Physics-informed neural networks (PINNs) have emerged as promising methods for solving partial differential equations (PDEs) by embedding physical laws within neural architectures. However, these classical approaches often require a large number of parameters to achieve reasonable accuracy, particularly for complex PDEs. In this paper, we present a quantum-classical PINNs (QCPINNs) that combines quantum and classical components, allowing us to solve PDEs with significantly fewer parameters while maintaining comparable accuracy and convergence to classical PINNs. We systematically evaluated two quantum circuit architectures across various configurations on five benchmark PDEs to identify optimal QCPINN designs. Our results demonstrate that the QCPINN achieves stable convergence and comparable accuracy, while using only 10%–30% of the trainable parameters required by classical PINNs. This approach also results in a significant reduction in the relative L2 error for Helmholtz, Klein–Gordon, and Convection–diffusion equations, with reduction ranging from 4% to 64% across various fields. These findings demonstrate the potential of parameter efficiency and solution accuracy in physics-informed machine learning, allowing for a substantial decrease in model complexity without compromising solution quality.
We introduce neural network architectures that combine physics-informed neural networks (PINNs) with the immersed boundary method (IBM) to solve fluid-structure interaction (FSI) problems. Our approach features two distinct architectures: a Single-FSI network with a unified parameter space, and an innovative Eulerian-Lagrangian network that maintains separate parameter spaces for fluid and structure domains. We study each architecture using standard Tanh and adaptive B-spline activation functions. Empirical studies on a 2D cavity flow problem involving a moving solid structure show that the Eulerian-Lagrangian architecture performs significantly better. The adaptive B-spline activation further enhances accuracy by providing locality-aware representation near boundaries. While our methodology shows promising results in predicting the velocity field, pressure recovery remains challenging due to the absence of explicit force-coupling constraints in the current formulation. Our findings underscore the importance of domain-specific architectural design and adaptive activation functions for modeling FSI problems within the PINN framework.
This study investigates the hemodynamic impacts of localized aortic valve calcification, utilizing immersed boundary-finite element (IBFE) method simulations with realistic inflow patterns of uniform and helical blood flow from the left ventricular outflow tract (LVOT). We modeled the aortic valve leaflets with varying grades of calcification, assessing their influence on valve performance, including transvalvular hemodynamics, wall shear stress (WSS) indices, and vortical structures. The findings highlighted that calcification significantly restricts leaflet motion, diminishes the orifice area, disrupts flow efficiency, and consequently increases the left ventricular workload. Advanced calcification resulted in elevated WSS, especially at the leaflet tips, which indicates a heightened risk of endothelial damage and further calcification. Asymmetrical calcifications redirect flow towards the ascending aorta wall, potentially inducing structural damage and increased stress on the remaining healthy leaflets. Calcification was also found to alter the naturally occurring helical blood flow patterns, affecting the system’s fluid transport efficiency and possibly contributing to cardiovascular disease progression. The study revealed a significant alteration in vortex formation, with calcification causing distorted and complex vortex structures, which may influence the dynamics of blood flow and valve function. These insights into the hemodynamic changes induced by calcification contribute to a better understanding of the progression of aortic valve diseases and could inform more effective diagnostic and treatment strategies.
28 September 2022 in Numerical Linear Algebra with Applications