Numerical Simulation Data Research Articles

Theory-guided deep neural network for boiler 3-D NOx concentration distribution prediction

Timely and accurate three-dimensional (3-D) NOx concentration distribution prediction is essential for achieving low-emission and efficient operation in power plants. This study proposed a theory-guided data-driven prediction method for the 3-D NOx concentration distribution prediction. Firstly, the method created a foundational dataset by fusing numerical simulation data from the computational fluid dynamics (CFD) with operational data from the distributed control system (DCS). Then, the data was classified into three load condition categories, and the center operating conditions for each category were computed separately. Subsequently, the K-means algorithm was employed to extract representative data to address the computational challenges associated with big data. Finally, a Theory-Guided Deep Neural Network model (TG-DNN) was established leveraging the principle of carbon element mass conservation and deep neural network. Experimental results demonstrate that the method effectively monitors the 3-D NOx concentration distribution, potentially facilitating efficient production processes.

Image Domain Multi-Material Decomposition Noise Suppression Through Basis Transformation and Selective Filtering.

Spectral CT can provide material characterization ability to offer more precise material information for diagnosis purposes. However, the material decomposition process generally leads to amplification of noise which significantly limits the utility of the material basis images. To mitigate such problem, an image domain noise suppression method was proposed in this work. The method performs basis transformation of the material basis images based on a singular value decomposition. The noise variances of the original spectral CT images were incorporated in the matrix to be decomposed to ensure that the transformed basis images are statistically uncorrelated. Due to the difference in noise amplitudes in the transformed basis images, a selective filtering method was proposed with the low-noise transformed basis image as guidance. The method was evaluated using both numerical simulation and real clinical dual-energy CT data. Results demonstrated that compared with existing methods, the proposed method performs better in preserving the spatial resolution and the soft tissue contrast while suppressing the image noise. The proposed method is also computationally efficient and can realize real-time noise suppression for clinical spectral CT images.

Turbulence modeling of stratified turbulence using a constrained artificial neural network

For large eddy simulations (LES) of stratified turbulence in the strongly stratified regime, an artificial neural network (ANN) with five hidden layers is used to construct a sub-grid scale (SGS) model. The ANN is assessed by comparing it to the Smagorinsky model, the dynamic Smagorinsky model, the gradient model, and filtered direct numerical simulation data. In the a priori test, the SGS model using ANN performed better than the Smagorinsky model and the gradient model in terms of the correlation coefficient and relative error of the energy transfer rate. However, the ANN does not provide sufficient energy dissipation when it is applied to LES with a larger filter width because it overpredicts backscatter. To address this problem, we also trained a constrained ANN using a custom loss function that penalizes excessive backscatter. It is shown that the constrained ANN successfully predicts less backscatter, maintaining the high correlation coefficient without ad hoc clipping. These results show that ANN is a promising tool for realizing a highly accurate and stable SGS model for stratified turbulence.

A modification of production term for the shear stress transport turbulence model based on a new wall-law for adverse pressure gradient

This paper presents an attempt to improve the predictive accuracy of the Menter shear stress transport (SST) turbulence model for adverse pressure gradient (APG) flows. The foundation of this work lies in the newly modified law of the wall and turbulent kinetic energy (TKE) transport equation within the APG turbulent boundary layer. The modification of the law of the wall is achieved through a newly proposed variable λ, which is derived theoretically and calibrated using direct numerical simulation data. Based on the new law of the wall, we analyze the TKE transport equation of the SST model at APG and demonstrate that the analytical solution satisfying the law of the wall cannot achieve equilibrium in the current SST model. Further theoretical analysis reveals that this imbalance is caused by the TKE production term Pk, and the decomposition of the wall friction coefficient Cf indicates that Pk directly affects the prediction of flow separation. Based on the analysis and the new law of the wall, an additional term for Pk is suggested to correct Pk and improve the model's predictive accuracy for APG flows. Numerical validation results show that the correction leads to an increase in Pk within the APG turbulent boundary layer, resulting in more accurate predictions of Cf and the mean velocity profile in the inner layer of the turbulent boundary layer.

