Two-phase Flow Phenomena Research Articles

Spatio-temporal reconstruction of droplet impingement dynamics by means of color-coded glare points and deep learning

Abstract The present work introduces a deep learning approach for the three-dimensional reconstruction of the spatio-temporal dynamics of the gas-liquid interface on the basis of monocular images obtained via optical measurement techniques.
The method is tested an evaluated at the example of liquid droplets impacting on structured solid substrates.
The droplet dynamics are captured through high-speed imaging in an extended shadowgraphy setup with additional glare points from lateral light sources that encode further three-dimensional information of the gas-liquid interface in the images.
A neural network is trained for the physically correct reconstruction of the droplet dynamics on a labelled dataset generated by synthetic image rendering on the basis of gas-liquid interface shapes obtained from direct numerical simulation.
The employment of synthetic image rendering allows for the efficient generation of training data and circumvents the introduction of errors resulting from the inherent discrepancy of the droplet shapes between experiment and simulation. 
The accurate reconstruction of the three-dimensional shape of the gas-liquid interface during droplet impingement on the basis of images obtained in the experiment demonstrates the practicality of the presented approach.
The introduction of glare points from lateral light sources in the experiments is shown to improve the reconstruction accuracy, which indicates that the neural network learns to leverage the additional three-dimensional information encoded in the images for a more accurate depth estimation.
By the successful reconstruction of obscured areas in the input images, it is demonstrated that the neural network has the capability to learn a physically correct interpolation of missing data from the numerical simulation.
Furthermore, the physically reasonable reconstruction of unknown gas-liquid interface shapes for drop impact regimes that were not contained in the training dataset indicates that the neural network learned a versatile model of the involved two-phase flow phenomena during droplet impingement.

Modeling of high-pressure transient gas-liquid flow in M-shaped jumpers of subsea gas production systems

Two-phase flow with low liquid loads is common in high-pressure natural gas offshore gathering and transmission pipelines. During gas production slowdowns or shutdowns, an accumulation of liquid in the lower sections of subsea pipelines may occur. This phenomenon is observed in jumpers that connect different units in deep-water subsea gas production facilities. The displacement of the accumulated liquid during production ramp-up induces temporal variations in pressure drop across the jumper and forces on its elbows, resulting in flow-induced vibrations (FIV) that pose potential risks to the structural integrity of the jumper. To bridge the gap between laboratory experiments and field conditions, transient 3D numerical simulations were conducted using the OpenFOAM software. These simulations facilitated the development of a mechanistic model to elucidate the factors contributing to increased pressure and forces during the liquid purging process. The study examined the influence of gas pressure level, pipe diameter, initially accumulated liquid amount, liquid properties, and gas mass flow rate on the transient pressure drop and the forces acting on the jumper's elbows. The critical gas production rate required for complete liquid removal of the accumulated liquid was determined, and scaling rules were proposed to predict the effects of gas pressure and pipe diameter on this critical value. The dominant frequencies of pressure and force fluctuations were identified, with low-pressure systems exhibiting frequencies associated with two-phase flow phenomena and high-pressure systems showing frequencies attributed to acoustic waves.

Computational fluid dynamics simulation of cryogenic vertical upflow boiling under Earth gravity

This study presents numerical simulations and their validation for flow boiling of liquid nitrogen (LN2) in a vertical upflow orientation, with a primary aim to understand the complex two-phase flow and heat transfer phenomena important to space applications. The computational fluid dynamics (CFD) model utilized the coupled level set volume of fluid (CLSVOF) method, incorporating additional source terms for bubble collision dispersion force and shear lift force in the momentum conservation equation to enhance simulation accuracy. The simulations were conducted for two mass velocities (G = 526 and 804 kg/m2s) and three different heat flux levels (approximately 10%, 30%, and 70% of critical heat flux (CHF) under Earth gravity. The model was validated against measured wall temperature data acquired from the authors’ previous experimental studies, demonstrating average deviations of less than 2.8 K across all operating conditions. The simulated two-phase flow contours illustrated various flow patterns, including bubbly, slug, churn, and annular. Both mass velocity and heat flux were observed to impact the onset of nucleate boiling (ONB), bubble nucleation, growth, and coalescence, and overall vapor structure. The simulations also offered insight into axial and radial void fraction and velocity profiles, revealing local flow acceleration trends synchronized with void fraction development. A comparison between predicted and measured bulk fluid temperature profiles showed excellent agreement, further validating the CFD model’s accuracy and practical usefulness for two-phase cryogenic flow boiling simulations in space applications.

