Pith. sign in

REVIEW 4 major objections 5 minor 27 references

Towards AC Feasibility of DCOPF Dispatch

T0 review · 4 major / 5 minor · reviewed 2026-08-03 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read A loss-augmented DCOPF followed by a structured AC power flow — distributed slack plus PV/PQ switching — restores AC-feasible operating points from DC dispatches, cutting cost deviation by 93% and mean error by 75% on a 13,659-bus system.

desk verdict Solid empirical ranking worth refereeing, but the abstract overclaims 'all constraints' when the paper's own tables show residual violations. read the letter →

arxiv 2511.14725 v2 pith:3TJFHYFV submitted 2025-11-18 eess.SY cs.SY

Michael A. Boateng , Russell Bent , Sidhant Misra , Parikshit Pareek , Pascal Van Hentenryck , Daniel Molzahn This is my paper · ORCID
classification eess.SYcs.SY
keywords DCoptimalpowerflowACfeasibilityrestorationdistributedslackbusPV/PQswitchingloss-augmentedDCOPFreactivelimitsNewton-Raphsonloaduncertainty
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

DC optimal power flow is fast and widely used, but its lossless, reactive-power-blind model produces dispatches that cannot satisfy the nonlinear AC power flow equations. The paper asks how often such dispatches can be repaired cheaply by solving an AC power flow from the DC setpoints, and which DC and AC formulations repair them best. Across more than 10,000 perturbed load scenarios on systems up to 13,659 buses, it finds that pairing a loss-augmented DC model — a convex quadratic line-loss formulation — with a structured AC power flow that shares real-power imbalance among generators by headroom and switches generator buses between PV and PQ modes with a tolerance deadband yields operating points that meet the AC power flow equations and all engineering limits. On the largest case, cost deviation and mean generator error drop by 93% and 75% relative to single-slack-bus baselines, and extreme-loading violations fall by a factor of 3 to 5. The paper's core claim is that AC infeasibility of DCOPF solutions, though theoretically guaranteed, is usually easy to restore in practice with the right pipeline.

What carries the argument

The load-bearing object is the headroom-based distributed slack model: each generator's share of the real-power mismatch is proportional to its remaining upward capacity, with a capacity-proportional fallback if no headroom exists. This replaces the conventional single slack bus, which would force one unit to absorb all DC-to-AC loss and mismatch and often exceed its limit. The second mechanism is tolerance-based PV/PQ switching, which lets generators leave voltage-control mode when reactive output saturates and return only when voltage moves beyond a deadband, preventing oscillatory chattering. The combination, ACSPF, is what allows the Newton–Raphson solver to converge on dispatches that a

What would settle it

Run the DCLQCP-to-ACSPF pipeline over thousands of load scenarios, record every sample including non-converged ones, and check each recovered setpoint with an independent AC power flow using a different, realistic slack-participation rule and tight tolerances; if a non-negligible share of scenarios fail to converge or show reactive, voltage, or thermal limit violations under that independent check, the feasibility claim is falsified.

Watch

Extended reading notes

Core claim

The central discovery is empirical: although DCOPF solutions are provably never AC-feasible, a structured ACPF can almost always restore feasibility with only small adjustments to generator setpoints, provided the ACPF models generator behavior realistically. The paper defines a four-variant ACPF family — single-slack, PV/PQ switching only, headroom-based distributed slack only, and the full structured model combining both — and couples each with four DCOPF variants: lossless, linear-loss-factor, convex quadratic line-loss, and linear outer-approximation. Evaluating violation counts, maxima, summed severities, cost deviation, mean absolute error, iterations, and solve time, the authors find

Load-bearing premise

The pipeline's feasibility claims rest on the structured ACPF itself being a faithful feasibility oracle: the same Newton–Raphson solver that recovers setpoints also certifies them, and only converged load samples are reported, so if non-converged scenarios or real generator imbalance-sharing behavior differ from the headroom-based slack model, the recovered operating points may not be AC-feasible in practice.

