novelty of the work. The strength of the work is to analyze the validation, training, and testing by using Levenberg-Marquardt artificial neural network (ANN). This ANN is verified by mean square error, error histogram, and regression analysis.

Backpropagation of Levenberg–Marquardt artificial neural networks for reverse roll coating process in the bath of Sisko fluid

Backpropagation of Levenberg–Marquardt artificial neural networks for reverse roll coating process in the bath of Sisko fluid

Experimental analysis and Levenberg-Marquardt artificial neural network predictions of heat transfer, friction factor, and efficiency of thermosyphon flat plate collector with MgO/water nanofluids

Experimental estimations have been made for the thermal efficiency, heat transfer coefficient, and friction factor of solar collector circulated of MgO/water nanofluid at thermosyphon and then predicted with artificial neural network-Levenberg-Marquardt algorithm. The trials were carried out between 10:00 a.m. and 16:30 p.m. at 0.1%, 0.5%, and 1.0% volume loadings, respectively. Results indicate that, under the Reynolds numbers of 143 and 345, a maximum rise in Nusselt number has been observed for 1.0% vol. of nanofluid is to be 21.48% and 37.28%, higher than water data. Maximum friction factor penalties of 1.14-fold and 1.27-fold at 1.0 vol% and at Reynolds numbers of 143 and 345, respectively, were simultaneously observed over water data. Using water and 1.0 vol% of nanofluid in a collector, the efficiencies are 57.15%–65.47%, respectively. The relative variances of the formulae created to assess the factor in friction and Nusselt numbers are within ±2.598%. Meanwhile, the used Levenberg-Marquardt algorithm predicts very high accurate values when compared with the experimental data. The obtained correlation coefficients for Nusselt number, transfer in heat, factor in friction, and efficiency is 0.98795, 0.98627, 0.9991, and 0.98512, respectively, these values indicate that, the used algorithm and its network construction is very congruent with the experimental data.

Intelligent computing with Levenberg–Marquardt artificial neural network for Carbon nanotubes-water between stretchable rotating disks

Hybrid Nano fluid has emerged to be an important field of study due to its better thermal performance compared to other Nano fluids. The problem of carbon nanotubes rotating between two stretchable discs while suspended in water is investigated in this research. Due to numerous uses of this problem, such as metal mining, drawing plastic films, and cooling continuous filaments, this problem is essential to industry. Considerations here include suction/injection, heat radiation, and the Darcy-Forchheimer scheme with convective boundary conditions. The partial differential equations are reduced to ordinary differential equations by using appropriate transformation. To examine the approximate solution validation, training and testing procedures are interpreted and the performance is verified through error histogram and mean square error results. To describe the behavior of flow quantities, several tabular and graphical representations of a variety of physical characteristics of importance are presented and discussed in detail. The basic aim of this research is to examine the behaviour of carbon nanotubes (nanoparticles) between stretchable disks while considering the heat generation/absorption parameter by using the Levenberg–Marquardt technique of artificial neural network. Heat transfer rate is accelerated by a decrease in velocity and temperature and an increase in the nanoparticle volume fraction parameter which is a significant finding of the current study.

Levenberg Marquardt artificial neural network model for self‐organising networks implementation in wireless sensor network

AbstractThe Wireless Sensor Network needs to become a dynamic and adaptive network to conserve energy stored in the wireless sensor network node battery. This dynamic and adaptive network sometimes are called SON (Self Organizing Network). Several SON concepts have been developed such as routing, clustering, intrusion detection, and other. Although several SON concepts already exist, however, there is no concept for SON in dynamic radio configuration. Therefore, the authors’ contribution to this field would be proposing a dynamic and adaptive Wireless Sensor Network node radio configuration. The significance of their work lies in the modelling of the SON network that builds based on our measurement in the real‐world jungle environment. The authors propose input parameters such as SNR, the distance between the transmitter and receiver, and frequency as the static parameter. For adaptive parameters, we propose bandwidth, spreading factor, and its most important parameter such as power for data transmission. Using the Levenberg Marquardt Artificial Neural Network (LM‐ANN) self‐organise Network model, power reduction and optimisation from 20 dBm to 14.9 dBm for SNR 3, to 11.5 dBm for SNR 6, and to 12.9 dBm for SNR 9 all within a 100‐m range can be achieved. With this result, the authors conclude that we can use LM‐ANN for the wireless sensor network SON model in the jungle environment.

