A Quantum Advantage Roadmap for Multi-Physics Simulations

Recently, the European quantum simulation company Quanscient announced the achievement of a major milestone in quantum-native multiphysics simulation, a milestone that Quanscient says marks the dawn of a new era in multiphysics simulation [1].

By quantum-native, we mean that the algorithm encodes the physics of the original problem in some sense directly into the quantum system. That is, in quantum-native simulations, we have a clear and direct analogy between the evolution of the quantum system and the process it is simulating.

In this study, they successfully ran computational fluid dynamics (CFD) simulations on a real quantum computer, using their quantum-native quantum lattice Boltzmann method (QLBM) algorithm to solve the one-dimensional advection diffusion equation with good accuracy. "We have fairly accurate results, not just noise."

Comparison of the one-dimensional advection-diffusion equations solved on a real quantum computer (Quantinuum Model H1-1) with ideal simulations of the Qiskit Aer and simulated H1-1 devices. These are preliminary results obtained using a circuit that is not fully optimized.

Although this is a small 1D problem with 16 computational data points, it marks a beginning. Today's NISQ devices can run macroscale physics simulations natively using Quanscient's quantum-native approach.

The question is, how far can we go?

To this end, Quanscient proposes a roadmap for quantum dominance of multiphysics simulations. This roadmap consists of six specific steps, of which the first two milestones have already been reached.

1) A prototype QLBM solver running on a quantum simulator √

2) Concrete evidence of quantum intrinsic macroscopic physics simulations on NISQ devices√

3) Extension to 2D and 3D simulations on real quantum computers

4) Quantum-native simulations scaled to be on par with the best classical hardware

5) Quantum acceleration

6) Simulation scales that cannot be solved on classical hardware

While this is an ongoing effort, each milestone marks a major breakthrough in quantum multiphysics simulations.

Quanscient says that their algorithm is exponentially related to the number of quantum bits, which means that if we have 100 quantum bits, the number of physics computation points that can be simulated is around 2^100. In principle, we could solve amazingly large systems right now.

Quanscient says they don't produce quantum hardware and don't have much to say about the error rate of the upcoming NISQ device. "All we can influence is how sensitive our algorithms are to noise: implementing deeper circuits and how to handle quantum bit connections."

Reference link：

https://quanscient.com/blog/2022/08/11/the-milestones-to-quantum-advantage-in-quantum-native-multiphysics-simulations/#what-resources-does-it-take