A hybrid quantum-classical framework for computational fluid dynamics

TL;DR

This paper proposes a hybrid quantum-classical computational framework for fluid dynamics, integrating quantum linear algebra solvers with classical CFD methods to enhance computational efficiency and scalability.

Contribution

It introduces a novel hybrid CFD framework combining quantum and classical computing, enabling scalable and precise solutions for fluid dynamics simulations.

Findings

Quantum linear solvers provide high-precision solutions.

The framework demonstrates feasibility through typical case studies.

Seamless integration of quantum algorithms into classical CFD is achieved.

Abstract

Great progress has been made in quantum computing in recent years, providing opportunities to overcome computation resource poverty in many scientific computations like computational fluid dynamics (CFD). In this work, efforts are made to exploit quantum potentialities in CFD, and a hybrid classical and quantum computing CFD framework is proposed to release the power of current quantum computing. In this framework, the traditional CFD solvers are coupled with quantum linear algebra libraries in weak form to achieve collaborative computation between classical and quantum computing. The quantum linear solver provides high-precision solutions and scalable problem sizes for linear systems and is designed to be easily callable for solving linear algebra systems similar to classical linear libraries, thus enabling seamless integration into existing CFD solvers. Some typical cases are performed to validate the feasibility of the proposed framework and the correctness of quantum linear algorithms in CFD.