End-to-End Pseudo-Measurement Learning for State Estimation under Limited Observability

A. Stratigakos; J. G. De la Varga; J. M. Morales; S. Pineda

arxiv: 2605.23525 · v1 · pith:VZAQRXVKnew · submitted 2026-05-22 · 🧮 math.OC

End-to-End Pseudo-Measurement Learning for State Estimation under Limited Observability

J. G. De la Varga , S. Pineda , A. Stratigakos , J. M. Morales This is my paper

Pith reviewed 2026-05-25 04:30 UTC · model grok-4.3

classification 🧮 math.OC

keywords distribution system state estimationpseudo-measurementsneural networksweighted least squaresend-to-end learninglimited observabilitypower flow model

0 comments

The pith

Embedding a weighted least squares estimator as a layer in a neural network lets data-driven pseudo-measurements directly reduce state estimation error while preserving the exact nonlinear power flow model.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper develops a distribution system state estimation framework that generates pseudo-measurements via a neural network and feeds them into a classical weighted least squares solver. The solver is placed inside the network as a differentiable layer so that estimation errors propagate back through implicit differentiation to update the network parameters. This end-to-end coupling replaces the usual decoupled approach in which pseudo-measurements are produced independently of the physical estimator. Numerical tests on the IEEE 30-bus and 33-bus systems show higher accuracy than existing methods across varied loads and measurement placements. The framework restores observability without altering the underlying AC power flow equations.

Core claim

The paper claims that embedding the WLS estimator as a layer inside the neural network and training via implicit differentiation produces pseudo-measurements whose effect on final estimation error is directly minimized, yielding more accurate state estimates than methods that generate pseudo-measurements separately from the physical solver.

What carries the argument

The end-to-end learning formulation that embeds the WLS estimator as a layer in the neural network, allowing implicit differentiation to back-propagate the impact of pseudo-measurements onto network parameters.

If this is right

The exact nonlinear AC power flow model remains unchanged inside the estimator.
State estimation accuracy improves relative to decoupled pseudo-measurement methods on the tested IEEE systems.
Performance holds across a wide range of loading conditions and measurement configurations.
Observability is restored in networks with limited and heterogeneous measurements.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

The same embedding could be applied to other optimization-based estimators that admit implicit differentiation.
If the learned pseudo-measurements prove robust, fewer physical sensors might suffice for acceptable accuracy in active distribution networks.
The method could be tested on time-series data to check whether the learned pseudo-measurements remain effective under temporal correlation not present in the static test cases.

Load-bearing premise

Implicit differentiation through the WLS layer remains numerically stable and the resulting gradients yield pseudo-measurements that generalize beyond training scenarios without biasing the physical model.

What would settle it

Numerical instability during implicit differentiation or degraded accuracy on loading conditions and measurement configurations outside the training set would falsify the claim that the coupled training improves estimation accuracy.

Figures

Figures reproduced from arXiv: 2605.23525 by A. Stratigakos, J. G. De la Varga, J. M. Morales, S. Pineda.

read the original abstract

Distribution System State Estimation (DSSE) is becoming increasingly important with the integration of Distributed Energy Resources (DERs) and the active operation of distribution networks (DNs), but it remains challenging due to the limited and heterogeneous monitoring infrastructure available in these networks. To address this challenge, this paper proposes a novel DSSE framework that restores observability through data-driven pseudo-measurements generated by a Neural Network (NN), while preserving the exact non-linear AC power flow model within a classical Weighted Least Squares (WLS) estimator. Unlike conventional approaches that generate pseudo-measurements independently of the physical estimation process, the proposed method explicitly couples both components through an end-to-end learning formulation. Specifically, the WLS estimator is embedded as a layer within the NN architecture, enabling implicit differentiation to propagate the impact of pseudo-measurements on the final estimation error back to the NN parameters. Extensive numerical experiments on the IEEE 30-bus and IEEE 33-bus systems demonstrate that the proposed framework consistently outperforms state-of-the-art methods in state estimation (SE) accuracy under a wide range of loading conditions and measurement configurations.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

The paper embeds a WLS estimator as a differentiable layer to train a NN for pseudo-measurements in DSSE, but the abstract shows no numbers and the implicit differentiation stability issue looks real.

