Artificial Neural Model Research Articles

Modeling and simulation of the enzymatic kinetics for the production of Galactooligosaccharides (GOS) using an Artificial Neural Network hybrid model

Due to the complexity of biochemical systems, the development of traditional phenomenological models is limited if the underlying mechanics are not entirely known. As an alternative, hybrid model frameworks, consisting of data-driven models complemented with first principles models like conservation law, are starting to be used for complex systems. In this work, a comparison of the modeling capabilities between a data-driven model and a hybrid model was developed. The enzymatic production of Galactooligosaccharides (GOS) with the effect of metallic ions was considered as case study. Compared with the experimental results, predictions from data-driven model achieve an R2 of 0.9188 in the best training fold, and the hybrid model an R2 of 0.9696 in the best training fold. Illogical predictions were avoided by including non-phenomenological first-principles constraints into the hybrid model. Finally, an optimization analysis was carried out to find the highest GOS productivity using the hybrid model, optimization results present a deviation of 5.99 % compared to the highest productivity found from experimental data.

Improvement of the classical artificial neural network simulation model of the parabolic trough solar collector outlet temperature and thermal efficiency using the conformable activation functions

The outlet temperature and the thermal efficiency predictions of the PTSC are essential parameters in the solar thermal power system. Therefore, it is crucial to have a prediction model that can predict its spatiotemporal behavior to the greatest extent possible. For that, the present study provides improved models of the classical ANN model by using ELU, swish, softplus, logsig, and tansig functions through the exponential transfer function to improve the outlet temperature prediction performance and thermal efficiency of PTSC. The PTSC’s outlet temperature simulation showed that the ANN model of topology 6-2-1 with the conformable swish transfer function (cswish) was the best model among nine other studied models, with an R2 and Radj2 of 0.9988, and MAE of 0.0052. The simulation of PTSC’s thermal efficiency proved that the conformable functions logsig, tansig, ELU and the classical ELU function involved in the ANN model for 6-3-1 achieved better results with approximately 95% precision. With the non-integer activation functions, was reduced a neuron in the hidden layer, which leads to the simplicity of the prediction model compared to the use of the classical activation functions. The proposed activation functions do not require high-performance features according to their execution times. Therefore, it could be a good alternative for engineers to apply in other thermal energy systems to make their prediction models more accessible and practical in the control and optimization process.

Comparative Analysis of Audio Processing Techniques on Doppler Radar Signature of Human Walking Motion Using CNN Models.

Artificial intelligence (AI) radar technology offers several advantages over other technologies, including low cost, privacy assurance, high accuracy, and environmental resilience. One challenge faced by AI radar technology is the high cost of equipment and the lack of radar datasets for deep-learning model training. Moreover, conventional radar signal processing methods have the obstacles of poor resolution or complex computation. Therefore, this paper discusses an innovative approach in the integration of radar technology and machine learning for effective surveillance systems that can surpass the aforementioned limitations. This approach is detailed into three steps: signal acquisition, signal processing, and feature-based classification. A hardware prototype of the signal acquisition circuitry was designed for a Continuous Wave (CW) K-24 GHz frequency band radar sensor. The collected radar motion data was categorized into non-human motion, human walking, and human walking without arm swing. Three signal processing techniques, namely short-time Fourier transform (STFT), mel spectrogram, and mel frequency cepstral coefficients (MFCCs), were employed. The latter two are typically used for audio processing, but in this study, they were proposed to obtain micro-Doppler spectrograms for all motion data. The obtained micro-Doppler spectrograms were then fed to a simplified 2D convolutional neural networks (CNNs) architecture for feature extraction and classification. Additionally, artificial neural networks (ANNs) and 1D CNN models were implemented for comparative analysis on various aspects. The experimental results demonstrated that the 2D CNN model trained on the MFCC feature outperformed the other two methods. The accuracy rate of the object classification models trained on micro-Doppler features was 97.93%, indicating the effectiveness of the proposed approach.