Numerical design of a flow restrictor for tanked liquid nitrogen undergoing reduced-gravity flights

NASA and the United States are designing multiple reduced-gravity cryogenic payloads to study various two-phase flow phenomena that will be ground-tested and subsequently flight-tested onboard parabolic flights. The setup will undergo 25+ parabolas per flight, during which liquid nitrogen will be cyclically or continuously demanded from a supply Dewar. The minimum gravity level during a parabolic maneuver is 0 ± 0.05 g, which can last around 20 seconds, and the maximum expected gravity is 2 g, which lasts for 40-50 seconds. This Computational Fluid Dynamics (CFD) study focuses on the design and optimization of a bi-directional perforated plate and a ring baffle to help ensure single-phase liquid nitrogen outflow during all flight phases, regardless of supply tank liquid level or gravity level. CFD analyses carried out using ANSYS FLUENT indicate that a double perforated plate with an open area percent equal or less than 0.63% can achieve high values of expulsion efficiency. Structural analyses performed using the ANSYS Static Structural solver determined the minimum plate thickness needed to withstand gravitational and hydrodynamic forces experienced during the flight.

Determination Of Two-Phase Flow Regimes Of A Condensing Water-Air Mixture In An Array Of Micro Channels

Proton exchange membrane (PEM) full cells (FCs) are a promising alternative to combustion engines for mobile applications. Their signature low operating temperature leads to the possible existence of product water in its liquid form. Liquid water formation and transport are important phenomena to understand and model in order to facilitate the development of new, more efficient PEMFCs. This contribution investigates the two-phase flow phenomena separately by introducing a gaseous air-vapor mixture into an optically accessible fuel cell flow plate with 14 parallel micro channels (confinement number Co = 7.3) and a carbon paper inlay representing the FC's gas diffusion layer and bringing it to the point of onset of condensation. Special care is taken to provide a controlled and reproducible air-vapor mixture with low pressure pulsation. One key feature of the experiment is that the flow plate used is designed to be as similar to a regular research fuel cell flow plate as possible, retaining its channel surface properties and electrical parameters. It can also be heated to a given temperature to mimic the fuel cell operating conditions. Therefore, the same plate can also be used in in-situ experiments in an operated fuel cell. An air mass flow rate of 500 - 1000 mLn / min with relative humidities of 75% - 95% at 65°C and 75°C was investigated. The emerging flow patterns were captured using a high-speed camera at 250 Hz and categorized using canonical two-phase flow regime nomenclature. It is shown that the present setup can produce the relevant two-phase flow patterns that can be found in operated fuel cells as reported by previous studies and is furthermore equipped to investigate flow states not easily reached with an operated fuel cell (e.g. flooded channels).

Simulation of the key phenomena in nuclear reactor two-phase flow and severe accident with particle method

Simulation of the key phenomena in nuclear reactor two-phase flow and severe accident with particle method

Development of physically based modelling of bubble behaviour for subcooled boiling applications

A methodology for the modelling of phase change phenomena in two-phase flow is presented, based on interface capturing simulation and the mechanistic modelling of interfacial heat and mass transfer. The volume of fluid approach is adopted, utilising a compressive scheme to maintain interface sharpness, allowing a continuum representation of mass and heat source terms to be used. The methodology is verified against analytical solutions for both evaporation and condensation phase change problems, demonstrating excellent agreement even in the case of a large phase density ratio and high rate of evaporation.

Validation of Universal Cryogenic Flow Boiling Correlations in Thermal Desktop for Liquid Helium

Understanding two-phase cryogenic propellant behavior is key to enabling technologies for future spaceflight missions. Developing accurate models of two-phase flow phenomena, particularly in the current work, flow boiling in the heating configuration, is relevant to the propellant transfer process both in 1-g and microgravity. Currently there is a need for more accurate, direct cryogenic data anchored models for various boiling phenomena. Recently, universal correlations for cryogens flowing in heated tubes have been developed for a wide variety of fluids, thermodynamic conditions, and various regimes across the boiling curve, and have been patched to provide a smooth, continuous predictive curve. This paper describes implementation and validation of these correlations into Thermal Desktop to improve predictive performance, with a focus on liquid helium. Results from Thermal Desktop using both the built-in and new correlations are validated against a historical dataset of flow boiling experiments in the heating configuration using liquid helium. Based on results, the new correlations show a substantial improvement over the original built-in flow boiling correlations in Thermal Desktop in predicting the wall temperature as a function of preponderant parameters for this quantum fluid at temperatures greater than the lambda temperature, T λ = 2.17K.