Experimental and Numerical Prediction of Slamming Impact Loads Considering Fluid–Structure Interactions

Slamming impacts on water are common occurrences, and the whipping induced by slamming can significantly increase the structural load. This paper carries out an experimental study of the water entry of rigid wedges with various deadrise angles. The drop height and deadrise angle are parametrically varied to investigate the effect of the entry velocity and wedge shape on the impact dynamics. A two-way coupled approach combing CFD method software STAR-CCM+12.02.011-R8 and the FEM method software Abaqus 6.14 is presented to analyze the effect of structural flexibility on the slamming phenomenon for a wedge and a ship model. The numerical method is validated through the comparison between the numerical simulation and experimental data. The slamming pressure, free surface elevation, and dynamic structural response, including stress and strain, in particular, are presented and discussed. The results show that the smaller the inclined angle at the bottom of the wedge-shaped body, the faster the entry speed into the water, resulting in greater impact pressure and greater structural deformation. Meanwhile, studies have shown that the bottom of the bow is an area of concern for wave impact problems, providing a basis for the assessment of ship safety design.

Hygrothermal Properties and Performance of Bio-Based Insulation Materials Locally Sourced in Sweden.

In recent years, there has been a paradigm shift in the building sector towards more sustainable, resource efficient, and renewable materials. Bio-based insulation derived from renewable resources, such as plant or animal fibres, is one promising group of such materials. Compared to mineral wool and polystyrene-based insulation materials, these bio-based insulation materials generally have a slightly higher thermal conductivity, and they are significantly more hygroscopic, two factors that need to be considered when using these bio-based insulation materials. This study assesses the hygrothermal properties of three bio-based insulation materials: eelgrass, grass, and wood fibre. All three have the potential to be locally sourced in Sweden. Mineral wool (stone wool) was used as a reference material. Hygrothermal material properties were measured with dynamic vapour sorption (DVS), transient plane source (TPS), and sorption calorimetry. Moisture buffering of the insulation materials was assessed, and their thermal insulation capacity was tested on a building component level in a hot box that exposed the materials to a steady-state climate, simulating in-use conditions in, e.g., an external wall. The tested bio-based insulation materials have significantly different sorption properties to stone wool and have higher thermal conductivity than what the manufacturers declared. The hot-box experiments showed that the insulating capacity of the bio-based insulators cannot be reliably calculated from the measured thermal conductivity alone. The results of this study could be used as input data for numerical simulations and analyses of the thermal and hygroscopic behaviour of these bio-based insulation materials.

Leakage current as a probe into the mechanics of carrier transport in insulating composite polymers

Amorphous composite polymers are widely used as insulators in microelectronics due to their high dielectric strength, mechanical robustness, and thermal stability. However, organic–inorganic composite systems suffer from undesirable performance and accelerated degradation due to leakage current (JTot). Unfortunately, the underlying mechanism of JTot and its components (e.g., ionic and electronic constituents) are inadequately understood, particularly in extreme use conditions (e.g., high humidity and temperature). In this study, we use numerical simulation and experimental JTot data (in amorphous epoxy polymer with silica fillers) to (i) unify the electrostatic model for JTot in composite polymers, (ii) illustrate that the early part of JTot (i.e., external current) is primarily due to the image charge associated with ion transport/ localization (Jion) near the metallic contacts, (iii) demonstrate that the accumulated counter-ions reduce the barrier for electronic charge injection (by band bending) and facilitate electronic injection from the metals (Jelec), and (iv) provide an algorithm for the in situ ion transport characterization of composite insulators by exploiting Jion. This work provides new insights regarding the leakage current mechanism and how it can be used as a probe into the complex transport mechanisms of the composite material.

Large eddy simulation of a turbulent channel flow using dynamic Smagorinsky subgrid scale model and differential equation wall model

A large eddy simulation (LES) is conducted in a plane channel flow which is turbulent. For understanding the mechanism of the turbulent motion more efficiently, the present study is an attempt to conduct the LES using a Dynamic Smagorinsky Model (DSM) and a Differential Equation Wall Model (DEWM). DSM is a subgrid-scale model and DEWM is for approximating the near wall layer. The simulation is performed in a domain of 2 π δ × 2 δ × π δwhich is discretized by 32 × 20 × 32grid points where δ is the channel half width. The discretization techniques have been used in the simulation, are the 3rd order low storage R-K method in time and 2nd order finite difference formulation in space. To assess the performance of the LES with DSM and DEWM, the simulation results are compared with direct numerical simulation (DNS) data and LES data using Standard Smagorinsky Model (SSM). Comparing the results in the essential turbulence statistical field it has been found that DSM performs better than SSM in most of the cases and shows better agreement with the DNS results. Flow structures in the turbulent flow field are compared in isosurfaces and contour plots, and found that tube-like vortical structures are randomly distributed over the entire flow field, and the number of the vortical structures are more for the DSM case than that of SSM.

