Traditional Computational Fluid Dynamics Research Articles

Evaluation of reduced-order models for the rapid aerodynamic analysis of supersonic and hypersonic bodies

Abstract This paper presents a comprehensive evaluation of reduced-order models (ROMs) for the determination of pressure coefficient distributions on supersonic and hypersonic bodies. The study investigates the limitations, aerodynamic precision and computational performance associated with various methodologies, ranging from simplistic Newtonian theory-based approaches to more advanced first and second-order shock-expansion theories. Validation is performed by comparing computed results with experimental and computational data for pressure distributions, drag and lift coefficients and centres of pressure for fundamental geometries and authentic vehicle design over a wide range of freestream conditions. The study also includes a comprehensive computational complexity analysis, demonstrating the superiority of finite-element ROM approaches over traditional finite-volume computational fluid dynamics (CFD) simulations. The primary objective of this paper is to scrutinise the extension of these methodological classes to the low supersonic regime. Hence, thermo-chemical reactions within the flow are disregarded, and the ideal gas law is adopted. A value of $\gamma = 1.4$ is chosen for consistency and comparability across the analyses. The proposed ROMs show remarkable potential for reducing high-speed simulation execution times by four orders of magnitude, maintaining accuracy within 20 per cent and as low as 1 per cent. The study unveils three key findings: first, the accuracy degradation of Newtonian-based theories for inclined elements, particularly around 45 degrees, and their reduced dependency on Mach number at large inclination. Secondly, the study presents novel insights into the impact of shock-wave-Mach-wave interactions on pressure distribution calculations, emphasising the Mach number as a crucial metric governing recompression effects. Lastly, the study demonstrates the exceptional accuracy of DeJarnette’s method, providing ${C_P}$ results within 2 per cent for a wide range of conditions, offering an attractive alternative to the Taylor-Maccoll equation.

Explainable and Reduced-Feature Machine learning models for shape and drag prediction of a freely moving drop in the sub-critical Weber number regime

Accurately predicting the shape and drag of a moving drop is crucial in many spray applications. However, due to the complex interaction between the drag force and drop shape deformation, accurate prediction of drop shape and drag from Computational Fluid Dynamics (CFD) simulation requires large computational resources and time. A novel data-driven approach using NARXNN (Non-linear Auto Regressive eXogenous input Neural Network) is proposed in this study, which recurrently predicts the drop shape and drag for a given Weber number (We) and Reynolds number (Re), in the sub-critical We regime where there is no drop breakup. The average error in radius and drag coefficient for the test cases, as low as 0.36% and 0.49%, respectively, is achieved. Most importantly, SHAP (SHapley Additive exPlanations) analysis is performed to identify important features for the predictions, hence enabling the explainability of the physics involved in the predictions of machine learning (ML) models for drop deformation and the interaction between the drop and gas flows. Global interpretation of SHAP values identifies We and Re as the most important features to predict the drop shape and drag. Based on the SHAP analysis, several reduced-feature models are developed. All the reduced-feature models can predict drop shape and drag coefficient within 4% average error in radius and drag coefficient for any We and Re combinations within the prescribed parameter space. The ML models developed in this study demonstrate a tremendous computational advantage over the traditional CFD simulations. The drop shape and drag coefficient evolution can be predicted within seconds for a single case, whereas the CFD simulation takes several days to a week on a single processor.

Physics-informed neural networks and higher-order high-resolution methods for resolving discontinuities and shocks: A comprehensive study

Addressing discontinuities in fluid flow problems is inherently difficult, especially when shocks arise due to the nonlinear nature of the flow. While handling discontinuities is a well-established practice in computational fluid dynamics (CFD), it remains a major challenge when applying physics-informed neural networks (PINNs). In this study, we compare the shock-resolving capabilities of traditional CFD methods with those of PINNs, highlighting the advantages of the latter. Our findings show that PINNs exhibit less dissipative behavior compared to conventional techniques. We evaluated the performance of both PINNs and traditional methods on linear and nonlinear test cases, demonstrating that PINNs offer superior shock-resolving properties. Notably, PINNs can accurately resolve inviscid shocks with just three grid points, whereas traditional methods require at least seven points. This suggests that PINNs are more effective at resolving shocks and discontinuities when using the same grid for both PINN and CFD simulations. However, it’s important to note that PINNs, in this context, are computationally more expensive than traditional methods on a given grid.

