Search

• Two-phase flow regime identification using ML with enhanced interpretability. • Evaluation of various dimensionality reduction techniques to optimize feature representation. • Kernel Fisher discriminant analysis is the best performing dimensionality reduction technique. • AutoML can impartially select the most effective ML classifier. • Our workflow can develop a virtual flow regime map to automate flow regime recognition. Accurate identification of flow regimes is paramount in several industries, especially in chemical and hydrocarbon sectors. This paper describes a comprehensive data-driven workflow for flow regime identification. The workflow encompasses: i) the collection of dynamic pressure signals using an experimentally verified numerical two-phase flow model for three different flow regimes: stratified, slug and annular flow, ii) feature extraction from pressure signals using Discrete Wavelet Transformation (DWT), iii) Evaluation and testing of 12 different Dimensionality Reduction (DR) techniques, iv) the application of an AutoML framework for automated Machine Learning classifier selection among K-Nearest Neighbors, Artificial Neural Networks, Support Vector Machines, Gradient Boosting, Random Forest, and Logistic Regression, with hyper-parameter tuning. Kernel Fisher Discriminant Analysis (KFDA) is the best DR technique, exhibiting superior goodness of clustering, while KNN proved to be the top classifier with an accuracy of 92.5 % and excellent repeatability. The combination of DWT, KFDA and KNN was used to produce a virtual flow regime map. The proposed workflow represents a significant step forward in automating flow regime identification and enhancing the interpretability of ML classifiers, allowing its application to opaque pipes fitted with pressure sensors for achieving flow assurance and automatic monitoring of two-phase flow in various process industries. [ABSTRACT FROM AUTHOR]

In this paper, we construct two deep neural network (DNN) approaches based on the hybridized discontinuous Galerkin (HDG) method for solving some elliptic problems. The main idea is to use an HDG scheme for spatial discretization and then estimate solutions using the deep learning idea. In fact, by employing DNN, we can achieve robust methods compared to classical methods for solving noisy and high-dimensional problems. A brief analysis shows that the loss functions corresponding to the proposed methods, which are called DNN-HDG-I and DNN-HDG-II, converge to zero as the mesh step size reduces. By testing several examples in one, two, and three dimensions, we demonstrate the performance of the proposed methods that show their apparent advantages especially compared to classical HDG methods. As we expected, the DNN-HDG methods can efficiently and accurately extract the pattern of the solutions in all three dimensions. To show the superiority of the proposed schemes, the DNN-HDG methods are compared with the classical HDG method for solving a problem involving some noisy data. Specifically, we employ the proposed methods for solving a problem which its exact solution is not accessible and compare it with the HDG solution. In this paper, the uniformly random data and Jacobi polynomial roots are considered, respectively, as the training data on the boundary and quadrature points for approximating integrals. Also, we find that the proposed methods are not required a high number of neurons (at most 30) and hidden layers (at most 10) for satisfactory results with acceptable accuracy. [ABSTRACT FROM AUTHOR]

In response to nonrenewable energy consumption and environmental pollution, the development of electric vehicles has accelerated. The safety of electric vehicles has garnered considerable attention. There are numerous incalculable foreign objects on the road that may collide with or scratch the battery-pack's frontal, resulting in battery-pack system damage or even explosion. This poses a significant risk to the safety of passengers and drivers. In this paper, a diversity of mechanical safety prediction models for battery-pack systems are proposed. These models support data-driven structural optimization of the battery-pack system by utilizing the numerical results of the bottom shell deformation. These simulation-based prediction models combine response surface method and machine learning algorithms. First, a nonlinear finite element model of the battery-pack system is established. The efficacy of the model is verified using constrained modal analysis in a variety of commercial software packages. Second, the collision simulations are executed and the data in various collision conditions are collected. Different sample sizes are used to develop various response surface models and machine learning models. The machine learning algorithms adopt support vector machine, Gaussian process regression, and neural network models. Third, the prediction accuracy of multiple prediction models is investigated according to error functions. The results show that the neural network model can predict the most accurate deformation under the condition of low speed frontal impact. The prediction average absolute percentage error of the neural network model is only 0.34% within the design domain, and only 2.54% outside the design domain. The proposed prediction model can be used for reliable design of the battery-pack system in electric vehicles. It also can be employed to design the early warning system of the battery-packs. [ABSTRACT FROM AUTHOR]

