Search

Computationally efficient modeling of gas turbine combustion is challenging due to the chaotic multi-scale physics and the complex non-linear interactions between acoustic, hydrodynamic, and chemical processes. A large-eddy simulation, referred to as the full order model (FOM), is performed for a gas turbine model combustor with turbulent combustion effects modeled using a flamelet-based method. Modal analysis reveals a high degree of correlation with averaged and instantaneous high-frequency particle image velocimetry fields. The dynamics of the precessing vortex core is quantitatively characterized using dynamic mode decomposition. The governing equations of the FOM are projected onto a low-dimensional linear manifold to construct a reduced-order model (ROM). A discretely-consistent least squares projection is used to guarantee global stability. The ROM provides an accurate reconstruction of the combustion dynamics within the training region, but faces a significant challenge in future state predictions. This limitation is mainly due to the increased projection error, which in turn is a direct consequence of the highly chaotic nature of the flow field, involving a wide range of dispersed coherent structures. This shortcoming is overcome using an adaptive basis method which yields accurate predictions of dynamics beyond the training region consistent with the FOM. Formal projection-based ROMs have not been applied to a problem of this scale and complexity, and achieving accurate and efficient ROMs is a grand challenge problem. A production-ready ROM method will significantly decrease the computational cost of the flame dynamics as well as the portability of this prediction to smaller-scale computers.

The main objective of this work is to develop a unified framework that can be used as a lens to quantitatively assess and augment a wide range of coarse-grained models of turbulence, viz. large eddy simulations (LES), hybrid Reynolds-averaged/LES methods and wall-modeled (WM)LES. Taking a turbulent channel flow as an example, optimality is assessed in the wall-resolved limit, the hybrid RANS/LES limit and the WMLES limit, via projections at different resolutions suitable for these approaches. These optimal a priori estimates are shown to have similar characteristics to existing a posteriori solutions reported in the literature. Consistent accuracy metrics are developed for scale-resolving methods using the optimal solution as a reference, and evaluations are performed. We further characterize the slip velocity in WMLES in terms of the near-wall under-resolution and develop a universal scaling relationship. Insights from the a-priori tests are used to augment existing slip-based wall models. Various a posteriori tests reveal superior performance over the dynamic slip wall model. Guidance for the development of improved slip-wall models is provided, including a target for the dynamic procedure., Comment: 25 pages, 14 figures

We introduce a new approach for adaptive mesh refinement in which adaptivity is driven by low rank decomposition and optimal sensing of the dynamically evolving flow field. This method seeks an ordered set of locations for mesh adaptation from the instantaneous data-driven basis of an online proper orthogonal decomposition of the velocity, which organizes features into sparse optimal orthogonal modes based on an energy norm. The sensing is achieved via a computationally expedient discrete empirical interpolation method using rank-revealing QR factorization (Drmac and Gugercin SIAM J Sci Comput 38(2):A631–A648, 2016). The methodology is applicable to a wide range of numerical discretizations, and is tested on a spatiotemporally evolving incompressible turbulent jet, a complex wind turbine wake, and supersonic flow over a forward-facing step. The proposed approach is demonstrated to predict accurate velocity statistics and yield significantly smaller grids in comparison to gradient-based methods. The algorithm is seen to focus refinement in the vicinity of dynamically significant regions such as those characterized by high turbulence kinetic energy, coherent structures and shock interactions. Moreover, the approach does not require parameters or thresholds, which may be difficult to obtain for complex flows, to be known a priori to facilitate mesh adaptation.

