The work reports an investigation of slug initiation in horizontal gas-liquid pipe flow induced by controlled interfacial disturbances. The idea is to investigate the limiting conditions for slug onset. To this end, disturbances with varying amplitudes were introduced at the fluid interface. An oscillating paddle was employed to generate controlled interfacial waves. These perturbations were introduced in the flow as short wavepackets of large spectral bandwidth. Thereby, disturbances in nearly all relevant wavenumbers are introduced in the flow hence providing seeds for instability mechanisms. The driving signal of the oscillating paddle was controlled and synchronized with acquisitions, enabling phase-locked measurements. The experiments were conducted at conditions close to the transition from stratified to slug/plug flow regimes. According to current findings, the disturbance amplitudes required to initiate slug increase for constant superficial gas velocity and decreasing liquid velocities. The variation of superficial liquid velocities required to initiate the slugs with respect to the disturbance amplitude displays two distinct behaviours. For small amplitudes (Amplitude/Diameter < 2%) there is a steep variation of critical liquid superficial velocity, USL. The flow becomes progressively less sensitive to disturbances as USL decreases. The slugs could not be initiated for USL values lower than approximately 20-30% of those required to initiate unforced slugs. A weakly nonlinear model based on the Landau equation was employed to predict the flow behaviour for disturbance amplitudes lower than 2% of the pipe diameter. Corresponding changes on the map of flow regimes due to weakly nonlinear effects are presented and discussed. The results suggest that weakly nonlinear corrections based on the Stuart-Landau equation are more relevant for the transition from stratified to plug flow regimes.

The present work reports an experimental characterization of linear and weakly nonlinear interfacial waves in a stratified air–water horizontal pipe flow. An oscillating paddle was employed to generate controlled waves at the liquid interface. The driving signal of the oscillating paddle was controlled and synchronized with image acquisitions, enabling phase-locked measurements and the application of ensemble averaging techniques. Velocity field measurements in the liquid and gas phases were performed simultaneously using an off-axis particle image velocimetry setup and shadowgraphy. The combined techniques allowed us to extract the coherent part of flow fluctuations related to the excited waves. This was done for a range of flow rates and wave frequencies. The selected conditions are close to the transition from stratified to slug/plug flow regimes. In the presence of linear waves, the coherent disturbances in both phases were weakly dependent on near-wall disturbances. Flow changes in the presence of weakly nonlinear waves were also investigated. In these cases, noticeable modifications in the mean flow and in turbulence distribution were observed near the interface, whereas close to the wall, the flow was weakly affected. This investigation follows the work of Farias et al. [“Characterization of interfacial waves in stratified turbulent gas-liquid pipe flow using Particle Image Velocimetry and controlled disturbances,” Int. J. Multiphhase Flow 161, 104381 (2023)], where the threshold for linear and weakly nonlinear waves was studied. Here, a clear comparison between wave-induced disturbances in linear and weakly nonlinear regimes is reported in the literature for the first time for stratified turbulent gas–liquid pipe flows. The methodology proposed is relatively simple and can contribute to describe wave-related phenomena in stratified pipe flows.

The transient flow regime related with the formation of laminar separation bubbles (LSB) is examined by time-resolved measurements of the velocity field. The investigated scenario aims at studying the flow in multiple stage turbines, where the wake of airfoils in previous compression stages induces periodic variation of turbulence level in the subsequent airfoils. This can lead to periodic removal and formation of LSB. This process is simulated here by exciting controlled disturbances with a vibrating ribbon. Experiments are carried out in a laminar water channel at PUC-Rio. Longitudinal PIV measurements are performed on a flat plate subjected to an adverse pressure gradient. The pressure gradient is set by false walls with adjustable geometry. Suction is applied on the false wall, in order to avoid the boundary layer separation at this surface. Bubble topology and disturbance growth along the streamwise direction are measured during the transient of bubble formation. Results show spatial amplification over a narrow frequency bandwidth. Dominant non-dimensional frequencies (St = fδ*s/U) at late stages of bubble formation are in close agreement with those reported in literature. During the transient the intensity of reverse flow is significantly changed, pointing out for possible changes on the stability mechanisms involved on the bubble reattachment.

