Fluid Mechanics Experiments Research Articles

Practical object and flow structure segmentation using artificial intelligence

This paper explores integrating artificial intelligence (AI) segmentation models, particularly the Segment Anything Model (SAM), into fluid mechanics experiments. SAM’s architecture, comprising an image encoder, prompt encoder, and mask decoder, is investigated for its application in detecting and segmenting objects and flow structures. Additionally, we explore the integration of natural language prompts, such as BERT, to enhance SAM’s performance in segmenting specific objects. Through case studies, we found that SAM is robust in object detection in fluid experiments. However, segmentations related to flow properties, such as scalar turbulence and bubbly flows, require fine-tuning. To facilitate the application, we have established a repository (https://github.com/AliRKhojasteh/Flow_segmentation) where models and usage examples can be accessed.

Physics-informed deep-learning applications to experimental fluid mechanics

High-resolution reconstruction of flow-field data from low-resolution and noisy measurements is of interest due to the prevalence of such problems in experimental fluid mechanics, where the measurement data are in general sparse, incomplete and noisy. Deep-learning approaches have been shown suitable for such super-resolution tasks. However, a high number of high-resolution examples is needed, which may not be available for many cases. Moreover, the obtained predictions may lack in complying with the physical principles, e.g. mass and momentum conservation. Physics-informed deep learning provides frameworks for integrating data and physical laws for learning. In this study, we apply physics-informed neural networks (PINNs) for super-resolution of flow-field data both in time and space from a limited set of noisy measurements without having any high-resolution reference data. Our objective is to obtain a continuous solution of the problem, providing a physically-consistent prediction at any point in the solution domain. We demonstrate the applicability of PINNs for the super-resolution of flow-field data in time and space through three canonical cases: Burgers’ equation, two-dimensional vortex shedding behind a circular cylinder and the minimal turbulent channel flow. The robustness of the models is also investigated by adding synthetic Gaussian noise. Furthermore, we show the capabilities of PINNs to improve the resolution and reduce the noise in a real experimental dataset consisting of hot-wire-anemometry measurements. Our results show the adequate capabilities of PINNs in the context of data augmentation for experiments in fluid mechanics.

A controllable sliding law for thin-film flows over slippery fluid-saturated substrates: theory and experiments

We develop a theoretical and experimental framework for generating slip underneath thin-film flows of viscous fluids in the laboratory, with the ability to control slip as desired. Such a framework is useful for large-scale fluid-mechanical experiments in which basal sliding is important. In particular, we consider the flow of a thin film of viscous fluid spreading over a structured, slippery substrate, involving a sequence of two-dimensional cavities that are prewetted with a fluid of smaller viscosity. By averaging over small-scale inhomogeneities, we demonstrate that such a substrate gives rise to a macroscopic linear sliding law, or Navier slip condition, that is effectively homogeneous on the large scale. The slip length, determining the slipperiness of the substrate, is proportional to the viscosity ratio and width of each cavity. As such, the slipperiness of the substrate can be controlled by altering the viscosity ratio, as desired. Two asymptotic regimes arise, describing flow over very slippery substrates and flow over no-slip substrates. The former regime is valid for early times, when the depth of the overlying fluid is much less than the slip length, and the latter is valid for late times, when the depth is much greater than the slip length. Solutions to the full model approach similarity solutions describing the two regimes for early and late times. We confirm our theoretical predictions by conducting a series of analogue laboratory experiments.

Robustness and stability of doubly periodic patterns of the focusing nonlinear Schrödinger equation.

The nonlinear Schrödinger equation possesses doubly periodic solutions expressible in terms of the Jacobi elliptic functions. Such solutions can be realized through doubly periodic patterns observed in experiments in fluid mechanics and optics. Stability and robustness of these doubly periodic wave profiles in the focusing regime are studied computationally by using two approaches. First, linear stability is considered by Floquet theory. Growth will occur if the eigenvalues of the monodromy matrix are of a modulus larger than unity. This is verified by numerical simulations with input patterns of different periods. Initial patterns associated with larger eigenvalues will disintegrate faster due to instability. Second, formation of these doubly periodic patterns from a tranquil background is scrutinized. Doubly periodic profiles are generated by perturbing a continuous wave with one Fourier mode, with or without the additional presence of random noise. Effects of varying phase difference, perturbation amplitude, and randomness are studied. Varying the phase angle has a dramatic influence. Periodic patterns will only emerge if the perturbation amplitude is not too weak. The growth of higher-order harmonics, as well as the formation of breathers and repeating patterns, serve as a manifestation of the classical problem of Fermi-Pasta-Ulam-Tsingou recurrence.