Artificial Neural Networks for Mineral Production Forecasting in the In Situ Leaching Process: Uranium Case Study

This study was conducted to assess the applicability of artificial neural networks (ANN) for forecasting the dynamics of uranium extraction over exploitation time during the process of In Situ Leaching (ISL). Currently, ISL process simulation involves multiple steps, starting with geostatistical interpolation, followed by computational fluid dynamics (CFD) and reactive transport simulation. While extensive research exists detailing each of these steps, machine learning techniques may offer the potential to directly obtain extraction curves (i.e., the concentration of the mineral produced over the exploitation time of the deposit), thereby bypassing these computationally expensive steps. As a basis, both an empirical experimental configuration and reactive transport simulations were used to generate training data for the neural network model. An ANN was constructed, trained, and tested on several test cases with different initial parameters, then the expected outcomes were compared to those derived from conventional modeling techniques. The results indicate that for the employed experimental configuration and a limited number of features, artificial intelligence technologies, specifically regression-based neural networks can model the recovery rate (or extraction degree) of the ISL process for mineral production, achieving a high degree of accuracy compared to traditional CFD and mass transport models.

On the design and fabrication of nanoliter-volume hanging drop networks

Hanging drop cultures provide a favorable environment for the gentle, gel-free formation of highly uniform three-dimensional cell cultures often used in drug screening applications. Initial cell numbers can be limited, as with primary cells provided by minimally invasive biopsies. Therefore, it can be beneficial to divide cells into miniaturized arrays of hanging drops to supply a larger number of samples. Here, we present a framework for the miniaturization of hanging drop networks to nanoliter volumes. The principles of a single hanging drop are described and used to construct the fundamental equations for a microfluidic system composed of multiple connected drops. Constitutive equations for the hanging drop as a nonlinear capacitive element are derived for application in the electronic-hydraulic analogy, forming the basis for more complex, time-dependent numerical modeling of hanging drop networks. This is supplemented by traditional computational fluid dynamics simulation to provide further information about flow conditions within the wells. A fabrication protocol is presented and demonstrated for creating transparent, microscale arrays of pinned hanging drops. A custom interface, pressure-based fluidic system, and environmental chamber have been developed to support the device. Finally, fluid flow on the chip is demonstrated to align with expected behavior based on the principles derived for hanging drop networks. Challenges with the system and potential areas for improvement are discussed. This paper expands on the limited body of hanging drop network literature and provides a framework for designing, fabricating, and operating these systems at the microscale.

Unsupervised Learning Bioreactor Regimes

Efficient operation of bioreactors is pivotal for the success of biomanufacturing. Traditional Computational Fluid Dynamics (CFD) simulations, while detailed, often suffer from long computation times and complexity, making them impractical for real-time applications. This study introduces a novel multivariate unsupervised learning algorithm for clustering bioreactors into coherent regions based on CFD-generated and real-world data. The resulting clusters can be used to determine regimes inside a reactor, or can work as an initial step for compartment models. Our method employs a custom k-means clustering algorithm that not only ensures spatial continuity within clusters but also optimizes the number of compartments based on clustering scores. This optimization process focuses on defining compartments clearly while retaining maximum information from the continuous body, as demonstrated by a Pareto front analysis. The efficacy and adaptability of this approach were validated through two case studies: a 202 m³ Rushton impeller bioreactor and an 840 m³ airlift reactor. The results underscore the benefits of 3-D compartmentalization in capturing the complex dynamics of fluid motion and cellular activities. By establishing a robust method for scaling down experiments and enhancing bioreactor design, this study paves the way for more efficient industrial applications.

Prediction of Turbulent Boundary Layer Flow Dynamics with Transformers

Time-marching of turbulent flow fields is computationally expensive using traditional Computational Fluid Dynamics (CFD) solvers. Machine Learning (ML) techniques can be used as an acceleration strategy to offload a few time-marching steps of a CFD solver. In this study, the Transformer (TR) architecture, which has been widely used in the Natural Language Processing (NLP) community for prediction and generative tasks, is utilized to predict future velocity flow fields in an actuated Turbulent Boundary Layer (TBL) flow. A unique data pre-processing step is proposed to reduce the dimensionality of the velocity fields, allowing the processing of full velocity fields of the actuated TBL flow while taking advantage of distributed training in a High Performance Computing (HPC) environment. The trained model is tested at various prediction times using the Dynamic Mode Decomposition (DMD) method. It is found that under five future prediction time steps with the TR, the model is able to achieve a relative Frobenius norm error of less than 5%, compared to fields predicted with a Large Eddy Simulation (LES). Finally, a computational study shows that the TR achieves a significant speed-up, offering computational savings approximately 53 times greater than those of the baseline LES solver. This study demonstrates one of the first applications of TRs on actuated TBL flow intended towards reducing the computational effort of time-marching. The application of this model is envisioned in a coupled manner with the LES solver to provide few time-marching steps, which will accelerate the overall computational process.

