Artificial Neuronal Network Research Articles

Extended human agency: towards a teleological account of AI

This paper analyzes human-machine interrelation concerning artificial neuronal networks (ANNs) from a teleological point of view. The paper argues that AI cannot be understood adequately in terms of subjectivity or objectivity but rather as a new kind of teleological relationship that holds between human and artificial performances of intelligence. Thereby, AI is understood as an enactivist extension of human agency, both in instrumental and moral terms. This hybrid account will be distinguished from four alternative accounts of human-machine relations: (i) the simulation account, according to which AI simulates human rationality; (ii) the instrumentalist account, according to which AI is just a tool; (iii) the anthropomorphic account, according to which AI is human-like; and (iv) the indifference account, according to which AI will merge with human rationality due to technological progress. Against these four accounts, the paper argues for a teleological account of AI as extended human agency that is part of the human lifeworld. By focusing on the teleological interrelation of socially grounded databases and algorithms, the paper finally develops an account of responsible AI that considers its specific relatedness with human actions, purposes, and intentions by means of language. Understanding human-machine relations in terms of extended agency finally allows to tackle the question of how to avoid the problems of AI bias and opacity.

Prediction of Rheological Properties of Flour From Physicochemical Properties Using Multiple Regression Techniques and Artificial Neuronal Networks

ABSTRACTThis study has two main objectives: (i) to determine the physicochemical and rheological properties of different flours and (ii) to estimate the alveograph parameters obtained as a result of experimental studies. In this context, physicochemical (protein, ash, falling number, wet gluten, gluten index, Zeleny, and delayed sedimentation) and alveograph parameters (P, L, G, W, P/L, and IE) of 150 different bread and pastry flours were determined. Multiple regression analysis (MRA) and artificial neural network (ANN) methods were then used to predict alveograph results from this experimentally obtained data set. Root mean square error (RMSE), mean absolute error (MAE), Nash‐Sutcliffe (NSEC), and relative error (RE) performance statistics were used to evaluate the CS prediction capabilities of the methods. It was found that the flours were in the range of 11.01%–13.82% protein, 325–403 s falling number, and 30–61 mL Zeleny and delayed sedimentation values. The ANN method showed better predictive performance than the regression‐based method. W was the best estimated parameter in the ANN model. This was followed by G, L, Ie, P/L, and P values. Considering the RMSE value of the W parameter, it was observed that the ANN method provided an improvement of 5.16, 1.76, and 2.15 times compared to the regression method for the training, validation, and test sets, respectively.

A predictive model for the estimation of industrial [formula omitted] emissions for IoT-based devices

The paper is devoted to solving the problem, associated with increasing the forecasting accuracy of pollutant concentration level based on PM2.5 dust. The LANN model (LANN – Lagrange powered Artificial Neuron Network), proposed in the paper, allows you to take into account the mutual influence of pollution sources on the concentration level at monitoring points, which, on the one hand, allows to manage the volume of emissions from industrial enterprises in such a way as to avoid penalties and, on the other hand, allows the regulatory agencies to determine points for environmental control, at which permissible standards will be most likely exceeded. The suggested solution makes it possible to use low-cost IoT-based air quality monitoring devices, which were primarily considered not applicable due to their low accuracy and exposure to the influence of short-term stochastic factors; besides, no models were available, which would be capable to provide the necessary accuracy and forecast horizon on the base of IoT data. At the same time, the obtained model has low requirements for computing resources compared to dispersion models and is dependent on the accuracy of weather information; besides, the model allows you to take into account trends and obtain better forecast values in terms of accuracy than the models, that use regression methods, neural network models and ML models, stochastic models and their ensembles. The application of the model improves well-known pollutant dispersion estimation calculation techniques through the use of measurement data flow and dynamic adaptation to changing conditions The model was trained on the dataset over the period 2022-2023 collected from 5 monitoring devices, which were installed in the catchment area of the group, incl. 13 industrial points of PM2.5 pollutant sources.