The transformative potential of machine learning for experiments in fluid mechanics

The transformative potential of machine learning for experiments in fluid mechanics

IMAGE PROCESSING OF LOW-CONTRAST NOISY GRAYSCALE IMAGES IN FLUID MECHANICS EXPERIMENTS

IMAGE PROCESSING OF LOW-CONTRAST NOISY GRAYSCALE IMAGES IN FLUID MECHANICS EXPERIMENTS

Commissioning of an MRI test facility for CFD-grade flow experiments in replicas of nuclear fuel assemblies and other reactor components

A new test facility is presented, enabling single-phase isothermal fluid mechanics experiments to support nuclear reactor safety studies. The applied measurement technique Magnetic Resonance Velocimetry (MRV) is based on the medical imaging modality Magnetic Resonance Imaging (MRI). The experiments in this laboratory aim at providing the mean velocity field in models of nuclear fuel assemblies using isothermal water. Additionally, turbulence statistics in terms of the Reynolds stress tensor can be acquired. Since MRV is a non-optical measurement technique, there are no requirements on optical properties, and the investigated models can be quickly built with additive manufacturing techniques. Even though MRV offers a relatively low temporal and spatial resolution compared to laser-optical techniques and has some limitations regarding the test section materials, e.g., no metals, the main gain is that MRV provides a full three-dimensional representation of the flow field in short times. Both the model preparation and the measurement time are significantly faster than with laser-optical measurement techniques. The full-field experimental data obtained with MRV is often referred to as 'CFD grade data' as it enables a full-field validation of numerical solutions obtained with Computational Fluid Dynamics (CFD). As a commissioning experiment, measurements inside a 5x5 fuel element model of a pressurized water reactor (PWR) are conducted and compared with CFD results. The measured mean velocity field inside the PWR replica comprised of 2.9 Million velocity vectors was acquired in 79 min. In addition, turbulence measurements in a periodic hill channel demonstrate how MRV can be used to obtain all components of the Reynolds stress tensor. Although the measurements in this proof-of-concept study suffer from some measurement errors, all major error contributions are well known, and validated methods exist to overcome these shortcomings. In conclusion, this study shows how the velocity field in a specific fuel assembly configuration can be obtained highly efficiently using a specialized MRI laboratory.

Large-Scale, Multidisciplinary Laboratory Teaching of Fluid Mechanics

The nature of fluid mechanics makes experimentation an important part of a course taught on the subject. Presented here is the application of a novel, large-scale multidisciplinary model of practical education in a fluids engineering laboratory. The advantages of this approach include efficiencies through the economy of scale leading to better pedagogy for students. The scale justifies dedicated academic resources to focus on developing laboratory classes and giving specific attention to designing activities that meet learning outcomes. Four examples of applying this approach to fluid mechanics experiments are discussed, illustrating tactics that have been developed and honed through many repeated instances of delivery. “The measurement lab” uses a flow measurement context to teach identifying and managing general experimental uncertainty. In this lab, new students, unfamiliar with fluid mechanics, are guided through a process to gain understanding that can be applied to all future experimental activities. The “pressure loss in pipes” lab discusses the advantage of and process for sharing equipment and teaching resources between multiple cohorts. Here, the provision for students is adapted for context, such as the degree program or year of study. The “weirs big and small” lab provides a methodology for teaching the power of dimensional analysis to mechanical engineers using a field of fluid mechanics that is outside their usual theoretical studies. Finally, the “spillway design” lab discusses mechanisms for delivering independent, open-ended student experiments at scale, without excessive staff resource requirements.

A novel test rig development for experimental study of fluid mechanics of hydraulic orifice and/or gaps under high or low temperature conditions

To meet the demand of ever operating in extreme environments of modern high-end vehicles or machineries, it is meaningful to perform basic research on fluid mechanics of hydraulic orifices and/or gaps under high or low temperature conditions. In this study, a novel compact experimental module which integrates the oil, the oil supply mechanism, the testing damping valve with orifices and/or gaps and several sensors was proposed, the experimental module has avoided the bulky method of using a full hydraulic system for oil supply and can be wholly placed into a high/low temperature chamber to undertake experiments; Based on the integrated experimental module, a test rig employing a variable-frequency-motor-driven ball screw driver to excite the testing module was developed. By using the virtual instrument technology, the hardware and software of automatic measurement and control system was further developed, so the automatic control and data collection of the test rig was realized. Final results show that the test rig can conveniently perform fluid mechanics experiments for various hydraulic orifices and/or gaps under high or low temperature conditions with high efficiency, so the test rig can be a very good platform for pertinent basic research works.

