Artificial Neural Network Research Articles

Adaptive AI-based surrogate modelling via transfer learning for DEM simulation of multi-component segregation

Segregation of granular materials is a critical challenge in many industries, often aimed at being controlled or minimised. The discrete element method (DEM) offers valuable insights into this phenomenon. However, calibrating DEM models is a crucial, albeit time-consuming, step. Recently, using machine learning (ML)-based surrogate models (SMs) in the calibration process has emerged as a promising solution. Nevertheless, developing such SMs is challenging due to the high number of DEM simulations required for training. Additionally, choosing a suitable ML model is not trivial. This study aims to develop SMs that effectively link particle-particle and particle-wall DEM interaction parameters to segregation of a multi-component mixture. We evaluate several ML models, ranging from artificial neural networks to ensemble learning, that are trained on a very cost-effective dataset, employing Bayesian optimisation with cross-validation to tune their hyperparameters. Next, we introduce a novel transfer learning (TL)-based approach that leverages knowledge from a few scenarios to handle new “unseen” ones. This method enables the construction of adaptive SMs for unseen scenarios, such as a new initial configuration (IC) of granular mixtures, without the need for a full-sized dataset. Our findings indicate that Gaussian process regression (GPR) efficiently builds accurate SMs on a very small dataset. We also demonstrate that only a few samples are required to build an accurate SM for the unseen IC, which significantly reduces the data preparation burden. By incorporating one and five samples from unseen scenarios to update the TL-GPR-based surrogate model, the SM’s performance (based on \\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}) on unseen scenarios improves by 17 and 47%, respectively. The insights and methodology presented in this study will facilitate and accelerate the development of accurate SMs for DEM calibration, assisting in developing reliable DEM models in a shorter timeframe.

Simulated synapse loss induces depression-like behaviors in deep reinforcement learning

Deep Reinforcement Learning is a branch of artificial intelligence that uses artificial neural networks to model reward-based learning as it occurs in biological agents. Here we modify a Deep Reinforcement Learning approach by imposing a suppressive effect on the connections between neurons in the artificial network—simulating the effect of dendritic spine loss as observed in major depressive disorder (MDD). Surprisingly, this simulated spine loss is sufficient to induce a variety of MDD-like behaviors in the artificially intelligent agent, including anhedonia, increased temporal discounting, avoidance, and an altered exploration/exploitation balance. Furthermore, simulating alternative and longstanding reward-processing-centric conceptions of MDD (dysfunction of the dopamine system, altered reward discounting, context-dependent learning rates, increased exploration) does not produce the same range of MDD-like behaviors. These results support a conceptual model of MDD as a reduction of brain connectivity (and thus information-processing capacity) rather than an imbalance in monoamines—though the computational model suggests a possible explanation for the dysfunction of dopamine systems in MDD. Reversing the spine-loss effect in our computational MDD model can lead to rescue of rewarding behavior under some conditions. This supports the search for treatments that increase plasticity and synaptogenesis, and the model suggests some implications for their effective administration.

Optimization of CED dry film thickness using artificial neural network for automobile industries

Optimization of CED dry film thickness using artificial neural network for automobile industries

Mechanical assessment for enhancing hybrid composite performance through silane treatment using RSM and ANN

The study utilized Response Surface Methodology (RSM) and Artificial Neural Networks (ANN) to identify the optimal combination of three key factors for producing polymer composites: nanoparticle percentage, silane concentration, and silane dipping duration. RSM and Analysis of Variance (ANOVA) were applied to explore the relationships between these variables and their effects on composite properties. Additionally, the ANN was used to analyze multiple factors, revealing a strong correlation between predicted and observed results. The findings highlighted that silane treatment was the most influential factor in enhancing the composite's flexural strength. Fiber-related properties, particularly the duration of silane dipping, significantly impacted both flexural strength and hardness. Nanoparticles further strengthened the fiber matrix by encouraging effective agglomeration. The ANN demonstrated 95% accuracy in predicting flexural strength and hardness, and its predictions were validated by comparing them with experimental data and the regression model's outcomes. Silane concentration also notably influenced flexural properties. The RSM analysis identified the optimal combination for maximizing flexural strength and hardness: 5% nanoparticles, 10% silane concentration, and a 20-minute silane dipping time.