The Koopman operator has emerged as a powerful tool for the analysis of nonlinear dynamical systems as it provides coordinate transformations to globally linearize the dynamics. While recent deep learning approaches have been useful in extracting the Koopman operator from a data-driven perspective, several challenges remain. In this work, we formalize the problem of learning the continuous-time Koopman operator with deep neural networks in a measure-theoretic framework. Our approach induces two types of models: differential and recurrent form, the choice of which depends on the availability of the governing equations and data. We then enforce a structural parameterization that renders the realization of the Koopman operator provably stable. A new autoencoder architecture is constructed, such that only the residual of the dynamic mode decomposition is learned. Finally, we employ mean-field variational inference (MFVI) on the aforementioned framework in a hierarchical Bayesian setting to quantify uncertainties in the characterization and prediction of the dynamics of observables. The framework is evaluated on a simple polynomial system, the Duffing oscillator, and an unstable cylinder wake flow with noisy measurements., Comment: 31 pages

In many-query applications such as design, optimization, and model-predictive control, reduced order models (ROMs) have the potential to significantly decrease the computational cost. In contrast to intrusive projection-based ROMs which require direct access to the high-fidelity model (HFM) operators, non-intrusive ROMs can be constructed in a data-driven fashion using the input-output data generated by the HFM. When used in the prediction of the system response in unseen parameter regimes, however, generalization capabilities have to be carefully assessed. In this study, we pursue a two-step approach to construct non-intrusive parametric ROMs: The first step involves the extraction of low-dimensional latent spaces using proper-orthogonal decomposition (POD) and convolutional neural network-based autoencoders (CNN-AE); and the second step uses regression for the latent variables in parameter space. We adapt a unique decomposition approach named Split-POD to enrich the low-dimensional subspace of the parametric ROM. The proposed methods are used to predict the flow field over a vehicle geometry as a function of vehicle speed and internal fan speed. Through orders of magnitude reduction in the degrees of freedom, the non-intrusive ROMs enable flow field prediction in real-time and bypass hours of computations by the HFM. The results show that the nonlinear encoding in CNN-AE can enhance ROM predictions given adequate data. POD-based ROMs are more robust and efficient, and in particular more accurate than CNN-AE in under-sampled regions of the parameter space. We further enhance the utility of non-intrusive ROMs by formulating a training process that minimizes the required number of high-fidelity simulations using Gaussian process regression to adaptively sample the input parameter space and optimize the estimated variance of the predictions.

Entropy-Stable (ES) schemes, specifically those built from [Tadmor \textit{Math. Comput.} 49 (1987) 91], have been gaining interest over the past decade, especially in the context of under-resolved simulations of compressible turbulent flows using high-order methods. These schemes are attractive because they can provide stability in a global and nonlinear sense (consistency with thermodynamics). However, fully realizing the potential of ES schemes requires a better grasp of their local behavior. Entropy-stability itself does not imply good local behavior [Gouasmi \textit{et al.} \textit{J. Sci. Comp.} 78 (2019) 971, Gouasmi \textit{et al.} \textit{Comput. Methd. Appl. M.} 363 (2020) 112912]. In this spirit, we studied ES schemes in problems where \textit{global stability is not the core issue}. In the present work, we consider the accuracy degradation issues typically encountered by upwind-type schemes in the low-Mach-number regime [Turkel \textit{Annu. Rev. Fluid Mech.} 31 (1999) 285] and their treatment using \textit{Flux-Preconditioning} [Turkel \textit{J. Comput. Phys.} 72 (1987) 277, Miczek \textit{et al.} \textit{A \& A} 576 (2015) A50]. ES schemes suffer from the same issues and Flux-Preconditioning can improve their behavior without interfering with entropy-stability. This is first demonstrated analytically: using similarity and congruence transforms we were able to establish conditions for a preconditioned flux to be ES, and introduce the ES variants of the Miczek's and Turkel's preconditioned fluxes. This is then demonstrated numerically through first-order simulations of two simple test problems representative of the incompressible and acoustic limits, the Gresho Vortex and a right-moving acoustic wave. The results are overall consistent with previous studies [...]