Deep neural networks (DNNs) for numerical solutions to partial differential equations (PDEs) have exhibited their remarkable merits of meshless methods, dimensionless features, and nonlinear approximation powers. The majority of conventional DNNs are mainly concentrated on the low- or high-dimensional PDE with smooth solutions. In singular problems the DNN does not behave so efficiently as in the smooth problem because of low regularities of the solution. In this paper, we propose a novel adaptive DNN for the PDE with corner singular solutions. In the corner singular problem gradients of solution vary rapidly around some particular vertices or lines. This poses a difficulty in development of DNN solvers. We design adaptive DNN techniques for the corner singularity, which incorporates (a) an adaptive loss function that assigns radial weights to local losses, (b) an adaptive activation function that varies the choice of activation functions from neuron to neuron, and (c) an adaptive sampling strategy that selectively focuses on sampling points near the singular locations. The adaptive DNN with these adaptive schemes greatly enhances approximation accuracy and computation efficiency. This is verified by a comprehensive series of numerical experiments of 2D and 3D corner singular problems. The comparison analyzes with the existing DNN solvers are also made to demonstrate effectiveness of the new method. The L 2 relative errors of the proposed adaptive DNN are reduced by an order of magnitude with respect to several conventional DNN methods. [ABSTRACT FROM AUTHOR]

This paper presents a novel interface technique employing Artificial Neural Networks (ANN) for efficient data transfer between incompressible fluid and deformable solid domains in a partitioned fluid–structure interaction (FSI) framework. Governing differential equations (GDE) with potential flow assumptions are solved to obtain pressure values at computational nodes in the fluid domain. Pressure values are then used to train an ANN-based model to interpolate pressure loads for application at non-collocated nodes on the solid boundary. Both finite element method (FEM) and meshfree element free Galerkin (EFG) formulations are used to calculate the solid deformation. Proposed methodology is applied to solve a two-dimensional (2D) unidirectional flow problem, consisting of a flexible cantilever beam immersed in a laminar, steady, and inviscid fluid. Overwhelming mathematical intricacies of computational solvers are avoided for the sake of simplistic interface treatment, flux transfer and associated phenomena. Dedicated partitioned solvers are developed for both solid and fluid domains, which are individually benchmarked for established problems from the literature. The presented ANN-based interpolation scheme provides an alternative to higher-order polynomial algorithms at the interface boundary. The proposed scheme is simple to employ, computationally efficient, and offers competitive accuracy as well as stability. • A novel ANN-based interface coupling approach for partitioned FSI problems. • Separate solvers, employing FDM for fluid domain, and FEM/EFG for solid domain. • Partitioned solvers are coupled at interface using ANN-based approximation function. • Steady unidirectional (one-way) FSI to couple FDM-ANN-FEM and FDM-ANN-EFG solvers. • ANN interpolate pressure from fluid to solid, better than polynomial-based fitting. • A generic ANN framework, extendable to problems with large deformations and velocities. [Display omitted] [ABSTRACT FROM AUTHOR]

• Second law aspects of hybrid nanofluid in a MCHS whit porous fins are examined. • The water based silver/graphene hybrid nanofluid is used as coolant. • Frictional entropy generation augments by boosting the nanoparticle concentration. • By rising the flow rate, thermal entropy generation declines. • Artificial neural network is employed to obtain a model for the entropy generation. This paper evaluates the characteristic of second law of thermodynamic, including Bejan number and entropy generation for hybrid nanofluid containing graphene-silver nanofluid through a MCHS whit porous fins. Finite-volume technique is utilized to solve the governing equations. To simulate the problem, different porous medium thicknesses, nanoparticle concentrations, and inlet mass flow rates are used while the heat flux remains constant. The minimum values of the frictional and thermal entropy generation are 5 × 10−4 and 6.25 × 10−2, while the maximum values are 3.2 × 10−4 and 9.75 × 10−2. With increasing nanoparticle concentration up to 0.06% wt at constant porous thickness t p =200 µm, frictional entropy generation rises up by 3 × 10−5 and heat transfer rate go up while, thermal entropy generation decreases by 1.5 × 10−2. In addition, by doubling the input mass flow rate and reaching 0.02% at constant nanoparticle concentration (0.06%), thermal entropy generation decreases by 2 × 10−2 while the frictional entropy generation increases by 2.4 × 10−4. The minimum magnitude of Bejan number is 0.994. This show that the irreversibility is derived significantly from thermal entropy generation rate. Finally, an artificial neural network is employed to obtain a model for entropy generation. [ABSTRACT FROM AUTHOR]