Analysing the Influence of Augmented Reality on Organization Performance via Supply and Logistics Value Chain Functions: A Hybrid ANN-PLS Model Assessment in the Gulf Cooperation Council Region

Background: Despite the resurgence of interest in augmented reality (AR) due to Industry 4.0 and its ability to resolve several challenges faced by current business models, comprehensive research examining the capabilities of AR in supply chain management (SCM) and logistics remains limited. This article aims to investigate the potential effects of AR technology on organizational performance through the mediation role of SCM and logistics value chain functions to address the existing knowledge gap. Methods: This research employed a cross-sectional design and an explanatory survey as a deductive approach for hypothesis development. The primary data collection method involved the self-administration of a questionnaire to furniture suppliers located in the Gulf Cooperation Council (GCC), including six countries. Of the 656 questionnaires submitted to suppliers, 483 were considered usable, yielding a response rate of 73.6%. The research utilized partial least squares structural equation modelling (PLS-SEM) and artificial neural network (ANN) techniques to evaluate the gathered data. Results: The current paper’s statistical evidence demonstrates that AR implementation has a positive impact on the supply and logistics value chain activities and organizational performance of furniture suppliers in the GCC region. Moreover, it illustrates that the design and planning variable of supply chain value dominates as the primary predictor of organization performance. The results indicated that the ANN strategy provided a more comprehensive explanation of internally generated constructs compared to the PLS-SEM technique. Conclusions: This study demonstrates its usefulness by advising furniture industry decision-makers on what to avoid and what aspects to consider when creating plans and regulations. The report also suggests operations managers apply machine learning (ANN) for prediction and decision-making in supply and operations value chains. This essay looks at how the AR and resource-based supply value chain view may affect company performance across countries, firm sizes, and ages.

Informatics-based design of polyaniline-carbon nanotube thermoelectric nanocomposite using ANN and GA

Abstract Conducting polymer and carbon nanotube (CNT) based nanocomposites have emerged as prospective thermoelectric (TE) materials due to their potential application in flexible electronics. Non-conventional charge and heat transport in these nanocomposites, presents the possibility to enhance the TE conversion efficiency, given by ZT. However, the highly non-linear and complex association of structure and composition with the overall TE properties have hindered the development of any general strategy to develop high ZT nanocomposites. Here, we implement artificial neural network (ANN) and genetic algorithm (GA) to develop data driven models followed by optimization to design high efficiency nanocomposites based on CNT dispersed in polyaniline (PANI) matrix. Our models suggest that CNT concentration plays the most crucial role in determining ZT. Non-dominated Pareto optimal solutions consisting of different combinations of design variables are obtained by multi-objective optimization. Although a range of optimal solutions span over different regions of the search space, in general we note that longer CNTs boost Seebeck coefficient (S) and electrical conductivity (σ), and smaller length lowers thermal conductivity (k), while higher diameter of CNTs increase S and lowers σ and k. The results provide a general guideline for developing CNT-PANI nanocomposites with enhanced TE figure of merit.

Rapid Classification of Sugarcane Nodes and Internodes Using Near-Infrared Spectroscopy and Machine Learning Techniques

Accurate and rapid discrimination between nodes and internodes in sugarcane is vital for automating planting processes, particularly for minimizing bud damage and optimizing planting material quality. This study investigates the potential of visible-shortwave near-infrared (Vis–SWNIR) spectroscopy (400–1000 nm) combined with machine learning for this classification task. Spectral data were acquired from the sugarcane cultivar Khon Kaen 3 at multiple orientations, and various preprocessing techniques were employed to enhance spectral features. Three machine learning algorithms, linear discriminant analysis (LDA), K-Nearest Neighbors (KNNs), and artificial neural networks (ANNs), were evaluated for their classification performance. The results demonstrated high accuracy across all models, with ANN coupled with derivative preprocessing achieving an F1-score of 0.93 on both calibration and validation datasets, and 0.92 on an independent test set. This study underscores the feasibility of Vis–SWNIR spectroscopy and machine learning for rapid and precise node/internode classification, paving the way for automation in sugarcane billet preparation and other precision agriculture applications.

