Energy Digest

Power grid state estimation is vulnerable to cyberphysical attacks, leading to significant deviations in system observations. A new framework identifies critical measurement nodes whose compromise results in severe system-level impact by evaluating coalition-based marginal contributions of compromised sensor subsets using permutation sampling. Simulation results show that adversarially identified nodes induce larger deviations in system parameters compared to randomly selected nodes.

Three-phase alternating current transmission is no longer unique in terms of power smoothness; ideal balanced 2-phase systems can also generate constant-magnitude rotating fields. The dominance of three phases in modern infrastructure can be attributed to advantages in conductor architecture, insulation stress, machine utilization, and technological path dependence. A six-phase demonstration showed that high-phase-order transmission was technically feasible and could improve corridor utilization, leading to identified benefits such as modular decomposition and increased power density.

Data-Enabled Predictive Control (DeePC) is a data-driven approach that constructs control actions directly from measured trajectories, enabling adaptation to changing operating conditions without requiring an explicit system model. DeePC has been shown to achieve superior damping of sustained forced oscillations in power systems compared to conventional power system stabilizers (PSSs). The effectiveness of DeePC depends on the quality and representativeness of the underlying dataset, highlighting its potential as a complementary or outperforming alternative to traditional stabilizers.

Frequency scans using electromagnetic transient (EMT) and root-mean-square (RMS) models reveal that RMS models are only valid at low frequencies. The current-controlled grid-forming inverter is more difficult to represent in RMS than the voltage-controlled grid-forming inverter, but accuracy improves with included inner voltage-control loops. Oscillatory modes and zeros of system impedance relate to peaks and dips in singular values of total system impedance.

Solid-State Transformers (SSTs) face a challenge when short-circuit faults occur, causing the SST to shut down due to excessive fault currents. A new control strategy has been proposed to limit DC current spikes in milliseconds using a closed-loop current controller and ramped recovery stage, enabling faster and more cost-effective fault response for data centers. The proposed strategy is embedded directly in the control of the SST's DC-DC stage without requiring additional hardware.

Virtual power plants (VPPs) aggregate flexible resources to provide frequency regulation for the grid, helping address supply-demand balance challenges. The proposed optimal operation strategy prioritizes low-cost resources while considering temporal coupling characteristics, resulting in reduced operational cost and increased profit. A fast disaggregation algorithm is also introduced to minimize online computation time.

Production scheduling identification (PSI) is an inverse-optimization approach that uses smart meter data to identify production scheduling parameters for industrial load modeling under incomplete information. PSI can accurately model industrial loads with acceptable accuracy, even without direct access to private data, and outperforms existing general-purpose models in certain cases. The method achieved a modeling error of up to 8.5% for one test case and 5.2% for another.

Expansion planning models can now optimize storage duration alongside other components, reducing annualized costs by 6.8% and long-run marginal cost by 20%. Pumped-storage hydropower (PSH) integration into the Brazilian Interconnected System reduces operating costs by 23.5%, thermal generation by 34 TWh/year, and VRE curtailment by 4.2 percentage points. The inclusion of PSH over battery energy storage systems results in a significant reduction in system costs.

A price-based distributed scheduling system optimizes energy consumption for households with renewable generation and deferrable EV charging, utilizing a bilevel stochastic dynamic program to find the joint optimal centralized policy, which involves a two-threshold policy on aggregate renewable generation. The Threshold Pricing Rule is proposed as a uniform, individually rational, and asymptotically optimal solution. Simulations confirm the rule's optimality and individual surplus gains under light-traffic conditions using synthetic and real-world data.