Editorial extensions

If this is right

  • Loss-augmented DC models, especially the convex quadratic line-loss formulation DCLQCP, produce DC setpoints that are much closer to AC feasibility than lossless DCOPF; the pipeline's success depends on this pre-accounting of losses.
  • Using distributed slack in the AC recovery eliminates active-power violations and sharply reduces cost deviation, because no single generator is forced to absorb the full DC-AC mismatch.
  • Reactive-power feasibility requires PV/PQ switching: distributed slack alone leaves hundreds of reactive-power-limit violations on large systems, while the structured model drives them to zero.
  • The DCLQCP-to-ACSPF pipeline can serve as a post-processing step after DCOPF-based market clearing or planning to produce deliverable, AC-feasible dispatches at a fraction of the full ACOPF solve cost.
  • Under extreme loading, the pipeline reduces inequality-constraint violations by a factor of 3 to 5, suggesting it is robust precisely in the stressed regimes where operators worry most.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • Because the paper reports only load samples that converged, the practical success rate under heavily stressed scenarios may be lower than the headline numbers suggest; counting and characterizing the non-converged samples would make the claim more operationally actionable.
  • If the empirical pattern generalizes, market software could adopt DCLQCP-to-ACSPF as a deliverability post-processor: keep the fast DCOPF for pricing and then run a structured AC power flow to certify that the cleared dispatch is physically realizable.
  • The headroom-based participation rule is a modeling choice; calibrating it to real automatic-generation-control shares would test whether the recovered setpoints match what generators would actually do when called upon.
  • The recovered setpoints are AC-feasible but not necessarily ACOPF-optimal; a small ACOPF warm-started from the ACSPF point could close the remaining cost gap while retaining feasibility.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 5 minor

Summary. The paper studies DCOPF-to-ACPF feasibility restoration pipelines. It defines four DCOPF variants (lossless DCBASE, linearized-loss LLLF, convex quadratic-loss LQCP, outer-approximation LLOA) and four ACPF 'recovery' variants (single-slack ACBASE, PV/PQ-switching ACBTS, distributed-slack ACDS, and the combined structured ACSPF with headroom-based distributed slack and tolerance-based PV/PQ switching). Using standard MATPOWER/ACTIVSg/PEGASE/RTE test systems and 1000 Gaussian load-perturbation samples per system (σ=5–15%), it reports violations of active/reactive power, voltage, and thermal limits, plus cost difference and mean-absolute-error metrics. The paper's central claim is that the structured ACPF, especially when paired with DCLQCP, restores AC feasibility and that this pipeline reduces cost difference and MAE by 93% and 75% on the 13,659-bus case compared to single-slack baselines. The stated most-effective pipeline is DCLQCP→ACSPF.

Significance. The empirical question addressed — how often DCOPF dispatches can be reconciled to AC-feasible operating points with realistic generator limits and distributed slack — is practically relevant, and the paper provides a broad comparison over standard large-scale test cases. The strengths are the scale of the study (10+ systems, 10,000+ scenarios), the use of standard benchmarks, and the systematic comparison of four DCOPF and four ACPF variants. If the claims are properly qualified by success rates and violation counts, the comparison would be a useful reference for practitioners. However, the headline claim that the structured ACPF yields solutions satisfying 'all engineering inequality constraints' is contradicted by the paper's own tables, and the lack of convergence-rate reporting prevents a reliable quantitative assessment. The contribution is therefore currently valuable but overstated.