The work reports an experimental investigation on stratified gas-liquid pipe flow characteristics in the presence of controlled interfacial waves. Studies of this flow regime with controlled interfacial waves are scarce in the literature. Here, the disturbances are excited at the liquid interface by an oscillating paddle. The waves are synchronized with image acquisitions, enabling the utilization of phase-locked measurements and ensemble averaging techniques. Off-axis Particle Image Velocimetry (PIV) and Shadowgraph techniques were applied to provide information about mean and wave-induced modifications on the velocity fields. Results show that mean flow velocities in the liquid and gas phases close to the pipe walls adhere well to the single-phase flow log-law profile. In the liquid layer, this agreement was observed up to half of the water depth. Controlled disturbances enabled the estimation of the wave amplitude thresholds for the appearance of relevant nonlinear wave effects on the flow field. Results suggest that such a threshold can be fairly represented by a constant value of the non-dimensional parameter proposed in the work of Kirby (2008). Within the non-linear wave regimes investigated, noticeable changes in the flow field were observed close to the interface. However, near the wall the flow was weakly affected by the presence of waves, suggesting that interfacial wave effects are weakly coupled with near-wall disturbances and might be modelled independently. Moreover, contributions to interfacial shear stress due to the presence of waves were obtained experimentally. The results presented here are useful for validation and improvement of models used to predict flow characteristics in stratified flows. In addition, they contribute to shed further light on the physical mechanisms involved in the phenomenon.

For several industrial applications involving two-phase gas-liquid flows, the one-dimensional Two-Fluid Model equations are typically solved for flow simulation. To ensure reliable predictions, it is very important to assess the stability properties and grid dependency of one-dimensional formulations, which are known to be largely affected by the closure relations and numerical schemes employed. In this work, a stability analysis of the transient one-dimensional Two-Fluid Model was performed for vertical annular flows. A viscous approach of differential and discretized formulations was analysed. The influence of the momentum flux parameter, inter-facial pressure jump due to surface tension and a dynamic pressure model were investigated. The analytical results were compared to those obtained by numerical solution of the model equations with the Finite Volume Method, for various experimental configurations taken from the literature. Results showed that closure models largely affect the wave frequencies and the growth rates captured by the model. The surface tension term introduced a cut-off frequency in the differential formulation rendering the model well-posed and effectively stabilizing short waves. The introduction of the dynamic pressure term considered here did not affect the cut-off values. However, the wave growth rates decreased compared to the case without this term. The liquid mo-mentum flux parameter greater than 1 presented a stronger influence on the evolution of interfacial waves in a broadband of frequencies than the other closures and leads to more regular waves and a more uniform flow field but much higher values may excessively damp the solution, which blocks the natural formation of larger waves in vertical annular flows. The numerical solution of the model equations showed very good agreement with the discrete stability analysis performed and was able to capture the wave frequencies and associated (linear) growth rates in the wave formation region. Concerning the momentum flux parameter, the bandwidth of the most damped disturbances in the linear and nonlinear region was similar. This suggests that the frequency response in this case was not very different from linear to nonlinear wave regimes. The methodology proved to be an important tool for further development of closure models and numerical schemes applied to vertical annular flows.

Computational fluid dynamics (CFD) is often applied to the study of combustion, enabling to optimize the process and control the emission of pollutants.This numerical methodology enables the analysis of different flame properties, such as the components of velocity, temperature, and mass fractions of chemical species.However, reproducing the behavior observed in engineering problems requires a high computational cost associated with memory and simulation time.Reduced order model (ROM) is a machine learning technique that has been applied to several engineering applications, aiming to develop models for complex systems with reduced computational cost.In this way, a high-fidelity model of complex systems is created from available data to learn its behavior and its main characteristics.In this work, different ROMs are created using CFD simulation data.The CFD model solves the mass, species, energy, and momentum conservation equations for a methane/air laminar diffusion flame, stabilized on the Gülder burner.Chemistry is modeled using a 19-species skeletal chemical kinetic mechanism.The static reduced order model uses the singular value decomposition (SVD) algorithm to decompose the CFD data and obtain the system's modes.Then, genetic aggregation response surface interpolation is applied on the higher SVD modes, creating the static ROM.This work analyzes the effect of different data preprocessing approaches on the ROM.The first analysis is the impact of reducing the number of learning data points, showing that this decrease does not directly impact the energy of the SVD modes, but, in the reconstruction field is possible to notice a degradation of the reconstruction.The second analysis is related to the effect of creating a ROM for each uncoupled flame property or treating the properties as a coupled system.The results of the coupled and uncoupled reduced order models are quite similar in terms of properties field reconstruction.However, in the energy analysis the coupled ROM converges rapidly, similarly to the uncoupled temperature ROM, while the uncoupled chemical species ROMs have a slower convergence.

