We present a monolithic hp space-time multigrid method (hp-STMG) for tensor-product space-time finite element discretizations of the incompressible Navier-Stokes equations. We employ mapped inf-sup stable pairs ℚ_r+1/ℙ_r^disc in space and a slabwise discontinuous Galerkin DG(k) discretization in time. The resulting fully coupled nonlinear systems are solved by Newton-GMRES preconditioned with hp-STMG, combining geometric coarsening in space with polynomial coarsening in space and time. Our main contribution is an hp-robust and practically efficient extension of space-time multigrid to Navier-Stokes: matrix-free operator evaluation is retained via column-wise, state-dependent spatial kernels; the nonlinear convective term is handled by a reduced, order-preserving time quadrature. Robustness is ensured by an inexact space-time Vanka smoother based on patch models with single time point evaluation. The method is implemented in the matrix-free multigrid framework of deal.II and demonstrates h- and p-robust convergence with robust solver performance across a range of Reynolds numbers, as well as high throughput in large-scale MPI-parallel experiments.

This work presents a goal-oriented a posteriori error estimator based on the Dual Weighted Residual (DWR) method together with space-time mesh adaptivity for the Navier–Stokes equations. The resulting nonlinear algebraic systems on the space-time slabs are solved by Newton's method with GMRES, preconditioned by a slab-wise geometric multigrid method. This combination yields reliable control of target quantities on computationally feasible space-time meshes together with a robust and efficient solution of the algebraic systems. The implementation is based on a MPI-parallel programming model in the deal.II library. Further ingredients are a discontinuous Galerkin discretization in time and inf-sup stable finite element pairs with discontinuous pressure on tensor-product meshes. The performance of the approach is investigated in benchmark computations with regard to accuracy, efficiency, and stability.

Stochastic Galerkin methods offer unexplored potential for the numerical simulation of parabolic problems with random variables, in particular if they are combined with variational discretizations of the space and time variables. Due to the high dimensionality, the solution of the arising algebraic systems do not become feasible without efficient solvers, preconditioners, and software architectures. A stochastic Galerkin discretization with an embedded slabwise finite element approximation of the space and time variables is proposed and analyzed numerically. For solving the linear systems, GMRES iterations are block-preconditioned by a geometric multigrid (GMG) technique using a local Vanka smoother for the space-time subsystems. Monte-Carlo methods are also used for solving random parabolic problems and studied here for the purpose of comparison. The Monte-Carlo approach is built on the space-time finite element formulation together with the GMRES-GMG solver technology. All algorithms have been implemented in a unified matrix-free framework based on the deal.II software library. Comparative numerical evaluations illustrate the performance properties of both approaches, including convergence of the discretizations and statistics of the algebraic solver. Superiority of the stochastic Galerkin approach is observed.

We propose and investigate space-time discontinuous Galerkin (DG) finite element approximations of the wave equation that is recast as an evolutionary first-order system. The DG gradient and DG divergence operator on broken polynomial spaces are defined carefully. The solution of the algebraic system by GMRES iterations with geometric multigrid (GMG) preconditioning is addressed. The performance properties of the DG approach are illustrated by numerical experiments.

We present a monolithic hp space-time multigrid method for tensor-product spacetime finite element discretizations of the Stokes equations. Geometric and polynomial coarsening of the space-time mesh is performed, and the entire algorithm is expressed through rigorous mathematical mappings. For the discretization, we use inf-sup stable pairs Qr+1/Pdisc r of elements in space and a discontinuous Galerkin (DG(k)) discretization in time with piecewise polynomials of order k. The key novelty of this work is the application of hp multigrid techniques in space and time, facilitated and accelerated by the matrix-free capabilities of the deal.II library. While multigrid methods are well-established for stationary problems, their application in space-time formulations encounter unique challenges, particularly in constructing suitable smoothers. To overcome these challenges, we employ space-time cell and vertex star patch based Vanka smoothers. Extensive tests on high-performance computing platforms demonstrate the efficiency of our hp multigrid approach on problem sizes exceeding a trillion degrees of freedom (dofs), sustaining throughputs of hundreds of millions of dofs per second.

Accurate flow simulations, especially at higher Reynolds numbers, are a complex challenge in computational science that often pushes classical methods to their limits. To address this, we have developed a hybrid deep neural network/finite element method. It utilizes classical finite element approximation techniques to efficiently represent large flow fields and deep neural networks to learn updates on finer meshes. Here, we evaluate the approach by comparisons across different FEM libraries. Extensive benchmarks underscore the scalability, efficiency, generalizability, and accuracy of DNN-MG.

We study a space-time finite element method for a system of poromechanics with memory effects that are modeled by a convolution integral. In the literature, the system is referred to as the Biot-Allard model. We recast the model as a first-order system in time, where the memory effects are transformed into an auxiliary differential equation. This allows for a computationally efficient numerical scheme. The system is discretized by continuous Galerkin methods in time and equal-order finite element methods in space. An optimal order error estimate is proved for the norm of the first-order energy of the unknowns of the system. The estimate is confirmed by numerical experiments.