Performance of different models in iron ore price prediction during the time of commodity price spike

Future prediction of commodity price based on available data is very important for mining investors and operators. Commodity prices cointegrate and show Granger causality to and from one another. This research reviewed five different estimation techniques which are Bivariate Non-Linear Regression (BNLR), Multiple Linear Regression (MLR), Multiple Non-Linear Regression (MNLR) as well as logsig and tansig model of Levenberg-Marquardt Artificial Neural Network modelling to simulate the future iron ore price based on 12 other monthly commodity prices and indices including LNG, aluminium, nickel, silver, Australian coal, zinc, gold, oil, tin, copper, lead, and Commodity Price Index (Metals).Six different models were tested in the paper to forecast the iron ore prices from 1 to 6 months over 10 months period. Linear model (purelin) using Levenberg-Marquardt technique was able to exhibit the best forecast result with average accuracy of 5.92% for 1 month ahead, 9.48% for 2 months, 11.21% for 3 months, etc. It is important to highlight that high accuracy is achieved (accuracy under 5% between forecasts and actuals in 40–50% cases) by purelin model for up to 2 months forecast for the period between July 2020 and April 2021. This indicates that prediction of iron ore price for the coming month is possible for up to 2 months period using the purelin model. It can be noted that the period tested was unstable for iron ore prices where rapid surge in iron ore price was observed. Same principle can be applied in the time of next commodity price cycle.

Machine Learning-Based Estimation of the Compressive Strength of Self-Compacting Concrete: A Multi-Dataset Study

This paper aims at performing a comparative study to investigate the predictive capability of machine learning (ML) models used for estimating the compressive strength of self-compacting concrete (SCC). Seven prominent ML models, including deep neural network regression (DNNR), extreme gradient boosting machine (XGBoost), gradient boosting machine (GBM), adaptive boosting machine (AdaBoost), support vector regression (SVR), Levenberg–Marquardt artificial neural network (LM-ANN), and genetic programming (GP), are employed. Four experimental datasets, compiled in previous studies, are used to construct the ML-based methods. The models’ generalization capabilities are reliably evaluated by 20 independent runs. Experimental results point out the superiority of the DNNR, which has excelled other models in three out of four datasets. The XGBoost is the second-best model, which has gained the first rank in one dataset. The outcomes point out the great potential of the utilized ML approaches in modeling the compressive strength of SCC. In more details, the coefficient of determination (R2) surpasses 0.8 and the mean absolute percentage error (MAPE) is always below 15% for all datasets. The best results of R2 and MAPE are 0.93 and 7.2%, respectively.

Prediction of Vehicular Traffic Flow Using Levenberg-Marquardt Artificial Neural Network Model: Italy Road Transportation System

In the last decades, the Italian road transport system has been characterized by severe and consistent traffic congestion and in particular Rome is one of the Italian cities most affected by this problem. In this study, a LevenbergMarquardt (LM) artificial neural network heuristic model was used to predict the traffic flow of non-autonomous vehicles. Traffic datasets were collected using both inductive loop detectors and video cameras as acquisition systems and selecting some parameters including vehicle speed, time of day, traffic volume and number of vehicles. The model showed a training, test and regression value (R2) of 0.99892, 0.99615 and 0.99714 respectively. The results of this research add to the growing body of literature on traffic flow modelling and help urban planners and traffic managers in terms of the traffic control and the provision of convenient travel routes for pedestrians and motorists.