Experimental investigation on optimizing wire-cut EDM process parameters of hybrid MMC(AA7475/ZrO2/Gr) using artificial neural network

Abstract Wire cut Electrical discharge machining (WCEDM) is a widely used method for machining complex shapes in advanced materials like metal matrix composites (MMCs) and hybrid metal matrix composites (HMMCs). To address these challenges, this study focusses on the wire-cut EDM (WCEDM) process of a workpiece made from zirconium dioxide and graphite-reinforced aluminium alloy 7475 with a molybdenum electrode. The effects of input process variables such as peak current (IP), pulse-on-time (TON), and flushing pressure (PF) on the output response features are investigated. These output responses include material removal rate (MRR), surface roughness (SR), and wire wear ratio (WWR). To optimise the process parameters, the Taguchi design technique is used. An artificial neural network (ANN) with a feed-forward back propagation architecture is utilised to find the best fit for the optimisation challenges. ANN predicted the results with an accuracy of 97.81% for MRR, 97.95% for SR and 95.865% for WWR. The results reveal that the WCEDM of AA7475/ZrO2/Gr with a molybdenum electrode achieved minimal WWR and SR, while maximizing the MRR.

Machine learning-based prediction of scour depth evolution around spur dikes

ABSTRACT Spur dikes are pivotal elements in river training, serving to mitigate the dynamic alterations induced by river degradation and aggradation. Traditionally, scour prediction models have relied on regression techniques, but the advent of soft computing and machine learning has offered opportunities for enhanced accuracy. This study focuses on the development of hybrid machine-learning models, including eXtreme Gradient Boosting (XGBoost), random forest (RF), convolutional neural network–long short-term memory, and artificial neural network, optimized using genetic algorithms to predict both temporal scour depth variation and maximum scour depth around the initial spur dike in a series. The analysis reveals strong associations between scour depth and various parameters such as non-dimensional time, spacing, channel width, time-averaged velocity, and densimetric Froude number. The models are established through an iterative process involving four predictor combinations. Results demonstrate XGBoost as the top-performing model, consistently exhibiting superior performance with R2 of 0.99, root mean square error (RMSE) of 0.012, and mean absolute error of 0.008 during training, and R2 of 0.96, RMSE of 0.044, and Kling–Gupta efficiency of 0.98 during testing for predicting temporal scour depth. For non-dimensional maximum scour depth, it reached R2 of 0.99 and RMSE of 0.005 in training, with R2 > 0.91 across all combinations during testing. Although RF showcases commendable accuracy, it slightly lags in precision compared to XGBoost.

Evaluating Advanced Machine Learning Models for Histopathological Diagnosis of Hansen Disease.

Leprosy is a neglected infectious disease caused by Mycobacterium leprae and Mycobacterium lepromatosis and remains a public health challenge in tropical regions. Therefore, the development of technological tools such as machine learning (ML) offers an opportunity to innovate strategies for improving the diagnosis of this complex disease. To validate the utility of different ML models for the histopathological diagnosis of Hansen disease. An observational study was conducted where 55 H&E-stained tissue slides from leprosy patients and 51 healthy skin controls were analyzed using microphotographs captured at various magnifications. These images were categorized based on histopathological findings and processed using the Cross-Industry Standard Process for Data Mining methodology for designing and training ML models. Five types of ML models were evaluated using standard metrics such as accuracy, sensitivity, and specificity, emphasizing data normalization as a fundamental step in optimizing model performance. The artificial neural network (ANN) model demonstrated an accuracy of 70%, sensitivity of 74%, and specificity of 65%, demonstrating its ability to identify leprosy cases with moderate precision. The receiver operating characteristic curve of the ANN model showed an area under the curve of 0.71. Conversely, models such as decision trees, logistic regression, and random forests showed similar accuracy results but with slightly lower sensitivity, potentially indicating a higher risk of false negatives in detecting leprosy-positive cases. The ANN model emerges as a promising alternative for leprosy detection. However, further refinement of these models is necessary to enhance their adaptability across different clinical settings and participation in patient care.

How Can Machine Learning Contribute to the Credit Scoring Process in SME Lending?

