Artificial Network Model Research Articles

Accelerated discovery of eutectic compositionally complex alloys by generative machine learning

Eutectic alloys have garnered significant attention due to their promising mechanical and physical properties, as well as their technological relevance. However, the discovery of eutectic compositionally complex alloys (ECCAs) (e.g. high entropy eutectic alloys) remains a formidable challenge in the vast and intricate compositional space, primarily due to the absence of readily available phase diagrams. To address this issue, we have developed an explainable machine learning (ML) framework that integrates conditional variational autoencoder (CVAE) and artificial neutral network (ANN) models, enabling direct generation of ECCAs. To overcome the prevalent problem of data imbalance encountered in data-driven ECCA design, we have incorporated thermodynamics-derived data descriptors and employed K-means clustering methods for effective data pre-processing. Leveraging our ML framework, we have successfully discovered dual- or even tri-phased ECCAs, spanning from quaternary to senary alloy systems, which have not been previously reported in the literature. These findings hold great promise and indicate that our ML framework can play a pivotal role in accelerating the discovery of technologically significant ECCAs.

Prediction of Mechanical Properties of Lattice Structures: An Application of Artificial Neural Networks Algorithms.

The yield strength and Young's modulus of lattice structures are essential mechanical parameters that influence the utilization of materials in the aerospace and medical fields. Currently, accurately determining the Young's modulus and yield strength of lattice structures often requires conduction of a large number of experiments for prediction and validation purposes. To save time and effort to accurately predict the material yield strength and Young's modulus, based on the existing experimental data, finite element analysis is employed to expand the dataset. An artificial neural network algorithm is then used to establish a relationship model between the topology of the lattice structure and Young's modulus (the yield strength), which is analyzed and verified. The Gibson-Ashby model analysis indicates that different lattice structures can be classified into two main deformation forms. To obtain an artificial neural network model that can accurately predict different lattice structures and be deployed in the prediction of BCC-FCC lattice structures, the artificial network model is further optimized and validated. Concurrently, the topology of disparate lattice structures gives rise to a certain discrete form of their dominant deformation, which consequently affects the neural network prediction. In conclusion, the prediction of Young's modulus and yield strength of lattice structures using artificial neural networks is a feasible approach that can contribute to the development of lattice structures in the aerospace and medical fields.

Breaking the mold of simulation-optimization: Direct forward machine learning methods for groundwater contaminant source identification

Groundwater Contaminant Source Identification (GCSI) is important for addressing environmental concerns. Currently, it is widely achieved using the Simulation/Optimization (S/O) method. However, the utilization of optimization techniques may cause high computation costs and parameter equifinality issues. We introduce two innovative GCSI methods, Direct Forward Machine Learning (DFML) and One-Hot Machine Learning (OHML), utilizing the classical Artificial Neuro Network (ANN) model in the machine learning field. Both new methods eliminate the need for an optimization algorithm in GCSI, thus reducing the construction effort and improving efficiency. The first method, DFML can directly estimate eight parameters, providing valuable insights into the contaminant location, release history, and aquifer properties. The second method, OHML can estimate the spatial probability distribution of the contaminant location through one-hot encoding, addressing uncertainties in the contaminant source location estimations realized by DFML.Evaluations demonstrate that both methods exhibit satisfying performances. DFML can estimate contaminant location and aquifer properties with high accuracy; and estimate the release history information with moderate accuracy. The OHML correctly assigns the higher contaminant probabilities to regions containing true contaminant locations. The combination of DFML and OHML offers a comprehensive framework. This study contributes valid methodologies to the GCSI field.

An Artificial Neural Network Model for a Comprehensive Assessment of the Production Performance of Multiple Fractured Unconventional Tight Gas Wells

The potential of unconventional hydrocarbon resources has been unlocked since the hydraulic fracturing technique in combination with long horizontal wells was applied to develop this type of reservoir economically. The design and optimization of the fracturing treatment and the stimulated reservoir volume and the forecasting of production performance are crucial for the development and management of such resources. However, the production performance of tight gas reservoirs is a complicated nonlinear problem, described by many parameters loaded with uncertainty. The complexity of the problem influences and inspires the sophistication of the solution to be used. This paper proposed an artificial network model that allows for fast, extended, and accurate analyses of the production performance of multiple fractured unconventional tight gas wells. In the comprehensive approach developed, the reservoir rock parameters, the drainage area, and the hydraulic fracture parameters are treated as a variable input to the model. The analysis is no longer constrained by fixed “shoes box” geometry, and the values of the parameters defining the reservoir and stimulated volume are not limited to a few discrete values. The numerical experiment used to construct a database for model development was designed using a genetically optimized Latin hypercube sampling technique. A special approach was used in the preparation of “blind data”, which are crucial for truly reliable model verification. In the result, a developed tool offers an extended rock-fluid description, flexible model, and stimulated reservoir volume dimensioning and parameterization, as well as a high degree of applicability in sensitivity analysis and/or optimization.

