Cost-aware distributed reactive power dispatch for voltage optimization in active distribution networks with multiple microgrids

Abstract

With the increasing penetration of distributed renewable energy resources, distribution networks are gradually evolving into interconnected systems composed of multiple microgrids. Microgrids are typically equipped with controllable reactive power resources and can provide voltage support to the distribution network through points of common coupling. However, directly dispatching microgrid reactive resources without explicitly quantifying boundary support responsibilities and local dispatch costs may result in inefficient reactive power allocation and excessive utilization of certain microgrid-side resources. To address this issue, this paper proposes a cost-aware distributed reactive power dispatch method for voltage optimization in active distribution networks with multiple microgrids. First, a PCC reactive-power reference model is established to quantify each microgrid’s baseline reactive support responsibility under rated-voltage operating conditions. Second, a unified reactive dispatch cost model is developed that considers the fixed, operating, and opportunity costs of reactive power resources. Based on these models, a multi-agent distributed reactive power scheduling framework is formulated, in which the distribution network and microgrids independently optimize their local objectives while coordinating through PCC boundary variables. The coupled optimization problem is solved using an adaptive consensus alternating direction method of multipliers (ADMM) algorithm combined with continuous relaxation and integer recovery for discrete voltage regulation devices. Case studies on a modified IEEE 69-bus distribution system demonstrate that the proposed method can effectively improve voltage profiles, reduce network losses and comprehensive operating costs, and enhance the coordinated utilization of microgrid-side reactive resources while preserving microgrid operational autonomy.