Field Of Fluid Dynamics Research Articles

Three-dimensional analysis of turbulent twin-swirling jets onto a heated rectangular prism in a channel

Purpose The purpose of this study is to investigate the impact of swirling jet flow on the cooling performance of a heated rectangular prism placed within a channel. The primary aim is to explore the influence of varying aspect ratios (AR) of the prism and different fluid Reynolds numbers (Re) on the cooling efficiency. Design/methodology/approach The numerical analysis is performed using a finite volume-based solver, which incorporates the large eddy simulations (LES) turbulence model. The setup consists of twin 45° swirling jets directed at isothermally heated bodies, with water used as the cooling medium. The rectangular prism is oriented perpendicularly to the channel flow direction, positioned one unit distance from the inlet. This study examines three distinct aspect ratios (AR = 0.5, 1 and 1.5) and a range of Reynolds numbers (6000 = Re = 20000). Findings The results indicate that cooling efficiency improves as the aspect ratio decreases and the Reynolds number increases. Higher Reynolds numbers enhance jet impingement and turbulent mixing, which are crucial for efficient heat transfer. Conversely, lower Reynolds numbers lead to diminished impingement and reduced cooling efficiency. Increasing the Reynolds number from 6000 to 20000 elevates the average Nusselt number by 35% (for AR = 0.5) and up to 45% (for AR = 1.5). It was observed that lower aspect ratios produce superior cooling effects due to intensified localized jet interactions. Originality/value This research significantly contributes to the fields of fluid dynamics and thermal engineering by elucidating the influence of swirling jet flows on the cooling of heated surfaces. The findings offer valuable insights for optimizing the design and performance of cooling systems across various industrial applications.

Analysis of heat transfer and AuNPs-mediated photo-thermal inactivation of E. coli at varying laser powers using single-phase CFD modeling

PurposeIn the wake of the COVID-19 pandemics, the demand for innovative and effective methods of bacterial inactivation has become a critical area of research, providing the impetus for this study. The purpose of this research is to analyze the AuNPs-mediated photothermal inactivation of E. coli. Gold nanoparticles irradiated by laser represent a promising technique for combating bacterial infection that combines high-tech and scientific progress. The intermediate aim of the work was to present the calibration of the model with respect to the gold nanorods experiment. The purpose of this work is to study the effect of initial concentration of E. coli bacteria, the design of the chamber and the laser power on heat transfer and inactivation of E. coli bacteria.Design/methodology/approachUsing the CFD simulation, the work combines three main concepts. 1. The conversion of laser light to heat has been described by a combination of three distinctive approximations: a- Discrete particle integration to take into account every nanoparticle within the system, b- Rayleigh-Drude approximation to determine the scattering and extinction coefficients and c- Lambert–Beer–Bourger law to describe the decrease in laser intensity across the AuNPs. 2. The contribution of the presence of E. coli bacteria to the thermal and fluid-dynamic fields in the microdevice was modeled by single-phase approach by determining the effective thermophysical properties of the water-bacteria mixture. 3. An approach based on a temperature threshold attained at which bacteria will be inactivated, has been used to predict bacterial response to temperature increases.FindingsThe comparison of the thermal fields and temporal temperature changes obtained by the CFD simulation with those obtained experimentally confirms the accuracy of the light-heat conversion model derived from the aforementioned approximations. The results show a linear relationship between maximum temperature and variation in laser power over the range studied, which is in line with previous experimental results. It was also found that the temperature inside the microchamber can exceed 55 °C only when a laser power higher than 0.8 W is used, so bacterial inactivation begins.Research limitations/implicationsThe experimental data allows to determinate the concentration of nanoparticles. This parameter is introduced into the mathematical model obtaining the same number of AuNPs. However, this assumption introduces a certain simplification, as in the mathematical model the distribution of nanoparticles is uniform.Practical implicationsThis work is directly connected to the use of gold nanoparticles for energy conversion, as well as the field of bacterial inactivation in microfluidic systems such as lab-on-a-chip. Presented mathematical and numerical models can be extended to the entire spectrum of wavelengths with particular use of white light in the inactivation of bacteria.Originality/valueThis work represents a significant advancement in the field, as to the best of the authors’ knowledge, it is the first to employ a single-phase computational fluid dynamics (CFD) approach specifically combined with the thermal inactivation of bacteria. Moreover, this research pioneers the use of a numerical simulation to analyze the temperature threshold of photothermal inactivation of E. coli mediated by gold nanorods (AuNRs). The integration of these methodologies offers a new perspective on optimizing bacterial inactivation techniques, making this study a valuable contribution to both computational modeling and biomedical applications.