Investigating green hydrogen production operated by redundant energy on a solar PV mini-grid through matlab simulation and artificial neural network

Green hydrogen is an important part of the transition to a net zero economy, due to its potential to deliver high amounts of energy without pollution. One important opportunity to produce green hydrogen is the use of redundant energy on solar PV mini-grids. Redundant energy is the potential generation of a mini-grid that is never utilized due to low demand and the mismatch of peak solar PV generation and peak demand. In this study, a typical rural community solar PV mini-grid is simulated using Matlab Simulink and incorporated with a water electrolyser for hydrogen production. Having run the simulation for 365 days of the year, it was revealed that the mini-grid can produce a daily average of 1.18–2.16 kg, a monthly average of 36.59–64.84 kg and a yearly total of 609.26 kg of hydrogen. Subsequently, an artificial neuron network was trained with data acquired from the simulation. With varying configurations of the number of neurons and hidden layers, it was found that the model successfully predicts hydrogen production on the mini-grid with the highest performance (RMSE = 39.85 g and R2 = 0.9979) seen when the number of hidden layers is 40 and the number of neurons per hidden layer is 3.

Développement et validation d’un outil de dépistage systématique des fragilités sociales des patients atteints de cancer : l’outil DEFCO

Literature suggests that patients from deprived backgrounds are less likely to adhere to their treatments, continue to expose themselves to risk factors and, as a result, have poorer health outcomes. It is therefore crucial to identify these vulnerable populations early on, in order to provide them with tailored and reinforced care. The primary aim of this research is to construct and validate a systematic screening tool for identifying patients at highest risk of social vulnerability due to deprivation, through the use of psychometric techniques. This tool is intended to be easily used by healthcare professionals, to provide tailored and targeted care throughout the patient's journey. This study involves the development and assessment of a screening tool, along with a self-questionnaire and a decision support tool incorporating an artificial neural network. It is a prospective, monocentric, 2-stage psychometric validation study. This study has demonstrated the successful development of the self-questionnaire using psychometric methodology. The tool was found a good performance in screening social vulnerabilities. This validated self-questionnaire is an easy-to-use tool, allowing systematic screening for social vulnerabilities for cancer patients. This early identification allows to reinforce patient's pathway in order to avoid disruption. The integration of the tool in an artificial neuron network system allows to automate and disseminate this method of deprived patients' detection, while limiting the workload for the staff.

REVOLUTIONIZING POWER TRANSFORMER FAULT DIAGNOSIS THROUGH COGNITIVE ARTIFICIAL INTELLIGENCE AND DISSOLVED GAS ANALYSIS INTEGRATION

The research introduces a cognitive artificial intelligence (CAI) model that leverages dissolved gas analysis (DGA) to investigate power transformer faults. Conventional fault interpretation methods using DGA are limited in accuracy and uncertainty. In response, the proposed CAI model utilizes cognitive learning and direct interaction to achieve remarkably accurate fault identification without the need for supervised training. By extracting fault features through key gas ratio limitations. However, the CAI model also has a gap in data perception due to the information sensory challenges. Using gas ratios based on the conventional fault interpretation methods in the latest study still limited data perception of the CAI model to only three or four gas ratios. Thus, this study aims to increase data perception by extracting fault features through ten gas ratio limitations. The proposed CAI model's performance is validated, outperforming traditional methods like the Duval triangle method, Duval pentagon method, Doernenburg ratio method, and Roger ratio method, as well as common AI approaches including artificial neuron network, long short-term memory, nearest neighbor classifiers, support vector machine, ensemble classifiers, and decision trees. Notably, the CAI model's success rate in fault type identification stands at an impressive 98.04%. A distinctive trait of the CAI model is its autonomous knowledge accumulation and enhancement, enabled by inferring-fusion information and sensor-based knowledge integration. This intrinsic learning ability further contributes to its exceptional fault diagnosis accuracy. The proposed CAI model showcases promising potential for revolutionizing power transformer fault investigation and diagnosis, mitigating unplanned outages, and ultimately bolstering power system reliability.

Mimicking Axons by Conductive Metal Lines Formed by Photocatalytically Grown Gold

