Distributed control of flexible assets in distribution networks considering personal usage plans

Abstract

In this article, a distributed Mixed-Integer Linear Programming (MILP)-based control scheme is proposed to coordinate flexible assets as a Virtual Storage Plant (VSP) for providing flexibility services to distribution networks. The VSP aggregates flexible assets, such as Heating, Ventilation, and Air Conditioning (HVAC) systems and battery storage systems, while considering their individual needs. It tracks a time-varying signal instructed by the Distribution System Operator (DSO) within the required time to support the host network. HVAC systems are treated as virtual batteries, considering indoor temperature comfort, while battery storage systems are managed according to users’ usage plans. To accurately model the flexible assets, binary variables are required, thus formulating an MILP problem. The MILP problem is then incorporated into a Model Predictive Control (MPC) scheme to manage system constraints. The formulated MILP-based MPC problem is finally solved using an accelerated primal decomposition method in a distributed fashion. Unlike existing distributed algorithms commonly proposed in the literature, the algorithm presented in this article is specifically developed for large-scale MILP problems, with guarantees of constraint satisfaction. The effectiveness of the proposed control scheme is evaluated through several case studies, which demonstrate that it ensures acceptable tracking precision. Furthermore, it supports plug-and-play functionality and enhances scalability.