BQP demonstrates the feasibility of large-scale fluid dynamics simulations with quantum computing

Internal compression

• BQP announced a major milestone in computational fluid dynamics (CFD) simulations using a hybrid quantum-classical solution, achieving a jet engine simulation with just 30 logical qubits.
• The BQPhy® platform demonstrated that quantum computing can outperform classical methods, which required 19.2 million compute cores, providing engineers with greater performance and accuracy.
• Advances in BQP may democratize large-scale CFD simulations with applications in aerospace and other fields.

Press Release – BQP, a startup leading the development of quantum-based engineering simulations, today announced a significant research milestone for computational fluid dynamics (CFD) simulations. This milestone was achieved using a hybrid quantum-classical solver that is part of BQP’s next-generation simulation platform, BQPhy®.

After conducting nearly 100,000 experiments, the BQP researchers published their work in a paper where they estimated that a large-scale CFD simulation could be achieved with 30 logical qubits on a quantum computer, leading to better accuracy, efficiency and costs. Current methods. A previous study that inspired the BQP team to undertake this research found that 19.2 million compute cores were needed to perform the same simulation with classical algorithms on state-of-the-art high-performance computers (HPCs).

As quantum computers become utility-scale, this research is critical to democratizing large-scale CFD simulation for every engineer,” said Abhishek Chopra, founder, CEO and chief scientific officer of BQP. “In the future, engineers will have easy access to—19.2 million HPC cores or 30-logical-qubit quantum Computers? I bet the latter.

“With continued research, we believe quantum computing has the potential to revolutionize the way simulations are conducted, allowing engineers to push the boundaries of design and engineering,” added Chopra.

“BQP’s results represent the introduction of rigorously high computing power to flow field analysis and simulation. This capability opens up new methods in aerospace development, enabling greater confidence during design and more proactive maintenance during the aircraft life cycle,” said Don Hart, senior aerospace administrator and member of the National Academy of Engineering. said..

For the research, BQP scientists evaluated the sociability, accuracy and stability of jet engine simulations using BQP’s Hybrid Quantum Classical Finite Method (HQCFM) solver. The study demonstrated the scalability of the HQCFM solver by simulating a nonlinear, time-dependent partial differential equation (PDE) from 4 qubits to 11 qubits.

The researchers found that accuracy and stability were comparable to classical computers, while HQCFM distinguished itself by running within a time loop on an unsteady problem, never propagating any error to the next time step. Achieving such high accuracy consistently is a significant advance towards more complex simulations beyond the capabilities of classical devices.

BQP believes that BQPhy’s solution will allow CFD engineers to simulate an entire aircraft for the first time, greatly improving flight patterns during turbulence. Based on current trends in supercomputing computational advances, simulating an entire aircraft with classical computing may not be possible until 2080.

BQPhy’s physics-based solution can also be used to solve other PDEs to capture interactions in gas dynamics, traffic flow, or flood tides in rivers. Combined with quantum algorithms, the technology can solve complex equations with reduced hardware demands compared to traditional high-performance computing (HPC) methods, while enabling efficient execution of sophisticated and complex simulations.

“Building on our successful collaborations with leading academic institutions, government research organizations such as AFRL, DARPA, industry pioneers and top academic institutions, BQP is excited to partner with organizations that share our vision of advancing quantum computing solutions,” said Chopra.