Managing Distributed Flexibility Under Uncertainty by Combining Deep Learning With Duality

Abstract

published version features the final layout of the paper including the volume, issue and page numbers. Link to publication General rightsCopyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights.• Users may download and print one copy of any publication from the public portal for the purpose of private study or research. • You may not further distribute the material or use it for any profit-making activity or commercial gain • You may freely distribute the URL identifying the publication in the public portal.If the publication is distributed under the terms of Article 25fa of the Dutch Copyright Act, indicated by the "Taverne" license above, please follow below link for the End User

Fair and Scalable Electric Vehicle Charging Under Electrical Grid Constraints

“…Updating the decisions of a charging station requires solving problem (23). Note that ( 23) is a MILP which needs to be solved iteratively by each station for every multiplier update.…”

“…With respect to the scalability of Algorithm 1 and the ADMM benchmark, we make the following remark Remark 1: In contrast to the ADMM approach, Algorithm 1 avoids the need to solve the non-convex problem (23) at each update of the multipliers. Instead, in its inner loop, it solves the much faster, approximate linear program (28) for each multiplier update and it only solves MILP (27) after the multipliers have converged, in order to check the choice of δ.…”

“…[20] and [21] propose different iterative auction procedures for allocating energy to users and determining their charging schedules), artificial intelligence (e.g. [22], [23] propose Reinforcement (and respectively Supervised) Learning algorithms for employing a transaction, price-based control of EV scheduling), convex optimization (e.g. in [24], EV users submit a convex utility function that represents their charging preferences, and the operator solves a convex optimization program to decide the charging schedules) and distributed optimization (e.g.…”

Stacking Revenues From Flexible DERs in Multi-Scale Markets Using Tri-Level Optimization

“…Markov Decision Process was used in [11] to stack revenues from energy arbitrage and frequency regulation in PV-Battery Energy Storage System (BESS) systems. Deep-Learning was used in [12] to accelerate the solution of the energy management problem of a community of DERs under uncertainty. Those techniques only address the problem of DERs providing transmission or local services, but not both at the same time.…”