Ulteriori informazioni
This book contains the proceedings of the Symposium on Systems Theory in Data and Optimization (SysDO) held in Stuttgart, Germany, from 30th September to 2nd October 2024. It addresses theoretical and practical research questions arising at the intersection of systems and control theory, data, and optimization. The increasing prevalence of cyber-physical systems sparks the need for new methods to integrate measured data and different forms of feedback, especially optimization-based feedback, inside the decision-making mechanism. There are distinctive challenges that arise in this scenario, such as the existence of different time-scales, the need to guarantee sufficient richness of the collected data, and the effect of suboptimal decisions under uncertainty. This book presents new methods and applications addressing these challenges.
This book is a valuable source on latest research findings spanning diverse topics including:
- data-driven and learning-based control;
- theory and applications of machine learning;
- model predictive control; and
- optimization.
Sommario
Part I. Data-Driven and Learning-Based Control.- Chapter 1. PACSBO: Probably Approximately Correct Safe Bayesian Optimization.- Chapter 2. Value of Communication: Data-Driven Topology Optimization for Distributed Linear Cyber-Physical Systems.- Chapter 3. Variance-Informed Model Reference Gaussian Process Regression: Utilizing Variance Information for Control in Nonlinear Systems.- Chapter 4. Data-Driven Dynamic Model and Model Reference Control of Inverter Based Resources.- Chapter 5. Adaptive Tracking MPC for Nonlinear Systems via Online Linear System Identification.- Part II: Machine Learning: Theory and Applications.- Chapter 6. Investigation of the Influence of Training Data and Methods on the Control Performance of MPC Utilizing Gaussian Processes.- Chapter 7. Wiener Chaos in Kernel Regression: Towards Untangling Aleatoric and Epistemic Uncertainty.- Chapter 8. A Universal Reproducing Kernel Hilbert Space for Learning Nonlinear Systems Operators.- Chapter 9. On Robust Reinforcement Learning with Lipschitz-Bounded Policy Networks.- Chapter 10. Solving Partial Differential Equations with Equivariant Extreme Learning Machines.- Chapter 11. Adaptive Robust L2 Loss Function using Fractional Calculus.- Chapter 12. Sparse Reconstruction of Forces, Torques and Velocity Signals for a Swimmer in a Wake.- Chapter 13. Control Theoretic Approach to Fine-Tuning and Transfer Learning.- Part III. Model Predictive Control.- Chapter 14. Accelerating Multi-Objective Model Predictive Control Using High-Order Sensitivity Information.- Chapter 15. On Discount Functions for Economic Model Predictive Control without Terminal Conditions.- Chapter 16. Multi-Parametric Programming with Constraint Telaxation for the Optimal Operation of Micro-Grids Integrating Renewables.- Chapter 17. Multi-Objective Learning Model Predictive Control.- Chapter 18. Terminal Set of Nonlinear Model Predictive Control with Koopman Operators.- Part IV: Optimization.- Chapter 19. Optimal Dynamic Pricing in Energy Markets: A Stackelberg Game Approach.- Chapter 20. Distributed Newton Optimization with ADMM-Based Consensus.- Chapter 21. Inexactness in Bilevel Nonlinear Optimization: A Gradient-free Newton's Method Approach.