read the letter

The core idea is to train a neural net that outputs pseudo-measurements for distribution system state estimation by back-propagating through an embedded weighted least squares solver using implicit differentiation. This keeps the exact nonlinear AC power flow equations inside the classical estimator instead of approximating them. On the IEEE 30-bus and 33-bus cases the abstract claims consistent gains over prior methods across different loads and measurement setups. That coupling of data-driven generation with the physical model is the part that stands out as new for this application. The experiments target a real operational problem with DERs and sparse monitoring, which matters for grid work. The soft spots are the lack of any reported error metrics, training details, or ablation results in the abstract, plus the open question of whether the implicit gradients remain stable. When observability is marginal the measurement Jacobian can be rank-deficient or ill-conditioned, and small changes in the pseudo-measurements could produce unreliable or exploding derivatives. The text supplies no evidence that condition numbers were monitored or that regularization was used to keep iterates in the valid region for the implicit function theorem. Without those checks the performance claim cannot be taken at face value. This paper is aimed at power systems researchers working on state estimation under limited data. A reader already familiar with differentiable optimization layers might extract the specific formulation and test setup, but the current evidence is too thin to rely on. I would send it for peer review only if the full manuscript adds quantitative results and directly addresses the gradient stability concern; otherwise it stays at the idea stage.

Referee Report

2 major / 1 minor

Summary. The paper proposes an end-to-end DSSE framework in which a neural network generates pseudo-measurements that are fed into a classical non-linear WLS estimator embedded as a differentiable layer; implicit differentiation is used to back-propagate estimation error to the NN parameters while preserving the exact AC power-flow model. Numerical experiments on the IEEE 30-bus and 33-bus systems are reported to show consistent outperformance versus state-of-the-art methods across loading conditions and measurement configurations.

Significance. The explicit coupling of data-driven pseudo-measurements with a physics-based WLS layer via implicit differentiation is a technically interesting direction for hybrid ML-physics state estimation under limited observability. If the numerical results are shown to be robust, the approach could influence subsequent work on differentiable power-system solvers.

major comments (2)

[Abstract] Abstract: the central claim of consistent outperformance is stated without any quantitative error metrics, training/validation splits, error bars, or ablation isolating the end-to-end component, preventing verification of the reported gains.
[Method (end-to-end formulation)] The implicit-differentiation step through the WLS layer (central to the end-to-end claim) presupposes that the measurement Jacobian remains well-conditioned and that the implicit-function theorem applies at every training iteration; the manuscript supplies no discussion of Jacobian regularization, condition-number monitoring, or safeguards against rank deficiency under the limited-observability regimes that are the paper’s target.

minor comments (1)

Figure captions and table headings should explicitly state the error metric (e.g., voltage magnitude RMSE) and the number of Monte-Carlo trials used.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive feedback. We address each major comment below, agreeing where the manuscript can be strengthened and outlining specific revisions.

read point-by-point responses

Referee: [Abstract] Abstract: the central claim of consistent outperformance is stated without any quantitative error metrics, training/validation splits, error bars, or ablation isolating the end-to-end component, preventing verification of the reported gains.

Authors: We agree that the abstract would be strengthened by including quantitative support. In the revised manuscript we will add specific metrics (e.g., average voltage-magnitude and angle errors on the IEEE 30- and 33-bus systems), note the train/validation/test splits employed, reference the ablation studies that isolate the end-to-end training benefit, and mention error bars obtained from repeated runs. revision: yes
Referee: [Method (end-to-end formulation)] The implicit-differentiation step through the WLS layer (central to the end-to-end claim) presupposes that the measurement Jacobian remains well-conditioned and that the implicit-function theorem applies at every training iteration; the manuscript supplies no discussion of Jacobian regularization, condition-number monitoring, or safeguards against rank deficiency under the limited-observability regimes that are the paper’s target.

Authors: The manuscript does not contain an explicit discussion of Jacobian conditioning or safeguards. Although the WLS solver converged reliably in all reported experiments, we acknowledge the referee’s point. We will insert a new subsection in the method that (i) recalls the conditions under which the implicit-function theorem applies to the WLS layer, (ii) describes monitoring of the Jacobian condition number during training, and (iii) outlines a lightweight safeguard (small diagonal regularization of the gain matrix when the condition number exceeds a preset threshold). Empirical condition-number statistics from the IEEE test cases will be added to the numerical-results section. revision: yes