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.

Comparing discriminating abilities of evaluation metrics in link prediction

Link prediction aims to predict the potential existence of links between two unconnected nodes within a network based on the known topological characteristics. Evaluation metrics are used to assess the effectiveness of algorithms in link prediction. The discriminating ability of these evaluation metrics is vitally important for accurately evaluating link prediction algorithms. In this study, we propose an artificial network model, based on which one can adjust a single parameter to monotonically and continuously turn the prediction accuracy of the specifically designed link prediction algorithm. Building upon this foundation, we show a framework to depict the effectiveness of evaluating metrics by focusing on their discriminating ability. Specifically, a quantitative comparison in the abilities of correctly discerning varying prediction accuracies was conducted encompassing nine evaluation metrics: Precision, Recall, F1-Measure, Matthews correlation coefficient, balanced precision, the area under the receiver operating characteristic curve (AUC), the area under the precision-recall curve (AUPR), normalized discounted cumulative gain (NDCG), and the area under the magnified receiver operating characteristic. The results indicate that the discriminating abilities of the three metrics, AUC, AUPR, and NDCG, are significantly higher than those of other metrics.

Retraction Note: A comparative study on parameters of leaf-shaped patch antenna using hybrid artificial intelligence network models

Retraction Note: A comparative study on parameters of leaf-shaped patch antenna using hybrid artificial intelligence network models

Arahan dan Strategi Perlindungan Lahan Sawah Untuk Mendukung Ketahanan Pangan Di Kabupaten Maros, Sulawesi Selatan

Integration of Maros Regency into the KSN Mamminasata and the increasing population has increased the conversion of rice fields. This research aims to (a) Analyze conversion of rice fields in Maros Regency period 2013-2042 (b) Analyze status of food balance in Maros Regency period 2023-2042, and (c) Formulate directions and strategies for protecting rice fields to support food availability in Maros Regency. The method used is visual interpretation of spot 5 satelite in 2012 and 2022 to obtain data on rice fields. The trend of paddy field conversion was obtained using spatial analysis. Paddy field data for 2042 was obtained based on prediction results using the Artificial Neurol Network (ANN) model using Land Change Modeler (LCM) application. Status of food balance in Maros Regency is obtained from the area of paddy fields in 2022 and 2042. The protection of paddy fields is based on 3 scenarios, using the rational method. Formulation of directions and strategies using the A'WOT method. The results of this research showed that land conversion in Maros Regency for the 2013-2023 period was 1,946 ha or 7%, while the 2023-2042 period had increased to 3,722 ha or 15%. The status of land balance in Maros Regency is a deficit from 2022 to 2042. The area of paddy fields must be protected to provide local food availability at 4,940 ha. The strategy for protecting rice fields is a strategy to improve the quality of rice fields to increase productivity, and strengthening policies that are more consistent and consequential.

INTEGRATING ANN AND CS TO IMPROVE TOOL WEAR PREDICTION IN HIGH-SPEED DRY TURNING OF SKD11 STEEL

This article presents the results of improving an artificial neural network (ANN) to predict the tool wear in high-speed dry turning of SKD11 steel. The original ANN was a back-propagation (BPN) model with the Gradient Descent algorithm (GD). In the improved model, so-called ANN-CS, some parameters were optimized by the Cuckoo search algorithm (CS). Both models were trained, validated, and tested with the same experimental machining dataset based on performance indices, such as R2, MSE, RMSE, and MAPE. The results show that the ANN-CS gives higher prediction accuracy in comparison with the BPN. Especially, the improvement is as high as 30% with the MAPE index. This research result has important implications in choosing artificial intelligence network models suitable for the nature and amount of data that is both large and different. Moreover, this result can help researchers have more basis to choose the training model with high accuracy.