Koopman decomposition is a non-linear generalization of eigen-decomposition, and is being increasingly utilized in the analysis of spatio-temporal dynamics. Well-known techniques such as the dynamic mode decomposition (DMD) and its linear variants provide approximations to the Koopman operator, and have been applied extensively in many fluid dynamic problems. Despite being endowed with a richer dictionary of nonlinear observables, nonlinear variants of the DMD, such as extended/kernel dynamic mode decomposition (EDMD/KDMD) are seldom applied to large-scale problems primarily due to the difficulty of discerning the Koopman invariant subspace from thousands of resulting Koopman eigenmodes. To address this issue, we propose a framework based on multi-task feature learning to extract the most informative Koopman invariant subspace by removing redundant and spurious Koopman triplets. In particular, we develop a pruning procedure that penalizes departure from linear evolution. These algorithms can be viewed as sparsity promoting extensions of EDMD/KDMD. Further, we extend KDMD to a continuous-time setting and show a relationship between the present algorithm, sparsity-promoting DMD, and an empirical criterion from the viewpoint of non-convex optimization. The effectiveness of our algorithm is demonstrated on examples ranging from simple dynamical systems to two-dimensional cylinder wake flows at different Reynolds numbers and a three-dimensional turbulent ship air-wake flow. The latter two problems are designed such that very strong nonlinear transients are present, thus requiring an accurate approximation of the Koopman operator. Underlying physical mechanisms are analyzed, with an emphasis on characterizing transient dynamics. The results are compared to existing theoretical expositions and numerical approximations., 48 pages

We establish subgrid-scale (SGS) characteristics of a finite volume vorticity-transport-based approach for large-eddy simulations. Modified equation analysis indicates that dissipation can be controlled locally via nonlinear limiting of the gradient employed for the vorticity reconstruction. The enstrophy budget highlights the remarkable ability of the truncation terms to mimic the true SGS dissipation and diffusion. Numerical dissipation in under-resolved simulations can be characterized by diffusion terms discovered in the modified equation analysis.

Balanced truncation (BT) is a model reduction method that uses a coordinate transformation to retain eigen-directions that are highly observable and reachable. To address realizability and scalability of BT applied to highly stiff and lightly damped systems, a non-intrusive data-driven method is developed for balancing discrete-time linear systems via the eigensystem realization algorithm (ERA). The advantage of ERA for balancing transformation makes full-state outputs tractable. Further, ERA enables balancing despite stiffness, by eliminating computation of balancing modes and adjoint simulations. As a demonstrative example, we create balanced reduced-order models (ROMs) for a one-dimensional reactive flow with pressure forcing, where the stiffness introduced by the chemical source term is extreme (condition number 10 13 ), preventing analytical implementation of BT. We investigate the performance of ROMs in prediction of dynamics with unseen forcing inputs and demonstrate stability and accuracy of balanced ROMs in truly predictive scenarios, whereas without ERA, proper orthogonal decomposition-Galerkin and least-squares Petrov–Galerkin projections fail to represent the true dynamics. We show that after the initial transients under unit impulse forcing, the system undergoes lightly damped oscillations, which magnifies the influence of sampling properties on predictive performance of the balanced ROMs. We propose an output domain decomposition approach and couple it with tangential interpolation to resolve sharp gradients at reduced computational costs. This article is part of the theme issue ‘Data-driven prediction in dynamical systems’.

The ability to extract generative parameters from high-dimensional fields of data in an unsupervised manner is a highly desirable yet unrealized goal in computational physics. This work explores the use of variational autoencoders for non-linear dimension reduction with the specific aim of disentangling the low-dimensional latent variables to identify independent physical parameters that generated the data. A disentangled decomposition is interpretable, and can be transferred to a variety of tasks including generative modeling, design optimization, and probabilistic reduced order modelling. A major emphasis of this work is to characterize disentanglement using VAEs while minimally modifying the classic VAE loss function (i.e., the Evidence Lower Bound) to maintain high reconstruction accuracy. The loss landscape is characterized by over-regularized local minima which surround desirable solutions. We illustrate comparisons between disentangled and entangled representations by juxtaposing learned latent distributions and the true generative factors in a model porous flow problem. Hierarchical priors are shown to facilitate the learning of disentangled representations. The regularization loss is unaffected by latent rotation when training with rotationally-invariant priors, and thus learning non-rotationally-invariant priors aids in capturing the properties of generative factors, improving disentanglement. Finally, it is shown that semi-supervised learning - accomplished by labeling a small number of samples (O (1%))–results in accurate disentangled latent representations that can be consistently learned.