major comments (4)
  1. [Abstract; Tables I–II; Section III-B] The abstract and conclusion assert that the structured ACPF yields solutions satisfying 'all engineering inequality constraints.' This is contradicted by the paper's own data. In Table I, the ACSPF rows show nonzero voltage violations for case_118/DCLLLF, case_1354/DCLLLF and DCLLOA, and case_2869/DCLLLF, and nonzero thermal violations for case_1354/DCLLLF and case_13659/DCLLLF. Table II reports a maximum thermal violation of 15.65% for case_1354/DCLLLF and case_13659/DCLLLF under ACSPF. Section III-B further states that 'Full feasibility was not restored only for case_2000.' Thus the universal feasibility claim is not supported by the reported results. The claim must be downgraded to a success-rate/violation-reduction statement, with per-case, per-pipeline counts of remaining violations.
  2. [Section III, paragraph beginning 'All results reported...'] The paper reports only samples that converged, without giving the number or proportion of diverged samples. A pipeline that fails on a nontrivial fraction of scenarios cannot be described as 'restoring AC feasibility' without that success rate. The duplicated text in Section II also states that the solver 'resolves 90% of stress-test scenarios that a flat-start, single-slack ACPF cannot solve,' which implies 10% non-convergence in some settings. Please report, for each DC variant, AC variant, and test case, the number of converged versus total samples, and the per-sample feasibility status. This is essential for a fair comparison and for the claimed 'most effective and consistent pipeline' conclusion.
  3. [Section II-B, Model 2] The recovery mechanism guarantees active-power balance by construction because the distributed-slack equations (5a–5c) adjust every generator's active-power setpoint. Zero active-power violations under ACDS/ACSPF is therefore partly a modeling artifact rather than an independent feasibility result. Meanwhile, voltage and thermal limits are not enforced by the ACPF and do remain violated in Tables I–II. The paper should characterize the pipeline as 'violation reduction' rather than 'restoration,' or at least define 'restoration' as 'restoration with respect to the constraints enforced by the structured ACPF models.' Otherwise the claim that the ACPF 'satisfies' all ACOPF inequality constraints is misleading.
  4. [Section II-B and Section III] The same custom Newton-Raphson ACPF is used both to perform the recovery and to certify AC feasibility, and the ACSPF model uses a headroom-based distributed slack (Eqs. 5a–5c) that is a modeling choice, not a standard market or operational convention. As a sanity check, and to strengthen the claim that the recovered setpoints are genuinely AC-feasible, the authors should evaluate the final setpoints under an independent single-slack ACOPF or a standard MATPOWER ACPF and report the resulting violation counts and magnitudes. This would address the concern that the recovered operating point may be feasible only for the custom distributed-slack/PV-PQ model, not for the standard ACOPF constraints used as the reference.
minor comments (5)
  1. [Tables I–II] The column alignment in Tables I and II is garbled in the manuscript, particularly for case_2869 and case_13659, where thermal-violation counts and values are difficult to parse. Please reformat the tables (e.g., one table per AC variant, or separate active/reactive/voltage/thermal sub-tables) so that each entry is unambiguous.
  2. [Section II and repeated text] Algorithms 1a/1b and 2, and the accompanying descriptions of bus-switching logic and distributed slack, appear twice in the manuscript with slight differences. Please consolidate these repeated passages into a single, complete description.
  3. [Section III-B, Table III] The test system is referred to as both 'Texas_2000' and 'case_2000' (with a footnote marker ⋄). Use a single name consistently. Also, the 'Improv. (%)' columns in Table III list three percentages (presumably relative to ACBASE, ACBTS, ACDS) but the column headers do not make this clear; please label them explicitly.
  4. [Abstract and Section III] The phrase 'all engineering inequality constraints' is used without specifying which constraints are included. The ACOPF constraints in Fig. 1 appear to be active-power limits, reactive-power limits, voltage limits, and thermal limits. Please state this explicitly and, in the numerical results, define what counts as a violation (e.g., tolerance, whether limits are enforced as hard or soft).
  5. [Fig. 6 caption] The caption says the bottom plots show 'mean reactive power violations,' but the y-axis label in the figure says 'Reactive Power Violations [Count]'. Clarify whether the plotted quantity is a count or a magnitude, and ensure the caption matches the axis.

Circularity Check

1 steps flagged · score 1.0 of 10

No significant circularity: empirical pipeline comparison grounded in independent ACOPF references and standard test cases; one minor self-citation of prior loss models is not load-bearing.

  1. self citation load bearing [Section II-A (DCOPF with Line Loss Approximation), Eqs. (2)-(4); references [16], [20]]
    "To improve the accuracy of the DCOPF while preserving tractability, several loss-augmented DCOPF variants have been developed in the literature, three of which are described next. 1) Linear Line Loss Factor Model (LLLF)[16], [20] ... 2) Line Loss Quadratic Convex Program (LQCP)[16] ... 3) Line Loss Outer Approximation (LLOA)[16]"

    The three loss-augmented DCOPF models whose relative performance drives the headline conclusion (DCLQCP->ACSPF is 'the most effective and consistent pipeline') are adopted from [16], prior work co-authored by P. Van Hentenryck, a co-author of this paper. This is a minor self-citation, but it is not load-bearing: the models are restated explicitly in Eqs. (2)-(4), the paper's contribution is the empirical DCOPF->ACPF pipeline comparison, and the pipeline ordering is evaluated against independently computed ACOPF references (PowerModels.jl) on standard MATPOWER/ACTIVSg/PEGASE/RTE test cases. No fitted parameter is renamed as a prediction, and no result is forced by the self-citation.

