Novel quantum algorithms for computational fluid dynamics: TU Delft and Fujitsu join forces
In a new chapter of an already fruitful collaboration, TU Delft and Fujitsu are bundling their expertise and resources over the next five years to cover new grounds in quantum computational fluid dynamics (QCFD). The project will have a special emphasis on developing novel quantum algorithms for simulating the complex behaviour of flows and testing them in the context of an aerodynamic design optimization workflow.
While the development of physical quantum computers is still in its infancy, with first prototype systems being available in the form of noisy intermediate-scale quantum (NISQ) computers, this projects already targets at the next generation of fault-tolerant quantum computers that will be able to execute much more complex quantum circuits on a few hundreds of logical qubits. With this technology, quantum CFD will be able to demonstrate clear advantage over classical CFD running on conventional high-performance computing systems like DelftBlue.
To leverage the foundational tenets of quantum mechanics, this projects takes a step back from the classical CFD workflow that consists in discretising the Navier-Stokes equations and solving the resulting nonlinear systems of equations iterative, and will focus on the lattice Boltzmann equation, which is more suitable for being solved start to end on a quantum computer.
The project will be led by associate professor Matthias Möller, and will in part entail four new PhD positions. Supported by a broader, multidisciplinary team, this collaboration promises not just incremental improvements but a novel approach toward shaping the future landscape of quantum CFD.