A projection-based formulation is presented for non-linear model reduction of problems with extreme scale disparity. The approach allows for the selection of an arbitrary, but complete, set of solution variables while preserving the structure of the governing equations. Least-squares-based minimization is leveraged to guarantee symmetrization and discrete consistency with the full-order model (FOM). Two levels of scaling are used to achieve the conditioning required to effectively handle problems with extremely disparate physical phenomena, characterized by extreme stiffness in the system of equations. The formulation -- referred to as model-form preserving least-squares with variable transformation (MP-LSVT) -- provides global stabilization for both implicit and explicit time integration schemes. To achieve computational efficiency, a pivoted QR decomposition is used with oversampling, and adapted to the MP-LSVT method. The framework is demonstrated in representative two- and three-dimensional reacting flow problems, and the MP-LSVT is shown to exhibit improved stability and accuracy over standard projection-based ROM techniques. Physical realizability and local stability are promoted by enforcing limiters in both temperature and species mass fractions. These limiters are demonstrated to be important in eliminating regions of spurious burning, thus enabling the ROMs to provide accurate representations of the heat release rate and flame propagation speed. In the 3D application, it is shown that more than two orders of magnitude acceleration in computational efficiency can be achieved, while also providing reasonable future-state predictions. A key contribution of this work is the development and demonstration of a comprehensive ROM formulation that targets highly challenging multi-scale transport-dominated problems., 47 pages, 26 figures

Data-driven decomposition techniques are presented for the analysis and development of reduced-order models of complex flow dynamics. The Proper Orthogonal Decomposition (POD) produces optimal representations of the dynamics in the sense of the energy norm. Alternatively, Dynamic Mode Decomposition (DMD) efficiently extracts coherent dynamics based on eigendecompositions of linearized dynamics. An extension to the latter, the Higher Order Dynamic Mode Decomposition (HODMD) method uses time delays to develop efficient reduced models to represent complex dynamics in a nonintrusive manner. High-fidelity simulation results of a laboratory-scale single-element gas turbine combustor are used to demonstrate and evaluate the capabilities of the aforementioned decomposition techniques.

A data-driven framework is proposed towards the end of predictive modeling of complex spatio-temporal dynamics, leveraging nested non-linear manifolds. Three levels of neural networks are used, with the goal of predicting the future state of a system of interest in a parametric setting. A convolutional autoencoder is used as the top level to encode the high dimensional input data along spatial dimensions into a sequence of latent variables. A temporal convolutional autoencoder (TCAE) serves as the second level, which further encodes the output sequence from the first level along the temporal dimension, and outputs a set of latent variables that encapsulate the spatio-temporal evolution of the dynamics. The use of dilated temporal convolutions grows the receptive field exponentially with network depth, allowing for efficient processing of long temporal sequences typical of scientific computations. A fully-connected network is used as the third level to learn the mapping between these latent variables and the global parameters from training data, and predict them for new parameters. For future state predictions, the second level uses a temporal convolutional network to predict subsequent steps of the output sequence from the top level. Latent variables at the bottom-most level are decoded to obtain the dynamics in physical space at new global parameters and/or at a future time. Predictive capabilities are evaluated on a range of problems involving discontinuities, wave propagation, strong transients, and coherent structures. The sensitivity of the results to different modeling choices is assessed. The results suggest that given adequate data and careful training, effective data-driven predictive models can be constructed. Perspectives are provided on the present approach and its place in the landscape of model reduction.

