Dynamic Flow Behavior Research Articles

Self-adaptive equation embedded neural networks for traffic flow state estimation with sparse data

The data-driven approach in intelligent traffic systems has achieved successive breakthroughs, thanks to the ever-increasing volume of traffic data. Nonetheless, in practical scenarios, the collected data often contain some issues, e.g., missing values, significantly impacting the accuracy and efficiency of the algorithms. To enhance the precision of traffic estimation utilizing the sparse data, we have developed a physics-informed neural network (PINN) based algorithm in the line with the traffic flow theory and deep learning principles. In contrast to the conventional PINNs, our approach uniquely incorporates a self-adaptive macro model for mixed flow into the network's architecture, serving as an embedded source of physics information. With this algorithm, we can capture the dynamic behavior of an entire traffic flow including its spatiotemporal evolution with sparse traffic data such as initial and boundary value information. To realize the model's adaptability, we have revised the macro model by inverting its parameters and incorporating a data-driven nonlinear element, which simplifies the intricate macro model structure. The network's effectiveness has been validated through the experiments conducted on a mixed traffic flow system experiencing local agglomeration and real-world data, demonstrating its capability for precise traffic simulation, efficient traffic flow prediction, and interpretability. Our study offers a novel insight for data-driven traffic flow state estimation.

Effects of ultrasound on bubble dynamic behavior of flow boiling in microchannel

Bubble dynamics is paramount in comprehending the heat transfer mechanisms of flow boiling in the microchannel within ultrasonic field, which is regarded as a promising method to confront challenges of thermal management posed by microelectronic devices. Nevertheless, the impact of ultrasound on bubble behaviors and its underlying mechanisms remain largely unexplored. This study first delves into the effect of ultrasonic parameters on bubble dynamic behaviors and associated mechanisms, subsequently further analyzing the forces acting on bubbles through the constructed force model. The findings suggest that although growth force serves as the significant resistance, the primary Bjerknes force dominates the rapid detachment of bubbles. The secondary Bjerknes force results in the bubble only sliding along the bottom wall rather than lifting off. Furthermore, the elevated ultrasonic pressure amplitude resulting from augmenting ultrasonic power induces a substantial increase in the critical detachment diameter and growth rate by 55.49% and 59.42%, respectively. The enhanced primary Bjerknes force, attributed to the rise in ultrasonic frequency, leads to a 71.42% increase in sliding velocity and a 46.45% reduction in growth time. The positive impacts arising from ultrasonic power and frequency are anticipated to notably enhance the thermal performance of microchannels. Besides, surface tension acts as the resistance and diminishes slightly with an augmentation of the boiling number, resulting in a moderate variation in sliding velocity and growth time.

Nusselt and Reynolds numbers correlation for oscillatory flow of thermoacoustics over heated tube banks

The back-and-forth motion characteristic of the oscillatory flow brings complexity in predicting the heat transfer nature for thermoacoustic wave conditions. Investigating this is crucial as the flow forms the backbone for the thermodynamic cycles of its operation. Often, the steady approximation in calculating heat transfer leads to ambiguity in designing the system. This concern is addressed in this paper via an in-depth analysis of the fluid dynamic behavior and heat transfer characteristics of oscillatory flow within a thermoacoustic framework, spanning experimental works for Reynolds numbers ranging from 300 to 24,000. The investigation focused on a tube bank heat exchanger composed of nine tubes arranged in inline and staggered configurations. Experiments were conducted at constant tube surface temperatures treated at 40 °C and 80 °C, respectively. The distinguishing difference in heat transfer behaviors for oscillatory flow is visualized by the Computational Fluid Dynamics (CFD) models, employing the Reynolds-Averaged Navier Stokes equation with the k-omega turbulence shear stress transport model. The flow never leaves the system, and it cyclically crosses the tubes back and forth with a travel distance that depends on the drive ratio of the flow. The unique nature of the flow forms the foundation for the experimental findings that show a linear relationship between the Nusselt and Reynolds numbers regardless of the configuration of tube banks and tube bank temperature. The Pearson ruler regression analysis was conducted using Matlab R2022b and a Nusselt correlation, Nu = 0.000853RePr⅓, is proposed with a confidence level of 95 %. Notably, the correlation aligns with the constant value known as the Colburn-j factor, with a value of 0.00083. The small Colburn-j value is shown to be the influence of the log-mean temperature difference characteristic of oscillatory flow. It shows a consistent heat transfer process between the heated tubes and surrounding fluid, which is important to sustain the thermoacoustic effect in the system. For future thermoacoustic design, the use of steady heat transfer correlation should be avoided or at least used with caution as the comparative analysis with published works concluded that the Nusselt and Reynolds correlation for steady flow tends to overpredict the heat transfer by two to threefold.