One-dimensional temperature measurement of supersonic jet flow by resonantly ionized photoemission thermometry of molecular nitrogen

As the field of fluid dynamics progresses, the demand for sophisticated diagnostic methods to accurately assess flow conditions rises. In this work, resonantly ionized photoemission thermometry (RIPT) has been used to directly target and ionize diatomic nitrogen (N2) to measure one-dimensional (1D) temperature profiles in a supersonic jet flow. This technique can be considered non-intrusive as the premise uses resonantly enhanced multiphoton ionization (REMPI) to target molecular nitrogen. This resonance excites N2 into absorption bands of the P, Q, and R rotational branches of N2 (b1Π u ). The ideal (3 + 1) REMPI scheme excites from the ground state and ionizes N2 (b1Π u ←X1Σ g +) where de-excitation results in photoemission from the first negative band of ionized N2+(B2Σu+→X2Σ g +) as nitrogen returns to the ground state. The resulting emission can be observed using an intensified camera, thus permitting inference of the rotational temperature of ground-state molecular nitrogen. A linearly regressive Boltzmann distribution is applied based on previous calibration data for this technique to quantify the temperature along the ionized line. This work applies this technique to a pure N2 supersonic jet in cross-flow and counter-flow orientations to demonstrate N2 RIPT’s applications in a supersonic flow. Temperature variations are observed at different locations downstream of the exit in cross-flow, and axisymmetric in counter-flow, to generate profiles characterizing the flow dynamics. Due to the collisional effects resulting from the number density of N2 at higher pressures, a (3 + 2) REMPI scheme is observed throughout this text.

On-site aerodynamics using stereoscopic PIV and deep optical flow learning

We introduce recurrent all-pairs field transforms for stereoscopic particle image velocimetry (RAFT-StereoPIV). Our approach leverages deep optical flow learning to analyze time-resolved and double-frame particle images from on-site measurements, particularly from the ‘Ring of Fire,’ as well as from wind tunnel measurements for fast aerodynamic analysis. A multi-fidelity dataset comprising both Reynolds-averaged Navier–Stokes (RANS) and direct numerical simulation (DNS) was used to train our model. RAFT-StereoPIV outperforms all PIV state-of-the-art deep learning models on benchmark datasets, with a 68 % error reduction on the validation dataset, Problem Class 2, and a 47 % error reduction on the unseen test dataset, Problem Class 1, demonstrating its robustness and generalizability. In comparison with the most recent works in the field of deep learning for PIV, where the main focus was the methodology development and the application was limited to either 2D flow cases or simple experimental data, we extend deep learning-based PIV for industrial applications and three-component two-dimensional (3C2D) velocity estimation. We believe that this study brings the field of experimental fluid dynamics one step closer to the long-term goal of having experimental measurement systems that can be used for fast flow field estimation.

Optimising Physics-Informed Neural Network Solvers for Turbulence Modelling: A Study on Solver Constraints Against a Data-Driven Approach

