Incentive-oriented Strategy for Optimizing Microgrid-enabled Distribution Network Voltage Regulation based on SQDDPG Algorithm

doi:10.1016/j.segan.2025.102000

Return

Incentive-oriented Strategy for Optimizing Microgrid-enabled Distribution Network Voltage Regulation based on SQDDPG Algorithm

2025-10-14

PRE

DOI：10.1016/j.segan.2025.102000

Original

Original request for help

Save

Abstract

En 中文

The continuous popularization of distributed energy and the increasing energy demand have led to more severe voltage violation problems in distribution networks. To address this challenge, grid-connected microgrids with sufficient flexible voltage regulation resources can be utilized to provide effective voltage support for the distribution network. However, microgrids typically operate as independent entities, and there are barriers to collaboration between distribution networks and microgrids. Consequently, this paper proposes a strategy based on incentivizing microgrids to regulate the voltage for the distribution network. First, the willingness for microgrids to participate in voltage regulation is enhanced by establishing an incentive-based voltage regulation scheme, which includes the cost savings of voltage regulation in the distribution network, the distributed generator disconnection risk in the distribution network, and the cost of voltage-dependent loads in the microgrids. The microgrids provide voltage support for the distribution network by adjusting the operation plan and obtaining the corresponding voltage regulation incentive. Second, to optimize the operation strategy of multi-microgrids while considering voltage regulation incentives, the Shapley Q-value deep deterministic policy gradient (SQDDPG) algorithm is proposed. The Shapley Q value is incorporated into the traditional multiagent deep deterministic policy gradient (MADDPG) algorithm for distributing the global reward to measure the contribution of different microgrids in the voltage regulation process, which allows the algorithm to converge to higher cumulative rewards. Finally, the simulation results for a modified IEEE 33-bus system show that the rate of the voltage violations of the distribution network is reduced by 51.52%, and the operational economy of microgrids has been improved by 9.12%. The efficiency of cooperation between distribution network and microgrids has been effectively improved.