Multi-Objective Optimization of Air-Cooled Perforated Micro-Pin Fin Heat Sink Via an Artificial Neural Network Surrogate Model Coupled With NSGA-II

Abstract This research aims to create an artificial neural network (ANN) regression model for predicting the performance parameters of the perforated micro-pin fin (MPF) heat sinks for various geometric parameters and inflow conditions. A three-dimensional computational fluid dynamics (CFD) simulation system is developed to generate dataset samples under different operational conditions, which are specified using Latin hypercube sampling (LHS). An ANN model is first obtained by optimizing the model hyper-parameters, which are then deployed to learn from the input feature space that consists of perforation diameter, perforation location, and inflow velocity. For accurate training of the ANN, the model is trained over a range of uniformly distributed data points in the input feature space. The developed multi-layer model predicted Nusselt number and friction factor with the mean absolute percentage error of 4.45% and 1.80%, respectively. Subsequently, the developed surrogate model is used in the optimization study to demonstrate the application of the surrogate model. A multi-objective non-dominated sorting genetic algorithm (NSGA-II) is used to perform the optimization of the perforation location, diameter, and inflow conditions. Negative of the Nusselt number and friction factor are chosen as objectives to minimize. A Pareto front is obtained from the optimization study that shows a set of optimal solutions. Thermal performance of the perforated MPF is increased between 11.5% and 39.77%. The optimizer selected a significantly smaller hole diameter at a higher location and a faster speed to maximize the Nusselt number and minimize the friction factor.

Artificial Neural Network Performance Modeling and Evaluation of Additive Manufacturing 3D Printed Parts

This research article presents a comprehensive study on the performance modeling of 3D printed parts using Artificial Neural Networks (ANNs). The aim of this study is to optimize the mechanical properties of 3D printed components through accurate prediction and analysis. The study focuses on the widely employed Fused Deposition Modeling (FDM) technique. The ANN model is trained and validated using experimental data, incorporating input parameters such as temperature, speed, infill direction, and layer thickness to predict mechanical properties including yield stress, Young's modulus, ultimate tensile strength, flexural strength, and elongation at fracture. The results demonstrate the effectiveness of the ANN model with an average error below 10%. The study also reveals the significant impact of process parameters on the mechanical properties of 3D printed parts and highlights the potential for optimizing these parameters to enhance the performance of printed components. The findings of this research contribute to the field of additive manufacturing by providing valuable insights into the optimization of 3D printing processes and facilitating the development of high-performance 3D printed components.

Revealing the diagnostic value and immune infiltration of senescence-related genes in endometriosis: a combined single-cell and machine learning analysis.

Introduction: Endometriosis is a prevalent and recurrent medical condition associated with symptoms such as pelvic discomfort, dysmenorrhea, and reproductive challenges. Furthermore, it has the potential to progress into a malignant state, significantly impacting the quality of life for affected individuals. Despite its significance, there is currently a lack of precise and non-invasive diagnostic techniques for this condition. Methods: In this study, we leveraged microarray datasets and employed a multifaceted approach. We conducted differential gene analysis, implemented weighted gene co-expression network analysis (WGCNA), and utilized machine learning algorithms, including random forest, support vector machine, and LASSO analysis, to comprehensively explore senescence-related genes (SRGs) associated with endometriosis. Discussion: Our comprehensive analysis, which also encompassed profiling of immune cell infiltration and single-cell analysis, highlights the therapeutic potential of this gene assemblage as promising targets for alleviating endometriosis. Furthermore, the integration of these biomarkers into diagnostic protocols promises to enhance diagnostic precision, offering a more effective diagnostic journey for future endometriosis patients in clinical settings. Results: Our meticulous investigation led to the identification of a cluster of genes, namely BAK1, LMNA, and FLT1, which emerged as potential discerning biomarkers for endometriosis. These biomarkers were subsequently utilized to construct an artificial neural network classifier model and were graphically represented in the form of a Nomogram.

Artificial Neural Networks-Based Route Selection Model for Multimodal Freight Transport Network During Global Pandemic

The global pandemic caused major disruptions in all supply chains. Road transport has been particularly affected by the challenges posed by the COVID-19 pandemic. The selection of an efficient and effective route in multimodal freight transport networks is a crucial part of transport planning to combat the challenges and sustain supply chain continuity in the face of the global pandemic. This study introduces a novel optimal route selection model based on integrated fuzzy logic approach and artificial neural networks. The proposed model attempts to identify the optimal route from a range of feasible route options by measuring the performance of each route according to transport variables including, time, cost, and reliability. This model provides a systematic method for route selection, enabling transportation planners to make smart decisions. A case study is conducted to exhibit the proposed model's applicability to real pandemic conditions. According to the findings of the study, the proposed model can accurately and effectively identify the best route and provides transportation planners with a viable option to increase the efficiency of multimodal transport networks. In conclusion, by proposing an innovative and efficient strategy for route selection in complex transport systems, our research significantly advances the field of transportation management.

