11/05/2021 – 16h30-17h30 – Online
Federated stochastic control of numerous heterogeneous energy storage systems
Abstract. We propose a stochastic control problem to control cooperatively Thermostatically Controlled Loads (TCLs) to promote power balance in electricity networks. We develop a method to solve this stochastic control problem with a decentralized architecture, in order to respect privacy of individual users and to reduce both the telecommuni- cations and the computational burden compared to the setting of an omniscient central planner. This paradigm is called federated learning in the machine learning community, therefore we refer to this problem as a federated stochastic control problem. The optimality conditions are expressed in the form of a high-dimensional Forward-Backward Stochastic Differential Equation (FBSDE), which is decomposed into smaller FBSDEs modeling the optimal behaviors of the aggregate population of TCLs of individual agents. In particular, we show that these FBSDEs fully characterize the Nash equilibrium of a stochastic Stackelberg differential game. In this game, a coor- dinator (the leader) aims at controlling the aggregate behavior of the population, by sending appropriate signals, and agents (the followers) respond to this signal by optimizing their storage system locally. A mean-field-type approxi- mation is proposed to circumvent telecommunication constraints and privacy issues. Convergence results and error bounds are obtained for this approximation depending on the size of the population of TCLs. A numerical illustration is provided to show the interest of the control scheme and to exhibit the convergence of the approximation. An implementation which answers practical industrial challenges to deploy such a scheme is presented and discussed.
Joint work with M. Grangereau.
Biography. Emmanuel Gobet is graduated from Ecole Polytechnique, Paris and got a PhD degree in probability and statistics at University Paris Diderot. He took different positions at University Pierre et Marie Curie, Grenoble Institute of Technology and he is currently Professor in applied mathematics at Ecole Polytechnique. His main topics of research are related to financial mathematics, stochastic modelling, Monte Carlo methods, machine learning, energy management. He has written about 80 papers in international journals, 3 books, supervised 15 PhD thesis, he is used to collaborate with private companies. He is the scientific coordinator of the Chair « Stress Test » with BNP Paribas.
Practical details. The seminar will take place online, using Microsoft Teams. If you want to attend this seminar (or any of the forthcoming online Automatiques seminars), and if you do not already have access to the AutoSeminar group on Teams, simply send an email and you will be invited. Please specify which email address the invitation must be sent to (this has to be the address associated with your Teams account). You will find the link to the seminar on the « General » AutoSeminar channel on Teams, approximately 15 minutes before the beginning.
The technical side of things: you can use Teams either directly from you web browser or using the « fat client », which is available for most platforms (Windows, Linux, Mac, Android & iOS). We strongly recommend the latter option whenever possible. Please give it a try before the seminar to anticipate potential problems