Multiscale modelling banner

Multiscale modelling combines different simulation techniques in order to model materials at different scales, both in terms of size and time.

Simulations performed at an atomistic scale are unable to model certain material properties that manifest at macroscopic scales over relatively long periods of time (microseconds), even when using the largest supercomputers. However, atomic simulations can return properties and information that can be used to inform the creation of finite element models that are designed to simulate these large time and size scales.

There is a range of methods that can be utilised, starting with quantum mechanical simulations, then atomic empirical potentials, followed by more coarse grained potentials that rely on rigid body approximations, to finite element models where materials are simulated based upon a continuum description of material properties. For example, the influence of dopants, and reactions can be investigated by atomic simulations, while the magnetic and electronic response of capacitor materials can be simulated using finite element methods.

Machine learning