Scalable Self-Supervised Learning for Multiphase AC-OPF in Distribution Systems with Topology Reconfiguration

Abstract

The proliferation of distributed energy resources (DERs) in distribution grids enables the active coordination of these assets to reduce costs and enable cleaner operations. Realizing this potential requires solving multiphase AC optimal power flow (AC-OPF) quickly across varying loads, DER availabilities, and topology reconfigurations, at much greater speed and scale than conventional nonlinear solvers. Learning-based surrogates can offer millisecond inference, yet existing methods target largely balanced transmission systems and do not scale to the multiphase, unbalanced, and reconfigurable nature of distribution feeders at utility scale. We present the Penalty + Sequential Linearized Feasibility Seeking (SLFS) algorithm, a self-supervised learning framework for multiphase distribution AC-OPF under switch-induced topology changes. Penalty+SLFS requires no labeled optimal solutions and trains directly from the AC-OPF objective and constraints through a differentiable fixed-point power flow solver, avoiding expensive label generation and admitting robust training procedures. Topology changes are handled efficiently using Sherman-Morrison-Woodbury updates of the admittance-matrix inverse, while an M-step Jacobian approximation accelerates differentiation through the power flow solver. At inference, SLFS repairs any infeasible predictions, providing feasibility guarantees with low computational overhead. On IEEE feeders ranging from 13 to 8,500 nodes, Penalty+SLFS achieves negligible optimality gaps and near-zero constraint violations, delivers up to three orders of magnitude speedups over IPOPT, and remains robust under large distributional shifts, demonstrating a viable path toward real-time, topology-aware AC-OPF for large-scale distribution grids.

Survival Reinforcement Learning: Toward Scalable Self-Supervised RL
Franki Nguimatsia-Tiofack, Fabian Schramm, Théotime Le Hellard +1
HAL18 Sep 2026PreprintCC BY-NC-SA 4.0Full textStatisticsHAL
Scalable partial information decomposition for symptom networks via supervised embeddings
Cillian Hourican, Eric Dignum, Rick Quax +1
arXiv1 Sep 2026PreprintCC BY 4.0Full textGeneral mathematicsarXiv
Fast and Scalable Multi-Agent Distribution Matching via Partitioned Optimal Transport
Kooktae Lee, Ruchika Singh
arXiv30 Sep 2026PreprintarXiv licenceFull textOptimisation & controlarXiv
Scalable and Adaptive Distribution Regression via Random Features
Yinsong Wang, Xiao Liu, Yu Ding
arXiv17 Oct 2026PreprintCC BY 4.0Full textStatisticsarXiv
Semi-supervised learning on singular data sets
Guillaume Valette
arXiv18 Oct 2026PreprintarXiv licenceFull textAlgebra & number theory StatisticsarXiv

Cite

BibTeX RIS

BibTeX

@misc{nguyen2026scalable,
  title = {{Scalable Self-Supervised Learning for Multiphase AC-OPF in Distribution Systems with Topology Reconfiguration}},
  author = {Hoang T. Nguyen and Shaohui Liu and Reetam Sen Biswas and Varsha Pendyala and Nurali Virani and Deepjyoti Deka and Priya L. Donti},
  howpublished = {arXiv},
  year = {2026},
  month = sep,
  doi = {10.48550/arxiv.2608.25095},
  eprint = {2608.25095},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2608.25095},
  note = {CC BY 4.0},
}

RIS

TY  - UNPB
TI  - Scalable Self-Supervised Learning for Multiphase AC-OPF in Distribution Systems with Topology Reconfiguration
AU  - Hoang T. Nguyen
AU  - Shaohui Liu
AU  - Reetam Sen Biswas
AU  - Varsha Pendyala
AU  - Nurali Virani
AU  - Deepjyoti Deka
AU  - Priya L. Donti
PY  - 2026
DA  - 2026/09/02
PB  - arXiv
DO  - 10.48550/arxiv.2608.25095
UR  - https://arxiv.org/abs/2608.25095
AB  - The proliferation of distributed energy resources (DERs) in distribution grids enables the active coordination of these assets to reduce costs and enable cleaner operations. Realizing this potential requires solving multiphase AC optimal power flow (AC-OPF) quickly across varying loads, DER availabilities, and topology reconfigurations, at much greater speed and scale than conventional nonlinear solvers. Learning-based surrogates can offer millisecond inference, yet existing methods target largely balanced transmission systems and do not scale to the multiphase, unbalanced, and reconfigurable nature of distribution feeders at utility scale. We present the Penalty + Sequential Linearized Feasibility Seeking (SLFS) algorithm, a self-supervised learning framework for multiphase distribution AC-OPF under switch-induced topology changes. Penalty+SLFS requires no labeled optimal solutions and trains directly from the AC-OPF objective and constraints through a differentiable fixed-point power flow solver, avoiding expensive label generation and admitting robust training procedures. Topology changes are handled efficiently using Sherman-Morrison-Woodbury updates of the admittance-matrix inverse, while an M-step Jacobian approximation accelerates differentiation through the power flow solver. At inference, SLFS repairs any infeasible predictions, providing feasibility guarantees with low computational overhead. On IEEE feeders ranging from 13 to 8,500 nodes, Penalty+SLFS achieves negligible optimality gaps and near-zero constraint violations, delivers up to three orders of magnitude speedups over IPOPT, and remains robust under large distributional shifts, demonstrating a viable path toward real-time, topology-aware AC-OPF for large-scale distribution grids.
ER  -

Abstract

Related

Cite