Circularity Check

0 steps flagged

No circularity in the derivation chain

full rationale

The paper describes a standard end-to-end training setup in which a neural network generates pseudo-measurements that are passed through an embedded WLS layer whose output error is back-propagated via implicit differentiation. The reported accuracy improvements on the IEEE 30-bus and 33-bus test cases are empirical outcomes of this supervised training process evaluated on held-out scenarios; they do not reduce by construction to any fitted quantity or self-citation. No uniqueness theorem, ansatz smuggled via prior work, or self-definitional mapping is invoked. The derivation chain therefore remains self-contained and externally falsifiable through standard train/test splits and comparison against independent baselines.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

Abstract-only review; the method implicitly assumes that the AC power flow equations remain exact inside the WLS layer and that the NN can be trained to produce pseudo-measurements consistent with those equations. No explicit free parameters or invented entities are named.

axioms (1)

domain assumption The non-linear AC power flow model inside the WLS estimator is treated as exact and known.
Stated in the abstract as preserving the exact non-linear AC power flow model.

pith-pipeline@v0.9.0 · 5738 in / 1216 out tokens · 18931 ms · 2026-05-25T04:30:46.433744+00:00 · methodology

discussion (0)

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

IndisputableMonolith/Cost/FunctionalEquation.lean washburn_uniqueness_aczel unclear

?

unclear
Relation between the paper passage and the cited Recognition theorem.

WLS-SE ˆx∈arg min_x (z−h(x))⊤W(z−h(x)); implicit diff yields ∂ˆx/∂z=(J⊤WJ)−1 J⊤W (Eq. 15, Gauss-Newton approx)
IndisputableMonolith/Foundation/RealityFromDistinction.lean reality_from_one_distinction unclear

?

unclear
Relation between the paper passage and the cited Recognition theorem.

end-to-end training via adjoint λ solving (J⊤WJ)λ=∇ˆxL; no recognition cost or distinction primitive

What do these tags mean?

matches: The paper's claim is directly supported by a theorem in the formal canon.
supports: The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends: The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses: The paper appears to rely on the theorem as machinery.
contradicts: The paper's claim conflicts with a theorem or certificate in the canon.
unclear: Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Reference graph

Works this paper leans on

35 extracted references · 35 canonical work pages

[1]

Graph-learning-assisted state estimation using sparse heterogeneous measurements,

H. Yue, W. Zhang, U. C. Yilmaz, T. Yildiz, H. Huang, H. Liu, Y. Lin, and A. Abur, “Graph-learning-assisted state estimation using sparse heterogeneous measurements,”Electric power systems research, vol. 235, p. 110644, 2024

work page 2024
[2]

A review on distribution system state estimation,

A. Primadianto and C. N. Lu, “A review on distribution system state estimation,” IEEE Transactions on Power Systems, vol. 32, no. 5, pp. 3875–3883, 2016

work page 2016
[3]

A survey of power system state estimation using multiple data sources: PMUs, SCADA, AMI, and beyond,

G. Cheng, Y. Lin, A. Abur, A. Gómez-Expósito, and W. Wu, “A survey of power system state estimation using multiple data sources: PMUs, SCADA, AMI, and beyond,”IEEE Transactions on Smart Grid, 2023

work page 2023
[4]

Distribution network state estimation based on attention-enhanced recurrent neural network pseudo-measurement modeling,

Y. Wang, J. Gu, and L. Yuan, “Distribution network state estimation based on attention-enhanced recurrent neural network pseudo-measurement modeling,”Pro- tection and Control of Modern Power Systems, vol. 8, no. 2, pp. 1–16, 2023

work page 2023
[5]

A game-theoretic data-driven approach for pseudo-measurement generation in distribution system state estima- tion,

K. Dehghanpour, Y. Yuan, Z. Wang, and F. Bu, “A game-theoretic data-driven approach for pseudo-measurement generation in distribution system state estima- tion,”IEEE Transactions on Smart Grid, vol. 10, no. 6, pp. 5942–5951, 2019. 17 Table B.5: Detailed results for IEEE 30-bus (mean±std over seeds) for different demand variability levels (Var.), measureme...

work page 2019
[6]

Decision-focused learning: Foundations, state of the art, benchmark and future opportunities,

J.Mandi, J.Kotary, S.Berden, M.Mulamba, V.Bucarey, T.Guns, andF.Fioretto, “Decision-focused learning: Foundations, state of the art, benchmark and future opportunities,”Journal of Artificial Intelligence Research, vol. 80, pp. 1623–1701, 2024

work page 2024
[7]

