Performance Guarantees in Adaptive Economic Model Predictive Control
Event details
| Date | 29.11.2024 |
| Hour | 11:00 › 12:00 |
| Speaker | Mr. Maximilian Degner, B.Sc. & M.Sc. in Mechanical Engineering, ETH Zürich. |
| Location | |
| Category | Conferences - Seminars |
| Event Language | English |
Abstract:
In this talk, we investigate the performance guarantees of an Adaptive Economic Model Predictive Control (MPC) framework for nonlinear systems that are subject to parametric uncertainty and disturbances. The proposed framework combines online model adaptation with the direct minimization of transient economic costs. We leverage the properties of a least-mean squares parameter estimator, artificial references, and a general robust MPC scheme to provide transient and asymptotic performance guarantees. The benefits of the proposed method are demonstrated in simulation using a typical chemical reactor example.
Biography :
Mr. Maximilian Degner recently graduated from ETH Zürich with a B.Sc. and M.Sc. in Mechanical Engineering. During his studies, he focused on optimization-based control systems and worked on methods that combine conventional reference tracking and economic optimization, such as Online Feedback Optimization and Economic Model Predictive Control. After completing his master’s thesis, he worked on the publication of the final results. His research interests lie at the intersection of online learning and control.
In this talk, we investigate the performance guarantees of an Adaptive Economic Model Predictive Control (MPC) framework for nonlinear systems that are subject to parametric uncertainty and disturbances. The proposed framework combines online model adaptation with the direct minimization of transient economic costs. We leverage the properties of a least-mean squares parameter estimator, artificial references, and a general robust MPC scheme to provide transient and asymptotic performance guarantees. The benefits of the proposed method are demonstrated in simulation using a typical chemical reactor example.
Biography :
Mr. Maximilian Degner recently graduated from ETH Zürich with a B.Sc. and M.Sc. in Mechanical Engineering. During his studies, he focused on optimization-based control systems and worked on methods that combine conventional reference tracking and economic optimization, such as Online Feedback Optimization and Economic Model Predictive Control. After completing his master’s thesis, he worked on the publication of the final results. His research interests lie at the intersection of online learning and control.
Practical information
- General public
- Free
Organizer
- Professor Giancarlo Ferrari Trecate