Processing, experimental characterization and thermal analysis for friction stir welded joint of 2219Al alloy

ABSTRACT Present study aimed at thermal modelling and numerical analysis of dynamic heat flow behaviour, for friction stir welded (FSW) joint of 2219Al alloy, further experimentally correlated with microstructural evolution and microhardness profile through different weld zones. FSW was performed on cast 2219Al alloy, with varying process parameters of welding and rotational speed of the tool, and transient temperatures measured at different distances from weld line. From microstructural analysis, three different heat affected weld zones having separate grain morphologies were identified, namely Weld Nugget Zone (WNZ), Thermo-Mechanical Affected Zone (TMAZ) and Heat Affected Zone (HAZ). Vickers microhardness was evaluated at varying distances from weld line, through different weld zones. To investigate transient temperature profiles over the welded plate, a 3-D Finite Element (FE) thermal model was developed, incorporating volumetric heat source as heat generation. Transient temperature profiles computed from FEM analysis, were successfully compared and validated with experimental results, with fairly good accuracy within 4.86 %. Individual influences of welding parameters, on transient temperature profile, weld zones, microstructural and mechanical property were investigated, to optimise and achieve desired weld quality. Present study established structure-property correlation, along with thermal analysis, to validate and optimise the potential application of FSW on 2219Al alloy.

Numerical modeling of dynamic heat flow behaviour during friction stir welding of Al-Cu alloy

Numerical modeling of dynamic heat flow behaviour during friction stir welding of Al-Cu alloy

Analysis of Multistability of Financial Risk Chaos Systems and Its Application to Voice Cryptography

In the chaos literature, the application of modeling and control of dynamic systems in chaos theory arising in several fields is investigated. In this article we analyze complex financial chaos systems with countries as interest rates, investment demand, and price indices. The proposed chaotic flow's dynamic behavior is examined using phase portraits, eigenvalues, bifurcation diagrams, and Lyapunov exponent spectra. A significant quantity of research on secure communication systems has been published in recent years as a result of the major advancements in communications equipment and encryption techniques. A new voice encryption algorithm design is given using a financial chaos model. An application for voice encryption is conducted using the suggested algorithm, and the outcomes are described.

Dynamic flow behaviors of an inlet isolator in embedded rocket-driven mode transitions

Dynamic flow behaviors of an inlet isolator in embedded rocket-driven mode transitions

Numerical study on steel flow dynamic behaviour in a round mould under the effect of a swirling flow brick design in tundish

Swirling flow submerged entry nozzle technology is one of the important methods to optimise the continuous casting of steel. In this study, a novel swirling flow brick design in tundish was proposed to achieve a swirling flow submerged entry nozzle. Numerical simulations were carried out to investigate steel flow dynamic behaviour in a round mould under the effect of new designs. The results showed that a rotational steel flow was generated inside the submerged entry nozzle by using both two-inlet and three-inlet brick designs. Due to the rotational flow momentum, molten steel from the submerged entry nozzle outlet moved towards the mould wall. Such a flow pattern eliminated the impinging flow in mould, which was generally formed in conventional single-port submerged entry nozzle casting. When using swirling flow brick designs, the whole flow field in the submerged entry nozzle and mould exhibited a dynamic change. As the side-inlet number of bricks decreased from three-inlet to two-inlet, the velocity fluctuation range was reduced from ± 63.3% to ± 8.9%. Moreover, the swirling flow intensity in the submerged entry nozzle was decreased by about 40%. In addition, a high-pressure region above atmospheric pressure was obtained near the submerged entry nozzle wall by using new designs, which is helpful to prevent air from being sucked into the submerged entry nozzle.

Dynamic stress prediction for a Pump-Turbine in Low-Load Conditions: Experimental validation and phenomenological analysis