A finite-volume scheme is developed for the three-dimensional, incompressible vorticity transport equations (VTE) using multi-dimensional upwinding, with the goal of efficient computations of vortex-dominated flows. By modifying the VTE with a term proportional to the divergence of the vorticity, a stable hyperbolic PDE system with a simple structure is revealed. The structure of the resulting eigensystem, including the vortex stretching term, makes the formulation and solution of the generalized Riemann problem and multi-dimensional upwinding more natural, when compared to the Euler equations. To reduce the computational costs of determining the transverse fluxes, a flux-based wave propagation approach is employed. In this approach, the transverse fluxes are computed via direct manipulation of the one-dimensional generalized Riemann problem with no additional Riemann problem solutions needed. The numerical scheme is implemented within an adaptive mesh refinement framework and evaluated on a series of canonical vortex-dominated flows. A translating vortex flow reveals that the multi-dimensional upwinding substantially outperforms one-dimensional schemes in terms of accuracy and computational time, especially when the vortex propagates oblique to cell surfaces. By including transverse fluxes in simulations of propagating vortex rings, key physical attributes including propagation velocity and impulse can be captured with better accuracy compared to one-dimensional schemes. Further vortex ring simulations demonstrate that the proposed multi-dimensional scheme can preserve vorticity with relatively coarse grids compared to simulations employing the incompressible Euler equations. Integrated quantities from leapfrogging vortex ring problems and energy spectra from decaying turbulent flows confirm that the multi-dimensional scheme can accurately reproduce flows with complex vortex stretching and multiple length scales.

A suite of computational tools capable of predicting in-plane low-frequency rotorcraft noise and its control using blade-tip geometry modifications is developed. The combined code, consisting of AV...

Derivation of reduced order representations of dynamical systems requires the modeling of the truncated dynamics on the retained dynamics. In its most general form, this so-called closure model has to account for memory effects. In this work, we present a framework of operator inference to extract the governing dynamics of closure from data in a compact, non-Markovian form. We employ sparse polynomial regression and artificial neural networks to extract the underlying operator. For a special class of non-linear systems, observability of the closure in terms of the resolved dynamics is analyzed and theoretical results are presented on the compactness of the memory. The proposed framework is evaluated on examples consisting of linear to nonlinear systems with and without chaotic dynamics, with an emphasis on predictive performance on unseen data., 33 pages

The development of reduced models for complex multiscale problems remains one of the principal challenges in computational physics. The optimal prediction framework of Chorin et al., which is a reformulation of the Mori-Zwanzig (M-Z) formalism of non-equilibrium statistical mechanics, provides a methodology for the development of mathematically-derived reduced models of dynamical systems. Several promising models have emerged from the optimal prediction community and have found application in molecular dynamics and turbulent flows. In this work, a new M-Z-based closure model that addresses some of the deficiencies of existing methods is developed. The model is constructed by exploiting similarities between two levels of coarse-graining via the Germano identity of fluid mechanics and by assuming that memory effects have a finite temporal support. The appeal of the proposed model, which will be referred to as the `dynamic-MZ-$\tau$' model, is that it is parameter-free and has a structural form imposed by the mathematics of the coarse-graining process (rather than the phenomenological assumptions made by the modeler, such as in classical subgrid scale models). To promote the applicability of M-Z models in general, two procedures are presented to compute the resulting model form, helping to bypass the tedious error-prone algebra that has proven to be a hindrance to the construction of M-Z-based models for complex dynamical systems. While the new formulation is applicable to the solution of general partial differential equations, demonstrations are presented in the context of Large Eddy Simulation closures for Burgers equation, decaying homogeneous turbulence, and turbulent channel flow. The performance of the model and validity of the underlying assumptions are investigated in detail., Comment: 28 pages. This is a pre-print of an article submitted to the Journal of Computational Physics

A modeling paradigm is developed to augment predictive models of turbulence by effectively using limited data generated from physical experiments. The key components of the current approach involve...

An axisymmetric Reynolds averaged Navier–Stokes solver is used along with an actuator disk model for the analysis of shrouded wind turbine flowfields. Following this, an efficient blade design technique that maximizes sectional power production is developed. These two techniques are incorporated into an optimization framework that seeks to design the geometry of the shroud and rotor to extract maximum power under thrust constraints. The optimal solution is also evaluated using a full three-dimensional Reynolds averaged Navier–Stokes solver, suggesting the viability of the design. The predicted optimal designs yield power augmentations well in excess of the Betz limit, even if the normalization of the power coefficient is performed with respect to the maximum shroud area. Copyright © 2016 John Wiley & Sons, Ltd.