An artificial intelligent network model to monitor the condition of a patient with a breast tumor based on fuzzy logic

An artificial intelligent network model to monitor the condition of a patient with a breast tumor based on fuzzy logic

Flow Rate Dependency of Steady-State Two-Phase Flows in Pore Networks: Universal, Relative Permeability Scaling Function and System-Characteristic Invariants

The phenomenology of steady-state two-phase flow in porous media is conventionally recorded by the relative permeability diagrams in terms of saturation. Yet, theoretical, numerical and laboratory studies of flow in artificial pore network models and natural porous media have revealed a significant dependency on the flow rates—especially when the flow regime is capillary to capillary/viscous and part of the disconnected non-wetting phase remains mobile. These studies suggest that relative permeability models should incorporate the functional dependence on flow intensities. In the present work, a systematic dependence of the pressure gradient and the relative permeabilities on flow rate intensity is revealed. It is based on extensive simulations of steady-state, fully developed, two-phase flows within a typical 3D model pore network, implementing the DeProF mechanistic–stochastic model algorithm. Simulations were performed across flow conditions spanning 5 orders of magnitude, both in the capillary number, Ca, and the flow rate ratio, r, and for different favorable /unfavorable viscosity ratio fluid systems. The systematic, flow rate dependency of the relative permeabilities can be described analytically by a universal scaling function along the entire domain of the independent variables of the process, Ca and r. This universal scaling comprises a kernel function of the capillary number, Ca, that describes the asymmetric effects of capillarity across the entire flow regime—from capillarity-dominated to mixed capillarity/viscosity- to viscosity-dominated flows. It is shown that the kernel function, as well as the locus of the cross-over relative permeability values, are single-variable functions of the capillary number; they are both identified as viscosity ratio invariants of the system. Both invariants can be correlated with the structure of the pore network, through a function of Ca. Consequently, the correlation is associated with the wettability characteristics of the system. Among the potential applications, the proposed, universal, flow rate dependency scaling laws are the improvement of core analysis and dynamic rock-typing protocols, as well as integration into field-scale simulators or associated machine learning interventions for improved specificity/accuracy.

Research on rainfall prediction based on RBF neural network model and stormwater inundation risk in scenic areas: A case study of the Yesanpo Scenic Area, Baoding, China

Research on stormwater inundation risk and rainwater management in scenic areas has a lot to do with rainfall during the flood season. When the measured rainfall data is limited, an artificial network model with nonlinear mapping capability can be applied to predict rainfall data during the flood season, which increases the sample size of rainfall data and improves the accuracy of research results. Based on a radial basis function (RBF)neural network model, this paper takes the Yesanpo Scenic Area in Baoding City, Hebei Province as an example to estimate the monthly maximum rainfall data during the flood season (July–September) of 2022, 2023, and 2024 in the study area. On this basis, the Pearson III frequency curve is used to calculate the design rainfall corresponding to the rainfall frequency of 20%, 5%, and 2%. With the help of SCS-CN model and ArcGIS spatial analysis tools, the stormwater inundation areas are simulated in the study area, which are divided into three risk levels: high, medium, and low, providing a reference for the stormwater management in the Yesanpo Scenic Area.

Flow-Dependent Relative Permeability Scaling for Steady-State Two-Phase Flow in Porous Media: Laboratory Validation on a Microfluidic Network

Conventionally, the relative permeabilities of two immiscible fluid phases flowing in porous media are considered and expressed as functions of saturation. Yet, this has been put into challenge by theoretical, numerical, and laboratory studies of flow in artificial pore network models and real porous media. These works have revealed a significant dependency of the relative permeabilities on the flow rates, especially when the flow regime is capillary to capillary-viscous dominated, and part of the disconnected nonwetting phase remains mobile. These studies suggest that relative permeability models should include the functional dependence on flow intensities. However, revealing the explicit form of such dependence remains a persistent problem. Just recently, a general form of dependence was inferred based on extensive simulations with the DeProF model for steady-state two-phase flows in pore networks. The simulations revealed a systematic dependence of the relative permeabilities on the local flow rate intensities. This dependence can be described analytically by a universal scaling functional form of the actual independent variables of the process, namely, the capillary number, Ca, and the flow rate ratio, r. The proposed scaling incorporated a kernel function, the intrinsic dynamic capillary pressure (IDCP) function, describing the transition between capillarity- and viscosity-dominated flow phenomena. In a parallel laboratory study, SCAL measurements provided a preliminary proof-of-concept on the applicability of the model. In the laboratory study presented here, we examine the applicability of the scaling model by taking extensive, ex-core measurements of relative permeabilities for steady-state co-injections of two immiscible fluids within an artificial microfluidic pore network, across different flow regimes in Ca and r. From these measurements, we calculate the values of the mobility ratio, and we compare these to the corresponding values of the flow rate ratio. We also extract the IDCP curve, the locus of critical flow conditions, whereby the process is more efficient in terms of energy utilization – accounted by the nonwetting phase flow rate per unit of total power provided to the process, as well as the locus of flow conditions of equal relative permeabilities. We show that the degree of consistency between flow rate ratio and mobility ratio values, the IDCP curve, the locus of critical flow conditions, and the locus of equal relative permeabilities, as well as some associated invariant characteristic values, can be used for assessing the extent of end effects and for characterizing the flow as capillary- or viscous-dominated. The proposed scaling introduces new opportunities for enhancing SCAL protocols and their associated applications. These include the characterization of systems and flow conditions, dynamic rock typing, evaluation of capillary end effects, as well as the advancement of more efficient field-scale simulators. Additionally, it paves the way in designing more energy-efficient EOR interventions.