Computationally effective machine learning approach for modular thermal energy storage design

This research presents an innovative approach that integrates computational fluid dynamics (CFD) and machine learning (ML) for the design and optimization of thermal energy storage (TES) systems. Heat discharging parametric analyses conducted using CFD serve as the basis for training ML models, including linear regression, K-nearest neighbor (KNN) regression, gradient boost regression (GBR), XGBoost, LightGBM, and neural network (NN). NN emerges as the most suitable for predicting time-dependent variations of concrete and heat transfer fluid (HTF) temperatures. The trained ML models offer an efficient alternative to traditional CFD simulations, enabling the prediction of temperatures in concrete thermal energy storage (CTES) modules under varying inlet conditions, velocities, and time. Leveraging these ML models, the research demonstrates the design of modular CTES cascaded systems with multiple modules in series and parallel configurations, significantly reducing computational cost and time by over 99% compared to full-scale CFD simulations. For instance, in predicting 4-hour time-dependent thermal behavior, CFD takes 97 s per data point and 238,500 s for a single module, compared to ML models’ 16-20 ms per data point and around 290 s per module, indicating their efficiency and scalability in predicting thermal discharge, especially for modular CTES system design and optimization. ML models also demonstrate computational efficiency for designing CTES systems involving multiple modules, taking approximately 765 s - 1047 s for various CTES system configurations, indicating their effectiveness over CFD in predicting thermal discharge for modular CTES systems. The integration of CFD and ML provides a streamlined workflow for designing and optimizing CTES systems, reducing computational efforts, cost, and time. Moreover, this workflow can be updated with additional training data to implement it for unique modular designs with different conditions. Such a generalization of this ML-based approach makes it applicable to a wide range of thermal energy storage designs and geometries, offering a promising avenue for future research and development in the field of thermal energy storage.

Optimising ship principal dimensions with a Dung Beetle Optimizer and random forest proxy model

ABSTRACT Shipping serves as an important component of global trade, with speed being a key attribute closely tied to navigational resistance. Amid rising marine pollution, the International Maritime Organisation (IMO) requires shipowners to modify ships to comply with existing requirements. Herein, focusing on the static water sailing resistance, whole-ship hull line optimisation was conducted for Japan Bulk Carrier (JBC). The optimisation included the bulbous bow, inflow and outflow sections, and the parallel mid-body. Initially, a random forest resistance proxy model was introduced to replace traditional Computational Fluid Dynamics (CFD) methods. The model’s hyperparameters were optimised using the Dung Beetle Optimizer algorithm. Subsequently, this algorithm was employed to optimise ship design parameters. Finally, comparisons confirmed the proxy model’s high precision, and the optimisation yielded substantial outcomes, achieving a 4.7% decrease in resistance.

Deep learning-aided active subspace exploration of free-stream effects for fan-shaped film cooling

Film cooling plays a critical role in protecting engine components from high temperatures that can influence safety and performance of gas turbines. However, the process is fraught with uncertainties due to complex inflow conditions and geometrical configurations. These uncertainties significantly impact cooling effectiveness, underscoring the importance of identifying the dominant factors in a quantified manner. Traditional methods, such as the Monte Carlo approach, encounter the “curse of dimensionality,” making them computationally intensive as the number of variable increases. This study tackles these challenges by employing a deep learning strategy with a convolutional neural network (CNN) model to predict film cooling effectiveness efficiently, reducing computational loads compared to traditional computational fluid dynamics (CFD) simulations. It enables the CNN model to act as a surrogate for the CFD simulation and provides gradient information for subsequent analyses. Additionally, this study employs active subspace (AS) analysis for dimensionality reduction, identifying dominant parameters based on the trained CNN model. This approach not only enhances the speed of simulations but also provides an effective way to analyze dominant parameters and carry out optimizations. Results demonstrate a strong correlation between the CNN predictions and detailed CFD simulations, with all the mean absolute errors below 0.08, validating the model's efficacy in capturing complex cooling dynamics. The dominant analysis based on the AS method gives the blowing ratio as the most important factors, and the subsequent parameter dimension reduction and suggested optimization region are also explored by applying the low-dimensional active variable. The tested optimal factors yield higher spatially averaged cooling effectiveness above 0.35 and show reasonable pattern changes compared to the maximum and minimum results in the sample.

