[HTML][HTML] SMGO-Δ: Balancing caution and reward in global optimization with black-box constraints

L Sabug Jr, F Ruiz, L Fagiano - Information Sciences, 2022 - Elsevier
In numerous applications across all science and engineering areas, there are optimization
problems where both the objective function and the constraints have no closed-form …

SMGO: A set membership approach to data-driven global optimization

L Sabug Jr, F Ruiz, L Fagiano - Automatica, 2021 - Elsevier
Many science and engineering applications feature non-convex optimization problems
where the objective function cannot be handled analytically, ie it is a black box. Examples …

[HTML][HTML] Iterative learning robust optimization-with application to medium optimization of CHO cell cultivation in continuous monoclonal antibody production

Y Wang, M Pasquini, V Chotteau, H Hjalmarsson… - Journal of Process …, 2024 - Elsevier
In the presence of uncertainty, the optimum obtained based on a nominal identified model
can neither provide any performance guarantee nor ensure that critical constraints are …

Trading-off safety, exploration, and exploitation in learning-based optimization: a Set Membership approach

L Sabug, F Ruiz, L Fagiano - 2021 60th IEEE Conference on …, 2021 - ieeexplore.ieee.org
We propose a technique for global optimization considering black-box cost function and
constraints, which have to be learned from data during the optimization process, arising for …

Direct control design using a Set Membership-based black-box optimization approach

R Galbiati, L Sabug, F Ruiz… - 2022 IEEE Conference …, 2022 - ieeexplore.ieee.org
The problem of controller tuning is a challenging question for practitioners, especially when
industrial plants are involved. In such plants, different physical mechanisms come into play …

Robust optimization with optimal experiment design-with application to continuous biopharmaceutical production

Y Wang, M Pasquini, V Chotteau, H Hjalmarsson… - IFAC-PapersOnLine, 2022 - Elsevier
Abstract Model-based optimization typically obtains the optimum based on a nominal
identified model. However, in the presence of uncertainty, the nominal optimum leads to …

Fast Calculation of Cross Ambiguity Function for Passive Radar Based on SMGO

D Zhang, C Jiang - 2021 China Automation Congress (CAC), 2021 - ieeexplore.ieee.org
For passive radar system, the distance and speed of target can be determined by locating
the peak of cross ambiguity function between the reference signal and the echo signal …

Global optimization of pulse patterns for an electrical drive via Set Membership methods

G Montecchio, M Alborghetti - 2022 - politesi.polimi.it
The optimization of pulse patterns is a crucial problem in the field of modulation techniques
in electrical drives. This problem is challenging due to its high dimensionality and non …