A trained neural network surrogate for distribution grid voltage constraints is encoded exactly as mixed-integer linear constraints inside optimal power flow, delivering sub-1 V voltage error and faster solves than nonlinear models on networks with PV, EVs, and heat pumps.
Ac opf in radial distribution networks–part i: On the limits of the branch flow convexification and the alternating direction method of multipliers,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
eess.SY 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Enhanced Optimal Power Flow Using a Trained Neural Network Surrogate for Distribution Grid Constraints
A trained neural network surrogate for distribution grid voltage constraints is encoded exactly as mixed-integer linear constraints inside optimal power flow, delivering sub-1 V voltage error and faster solves than nonlinear models on networks with PV, EVs, and heat pumps.