Dynamic causes contribute to the increasing trend of red tides in the east China sea during 2020–2022

Red tide is a marine phenomenon caused by the excessive growth of microscopic algae in the ocean. This study aims to analyze the development trends of red tides in the past 20 years and the dynamic external causes that induce red tides based on existing satellite remote sensing and numerical simulation data. And the changes in dominant species of red tides in different seasons are analyzed. The results show significant temperature fluctuations within the week before the red tide occurs, with an average increase of 1.42 °C. In contrast, the change in salinity is relatively small. Meanwhile, ocean fronts are areas in the ocean where different water masses meet and form boundaries. The average strength of ocean fronts increased by 3.7%, indicating enhanced ocean mixing over a short period of time. Under the combined influence of these factors, the probability of a red tide outbreak in the East China Sea increases rapidly. Therefore, this study has important reference value for further research on the causes of red tides and their response to ocean dynamic changes.

Effect of ultrasound geometry on the production efficiency of damaged starch: Determining rheology parameters, and non-isothermal reaction kinetics

Present study investigates the effects of probe size geometry on thermodynamic kinetics, rheology, and microstructure of wheat and tapioca starch. Ultrasound treatment using different probe diameters (20 mm and 100 mm) significantly influenced the gelatinization process. Results showed reduced enthalpy (ΔH) and Gibbs energy (ΔG), indicating enhanced gelatinization efficiency. According to the results, using a 20 mm and 100 mm probe leads to a reduction of 52.7 % and 68.6 % in reaction enthalpy for wheat starch compared to native starch, respectively. Microstructure analysis revealed structural changes, with ultrasound treatment leading to granular fractures and a sheet-like structure with air bubbles. The rheological behavior of the starches is found to exhibit shear thinning behavior, with the Casson model providing the best fit for the experimental data. Moreover, rheology modeling using Herschel-Bulkley and power law models showed increased viscosity and shear stress in larger probes. Numerical simulation data demonstrated that probe size influenced ultrasonic pressure, sound pressure level, and thermal power dissipation density, affecting fluid motion and velocity field components. Moreover, the maximum dissipated power decreases from 8.43 to 0.655 mW/m3 with an increase in probe diameter from 20 to 100 mm. The average yield shear stress values are calculated as 3.36 and 3.14 for wheat and tapioca starches, respectively. The larger probe diameter leads to greater entropy increases, with tapioca starch showing a 4.72 % increase and wheat starch a 4.97 % increase, compared to 2.56 % and 3.11 %, respectively, with the smaller probe. Additionally, the Keller-Miksis model provided insights into bubble dynamics, revealing increased pressure and temperature with higher pressure amplitudes.

Forward and Inverse Energy Cascade in Fluid Turbulence Adhere to Kolmogorov's Refined Similarity Hypothesis.

We study fluctuations of the local energy cascade rate Φ_{ℓ} in turbulent flows at scales (ℓ) in the inertial range. According to the Kolmogorov refined similarity hypothesis (KRSH), relevant statistical properties of Φ_{ℓ} should depend on ε_{ℓ}, the viscous dissipation rate locally averaged over a sphere of size ℓ, rather than on the global average dissipation. However, the validity of KRSH applied to Φ_{ℓ} has not yet been tested from data. Conditional averages such as ⟨Φ_{ℓ}|ε_{ℓ}⟩ as well as of higher-order moments are measured from direct numerical simulations data, and results clearly adhere to the predictions from KRSH. Remarkably, the same is true when considering forward (Φ_{ℓ}>0) and inverse (Φ_{ℓ}<0) cascade events separately. Measured ratios of forward and inverse cascade probability densities conditioned on ε_{ℓ} also confirm the applicability of the KRSH to analysis of the fluctuation relation from nonequilibrium thermodynamics.

A-priori evaluation of data-driven models for large-eddy simulations in Rayleigh–Bénard convection

Natural convection is a commonly occurring heat-transfer problem in many industrial flows and its prediction with conventional large eddy simulations (LES) at higher Rayleigh numbers using progressively coarser grids leads to increasingly inaccurate estimates of important performance indicators, such as Nusselt number (Nu). Thus, to improve the heat transfer predictions, we utilize Gene Expression Programming (GEP) to develop sub-grid scale (SGS) stress and SGS heat-flux models simultaneously for LES. The models’ development is performed using reference direct numerical simulation data of a typical natural convection case, i.e. the Rayleigh–Bénard convection (RBC). The training frameworks are built by using alignment as the evaluation metrics of the basis functions when comparing to the Gaussian-filtered SGS stress and SGS heat-flux. The trained models in isotropic form, by utilizing the L2 norm of the grid cell ΔL2 as the length scale, demonstrate good performance in the bulk region, but less improved performance in the near wall region. It is shown, that for LES of wall-bounded flow, the GEP models in anisotropic form, i.e. using different grid length scale for the different spatial directions, are required to obtain generalized models suitable for different regions. Consequently, the a-priori results demonstrate a significant improvement in the prediction of both instantaneous and mean quantities for a wide range of filter widths.