An integrated modelling framework for multiple pollution source identification in surface water

Pollution source identification is vital in water safety management. An integrated simulation-optimization modelling framework comprising a process-based hydrodynamic water quality model, artificial neural network surrogate model and particle swarm optimization (PSO) was proposed to achieve rapid, accurate and reliable pollution source identification. In this study, the hydrodynamics and water quality processes in a straight lab-based flume were simulated to test pollution source identification under steady flow conditions. Additionally, the pollution source identification in the unsteady flow conditions was examined using a real-life estuary, specifically the Yangtze River estuary. First, we developed two process-based models to simulate hydrodynamics and water quality in the flume and estuary. Then, the data generated from the process-based models were used to develop surrogate models. Three typical artificial neural networks (ANNs) algorithms: backpropagation (BP), radial basis function (RBF) and general regression neural networks (GRNN) were selected to develop surrogates for process-based models (PBMs), and they were coupled with PSO algorithm to achieve the hybrid modelling framework for pollution source identification. Our results showed that hybrid PBM-ANNs-PSO models could be applied to identify the pollution source and quantify release intensity in spatial distribution when the discharge type was assumed as the point source with a continuous release. Multiple-performance criteria metrics, in terms of the coefficient of determination, root-mean-square error, mean absolute error, evaluated the model performance as “Excellent prediction”. The BP-PSO models consistently appear to be the top-performing source identification model within the developed models, with most cases of relative error (RE) values lower than 5%. The new insights from the hybrid modelling framework would provide useful information for the local government agency to make reasonable decisions regarding pollution source identification issues.

Digital Prediction of the Purchase Price of Fresh Tea Leaves of Enshi Yulu Based on Near-Infrared Spectroscopy Combined with Multivariate Analysis.

In this study, near-infrared spectroscopy (NIRS) combined with a variety of chemometrics methods was used to establish a fast and non-destructive prediction model for the purchase price of fresh tea leaves. Firstly, a paired t-test was conducted on the quality index (QI) of seven quality grade fresh tea samples, all of which showed statistical significance (p < 0.05). Further, there was a good linear relationship between the QI, quality grades, and purchase price of fresh tea samples, with the determination coefficient being greater than 0.99. Then, the original near-infrared spectra of fresh tea samples were obtained and preprocessed, with the combination (standard normal variable (SNV) + second derivative (SD)) as the optimal preprocessing method. Four spectral intervals closely related to fresh tea prices were screened using the synergy interval partial least squares (si-PLS), namely 4377.62 cm-1-4751.74 cm-1, 4755.63 cm-1-5129.75 cm-1, 6262.70 cm-1-6633.93 cm-1, and 7386 cm-1-7756.32 cm-1, respectively. The genetic algorithm (GA) was applied to accurately extract 70 and 33 feature spectral data points from the whole denoised spectral data (DSD) and the four characteristic spectral intervals data (FSD), respectively. Principal component analysis (PCA) was applied, respectively, on the data points selected, and the cumulative contribution rates of the first three PCs were 99.856% and 99.852%. Finally, the back propagation artificial neural (BP-ANN) model with a 3-5-1 structure was calibrated with the first three PCs. When the transfer function was logistic, the best results were obtained (Rp2 = 0.985, RMSEP = 6.732 RMB/kg) by 33 feature spectral data points. The detection effect of the best BP-ANN model by 14 external samples were R2 = 0.987 and RMSEP = 6.670 RMB/kg. The results of this study have achieved real-time, non-destructive, and accurate evaluation and digital display of purchase prices of fresh tea samples by using NIRS technology.