Agrawal, B

A. Agrawal, B. Amos, S. Barratt, S. Boyd, S. Diamond, and J. Z. Kolter, “Differen- 18 Table B.6: Detailed results for IEEE 33-bus (mean±std over seeds) for different demand variability levels (Var.), measurement configurations (Conf.) and all the different approaches (App.). Var. Var. Var. Conf. Conf. Conf. App. App. App. RMSEVRMSEVRMSEV RMSEθRMSEθRMSEθ R...

work page 2019
[8]

Bayesian state estimation for un- observable distribution systems via deep learning,

K. R. Mestav, J. Luengo-Rozas, and L. Tong, “Bayesian state estimation for un- observable distribution systems via deep learning,”IEEE Transactions on Power Systems, vol. 34, no. 6, pp. 4910–4920, 2019. 19

work page 2019
[9]

Linear single- and three-phase voltage forecasting and bayesian state estimation with limited sensing,

R. Dobbe, W. van Westering, S. Liu, D. Arnold, D. Callaway, and C. Tomlin, “Linear single- and three-phase voltage forecasting and bayesian state estimation with limited sensing,”IEEE Transactions on Power Systems, vol. 35, no. 3, pp. 1674–1683, May 2020

work page 2020
[10]

Arobustdata- driven gaussian process regression-enabled distribution system state estimation using partial node data,

J.Zhang, Y.Liu, A.Rouhani, J.Zhao, G.J.Warchol, andK.Scott, “Arobustdata- driven gaussian process regression-enabled distribution system state estimation using partial node data,” in2024 IEEE Power & Energy Society General Meeting (PESGM), Jul. 2024, pp. 1–5, iSSN: 1944-9933

work page 2024
[11]

Real-timepowersystemstateestimation and forecasting via deep unrolled neural networks,

L.Zhang, G.Wang, andG.B.Giannakis, “Real-timepowersystemstateestimation and forecasting via deep unrolled neural networks,”IEEE Transactions on Signal Processing, vol. 67, no. 15, pp. 4069–4077, 2019

work page 2019
[12]

Physics-informed graphical neural network for pa- rameter & state estimations in power systems,

L. Pagnier and M. Chertkov, “Physics-informed graphical neural network for pa- rameter & state estimations in power systems,”arXiv preprint arXiv:2102.06349, 2021

work page arXiv 2021
[13]

Powersystemstateforecastingusing machine learning techniques,

D.Mukherjee, S.Chakraborty, andS.Ghosh, “Powersystemstateforecastingusing machine learning techniques,”Electrical Engineering, vol. 104, no. 1, pp. 283–305, 2022

work page 2022
[14]

A survey on state estimation techniques and challenges in smart distribution systems,

K. Dehghanpour, Z. Wang, J. Wang, Y. Yuan, and F. Bu, “A survey on state estimation techniques and challenges in smart distribution systems,”IEEE Trans- actions on Smart Grid, vol. 10, no. 2, pp. 2312–2322, 2018

work page 2018
[15]

Distribution system state estimation through gaus- sian mixture model of the load as pseudo-measurement,

R. Singh, B. Pal, and R. Jabr, “Distribution system state estimation through gaus- sian mixture model of the load as pseudo-measurement,”IET generation, trans- mission & distribution, vol. 4, no. 1, pp. 50–59, 2010

work page 2010
[16]

Distribution system state estima- tionusinganartificialneuralnetworkapproachforpseudomeasurementmodeling,

E. Manitsas, R. Singh, B. C. Pal, and G. Strbac, “Distribution system state estima- tionusinganartificialneuralnetworkapproachforpseudomeasurementmodeling,” IEEE Transactions on power systems, vol. 27, no. 4, pp. 1888–1896, 2012

work page 2012
[17]

Robust forecasting aided power system state estimation considering state correlations,

J. Zhao, G. Zhang, Z. Y. Dong, and M. La Scala, “Robust forecasting aided power system state estimation considering state correlations,”IEEE Transactions on Smart Grid, vol. 9, no. 4, pp. 2658–2666, 2016

work page 2016
[18]

Physics-guided deep learning for power system state estimation,

L. Wang, Q. Zhou, and S. Jin, “Physics-guided deep learning for power system state estimation,”Journal of Modern Power Systems and Clean Energy, vol. 8, no. 4, pp. 607–615, 2020

work page 2020
[19]

Robust interval state estimation for distribution systems considering pseudo-measurement interval prediction,

X. Zhang, W. Yan, M. Huo, and H. Li, “Robust interval state estimation for distribution systems considering pseudo-measurement interval prediction,”Journal of Modern Power Systems and Clean Energy, vol. 12, no. 1, pp. 179–188, 2023. 20

work page 2023
[20]