MLFV: a novel machine learning feature vector method to predict characteristics of turbulent heat and fluid flow

PurposeThis study aims to introduce a novel machine learning feature vector (MLFV) method to bring machine learning to overcome the time-consuming computational fluid dynamics (CFD) simulations for rapidly predicting turbulent flow characteristics with acceptable accuracy.Design/methodology/approachIn this method, CFD snapshots are encoded in a tensor as the input training data. Then, the MLFV learns the relationship between data with a rod filter, which is named feature vector, to learn features by defining functions on it. To demonstrate the accuracy of the MLFV, this method is used to predict the velocity, temperature and turbulent kinetic energy fields of turbulent flow passing over an innovative nature-inspired Dolphin turbulator based on only ten CFD data.FindingsThe results indicate that MLFV and CFD contours alongside scatter plots have a good agreement between predicted and solved data with R2 ≃ 1. Also, the error percentage contours and histograms reveal the high precisions of predictions with MAPE = 7.90E-02, 1.45E-02, 7.32E-02 and NRMSE = 1.30E-04, 1.61E-03, 4.54E-05 for prediction velocity, temperature, turbulent kinetic energy fields at Re = 20,000, respectively.Practical implicationsThe method can have state-of-the-art applications in a wide range of CFD simulations with the ability to train based on small data, which is practical and logical regarding the number of required tests.Originality/valueThe paper introduces a novel, innovative and super-fast method named MLFV to address the time-consuming challenges associated with the traditional CFD approach to predict the physics of turbulent heat and fluid flow in real time with the superiority of training based on small data with acceptable accuracy.

Numerical Method for Predicting Transient Aerodynamic Heating in Hemispherical Domes

In this research, a streamlined numerical approach designed for the quick estimation of temperature profiles across the finite thickness of a hemispherical dome subjected to aerodynamic heating is introduced. Hemispherical domes, with their advantageous aerodynamic, structural, and optical properties, are frequently utilized in the front sections of objects traveling at supersonic velocities, including missiles or vehicles. The proposed method relies on one-dimensional analyses of fluid dynamics and flow characteristics to approximate the local heat flux across the exterior surface of the dome. By calculating these local heat flux values, it is also possible to predict the temperature variations within the thickness of the dome by employing the finite difference technique, to solve the heat conduction equation in spherical coordinates. This process is iterated over successive time intervals, to simulate the entire flight duration. Unlike traditional Computational Fluid Dynamics (CFD) simulations, the proposed strategy offers the benefits of significantly lower computational time and resource demands. The primary objective of this work is to provide an efficient numerical tool for evaluating aerodynamic heating impact and temperature gradients on hemispherical domes under specific conditions. The effectiveness of the proposed method will be validated by comparing the temperature profiles derived for a standard flight scenario against those obtained from 2-D axisymmetric transient CFD simulations performed using ANSYS-Fluent 2022 R2.

Fast flow field prediction of pollutant leakage diffusion based on deep learning.

Predicting pollutant leakage and diffusion processes is crucial for ensuring people's safety. While the deep learning method offers high simulation efficiency and superior generalization, there is currently a lack of research on predicting pollutant leakage and diffusion flow field using deep learning. Therefore, it is necessary to conduct further studies in this area. This paper introduces a two-level network method to model the flow characteristics of pollutant diffusion. The proposed method in this study demonstrates a significant enhancement in flow field prediction accuracy compared to traditional deep learning methods. Moreover, it improves computational efficiency by over 800 times compared to traditional computational fluid dynamics (CFD) methods. Unlike conventional CFD methods that require grid expansion to calculate all operation conditions, the deep learning method is not confined by grid limitations. While deep learning methods may not entirely replace CFD methods, they can serve as a valuable supplementary tool, expanding the versatility of CFD methods. The findings of this research establish a robust foundation for incorporating deep learning methods in addressing pollutant leakage and diffusion challenges.

