Estimados alumnos,
Están todos cordialmente invitados a una nueva Charla de Postgrado, la que se realizará el próximo martes 14 de Septiembre a las 13:00 hrs.
Relator: PABLO MARTÍN POBLETE DURRUTY, alumno de Doctorado del área Energía, profesor supervisor Javier Pereda.
Titulo: "Predictive Control for Modular Multilevel Converters".
Resumen: The modular multilevel converter (MMC) has been under intense research and development for the last decades due to its high scalability, fault tolerance capacity and low harmonic distortion. Accordingly, the MMC has received a wide acceptance in industrial applications such as high-voltage direct current systems (HVDC) and high power electric drives. This presentation will cover the modelling of the MMC, along with suitable control strategies for governing its multiple internal states. In addition, part of the experimental work carried out in the PecLab on new predictive control strategies with a reduced computational burden for the MMC will be presented. These strategies provide desirable advantages such as a fast dynamic response and flexibility for prioritizing the MMC size/weight reduction or energy conversion efficiency. Consequently, the proposed control scheme can be a promising alternative to govern MMCs, offering high performance and full scalability.
Relator: GERMÁN EDUARDO PIZARRO LORCA, alumno de Magíster del área Automatización, Instrumentación y Robótica, profesor supervisor Felipe Núñez.
Titulo: "Coordination of Autonomous Vehicles: Graph-based Distributed Motion Planning in Tight Multi-lane Platoons".
Resumen: Platooning of connected and automated vehicles (CAVs) has received extensive interest due to its potential to improve road traffic. In this context, multi-lane platoons have appeared as a generalized version of the classical train-like platoon structure used in the early platoon implementations from the 80s. Multi-lane platoons add structural flexibility, at the price of more intricate internal dynamics. This work addresses motion planning in tight multi-lane platoons using distributed techniques that make use of a graph theoretical formulation of platoon formations, for reference generation and distributed non-linear model predictive control, for vehicle command. The effectiveness of the proposed approaches is tested by simulating a variety of scenarios. Finally, the presentation ends with a discussion about the future lines of research: Distributed Control and Multi-Agent Reinforcement Learning interaction to improve road throughput.
Los esperamos a las 13:00 hrs. Si quieres participar, favor inscríbete en el adjunto.
Unirse a la reunión Zoom: zoom.com/my/abusleme
Atentos saludos,
MATIAS NEGRETE