This research investigates the potential of machine learning to enhance the credit scoring process in small and medium-sized enterprise (SME) lending. SMEs are vital to global economies, accounting for over 90\% of businesses and more than half of total employment worldwide, according to The World Bank (2023). In emerging economies, SME lending is particularly crucial, providing access to capital that supports economic growth and development. However, assessing the credit risk of SMEs poses unique challenges due to limited financial histories and variable cash flows, making accurate credit scoring essential for lenders. Machine learning offers promising solutions for the risk assessment process in SME lending, leveraging algorithms and large datasets to identify patterns and improve decision-making. By applying techniques such as artificial neural networks, decision tree learning, gradient boosting, logistic regression, random forest, and support vector machines, machine learning can enhance the accuracy and efficiency of credit scoring models. This research explores how these machine learning methods can reduce biases, improve risk prediction, and ultimately support sustainable lending practices by enabling lenders to make more informed credit decisions. The findings aim to illustrate the transformative potential of machine learning in SME credit scoring, offering a framework for financial institutions seeking to integrate advanced data-driven technologies in the credit assessment process.

Machine learning-based clustering analysis of the banking green credit for the renewable energy industry

Credit policies for clean and renewable energy businesses play a crucial role in supporting carbon neutrality efforts to combat climate change. Clustering the credit capacity of these companies to prioritize lending is essential given the limited capital available. Support Vector Machine (SVM) and Artificial Neural Network (ANN) are two robust machine learning algorithms for addressing complex clustering problems. Additionally, hyperparameter selection within these models is effectively enhanced through the support of a robust heuristic optimization algorithm, Particle Swarm Optimization (PSO). To leverage the strength of these advanced machine learning techniques, this paper aims to develop SVM and ANN models, optimized with the PSO, for the clustering problem of green credit capacity in the renewable energy industry. The results show low Mean Square Error (MSE) values for both models, indicating high clustering accuracy. The credit capabilities of wind energy, clean fuel, and biomass pellet companies are illustrated in quadrant charts, providing stakeholders with a clear view to adjust their credit strategies. This helps ensure the efficient operation of banking green credit policies.

Revolutionizing spinal interventions: a systematic review of artificial intelligence technology applications in contemporary surgery

Leveraging its ability to handle large and complex datasets, artificial intelligence can uncover subtle patterns and correlations that human observation may overlook. This is particularly valuable for understanding the intricate dynamics of spinal surgery and its multifaceted impacts on patient prognosis. This review aims to delineate the role of artificial intelligence in spinal surgery. A search of the PubMed database from 1992 to 2023 was conducted using relevant English publications related to the application of artificial intelligence in spinal surgery. The search strategy involved a combination of the following keywords: "Artificial neural network," "deep learning," "artificial intelligence," "spinal," "musculoskeletal," "lumbar," "vertebra," "disc," "cervical," "cord," "stenosis," "procedure," "operation," "surgery," "preoperative," "postoperative," and "operative." A total of 1,182 articles were retrieved. After a careful evaluation of abstracts, 90 articles were found to meet the inclusion criteria for this review. Our review highlights various applications of artificial neural networks in spinal disease management, including (1) assessing surgical indications, (2) assisting in surgical procedures, (3) preoperatively predicting surgical outcomes, and (4) estimating the occurrence of various surgical complications and adverse events. By utilizing these technologies, surgical outcomes can be improved, ultimately enhancing the quality of life for patients.

Nordic environmental resilience: balancing air quality and energy efficiency by applying artificial neural network

Maintaining public health and environmental safety in the Nordic nations calls for a strict plan to define exact benchmarks on air quality and energy efficiency. This study investigates the complicated interaction of decentralized energy production (DEP) with energy efficiency, and air quality index in the Nordic nations from 1990 to 2022 using System GMM and Artificial Neural Network (ANN) approach. Our research explored positive role of decentralized energy production and technological advancement to propel notable increases in energy efficiency, hence lowering pollution expressed as PM2.5 level. Our research indicates, however, that although international trade, GDP and urbanization assist to enhance energy efficiency, they also contribute to pollution by raising PM2.5 Level by higher energy usage. Furthermore damaging to environmental quality is the persistent link shown by economic disparity and the energy price index with increased degrees of pollution and less energy efficiency. Policy frameworks must devised sustainable development policy (decentralized energy production) to significantly improve energy efficiency and lower the amount of pollution. This calls for proper urban planning and a close observation of the possible drawbacks of growing GDP, trade, economic disparity, and energy expenses.

Durability of Forging Tools Used in the Hot Closed Die Forging Process—A Review

The article presents the classification of the wear mechanisms of forging tools. The durability of dies can be enhanced through a variety of methods, including the selection of appropriate hot working tool steel, the application of effective heat treatment, the utilization of advanced surface engineering techniques, and the incorporation of lubricating and cooling agents. Two popular methods of tool regeneration, such as re-profiling and laser regeneration, are presented. The issue of numerical wear prediction based on the Archard model, the correlation of this model with experimental results, low-cycle fatigue (HTLCF), and an alternative method based on artificial neural networks are discussed. The paper aims to present currently known wear mechanisms and the methods of increasing and predicting tool durability.