General route to design polymer molecular weight distributions through flow chemistry

The properties of a polymer are known to be intrinsically related to its molecular weight distribution (MWD); however, previous methodologies of MWD control do not use a design and result in arbitrary shaped MWDs. Here we report a precise design to synthesis protocol for producing a targeted MWD design with a simple to use, and chemistry agnostic computer-controlled tubular flow reactor. To support the development of this protocol, we constructed general reactor design rules by combining fluid mechanical principles, polymerization kinetics, and experiments. The ring opening polymerization of lactide, the anionic polymerization of styrene, and the ring opening metathesis polymerization are used as model polymerizations to develop the reactor design rules and synthesize MWD profiles. The derivation of a mathematical model enables the quantitative prediction of the experimental results, and this model provides a tool to explore the limits of any MWD design protocol.

Novel Environment Enables PIV Measurements of Turbulent Flow around and within Complex Topographies

AbstractA novel flow facility developed to enable a broad spectrum of fluid mechanics experiments involving complex geometries and employing high-fidelity optical diagnostics is described in this p...

Computational Model of Unsteady Hydromechanics of Large Amplitude Gerstner Waves

The computational experiments in the ship fluid mechanics involve the nonstationary interaction of a ship hull with wave surfaces that include the formation of vortices, surfaces of jet discontinuities, and discontinuities in the fluid under the influence of negative pressure. These physical phenomena occur not only near the ship hull, but also at a distance where the waves break as a result of the interference of the sea waves with waves reflected from the hull. In the study reported here we simulate the wave breaking and reflection near the ship hull. The problem reduces to determining the wave kinematics on the moving boundary of a ship hull and the free boundary of the computational domain. We build a grid of large particles having the form of a parallelepiped and, in the wave equation instead of the velocity field we integrate streams of fluid represented by functions as smooth as the wave surface elevation field. We assume that within the boundaries of the computational domain the waves do not disperse, i.e. their length and period stay the same. Under this assumption, we simulate trochoidal Gerstner waves of a particular period. This approach allows to simulate the wave breaking and reflection near the ship hull. The goal of the research is to develop a new method of taking the wave reflection into account in the ship motion simulations as an alternative to the classic method which uses added masses.

Adaptation to life in aeolian sand: how the sandfish lizard, Scincus scincus, prevents sand particles from entering its lungs.

ABSTRACTThe sandfish lizard, Scincus scincus (Squamata: Scincidae), spends nearly its whole life in aeolian sand and only comes to the surface for foraging, defecating and mating. It is not yet understood how the animal can respire without sand particles entering its respiratory organs when buried under thick layers of sand. In this work, we integrated biological studies, computational calculations and physical experiments to understand this phenomenon. We present a 3D model of the upper respiratory system based on a detailed histological analysis. A 3D-printed version of this model was used in combination with characteristic ventilation patterns for computational calculations and fluid mechanics experiments. By calculating the velocity field, we identified a sharp decrease in velocity in the anterior part of the nasal cavity where mucus and cilia are present. The experiments with the 3D-printed model validate the calculations: particles, if present, were found only in the same area as suggested by the calculations. We postulate that the sandfish has an aerodynamic filtering system; more specifically, that the characteristic morphology of the respiratory channel coupled with specific ventilation patterns prevent particles from entering the lungs.

Characterization of zebrafish larvae suction feeding flow using μPIV and optical coherence tomography

The hydrodynamics of suction feeding is critical for the survival of fish larvae; failure to capture food during the onset of autonomous feeding can rapidly lead to starvation and mortality. Fluid mechanics experiments that investigate the suction feeding of suspended particles are limited to adult fishes, which operate at large Reynolds numbers. This manuscript presents the first literature results in which the external velocity fields generated during suction feeding of early zebrafish larvae (2500–20,000 μm total length) are reported using time-resolved microscopic particle image velocimetry. For the larval stages studied, the maximum peak suction velocity of the inflow bolus is measured at a finite distance from the mouth tip and ranges from 1 to 8 mm/s. The average pressure gradient and the velocity profile proximal to the buccal (mouth) cavity are calculated, and two distinct trends are identified. External recirculation regions and reverse flow feeding cycles are also observed and quantified. One of the unresolved questions in fish suction feeding is the shape and dynamics of the buccal cavity during suction feeding; optical coherence tomography imaging is found to be useful for reconstructing the mouth kinematics. The projected area of the mouth cavity during the feeding cycle varies up to 160 and 22 % for the transverse and mid-sagittal planes, respectively. These findings can inspire novel hydrodynamically efficient biomedical and microfluidic devices.

Coordinate Systems, Numerical Objects and Algorithmic Operations of Computational Experiment in Fluid Mechanics

The paper deals with the computer implementation of direct computational experiments in fluid mechanics, constructed on the basis of the approach developed by the authors. The proposed approach allows the use of explicit numerical scheme, which is an important condition for increasing the effciency of the algorithms developed by numerical procedures with natural parallelism. The paper examines the main objects and operations that let you manage computational experiments and monitor the status of the computation process. Special attention is given to a) realization of tensor representations of numerical schemes for direct simulation; b) realization of representation of large particles of a continuous medium motion in two coordinate systems (global and mobile); c) computing operations in the projections of coordinate systems, direct and inverse transformation in these systems. Particular attention is paid to the use of hardware and software of modern computer systems.

