Rapid Active Power Support by Regional PV Aggregation Considering Physical Topological Correlation and Communication Delay

Harnessing the rapid active power regulation potential of aggregated distributed photovoltaics (PVs) to support the bulk power system is crucial under high-penetration PV scenarios. In this article, based on a hierarchical architecture, the optimization and control strategies of distributed PVs are proposed to realize rapid active power support. In the upper level, to achieve reliable and rapid active power allocation, an optimal active power flow model incorporating the prediction uncertainty of distributed PVs and active power transmission loss is constructed, which can be solved in real-time by Chebyshev graph convolutional network (ChebNet). In the lower level, given the electrical interaction between distributed PVs, the optimal active power tracking of the distributed PVs is realized through reference voltage and current tracking control. Further, a delay-dependent distributed coordinated
$ H_{\infty }$
controller is designed to achieve stable tracking in the presence of communication delays and load perturbations. Finally, simulations are performed on the modified IEEE 33-bus and IEEE 118-bus test systems, and the results verify the effectiveness and superiority of ChebNet, as well as the good accuracy, tracking speed, and robustness of the proposed control method.