Bayesian optimisation (BO) protocol based on Active Learning (AL) principles has garnered significant attention due to its ability to efficiently optimise black-box objective functions. This capability is a prerequisite for guiding autonomous and high-throughput materials design and discovery processes. However, its application in the materials science domain, particularly for novel alloy designs with targeted properties, remains limited. This limitation is due to the computational complexity and the lack of reliable and robust acquisition functions for multi-objective optimisation. In recent years, expected hypervolume-based geometrical acquisition functions have demonstrated superior performance and speed compared to other multi-objective optimisation algorithms, such as Thompson Sampling Efficient Multiobjective Optimisation (TSEMO), Pareto Efficient Global Optimisation (parEGO), etc. In this work, we introduce a sparse, axis-aligned subspace BO framework that utilises the q-expected hypervolume improvement (qEHVI) acquisition function. This efficient, parallel variant of the expected hypervolume-based method is employed for the multi-objective optimisation of physical properties in multi-component alloys. We demonstrate the impressive performance of the qEHVI acquisition function in finding the optimum Pareto front in both 1- and 2-objective optimisation problems. Furthermore, by conducting several computational and experimental campaigns in each iteration within the pool-based active learning protocol, we can further accelerate the discovery process. This approach is particularly advantageous for use in massively parallel high-throughput synthesis facilities and computer architectures.

In this work, we present an anisotropic multi-goal error control based on the Dual Weighted Residual (DWR) method for time-dependent convection-diffusion-reaction (CDR) equations. This multi-goal oriented approach allows for an accurate and efficient error control with regard to several quantities of interest simultaneously. Using anisotropic interpolation and restriction operators, we obtain elementwise error indicators in space and time, where the spatial indicators are additionally separated with respect to the single directions. The directional error indicators quantify anisotropy of the solution with respect to the goals, and produce adaptive, anisotropic meshes that efficiently capture layers. To prevent spurious oscillations the streamline upwind Petrov-Galerkin (SUPG) method is applied to stabilize the underlying system in the case of high Péclet numbers. Numerical examples show efficiency and robustness of the proposed approach for several goal quantities using established benchmarks for convection-dominated transport.

We provide an error analysis for the solution of the nonstationary Stokes problem by a variational method in space and time. We use finite elements of higher order for the approximation in space and a Galerkin-Petrov method with first order polynomials for the approximation in time. We require global continuity of the discrete velocity trajectory in time, while allowing the discrete pressure trajectory to be discontinuous at the endpoints of the time intervals. We show existence and uniqueness of the discrete velocity solution, characterize the set of all discrete pressure solutions and prove an optimal second order estimate in time for the pressure error in the midpoints of the time intervals. The key result and innovation is the construction of approximations to the pressure trajectory by means of post-processing together with the proof of optimal order error estimates. We propose two variants for a post-processed pressure within the set of pressure solutions based on collocation techniques or interpolation. Both variants guarantee that the pressure error measured in the L2-norm converges with optimal second order in time and optimal order in space. For the discrete velocity solution, we prove error estimates of optimal order in time and space. We present some numerical tests to support our theoretical results.

We study the numerical approximation of a coupled hyperbolic-parabolic system by a family of discontinuous Galerkin space-time finite element methods. The model is rewritten as a first-order evolutionary problem that is treated by the unified abstract solution theory of R. Picard. For the discretization in space, generalizations of the distribution gradient and divergence operators on broken polynomial spaces are defined. Since their skew-selfadjointness is perturbed by boundary surface integrals, adjustments are introduced such that the skew-selfadjointness of the first-order differential operator in space is recovered. Well-posedness of the fully discrete problem and error estimates for the discontinuous Galerkin approximation in space and time are proved.

We present a monolithic $hp$ space-time multigrid method for tensor-product space-time finite element discretizations of the Stokes equations. Geometric and polynomial coarsening of the space-time mesh is performed, and the entire algorithm is expressed through rigorous mathematical mappings. For the discretization, we use inf-sup stable pairs $\mathbb Q_{r+1}/\mathbb P_{r}^{\text{disc}}$ of elements in space and a discontinuous Galerkin (DG$(k)$) discretization in time with piecewise polynomials of order $k$. The key novelty of this work is the application of $hp$ multigrid techniques in space and time, facilitated and accelerated by the matrix-free capabilities of the deal.II library. While multigrid methods are well-established for stationary problems, their application in space-time formulations encounter unique challenges, particularly in constructing suitable smoothers. To overcome these challenges, we employ a space-time cell-wise Vanka smoother. Extensive tests on high-performance computing platforms demonstrate the efficiency of our $hp$ multigrid approach on problem sizes exceeding a trillion degrees of freedom (dofs), sustaining throughputs of hundreds of millions of dofs per second.

The numerical approximation of dynamic poroelasticity, modeling flow in deformable porous media, by a family of continuous space-time finite element methods is investigated. Equal order approximation in space without any further stabilization is used for the displacement and pore pressure variable. Optimal order L^∞(L^2) error estimates are proved and numerically confirmed.