full rationale

The paper's derivation chain is empirical rather than theoretical: four DCOPF variants produce setpoints, a structured Newton-Raphson ACPF (distributed slack, PV/PQ switching) reconciles them, and the results are scored by violations, cost difference (CD), and mean absolute error (MAE) against independently computed ACOPF solutions via PowerModels.jl on standard public test systems. The central ranking (DCLQCP -> ACSPF is most consistent) is therefore grounded in an external benchmark, not in the paper's own definitions. The only self-citation of note is [16] for the LLLF/LQCP/LLOA loss models; since these are stated in the paper and the comparison is external, this is a minor, non-load-bearing citation and does not constitute circularity. Several limitations were flagged in-scope but do not amount to circular derivation: (i) the abstract claims solutions satisfy 'all engineering inequality constraints,' yet Tables I-II show nonzero thermal and voltage violations for ACSPF (e.g., case_1354/DCLLLF thermal = 13; case_2869/DCLLLF thermal = 61; Table II shows 15.65% max thermal violation on case_13659/DCLLLF), and Table III's case_2000 note says it 'incurred violations of inequality constraints,' with Section III-B admitting 'Full feasibility was not restored only for case_2000'; (ii) the same custom ACPF serves as both recovery mechanism and feasibility certifier, while 'All results reported correspond to samples that converged' without reporting divergence counts, so success rates are not quantified. These are correctness/validity and reporting concerns, not circularity of the derivation chain.

Assumptions & free parameters 4 free parameters · 6 assumptions · 0 invented entities

The paper is empirical and rests on standard power-flow models. The main extra assumptions are the realism of the custom ACPF's distributed-slack/PV-PQ model and representativeness of the sampled scenarios; no new physical entities are introduced.

free parameters (4)
  • LLLF/LLOA linearization reference flows p_ref = reference DC operating point
    Loss approximations in LLLF and LLOA are expanded around a reference DC flow (Section II-A, Eqs. 2 and 4); the choice affects loss estimates and thus the dispatch.
  • PV/PQ switching tolerances epsilon_q, epsilon_v = 1e-4 p.u., 1e-5 p.u.
    Set by hand in Section III; the deadband determines when reactive/voltage violations count and prevents switching chatter.
  • Load-perturbation scale sigma and power factor range = sigma in {5%,15%}, pf in [0.95,1.0]
    Used to generate the >10,000 scenarios; claims about 'extreme loading' generalize only if this distribution is representative of real operating stress.
  • Newton solver convergence tolerance = 1e-6 p.u.
    Controls when a power-flow solution is deemed converged and therefore when feasibility is certified.
assumptions (6)
  • domain assumption Nonzero-loss DCOPF solutions are AC-infeasible (Baker [14])
    Motivates the need for recovery; cited from [14], not re-derived in this paper.
  • domain assumption Newton-Raphson ACPF with distributed slack and PV/PQ switching converges to the relevant AC power flow solution
    The custom solver is the feasibility oracle; Section II-B and Section III.
  • ad hoc to paper Headroom-based slack sharing is a realistic generator response model
    Model 2 Eqs. (5a)-(5c); a modeling choice with no field validation.
  • domain assumption Sampled load perturbations represent practically relevant stress conditions
    Gaussian sigma in {5%,15%} and pf in [0.95,1.0] are chosen by the authors; sensitivity claims rely on this.
  • domain assumption PowerModels.jl ACOPF solutions are valid ground-truth references
    ACOPF is nonconvex; the solver's local solution may not be global, and MAE/cost-difference metrics depend on this reference.
  • ad hoc to paper Excluded non-converged samples do not bias the feasibility results
    The paper says 'All results reported correspond to samples that converged' but gives no convergence rates.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Towards AC Feasibility of DCOPF Dispatch." pith.science (2026). https://pith.science/paper/3TJFHYFV

@misc{pith2026251114725,
  author       = {Pith},
  title        = {Pith review of: Towards AC Feasibility of DCOPF Dispatch},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/3TJFHYFV}},
  note         = {Machine review of arXiv:2511.14725}
}
read the original abstract

DC Optimal Power Flow (DCOPF) is widely utilized in power system operations due to its simplicity and computational efficiency. However, its lossless, reactive power-agnostic model often yields dispatches that are infeasible under practical operating scenarios such as the nonlinear AC power flow (ACPF) equations. While theoretical analysis demonstrates that DCOPF solutions are inherently AC-infeasible, their widespread industry adoption suggests substantial practical utility. This paper develops a unified DCOPF-ACPF pipeline to recover AC feasible solutions from DCOPF-based dispatches. The pipeline uses four DCOPF variants and applies AC feasibility recovery using both distributed slack allocation and PV/PQ switching. The main objective is to identify the most effective pipeline for restoring AC feasibility. Evaluation across over 10,000 dispatch scenarios on various test cases demonstrates that the structured ACPF model yields solutions that satisfy both the ACPF equations, and all engineering inequality constraints. In a 13,659 bus case, the mean absolute error and cost differences between DCOPF and ACOPF are reduced by 75% and 93%, respectively, compared to conventional single slack bus methods. Under extreme loading conditions, the pipeline reduces inequality constraint violations by a factor of 3 to 5.