Degradation performance rapid prediction and multi-objective operation optimization of gas turbine blades

The performance degradation of high-temperature blades will affect the safety and efficiency of operation. In response to issues such as ignoring strength characteristics, slow degradation performance prediction, insufficient applicability and accuracy of traditional proxy models, and long operation optimization cycle, a degradation performance prediction model of turbine blades based on the back propagation neural network is established in this research. Furthermore, the efficiency-stress multi-objective operation optimization method based on genetic algorithm is proposed. The results show that the R-square value of the degradation performance neural network prediction model is greater than 0.9 for predicting characteristic parameters. The prediction speed is improved by 104 orders of magnitude compared to the traditional computational fluid dynamics method. The influence mechanisms of turbine outlet flow rate, cooling gas inlet total temperature, and cooling gas flow rate on the degradation performance are revealed. Under fouling extreme conditions and corrosion/wear extreme conditions, the operating parameters of turbine blades are optimized. The maximum values of the efficiency improvement, power increase and maximum von Mises stress reduction are 0.74 %, 2.13 MW and 43 MPa, respectively. This study can provide an effective and novel method for real-time degradation performance prediction and rapid operating condition optimization of turbine blades.

Deep neural network model for predicting thermal-hydraulic performance of a solar air heater with artificial roughness: Sensitivity, generalization capacity, and computational efficiency

A novel coupled computational fluid dynamics-deep neural network approach is proposed to accurately predict the performance of an artificially roughened solar air heater. Computational fluid dynamics data sets are used to develop an optimized neural network model for the prediction of thermal-hydraulic performance factor. The optimized neural network model with softmax transfer function in hidden layers has an architecture of 3‐37‐37‐1 and it predicts the thermal-hydraulic performance factor of the artificially roughened solar air heater with maximum, mean, and minimum error of 2.54 %, 0.245 %, and 0.0009 %, respectively. The concept of Shapley values is adopted to evaluate the global sensitivity analysis of the optimized deep neural network model and it is found that rib height is the most influential design parameter for the artificially roughened solar air heater. The generalization capacity of the neural network model is evaluated by comparing its predictions with experimental data, demonstrating superior predictive performance over traditional computational fluid dynamics simulations. The optimized artificial neural network model achieves a computational time gain of approximately 100 % against computational fluid dynamics.

Generalization of single wake surrogates for multiple and farm-farm wake analysis

Wind farm efficiency is influenced by atmospheric turbulence and wake interactions from preceding turbines. Optimal performance necessitates effective control strategies, encompassing collective/individual pitch (and/or torque) control, yaw control, and innovative techniques, significantly boosting energy capture. Wake effects are crucial, impacting downstream turbines and reducing overall energy extraction. For this purpose, the study of wind farm flow control (WFFC) holds significant relevance in this context. In this study, a Deep Neural Network predicts downstream flow features of a wind turbine under diverse control scenarios, including varying thrust coefficient and yaw control. A feed-forward neural network models the deficit and added turbulent intensity of a single wake, trained using Computational Fluid Dynamics (CFD) simulations as reference data. Linear superposition establishes the wind farm flow field, allowing examination of downstream effects. Another feedforward neural network encompasses wind farm physics such as blockage and wake recovery in and around wind turbine arrays which are overlooked by the single wake model. The methodology employed in this study yields results that are more time-efficient compared to traditional CFD models while maintaining a higher level of accuracy (ideally) than the engineering models, especially when implemented for WFFC without calibration.

Strategies for multi-case physics-informed neural networks for tube flows: a study using 2D flow scenarios

Fluid dynamics computations for tube-like geometries are crucial in biomedical evaluations of vascular and airways fluid dynamics. Physics-Informed Neural Networks (PINNs) have emerged as a promising alternative to traditional computational fluid dynamics (CFD) methods. However, vanilla PINNs often demand longer training times than conventional CFD methods for each specific flow scenario, limiting their widespread use. To address this, multi-case PINN approach has been proposed, where varied geometry cases are parameterized and pre-trained on the PINN. This allows for quick generation of flow results in unseen geometries. In this study, we compare three network architectures to optimize the multi-case PINN through experiments on a series of idealized 2D stenotic tube flows. The evaluated architectures include the ‘Mixed Network’, treating case parameters as additional dimensions in the vanilla PINN architecture; the “Hypernetwork”, incorporating case parameters into a side network that computes weights in the main PINN network; and the “Modes” network, where case parameters input into a side network contribute to the final output via an inner product, similar to DeepONet. Results confirm the viability of the multi-case parametric PINN approach, with the Modes network exhibiting superior performance in terms of accuracy, convergence efficiency, and computational speed. To further enhance the multi-case PINN, we explored two strategies. First, incorporating coordinate parameters relevant to tube geometry, such as distance to wall and centerline distance, as inputs to PINN, significantly enhanced accuracy and reduced computational burden. Second, the addition of extra loss terms, enforcing zero derivatives of existing physics constraints in the PINN (similar to gPINN), improved the performance of the Mixed Network and Hypernetwork, but not that of the Modes network. In conclusion, our work identified strategies crucial for future scaling up to 3D, wider geometry ranges, and additional flow conditions, ultimately aiming towards clinical utility.