Physics-informed neural networks (PINNs) have emerged as a promising approach for simulating nonlinear physical systems, particularly in the field of fluid dynamics and turbulence modelling. Traditional turbulence models often rely on simplifying assumptions or closed numerical models, which simplify the flow, leading to inaccurate flow predictions or long solve times. This study examines solver constraints in a PINNs solver, aiming to generate an understanding of an optimal PINNs solver with reduced constraints compared with the numerically closed models used in traditional computational fluid dynamics (CFD). PINNs were implemented in a periodic hill flow case and compared with a simple data-driven approach to neural network modelling to show the limitations of a data-driven model on a small dataset (as is common in engineering design). A standard full equation PINNs model with predicted first-order stress terms was compared against reduced-boundary models and reduced-order models, with different levels of assumptions made about the flow to monitor the effect on the flow field predictions. The results in all cases showed good agreement against direct numerical simulation (DNS) data, with only boundary conditions provided for training as in numerical modelling. The efficacy of reduced-order models was shown using a continuity only model to accurately predict the flow fields within 0.147 and 2.6 percentage errors for streamwise and transverse velocities, respectively, and a modified mixing length model was used to show the effect of poor assumptions on the model, including poor convergence at the flow boundaries, despite a reduced solve time compared with a numerically closed equation set. The results agree with contemporary literature, indicating that physics-informed neural networks are a significant improvement in solve time compared with a data-driven approach, with a novel proposition of numerically derived unclosed equation sets being a good representation of a turbulent system. In conclusion, it is shown that numerically unclosed systems can be efficiently solved using reduced-order equation sets, potentially leading to a reduced compute requirement compared with traditional solver methods.

The Aerodynamic Assessment and Shot Point Position Accuracy of UAV Seismic Source

An unmanned aerial vehicle (UAV) seismic source, suitable for complex terrains due to its eco-friendly, cost-effective, and efficient nature, operates by hovering and releasing impact objects to generate seismic waves. The aerodynamic behavior of these impact objects is vital for shotpoint position accuracy and drag magnitude. Excessive drag will result in the loss of seismic source energy, whereas the deviations of the shotpoint position will disrupt the geometric structure of the observation system, significantly impacting the quality of seismic data acquisition. Through theoretical calculations, the influencing factors of air resistance and cross-wind deflection are analyzed. Air resistance increases with the increase in mass, cross-sectional area, speed, and drag coefficient. Increasing the mass and throwing height can increase the upper limit of the output, but after the air resistance and gravity are balanced, the output kinetic energy of the drone’s seismic source will no longer increase when the throwing height is increased. Using computational fluid dynamics simulations and field tests, this study examines the aerodynamics of four prevalent impact object designs. The results show that shuttle and spherical structures have the least drag during descent and are less influenced by crosswinds, thus being optimal for this application. Field measurement data are also presented for reference.

Impact of Reynolds and Prandtl Numbers Coupled with Viscous Dissipation on Mixed Convection in A Square Cavity

Efficient thermal management in industrial manufacturing and electronic cooling systems can be achieved by comprehending characteristics of Reynolds and Prandtl numbers in mixed convection scenarios, which aids in the optimization of heat transfer systems through the utilization of forced and natural convection effects. The impact of Reynolds and Prandtl numbers on mixed convection in a square cavity connected to a moving heated horizontal plate is investigated in this research paper. The convective behavior under different conditions was evaluated by discretizing the flow governing equations, which included the momentum and energy equations, using the finite difference method. The research assessed various fluids, including air (Pr = 0.7), liquid metal (Pr = 0.01), and oil (Pr = 10). The Reynolds number ranged from 0.001 to 100, the Eckert number ranged from 0.01 to 40, and a constant Richardson number (Ri = 1) was maintained throughout. The results revealed that the Reynolds number substantially impacts the velocity and temperature characteristics, especially when coupled with a Prandtl number over one and when viscous dissipation remains constant. The utmost velocity that can be attained within the cavity is significantly diminished as the Reynolds number rises, underscoring the critical importance of dynamic fluid properties in determining heat transfer efficiency and fluid flow characteristics. The research unveiled the critical importance of Reynolds and Prandtl numbers in the field of fluid dynamics concerning enhancing heat transfer attributes for engineering purposes, thereby guaranteeing the effectiveness of thermal systems.