Exact modeling of non- gaussian measurement uncertainty in distribution system state estimation,

M. Vanin, T. Van Acker, R. D’hulst, and D. Van Hertem, “Exact modeling of non- gaussian measurement uncertainty in distribution system state estimation,”IEEE Transactions on Instrumentation and Measurement, vol. 72, pp. 1–11, 2023

work page 2023
[21]

Matrix completion for low-observability voltage estimation,

P. L. Donti, Y. Liu, A. J. Schmitt, A. Bernstein, R. Yang, and Y. Zhang, “Matrix completion for low-observability voltage estimation,”IEEE Transactions on Smart Grid, vol. 11, no. 3, pp. 2520–2530, 2020

work page 2020
[22]

Data en- hanced robust distribution system state estimation for realistic system with pa- rameter corrections,

Y. Liu, A. Rouhani, J. Zhao, J. Zhang, G. J. Warchol, and K. Scott, “Data en- hanced robust distribution system state estimation for realistic system with pa- rameter corrections,”IEEE Transactions on Power Systems, pp. 1–12, 2025

work page 2025
[23]

A novel state estimation method for distribution networks with low observability based on linear AC optimal power flow model,

M. Zeraati, M. R. Sheibani, F. Jabari, and E. Heydarian-Forushani, “A novel state estimation method for distribution networks with low observability based on linear AC optimal power flow model,”Electric Power Systems Research, vol. 228, p. 110085, Mar. 2024

work page 2024
[24]

WLS-based state estimation for unobservable distri- bution grids through allocation factors evaluation,

M. Pau and P. A. Pegoraro, “WLS-based state estimation for unobservable distri- bution grids through allocation factors evaluation,”IEEE Transactions on Instru- mentation and Measurement, vol. 73, pp. 1–13, 2024

work page 2024
[25]

Power system robust state estimation as a layer: A novel end-to-end learning approach,

Y. Ding, W. Shi, M. Duan, Y. Zhao, J. Ruan, J. Zhao, and Z. Xu, “Power system robust state estimation as a layer: A novel end-to-end learning approach,”arXiv preprint arXiv:2511.22836, 2025

work page arXiv 2025
[26]

Abur and A

A. Abur and A. Gómez-Expósito,Power system state estimation: theory and im- plementation, ser. Power engineering. New York, NY: Marcel Dekker, 2004

work page 2004
[27]

Robust and scalable power system state estimationviacompositeoptimization,

G. Wang, G. B. Giannakis, and J. Chen, “Robust and scalable power system state estimationviacompositeoptimization,”IEEE Transactions on Smart Grid, vol.10, no. 6, pp. 6137–6147, Nov. 2019

work page 2019
[28]

State estimation in power systems based on a mathe- matical programming approach,

E. Caro and A. Hernández, “State estimation in power systems based on a mathe- matical programming approach,” inAdvances in Electric Power and Energy. John Wiley & Sons, Ltd, 2020, pp. 23–57

work page 2020
[29]

Power system observability with minimal phasor measurement placement,

T. Baldwin, L. Mili, M. Boisen, and R. Adapa, “Power system observability with minimal phasor measurement placement,”IEEE Transactions on Power Systems, vol. 8, no. 2, pp. 707–715, 1993

work page 1993
[30]

Learning-based state estima- tion in distribution systems with limited real-time measurements,

J. De la Varga, S. Pineda, J. Morales, and A. Porras, “Learning-based state estima- tion in distribution systems with limited real-time measurements,”Electric Power Systems Research, vol. 241, p. 111268, 2025

work page 2025
[31]

State estimation in two time scales for smart distribution systems,

A. Gómez-Expósito, C. Gómez-Quiles, and I. Džafić, “State estimation in two time scales for smart distribution systems,”IEEE Transactions on Smart Grid, vol. 6, no. 1, pp. 421–430, Jan. 2015. 21

work page 2015
[32]

NewYork: Springer, 2006

J.NocedalandS.J.Wright,Numerical Optimization, 2nded. NewYork: Springer, 2006

work page 2006
[33]

A protec- tion coordination index for evaluating distributed generation impacts on protection for meshed distribution systems,

H. H. Zeineldin, Y. A.-R. I. Mohamed, V. Khadkikar, and V. R. Pandi, “A protec- tion coordination index for evaluating distributed generation impacts on protection for meshed distribution systems,”IEEE Transactions on Smart Grid, vol. 4, no. 3, pp. 1523–1532, 2013

work page 2013
[34]