Prediction of deck grillages lifting deformation using a data-driven Def-GAN network

The thin-walled deck grillages are prone to deformation in lifting process of ship blocks construction due to huge rope tension loads. Optimizing hoisting scheme with the aid of CAE analysis techniques before implementing is conductive to controlling structure deformation. A deformation prediction method based on generative adversarial network (Def-GAN) is developed in this paper to perform quick-response and precise deformation predictions under a given hoisting scheme of deck grillages. This artificial intelligence network model has an adversarial architecture and comprises a pair of submodels, namely generator and discriminator. The datasets including lifting conditions and deformation distributions of deck grillages are set up to train and test the Def-GAN model. The deformation distributions in datasets are obtained from a series of FEM analyses. An image coding and mapping algorithm is employed to deal with input data of lifting conditions and output data of deformation distributions respectively. The results show that the predictive model achieved an R-squared accuracy of 0.997 and a mean squared error of 0.005 compared to numerical results for the test dataset. In addition to predicting deformation of grillages lifting, the Def-GAN model is a promising candidate tool for hoisting scheme optimization.

Ontologies for Supporting Traffic Behaviour, Critical Gap, and Conflict Models at Unsignalized Intersection Routes

Background: The challenge arises from management of big data in transportation can be identified by ontological approach. Road accidents are regarded as a significant cause of both harm to humans and financial damage in many countries. In Malaysia, there were 567,516 accidents in 2020 alone. This translates into a daily average of 13 people killed due to traffic accidents in the country. Here, our work focuses on gap patterns, critical gap analysis, gap acceptance of right-turning motorists (RTMs) and serious conflict model on Malaysia Rural Roadways at three-leg unsignalized intersections. Methods: In early stage, traffic volume, motorist turning manoeuvre and speed study are implemented to identify the traffic behaviour at selected intersection. Three unsignalized intersection (UI) was involved namely (UI2), (UI 9), and (UI10). In the development of logistic regression models, five different datasets were used in this study: right-turning motorist (RTM) at unsignalized intersection UI2 (259 dataset), right-turning motorist at UI 9 (239 dataset), right turning motorist at UI 10 (314 dataset), right turning motorist combined model (812 dataset) and serious conflict lane change (351 dataset). Determination of critical gap was carried out at each unsignalized intersection 2, 9 and 10. Meanwhile gap-pattern analysis at each intersection used visualization spatial plot. In addition, this work investigated logistic regression method, artificial neuron network and structural equation modelling. Results: Gap pattern three was discovered to be a vulnerable gap pattern. Furthermore, this research reveals that the attributes of the gap three pattern, motorcycles rider, speed limit exceeding 50 kph and RTMs where motorcycles stop near passenger cars in minor roads encourage serious conflict. Moreover, this study proposes novelty tool for identifying hazardous unsignalized intersection by using visualization spatial plot between approach speed and gap.

Nowcasting of fecal coliform presence using an artificial neural network

At least 2 billion people worldwide use drinking water sources that are contaminated with feces, causing waterborne diseases; poor sanitation, poor hygiene, and unsafe drinking water result in a daily death rate of more than 800 children under 5 years of age from diarrheal diseases. This study shows the feasibility of a novel method to nowcast fecal coliforms' (FC) presence in drinking water sources by applying a multilayer perceptron artificial neuron network (MLP-ANN) model. The model gives a binary answer for FC presence or absence in drinking water sources using a minimum of water quality and geographical parameters, which can be monitored in real-time as predictors with low-cost and in-situ equipment. Using 51,400 samples to train, validate and test the model with temperature, pH, electrical conductivity, turbidity, dissolved oxygen, and total dissolved solids (TDS) as water-quality inputs and the water source type and location (as districts in India) as geographical inputs. The model achieved a total accuracy of 92.8% and a sensitivity of 98.2%, meaning that most FC-contaminated samples were classified correctly. In addition, precision reached 93.1%, meaning that most FC-contamination classifications were actually contaminated. The MLP-ANN performed better than the Linear Regression and K-Nearest Neighbors models, with lower accuracies of 90.2% and 91.0%, respectively. The MLP-ANN model could characterize the water quality geospatially, learn from the parameters whether the water is contaminated by FC, and predict with high accuracy on new testing data. This method can be used as a part of a sensor for FC monitoring and management in water, reducing the time gaps between routine lab testing and thus improving drinking water quality and addressing the SDG 6 targets.