Machine Learning-Based Predictions of Flow and Heat Transfer Characteristics in a Lid-Driven Cavity with a Rotating Cylinder

Machine learning-based predictions of heat transfer characteristics in lid-driven cavities are transforming the field of computational fluid dynamics (CFD). Lid-driven cavities are a fundamental problem in fluid mechanics, characterized by the motion of a fluid inside a square cavity driven by the motion of one of its walls. The goal of this study was to develop multiple machine-learning regression models and highlight the discrepancies between the predicted and actual average Nusselt numbers. Additionally, the study utilized physics-informed neural networks (PINNs) to model the flow and thermal behavior at both low and high Reynolds numbers. The results were compared among actual data from computational fluid dynamics (CFD) simulations, PINN models trained with CFD data, and purely PINN models created without any prior data input. The findings of this study showed that the random forest model exhibited an exceptional stability in its predictions, consistently maintaining low errors even as the Reynolds number increased compared with other machine-learning regression models. Further, the results of this study in terms of flow and thermal behavior within the cavity were found to depend significantly on the PINN method. The data-driven PINN exhibited a much lower mean average errors at both Reynolds numbers, while the physics-based PINN showed lower physics loss.

Effects of velocity regularization on neural network performance in processing particle images

Recent studies have witnessed remarkable progress in harnessing convolutional neural networks (CNNs) to overcome the inherent limitations of conventional particle image velocimetry (PIV) methods. Traditional PIV techniques often suffer from compromised resolution and precision, hindering their ability to capture the complexities of fluid dynamics within the observation frame. While CNNs offer promise in addressing these challenges, they face obstacles such as limited accuracy, weak generalization, and a dearth of physical interpretability. In our prior research, we presented a CNN architecture that incorporates optical flow algorithms as supplementary physical constraints, thereby bolstering the model interpretability and precision. Nevertheless, the practical implications of this approach, especially when dealing with multi-dimensional, low-quality particle image data and restricted training sets, have yet to be fully explored. To address this knowledge gap, we have assembled a comprehensive dataset that simulates a wide array of experimental scenarios. We have systematically assessed the influence of velocity regularization on neural network performance, taking into account variations in image quality and the size of training datasets. The results underscore the pivotal importance of velocity regularization in enhancing the predictive prowess of neural networks, particularly when dealing with poor image quality and smaller data sizes. This study provides useful insights into the effective application of CNNs with velocity regularization in the field of experimental fluid dynamics.

A comprehensive review of advances in physics-informed neural networks and their applications in complex fluid dynamics

Physics-informed neural networks (PINNs) represent an emerging computational paradigm that incorporates observed data patterns and the fundamental physical laws of a given problem domain. This approach provides significant advantages in addressing diverse difficulties in the field of complex fluid dynamics. We thoroughly investigated the design of the model architecture, the optimization of the convergence rate, and the development of computational modules for PINNs. However, efficiently and accurately utilizing PINNs to resolve complex fluid dynamics problems remain an enormous barrier. For instance, rapidly deriving surrogate models for turbulence from known data and accurately characterizing flow details in multiphase flow fields present substantial difficulties. Additionally, the prediction of parameters in multi-physics coupled models, achieving balance across all scales in multiscale modeling, and developing standardized test sets encompassing complex fluid dynamic problems are urgent technical breakthroughs needed. This paper discusses the latest advancements in PINNs and their potential applications in complex fluid dynamics, including turbulence, multiphase flows, multi-field coupled flows, and multiscale flows. Furthermore, we analyze the challenges that PINNs face in addressing these fluid dynamics problems and outline future trends in their growth. Our objective is to enhance the integration of deep learning and complex fluid dynamics, facilitating the resolution of more realistic and complex flow problems.

The Effect of Reynolds Numbers on Flow-Induced Vibrations: A Numerical Study of a Cylinder on Elastic Supports

In the field of fluid dynamics, the Reynolds number is a key parameter that influences the flow characteristics around bluff bodies. While its impact on flow around stationary cylinders has been extensively studied, systematic research into flow-induced vibrations (FIVs) under these conditions remains limited. This study utilizes numerical simulations to explore the FIV characteristics of smooth cylinders and passive turbulence control (PTC) cylinders supported elastically within a Reynolds number range from 0.8 × 104 to 1.1 × 105. By comparing the vibration responses, lift coefficients, and wake structures of these cylinders across various Reynolds numbers, this paper aims to elucidate how Reynolds numbers affect the flow and vibration characteristics of these structures. The research employs images of instantaneous lift changes and vortex shedding across multiple sections to visually demonstrate the dynamic changes in flow states. The findings are expected to provide theoretical support for optimizing structural design and vibration control strategies in high-Reynolds-number environments, emphasizing the importance of considering Reynolds numbers in structural safety and design optimization.