To meet the increasing demand for higher flexibility in hydropower, an operating range extension in an existing hydroelectric powerplant is considered. To this end, the dynamic flow and structural behavior of the low-head pump-turbine is investigated in turbine mode in partial load and deep partial load conditions, to evaluate the feasibility of this flexibilization due to the increased high cycle fatigue damage. In a first step, different numerical simulation approaches that combine fluid dynamics and structural dynamics are applied and compared to in-situ dynamic strain and pressure measurements on the impeller. As a result, the modelling requirements for a numerical workflow that leads to unmatched accuracy in predicting the broad-band dynamic stresses of deep partial load are identified. In a second step, the operating-point-dependent fatigue crack initiation spots are identified based on dynamic stresses. Herein, the emergence of an additional dominant critical spot near the blade leading edge is observed, which is unexpected for low-head pump turbines. Thanks to the detailed numerical flow and structural analyses, the root cause for the dynamic stress concentration is studied and the emergence of the new critical spot is understood. It turns out in deep partial load, that the main source of pressure oscillations on the impeller blade surfaces is the vortex-dominated flow detachment zone around the blade leading edge, which, in combination with the geometry of the investigated impeller, translates into the main stress oscillations. The results of this work were later used to extend the operating range of the hydroelectric powerplant from initially 25% to 100% power output to now 0% to 100%.

Dynamics of Jeffrey fluid flow and heat transfer: A Prabhakar fractional operator approach

The primary goal of current study is to formulate a novel mathematical model employing the Prabhakar fractional operator, aimed at examining the dynamic behaviour of Jeffrey fluid flow and heat transfer phenomena under ramped wall temperature conditions. Our investigation entails a meticulous investigation of magnetohydrodynamic (MHD) natural convective flow within Jeffrey fluids, where we derive precise analytical solutions utilizing the Prabhakar fractional derivative, characterized by a non-singular type kernel. Furthermore, our study integrates fundamental principles such as Fick’s and Fourier’s laws into the model, leveraging a multi-parameter Mittag-Leffler kernel associated with the fractional operator. We delve into the intricacies of fluid flow near an infinitely vertical plate, considering characteristics such as velocity u0. To address the complexities of the problem, we express it in context the of partial differential equations along side generalized boundary conditions, employing a set of appropriate variables to transform these equations into a dimensionless form. Utilizing Laplace transform, we analyse the equations of fractional system, presenting outcomes both in the form of series and through specialized functions. We systematically explore the influence of key parameters α, Pr, β, Gm, Sc, γ, Gr on fluid flow dynamics, unveiling significant insights. Our comparative analysis reveals the superior performance of the Prabhakar-like non-integer approach over existing operators, substantiated by graphical representations of the results. Additionally, we extend our investigation to various limiting cases, including Newtonian fluids and second-grade, in both fractionalized and classical forms, thus highlighting the versatility and applicability of our proposed model within fluid dynamics research.

Exploring the load characteristics and structural responses of a high-speed vehicle entering water

The process of a trans-medium vehicle crossing from air into water is referred to as water entry. It involves the interplay of air, water, and the vehicle and is a non-stationary process. In this study, we use the coupled Eulerian–Lagrangian method, along with the constitutive Johnson–Cook model and the model of cumulative damage-induced failure, to describe the dynamic plastic flow and fracture-related behavior of the vehicle shell, and use it to develop a method to numerically simulate the process of a high-speed vehicle entering water. When it contacts with water, the elasticity of the medium prompted a significant deflection and deformation in the central area of the head of the vehicle shell. As deformation approached its limit, tensile fractures occurred that caused the shell of the head to separate from the main body. Changes in its angle of water entry influenced the fracture process of the shell. The symmetric, parabolic bending deformation of the head of the vehicle shell occurred around its central axis. The time taken by different types of vehicle heads to fail varied significantly, leading to marked differences in their peak deformation. We determined the quantitative relationship between the dimensionless factor χ and the velocity of water entry, using it to estimate the ultimate water entry velocity for vehicles of different sizes but composed of the same material.

Fault mechanism and dynamic two-phase flow behavior of liquid slugging in reciprocating compressors

Liquid slugging is a fatal fault for large process compressors, leading to transient overpressure, the deformation and fracture of vital pressure-bearing parts, and even gas leakage or explosion. In the study reported here, to reveal the mechanism of overpressure formation, numerical simulations were conducted by means of the volume-of-fluid method to explore the dynamic evolution characteristics of the two-phase flow pattern. Then, high-speed photography was applied to capture the dynamic changes of the liquid boundary in the modified cylinder from different views, thus realizing the validation of the numerical model. This study reveals the significant influence of increased rotational speed on fluid flow patterns, impeding liquid discharge and exacerbating overpressure events. Additionally, changes in pressure waveform and a distinctive waveform feature were identified as effective diagnostic indicators for detecting fluid slugging. Next, a nondestructive pressure monitoring reconstruction method based on measuring bolt strain was proposed. The strain-based pressure showed good agreement with the simulated results, thereby validating its effectiveness and feasibility as an early warning indicator for liquid slugging. This study offers new perspectives on the failure mechanism of liquid slugging in reciprocating compressors by delving into the behavior of two-phase flow, with the potential to enhance the theoretical foundation of compressor condition monitoring and fault diagnosis.