Assessing the accuracy of sensitivity analysis: an application for a cellular automata model of Bogota’s urban wetland changes

This study analyzes the outcomes of Cellular Automata (CA) with different neighborhood sizes and spatial resolution configurations on the performance of the Future Land Use Simulation (FLUS) model. The analysis is executed using three analogic images to extract the land use/land cover in Bogota, Colombia, for three years: 1998, 2004, and 2010. The FLUS model has an Artificial Neuronal Network model, which was used for calculating the relationships between the land uses and the associated drivers and to estimate the probability of occurrence of each land use. Whenever a CA is used to model and simulate, sensitivity analysis (SA) becomes a crucial step in CA modeling to understand better the influence of parameters’ changes in the simulation outcomes. Therefore, the SA is conducted by varying the neighborhood sizes between 3 × 3, 5 × 5, and 7 × 7 for 5 and 30 meters. In addition, cross-classification maps, Area Under the Curve (AUC) of the Total Operating Characteristic, landscape metrics, the figure of merit, Fuzzy Kappa, and disagreement metrics were calculated to assess how well the model performed. High AUC values and low disagreement results show that, in general, the model performed well, and the accuracy of the outputs improves with a 3 × 3 neighborhood size and 5 meters spatial resolution. This study provides a broad assessment approach to the different methods that must be considered to evaluate the sensitivity of CA models in the simulation of urban wetlands’ spatial-temporal evolution.

Artificial neural network models for prediction of standardized precipitation index in central Mexico

Some of the effects of climate change may be related to a change in patterns of rainfall intensity or rainfall scarcity. So, humanity is facing environmental challenges due to an increase in the occurrence of droughts. Forecasting of droughts based on cumulative influence of rainfall could be greatly beneficial for mitigating adverse consequences on water-sensitive sectors such as agriculture. Then, predictive models of drought indices could help in assessing water scarcity situations, droughts identification and their severity characterization. In this paper, we tested the feasibility of the Artificial Neural Network as a data-driven model for predicting the monthly Standardized Precipitation Index in 4 regions (Semi-desert, Highlands, Canyons and Mountains) of north-central México using predictive variable data from 1965 to 2004 as training data and simulated data for the period 2005-2014. The best model was found using the Hyperbolic Tangent as activation function and the Adaptive Moment Estimation (Adam) algorithm as optimization method. The best model was set to the following architecture: 26-12-1 network with 4 weights and 365 trainable parameters. Based on analysis of scatter plot between predicted and observed Standardized precipitation Index values for the test dataset, the Coefficient of Determination was between 0.84 and 0.88. In terms of quantitative statistics averaged over the test set, Artificial Network Model performed very well in predicting Standardized Precipitation Index at the four analyzed regions. This was verified by all-region average value of performance statistics Mean Absolute Error (0.0805), Mean Square Error (0.0144) and the Coefficient of Determination (0.8671). In a nutshell we summarize that the Artificial Network models developed and tested in this study had good prediction skills of the monthly Standardized Precipitation Index for stations and its drought-related properties in the study region.

Comparative analysis of fouling resistance prediction in shell and tube heat exchangers using advanced machine learning techniques

Heat exchangers are utilized in a vast region of the process industry for heating and cooling. Long-term operation of heat exchangers results in decreased efficiency due to many problems, such as fouling. Therefore, the object of this research paper is to use three artificial intelligence techniques (feedforward neural networks-multilayer perceptron (FNN-MLP), nonlinear autoregressive networks with exogenous inputs (NARX), and support vector machines (SVM-RBF)) for predicting the fouling resistance in the tube and the shell heat exchanger in the preheating circuit of atmospheric distillation. The results summarize the high training as well as the predictive capacity of the "FFNN-MLP" model for predicting the fouling resistance in the heat exchanger with the highest coefficient of correlation (R = 0.99961) and the lowest root-mean-squared error (nRMSE = 1.0031%) for the testing phase, where the FNN-MLP network is superior to that provided using the SVM model (R = 0.9955 and nRMSE = 3.8652%). All the models of artificial networks and machine learning techniques used in the current work can be used to predict the fouling resistance in heat exchanger data with high accuracy. Despite this, the FNN-MLP model is the preferred model compared with the other proposed models, followed by the NARX model.