Numerical study of hybrid nanofluid and thermal transport in sun-powered energy ship within the application of parabolic trough solar collectors

PurposeRecent advancements in technology have led to the exploration of solar-based thermal radiation and nanotechnology in the field of fluid dynamics. Solar energy is captured through sunlight absorption, acting as the primary source of heat. Various solar technologies, such as solar water heating and photovoltaic cells, rely on solar energy for heat generation. This study focuses on investigating heat transfer mechanisms by utilizing a hybrid nanofluid within a parabolic trough solar collector (PTSC) to advance research in solar ship technology. The model incorporates multiple effects that are detailed in the formulation.Design/methodology/approachThe mathematical model is transformed using suitable similarity transformations into a system of higher-order nonlinear differential equations. The model was solved by implementing a numerical procedure based on the Wavelets and Chebyshev wavelet method for simulating the outcome.FindingsThe velocity profile is reduced by Deborah's number and velocity slip parameter. The Ag-EG nanoparticles mixture demonstrates less smooth fluid flow compared to the significantly smoother fluid flow of the Ag-Fe3O4/EG hybrid nanofluids (HNFs). Additionally, the Ag-Ethylene Glycol nanofluids (NFs) exhibit higher radiative performance compared to the Ag-Fe3O4/Ethylene Glycol hybrid nanofluids (HNFs).Practical implicationsAdditionally, the Oldroyd-B hybrid nanofluid demonstrates improved thermal conductivity compared to traditional fluids, making it suitable for use in cooling systems and energy applications in the maritime industry.Originality/valueThe originality of the study lies in the exploration of the thermal transport enhancement in sun-powered energy ships through the incorporation of silver-magnetite hybrid nanoparticles within the heat transfer fluid circulating in parabolic trough solar collectors. This particular aspect has not been thoroughly researched previously. The findings have been validated and provide a highly positive comparison with the research papers.

Assessment of RANS turbulence models based on the cell-based smoothed finite element model for prediction of turbulent flow

There is a growing body of literature that recognizes the importance of Smoothed Finite Element Method (S-FEM) in computational fluid dynamics (CFD) fields and, to a lesser extent, in complex turbulent flow problems. This study evaluates the performance of Reynolds-averaged Navier-Stokes (RANS) turbulence models within the S-FEM framework for predicting incompressible turbulent flows. Our assessment of three turbulence models based on the cell-based S-FEM (CS-FEM) is convincingly supported by testing on three flow problems. It is found that the CS-FEM exhibits superior mesh robustness compared to the Finite Volume Method (FVM) and achieves higher computational accuracy than the Finite Element Method (FEM). Notably, the CS-FEM combined with the standard k-epsilon model (CS-FEM-SKE) and the realizable k-epsilon model (CS-FEM-RKE) demonstrate robust performance in handling severely distorted meshes, with CS-FEM-RKE outperforming in regions of strong flow separation and convection. The Spalart-Allmaras model with CS-FEM (CS-FEM-SA) offers faster computational speed but shows poor mesh robustness. The hexcore mesh based on CS-FEM-RKE is employed to evaluate the aerodynamic performance of High-speed train (HST), resulting in enhanced computational efficiency. The outcomes show good agreement with other numerical studies and experimental data. Overall, it also highlights the latent capability of CS-FEM in solving complex engineering problems.

Variable coefficient-informed neural network for PDE inverse problem in fluid dynamics

Variable-coefficient equations are crucial in the field of fluid dynamics as they accurately capture the spatial and temporal variations of fluid properties. In many cases, there exist some constraints among multiple coefficients and embedding these constraints into neural networks poses a challenge. In this paper, we design a variable coefficient-informed neural network (VCINN) to address the inverse problem of variable-coefficients partial differential equation in fluid dynamics. The VCINN framework integrates the physics-informed neural network (PINN) with the constraints among multiple coefficients, encoding both constraints and physics information into the neural networks. Compared to classical PINN, VCINN enjoys such several advantages as parallelization capacity, embedding constraint information and efficient hyperparameter adjustment. Through series of examples, the capability of the approach to recover coefficients from observations has been demonstrated. Numerical results indicate that the present method achieves higher accuracy and lower training error compared to classical PINN.

