Adaptive Coalition Formation-Based Coordinated Voltage Regulation in Distribution Networks

TL;DR

This paper introduces a distributed, coalition-based control strategy for PV inverters to effectively regulate voltage in distribution networks with high PV penetration, improving voltage stability and fairness.

Contribution

It proposes a novel dynamic coalition formation scheme combined with a leader-follower consensus algorithm for coordinated voltage regulation.

Findings

Effective voltage regulation demonstrated in case studies.

Coalition formation offers flexibility over centralized schemes.

Strategy shows robustness and adaptability in realistic scenarios.

Abstract

High penetrations of photovoltaic (PV) systems can cause severe voltage quality problems in distribution networks. This paper proposes a distributed control strategy based on the dynamic formation of coalitions to coordinate a large number of PV inverters for voltage regulation. In this strategy, a rule-based coalition formation scheme deals with the zonal voltage difference caused by the uneven integration of PV capacity. Under this scheme, PV inverters form into separate voltage regulation coalitions autonomously according to local, neighbor as well as coalition voltage magnitude and regulation capacity information. To coordinate control within each coalition, we develop a feedback-based leader-follower consensus algorithm which eliminates the voltage violations caused by the fast fluctuations of load and PV generation. This algorithm allocates the required reactive power contribution among the PV inverters according to their maximum available capacity to promote an effective and fair use of the overall voltage regulation capacity. Case studies based on realistic distribution networks and field-recorded data validate the effectiveness of the proposed control strategy. Moreover, comparison with a centralized network decomposition-based scheme shows the flexibility of coalition formation in organizing the distributed PV inverters. The robustness and generalizability of the proposed strategy are also demonstrated.