Soft computing procedure to optimize the electrospinning parameters of polyacrylonitrile nanofibrous air filter

Following the previous work, this research aims to highlight the applicability of a new strategy for engineering morphological properties of nanofibrous filters using a model consisting of artificial neural analytical model, and genetic algorithm. The main idea is to engineer the morphological properties of nanofibrous filters before mass production. To collect data, a central composite design of experiment considering parametersch as polymer concentration (10, 12, and 14 wt % ), applied voltage (14, 18, and 22 kV ), distance (10, 15, and 20 cm ), and time (15, 30, and 45 min ) is employed. Results showed that reduced the cost value from 0.1477 to 0.1183 incororating a polymer concentration of 10.08 wt % , an applied voltage of 16.09 kV , a distance of 20 cm , and a time of 27.17 min . The optimized filter possessed a filtration efficiency of 95% and a pressure drop of 169.2530 Pa for a particle size of 300 nm.

The complexity of financial development and economic growth nexus in Syria: A nonlinear modelling approach with artificial neural networks and NARDL model

This study investigates the complex interaction between financial development (FD) and economic growth (EG) in Syria from 1980 to 2018 using advanced nonlinear modeling techniques including artificial neural network VAR models, nonlinear causality tests, and nonlinear autoregressive distributed lag (NARDL) models. The results indicate that linear models are inadequate to capture the data patterns, necessitating nonlinear approaches. The artificial neural network VAR model reveals a nonlinear connection between FD and EG. The nonlinear causality test confirms that FD causes EG in a nonlinear manner. The NARDL (1, 1, 0, 1, 1) model is selected based on Akaike information criterion and diagnostics. The findings show a long-run equilibrium and short-run dynamics between FD and EG in Syria. Moreover, positive changes in FD have stronger, more persistent effects on EG compared to negative changes, implying asymmetry. Additionally, the impact of FD on EG is nonlinear, varying with FD levels. These results support recent studies suggesting a nonlinear nexus between FD and EG. They also lend support to the finance-led growth theory while opposing the “too much finance harms growth” hypothesis. The study offers policy implications for Syria to create conditions conducive to positive FD shocks and adopt a long-term perspective regarding FD-EG policies.

GIS-based multicriteria and artificial neural network (ANN) investigation for the assessment of groundwater vulnerability and pollution hazard in the Braga shallow aquifer (Central Tunisia): A critical review of generic and modified DRASTIC models

Groundwater vulnerability and pollution hazard in the Braga shallow aquifer were assessed through an integrated GIS-based multicriteria analysis and Artificial Neural Network (ANN) approach, using DRASTIC and DRASTIC-LU models. The DRASTIC model integrates seven geological parameters. The DRASTIC-LU model includes an eighth parameter in addition to the previous ones. This parameter is the land use that represents the human source of groundwater pollution. The DRASTIC map showed four classes: very low (12.06%), low (81.88%), moderate (5.16%) and high (0.9%), where the vulnerability index ranged between 43 and 159. The DRASTIC-LU vulnerability index ranged between 53 and 204 and showed five classes: very low (3.10%), low (14.06%), moderate (17.11%), high (27.08%) and very high (38.65%). The DRASTIC-LU vulnerability map indicated that the high pollution risk is imposed by the intensive vegetable cultivation and the domestic wastewater. The pollution hazard index (PHI) was calculated based on the ANN modelling, using the land-use as an input and the vulnerability as a hidden layer. The DRASTIC model-based PHI map showed six classes: rare hazard (8.6%), very low (30.97%), low (6.18%), moderate (51.45%), high (2.43%) and very high (0.37%). While, The DRASTIC-LU model-based PHI map (PHILU) showed seven classes: rare hazard (2.91%), very low (11.9%), low (12.33%), moderate (13.78%), high (9.23%), very high (15.46%) and extremely hazardous (34.39%). The validation of these maps indicated that the DRSTIC-LU-based PHI is more reliable as it accurately identifies the hazardous zones.

