Qifeng Hu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Qifeng Hu has a total of 3 co-authors. Here are the top co-authors by number of shared publications:
- Inanc Senocak
University of Pittsburgh
4 shared publications Shamsulhaq Basir
1 shared publication- Shamsulhaq Basir
Department of Mechanical Engineering and Materials Science University of Pittsburgh 3700 O'Hara St. 15261 Pittsburgh PA USA
1 shared publication
Latest Publications
- Open Access
- 14 May 2026
- Published by Elsevier in SSRN Electronic Journal
Unsupervised simulation of incompressible flows with physics‑ and equality‑constrained artificial neural networks
Physics-informed neural networks (PINNs) have shown considerable promise for solving partial differential equations, yet they have achieved limited success in simulating incompressible flows, particularly at high Reynolds numbers. Existing approaches rely on auxiliary labeled data, supervised pretraining, or reference solutions, and no purely unsupervised method comparable to conventional finite-difference or finite-volume solvers has been demonstrated to date. We attribute this gap to the absence of a robust mechanism for enforcing the divergence-free constraint and boundary conditions to strict tolerances — a capability that is fundamental to the success of conventional incompressible flow solvers. To address this limitation, we adopt the physics- and equality-constrained artificial neural network (PECANN) framework in conjunction with a conditionally adaptive augmented Lagrangian method (CA-ALM), and introduce a pressure-Poisson-based objective function. In the proposed formulation, the residual of the pressure Poisson equation is minimized subject to the momentum and continuity equations and boundary conditions on the primitive variables as equality constraints, with CA-ALM enforcing all constraints to strict tolerances. To enhance robustness for advection-dominated, high-Reynolds-number flows, we further propose an adaptive vanishing entropy viscosity that stabilizes early training without influencing the final predictions. A baseline formulation that omits the pressure Poisson equation and uses the residual of the momentum equations as the objective proves ineffective even with the same constraint-enforcement machinery and network architecture, underscoring the critical role of the proposed pressure-Poisson-based objective function. The method is assessed on several benchmark problems — lid-driven cavity flow at Reynolds numbers up to Re=7,500, three-dimensional unsteady Beltrami flow, and both steady and unsteady flow past a circular cylinder with general inflow–outflow boundary conditions, including an ablation study that identifies admissible conditions on velocity and pressure at the outlet — all without labeled data or supervised pretraining. Most remarkably, the method captures the spontaneous onset of periodic vortex shedding in unsteady cylinder flow without external perturbations, starting from a randomly initialized network.
- 14 May 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessUnsupervised simulation of incompressible flows with physics‑ and equality‑constrained artificial neural networks
Physics-informed neural networks (PINNs) have shown considerable promise for solving partial differential equations, yet they have achieved limited success in simulating incompressible flows, particularly at high Reynolds numbers. Existing approaches rely on auxiliary labeled data, supervised pretraining, or reference solutions, and no purely unsupervised method comparable to conventional finite-difference or finite-volume solvers has been demonstrated to date. We attribute this gap to the absence of a robust mechanism for enforcing the divergence-free constraint and boundary conditions to strict tolerances — a capability that is fundamental to the success of conventional incompressible flow solvers. To address this limitation, we adopt the physics- and equality-constrained artificial neural network (PECANN) framework in conjunction with a conditionally adaptive augmented Lagrangian method (CA-ALM), and introduce a pressure-Poisson-based objective function. In the proposed formulation, the residual of the pressure Poisson equation is minimized subject to the momentum and continuity equations and boundary conditions on the primitive variables as equality constraints, with CA-ALM enforcing all constraints to strict tolerances. To enhance robustness for advection-dominated, high-Reynolds-number flows, we further propose an adaptive vanishing entropy viscosity that stabilizes early training without influencing the final predictions. A baseline formulation that omits the pressure Poisson equation and uses the residual of the momentum equations as the objective proves ineffective even with the same constraint-enforcement machinery and network architecture, underscoring the critical role of the proposed pressure-Poisson-based objective function. The method is assessed on several benchmark problems — lid-driven cavity flow at Reynolds numbers up to Re=7,500, three-dimensional unsteady Beltrami flow, and both steady and unsteady flow past a circular cylinder with general inflow–outflow boundary conditions, including an ablation study that identifies admissible conditions on velocity and pressure at the outlet — all without labeled data or supervised pretraining. Most remarkably, the method captures the spontaneous onset of periodic vortex shedding in unsteady cylinder flow without external perturbations, starting from a randomly initialized network.
