A Hierarchical Volt–Var Optimization Strategy for High-Penetration PV Networks Leveraging Adaptive Weight-Partitioned Inverter Control and Magnetically Controlled Reactors

The high penetration of distributed photovoltaic (PV) systems introduces significant voltage fluctuations in distribution networks due to the stochastic nature of PV generation. To address the limitations of conventional volt–var regulation, this paper proposes a novel two-layer hierarchical framework driven by two core innovations: a robust globally scheduled magnetically controlled reactor (MCR) and an autonomous adaptive control strategy for local PV inverters. At the local layer, an adaptive five-region weighting strategy enables PV inverters to rapidly mitigate minor voltage fluctuations without relying on communication networks. At the global layer, an improved particle swarm optimization (IPSO) algorithm is employed to coordinate MCR reactive power scheduling, thereby mitigating severe voltage violations and reducing active power losses. The proposed framework is validated on a modified IEEE 33-bus distribution system. Simulation results show that the adaptive local control of PV inverters effectively reduces node voltage deviations compared with conventional control methods. Furthermore, the two-layer coordinated optimization significantly improves overall system performance by reducing both the objective function value and the maximum voltage deviation compared with single-layer control strategies. Compared with other optimization algorithms, IPSO demonstrates strong robustness and stable convergence in the proposed optimization problem. Overall, the proposed hierarchical framework provides a reliable, scalable, and cost-effective solution for real-time voltage regulation in modern active distribution networks.