Analysis the securable operation and protection coordination indices on multi-objective sitting and sizing of synchronous distributed generations on distribution networks,

H. Shad, M. Gandomkar, and J. Nikoukar, “Analysis the securable operation and protection coordination indices on multi-objective sitting and sizing of synchronous distributed generations on distribution networks,”IET Generation, Transmission & Distribution, vol. 17, no. 9, pp. 2169–2181, 2023

work page 2023
[35]

Network reconfiguration in distribution systems for loss reduction and load balancing,

M. Baran and F. Wu, “Network reconfiguration in distribution systems for loss reduction and load balancing,”IEEE Transactions on Power Delivery, vol. 4, no. 2, pp. 1401–1407, 1989. 22

work page 1989

[1] [1]

Graph-learning-assisted state estimation using sparse heterogeneous measurements,

H. Yue, W. Zhang, U. C. Yilmaz, T. Yildiz, H. Huang, H. Liu, Y. Lin, and A. Abur, “Graph-learning-assisted state estimation using sparse heterogeneous measurements,”Electric power systems research, vol. 235, p. 110644, 2024

work page 2024

[2] [2]

A review on distribution system state estimation,

A. Primadianto and C. N. Lu, “A review on distribution system state estimation,” IEEE Transactions on Power Systems, vol. 32, no. 5, pp. 3875–3883, 2016

work page 2016

[3] [3]

A survey of power system state estimation using multiple data sources: PMUs, SCADA, AMI, and beyond,

G. Cheng, Y. Lin, A. Abur, A. Gómez-Expósito, and W. Wu, “A survey of power system state estimation using multiple data sources: PMUs, SCADA, AMI, and beyond,”IEEE Transactions on Smart Grid, 2023

work page 2023

[4] [4]

Distribution network state estimation based on attention-enhanced recurrent neural network pseudo-measurement modeling,

Y. Wang, J. Gu, and L. Yuan, “Distribution network state estimation based on attention-enhanced recurrent neural network pseudo-measurement modeling,”Pro- tection and Control of Modern Power Systems, vol. 8, no. 2, pp. 1–16, 2023

work page 2023

[5] [5]

A game-theoretic data-driven approach for pseudo-measurement generation in distribution system state estima- tion,

K. Dehghanpour, Y. Yuan, Z. Wang, and F. Bu, “A game-theoretic data-driven approach for pseudo-measurement generation in distribution system state estima- tion,”IEEE Transactions on Smart Grid, vol. 10, no. 6, pp. 5942–5951, 2019. 17 Table B.5: Detailed results for IEEE 30-bus (mean±std over seeds) for different demand variability levels (Var.), measureme...

work page 2019

[6] [6]

Decision-focused learning: Foundations, state of the art, benchmark and future opportunities,

J.Mandi, J.Kotary, S.Berden, M.Mulamba, V.Bucarey, T.Guns, andF.Fioretto, “Decision-focused learning: Foundations, state of the art, benchmark and future opportunities,”Journal of Artificial Intelligence Research, vol. 80, pp. 1623–1701, 2024

work page 2024

[7] [7]

Agrawal, B

A. Agrawal, B. Amos, S. Barratt, S. Boyd, S. Diamond, and J. Z. Kolter, “Differen- 18 Table B.6: Detailed results for IEEE 33-bus (mean±std over seeds) for different demand variability levels (Var.), measurement configurations (Conf.) and all the different approaches (App.). Var. Var. Var. Conf. Conf. Conf. App. App. App. RMSEVRMSEVRMSEV RMSEθRMSEθRMSEθ R...

work page 2019

[8] [8]

Bayesian state estimation for un- observable distribution systems via deep learning,

K. R. Mestav, J. Luengo-Rozas, and L. Tong, “Bayesian state estimation for un- observable distribution systems via deep learning,”IEEE Transactions on Power Systems, vol. 34, no. 6, pp. 4910–4920, 2019. 19

work page 2019

[9] [9]

Linear single- and three-phase voltage forecasting and bayesian state estimation with limited sensing,

R. Dobbe, W. van Westering, S. Liu, D. Arnold, D. Callaway, and C. Tomlin, “Linear single- and three-phase voltage forecasting and bayesian state estimation with limited sensing,”IEEE Transactions on Power Systems, vol. 35, no. 3, pp. 1674–1683, May 2020

work page 2020

[10] [10]

