arXiv Science⌕ Search

arXiv · 2610.06990

A density-based topology optimization framework for steady Navier-Stokes flow with design-dependent wall actuation

Abstract

Design-dependent wall velocities are difficult to treat in density-based fluid topology optimization. This difficulty arises because the fluid-solid interface is represented implicitly and evolves with the design. This paper proposes a diffuse-interface formulation for topology optimization of steady incompressible Navier-Stokes flows with design-dependent wall actuation. A continuous topological description field parameterized by compactly supported radial basis functions (CS-RBFs) provides the local interface orientation, and its regularized projection defines the pseudo-density for Brinkman flow analysis. The wall velocity is imposed on a fixed mesh through an equivalent volumetric momentum source, which is localized by a diffuse-interface measure constructed from the pseudo-density and the gradient of the description field. Assuming a locally planar interface and uniformly distributed Wendland radial basis functions, the source coefficient is determined analytically by balancing the equivalent momentum input with the Brinkman resistance at the interface center. These assumptions also yield a characteristic transition-width estimate in terms of the projection steepness and support radius. Continuous adjoint analysis is performed to derive the design sensitivities. Two- and three-dimensional examples formulated using the analogy equation demonstrate the effectiveness of the proposed topology optimization method for flow-rate maximization. Body-fitted re-simulations verify the flow fields and transport performance of the optimized designs. The formulation is further extended to electroosmotic transport under the thin electric double layer (EDL) assumption using the Helmholtz-Smoluchowski relation.

Explore related subjects

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Limin Chen, Yuan Liang, Yongbo Deng, Chong Wang, Youming Zhang. 2026-10-04. A density-based topology optimization framework for steady Navier-Stokes flow with design-dependent wall actuation. https://arxiv.org/abs/2610.06990

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related papers

Quantum-inspired space-time PDE solver and dynamic mode decomposition

The numerical solution of time-dependent PDEs posed over the combined space-time domain, instead of treating time sequentially, leads to space-time methods. Treating time as another spatial dimension allows parallelization across time and simultaneous adaptivity in spatial and temporal discretizations. However, space-time formulations suffer from the curse of dimensionality due to the (d+1)-dimensional domain. Time-series forecasting methods, such as dynamic mode decomposition (DMD), also suffer the same scaling for storing and processing measurement data. In this work, we investigate whether quantum-inspired tensor network methods can mitigate this curse for both solving PDEs and data-driven modeling. We formulate a space-time matrix product state (MPS) encoding to represent either the space-time solution or time-series data. Using this representation, we develop a unified framework that combines an MPS space-time PDE solver with an MPS-DMD algorithm. We benchmark our MPS space-time solver for linear and nonlinear PDEs in both (1+1) and (2+1) dimensions, including non-diffusive regimes such as shockwave formation and Allen-Cahn phase separation. We observe that the MPS ansatz efficiently captures the spatio-temporal correlations of the solutions, with a 99% reduction in parameters with respect to the grid size. We demonstrate the MPS-DMD algorithm in two cases. First, the MPS space-time solver generates the MPS encoding of the time-series data. We then apply MPS-DMD directly to predict the dynamics without solving PDEs. Second, for the general case where data is available as dense arrays, MPS-DMD needs an intermediate compression step. We implement a streaming-like compression algorithm and observe a 97% reduction in memory, with MPS-DMD predictions matching those of exact DMD.

physics.comp-ph↗

CATAPULT: A CUDA-Accelerated Timestepper for Alpha Particles Using Local Tricubics

We introduce a CUDA-Accelerated Timestepper for Alpha Particles Using Local Tricubics (CATAPULT) for use in Monte Carlo calculations of alpha particle confinement in stellarators. Our GPU implementation is significantly faster than CPU-parallelized implementations of the same field representation such as SIMSOPT, and handles both equilibrium magnetic fields and Shear Alfven Waves. We test our implementation on several example stellarators to exhibit both the speed and correctness of our code. The source code is included in the firm3d Python package.

physics.comp-ph↗

Signature of Short-Range Order in Static Response of the Three-Dimensional Electron Gas

The three-dimensional electron gas is a fundamental model in condensed matter physics and quantum chemistry, and the exchange-correlation energy derived from it serves as the starting point of \textit{ab initio} computations of materials. However, the wave-vector-dependent response, and hence the static local field factor $G(q)$, has remained without accurate ground-state benchmark in the strongly coupled regime for three decades. Using diffusion Monte Carlo, we calculate the static density-density response function and $G(q)$ across the liquid phase and find a pronounced structure in $G(q)$ at intermediate wave vectors, already visible at metallic densities and growing with increasing interaction strength. We identify it as a fingerprint of short-range order by showing that it is required to reproduce the static structure factor. Our parametrization, valid in the entire liquid phase, predicts a low-energy resonant mode inside the particle-hole continuum.

physics.comp-ph↗