Development of an artificial neural network EMMS drag model for the simulation of fluidized beds in chemical looping combustion

The current work presents an Artificial Neural Network EMMS (ANN-EMMS) drag scheme, developed specifically for applications related to fluidized bed (FB) reactors in chemical looping combustion (CLC); the examined particles correspond to the main bed materials encountered in CLC: ilmenite, hematite and titanium oxide. The data that feed the model are generated by a custom-built MATLAB code that solves the EMMS equations. The developed ANN-EMMS model, along with three other drag models, i.e. the Gidaspow, the Wen-Yu and the conventional EMMS, are tested for their performance in CFD simulations for the conditions of the 1 MWth pilot scale FB air reactor of TU Darmstadt. The results of the simulations utilizing the ANN-EMMS followed by those of the conventional EMMS are the closest to published experimental data for the pressure profile in the reactor, while the ilmenite conversion as predicted by both models is in good agreement with published numerical data.

Learning heterogeneous delays in a layer of spiking neurons for fast motion detection.

The precise timing of spikes emitted by neurons plays a crucial role in shaping the response of efferent biological neurons. This temporal dimension of neural activity holds significant importance in understanding information processing in neurobiology, especially for the performance of neuromorphic hardware, such as event-based cameras. Nonetheless, many artificial neural models disregard this critical temporal dimension of neural activity. In this study, we present a model designed to efficiently detect temporal spiking motifs using a layer of spiking neurons equipped with heterogeneous synaptic delays. Our model capitalizes on the diverse synaptic delays present on the dendritic tree, enabling specific arrangements of temporally precise synaptic inputs to synchronize upon reaching the basal dendritic tree. We formalize this process as a time-invariant logistic regression, which can be trained using labeled data. To demonstrate its practical efficacy, we apply the model to naturalistic videos transformed into event streams, simulating the output of the biological retina or event-based cameras. To evaluate the robustness of the model in detecting visual motion, we conduct experiments by selectively pruning weights and demonstrate that the model remains efficient even under significantly reduced workloads. In conclusion, by providing a comprehensive, event-driven computational building block, the incorporation of heterogeneous delays has the potential to greatly improve the performance of future spiking neural network algorithms, particularly in the context of neuromorphic chips.

РОЛЬ ГЕНЕРАТИВНЫХ МОДЕЛЕЙ ИИ И ЯВЛЕНИЕ НЕЙРОННЫХ ЯЗЫКОВЫХ МОДЕЛЕЙ ТИПА GPT В ДЕТЕРМИНАЦИИ ЦИФРОВОЙ ЭКОНОМИКИ

LLM such as ChatGPT have the potential to revolutionize research in economics and other sciences. New technologies tend to save labor and resources, stimulating the optimization of most economic processes. The article reviews the main prospects of development of generative models of artificial intelligence and neural language models. An analysis of examples of the use of artificial intelligence tools in economics is conducted. Particular attention is paid to recent trends and current problems in the development of machine learning, artificial intelligence and neural networks. Since the digital transformation of the economy is all-encompassing, AI developments are making it much easier for humans to interact with advanced technologies. This reduces the time it takes to perform various tasks and increases productivity.

Determination of moisture content and contaminated blank dried figs (Ficus carica L.) using dielectric property and artificial neural network

AbstractDried figs are a garden produce that must be graded after harvesting. Moisture levels and contaminated blank are two of the most critical effective elements on the marketability of dried figs, and they are highly related to fig quality. In the present research, an intelligent system was employed to classify dried figs based on moisture content levels and infected blank fruits. Capacitance characteristics, average diameter, and fruit area were all taken into account in this study. The dried fig dielectric constant was measured at six different frequency levels: 12, 22, 32, 42, 52, and 62 MHz. The best frequency was then chosen using the improved distance evaluation feature selection approach. Image processing was also used to determine the average diameter and area of the figures. Following that, the dielectric constant of the most effective frequency, the average diameter, and the area of the fruit were used as input parameters in the artificial neural network classification model to classify and describe the moisture and porosity level of the dried fig. The most essential dielectric constant information relating to moisture and porosity level was at frequencies of 22 and 52 MHz, respectively. Finally, classification accuracy of 95.7% for moisture and 91.3% for porosity level was attained. The results demonstrated the excellent performance and capabilities of the proposed approach for rating the internal quality of dried figs.