Arobustdata- driven gaussian process regression-enabled distribution system state estimation using partial node data,

J.Zhang, Y.Liu, A.Rouhani, J.Zhao, G.J.Warchol, andK.Scott, “Arobustdata- driven gaussian process regression-enabled distribution system state estimation using partial node data,” in2024 IEEE Power & Energy Society General Meeting (PESGM), Jul. 2024, pp. 1–5, iSSN: 1944-9933

work page 2024

[11] [11]

Real-timepowersystemstateestimation and forecasting via deep unrolled neural networks,

L.Zhang, G.Wang, andG.B.Giannakis, “Real-timepowersystemstateestimation and forecasting via deep unrolled neural networks,”IEEE Transactions on Signal Processing, vol. 67, no. 15, pp. 4069–4077, 2019

work page 2019

[12] [12]

Physics-informed graphical neural network for pa- rameter & state estimations in power systems,

L. Pagnier and M. Chertkov, “Physics-informed graphical neural network for pa- rameter & state estimations in power systems,”arXiv preprint arXiv:2102.06349, 2021

work page arXiv 2021

[13] [13]

Powersystemstateforecastingusing machine learning techniques,

D.Mukherjee, S.Chakraborty, andS.Ghosh, “Powersystemstateforecastingusing machine learning techniques,”Electrical Engineering, vol. 104, no. 1, pp. 283–305, 2022

work page 2022

[14] [14]

A survey on state estimation techniques and challenges in smart distribution systems,

K. Dehghanpour, Z. Wang, J. Wang, Y. Yuan, and F. Bu, “A survey on state estimation techniques and challenges in smart distribution systems,”IEEE Trans- actions on Smart Grid, vol. 10, no. 2, pp. 2312–2322, 2018

work page 2018

[15] [15]

Distribution system state estimation through gaus- sian mixture model of the load as pseudo-measurement,

R. Singh, B. Pal, and R. Jabr, “Distribution system state estimation through gaus- sian mixture model of the load as pseudo-measurement,”IET generation, trans- mission & distribution, vol. 4, no. 1, pp. 50–59, 2010

work page 2010

[16] [16]

Distribution system state estima- tionusinganartificialneuralnetworkapproachforpseudomeasurementmodeling,

E. Manitsas, R. Singh, B. C. Pal, and G. Strbac, “Distribution system state estima- tionusinganartificialneuralnetworkapproachforpseudomeasurementmodeling,” IEEE Transactions on power systems, vol. 27, no. 4, pp. 1888–1896, 2012

work page 2012

[17] [17]

Robust forecasting aided power system state estimation considering state correlations,

J. Zhao, G. Zhang, Z. Y. Dong, and M. La Scala, “Robust forecasting aided power system state estimation considering state correlations,”IEEE Transactions on Smart Grid, vol. 9, no. 4, pp. 2658–2666, 2016

work page 2016

[18] [18]

Physics-guided deep learning for power system state estimation,

L. Wang, Q. Zhou, and S. Jin, “Physics-guided deep learning for power system state estimation,”Journal of Modern Power Systems and Clean Energy, vol. 8, no. 4, pp. 607–615, 2020

work page 2020

[19] [19]

Robust interval state estimation for distribution systems considering pseudo-measurement interval prediction,

X. Zhang, W. Yan, M. Huo, and H. Li, “Robust interval state estimation for distribution systems considering pseudo-measurement interval prediction,”Journal of Modern Power Systems and Clean Energy, vol. 12, no. 1, pp. 179–188, 2023. 20

work page 2023

[20] [20]

Exact modeling of non- gaussian measurement uncertainty in distribution system state estimation,

M. Vanin, T. Van Acker, R. D’hulst, and D. Van Hertem, “Exact modeling of non- gaussian measurement uncertainty in distribution system state estimation,”IEEE Transactions on Instrumentation and Measurement, vol. 72, pp. 1–11, 2023

work page 2023

[21] [21]

Matrix completion for low-observability voltage estimation,

P. L. Donti, Y. Liu, A. J. Schmitt, A. Bernstein, R. Yang, and Y. Zhang, “Matrix completion for low-observability voltage estimation,”IEEE Transactions on Smart Grid, vol. 11, no. 3, pp. 2520–2530, 2020

work page 2020

[22] [22]

Data en- hanced robust distribution system state estimation for realistic system with pa- rameter corrections,