Effects of pressure ratio allocation on the flow characteristics in a five-stage ionic liquid compressor

Ionic liquid compressors are a new type of hydrogen compressor technology with significant advantages when applied to high-pressure hydrogen refueling stations, where multi-stage compression is an ideal solution. The utilization of a liquid piston introduces complex two-phase flow phenomena within the cylinders, and understanding the characteristics of two-phase flow is crucial for the thermodynamic performance of the compressor. For the multi-stage compressors, pressure ratio allocation has a significant influence on the structural design and energy conversion efficiency of the compressor, however, there is a lack of relevant research in the existing studies. In this paper, the compression cylinder structures of a five-stage ionic liquid compressor under three pressure ratio allocation strategies are designed, and then the two-phase flow characteristics inside the cylinder are numerically investigated. Based on the analysis of the two-phase flow pattern in the cylinder, the minimum initial liquid heights in each stage that ensure the solid piston is not exposed are determined, and the escape characteristics of ionic liquid and the energy efficiency of the ionic liquid compressor were then analyzed. Under the same total pressure ratio and flow rate, when the consumption cost of ionic liquid was considered as a secondary factor, the strategy of equal pressure ratio allocated for five stages was recommended as the compressor efficiency was higher with better heat transfer performance.

Alternative Approach for Thermo-Hydraulic Modeling of Direct Steam Generation in Parabolic Trough Solar Collectors

Abstract The implementation of direct steam generation in linear concentrators is limited mainly by the complexity and the high demand for computational resources of the models developed to predict the installation behavior. With this in mind, we introduce an innovative methodology to characterize the thermo-hydraulic behavior of direct steam generation in parabolic trough solar collectors, with a strong focus on two-phase flow phenomena. Our proposed approach has resulted in a generalized function that eliminates the need for the convective coefficient h and enables accurate prediction of the flow pattern within the receiver. By comparing our model with experimental data from the literature, we achieved relative squared errors (RSEs) values of less than 3% for temperature and pressure calculations, thus validating the robustness of our methodology. The Taitel and Dukler diagram confirms an appropriate flow pattern, while intermittent flow is observed initially during boiling; the pressure drop, although slightly elevated compared to direct solar steam (DISS) loop results, remains within acceptable limits; and the model demonstrates suitability for assessing liquid water, phase change, and superheated steam temperature evolution along the loop. Moreover, we further showcased the practical application of our developed model by applying it to a specific case study conducted in Agua Prieta, Sonora (Northwest México). The validated model exhibits versatility and is applicable to various cases, including both concentrating systems for electricity production and solar heat for industrial processes.

A two-field formulation for surfactant transport within the algebraic volume of fluid method

Surfactant transport plays an important role in many technical processes and industrial applications such as chemical reactors, microfluidics, printing and coating technology. High fidelity numerical simulations of two-phase flow phenomena reveal rich insights into the flow dynamics, heat, mass and species transport. In the present study, a two-field formulation for surfactant transport within the algebraic volume of fluid method is presented. The slight diffuse nature of representing the interface in the algebraic volume of fluid method is utilized to track the concentration of surfactant at the interface as a volumetric concentration. Transport of insoluble and soluble surfactants is investigated by tracking two different concentrations of the surfactant, one within the bulk of the liquid and the other one at the interface. These two transport equations are in turn coupled by source terms considering the ad-/desorption processes at a liquid–gas interface. Appropriate boundary conditions at a solid–fluid interface are formulated to ensure surfactant conservation, while also enabling to study the ad-/desorption processes at a solid–fluid interface. The developed numerical method is verified by comparing the numerical simulations with well-known analytical and numerical reference solutions and applied to a surfactant-laden drop in shear flow. The presented numerical methodology offers a seamless integration of surfactant transport into the algebraic volume of fluid method, where the latter has many advantages such as volume conservation and an inherent ability of handling large interface deformations and topological changes.

Two-Phase Crude Oil-Water Flow Through Different Pipes: An Experimental Investigation Coupled with Computational Fluid Dynamics Approach.