The evolution of interfacial waves is investigated for a stratified laminar-laminar flow in a plane channel using numerical simulations based on the Volume of Fluid (VOF) method. Different nonlinear instability mechanisms that can promote saturation or rapid amplification of interfacial waves are investigated. Controlled disturbances are introduced at the interface between the two fluids, to assess five different scenarios of nonlinear interactions including harmonic excitation, subharmonic resonance, interaction between a short and a long wave, and two kinds of modulated wavepackets. Present simulations suggest that the same non-resonant type of wave interaction dominates the nonlinear stages of evolution in all the scenarios, for the fluid properties and parameters covered in this work. This mechanism involves the amplification of harmonics by quadratic nonlinearities and leads to the finite-amplitude saturation of interfacial waves, which results in a wavy flow. Reduced order models based on interaction of few wavelengths and described by the Stuart–Landau equation are proposed for the prediction of the nonlinear flow dynamics of laminar liquid-liquid two-phase flows.

A novel low cost sensor is proposed to measure simultaneously viscosity and density of Newtonian fluids. In the proposed configuration, the sensor is built with two piezoelectric buzzers disks. One disk is excited with a step signal while the other is used to read the sensor response. Damping factor and resonant frequencies are measured when the sensor is immersed in fluids of different densities and viscosities. Excitation signal and data acquisitions are phase locked, therefore ensemble averaging techniques are employed to improve signal to noise ratio. Binary mixtures of glycerol and ethanol are selected for assessment of sensor capabilities. Due to the high temperature dependency of fluid viscosities a parametric variation of the fluid properties could be achieved by varying the fluid temperature. Dynamic response of the sensor is calibrated against the fluid properties for viscosities up to 200 cP and densities up to 1.23 g/cm(3). Estimated uncertainties are +/- 8 cP and +/- 0.03gm/cm(3), with 95% of confidence, for viscosity and specific gravity, respectively. These uncertainties include sensor sensitivity to temperature. Results are promising and show a proof of concept. Thus, disposable and very low cost devices can be developed based on the proposed sensor. (C) 2019 Elsevier Ltd. All rights reserved.

The present work aims at demonstrating the applicability of using tubes made of the polymer PVDF as intrinsic flowmeters, considering the peculiar piezoelectric properties of the PVDF-beta polymorph and using as flow measurement method the Flow-Induced Vibration (FIV) technique. The work methodology consisted of initially characterizing the material of a commercial PVDF tube spectroscopically, using the Fourier Transform Infrared Spectroscopy (FTIR) technique, to confirm the presence of PVDF-beta. Then, the PVDF tube was installed in a hydraulic test circuit and electromechanically tested, to identify the levels of electrical voltages generated by the vibration caused by the passage of water through the tube. To compare the vibration signals, simultaneous measurements were made with commercial accelerometers and with a commercial PVDF film built specifically as a piezoelectric transducer. The results of the FIV tests allowed confirming that the standard deviation of the voltage signal measured by the PVDF tube is related to the flow rate. The uncertainties associated with flow rate measurement by the tube showed a considerable reduction in the higher flow rates. On the other hand, in the lowest flow rate levels, a high instability was observed, possibly due to the process of initial mechanical accommodation of the tube. Despite this, a strong relationship between the signal generated by the PVDF tube and the flow rate that induced this vibration has proven the potential applicability of the PVDF tube as a water flowmeter. (C) 2019 Elsevier Ltd. All rights reserved.