Intelligent computing based supervised learning for solving nonlinear system of malaria endemic model

<abstract><p>A repeatedly infected person is one of the most important barriers to malaria disease eradication in the population. In this article, the effects of recurring malaria re-infection and decline in the spread dynamics of the disease are investigated through a supervised learning based neural networks model for the system of non-linear ordinary differential equations that explains the mathematical form of the malaria disease model which representing malaria disease spread, is divided into two types of systems: Autonomous and non-autonomous, furthermore, it involves the parameters of interest in terms of Susceptible people, Infectious people, Pseudo recovered people, recovered people prone to re-infection, Susceptible mosquito, Infectious mosquito. The purpose of this work is to discuss the dynamics of malaria spread where the problem is solved with the help of Levenberg-Marquardt artificial neural networks (LMANNs). Moreover, the malaria model reference datasets are created by using the strength of the Adams numerical method to utilize the capability and worth of the solver LMANNs for better prediction and analysis. The generated datasets are arbitrarily used in the Levenberg-Marquardt back-propagation for the testing, training, and validation process for the numerical treatment of the malaria model to update each cycle. On the basis of an evaluation of the accuracy achieved in terms of regression analysis, error histograms, mean square error based merit functions, where the reliable performance, convergence and efficacy of design LMANNs is endorsed through fitness plot, auto-correlation and training state.</p></abstract>

Employing artificial neural network for effective biomass prediction: An alternative approach

Wood products and energy production originating from harnessing the tree biomass require optimizing the forest management process so as to ensure the sustainability of the forest ecosystems. This optimization can also act as a preventive factor towards limiting the consequences of climate change given it is a contributing factor for maintaining healthy ecosystems. To that end, the need to develop methodologies that enable accurate prediction of biomass is more than evident. Nonlinear seemingly unrelated regressions, Dirichlet regressions, and the Levenberg-Marquardt artificial neural network (LMANN) modeling techniques have been applied for whole tree (above and below ground) biomass prediction as well as its components. We conducted a comparative analysis of these approaches using destructively sampled black pine (Pinus nigra Arnold.) trees. Results showed that the LMANN models are flexible and fit tree biomass data with the highest accuracy. Inherent deviations of the biomass data from regression assumptions further support the use of LMANN models as a reliable and promising alternative to the other modeling approaches.

Axial Capacity Prediction of Concrete-Filled Steel Tubular Short Members Using Multiple Linear Regression and Artificial Neural Network

The estimation of the ultimate capacity of rectangular or circular shaped steel tubular members filled with concrete, such as columns, beams, and beam-column connections, requires a detailed structural study to be carried out. Therefore, identify the concrete strength the member subjected to axial-load only. Using the Levenberg-Marquardt artificial neural network, this paper investigates the concrete-filled steel tubular (CFT) members axial strength. 201 experimental specimens were collected from the literature to obtain the best results, and a wide range of geometric and material properties of CFT members were included. The proposed design and specimens illustrate the practicality and effectiveness of the chosen CFT column approach to classify real structural results.

Backpropagation of Levenberg Marquardt artificial neural networks for wire coating analysis in the bath of Sisko fluid

In the artificial neural networks domain, the Levenberg-Marquardt technique is novel with convergent stability and generates a numerical solution of the wire coating system for Sisko fluid flow (WCS-SFF) through regression plots, histogram representations, state transition measures, and means squared errors. In this paper, the analysis of fluid flow problem based on WCS-SFF is studied with a new application of intelligent computing system via supervised learning mechanism using the efficacy of neural networks trained by Levenberg-Marquardt algorithm (NN-TLMA). The original mathematical formulation in terms of PDEs for WCS-SFF is converted into dimensionless nonlinear ODEs. The data collection for the projected NN-TLMA is produced for parameters associated with the system model WCS-SFF influencing the velocity using the explicit Runge-Kutta technique. The training, validation, and testing processes of NN-TLMA are utilized to evaluate the obtained results of WCS-SFF for various cases, and a comparison of the obtained results is performed with reference data set to check the accuracy and effectiveness of the proposed algorithm NN-TLMA for the analysis of non-Newtonian fluid problem-related WCS-SFF. The proposed NN-TLMA for solving the WCS-SFF is effectively confirmed through state transition dynamics, mean square error, regression analyses, and error histogram studies. The powerful consistency of suggested outcomes with reference solutions indicates the validity of the framework, and the accuracy of 10-8 to 10-6 is also achieved.

