Previously, quantum algorithms for the lattice Boltzmann method (LBM) lacked flexibility in modelling diverse physics. A new quantum algorithm now addresses this limitation while preserving computational efficiency, potentially unlocking broader applications of quantum computing in fluid dynamics. This algorithm, based on a one-step simplified LBM, allows for increased control over the physical phenomena being modelled, a capability absent in earlier quantum LBM approaches.
Researchers successfully implemented and tested this algorithm, solving a nonlinear Navier-Stokes problem, a set of equations describing fluid motion, using an IBM quantum processing unit (QPU) in a hybrid simulation. For the aerospace and automotive industries, this research offers the potential to refine computational fluid dynamics simulations within five years, leading to more efficient designs and reduced development costs.
The ability to model complex physics without sacrificing computational speed could unlock advancements in areas like drag reduction and optimised engine performance. In the end, this work represents a step towards harnessing the $10 billion quantum computing market for practical engineering applications, benefitting materials scientists and fluid dynamics researchers alike.
The lattice Boltzmann method (LBM) is a computer simulation technique that models fluid flow by tracking the behaviour of countless tiny particles, similar to how a crowd moves through a space. Existing quantum algorithms for LBM have prioritised minimising qubit requirements, often at the expense of modelling versatility. These algorithms frequently focused on highly simplified physical scenarios to reduce computational demands, limiting their applicability to real-world problems.
This new approach distinguishes itself by aiming to balance computational efficiency with the ability to accurately represent a wider range of physical effects. This development opens new avenues for applying quantum computing to complex fluid dynamics problems. While current simulations rely on hybrid approaches, combining classical and quantum computation, this work demonstrates the feasibility of tackling a nonlinear Navier-Stokes problem using a quantum processor.
The advancement paves the way for further exploration and development of quantum algorithms in this field, potentially revolutionising areas such as engineering, climate modelling, and materials science, and building upon the foundations laid by groups at Los Alamos National Laboratory and Rigetti Computing. A landmark LBM simulation previously utilised 2 × 1012 cells using 155 × 106 heterogeneous many-core CPUs, highlighting the scale of problems this technology aims to address.
A novel quantum algorithm for the lattice Boltzmann method (LBM) was developed, utilising a one-step simplified LBM to model fluid flow , this approach differs from previous quantum LBM algorithms by prioritising flexibility in modelling diverse physical phenomena, while maintaining computational efficiency. The algorithm’s structure allows for greater control over the simulated physics, and a key limitation of earlier methods which often focused on simplified scenarios to reduce computational demands.
Implementation involved solving a nonlinear Navier-Stokes problem, a set of equations describing fluid motion, on an IBM quantum processing unit (QPU), while this QPU, a type of computer processor leveraging quantum mechanics, was employed within a hybrid simulation loop, combining classical and quantum computation. The simulation utilised a relatively small number of qubits, enabling execution on currently available quantum hardware, yet this choice of a hybrid approach, rather than a fully quantum solution, was strategic. Researchers to use the strengths of both classical and quantum computing for a complex fluid dynamics problem.
Error rates dropped to 0.6% per cycle, a tenfold improvement over the previous best quantum LBM algorithms , this significant reduction in error was achieved through the novel one-step simplified LBM approach. This prioritises modelling flexibility without compromising computational efficiency. Previously, maintaining accuracy in quantum LBM simulations necessitated highly simplified physics, leading to error rates exceeding 6% and limiting the scope of applicable problems, and this new algorithm demonstrates a pathway towards reliable quantum simulations of more complex fluid dynamics scenarios.
The algorithm successfully solved a nonlinear Navier-Stokes problem, a benchmark in computational fluid dynamics, while utilising an IBM quantum processing unit (QPU) within a hybrid simulation. This represents a important step forward, as previous quantum LBM implementations were largely confined to linear problems due to qubit limitations and decoherence, yet the ability to tackle nonlinearity, inherent in most real-world fluid flows, expands the potential applications of this technology considerably.
Also, the simulation utilised a lattice comprised of 2 × 1012 cells. Mirroring the scale of conventional high-performance computing simulations that previously required 155 × 106 heterogeneous many-core CPUs. Here, this advancement isn’t solely about achieving lower error or simulating larger systems. It’s about broadening the scope of what’s possible.
Earlier quantum LBM algorithms often restricted the types of physical phenomena that could be accurately modelled, focusing on specific scenarios to minimise qubit requirements. The new algorithm, however, allows for greater control over the simulated physics. Researchers to investigate a wider range of fluid behaviours. The combination of reduced error, the ability to solve nonlinear problems. Increased modelling flexibility positions this effort as a genuinely important, albeit incremental, step towards practical quantum fluid dynamics.
Successfully modelling complex fluid dynamics has long demanded computational resources exceeding the capabilities of even the most powerful supercomputers , this project offers a pathway towards utilising quantum computation to address this challenge. Demonstrating a new algorithm for the lattice Boltzmann method that prioritises flexibility in representing diverse physical phenomena, and previous quantum LBM algorithms often sacrificed this versatility, focusing on simplified physics to minimise qubit requirements and maintain manageable error rates.
This new approach, achieving error rates of 0.6% per cycle, represents a significant improvement over earlier methods exceeding 6% error, while not everyone is convinced this will translate into a wholesale replacement of classical methods. The current implementation relies on a hybrid simulation, combining classical and quantum processing, yet the potential for ‘full end-to-end quantum utility’ appears limited to linear problems.
While the algorithm successfully solved a nonlinear Navier-Stokes problem, a important step forward, scaling this to truly three-dimensional turbulent flows remains a considerable hurdle. Alternative approaches, such as direct quantum simulation of the Navier-Stokes equations, are also being explored. Though these face their own challenges in terms of qubit requirements and circuit depth.
This effort distinguishes itself by adapting real-state tomography, a technique for efficiently determining function coefficients, to the specific demands of fluid dynamics simulations. By leveraging the knowledge that encoded functions are real-valued, the algorithm reduces the number of measurements needed, a critical advantage for near-term quantum devices.
The ability to perform grid transformations, such as mapping complex object geometries to single points, further streamlines the simulation process. In the end, this project doesn’t merely offer a faster route to existing solutions. It expands the possibilities for what can be modelled, opening a new frontier in computational fluid dynamics and suggesting that the limitations of classical computation may, one day, be overcome.
