In this paper, a novel distributed coordinated control framework is proposed to handle the uncertain voltage violations in active distribution networks. It addresses the problem of coordination of different types of devices in a distributed manner. In our control design, on-load tap changers (OLTCs) are firstly employed to handle the potential voltage violations based on the prediction of renewable outputs and load variations. During real-time operation, once an unmanageable voltage violation is detected, the reactive power of distributed energy resources (DERs) will be coordinated immediately to provide fast corrective control. The control schedules of OLTCs are calculated by solving a multitime-step constrained optimization problem via the alternating direction method of multipliers, whereas the reactive power injections of DERs are determined by a novel online distributed algorithm. The effectiveness of the proposed control framework is verified on the modified IEEE 34-bus and 123-bus test feeders.

This letter considers the problem of voltage regulation of distribution networks by optimally setting the reactive power of distributed energy resources (DERs). Based on the linearized DistFlow model, the problem is first formulated as a convex quadratic optimization problem with linear constraints. Then, a distributed accelerated dual descent algorithm (DADD) is proposed to solve the optimization problem by employing the dual decomposition and accelerated gradient projected techniques.

In this letter, a novel consensus-based approach is proposed to solve the energy management problem for islanded microgrids. By taking the incremental cost of each agent as the consensus variable and through limited communication between neighboring agents, the proposed approach can quickly converge to the optimal solutions.

The increasing penetration of renewables and growing number and multiple operating modes of microgrids (MGs) bring challenges to designing a frequency control scheme for the networked microgrid system (NMS) that can work in all possible scenarios. This paper proposes a distributed model predictive control (DMPC)-based controller to regulate the frequency while maintaining the voltage constraints of all buses in the NMS with different system topologies and operating modes. In the proposed scheme, the frequency is regulated by adjusting the voltage-sensitive load voltages through MPC. By applying a consensus version of the alternating direction method of multipliers algorithm, the proposed controller coordinates MGs in the NMS using limited communication between neighboring MGs to achieve better privacy and scalability. Simulation results on extensive scenarios show the proposed controller successfully mitigates the frequency deviation caused by renewables’ fluctuation with voltage constraints considered and performs comparably well with the centralized controller. Some general issues regarding frequency control of the NMS are also discussed.

The share of inverter-connected renewable energy resources (RESs) is increasing in the grid, with these resources partially displacing conventional synchronous generators. This has resulted in increased variability of active power supply, reduced overall inertia, and increased spatial heterogeneity of inertia, leading to faster system frequency dynamics along with larger and more frequent frequency control events. These effects are expected to become increasingly more important in power system control in next-generation grids, which may conceivably be made up entirely of RESs. To mitigate these challenges, a fast, area-based hierarchical control strategy is proposed. This scheme partitions the power system into small areas, estimates local power imbalances, and corrects them by utilizing local inverter-based resources (IBRs). In cases where sufficient resources are not available locally, power is preferentially sourced from electrically close neighbours using an iterative distributed optimization scheme which preserves information privacy between areas. The proposed frequency control architecture can be retrofit onto existing control systems, and allows for flexibility in the amount of model information available to the designer. The control strategy is validated on two detailed multi-area power system models. Simulation results show that the strategy provides fast and localized frequency control.

Larger-scale stochastic power systems urge the development of real-time alternating current optimal power flow, artificial intelligence (AI) thus becomes an alternative. However, traditional AI only imitates experiences, and cannot follow in-depth physics. This may cause an undesired nongeneralizability and topology intractability. To address this issue, a physics-guided graph neutral network (PG-GNN) is proposed. The PG-GNN firstly capture the physical constraints by a dual Lagrangian. Besides, the branch features of power grids are fully exploited to allow the PG-GNN to master tremendous topological patterns. To further manage the out-of-distribution topology, stability property of the PG-GNN is proved, then upon this evidence, an online transfer learning is proposed to allow the PG-GNN to fast master the unexpected topology. Numerical tests on benchmarks show that, the proposed method holds well topology-transferability, enables near or even better solutions than conventional optimizer, but merits much more than 100 times efficiency.