The present study deals with two-phase non-Newtonian pseudoplastic crude oil and water flow inside horizontal pipes simulated by ANSYS. The study helps predict velocity and velocity profiles, as well as pressure drop during two-phase crude-oil-water flow, without complex calculations. Computational fluid dynamics (CFD) analysis will be very important in reducing the experimental cost and the effort of data acquisition. Three independent horizontal stainless steel pipes (SS-304) with inner diameters of 1 in., 1.5 in., and 2 in. were used to circulate crude oil with 5, 10, and 15% v/v water for simulation purposes. The entire length of the pipes, along with their surfaces, were insulated to reduce heat loss. A grid size of 221,365 was selected as the optimal grid. Two-phase flow phenomena, pressure drop calculations, shear stress on the walls, along with the rate of shear strain, and phase analysis were studied. Moreover, velocity changes from the wall to the center, causing a velocity gradient and shear strain rate, but at the center, no velocity variation (velocity gradient) was observed between the layers of the fluid. The precision of the simulation was investigated using three error parameters, such as mean square error, Nash-Sutcliffe efficiency, and RMSE-standard deviation of observation ratio. From the simulation, it was found that CFD analysis holds good agreement with experimental results. The uncertainty analysis demonstrated that our CFD model is helpful in predicting the rheological parameters very accurately. The study aids in identifying and predicting fluid flow phenomena inside horizontal straight pipes in a very effective way.

Investigation of cavitation characteristics in an aircraft centrifugal fuel pump

Cavitation is a classic gas–liquid two-phase flow phenomenon that needs to be restricted for the hydraulic performance enhancement of centrifugal pumps. However, due to the particularity of fuel mediums and rigorous aviation conditions, the characteristics of cavitation occurring in aircraft centrifugal fuel pumps have not been elaborated. In this research, cavitation characteristics in an aircraft centrifugal fuel pump with a high rotational speed of 14,000 rpm were investigated by considering the influence of the flow rate ranging from 0.2Qd to 1.2Qd and the fuel temperature ranging from 20 °C to 100 °C. An experiment for testing the hydraulic performance of an aircraft centrifugal fuel pump was built, and a corresponding numerical simulation of gas–liquid two-phase flow was employed. The distributions of the pressure, streamlines, and turbulence kinetic energy in the centrifugal pump for different flow rates and fuel temperatures were comparatively analyzed. The net positive suction head available (NPSHa) was defined to represent the cavitation degree, and the net positive suction head required (NPSHr) was introduced to characterize the critical cavitation occurrence. It was also found that a large area of cavitation bubbles was generated in the inducer and that this area grew to block the flow channel of the impeller when the NPSHa was close to or lower than the NPSHr, which was the cavitation mechanism of hydraulic performance degradation. Additionally, the developments of the NPSHr with the flow rate and fuel temperature were determined adequately using natural exponential curve fitting. It was deduced that the occurrence of cavitation was more sensitive to higher flow rates, especially when the flow rate was higher than the designed flow rate. The effect of fuel temperature variations should also be considered for the long-time high-speed operation of centrifugal fuel pumps in the aerospace industry.

High-order relaxation methods for nonequilibrium two-phase flow equations

PurposeThe purpose of this paper is to investigate the two-phase flow problems involving gas–liquid mixture.Design/methodology/approachThe governed equations consist of a range of conservation laws modeling a classification of two-phase flow phenomena subjected to a velocity nonequilibrium for the gas–liquid mixture. Effects of the relative velocity are accounted for in the present model by a kinetic constitutive relation coupled to a collection of specific equations governing mass and volume fractions for the gas phase. Unlike many two-phase models, the considered system is fully hyperbolic and fully conservative. The suggested relaxation approach switches a nonlinear hyperbolic system into a semilinear model that includes a source relaxation term and characteristic linear properties. Notably, this model can be solved numerically without the use of Riemann solvers or linear iterations. For accurate time integration, a high-resolution spatial reconstruction and a Runge–Kutta scheme with decreasing total variation are used to discretize the relaxation system.FindingsThe method is used in addressing various nonequilibrium two-phase flow problems, accompanied by a comparative study of different reconstructions. The numerical results demonstrate the suggested relaxation method’s high-resolution capabilities, affirming its proficiency in delivering accurate simulations for flow regimes characterized by strong shocks.Originality/valueWhile relaxation methods exhibit notable performance and competitive features, as far as we are aware, there has been no endeavor to address nonequilibrium two-phase flow problems using these methods.