Socio-economic correlation analysis and hybrid artificial neural network model development for provincial waste electrical and electronic equipment generation forecasting in China

Ability to forecast the waste electrical and electronic equipment (WEEE) generation can help formulate a robust future WEEE management system. Previous studies applied forecasting models, such as the grey model and artificial neural networks, to predict WEEE generation from a country perspective, leading to less accurate forecasts due to huge socio-economic differences in rural and urban areas. Additionally, there has been the incompatibility of a single forecasting model for all WEEE types, and this remained a research gap. Taking advantage of respective forecasting models, this study presents a hybrid model, Grey Artificial Neural Network, to forecast the WEEE generation of 31 province-level regions in China while evaluating the socio-economic analysis of seven WEEE types via Pearson correlation analysis. More than 70% of WEEE from province-level regions strongly correlates (R > ±0.8) with the gross domestic product and the population, whereas some top WEEE-generating province-level regions (i.e., Tianjin and Shanghai) correlate weakly to moderately. The root mean square error and mean absolute error of the developed Grey Artificial Neural Network hybrid model are the lowest at 8.29 and 6.48, compared to the grey model (13.53 and 11.13) and back-propagation neural network (9.21 and 7.22). Though the Grey Artificial Neural Network hybrid model has the lowest error, posterior mean square deviation ratio analysis indicates that this hybrid model is only suitable for washing machines, refrigerators, color televisions, and personal computers (urban area), while the back-propagation neural network is suitable for monochrome televisions, air conditioners, and personal computers (rural area). Compared to 2019, it is projected that an additional 32.92 M units of WEEE will be generated by 2025, suggesting that China should build at least 15 extra recycling centers (14% more based on 2016) to handle the increased WEEE generation. This study provides policy implications for effective WEEE monitoring and collection systems to build resilient WEEE management.

Artificial Neural Networks-Based Fault Diagnosis Model for Distribution Network

With many branch lines in radiant distribution networks, diagnosing faults in a distribution network is very difficult. It is of great significance to identify different types of faults quickly and accurately for the stable operation of the power grid. This research presents a fault identification model for a distribution network based on artificial neural networks. The principal component analysis first extracts features from transitory data in a distribution network. The resulting low-dimensional data is subsequently used to update the artificial neural network model. The artificial neural network may also identify the type of fault. The proposed model’s fault detection accuracy is improved over the traditional approach by examining distribution network fault data during the simulation test.

Evaluation of 5G and Fixed-Satellite Service Earth Station (FSS-ES) Downlink Interference Based on Artificial Neural Network Learning Models (ANN-LMS).

Fifth-generation (5G) networks have been deployed alongside fourth-generation networks in high-traffic areas. The most recent 5G mobile communication access technology includes mmWave and sub-6 GHz C-bands. However, 5G signals possibly interfere with existing radio systems because they are using adjacent and co-channel frequencies. Therefore, the minimisation of the interference of 5G with other signals already deployed for other services, such as fixed-satellite service Earth stations (FSS-Ess), is urgently needed. The novelty of this paper is that it addresses issues using measurements from 5G base stations (5G-BS) and FSS-ES, simulation analysis, and prediction modelling based on artificial neural network learning models (ANN-LMs). The ANN-LMs models are used to classify interference events into two classes, namely, adjacent and co-channel interference. In particular, ANN-LMs incorporating the radial basis function neural network (RBFNN) and general regression neural network (GRNN) are implemented. Numerical results considering real measurements carried out in Malaysia show that RBFNN evidences better accuracy with respect to its GRNN counterpart. The outcomes of this work can be exploited in the future as a baseline for coexistence and/or mitigation techniques.

Symbiosis of an artificial neural network and models of biological neurons: Training and testing

In this paper, we show the possibility of creating and identifying the features of an artificial neural network (ANN), which consists of mathematical models of biological neurons. The FitzHugh–Nagumo (FHN) system is used as a paradigmatic model demonstrating basic neuron activities. First, in order to reveal how biological neurons can be embedded within an ANN, we train the ANN with nonlinear neurons to solve a basic image recognition problem with an MNIST database; next, we describe how FHN systems can be introduced into this trained ANN. After all, we show that an ANN with FHN systems inside can be successfully trained with improved accuracy comparing with first trained ANN and then with inserted FHN systems. This approach opens up great opportunities in terms of the direction of analog neural networks, in which artificial neurons can be replaced by more appropriate biological ones.