A fast and accurate simulation methodology enabled by a physics-informed learning method, using Proper Orthogonal Decomposition (POD) and Galerkin projection, is demonstrated for photonic crystals and periodic quantum nanostructure. POD is a projection-based method with its basis functions trained by solution data collected from Direct Numerical Simulations (DNSs) of the wave equation, which offers the best least squares fit to the solution data. The Galerkin projection of the wave equation onto POD basis functions is then performed to close the model. This projection also incorporates physical principles during POD-Galerkin simulation guided by the wave equation, which thus enables the extrapolation capability beyond the training conditions. Such a capability is difficult to achieve using neural-network-based methods for physics simulation, where no physical principle is enforced during simulation. Applications of the POD-Galerkin methodology to a 2D photonic lattice and a 2D periodic quantum nanostructure demonstrate a computing speedup near two orders of magnitude with high accuracy, compared to DNS, if the wave solution and band structure are both needed. If only the band structure is of interest, a four-order improvement in computational efficiency can be achieved.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.