Entropy optimization of a FENE-P viscoelastic model: a numerically guided comprehensive analysis

The influence of polymers on entropy generation processes is substantial, particularly in the fields of fluid dynamics and rheology. The FENE-P (Finitely Extensible Nonlinear Elastic-Peterlin) model describes the polymer’s dynamics as a result of the interaction between the stretching caused by the velocity gradient and the elastic force that restores the polymer to its equilibrium position. Models such as FENE-P aid in understanding and predicting polymer flow behaviour allowing for the reduction of entropy generation by optimizing system designs. A continuum approach is employed to express the heat flux vector and polymer confirmation tensor of the model. The study investigates the complex relationship between polymer conformation, flow dynamics, and heat transfer taking into account the thermophoresis (Soret effect) and mass diffusion-thermal diffusion coupling (Dufour effect) phenomena to optimize processes by reducing entropy. This study illuminates polymer’s significance in entropy minimization, improving engineering design methodologies and applications in materials science, chemical engineering, and fluid dynamics. As result, the presence of polymers leads to a substantial decrease in the total entropy of the system. This understanding provides opportunities for enhancing heat transfer systems, thereby facilitating the development of more efficient and sustainable technology.

Stochastic neuro-swarming intelligence paradigm for the analysis of magneto-hydrodynamic Prandtl–Eyring fluid flow with diffusive magnetic layers effect over an elongated surface

In recent years, the integration of stochastic techniques, especially those based on artificial neural networks, has emerged as a pivotal advancement in the field of computational fluid dynamics. These techniques offer a powerful framework for the analysis of complex fluid flow phenomena and address the uncertainties inherent in fluid dynamics systems. Following this trend, the current investigation portrays the design and construction of an important technique named swarming optimized neuro-heuristic intelligence with the competency of artificial neural networks to analyze nonlinear viscoelastic magneto-hydrodynamic Prandtl–Eyring fluid flow model, with diffusive magnetic layers effect along an extended sheet. The currently designed computational technique is established using inverse multiquadric radial basis activation function through the hybridization of a well-known global searching technique of particle swarm optimization and sequential quadratic programming, a technique capable of rapid convergence locally. The most appropriate scaling group involved transformations that are implemented on governing equations of the suggested fluidic model to convert it from a system of nonlinear partial differential equations into a dimensionless form of a third-order nonlinear ordinary differential equation. The transformed/reduced fluid flow model is solved for sundry variations of physical quantities using the designed scheme and outcomes are matched consistently with Adam's numerical technique with negligible magnitude of absolute errors and mean square errors. Moreover, it is revealed that the velocity of the fluid depreciates in the presence of a strong magnetic field effect. The efficacy of the designed solver is depicted evidently through rigorous statistical observations via exhaustive numerical experimentation of the fluidic problem.

Flow dynamics and heat transfer in the wake of an rotary oscillating circular cylinder with an isothermal control plate

This research explores the flow dynamics and heat transfer within the wake of a rotary oscillating circular cylinder in the presence of an isothermal control plate. Two-dimensional, unsteady, viscous and laminar flow of a Newtonian fluid is considered using the Higher Order Compact Scheme (HOC) to discretize the governing equations and the Bi-Conjugate Gradient Stabilized method to solve the resulting system. Simulations are performed for various gap ratios, maximum angular velocities, and frequency ratios of oscillation at Prandtl number 0.7 and Reynolds number 150 using an in-house code. Results show a significant increase in heat transmission for d / R 0 = 0.5 and f / f 0 = 0.5 for all αm . Drag and lift coefficients are also analyzed, with the maximum peak of the drag coefficient decreasing by 9.88% for d / R 0 = 3 . This research offers valuable insights for advancing and optimizing aerodynamic forces and heat transfer processes, particularly in the field of fluid dynamics. The study focuses on enhancing the efficiency of heat transfer mechanisms around rotary oscillating circular cylinders, contributing to the development of cutting-edge technologies in aerodynamics and thermal management.

