Energy Digest

bayesgrid, an open-source Python tool, generates synthetic power transmission-distribution grids using Bayesian Hierarchical Models trained on existing distribution network databases, allowing for probability-based generation of multiple grid instances worldwide. The generated networks contain detailed information on nodal demand distributions and critical reliability indices, making them suitable for reliability-related studies. The tool can be applied to any geographical location worldwide and saves generated networks in open-source platforms like PandaPower and OpenDSS.

The article studies enhanced droop-free control strategies for islanded microgrids to achieve effective active power sharing and maintain frequency stability. A theoretical proof is provided for the asymptotic stability of two NAPC-based droop-free control schemes, showing that all effective eigenvalues have negative real parts. The analysis reveals that the average available capacity of controllable DERs has a decisive influence on the stability margin of NAPC-based control schemes.

A new power grid optimization method is proposed to reconfigure substation topology while considering security constraints, including busbar and coupler contingencies. The approach uses a heuristic master problem with multiple independent substation problems to reduce computational complexity and improve scalability for large-scale systems. It successfully mitigates the impact of potential faults and balances system security and cost in case studies on three different power grid systems.

A novel harmonic-domain framework for systems with variable fundamental frequency has been developed, introducing a new sliding Fourier decomposition and necessary conditions for time-domain realizability. An exact differential model has been derived for nonlinear systems under variable frequency, allowing for explicit parameter-varying approximations and stability analysis and control synthesis without approximation or assumptions on frequency variation within a prescribed interval. This framework reduces problems to harmonic Lyapunov inequalities evaluated at extreme frequency values, providing a convex LMI characterization.

LUMINA is a framework for developing foundation models that can balance accuracy with physical constraints in complex optimization problems like ACOPF, optimizing learning of physics-invariant representations while ensuring constraint satisfaction. The framework aims to accelerate scientific computation by leveraging reusable representations across problem instances. It provides data processing and training pipelines to support reproducible research on physics-informed foundation models.

This work presents a general framework for operationally driven optimal siting and sizing of battery energy storage systems in power transmission networks to enhance their resource adequacy. A reformulated mixed-integer second-order cone programming problem is solved via Generalized Benders Decomposition with feasibility cuts, enabling efficient computation while ensuring convergence guarantees. The approach considers multi-period planning horizons and targets large-scale meshed systems with high temporal resolution.

Utility-scale off-grid renewable power-to-hydrogen (ReP2H) systems coordinating heterogeneous electrolyzers can be optimized to balance energy production and frequency security. A new co-optimization framework is developed to enable coordinated on/off switching and load allocation across electrolyzers, allowing for more efficient use of hydrogen output under uncertain renewable power input. The proposed approach results in significant reductions in reliance on conventional reserves and increases in annual net profit.