Hypersonic inlet flow field reconstruction dominated by shock wave and boundary layer based on small sample physics-informed neural networks

High Mach number inlets face complex challenges such as shock wave/boundary layer interference, adversely impacting the aerodynamic characteristics and stable operating boundaries. Traditional computational fluid dynamics (CFD) simulations for performance and flow field analysis are slow, and data-driven deep learning technologies struggle with predicting flow fields in complex aerodynamic phenomena like shock wave/boundary layer interactions, boundary layer separation, and Mach disks. This study introduces a rapid, robust method for reconstructing hypersonic viscous inlet flow fields using inlet geometry design parameters. This method is called shock wave and boundary layer dominated two-dimensional multiphysical field reconstruction model based on multi-scale sensory field fusion residual physical information neural network (PIMSRF_ResNet), ensuring high-precision training data. Using the optimal Pareto solution set from multi-objective optimization with a small sample intelligence method, 150 sets of multi-physics fields under Ma10 conditions with varying geometric design parameters are calculated. Compared to traditional pure data-driven models, PIMSRF_ResNet's structural similarity in the test set reaches 0.9773, with a peak signal-to-noise ratio close to 30 dB and a correlation coefficient exceeding 98%. These experimental results demonstrated that the proposed method is a promising tool for real-time prediction of complex internal flow fields dominated by high Mach number shock waves and boundary layers.

A deep learning computational fluid dynamics solver for simulating liquid hydrogen jets

Modeling and simulating the sudden depressurization of liquids inside nozzles is a significant challenge because of the plethora of the associated complex phenomena. This pressure drop together with the rapid phase change of the liquid is important characteristics of flash boiling. Computational fluid dynamics (CFD) multiscale simulations of flashing jets usually deploy additional models for modeling heat and mass transfer with long computational times. Intermediate steps such as volumetric meshing in mesh-based methods can also significantly increase the computational cost. This paper aims at providing academia and industry with a modeling tool to simulate and investigate the complex multi-facet phenomenon of flash-boiling atomization deploying a machine-learning method that could save thousand Central Processing Unit hours offering instantaneous CFD predictions. The presented machine-learning CFD method completely replaces the traditional CFD simulations workflow and requires little simulation expertise from the end-user. Notably, this is a novel model that couples for the first time the thermodynamic non-equilibrium with convolutional neural networks to simulate flashing liquid hydrogen jets thousand times faster than the standalone CFD solver. The accuracy of the novel approach is evaluated, demonstrating adequate accuracy compared to different unseen simulations and experiments. This work offers the groundwork for further accelerating CFD predictions in multiphase flow problems and could significantly improve testing flash-boiling scenarios in various industrial settings.

Graph Neural Network for domain segmentation to predict regions of non-ideal mixing in two-dimensional baffle flow systems

This paper presents a novel approach to address computational challenges in predicting flow features by employing a Graph Neural Network (GNN), which is proficient in predicting flow domain values. Traditional Computational Fluid Dynamics (CFD) simulations, although effective, often require substantial computational resources and time, limiting their applicability in time-sensitive scenarios and optimization studies necessitating extensive case studies. The main objective was to evaluate the feasibility of employing node classification on a graph generated from a 2D baffle flow system to segment the domain based on relative fluid age. A second objective was to compare the computational time required for CFD simulations with the inference time of the network to quantify the efficiency gains achieved by utilizing the network. Results demonstrate the potential of utilizing graph convolutional networks for domain segmentation to predict regions of holdup and bypass. The GNN achieved 97% and 92% accuracy in predicting recirculation regions in single and double baffle cases, particularly excelling in high Reynolds number scenarios. Importantly, the proposed GNN-based approach reduces computation time by over 2100%, showcasing significant efficiency gains. Results here highlight the promise of employing graph convolutional networks for flow feature prediction, offering substantial computational efficiency improvements over traditional CFD simulations.