Computational analysis of radon progeny deposition patterns in the human respiratory system

In the last year, the use of computational fluid dynamics (CFD) techniques has gained prominence as a powerful tool for modeling biological phenomena and influencing the design of biomedical devices. In this study, we utilized a computational fluid dynamics (CFD) model to simulate airflow and the deposition of aerosol particles within the human respiratory tract. To achieve this, we meticulously constructed a 3D model of the human tracheobronchial airways using SolidWorks software. Our computational analyses encompassed a range of breathing conditions, ranging from 15 to 60 (L/min). Through the application of discrete phase modeling (DPM), we investigate the behavior of two-phase flow dynamics. Our focus lies in the examination of aerosol particles, with diameters ranging from 1 to 10 (μm), in order to evaluate the influence of aerosol particle size on deposition rates. Our findings encompass velocity contour maps, deposition rates of aerosol particles, and insights into the process of aerosol particle entrapment at various locations within the respiratory tract. Our study reveals a direct correlation between higher inhalation rates and larger aerosol particle sizes, resulting in increased deposition rates. Additionally, we observe a heightened deposition of aerosol-particles at bronchi region. These computational results hold significant value in estimating the distribution of doses resulting from radon progeny exposure in distinct anatomical regions of the respiratory tract.

Multi-layer visibility graph-based ordinal network for revealing gas-liquid nonlinear dynamic flow behaviors

Multi-layer visibility graph-based ordinal network for revealing gas-liquid nonlinear dynamic flow behaviors

Dynamic Mechanical Behavior and Energy Release Effect of a Novel Nb17Zr33Ti17W33 High-Entropy Alloy under Impact Load

The present study successfully demonstrates the fabrication of a novel class of high-entropy alloy, namely Nb17Zr33Ti17W33, through suspension melting and casting technique. To investigate the dynamic mechanical behavior and energy release effects of the alloy under high-speed impact loads, various techniques were employed, including split Hopkinson pressure bar (SHPB), X-ray diffractometer (XRD), scanning electron microscope (SEM), and high-speed photography. These methods were utilized to acquire crucial data, such as crystal structure analysis, stress–strain curves, and microstructural examination of failed specimens. The modified Johnson–Cook (J-C) model was employed to elucidate the dynamic flow behavior of the alloy, while investigating the failure mechanism and energy release phenomenon during the process of dynamic compression. The experimental results demonstrate that the alloy material exhibits a dual-phase (BCC1 + BCC2) structure, exhibiting ductile fracture behavior under dynamic compression conditions. On the fracture surface, typical dimple structures along with evidence of shear slip and melting traces were observed, indicating an energy-releasing failure process. The newly developed alloy exhibited exceptional strength, high density, remarkable plasticity, and outstanding energy release properties, rendering it highly promising for applications under extreme loads.

Uncovering flow dynamic behaviors underlying oil–gas–water three phase flow using multivariate synchrosqueezing transform

Abstract Oil-gas-water three-phase flow is distinguished by its intricate flow pattern. The analysis of experimental observations to reveal the oil-gas-water three phase flow’s dynamic behavior remains a challenging task. In this paper, firstly, a simulation investigation to compare the multivariate pseudo-winger distribution (MPWD) and multivariate synchrosqueezing transform (MSST) is presented. The cross term may be suppressed while maintaining high time-frequency concentration, according to our research on multivariate synchrosqueezing transform. The time-frequency analysis of various vertical oil–gas–water three phase flow patterns is then conducted utilizing MSST. The results from this study reveal that in various frequency bands, slug flow, bubble flow, and churn flow exhibit considerable temporal frequency variances. The MSST can effectively uncover the intrinsic connection between signal fluctuations and flow structure, and promote the understanding of various patterns of flow.

Numerical investigation of departure from nucleate boiling for subcooled flow boiling in a rolling tube

A crucial issue in floating nuclear plants is the critical heat flux (CHF) under oscillatory conditions by ocean waves. CHF must be predicted considering the dynamic motion of the floating platform. We present numerical simulations for predicting the DNB (departure-from-nucleate-boiling)-type CHF of subcooled R134a flow in a rolling tube and provide physical insight into the dynamic flow behavior regarding the DNB, considering rolling amplitudes of 15°, 30°, and 40° and a rolling period of 6 s. The local wall temperature and void fraction depended mainly on the dynamic changes in the gravitational direction. Gravity had the effect of lowering DNB. The Euler force created a secondary liquid flow that somewhat reduced the liquid temperature at the heating wall during the rolling period, thereby enhancing the convective and quenching heat fluxes. Furthermore, the secondary flow rapidly increased the liquid area fraction at the overheated wall and decreased the temporarily increased wall temperature before DNB occurred. The secondary flow had the effect of increasing DNB. In this study, the effect of increasing the DNB value is greater than the effect of reducing the DNB value. The occurrence of DNB depended on sufficient cooling of the downward-facing wall during the rolling period.