Nanoionic memristive devices based on the field−driven migration of metal cations and oxygen vacancies have gathered high interest in the scientific community for their capability to technically mimic synaptic connections. Less attention has been given to long−range global plasticity in artificial neuronal networks. The aim of our research is to use light−stimulated growth of gold metal lines to technically mimic the growth of long−range axonal connections for neuromorphic computing architectures. For this purpose, we investigate the formation of metal lines by UV−light−stimulus−driven photocatalytic deposition of gold on titanium dioxide layers. In this study, we present a significant contribution that stems from our active involvement in CRC 1461- Neurotronics, a pivotal initiative funded by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation). Our research falls under Project-ID 434434223 – SFB 1461, within the broader context of Bio-inspired information pathways.0.8-mm wide photoactive titanium dioxide lines are defined by photolithography on a Sodalime glass substrate. We deposited a 70 nm thick titanium dioxide line on a 10 nm thick indium tin oxide adhesion layer by DC sputtering. The substrate was then heat treated for 90 minutes at 400°C and subsequently cooled on a metal plate to achieve the photocatalytically active anatase phase of titanium dioxide [1]. The growth process is conducted in a beaker by reduction from an aqueous gold chloride precursor solution (HAuCl4). In each step of the growth process, the substrate is placed in a beaker containing 15 ml of the precursor. The solution is obtained by dissolving 15 mg of HAuCl4 powder in 60 ml of deionized water. We utilize a UV LED with the peak wavelength of 365 nm and an approximate intensity of 1.8 - 2.7 mW/cm² to illuminate the sample in the beaker. The illumination time and UV intensity are crucial for this dynamic process. Here, we illuminated the sample for 12 hours and refreshed the solution every 2 hours. After each refreshment of the solution, the substrate is washed thoroughly with deionized (DI) water and is dried under nitrogen gas. We analyze the surface properties of grown gold utilizing scanning electron microscopy (SEM) and measure the conductance of the grown lines by applying voltage via the needles of a hard contact probe station.In the first experiments the growth of gold clusters on the surface is observed instead of the homogeneous gold coverage of the photolithographically defined titanium dioxide lines, which are needed for long-range electrical connections. Then, we coincidentally observed that the presence of a scratch on the titanium dioxide line led to the growth of continuous gold lines on the damaged area. Subsequently, deliberate scratches were introduced manually using tweezers, varying in both size and thickness. SEM images depict islands and covered areas, with notable growth observed on the scratched regions. Remarkably, wider scratches exhibit a proportionally increased width of the grown lines, and a uniform line devoid of any gaps is distinctly visible on broader scratches. We observed particle growth on thinner lines. However, unlike the uniform alignment observed in wider scratches, these particles did not coalesce into a continuous line. Instead, they remained separated from each other, exhibiting small gaps between individual particles. The morphological analysis of the grown gold line using SEM reveals a crystalline morphology of particles.As the primary objective of this research is to mimic axons with gold lines, ensuring conductivity is crucial for obtaining long−range electrical connections mimicking axons. To achieve this, voltage was applied using the needles of a hard contact probe station, and the resulting current was measured. The linear behavior observed in the I–V diagram of the wider line serves as a confirmation of its conductance (≈0.003 S).

Prediction of time-dependent concrete mechanical properties based on advanced deep learning models considering complex variables

Evaluating the mechanical properties such as compressive strength, tensile strength, and modulus of elasticity (MOE) of concrete is crucial for the design, construction, and quality control of engineering projects. Developing a comprehensive theoretical-based numerical model to predict the concrete mechanical properties is considered challenging, owing to the diversity of admixtures and curing environment affecting the performance of modern concrete. The deep learning (DL) model gives a feasible approach as data-driven methods for accurate prediction of various material properties these years. However, predicting time-dependent mechanical properties of modern concrete with complex mix constituents by DL algorithms is still confined so far. In this article, artificial neuron network (ANN) and long short-term memory (LSTM) based DL models were established to predict concrete compressive strength, tensile strength, and MOE at different ages, considering the mix design, curing condition, curing time and tension testing method as input variables. LSTM outperformed ANN in terms of the statistical indicators for accuracy evaluation with high training and testing time costs. Bidirectional modification and multi-head attention mechanism were implemented to further improve the prediction accuracy of LSTM. It is shown that the multi-head attention bidirectional LSTM had more precise predicting results (R2=0.9355 for compressive strength, R2=0.8930 for tensile strength, R2=0.9584 for MOE) and fitting of the mechanical property-age curves. The LSTM-based models provided a promising prediction tool for time-dependent concrete mechanical properties with complex input parameters.

Conversion Factor Estimation of Stacked Eucalypt Timber Using Supervised Image Classification with Artificial Neural Networks

Stacked timber is quantified in-store units and then adjusted with a conversion factor for volume estimation in cubic meters, which is important for the wood trade in South America. However, measuring large quantities accurately can be challenging. Digital image processing and artificial intelligence advancements offer promising solutions, making research in this area increasingly attractive. This study aims to estimate conversion factors of stacked Eucalyptus grandis timber using supervised image classification with Artificial Neuronal Network (ANN). Measured data and photographs from an experiment involving thirty stacks of timber were used to achieve this. The conversion factor was determined using photographic methods that involved the applications of equidistant points and ANN and subsequently validated with values observed through the manual method. The ANN method produced more accurate conversion factor estimates than the equidistant points method. Approximately 97% of the ANN estimates were within the ±1% error class, even when using low-resolution digital photographs.