Performance of smoothly clipped absolute deviation as a variable selection method in the artificial neural network‐based QSAR studies

AbstractA hybrid of smoothly clipped absolute deviation (SCAD) and Levenberg‐ Marquardt artificial neural network (LM‐ANN) was used as an efficient approach (SCAD‐LM‐ANN) in the ANN‐based quantitative structure–activity relationship (QSAR) studies. The proposed technique (SCAD‐LM‐ANN) exploits the useful shrinkage nature of SCAD in the reduction of high dimensional data prior to modeling by the robust LM‐ANN method. The performance of the method was examined by establishing a QSAR model between about 3224 Dragon‐derived descriptors and biological activities (pEC50) for a set of thioacetamide/acetanilide derivatives as HIV inhibitors. SCAD method with the tenfold random splitting of data was applied to the data set in the absence of compounds in the external test set. A number of 11 descriptors were selected at a λ with the lowest cross‐validation error (λmin). The selected descriptors were used as inputs of the LM‐ANN modeling method. All parameters affecting model performance were optimized, and the LM‐ANN model with the architecture of 5‐5‐1 was selected as the optimal QSAR model. Several statistical parameters such as determination coefficient (R2) and mean square error (MSE) were calculated for the predicted pEC50 values for the external test set and the whole data set through the leave one out (LOO) technique. The results ( = 0.92, MSEtest = 0.12, = 0.81, and MSELOO = 0.12) prove the generalizability and predictability of the proposed SCAD‐LM‐ANN model. According to the established relationship in the recommended QSAR model, new derivatives were designed and suggested as new active HIV inhibitors for further studies. The accuracy of the suggested compounds was studied and confirmed by analyzing the ligand–receptor (LR) interactions derived from the molecular docking studies.

Intelligent computing with Levenberg-Marquardt artificial neural networks for nonlinear system of COVID-19 epidemic model for future generation disease control.

The aim of this work is to design an intelligent computing paradigm through Levenberg–Marquardt artificial neural networks (LMANNs) for solving the mathematical model of Corona virus disease 19 (COVID-19) propagation via human to human interaction. The model is represented with systems of nonlinear ordinary differential equations represented with susceptible, exposed, symptomatic and infectious, super spreaders, infection but asymptomatic, hospitalized, recovery and fatality classes, and reference dataset of the COVID-19 model is generated by exploiting the strength of explicit Runge–Kutta numerical method for metropolitans of China and Pakistan including Wuhan, Karachi, Lahore, Rawalpindi and Faisalabad. The created dataset is arbitrary used for training, validation and testing processes for each cyclic update in Levenberg–Marquardt backpropagation for numerical treatment of the dynamics of COVID-19 model. The effectiveness and reliable performance of the design LMANNs are endorsed on the basis of assessments of achieved accuracy in terms of mean squared error based merit functions, error histograms and regression studies.

QSAR modeling of anti-HIV activity for DAPY-like derivatives using the mixture of ligand-receptor binding information and functional group features as a new class of descriptors

An accurate QSAR model was developed for the prediction of the anti-HIV activities of a set of DAPY-like derivatives as new non-nucleoside reverse transcriptase inhibitors (NNRTIs). The ligand–receptor (LR) interactions for all compounds were studied by the docking of compounds in the active site of appropriate receptors. The binding information of LR complexes at the best pose was called the molecular docking descriptors (MDDs). The mixture of 10 MDDs with about 154 simple, functional group counts was used as a new group of descriptors in the QSAR study of DAPY-like compounds. Among the 164 mixed descriptors, seven descriptors were selected as the most effective descriptors using the linear stepwise regression (SR) method and used as inputs of the artificial neural network (ANN) model. Levenberg–Marquardt (LM) method was used for the training of ANN through the backpropagation of errors algorithm. The Levenberg–Marquardt artificial neural network (LM-ANN) model with the architecture of 6-4-1 was selected as the optimal model. The predictability of the LM-ANN model was estimated by applying the external test set and the leave-one-out (LOO) method. The mean square errors (MSEs) and coefficient of determination (R2) values for the test set were 0.16 and 0.89, respectively. The Q2 value for the LOO method was calculated as 0.72. The results obtained demonstrated that a mixture of MDDs and functional group counts provided the required information for the construction of a QSAR model with acceptable prediction ability.

