NEO-Grid trains ReLU networks as power-flow surrogates and applies deep equilibrium models for closed-loop volt-var optimization and control, reporting better voltage regulation than linear and heuristic baselines on the IEEE 33-bus test system.
Data-driven modeling of linearizable power flow for large-scale grid topology optimization
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
eess.SY 2verdicts
UNVERDICTED 2representative citing papers
LACE-S uses a neural representation with a projection layer and Jacobian regularization to produce locational carbon emission metrics that remain consistent with total emissions and sensitivities, leading to reliable system-wide emission reductions in load-shifting tests unlike prior metrics.
citing papers explorer
-
NEO-Grid: A Neural Approximation Framework for Optimization and Control in Distribution Grids
NEO-Grid trains ReLU networks as power-flow surrogates and applies deep equilibrium models for closed-loop volt-var optimization and control, reporting better voltage regulation than linear and heuristic baselines on the IEEE 33-bus test system.
-
LACE-S: Toward Sensitivity-consistent Locational Average Carbon Emissions via Neural Representation
LACE-S uses a neural representation with a projection layer and Jacobian regularization to produce locational carbon emission metrics that remain consistent with total emissions and sensitivities, leading to reliable system-wide emission reductions in load-shifting tests unlike prior metrics.