Application of artificial neuronal networks for the prediction of monthly photovoltaic power using tilt angle as input in Rabat city, Morocco

The use of artificial neural networks (ANNs) is attracting interest in the field of renewable energies, particularly for predicting operational parameters of photovoltaic (PV) systems. Thus, the present study aims to develop an ANN model for predicting the output power of a photovoltaic module. In this context, two ANN models with two and three hidden layers, respectively, were trained and validated to predict the average monthly photovoltaic power Pp produced by a 0. 325 kW photovoltaic module by varying the tilt angle from 0° to 90° of Rabat city, Morocco. The training data used in this study include longitude, latitude, calculated photovoltaic power, tilt angle, and months as inputs, while the predicted Pp is the target. These two proposed models are trained using the Levenberg-Marquardt (LM) optimiser. Furthermore, to confirm the reliability of each of these two proposed ANN models, existing photovoltaic power data were used to validate the two proposed models by predicting the average monthly power of the photovoltaic module. For the tilt angle of 34°, which corresponds to the latitude of the city of Rabat, the ANN model with three hidden layers performed better with an R2 value of 0.9997.

An artificial neuronal network coupled with a genetic algorithm to optimise the production of unsaturated fatty acids in Parachlorella kessleri

In this study, an Artificial Neural Network-Genetic Algorithm (ANN-GA) approach was successfully applied to optimise the physicochemical factors influencing the synthesis of unsaturated fatty acids (UFAs) in the microalgae P. kessleri UCM 001. The optimized model recommended specific cultivation conditions, including glucose at 29 g/L, NaNO3 at 2.4 g/L, K2HPO4 at 0.4 g/L, red LED light, an intensity of 1000 lx, and an 8:16-h light-dark cycle. Through ANN-GA optimisation, a remarkable 66.79% increase in UFAs production in P. kessleri UCM 001 was achieved, compared to previous studies. This underscores the potential of this technology for enhancing valuable lipid production. Sequential variations in the application of physicochemical factors during microalgae culture under mixotrophic conditions, as optimized by ANN-GA, induced alterations in UFAs production and composition in P. kessleri UCM 001. This suggests the feasibility of tailoring the lipid profile of microalgae to obtain specific lipids for diverse industrial applications. The microalgae were isolated from a high-mountain lake in Colombia, highlighting their adaptation to extreme conditions. This underscores their potential for sustainable lipid and biomaterial production. This study demonstrates the effectiveness of using ANN-GA technology to optimise UFAs production in microalgae, offering a promising avenue for obtaining valuable lipids. The microalgae's unique origin in a high-mountain environment in Colombia emphasises the importance of exploring and harnessing microbial resources in distinctive geographical regions for biotechnological applications.

A ratiometric SERS strategy for the prediction of cancer cell proportion and guidance of glioma surgical resection

Rapid and accurate identification of tumor boundaries is critical for the cure of glioma, but it is difficult due to the invasive nature of glioma cells. This paper aimed to explore a rapid diagnostic strategy based on a label-free surface-enhanced Raman scattering (SERS) technique for the quantitative detection of glioma cell proportion intraoperatively. With silver nanoparticles as substrate, an in-depth SERS analysis was performed on simulated clinical samples containing normal brain tissue and different concentrations of patient-derived glioma cells. The results revealed two universal characteristic peaks of 655 and 717 cm−1, which strongly correlated with glioma cell proportion regardless of individual differences. Based on the intensity ratio of the two peaks, a ratiometric SERS strategy for the quantification of glioma cells was established by employing an artificial neuron network model and a polynomial regression model. Such a strategy accurately estimated the proportion of glioma cells in simulated clinical samples (R2 = 0.98) and frozen samples (R2 = 0.85). More importantly, it accurately facilitated the delineation of tumor margins in freshly obtained samples. Taken together, this SERS-based method ensured a rapid and more detailed identification of tumor margins during surgical resection, which could be beneficial for intraoperative decision-making and pathological evaluation.