Y. Liu, A. Rouhani, J. Zhao, J. Zhang, G. J. Warchol, and K. Scott, “Data en- hanced robust distribution system state estimation for realistic system with pa- rameter corrections,”IEEE Transactions on Power Systems, pp. 1–12, 2025

work page 2025

[23] [23]

A novel state estimation method for distribution networks with low observability based on linear AC optimal power flow model,

M. Zeraati, M. R. Sheibani, F. Jabari, and E. Heydarian-Forushani, “A novel state estimation method for distribution networks with low observability based on linear AC optimal power flow model,”Electric Power Systems Research, vol. 228, p. 110085, Mar. 2024

work page 2024

[24] [24]

WLS-based state estimation for unobservable distri- bution grids through allocation factors evaluation,

M. Pau and P. A. Pegoraro, “WLS-based state estimation for unobservable distri- bution grids through allocation factors evaluation,”IEEE Transactions on Instru- mentation and Measurement, vol. 73, pp. 1–13, 2024

work page 2024

[25] [25]

Power system robust state estimation as a layer: A novel end-to-end learning approach,

Y. Ding, W. Shi, M. Duan, Y. Zhao, J. Ruan, J. Zhao, and Z. Xu, “Power system robust state estimation as a layer: A novel end-to-end learning approach,”arXiv preprint arXiv:2511.22836, 2025

work page arXiv 2025

[26] [26]

Abur and A

A. Abur and A. Gómez-Expósito,Power system state estimation: theory and im- plementation, ser. Power engineering. New York, NY: Marcel Dekker, 2004

work page 2004

[27] [27]

Robust and scalable power system state estimationviacompositeoptimization,

G. Wang, G. B. Giannakis, and J. Chen, “Robust and scalable power system state estimationviacompositeoptimization,”IEEE Transactions on Smart Grid, vol.10, no. 6, pp. 6137–6147, Nov. 2019

work page 2019

[28] [28]

State estimation in power systems based on a mathe- matical programming approach,

E. Caro and A. Hernández, “State estimation in power systems based on a mathe- matical programming approach,” inAdvances in Electric Power and Energy. John Wiley & Sons, Ltd, 2020, pp. 23–57

work page 2020

[29] [29]

Power system observability with minimal phasor measurement placement,

T. Baldwin, L. Mili, M. Boisen, and R. Adapa, “Power system observability with minimal phasor measurement placement,”IEEE Transactions on Power Systems, vol. 8, no. 2, pp. 707–715, 1993

work page 1993

[30] [30]

Learning-based state estima- tion in distribution systems with limited real-time measurements,

J. De la Varga, S. Pineda, J. Morales, and A. Porras, “Learning-based state estima- tion in distribution systems with limited real-time measurements,”Electric Power Systems Research, vol. 241, p. 111268, 2025

work page 2025

[31] [31]

State estimation in two time scales for smart distribution systems,

A. Gómez-Expósito, C. Gómez-Quiles, and I. Džafić, “State estimation in two time scales for smart distribution systems,”IEEE Transactions on Smart Grid, vol. 6, no. 1, pp. 421–430, Jan. 2015. 21

work page 2015

[32] [32]

NewYork: Springer, 2006

J.NocedalandS.J.Wright,Numerical Optimization, 2nded. NewYork: Springer, 2006

work page 2006

[33] [33]

A protec- tion coordination index for evaluating distributed generation impacts on protection for meshed distribution systems,

H. H. Zeineldin, Y. A.-R. I. Mohamed, V. Khadkikar, and V. R. Pandi, “A protec- tion coordination index for evaluating distributed generation impacts on protection for meshed distribution systems,”IEEE Transactions on Smart Grid, vol. 4, no. 3, pp. 1523–1532, 2013

work page 2013

[34] [34]

Analysis the securable operation and protection coordination indices on multi-objective sitting and sizing of synchronous distributed generations on distribution networks,

H. Shad, M. Gandomkar, and J. Nikoukar, “Analysis the securable operation and protection coordination indices on multi-objective sitting and sizing of synchronous distributed generations on distribution networks,”IET Generation, Transmission & Distribution, vol. 17, no. 9, pp. 2169–2181, 2023

work page 2023

[35] [35]

Network reconfiguration in distribution systems for loss reduction and load balancing,

M. Baran and F. Wu, “Network reconfiguration in distribution systems for loss reduction and load balancing,”IEEE Transactions on Power Delivery, vol. 4, no. 2, pp. 1401–1407, 1989. 22

work page 1989