Energy Digest

Shadow pricing in power systems reveals how stability constraints alter economic dispatch, producing price signals and settlement implications that remain poorly characterized. Shadow prices derived from transient stability-constrained optimal power flow decompose into nodal energy and stability components, allowing for asset-specific analysis. These prices enable fair settlement without side payments by charging assets for reduced stability margins and paying those that improve them.

This paper assesses the impact of security modeling fidelity on hyperscale data center hosting capacity in transmission grids, finding that the N-SCUC model yields an average hosting capacity of 444 MW, while omitting N-1 security results in a 19.8% overestimation. Other models, including DC-OPF and basic SCUC, further overestimate hosting capacity. The impact of unit commitment, reserve requirements, and N-1 security on bus-level hosting capacity estimates is quantified.

A new AI model called PE-TSFM achieves an accuracy of 87% in degradation classification tasks for power module health monitoring, outperforming five conventional models. The model is trained on over 143 million unlabeled timestamps collected from an operating traction converter, showcasing its strong few-shot learning capability. PE-TSFM demonstrates potential for advancing model performance in power electronics converters.

Dynamic wireless charging (DWC) could fundamentally alter transportation electrification by enabling high-throughput charging in motion and reducing reliance on large onboard batteries, with its share rising sharply from 48% at a project lifetime of 30 years to 100% at 32 years. DWC can provide over six times the rated charging capacity of traditional stationary charging stations (SCS) while requiring 46% less investment in new distribution-grid infrastructure. A combined DWC and SCS portfolio reaches a maximum lifetime value of USD 60 billion, 63% above the case with only SCS.

A proposed method for distribution system operators to design spatially fair dynamic price signals for flexible prosumers, accounting for distribution grid constraints and prosumer stochastic behavior, involves solving a scenario-based optimal power flow problem to derive price signals. The method includes mechanisms to enforce price fairness across the network and evaluates its performance through simulations on a model distribution grid. The approach aims to reduce price unfairness and cost variance among prosumers.

The proposed control method for grid-forming PV systems uses control barrier functions (CBFs) to account for constraints of voltage source converters (VSCs) and PV resources, ensuring reliable and safe operation under grid faults, sudden changes in irradiance, and load perturbations. The control combines dual-port grid-forming control with CBFs to restrict its dynamics within the VSC and PV limits. A real-time optimization process is used to minimize a cost function, making the least modification to the control state to remain within the safe set encoded by the CBFs.

A 500 MW data center load, equivalent to 17.5% of the base-year peak, significantly impacts grid expansion planning, increasing the 15-year net present value of total system cost by 1,794 million USD and system-average locational marginal price by 45.2%. The added demand is served almost entirely by incremental thermal generation, with renewable curtailment remaining negligible. Strategically sited generation can substitute for network reinforcement, shifting the optimal plan from transmission-only to generation-only.

Conventional distributionally robust optimization methods based on the Wasserstein distance tend to amplify the probabilities of low-probability extreme events, leading to over-conservative investment decisions. A novel ambiguity set is proposed that incorporates a Kullback-Leibler divergence regularization term to curb this over-conservatism. The proposed method is validated through numerical results on two power systems.

The proposed method applies the stress strength method to estimate two key resource adequacy indices, LOLE and EENS, in renewable-rich power systems by modeling the cumulative impact of uncertainties on resource adequacy using the central limit theorem. The approach improves accuracy with a cumulant-based Gram Charlier expansion, achieving a substantial reduction in computational time without sacrificing accuracy. The method is particularly suitable for future power systems operation and planning, particularly with increasing renewable penetration.

A data-driven surrogate-modeling framework is presented for real-time control of floating offshore wind farms, using mid-fidelity simulations and machine learning techniques to estimate Damage Equivalent Load (DEL) datasets of floating wind turbines. The proposed methodology achieves accuracy errors below 15% for various load parameters and maintains computational times suitable for real-time applications. The study highlights the challenges of generalizing the load-surrogate model under varying environmental and control conditions.