- Open Access
- 3 April 2026
- Published by Elsevier in Computer Methods in Applied Mechanics and Engineering
- Vol. 456
Conditionally adaptive augmented Lagrangian method for physics-informed learning of forward and inverse problems
- 3 April 2026
- Published by Elsevier in Computer Methods in Applied Mechanics and Engineering
- Vol. 456
Open AccessConditionally adaptive augmented Lagrangian method for physics-informed learning of forward and inverse problems
- Open Access
Solution of Incompressible Flow Equations with Physics and Equality Constrained Artificial Neural Networks
We present a meshless method for the solution of incompressible Navier-Stokes equations in advection-dominated regimes using physics- and equality-constrained artificial neural networks combined with a conditionally adaptive augmented Lagrangian formulation. A single neural network parameterizes both the velocity and pressure fields, and is trained by minimizing the residual of a Poisson's equation for pressure, constrained by the momentum and continuity equations, together with boundary conditions on the velocity field. No boundary conditions are imposed on the pressure field aside from anchoring the pressure at a point to prevent its unbounded development. The training is performed from scratch without labeled data, relying solely on the governing equations and constraints. To enhance accuracy in advection-dominated flows, we employ a single Fourier feature mapping of the input coordinates. The proposed method is demonstrated for the canonical lid-driven cavity flow up to a Reynolds number of 7,500 and for laminar flow over a circular cylinder with inflow-outflow boundary conditions, achieving excellent agreement with benchmark solutions. We further compare the present formulation against alternative objective-function constructions based on different arrangements of the flow equations, thereby highlighting the algorithmic advantages of the proposed formulation centered around the Poisson's equation for pressure.
Open AccessSolution of Incompressible Flow Equations with Physics and Equality Constrained Artificial Neural Networks
We present a meshless method for the solution of incompressible Navier-Stokes equations in advection-dominated regimes using physics- and equality-constrained artificial neural networks combined with a conditionally adaptive augmented Lagrangian formulation. A single neural network parameterizes both the velocity and pressure fields, and is trained by minimizing the residual of a Poisson's equation for pressure, constrained by the momentum and continuity equations, together with boundary conditions on the velocity field. No boundary conditions are imposed on the pressure field aside from anchoring the pressure at a point to prevent its unbounded development. The training is performed from scratch without labeled data, relying solely on the governing equations and constraints. To enhance accuracy in advection-dominated flows, we employ a single Fourier feature mapping of the input coordinates. The proposed method is demonstrated for the canonical lid-driven cavity flow up to a Reynolds number of 7,500 and for laminar flow over a circular cylinder with inflow-outflow boundary conditions, achieving excellent agreement with benchmark solutions. We further compare the present formulation against alternative objective-function constructions based on different arrangements of the flow equations, thereby highlighting the algorithmic advantages of the proposed formulation centered around the Poisson's equation for pressure.
- Open Access
- 31 December 2024
- Published by Elsevier in Computer Methods in Applied Mechanics and Engineering
- Vol. 436
Non-overlapping, Schwarz-type domain decomposition method for physics and equality constrained artificial neural networks
- 31 December 2024
- Published by Elsevier in Computer Methods in Applied Mechanics and Engineering
- Vol. 436
Open AccessNon-overlapping, Schwarz-type domain decomposition method for physics and equality constrained artificial neural networks