No-Go Theorems Face Background-Based Theories for Quantum Mechanics

Recent experiments have shown that certain fluid-mechanical systems, namely oil droplets bouncing on oil films, can mimic a wide range of quantum phenomena, including double-slit interference, quantization of angular momentum and Zeeman splitting. Here I investigate what can be learned from these systems concerning no-go theorems as those of Bell and Kochen-Specker. In particular, a model for the Bell experiment is proposed that includes variables describing a 'background' field or medium. This field mimics the surface wave that accompanies the droplets in the fluid-mechanical experiments. It appears that quite generally such a model can violate the Bell inequality and reproduce the quantum statistics, even if it is based on local dynamics only. The reason is that measurement independence is not valid in such models. This opens the door for local 'background-based' theories, describing the interaction of particles and analyzers with a background field, to complete quantum mechanics. Experiments to test these ideas are also proposed.

Episodic Earth evolution

Abstract U–Pb ages of zircons from Precambrian granitoids and major rivers are grouped into a series of major peaks at about 2.7, 2.5, 2.1, 1.9 and 1.1 Ga. Recently these peaks have been interpreted as times of enhanced preservation of the continental crust associated with the assembly of supercontinents. An older interpretation, which we support, is that they correspond instead to periods of accelerated crustal growth related to episodic convection of the mantle. In this paper we use fluid mechanics experiments to develop a new model of mantle convection and crustal growth. A dense layer at the base of the mantle persists until 2.7 Ga when it destabilizes and generates large domes that rise into the upper mantle. There they cause a large increase in the rate of subduction which leads to enhanced granite magmatism at convergent margins and thus to a pulse of crustal growth. The domes heat the upper mantle which partially melts at mid-ocean ridges to produce thick oceanic crust that resists subduction. The subsequent period of subdued plate motion is broken by the next generation of mantle domes. Before and after the Archean–Proterozoic period of episodic crustal growth, plate tectonics operated quasi-continuously.

Dynamic measurement of microfilms and nanofilms of fluids using fluorescence microscopy

The noninvasive measurement of submicron liquid film thicknesses has been one of the greatest challenges in fluid mechanics experiments. Most applications require the time-dependent measurement of thin film thickness as it varies from macroscopic to microscopic dimensions. The current techniques lack resolution and range to capture the complete characteristics of these phenomena. Here, we use fluorescence microscopy as an optical technique to investigate the thickness of micro- and nanofilms as they evolve with time. We show that this technique is capable of measuring film thicknesses on the order of nanometers to millimeters with very high spatial resolution. Experimental examples are provided to demonstrate the various potential applications of this technique.

Vortices and vortex sources of multiple vortex interaction systems

In this paper, a multiple vortex interaction model (MVIM) is proposed to describe the possible stream–vorticity interaction (SVI) and the vorticity–vorticity interaction (VVI) among vortices. The symmetries and conservation laws of the MVIM show that the SVI preserves the momenta, the vortex momenta and the energies of every vortex, and the interaction energies of every two vortices. However, the VVI destroys the energy conservation property for every vortex. Some special types of exact vortices and vortex source solutions including multiple point vortices, vortex dipoles, vortex multi-poles, fractal cyclons, fractal cyclon dipoles and Bessel vortices (BV) are presented. A special theoretical solution, the first BV is just the usual modon solution which can be used to describe the so-called atmospheric blocking. The second BV is supported by an atmospheric observation, saddle field, occurred over the North Pacific on 26 March, 2009. The characteristic features of vortex interactions are discussed under the MVIM without the β-effect via numerical simulations. Several interaction patterns such as merging, separation, mutual orbiting and absorption are reported. Those interaction behaviors are well consistent with some known fluid mechanical experiments and meteorologic observations.

Stick-slip control of the Carbopol microgels on polymethyl methacrylate transparent smooth walls

Microgel suspensions such as Carbopol are commonly used as viscoplastic fluid models in fluid mechanics experiments. Polymethyl methacrylate (PMMA) walls are also frequently used in these experiments since they enable flow visualizations via their transparency properties. When sheared with smooth surfaces, the microgels slip on the surfaces and an apparent motion of the fluid occurs below the bulk yield stress τy. This motion stops at a sliding yield stress τs. The occurrence of this apparent slip below τy can significantly modify the flow features. This article presents a PMMA treatment to inhibit the slip of the Carbopol microgels. The method is simple to perform, independent of the treatment conditions and stable over time.