In this work, the space of admissible entropy functions for the compressible multicomponent Euler equations is explored, following up on Harten (J. Comput. Phys. 49 (1983) 151–164). This effort allows us to prove a minimum entropy principle on entropy solutions, whether smooth or discrete, in the same way it was originally demonstrated for the compressible Euler equations by Tadmor (Appl. Numer. Math. 49 (1986) 211–219).

We present applications of modal analysis techniques to study, model, and control canonical aerodynamic flows. To illustrate how modal analysis techniques can provide physical insights in a complementary manner, we selected four fundamental examples of cylinder wakes, wall-bounded flows, airfoil wakes, and cavity flows. We also offer brief discussions on the outlook for modal analysis techniques, in light of rapid developments in data science., 37 pages, 19 figures, 2 tables

In trajectory planning and control design for unmanned air vehicles, highly simplified models are typically used to represent the vehicle dynamics and the operating environment. The goal of this work is to perform real-time, but realistic flight simulations and trajectory planning for quad-copters in low altitude (, Preprint submitted to AIAA Journal

Simulation of multiscale problems remains a challenge due to the disparate range of spatial and temporal scales and the complex interaction between the resolved and unresolved scales. This work develops a coarse-grained modeling approach for the Continuous Galerkin discretizations by combining the Variational Multiscale decomposition and the Mori-Zwanzig (M-Z) formalism. An appeal of the M-Z formalism is that - akin to Greens functions for linear problems - the impact of unresolved dynamics on resolved scales can be formally represented as a convolution (or memory) integral in a non-linear setting. To ensure tractable and efficient models, Markovian closures are developed for the M-Z memory integral. The resulting sub-scale model has some similarities to adjoint stabilization and orthogonal subscale models. The model is made parameter free by adaptively determining the memory length during the simulation. To illustrate the generalizablity of this model, it is employed in coarse-grained simulations for the one-dimensional Burgers equation and in incompressible turbulence problems., Comment: 37 pages, 18 figures

The impact of chemical reactions on the robustness and accuracy of projection-based Reduced-Order Models (ROMs) of fluid flows is investigated. Both Galerkin and Least-Squares Petrov Galerkin ROMs are shown to be less robust in reacting flows as compared to non-reacting flows. In particular, reacting flow ROMs show a strong sensitivity to \st{the} resolution and are often unstable. To identify the main underlying causes, a representative problem that contains the essential physics encountered in typical combustion dynamics problems is chosen. Comparisons with non-reacting solutions are used to assess the impact of reactions. Investigations are focused on three potential areas of significance: 1) preservation of conservation laws; 2) loss of dissipation; and 3) existence of unphysical local phenomena. Results indicate that conservation is relatively well-controlled and the global dissipation in the ROMs is actually larger than that in the underlying CFD solutions. Spurious local phenomena are, however, highly deleterious. Specifically, the steep temperature gradients that characterize combustion can lead to oscillations in local temperatures even in the absence of reactions. Representative calculations with physics-based temperature constraints verify that eliminating such excursions results in considerable improvement in both stability and future-state prediction capability.

An approximation model based on convolutional neural networks (CNNs) is proposed for flow field predictions. The CNN is used to predict the velocity and pressure field in unseen flow conditions and geometries given the pixelated shape of the object. In particular, we consider Reynolds Averaged Navier–Stokes (RANS) flow solutions over airfoil shapes as training data. The CNN can automatically detect essential features with minimal human supervision and is shown to effectively estimate the velocity and pressure field orders of magnitude faster than the RANS solver, making it possible to study the impact of the airfoil shape and operating conditions on the aerodynamic forces and the flow field in near-real time. The use of specific convolution operations, parameter sharing, and gradient sharpening are shown to enhance the predictive capabilities of the CNN. We explore the network architecture and its effectiveness in predicting the flow field for different airfoil shapes, angles of attack, and Reynolds numbers.