The influence of a single roughness element on the evolution of two-dimensional (2-D) Tollmien–Schlichting (TS) waves is investigated experimentally. Experiments are carried out in a region of zero pressure gradient of an airfoil section. Downstream from the disturbance source, TS waves interact with a cylindrical roughness element with a slowly oscillating height. The oscillation frequency of the roughness was approximately 1500 times lower than the wave frequency and approximately 250 times slower than the characteristic time of flow passing the region of transition development. Therefore, the roughness behaved as a quasi-steady disturbance. The set-up enabled us to perform hot-wire measurements phase locked to the waves and to the roughness movement. Experimental results show a scattering of the 2-D waves into oblique ones and a relatively weak distortion of the mean flow for roughness heights as large as 0.2 times the boundary layer displacement thickness ( $\unicode[STIX]{x1D6FF}^{\ast }$ ). Transfer functions for TS wave scattering at the roughness are obtained. Results show an unexpected coincidence in shape with acoustic receptivity functions found in Würz et al. (J. Fluid Mech., vol. 478, 2003, pp. 135–163) for the problem of excitation of TS waves by scattering of acoustic waves at surface roughness. In the present work, the ratio between the incoming 2-D wave amplitude to the amplitude of the scattered oblique waves scaled linearly with the roughness height only for very shallow roughness. For roughness elements higher than $0.08\unicode[STIX]{x1D6FF}^{\ast }$ and below $0.2\unicode[STIX]{x1D6FF}^{\ast }$ , the wave scattering exhibited a quadratic variation with respect to the roughness height. In addition, this feature did not vary significantly with respect to TS wave frequency. An analysis of the weakly nonlinear interactions triggered by the roughness element is also carried out, assisted by numerical solution of nonlinear parabolized stability equations, performed for a two-dimensional Blasius boundary layer. A comparison between experiments and simulations reveals that the weakly nonlinear interactions observed are not substantially affected by mean flow distortions that could be produced in the wake of the small and medium sized roughness elements ( $h<0.2\unicode[STIX]{x1D6FF}^{\ast }$ ). From a practical perspective, results suggest that scattering coefficients might be employed to include the effect of isolated and medium sized roughness elements in transition prediction tools developed for smooth surfaces.

The evolution of interfacial waves on a stratified flow was investigated experimentally for air water flow in a horizontal pipe. Waves were introduced in the liquid level of stratified flow near the pipe entrance using an oscillating plate. Mean height of liquid layer and fluctuations superimposed to this mean level were captured using high speed cameras. Digital image processing techniques were used to detect instantaneous interfaces along the pipe. The driving signal of the oscillating plate was controlled by a D/A board that was synchronized with the acquisitions. This enabled to perform phase locked acquisitions and to use ensemble average procedures. Thereby, it was possible to measure the temporal and the spatial evolution of the disturbances introduced in the flow. In addition, phase locked measurements of the velocity field in the liquid layer were performed using standard planar PIV. The velocity fields were extracted at a fixed stream wise location, whereas the measurements of the liquid level were performed at several locations along the pipe. The assessment of the set-up was important for validation of the methodology proposed in this work since it aimed at providing results for further comparisons with theoretical models and numerical simulations. Results show that linear waves were observed for liquid level oscillations lower than about 1.5% of the pipe diameter. Eigenfunctions in the liquid layer related to interfacial modes were measured experimentally for the first time. For moderate holdup levels, the eigenfunctions clearly show that interfacial modes are decoupled from inner modes which are related to wall turbulence.

The evolution of interfacial waves on a stratified flow was investigated experimentally for air-water flow in a horizontal pipe. Waves were introduced in the liquid level of stratified flow near the pipe entrance using an oscillating plate. The mean height of liquid layer and the fluctuations superimposed on this mean level were captured using high speed cameras. Digital image processing techniques were used to detect instantaneous interfaces along the pipe. The driving signal of the oscillating plate was controlled by a D/A board that was synchronized with acquisitions. This enabled to perform phase-locked acquisitions and to use ensemble average procedures. Thereby, it was possible to measure the temporal and spatial evolution of the disturbances introduced in the flow. In addition, phase-locked measurements of the velocity field in the liquid layer were performed using standard planar Particle Image Velocimetry (PIV). The velocity fields were extracted at a fixed streamwise location, whereas the measurements of the liquid level were performed at several locations along the pipe. The assessment of the setup was important for validation of the methodology proposed in this work, since it aimed at providing results for further comparisons with theoretical models and numerical simulations. Therefore, the work focuses on validation and characterization of interfacial waves within the linear regime. Results show that under controlled conditions, the wave development can be well captured and reproduced. In addition, linear waves were observed for liquid level oscillations lower than about 1.5% of the pipe diameter. It was not possible to accurately define an amplitude threshold for the appearance of nonlinear effects because it strongly depended on the wave frequency. According to the experimental findings, longer waves display characteristics similar to linear waves, while short ones exhibit a more complex evolution, even for low amplitudes.