Mean Field Control and Disorder for Efficient Mixing of Energy Loads

Virtual Informal Systems Seminar (VISS) Centre for Intelligent Machines (CIM) and Groupe d'Etudes et de Recherche en Analyse des Decisions (GERAD)

Speaker: David Métivier

Abstract: 

The introduction of renewable energies in power systems has forced electricity management to become more flexible both for generation and consumption. Demand Response (DR) is a control strategy aiming to address this challenge by adapting and controlling in real-time resources to the demand. Ensembles of cycling electrical devices are good candidates to be used in DR schemes. Approached from the standpoint of Statistical Physics, an ensemble of cycling devices represents a non-equilibrium system driven away from its natural steady-state by DR perturbations. After introducing the model describing large aggregate of devices via coupled Fokker-Planck equations, we will explore the following points: i) how randomness makes the system resilient, mixing it toward a steady-state; ii) how a Mean-Field Control, simple and private, can be implemented to improve the ensemble resiliency even more; iii) how disorder (variability) in the devices ensemble affects the system.

Bio:
David Métivier received his PhD in 2017 at University of Nice Côte d'Azur in France working on the dynamics of mean field particles systems. He did a 3 year postdoc at Los Alamos National Laboratory where he used methods from statistical physics (and mean field dynamics) applied to sustainable devlopment thematics. In particular he worked on the dynamics of thermostatic controlled loads and uncertainty quantification in electrical networks. He is since September a postdoc at the CMAP (applied mathematics) of École Polytechnique in France where he work on the resiliency of the French production system against climate change.