This work presents a numerical investigation of different approximation techniques for the temporal weights used in the Dual Weighted Residual (DWR) method applied to a time-dependent convection-diffusion equation which is assumed to be convection-dominated. It is a continuation of a previous work by the authors where spatial weights were compared for a steady-state case. A higher-order finite elements approach is compared to a more cost-efficient higher-order reconstruction approach. Numerical examples point out the results regarding accuracy, efficiency and stability reasons.

We consider the fully dynamic Biot-Allard model, which includes memory effects. Convolution integrals in time model the history of the porous medium. We use a series representation of the dynamic permeability in the frequency domain to rewrite the equations in a coupled system without convolution integrals, suitable for the design of efficient numerical approximation schemes. The main result is the well-posedness of the system, proved by the abstract theory of R. Picard for evolutionary problems.

We consider Biot’s equations of poroelasticity where the development of viscous boundary layers in the pores is allowed for by using a dynamic permeability convolution operator in the time domain. This system with memory effects is also referred to as the dynamic Biot–Allard model. We use a series representation of the dynamic permeability in the frequency domain to rewrite the equations in the time domain in a coupled system without convolution integrals, which is also suitable for designing efficient numerical approximation schemes. The main result here is the well-posedness of the system, rewritten in evolutionary form, which is proved by an abstract theory for evolutionary problems.

This paper presents a mathematical foundation for physical models in nonlinear optics through the lens of evolutionary equations. It focuses on two key concepts: well-posedness and exponential stability of Maxwell equations, with models that include materials with complex dielectric properties, dispersion, and discontinuities. We use a Hilbert space framework to address these complex physical models in nonlinear optics. While our focus is on the first-order formulation in space and time, higher solution regularity recovers and equates to the second-order formulation. We incorporate perfectly matched layers (PMLs), which model absorbing boundary conditions, to facilitate the development of numerical methods. We demonstrate that the combined system remains well-posed and exponentially stable. Our approach applies to a broad class of partial differential equations (PDEs) and accommodates materials with nonlocal behavior in space and time. The contribution of this work is a unified framework for analyzing wave interactions in advanced optical materials.

We present an approach for solving optimal Dirichlet boundary control problems of nonlinear optics by using deep learning. For computing high resolution approximations of the solution to the nonlinear wave model, we propose higher order space-time finite element methods in combination with collocation techniques. Thereby, $C^{l}$-regularity in time of the global discrete is ensured. The resulting simulation data is used to train solution operators that effectively leverage the higher regularity of the training data. The solution operator is represented by Fourier Neural Operators and Gated Recurrent Units and can be used as the forward solver in the optimal Dirichlet boundary control problem. The proposed algorithm is implemented and tested on modern high-performance computing platforms, with a focus on efficiency and scalability. The effectiveness of the approach is demonstrated on the problem of generating Terahertz radiation in periodically poled Lithium Niobate, where the neural network is used as the solver in the optimal control setting to optimize the parametrization of the optical input pulse and maximize the yield of $0.3\,$THz-frequency radiation. We exploit the periodic layering of the crystal to design the neural networks. The networks are trained to learn the propagation through one period of the layers. The recursive application of the network onto itself yields an approximation to the full problem. Our results indicate that the proposed method can achieve a significant speedup in computation time compared to classical methods. A comparison of our results to experimental data shows the potential to revolutionize the way we approach optimization problems in nonlinear optics.

We present and analyze computationally Geometric MultiGrid (GMG) preconditioning techniques for Generalized Minimal RESidual (GMRES) iterations to space-time finite element methods (STFEMs) for a coupled hyperbolic–parabolic system modeling, for instance, flow in deformable porous media. By using a discontinuous temporal test basis, a time marching scheme is obtained. Higher order approximations that offer the potential to inherit most of the rich structure of solutions to the continuous problem on computationally feasible grids increase the block partitioning dimension of the algebraic systems, comprised of generalized saddle point blocks. Our V-cycle GMG preconditioner uses a local Vanka-type smoother. Its action is defined in an exact mathematical way. Due to nonlocal coupling mechanisms of 348 unknowns, the smoother is applied on patches of elements. This ensures damping of higher order error frequencies. By numerical experiments of increasing complexity, the efficiency of the solver for STFEMs of different polynomial order is illustrated and confirmed. Its parallel scalability is analyzed. Beyond this study of classical performance engineering, the solver’s energy efficiency is investigated as an additional and emerging dimension in the design and tuning of algorithms on the hardware.

We study the numerical approximation by space-time finite element methods of a coupled hyperbolic-parabolic system modeling, for instance, poro- and thermoelasticity. The equations are rewritten as a first-order system in time. Discretizations by continuous Galerkin methods in time and inf-sup stable pairs of finite element spaces for the spatial variables are investigated. Optimal order error estimates are proved for the first-order energy of the system's variables. An important ingredient of the estimates is the control of the coupling terms by a tailored testing strategy. The techniques developed here can be generalized to other families of space-time finite element discretizations and related models. The error estimates are confirmed by numerical experiments, also for higher order piecewise polynomials in space and time. The latter lead to algebraic systems with complex block structure and put a facet of challenge on the design of iterative solvers. An efficient solution technique is referenced.