New alternatives to traditional petrol vehicles are required to reform the transport industry, which currently contributes nearly 30% of all greenhouse gas emissions in North America. One such alternative is Fuel Cell Electric Vehicles. Fuel cells directly convert chemical energy contained in fuels, such as hydrogen, into electricity, and hence provide a hybrid solution that combines the advantages of internal combustion engines and battery. However, the higher cost, lower durability, and lack of fueling infrastructure have typically hindered their commercialization. In order to address the first two challenges, it is crucial to understand the electrochemical behavior of fuel cells via mathematical models. Such models range from lumped models, which provide simple formulas for voltage-current response curves, to full CFD simulations, which can resolve complex physical phenomena in space and time but come at a high computational cost. The aim of this work is to provide intermediate models for proton-exchange membrane (PEM) fuel cells that are as accurate as possible while solvable in real-time. In this work, we begin with a two-dimensional model of a PEM fuel cell based on Refs. [1] and [2]. The two dimensions are the “along-the-channel” dimension, and the “through-cell” dimension, which is much thinner. This model characterizes mass- and heat-transport related phenomena and hence facilitates water management in cells and membrane electrode assemblies. By nondimensionalizing the model and considering various physically relevant parameter limits, we derive a simplified reduced-order model that accurately reproduces the results of the full model at a much smaller computational cost. The reduced-order model has distributed states along the channel and lumped states in the through-plane dimension. We compare our final reduced-order model to experimental measurements of spatial channel and GDL water distributions obtained using neutron imaging [3]. We find generally good agreement but identify some innate discrepancies. In order to improve the accuracy of our model, we use the Field Inversion and Machine Learning (FIML) modeling paradigm [4, 5]. This consists of inverse modeling to identify the spatial distribution of model discrepancies, and machine learning to learn the dependence of these fields on other model states and hence allow predictive modeling. A key advantage over purely data-driven methods is that the solutions must always satisfy physical laws enforced by the mathematical model, and so only limited data is required. We apply the FIML methodology to our PEM fuel cell model, and hence learn corrections that reduce discrepancies with the experimental data. We show that improved agreement with the experimental data can be obtained in operating conditions that were not used to train the model. The authors acknowledge funding from Toyota Motor Engineering and Manufacturing North America. References [1] R. Vetter and J. O. Schumacher, “Free open reference implementation of a two-phase PEM fuel cell model,” Computer Physics Communications, vol. 234, pp. 223–234, 2019. [2] J. B. Siegel, S. V. Bohac, A. G. Stefanopoulou, and S. Yesilyurt. Nitrogen Front Evolution in Purged Polymer Electrolyte Membrane Fuel Cell with Dead-Ended Anode. Journal of the Electrochemical Society (2010) 157(7): B1081-B1093. [3] J. B. Siegel, D. A. McKay, A. G. Stefanopoulou, D. S. Hussey, D. L. Jacobson, “Measurement of Liquid Water Accumulation in a PEMFC with Dead-Ended Anode,” Journal of The Electrochemical Society (2008), 155 (11) B1168-B1178. DOI: 10.1149/1.2976356 [4] E.J. Parish and K. Duraisamy. A paradigm for data-driven predictive modeling using field inversion and machine learning. Journal of Computational Physics 305 (2016): 758-774. [5] A.P. Singh, S. Medida and K. Duraisamy. Machine-learning-augmented predictive modeling of turbulent separated flows over airfoils. AIAA Journal (2017): 2215-2227.

The Algebraic Structure-based Model (ASBM) is an engineering model of turbulence that incorporates information about the morphology of turbulent structures, and thus characterizes additional turbulence information beyond what is provided solely by the Reynolds stresses. In this work, a new formulation of the ASBM is developed that exhibits three defining characteristics. The first is a near-wall correction segregated from the homogeneous component of the model, which allows for a new paradigm for model development and comparison. The second characteristic is a set of fully-explicit equations that replace the original implicit formulation and make evident the highly nonlinear nature of the model. Finally, the last defining characteristic is based on the coupling of the ASBM with turbulent transport equations that provide the required turbulent scales. Using arguments of consistency, stability, and accuracy we arrive at a model that uses transport equations for k and ϵ. The resulting model is simpler to implement, more robust, and amenable to further development. The analysis of the new ASBM formulation is carried out by computing fully-developed channel, pipe and squared duct flows.