Estimation of load for tunnel lining in elastic soil using physics‐informed neural network

AbstractA reverse calculation method termed soil and lining physics‐informed neural network (SL‐PINN) is proposed for the estimation of load for tunnel lining in elastic soil based on radial displacement measurements of the tunnel lining. To achieve efficient and accurate calculations, the framework of SL‐PINN is specially designed to consider the respective displacement characteristics of surrounding soil and tunnel lining. A multistep training method based on the meshless characteristics of SL‐PINN is established to promote calculation efficiency. The multistep training method involves increasing the number of collocation points in each calculation step while decreasing the learning rate after scaling of SL‐PINN. The feasibility of SL‐PINN is verified by numerical simulation data and field data. Compared to other inverse calculation methods, SL‐PINN has lower precision requirements for the measurement instrument with the same level of calculation accuracy.

New momentum integral equation applicable to boundary layer flows under arbitrary pressure gradients

By incorporating the traditionally overlooked advective term in the wall-normal momentum equation, a new momentum integral equation is developed for two-dimensional incompressible turbulent boundary layers under arbitrary pressure gradients. The classical Kármán's integral arises as a special instance of the new momentum integral equation when the pressure gradient is weak. The new momentum integral equation's validity is substantiated by direct numerical simulation data. Unlike the classical Kármán's integral, which is limited to predicting wall shear stress within mild pressure gradients, the new momentum integral equation accurately computes wall shear stress across a broad range of pressure gradients, even in the presence of strong adverse pressure gradients that lead to flow separation. Moreover, a new pressure parameter $\beta _\kappa$ is introduced through examining terms in the new momentum integral equation. This parameter naturally quantifies the pressure gradient's influence on turbulent boundary layers and offers guidance for applying the classical Kármán's integral. Additionally, to facilitate experimental determination of wall shear stress under strong pressure gradients, an approximate integral equation is proposed that relies solely on easily measurable variables. Validation against direct numerical simulation data demonstrates that this simplified equation provides reasonably accurate estimates of wall shear stress in turbulent boundary layers experiencing strong pressure gradients.

Numerical investigation on upper layer thickness and temperature rise in a fire in two rooms connected through a side vent

Upper layer thickness and temperature rise in a fire in two rooms connected through a side vent were numerically investigated. Side vents of SV-1 and SV-2 were installed between two rooms (R-1 and R-2) and between R-2 and the outside, respectively. Situations in which fires occurred in R-1 and R-2, referred to as R-1-F and R-2-F, respectively, were tested. Effects of heat release rate (HRR), SV-1 width (W1), SV-2 width (W2), and room boundary wall material (hk) on the upper layer thickness and temperature rise in the fire and adjacent rooms were examined using the velocity profile and mass flow rate of vent flow. W1 and W2 had the greatest impact on the upper layer thickness in R-1 of R-1-F and in the other rooms, respectively, whereas the effect of hk was the smallest. HRR and hk had the greatest and smallest impacts, respectively, on the upper layer temperature rise. Bias factors and model's relative standard deviations of upper layer thickness, upper layer temperature rise, and vent flow velocity were evaluated. Based on the numerical simulation data and bias factor, correlations for predicting the upper layer temperature rises in the fire and adjacent rooms of R-1-F and R-2-F were proposed.

Theoretical analysis and numerical study of natural convection inside combined solar chimney

AbstractSolar chimneys may induce natural ventilation through solar radiation. However, sufficient theoretical studies are needed as a basis to fully exploit passive design in practical green building design. In this work, we investigate the heat transfer properties of turbulent natural convective flows in a combined solar chimney with a thermal flux at the absorption wall by means of theoretical analysis and numerical simulations. Two different flow patterns have been found, one with a clear thermal boundary layer flow pattern and the other without, based on high Rayleigh numbers. For flow development in these two flow regimes, the transient scaling analysis is performed separately and the control mechanism for each phase is presented. Some new scale relationships are established to characterize the ventilation performance of solar chimneys, including thermal boundary layer thickness δT, velocity vT, mass flow m, and so on. For the distinct thermal boundary layers, δT,s ~ HΓ2/5/Bo1/5κ2/5, vT,s ~ Bo2/5κ4/5Γ1/5/H, m ~ ρBo1/5κ2/5Γ3/5. For nonobvious thermal boundary layers, vT,f ~ Bo1/3κ/H2/3W1/3, m ~ ρBo1/3κ/A2/3. The important scale relationships are validated using corresponding numerical simulation data, such as the mass flow rate scale M ~ (Γ/κ)3/5Bo1/5 in the distinct thermal boundary layer flow state, and so on. The air changes per hour and heat exchange efficiencies are calculated for a solar chimney with a fixed height‐to‐width ratio to provide a basis for the design of a solar chimney.

