JavaScript is Disabled
Your browser's JavaScript functionality is disabled. It has to be enabled to use this function of ConfTool.
Here you can find information on how to enable JavaScript
If you have any problems, please contact the organizers at karsch@saw-tagungsmanagement.com.

Conference Agenda

The sessions of the sections are highlighted in blue, those of the mini-symposia in yellow.

Please select a date or location to show only sessions at that day or location. If you click the selected day again, you return to the agenda overview.

You can also filter by sections or mini-symposia (topics). Please select a single session for detailed view with abstracts.

As participant you can create your own personal agenda. To do so, log into your account first. Then go to the agenda and click on the plus symbol to add sessions to your personal agenda.

Daily Overview

Session

OC3: Optimisation and Control

Time:

Tuesday, 08/Sept/2026:

1:30pm - 3:30pm

Session Chair: Behzad Azmi

Location: A702

Presentations

Well-posedness of adaptive sliding mode boundary controlled parabolic systems

Richard Löscher, Paul Mayr

TU Graz, Austria

Sliding mode control is a robust control strategy, especially suited for systems subject to parameter uncertainties and external disturbances. Motivated by the stabilization of the temperature distribution in a metallic beam by thermoelectric modules acting on the boundaries, we study a parabolic problem with adaptive sliding mode boundary control. The resulting closed-loop system leads to nonlinear, non-autonomous, and state-dependent boundary conditions.
While the control method is already employed in laboratory experiments, a rigorous mathematical analysis of the underlying control scheme has so far been missing. In this talk, we establish well-posedness of the closed-loop system and discuss the analytical challenges arising from the adaptive, discontinuous boundary feedback. We further comment on Lyapunov-based stability properties and the relation to the motivating application and numerical results.

Observer-Based Stabilization of Linear Water Waves from Partial Free-Surface Observations

Lucas Perrin¹, Behzad Azmi¹, Pierre Lissy²

1: Universität Konstanz, Germany; 2: CERMICS, ENPC, France

We investigate the stabilization of a linearized water-wave equation from partial observations of the free surface. The model under consideration combines the rectangular-tank formulation introduced by Su, Tucsnak and Weiss [1] with observer-based reconstruction techniques inspired by the recent work of Perrin and Lissy [2].

Starting from the linearized gravity water-wave system, we study a finite-dimensional modal approximation based on the eigenstructure of the Dirichlet-to-Neumann operator. Particular attention is devoted to the construction of Luenberger-type observers using localized free-surface measurements.
Using these variables, we introduce observer-based boundary feedback controls and investigate the convergence properties of the resulting coupled state-observer system. The analysis is complemented by numerical simulations illustrating stabilization, observer convergence, and the possible influence of projected observation operators on low- and high-frequency modes.
[1] Su, Tucsnak and Weiss, Stabilizability properties of a linearized water waves system, 2026

[2] Lissy and Perrin,Theoretical and numerical study of the convergence of Luenberger observers for a linearized water wave model, 2025

Event-triggered control and observer design for infinite-dimensional systems

Nail Baloul¹, Lucie Baudouin², Sylvain Ervedoza³, Amaury Hayat¹, Thibault Liard⁴, Pierre Lissy¹

1: CERMICS, ENPC, Institut Polytechnique de Paris, CNRS; 2: LAAS-CNRS, Université de Toulouse; 3: Institut de Mathématiques de Bordeaux, Université de Bordeaux, CNRS; 4: XLIM, Université de Limoges

We investigate event-triggered observer design for infinite-dimensional dynamical systems governed by partial differential equations. The proposed framework combines a Lyapunov-based analysis with careful estimates on the observer dynamics, allowing us to derive uniform bounds on the state reconstruction error. A central difficulty lies in the coupling between the observer and the triggering mechanism, which may lead to an accumulation of triggering times. We address this issue by establishing bounds that are independent of the triggering sequence, ensuring a strictly positive minimum inter-event time.

We consider a class of linear systems for which a Luenberger-type observer is available and propose a dynamic event-triggering mechanism that determine when the control input should be updated.

Unlike approaches that rely on state-dependent triggering rules, our conditions are formulated using the observer state only, making them implementable in practice when full-state measurements are unavailable.

The dynamic rule incorporates an internal variable that provides additional flexibility in preventing event accumulation.

Under suitable assumptions, we prove exponential stability of the closed-loop system and we give sufficient conditions for the triggering times not to accumulate on finite-time horizon excluding Zeno behaviour.

Furthermore, the operator generating the system dynamics is not assumed to be skew-adjoint, which broadens the applicability of our approach to systems with quasi-dissipative or non-conservative behaviour.

Beyond the theoretical contribution, we mention how these results provide a natural and efficient tool for real-time state reconstruction and control of large-scale dynamical systems, in particular transportation networks.

Stabilization of Stochastic Parabolic Equations by Finite-Dimensional Feedback

Jonas von der Heydt¹, Behzad Azmi¹, Sérgio Rodrigues²

1: Dep. Math. Stat., Univ. Konstanz, Germany; 2: Inst. Comput. Appl. Math., öAW, Austria

We present recent results on stabilizing nonlinear, time-dependent stochastic parabolic equations by finite-dimensional feedback. The feedback acts through a finite number of localized indicator-type actuators whose supports may cover an arbitrarily small fraction of the domain and is constructed via oblique projections onto finite-dimensional subspaces. Its offline computation requires only the actuator geometry, independently of the equation's coefficients. Our main result shows that, for any prescribed exponential decay rate $\mu > 0$, a suitable actuator configuration and feedback gain achieve mean-square exponential stabilization, with almost-sure exponential decay in the pure multiplicative noise case. We complement the theoretical findings with numerical experiments illustrating the influence of the number of actuators, noise intensity, and nonlinear effects on stabilization behavior.