Estimation of the axial capacity of high-strength concrete-filled steel tube columns using artificial neural network, random forest, and extreme gradient boosting approaches

Estimation of the axial capacity of high-strength concrete-filled steel tube columns using artificial neural network, random forest, and extreme gradient boosting approaches

Landslide susceptibility mapping using logistic regression, random forests, and artificial neural networks: a case study in Mariana/MG, Brazil

Landslide susceptibility mapping using logistic regression, random forests, and artificial neural networks: a case study in Mariana/MG, Brazil

Efficient High Heating Value estimation using Latin Hypercube Sampling and Artificial Neural Network-based approach.

To maximize energy recovery in waste-to-energy (WTE) systems, the High Heating Value (HHV) of municipal solid waste (MSW) must be accurately estimated. To forecast the HHV of MSW, this study proposes a unique method that combines an Artificial Neural Network (ANN) model with Latin Hypercube Sampling (LHS), with a focus on Solan City, Himachal Pradesh, India. In the present study, the elemental characteristics of waste have been used to deal with uncertainty and to find the suitable parameters responsible for the HHV of the MSW. Initially, Latin Hypercube Sampling (LHS) has been used to deal with uncertainty in the elemental composition of MSW, which includes carbon (C), hydrogen (H), nitrogen (N), sulfur (S), and oxygen (O) content. This elemental composition has been used as input parameters to the ANN model for predicting the HHV of MSW. The network 5-28-5-1 offered a minimum MAPE value of 2.18%, MSE, RMSE and R2 values are 0.012, 0.107 and 0.767, respectively. Thereafter, a synaptic weight approach was used to find the most significant parameters responsible for HHV in MSW. It was observed that carbon is the most suitable parameter for HHV of MSW. By dealing with the uncertainty in MSW characteristics, the integration of LHS strengthens the robustness of the model. The results offer an accurate and economical approach for HHV estimation, which will be useful for improving the MSW management and WTE conversion procedures.

Discovery of Alloy Catalysts for Ammonia Decomposition by Machine Learning-Based Prediction of Adsorption Energies

Ammonia decomposition has gained significant attention as an eco-friendly method for hydrogen production because it creates no carbon dioxide emissions. While Ru catalysts are known for their high activity in ammonia decomposition, their high cost makes them uneconomical for commercial use. Therefore, it is essential to explore novel alloy catalysts composed of inexpensive elements with high catalytic performance. Nitrogen adsorption energies serve as key descriptors indicating the catalytic performance for ammonia decomposition, and first-principle calculations can compute these energies. However, the screening of numerous alloy catalyst candidates through extensive first-principle calculations and experimental validations remains time-consuming due to the vast number of potential candidates. To address this, artificial intelligence and machine learning models are being developed to quickly predict catalyst performance, efficiently searching for promising catalyst candidates. In this study, we developed a machine-learning-based method to rapidly predict nitrogen adsorption energies using a graph-based artificial neural network, thereby efficiently searching for novel catalysts for ammonia decomposition. Our training dataset included the nitrogen adsorption energies of 30 pure transition metal catalyst candidates, as well as binary alloy catalyst candidates, including core-shell and intermetallic compounds. As a result, we successfully identified 12 catalyst candidates composed of inexpensive elements that are likely to exhibit catalytic performance comparable to Ru catalysts.

Structural-functional organization of the visual system in ensuring goal-directed activity

A review of the results of half a century of research into the visual system as a hierarchical structure is presented: a multichannel, multilayer “pyramid”, each layer of which has a different spatiotemporal resolution, but together provides an invariant description of images for their classification, decision making, organization of eye movements and target search. An analysis of the multichannel organization of the human visual system was carried out, as the most effective and most economical. The “periscope and telescopic vision” systems, unique in their morphological and functional characteristics, are identified, providing gaze translation and recognition when searching and achieving a goal. Models of the pyramidal organization of the visual system have justified their existence by having an exceptional influence on the development of engineering solutions for the design of recognition systems operating in real time and the creation of artificial neural networks.