Purpose of this paper is the presentation of a novel Machine Learning (ML) technique for nanoscopic study of thin nanoplates. The second-order strain gradient theory is used to derive the governing equations and account for size effects. The ML framework is based on Physics-Informed Neural Networks (PINNs), a new concept of Artificial Neural Networks (ANNs) enriched with the mathematical model of the problem. Training of PINNs is performed using a highly efficient learning algorithm, known as Extreme Learning Machine (ELM). Two applications of this ANNs-based method are illustrated: solution of the Partial Differential Equations (PDEs) modeling the flexural response of thin nanoplates (direct problem), and identification of the length scale parameter of the nanoplate mathematical model with the aid of measurement data (inverse problem). Comparison with analytical and Finite Element (FE) solutions demonstrate the accuracy and efficiency of this ML framework as meshfree solver of high-order PDEs. The stability and reliability of the present method are verified through parameter studies on hyperparameters, network architectures, data noise and training initializations. The results presented give evidence of the effectiveness and robustness of this new ML approach for solving both direct and inverse nanoplate problems. [ABSTRACT FROM AUTHOR]

In this paper, a multilayer perceptron (MLP) artificial neural network (ANN) with a back-propagation (BP) training algorithm is applied for modeling thermophysical properties and subcooled flow boiling performance of Al 2 O 3 /water nanofluid in a horizontal tube. The influence of nanofluid concentration, heat flux, and flow rate on different thermophysical parameters, including thermal conductivity, thermal conductivity enhancement, viscosity, viscosity enhancement, and heat transfer coefficient, are investigated. Specifically, flow boiling of Al 2 O 3 /water nanofluid in a horizontal tube is modeled with the MLP neural network optimized by three novel swarm-based optimization algorithms: namely, Equilibrium Optimizer (EO), Marine Predators Algorithm (MPA), and Slime Mould Algorithm (SMA). To evaluate the effectiveness of different models, the MSE (Mean-Square Error) of the ANN model with varying optimization algorithms is calculated and compared. Additionally, the optimal network and regression values for each parameter are determined. The results show that the applied neural network and optimization algorithms could model the thermal conductivity, thermal conductivity enhancement, and viscosity better than the viscosity enhancement and heat transfer coefficient. The MSE of the best network for the thermal conductivity is 2.693 × 10−7, while the MSE of the best network for the viscosity enhancement is 0.0598. Also, the EO algorithm achieves the best optimization for the first three outputs, thermal conductivity, thermal conductivity enhancement, and viscosity. In comparison, the MPA algorithm extracts the optimal network for the other two outputs, viscosity enhancement, and heat transfer coefficient. [ABSTRACT FROM AUTHOR]

This paper presents the numerical simulations and machine learning-based prediction of the transient melting process of phase change material (PCM) in latent heat thermal storage (LHTS) units. The storage units are rectangular enclosures equipped with fins of different heights and numbers. For all enclosures, the volume of fins and PCM are kept constant. Melting processes of PCM in different storage units are simulated using computational fluid dynamics (CFD) to determine the impacts of fin parameters on the thermal behavior of the LHTS unit. Transient variation of liquid fraction and stored energy in the different storage units are obtained. Then, the group method of data handling (GMDH) type of artificial neural networks (ANNs) is employed and trained through numerical findings to develop correlations for predicting the instantaneous liquid fractions and stored energy in the finned enclosures. To evaluate the effectiveness of the prediction model, mean square, root mean square, and standard deviation errors as well as correlation coefficient have been calculated and proved the accuracy of the proposed correlations. [ABSTRACT FROM AUTHOR]

The study of the impact of random parameters on the load-carrying capacity of imperfect cylindrical stiffened shells remains limited, due to the expensive cost of experimental testing. In this study, a post-buckling analysis model to numerically determine the collapsed load is first introduced. However, it is challenging to analyze the probability characteristics of a shell considering multiple random parameters. Thus, in this paper, the direct probability integral method (DPIM), as a novel stochastic analysis method, is extended to address this issue. Given the lack of enough quantity of statistics about uncertainty factors, a back-propagation neural network improved by particle swarm optimization (BPNN-PSO) model for predicting the collapse load is established. Building upon this, a novel neural network-based DPIM is proposed. In the numerical example, we compare the calculated results with those using the analytical solution, path integral method, and Monte Carlo simulation, demonstrating the high accuracy and efficiency of DPIM. Finally, an efficient physical-based uncertainty quantification of imperfect stiffened shells is implemented by BPNN-PSO-based DPIM. The results reveal the effects caused by different random parameters on the load-carrying of imperfect cylindrical stiffened shells. In particular, the change in stiffened height will bring a huge reduction in structural reliability. [ABSTRACT FROM AUTHOR]

