Photonic Inverse Design using the Adjoint Method
In this webinar, we will explore inverse design – a design process where given desired target performance, computational methods are employed to find the optimal device geometry – to photonics. The promise of such inverse design techniques is that many parameters can be optimized against a set of different manufacturing and operating conditions to ensure that resulting design are truly optimal and robust against, for example, temperature variations and manufacturing imperfections.
We begin with a review of different techniques, including so-called shape-based design where parameterized shapes are specified and optimized, to topology optimization where computational methods are used to find optimal, nonintuitive designs that typically offer superior performance to shape-based methods. Using Lumerical’s nanophotonic simulator FDTD Solutions, we will demonstrate the power of these advanced optimization techniques as applied to typical structures and design goals encountered in designing components for integrated photonics.
February 19, 2019
5:00pm (America/Vancouver -08:00 GMT)