Filters

2017 - 201912020 - 20221

More filters

2019 - 201912020 - 20221
Reset filters

Settings

Author: Hintermüller, Michael × Subject: Optimal control ×

Search Results

Now showing 1 - 2 of 2
  • Thumbnail Image
    Item
    Optimization with learning-informed differential equation constraints and its applications
    (Les Ulis : EDP Sciences, 2022) Dong, Guozhi; Hintermüller, Michael; Papafitsoros, Kostas
    Inspired by applications in optimal control of semilinear elliptic partial differential equations and physics-integrated imaging, differential equation constrained optimization problems with constituents that are only accessible through data-driven techniques are studied. A particular focus is on the analysis and on numerical methods for problems with machine-learned components. For a rather general context, an error analysis is provided, and particular properties resulting from artificial neural network based approximations are addressed. Moreover, for each of the two inspiring applications analytical details are presented and numerical results are provided.
  • Thumbnail Image
    Item
    On the consistency of Runge-Kutta methods up to order three applied to the optimal control of scalar conservation laws
    (Berlin : Weierstraß-Institut für Angewandte Analysis und Stochastik, 2017) Hintermüller, Michael; Strogies, Nikolai
    Higher-order Runge-Kutta (RK) time discretization methods for the optimal control of scalar conservation laws are analyzed and numerically tested. The hyperbolic nature of the state system introduces specific requirements on discretization schemes such that the discrete adjoint states associated with the control problem converge as well. Moreover, conditions on the RK-coefficients are derived that coincide with those characterizing strong stability preserving Runge-Kutta methods. As a consequence, the optimal order for the adjoint state is limited, e.g., to two even in the case where the conservation law is discretized by a third-order method. Finally, numerical tests for controlling Burgers equation validate the theoretical results.