Real-time vacuum plume flow field reconstruction during lunar landings based on deep learning

In space missions, the vacuum plume generated by rocket engines can negatively impact spacecraft. Therefore, researching the vacuum plume is crucial to guarantee the regular operation of spacecraft. The conventional numerical simulation methodology, the direct simulation Monte Carlo (DSMC) method, is time-consuming and lacks real-time calculation capabilities. Recently, deep learning (DL) methods have emerged in the field of fluid dynamics. In this study, a DL model trained by a convolutional neural network with multiple decoders is introduced to predict the vacuum plume flow field during lunar landings. The network processes shape topology information and boundary conditions as inputs, yielding flow field data including velocity and pressure fields as outputs. Meanwhile, the flow field prediction results under different conditions and training methods are discussed. The results show that the predicted flow field under different lunar surface conditions is in accord with the DSMC results. The maximum mean and standard deviation errors of the data distribution of each flow field do not exceed 9.72% and 9.07%, respectively. Different training methods with flat and inclined lunar surfaces also have an impact on the prediction results. Compared with the DSMC method, the DL method exhibits higher efficiency with a speedup of about four orders of magnitude, indicating that the DL-based flow field reconstruction method has strong application prospects in the real-time computation of vacuum plume flow fields.

Experimental and numerical simulation of the attenuation effect of blast shock waves in tunnels at different altitudes

Traffic engineering such as tunnels in various altitudinal gradient zone are at risk of accidental explosion, which can damage personnel and equipment. Accurate prediction of the distribution pattern of explosive loads and shock wave propagation process in semi-enclosed structures at various altitude environment is key research focus in the fields of explosion shock and fluid dynamics. The effect of altitude on the propagation of shock waves in tunnels was investigated by conducting explosion test and numerical simulation. Based on the experimental and numerical simulation results, a prediction model for the attenuation of the peak overpressure of tunnel shock waves at different altitudes was established. The results showed that the peak overpressure decreased at the same measurement points in the tunnel entrance under the high altitude condition. In contrast, an increase in altitude accelerated the propagation speed of the shock wave in the tunnel. The average error between the peak shock wave overpressure obtained using the overpressure prediction formula and the measured test data was less than 15%, the average error between the propagation velocity of shock waves predicted values and the test data is less than 10%. The method can effectively predict the overpressure attenuation of blast wave in tunnel at various altitudes.

Machine learning approach to predict flow fields induced by internal solitary waves acting on mid-water structures based on particle image velocimetry experiments

Mid-water structures (MWSs) have important applications in the ocean to minimize the effects of environmental factors in the surface ocean and to avoid the effects of high pressure on the seabed. However, their operational positions are vulnerable to internal solitary waves (ISWs), which poses a significant challenge to the safe functioning of MWSs. Consequently, there is a pressing need to explore the flow field characteristics surrounding MWSs under the action of ISWs to establish a reliable foundation for structural design. In this study, experimental investigations were conducted to analyze the flow fields induced by ISWs acting on MWSs, employing the particle image velocimetry (PIV) method. The intricate details of the flow fields around the structure were thoroughly analyzed, and the impact of the structure's submerged depth on the ISW-induced flow fields was discussed. A machine learning approach is utilized to develop an Artificial Neural Network (ANN) model aimed at predicting the flow fields of ISWs. The experimental results are compared and analyzed alongside the predicted outcomes, focusing on the anomaly region, the tail field region, and the occluded region. The results show that the presence of the structure complicates the flow characteristics in the tail flow regions, and the flow patterns are constantly changing. The characteristics of the ISW flow fields around the structure vary with the different submerged depths of the MWS. Positioning the MWS above the wave trough results in an ascending trend followed by a descending trend in the strength of the surrounding flow field with increasing submerged depth. The ANN predictions for different types of regions of the flow fields show satisfactory accuracy. The model closely matches the PIV results in predicting the extremes and positions of horizontal velocities in the tail flow fields at different stages of ISW trough propagation. The ANN model can be recommended for predicting the flow fields of ISWs acting on MWSs. This study offers comprehensive insights into the characteristics of the flow fields induced by ISWs acting on MWSs, emphasizing the valuable contribution of machine learning methods to the field of fluid dynamics.