Artificial neural network modelling of natural gas dehydration process

A multi-input–multi-output artificial neuron network (MIMO-ANN) model has been developed for process monitoring and improvement on a natural gas glycol dehydration process. The MIMO-ANN model was based on a steady-state process simulation model constructed in commercial software Aspen HYSYS. A set of training data was generated with the converged simulation model for the training of the MIMO-ANN model in Python. The process input of the model includes lean glycol recirculation rate and purity along with wet gas inlet pressure. On the other hand, the process output considered includes dehydrated gas water and aromatics content, dehydrated gas hydrate formation temperature and water dew point, stripping gas flow rate as well as reboiler duty. The overall mean squared error (MSE) of the MIMO-ANN model was calculated as 1.79. The best-fit line that highly overlaps with a 45° diagonal line is constructed with a correlation coefficient (R2) score of more than 0.999 for all studied process output testing datasets. The mean absolute percentage error (MAPE) of the predicted outputs is generally less than 1%, except for dehydrated gas water dew point (16.63%) and stripping gas flow rate (11.55%), due to error predictions attempted on pseudo-zero values. This successfully displays an exceptional predictive performance of the MIMO-ANN model developed in this work which can be further deployed as an online dashboard for real-time monitoring tool.

Machine learning approaches for simultaneous spectrophotometric determination of heavy metal ions in water samples

In this study, the simultaneous determination of Co, Cd, Ni, Cu, and Pb was carried out as a color complex with 4-(2-pyridylazo) resorcinol in an aqueous solution under the assesting of machine learning. A partial least-squares multivariate linear regression and artificial neuron network for the analysis of mixtures of metals were developed. MATLAB is a powerful software machine learning program that was used to support matrix calculations and displays. The benefit of MATLAB in the construction of the machine learning model allows the development of a rapid and highly effective analysis of multiple components in the mixtures without separation and enrichment. For individual determinations, the working ranges were discovered as the important information for choosing the initial concentration of each heavy metal in a mixture, r. The results of analysis of Ni2+, Pb2+, and Cd2+ by two methods Partial Least Squares - PLS and Artificial Neural Networks - ANN are sensitive and accurate for simultaneous determination of the concentration of these ions in the synthesis mixture with a high regression coefficient of 0.993, respectively, 0.997, 0.997 for Ni2+, Pb2+ and Cd2+. As for Cu2+ and Co2+, the accuracy is higher when using the ANN method.

Enhanced adsorption of Bismark Brown R dye by chitosan conjugated magnetic pectin loaded filter mud: A comprehensive study on modeling and mechanisms

Herein, a polymer-based bioadsorbent was prepared by cross-linking chitosan to filter mud and magnetic pectin (Ch-mPC@FM) for the removal of Bismark Brown R dye (BB-R) from wastewater. Morphological characterization analysis indicated that Ch-mPC@FM had a higher surface area and better pore structure than its components. The Artificial Neuron Network (ANN) and Adaptive Neuro-Fuzzy Inference System (ANFIS) were employed to evaluate the simulation and prediction of the adsorption process based on input variables like temperature, pH, dosage, initial BB-R dye concentration, and contact time. ANFIS and ANN demonstrated significant modeling and predictive accuracy, with R2 > 0.93 and R2 > 0.96, and root mean square error < 0.023 and <0.020, respectively. The Langmuir isotherm and the pseudo-second-order kinetic models provided the best fits to the equilibrium and kinetic data. The thermodynamic assessment showed spontaneous and endothermic adsorption with average entropy and enthalpy changes of 119.32 kJ mol−1 K and 403.47 kJ mol−1, respectively. The study of BB-R dye adsorption on Ch-mPC@FM revealed multiple mechanisms, including electrostatic, complexation, pore filling, cation-π interaction, hydrogen bonding, and π-π interactions. The approximate production cost of US$ 5.809 Kg−1 and excellent adsorption capability render Ch-mPC@FM an inexpensive, pragmatic, and ecologically safe bioadsorbent for BB-R dye removal from wastewater.

Application of machine learning models for green andsimultaneous determination of asiaticoside andmadecassoside in Centella asiatica

The artificial neuron network (ANN), which is a subfield of machine learning, has been widely applied in analytical chemistry for classification/pattern recognition, prediction, and modeling. ANN combined with UV spectrometry can be used to tackle the problem of overlapping spectra of a complicated matrix of herbal medicine. In the present study, ANN has been used for simultaneous determination of asiaticoside andmadecassoside in Centella asiatica collected from various provinces in Viet Nam based on UV spectra of standard reference and spiked samples. The absorption spectra of 108 C. asiatica samples were recorded in (190-250) nm of wavelength with an interval of 1 nm (61 variables) were used for data acquisition. An ANN model using traincgb function with 40 neural hidden layers was trained. The correlation coefficients were all higher than 0.9999. The concentration of asiaticoside and madecassoside in all tested samples revealed a good recovery, as compared to the referenced values. TheANN model can be considered as effective, time-saving and environment-friendly quantitative analysis tool for herbal medicine ® 2024 Journal of Science and Technology - NTTU