LM-ANN-based QSAR model for the prediction of pEC50 for a set of potent NNRTI using the mixture of ligand–receptor interaction information and drug-like indexes

A combination of ligand–receptor interactions and drug-like indexes have been used to develop a quantitative structure–activity relationship model to predict anti-HIV activity (pEC50) of 73 azine derivatives as non-nucleoside reverse transcriptase inhibitors. Ligand–receptor interactions were derived from the best position (best pose) of studied compounds, as ligands, in the active site of receptors using Autodock 4.2 software and named as molecular docking descriptors. The drug-like indexes were calculated using DRAGON 5.5 software. Two groups of descriptors were mixed, and the stepwise regression method was used for the selection of the most relevant descriptors. Four selected descriptors were subsequently used to construct the quantitative structure–activity relationship model using the Levenberg–Marquardt artificial neural network method. Dataset was randomly divided into the train (53 compounds), validation (10 compounds) and test set (10 compounds). The best model was selected according to the lowest mean square error value of the validation set. The accuracy and predictability of the model were evaluated using test and validation sets and the leave-one-out technique. According to the predicted results, the coefficient of determination of the test set (R2 = 0.86) and all data ( $${Q}_{LOO}^{2}$$ = 0.73) were acceptable. The mean square error value for the test set was equal to 0.11. The obtained results emphasized the good prediction ability and generalizability of the developed model in the prediction of pEC50 values for new compounds.

Mobility Prediction Assisted Call Admission Control Model for Mobile Heterogeneous Cellular Network

The exponential rise in mobile wireless networks and allied demands has revitalised academia-industries to achieve more efficient and quality of service (QoS) centric communication systems. On the other hand, high pace up surge of mobile communication and internet services have alarmed industries to ensure seamless QoS provision to the mobile user's (MUs) irrespective of geographical locations and movement pattern. However, majority of conventional approaches apply either received signal strength or bandwidth availability to make CAC decision, which seems limited especially under mobile user scenario. In this paper a highly robust mobility prediction assisted (QoS demand centric) handoff management model is developed for call admission control in heterogeneous cellular network. Our proposed method employs Levenberg-Marquardt artificial neural network (LM-ANN)-based mobility prediction, target cell prioritisation and optimal selection for QoS centric CAC decision. Simulation results have shown low handoff and reduced call blocking probability over that makes it suitable for HCNs with constrained network conditions.

Prediction of two-dimensional gas chromatography time-of-flight mass spectrometry retention times of 160 pesticides and 25 environmental organic pollutants in grape by multivariate chemometrics methods

A quantitative structure–retention relation (QSRR) study was conducted on the retention times of 160 pesticides and 25 environmental organic pollutants in wine and grape. The genetic algorithm was used as descriptor selection and model development method. Modeling of the relationship between selected molecular descriptors and retention time was achieved by linear (partial least square; PLS) and nonlinear (kernel PLS: KPLS and Levenberg-Marquardt artificial neural network; L-M ANN) methods. The QSRR models were validated by cross-validation as well as application of the models to predict the retention of external set compounds, which did not have contribution in model development steps. Linear and nonlinear methods resulted in accurate prediction whereas more accurate results were obtained by L-M ANN model. The best model obtained from L-M ANN showed a good R2 value (determination coefficient between observed and predicted values) for all compounds, which was superior to those of other statistical models. This is the first research on the QSRR of the compounds in wine and grape against the retention time using the GA-KPLS and L-M ANN.