Data-driven model for Lagrangian evolution of velocity gradients in incompressible turbulent flows

Velocity gradient tensor, $A_{ij}\equiv \partial u_i/\partial x_j$ , in a turbulence flow field is modelled by separating the treatment of intermittent magnitude ( $A = \sqrt {A_{ij}A_{ij}}$ ) from that of the more universal normalised velocity gradient tensor, $b_{ij} \equiv A_{ij}/A$ . The boundedness and compactness of the $b_{ij}$ -space along with its universal dynamics allow for the development of models that are reasonably insensitive to Reynolds number. The near-lognormality of the magnitude $A$ is then exploited to derive a model based on a modified Ornstein–Uhlenbeck process. These models are developed using data-driven strategies employing high-fidelity forced isotropic turbulence data sets. A posteriori model results agree well with direct numerical simulation data over a wide range of velocity-gradient features and Reynolds numbers.

Sensitivity and resolution response of optical flow-based background-oriented schlieren to speckle patterns

The degraded resolution and sensitivity characteristics of background-oriented schlieren (BOS) can be recovered by utilizing an optical flow (OF)-based image processing scheme. However, the background patterns conventionally employed in BOS setups suit the needs of the cross-correlation approach, whereas OF is based on a completely different mathematical background. Thus, in order to characterize the resolution and sensitivity response of OF-based BOS to the background generation configurations, a parametric study is performed. First, a synthetic assessment based on an analytical solution of a one-dimensional shock tube problem is conducted. Then, a numerical assessment utilizing direct numerical simulation data of density-driven turbulence is performed. Finally, the applicability of the documented conclusions in realistic scenarios is tested through an experimental assessment over a plume of a swirling heated jet.

Numerical simulation and experimental study on seafloor sampling of an anchor-type mud collector based on SPH-FEM coupling method

Marine sediments are important for research in scientific fields such as marine geology, environmental testing of waters, marine biology and seabed resource exploration. Among them, mud miner is an important way to obtain sediments. However, due to the complexity of the marine environment, the seabed sampling operation is a relatively difficult and complicated project. The structural design of the sampler, the operation mode, and the interaction between the sampler and the sediment affect the sampling effect, which leads to the low efficiency of the sampler operation. In order to investigate the main factors affecting the drag force of the sampler during seafloor sampling, this paper takes a simple and portable anchor-type mud collector surface sediment sampler as a study. This paper introduces the mechanical structure and working principle of an anchor-type mud collector, establishes a mechanical model of the mud collector seafloor sampling process and derives the main factors affecting the dragging force: internal friction angle; the horizontal angle of the bi-directional shaft rod; undercut angle [Formula: see text] of bottom cover and dragging speed. A FEM-SPH method based on the coupling of the finite element method (FEM) and the smooth particle hydrodynamics method (SPH) was proposed to simulate the dynamic process of mud collector subsea sampling, and the mechanical data of fine sand and clay were obtained through land-based experiments. Based on the comparison between experimental data and numerical simulation data, the simulation validity of the FEM-SPH method was verified. The results show that the drag force of the clay was greater than that of the fine sand in the experiment of cutting the lower cover of the mud collector into the fine sand and clay, the internal friction angle of the clay is greater than that of the fine sand; with the increase in horizontal inclination angle and the decrease in undercut angle, the dragging force gradually increased; The dragging speed ranged from 0.2 to 0.6 m/s, with an increase in the mass of clay and fine sand collected with increasing speed, resulting in a gradual increase in dragging force. This paper provides a new method to study the force of the mud collector, provides a theoretical method to reduce the intensity and difficulty of manual work in the sampling process and increase the efficiency of sampling.

Prediction of turbulent energy based on low-rank resolvent modes and machine learning

A modelling framework based on the resolvent analysis and machine learning is proposed to predict the turbulent energy in incompressible channel flows. In the framework, the optimal resolvent response modes are selected as the basis functions modelling the low-rank behaviour of high-dimensional nonlinear turbulent flow-fields, and the corresponding weight functions are determined by data-driven neural networks. Turbulent-energy distribution in space and scales, at the friction Reynolds number 1000, is predicted and compared to the data of direct numerical simulation. Close agreement is observed, suggesting the feasibility and reliability of the proposed framework for turbulence prediction.