Slug flow parameters applied to dynamic riser analysis

Slug flow is a typical pattern in the petroleum production system. Its main feature is intermittency. This feature influences the riser dynamic behavior due to the coupling of its geometric configuration and time and space variation of internal fluid parameters. Although the studies that illustrate slug flow intermittency and dynamic riser behaviors have been growing recently, this discussion still needs to be expanded. Most slug models proposed in the literature are simplified and do not adequately capture the intermittency to a proper riser dynamic influence determination. Previous models neglected at least one of the following slug flow mechanisms: random features, bubble shape, liquid slug aeration, wake, and pipeline inclination effects. This paper aims to cover all these phenomena and present a model that provides realistic slug flow data for any riser analysis program. The dynamic riser analysis is not the scope of this paper. This article used a new slug tracking model that captures the intermittency to study the influence of the internal flow over the dynamic riser behavior. The model considers the liquid slug hold-up and its change over the pipe, the oil and gas thermodynamic properties, and the geometrical transition from typically horizontal plane interfaces to vertical concentric bubble shapes. The slug tracking model tests use three gas and liquid ratios (0.55, 0.7, and 0.82) with a 2.05 m/s constant mixture velocity. The maximum relative pressure deviation was 24.1. Time and space analysis of the slug flow parameters demonstrates the slug tracking capacity to reproduce the intermittency of the bubble and liquid slug lengths, void fractions, velocities, and pressures. This intermittence is essential to describe the internal slug flow influence on the structure behavior.

Investigation of electrical, electromagnetic interference shielding and tensile properties of 3D-printed acrylonitrile butadiene styrene/carbon nanotube composites

Conductive polymer composites (CPCs) are manufactured by compounding conductive particles with a polymer matrix. There are many applications where they are used, including light-weight applications requiring both electrical and heat conductivity. Carbon nanotubes (CNTs) have been accepted as common materials to accomplish this goal. As part of this research, a material extrusion additive manufacturing (AM) process was utilized to create nanocomposites by a 3D-printing technique. Multi-wall carbon nanotubes (MWCNTs) were mixed in 1, 3 and 6 wt fractions with acrylonitrile butadiene styrene (ABS) and extruded in filament form. Three-dimensionally printed specimens were used to evaluate electrical, electromagnetic interference shielding effectiveness (EMI SE) and tensile properties. The electrical conductivity of the material was 26 times greater than that of ABS. In the X-band of electromagnetic waves, EMI SE's reflected and absorbed portions increased respectively 4 and 16 times. The tensile strength and modulus were enhanced by 15% and 9%, respectively. On composite specimens, microwave heat treatment was applied. There is less void space between the rasters and layers, which helps improve tensile properties. Additionally, 3D-printed specimens were tested for melt flow rate (MFR) and dynamic mechanical behaviour. The nozzle has experienced some wear due to the intrinsic abrasive nature of CNTs.

Etude numérique de l’effet des générateurs de vortex longitudinaux sur le transfert thermique d’un écoulement laminaire traversant un micro-canal

We conducted, in this work, a three-dimensional numerical study of forced convection heat transfer of a laminar flow of water passing through micro-channels with and without longitudinal vortex generator, using the CFD code “Ansys- Fluent”. This work aimed to elucidate the effect of vortex generators on the dynamic and thermal behavior of the micro-fluidic flow. The results obtained show that the increase in the Reynolds number leads to an improvement in the quality of heat transfer in both cases of the study. Rectangular micro-channel with LVG can improve heat transfer compared to smooth rectangular micro-channel while consuming more pressure drop. In the range of Reynolds numbers between 200 and 1200, a 2-21% increase in the mean Nusselt number was observed for micro-channels with LVG compared to smooth micro-channels. This occurs through better mixing of the fluid, reduction in the thickness of the thermal boundary layer, and increased heat transfer area. In addition, the friction factor has been increased by more than 50% compared to smooth micro-channels, due to the local resistance of LVGs and the presence of secondary flows.