Abstract
Electric power systems are currently experiencing a profound change, as increasing amounts of Renewable Energy Sources (RESs) displace conventional forms of generation. This development has gone hand-in-hand with an increasing share of distributed power generation being connected directly to the Distribution Network (DN), and the widespread of other types of Distributed Energy Resources (DERs), such as Energy Storage Sytems (ESSs), Electric Vehicles (EVs), and active (flexible) consumers. As these trends are expected to continue, this will require a profound revision of the way Transmission System Operators (TSOs) and Distribution System Operators (DSOs) interact with each other to fully benefit from the growing flexibility that is available at the DN level. In this work we propose a new tool for the coordinated operational planning of transmission and distribution systems, considering the existence of shared resources that can be simultaneously used by TSO and DSOs for the optimal operation of their networks. The tool uses advanced distributed optimization techniques, namely the Alternating Direction Method of Multipliers (ADMM) in order to maintain data privacy of the several agents involved in the optimization problem, and keep the tractability of the problem. The proposed tool is applied to modified IEEE test systems, and the results obtained highlight the benefits of the proposed coordination mechanism to solve problems occurring simultaneously at the transmission and DN-levels.

Abstract
This work models a distributed community-based market with diverse assets (photovoltaic generators and energy storage systems), accounting for network constraints and adopting the relaxed branch flow model. The market is modeled in a single and fully distributed approach, employing the alternating direction method of multipliers (ADMM) to prevent voltage and line capacity problems in the community network and improve data privacy and reduce the communication burden. Different scenarios, based on the penalty term and the agents' number, are tested to study the efficiency of the algorithm and the convergence rate of the ADMM distributed model. The proposed method is tested on 10-bus, 22-bus, and 33-bus medium voltage radial distribution networks, where each node contains a large prosumer with one or several assets. One important conclusion is that the implemented residual balancing technique improves the efficiency of the ADMM distributed algorithm by increasing the convergence rate and reducing the computational time.