Optimization of PDEs with Uncertain Inputs

Abstract

Uncertainty pervades nearly all science and engineering applications including the optimal control and design of systems governed by partial differential equations (PDEs). In many applications, it is critical to determine optimal solutions that are resilient to the inherent uncertainty in unknown boundary conditions, inaccurate coefficients, and unverifiable modeling assumptions. In this tutorial, we develop a general theory for PDE-constrained optimization problems in which inputs or coefficients of the PDE are uncertain. We discuss numerous approaches for incorporating risk preference and conservativeness into the optimization problem formulation, motivated by concrete engineering applications. We conclude with a discussion of nonintrusive solution methods and numerical examples.