Operando Identification of Temperature-Dependent Pore-Scale Gas Saturations in PEM Water Electrolyzers

Polymer electrolyte membrane (PEM) water electrolysis is a promising solution for the chemical storage of intermittent renewable energy in the form of clean hydrogen (1). For the widespread adoption of this technology, the efficiency of PEM electrolyzers must be improved through informed material design and tailored operating conditions to compete with incumbent carbon-emitting technologies (2). A main source of inefficiency is related to mass transport losses, where oxygen gas, a by-product of the anode reaction, accumulates in the anode porous transport layer (PTL) and hinders reactant liquid water delivery to reaction sites (3). Although extensive studies have explored this two-phase flow phenomenon, the consideration of operating temperature on gas accumulation in the PTL adds further complexity. A duality in temperature effects has been reported in the literature where increasing the operating temperature may encourage gas coverage which obstructs liquid water transport, but it may also enhance liquid water delivery due to favorable fluid property changes in viscosity and surface tension (4). Efforts have been made previously to reconcile these competing observations using 2D visualization methods (5), however a closer investigation using 3D visualization techniques will provide valuable pore-scale insights in the comprehensive understanding of the impact of operating temperature on gas transport in the PTL, and subsequently mass transport losses in the PEM electrolyzer.In this work, we employed operando X-ray computed tomography (CT) to visualize three-dimensional (3D) pore-scale gas saturations in the PTL at various industrially relevant operating conditions. Specifically, a custom in-house electrolyzer was imaged with synchrotron CT at five different temperatures (40°C, 50°C, 60°C, 70°C, 80°C) and current densities ranging from 0.5 A/cm² to 4.0 A/cm² with 0.5 A/cm² incremental steps. Electrochemical impedance spectroscopy and Tafel measurements were also acquired to characterize the relationship between gas distributions and corresponding mass transport overpotentials. Considering X-ray transparency, a model PTL made of carbon fibers was used to achieve sufficient contrast to quantify water and gas in the obtained images. Using precise volume registration and subtraction methods, pore-scale gas distributions in the PTL were quantified to reveal changes in bubble accumulation and coverage area at different operating temperatures and over operation time. The novel insights from this work will aid in our understanding of the complex relationship between operating temperature and multiphase flow phenomena in the PTL, which is vital for tailoring operating conditions of PEM electrolyzers to curtail the costs of green hydrogen production.

Snap-Off during Imbibition in Porous Media: Mechanisms, Influencing Factors, and Impacts

The phenomenon of snap-off during imbibition in porous media, a fundamental two-phase fluid flow phenomenon, plays a crucial role in both crude oil production and carbon dioxide (CO2) utilization and storage. In porous media where two phases coexist, the instability of the phase interface may give rise to various displacement phenomena, including pore–body filling, piston-like displacement, and snap-off. Snap-off, characterized by the generation of discrete liquid droplets or gas bubbles, assumes paramount significance. This study provides a comprehensive overview of snap-off mechanisms, influencing factors, and impacts. Snap-off initiation arises from variations in the curvature radius at the interface between two phases within narrow regions, primarily influenced by capillary pressure. It can be influenced by factors such as the characteristics of multiphase fluids, the wettability of porous media, as well as the pore–throat geometry and topology within porous media. In turn, snap-off exerts a discernible influence on the fluid dynamics within the porous medium, resulting in impacts that encompass unrecoverable oil droplet formation, the oil bridging effect, drainage–imbibition hysteresis, strong foam generation and transient/dynamic effects. Although the snap-off phenomenon exerts detrimental effects during the conventional waterflooding in oil production, its potential is harnessed for beneficial outcomes in CO2-EOR and CO2 storage. This study significantly advances our understanding of snap-off and its multifaceted roles in multiphase fluid dynamics, offering vital insights for the precise prediction of fluid flow behavior and strategic control. These valuable insights can serve as a theoretical foundation to guide our deliberate modulation of snap-off phenomena, aiming at optimizing oil-recovery processes and enhancing the safety and stability of CO2 storage.

PIV-based experiments on reverse flow characteristics in an equal-height-difference passive heat removal system for ocean nuclear power plants

The equal-height-difference passive heat removal system (EHDPHRS) has a large potential in floating nuclear power plants (FNPPs) due to its ability to fully utilize the marine environment to achieve residual heat removal. However, the horizontal connection between the heat sink and the heat source could easily lead to the subcooled-water reverses flow and thus trigger serious flow instabilities. In the present paper, a fully visualized experimental set was built, combined with a particle image velocity (PIV) system to investigate the subcooled-water reverses flow in the EHDPHRS. First, PIV observations clearly highlighted five flow stages during a complete natural circulation, including the reverse flow occurred in the single-phase and two-phase flow. Then, it was shown that the single-phase reverse velocity is proportional to heating power and inversely proportional to subcooled-water temperature. While, the two-phase reverse velocity increases first and then decreases with the increase in heating power. Moreover, the two-phase reverse flow phenomenon shows obvious periodicity. Finally, the buoyancy force (Fb) caused by the density contrast, and the pressure difference (FΔP) caused by the direct contact condensation (DCC) are the triggering forces for the reverse flow, while the flow inertia force (Fi) caused by the forward flow represents the restraining force.