In this paper, the effect of using an elliptical chamber around a battery pack with three lithium-ion battery cells is studied numerically. The cylindrical batteries are located in the middle of the oval chamber, and the elliptical chamber, which is full of Phase-Change Materials (PCM), is inside an air duct. Numerical solution was performed using finite element method and the obtained values were performed using network training and artificial neural network. The possible error of this network is less than 3, which indicates the high accuracy of the network. Another artificial grid with one input layer, 32 hidden layers, and one output layer is the best structure to predict the amount of energy consumed in the battery pack. The results of this study showed that in a smaller PCM compartment, the freezing time of all PCMs around batteries is much shorter than in a larger PCM compartment. The temperature of the left battery is lower than the middle battery, especially at times longer than 3,000 s, and both are lower than the right battery in the battery pack. The temperature of the battery cell on the right, in particular, is influenced by the size of the PCM compartment. [ABSTRACT FROM AUTHOR]

• A hybrid RBF-MLP model was tested to achieve a highly nonlinear approximation. • A POD technique was employed to define a low-dimensional reduced-order system. • A POD-ANN model was developed to accurately reconstruct the principal flow features. • The combination of POD and machine learning can reveal the underlying physical mechanisms of flow features. The development of mathematical models, and the associated numerical simulations, are challenging in higher-dimensional systems featuring flows of supercritical fluids in various applications. In this paper, a data-driven methodology is presented to achieve system order reduction, and to identify important physical information within the principal flow features. Firstly, a new hybrid neural network based on radial basis function (RBF) and multi-layer perceptron (MLP) methods, namely RBF-MLP, is tested to achieve a highly nonlinear approximation. When provided with 1000 nonlinear test samples, this model provides an excellent prediction accuracy with a maximum regression coefficient (R) of 0.99 and a minimum root mean square error (RMSE) below 1%. Furthermore, the model is also proven to be flexible enough to capture accurately the turbulent fluctuation characteristics, even at significant nonlinear buoyancy conditions. Secondly, the high-dimensional buoyancy data is collected and integrated into a matrix database. Subsequently, a proper orthogonal decomposition (POD) approach is employed to reduce the high-dimensional database, and to obtain a set of low-dimensional POD basis-spanned space, which defines a reduced-order system with low-dimensional basis vectors. The results reveal that the first five order modes contain dominant flow features, accounting for 93% of the total mode energy, which can be selected to reconstruct the physical flow field. Thirdly, a new data-driven POD-ANN model is established to construct the nonlinear mapping between the full-field buoyancy data and decomposed basis vectors. It is also demonstrated that the POD-ANN model reconstructs the principal flow features accurately and reliably. This POD-ANN model can be used to provide new insights for reduced-order modelling and for reconstructing physical fields of higher-dimensional nonlinear flow cases. [ABSTRACT FROM AUTHOR]

Abstract: The idea of electrical impedance tomography (EIT) is to evaluate conductivity or permittivity distribution inside the examined object by measuring the voltages between electrodes placed on its surface. In this paper, EIT as a default 3D diagnostic method of the breast cancer is suggested. The breast was modelled as a hemisphere consists of two spatially homogenous areas with different conductivity. In order to determine the distribution of potential in the breast model, a multi-region boundary element method (BEM) was implemented. In this paper, a multi-region BEM with quadratic interpolation function for the flat, triangular surface elements was introduced. The inverse problem solution provided the identification of the size and the position of the anomalies in the breast tissue. For this purpose the efficient method based on principal component analysis (PCA) and the artificial neural network (ANN) was used. PCA applied to EIT data allows reducing dimensionality of measured data for 3D space and removing the unused part of information, usually corresponding to noise and interrelated variables. ANN method allows to obtain the results of inverse problem solution in real-time. [Copyright &y& Elsevier]

The computer simulation is replacing mechanical experiments in many cases due to its cost-effectiveness and improved accuracy. Nevertheless, its application fields are still limited to elastic analysis, as there exists a significant amount of model error in present inelastic material models. In this paper, we first propose a material model using neural networks, which has the ability to describe plasticity and cyclic plasticity. The proposed model was first created with the material data of 2 1/4Cr-1 Mo steel, and the results show that the model can represent the actual cyclic plastic material behavior within a 2% error. The model is then implemented in a commercially available finite element analysis package and the cyclic plastic deformation behavior of a test specimen of the material is predicted within a 3% error. [Copyright &y& Elsevier]