Generating random complex networks with network motifs using evolutionary algorithm-based null model

Network motifs in complex networks signify critical patterns of connections essential for deciphering system dynamics. Identifying and understanding these rare and elusive motifs is crucial for analyzing complex network behaviors. Our previous research has established a significant positive correlation between the occurrence of motifs and two network properties at the micro level, namely Assortativity Degree (ρ) and Local Clustering Coefficient (CCl) (Mursa, 2019b; Mursa, 2019c; Mursa, 2019a; Mursa, 2021). Making use of these findings, in this paper, we present a novel null model, MuAn, which leverages Evolutionary Algorithms to produce synthetic networks with topologies that are characterized by high values of ρ and CCl, facilitating the emergence of network motifs. A comprehensive experiment validates our proposed null model quality in generating networks with motifs through fine-tuning the EA configuration for precise and optimal outcomes and a comparative analysis with other well-known null models. The proposed model offers a promising avenue to advance motif analysis by enabling the generation of synthetic networks with motifs, expediting investigations on gene regulatory networks, transcriptional networks, drug-target networks, artificial neuronal networks, and other domains wherein motif-driven analysis is pivotal to unraveling their intrinsic complexities.

Field Telemetry Drilling Dataset Modeling with Multivariable Regression, Group Method Data Handling, Artificial Neural Network, and the Proposed Group-Method-Data-Handling-Featured Artificial Neural Network

This paper presents data-driven modeling and a results analysis. Group method data handling (GMDH), multivariable regression (MVR), artificial neuron network (ANN), and new proposed GMDH-featured ANN machine learning algorithms were implemented to model a field telemetry equivalent mud circulating density (ECD) dataset based on surface and subsurface drilling parameters. Unlike the standard GMDH-ANN model, the proposed GMDH-featured ANN utilizes a fully connected network. Based on the considered eighteen experimental modeling designs, all the GMDH regression results showed higher R-squared and minimum mean-square error values than the multivariable regression results. In addition, out of the considered eight experimental designs, the GMDH-ANN model predicts about 37.5% of the experiments correctly, while both algorithms have shown similar results for the remaining experiments. However, further testing with diverse datasets is necessary for better evaluation.

VCSEL-based photonic spiking neural networks for ultrafast detection and tracking

Inspired by efficient biological spike-based neural networks, we demonstrate for the first time the detection and tracking of target patterns in image and video inputs at high-speed rates with networks of multiple artificial spiking optical neurons. Using photonic systems of in-parallel spiking vertical cavity surface emitting lasers (VCSELs), we demonstrate the implementation of multiple convolutional kernel operators which, in combination with optical spike signalling, enable the detection and tracking of target features in images/video feeds at an ultrafast photonic operation speed of 1 ns per pixel. Alongside a single layer optical spiking neural network (SNN) demonstration, a multi-layer network of photonic (GHz-rate) spike-firing neurons is reported where the photonic system successfully tracks a large complex feature (Handwritten Digit 3). The consecutive photonic layers perform spike-enabled image reduction and convolution operations, and interact with a software-implemented SNN, that learns the feature patterns that best identify the target to provide a high detection efficiency even in the presence of a distractor feature. This work therefore highlights the effectiveness of combining neuromorphic photonic hardware and software SNNs, for efficient learning and ultrafast operation, thanks to the use of spiking light signals, towards tackling complex AI and computer vision problems.

Simulation of Rarefied Gas Flow in a Channel Applying Artificial Neuron Network

The purpose of the present paper is to confirm numerically applying our ANN algorithm for transitional gas flow between continuum and free-molecular flow. The main advantage of this approach is substantial computing time economy compared to usual DSMC procedure. The learning of ANN is based on well known rarefied gas flow examples, analytic solutions in limit cases, calculations by lattice Boltzmann method (LBM), and on quite few Monte Carlo simulations. Our previous results obtained for large Knudsen numbers have been applied near the free-molecular regime. In this region the flow has demonstrated the properties of unstable nonlinear dynamic system. The flow of nitrogen in a microchannel with aspect ratio L/h = 20 and by pressure ratio P between 1.84 and 4.14 is simulated by ANN. This example has shown a good coincidence of ANN and DSMC calculations even in the case when only a small part of original DSMC computation has been applied for learning ANN. Thus, computing time is reduced up to 10 times in this case.