A Brief on Optical-based Investigation towards The Interfacial Behaviors during High Viscous Liquid / Gas Countercurrent Two-Phase Flow in a Complex Conduit Representing 1/30 Down-Scaled of PWR Hot Leg Geometry

The present work briefly investigates liquid/gas countercurrent two-phase flow phenomena which can be specifically found in a mitigation during an accidental scenario in the operation of a nuclear reactor. A comprehensive knowledge on the corresponding phenomena is obviously important to avoid the failure on the cooling mechanism. Here, a pair of fluid containing high viscous liquid/gas flows through a complex conduit representing 1/30 scaled-down of PWR hot leg’s typical geometry. Furthermore, the flow structures were visually observed, while the film thicknesses are extracted by an image processing algorithm through the corresponding optical-tabulated data. The obtained results reveal that a rather sharp decrease in liquid film corresponds to the flow regime transition.

Percolation transitions of spontaneous imbibition in fractional-wet porous media

Percolation transition is an important scientific question for characterizing complex two-phase flow phenomena in fractional-wet porous media composed of different wetting grains. Recent studies have shown significant transitions of fluid distribution and pressure system in fractional-wet porous media by changing wettability proportion. However, very few studies have systematically investigated controlling factors of percolation transition in spontaneous imbibition. Additionally, there is a lack of comprehensive knowledge and experimental observations about imbibition recovery and dynamic imbibition phenomena (including imbibition time, imbibition rate, and induction time) in fractional-wet porous media. In this study, these knowledge gaps are filled through a series of water imbibition experiments for fractional-wet artificial porous media with different water-wet proportions, three pore diameters, and two wettability heterogeneities. The results show that pore diameter and wettability heterogeneity are influential factors for percolation transition. Decreasing water-wet proportion results in the emergence of induction time, as well as the absences of transition and late imbibition stage, small pore radius enhances these negative impacts. Under strong wettability heterogeneity, relationship of water-wet proportion and imbibition recovery follows the classical percolation theory containing two percolation transitions. Small pore diameter retards percolation transitions, and weakens fluid displacements of the transition stage and late stage in the partial imbibition. Moreover, effects of water-wet proportion on dynamic imbibition phenomena exhibit exponential relationship resulted from transition of pathway type. Additionally, through decreasing wettability heterogeneity, three percolation transitions and an optimal imbibition ability are achieved, thus hydrophilization of hydrophobic pore is advantageous for water imbibition. These intriguing findings can broaden current knowledge about spontaneous imbibition in fractional-wet porous media.

Artificial neural network modelling for predicting classical air pressure profile curves in building drainage systems

Water trap seal loss can lead to the ingress of foul odours from the sewer network carrying harmful pathogens which can affect the health of building occupants. This loss is due to excessive negative or positive pressures within the system as a consequence of rapidly changing flow due to the unsteady nature of air and water flow discharges from appliances. This pressure regime within the system is often represented by a pressure profile along the height of the main vertical stack in the Building Drainage System (BDS). Experimental data from peer-reviewed literature and data from a unique 34-storey drainage test rig have been used as pressure profile data (Test data) for an Artificial Neural Network (ANN) model. Discharge input height (storey number) and discharge flow rate are considered to be the two independent input parameters and the pressure along the vertical stack is considered to be the output. In this work, both a Feed Forward and Back Propagation (FFBP) ANN model and a Radial Basis Function (RBF) ANN model have been used to train the algorithm. The work has confirmed the applicability of the FFBP-ANN model for steady two-phase fluid flow phenomena in BDS and allows for the prediction of pressures in a system for which no pressure data exists, by the prediction of modelled weights, based only on its physical and flow characteristics. Practical Application: Of great concern to designers of building drainage systems (BDS) is the control of pressure fluctuations within the system to prevent water trap seal loss. Prediction of a pressure profile based on the characteristics of the building – e.g. height, location of water discharge, discharge flow rate and ventilation type, would therefore increase confidence in designs, particularly in tall buildings. The work presented here addresses, for the first time, the applications of two ANN models for predicting the pressure profile in the BDS vertical stack of multi-storey buildings.