Figures

Figures reproduced from arXiv: 2511.14725 by the authors.

Figure 1
Figure 1. Pipeline of the DCOPF→ACPF model. The DCOPF gives p sp g and θ dc from input pd. The ACPF performs feasibility checks with distributed slack using participation factors (πg). Voltage initialization (v int) aids in convergence. ACOPF constraints are marked with red. with distributed slack and PV/PQ switching. The stages are coupled: the DCOPF provides generator setpoints, and the ACPF applies minimal adjustments to o… view at source ↗
Figure 3
Figure 3. Comparison of the sum total of violations in different AC power flow formulations across various cases: [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figure 5
Figure 5. Pipeline comparison across different DC and AC variants, over load uncertainty with [PITH_FULL_IMAGE:figures/full_fig_p007_5.png] view at source ↗
Figures from the paper (1 more)
Figure 6
Figure 6. Figure 6: Interplay between voltage magnitude and reactive power for various pipelines. The plots are taken over load uncertainty with σ = 5%, for 1000 samples per case. The top box plots (mean with min–max extremes) show the voltage distributions for three cases (case_118, case…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

27 extracted references · 1 linked inside Pith

  1. [1]

    A comparison of AC and DC power flow models for contingency and reliability analysis,

    H. Kile, K. Uhlen, L. Warland, and G. Kjølle, “A comparison of AC and DC power flow models for contingency and reliability analysis,” in 18th Power Systems Computation Conference (PSCC), Poland, 2014. 24th Power Systems Computation Conference PSCC 2026 Limassol, Cyprus — June 8-12, 2026 case 118 case 1354 case 2869 case 9241 case 13659Test Cases 101 102 1...

    2014

  2. [2]

    Effectiveness of the distribution factor approxima- tions used in congestion modeling,

    M. Liu and G. Gross, “Effectiveness of the distribution factor approxima- tions used in congestion modeling,” in14th Power Systems Computation Conference (PSCC), Seville, Spain, June 2002

    2002

  3. [3]

    A comparison of the AC and DC power flow models for LMP calculations,

    T. J. Overbye, X. Cheng, and Y . Sun, “A comparison of the AC and DC power flow models for LMP calculations,” in37th Hawaii International Conference on System Sciences (HICSS), January 2004

    2004

  4. [4]

    Usefulness of DC power flow for active power flow analysis,

    K. Purchala, L. Meeus, D. Van Dommelen, and R. Belmans, “Usefulness of DC power flow for active power flow analysis,” inIEEE Power & Energy Society General Meeting, June 2005, pp. 454–459

    2005

  5. [5]

    DCOPF-based LMP simulation: Algorithm, comparison with ACOPF, and sensitivity,

    F. Li and R. Bo, “DCOPF-based LMP simulation: Algorithm, comparison with ACOPF, and sensitivity,”IEEE Transactions on Power Systems, vol. 22, no. 4, pp. 1475–1485, November 2007

    2007

  6. [6]

    DC power flow revisited,

    B. Stott, J. Jardim, and O. Alsa c ¸, “DC power flow revisited,”IEEE Transactions on Power Systems, vol. 24, no. 3, pp. 1290–1300, 2009

    2009

  7. [7]

    Impact of assumptions on DC power flow model accuracy,

    Y . Qi, D. Shi, and D. Tylavsky, “Impact of assumptions on DC power flow model accuracy,” inNorth American Power Symposium (NAPS), September 2012

    2012

  8. [8]

    Optimizing parameters of the DC power flow,

    B. Taheri and D. K. Molzahn, “Optimizing parameters of the DC power flow,” vol. 235, no. 110719, October 2024, presented at the23rd Power Systems Computation Conference (PSCC)

    2024

Show all 27 references
  1. [9]

    Accurate load and generation scheduling for linearized DC models with contingencies,

    C. Coffrin, P. Van Hentenryck, and R. Bent, “Accurate load and generation scheduling for linearized DC models with contingencies,” inIEEE Power & Energy Society General Meeting, July 2012

    2012

  2. [10]

    Comparing the effects of failures in power grids under the AC and DC power flow models,

    H. Cetinay, S. Soltan, F. A. Kuipers, G. Zussman, and P. Van Mieghem, “Comparing the effects of failures in power grids under the AC and DC power flow models,”IEEE Transactions on Network Science and Engineering, vol. 5, no. 4, pp. 301–312, October 2018

    2018

  3. [11]

    Analysis of linearized decoupled power flow approximations for steady-state security assessment,

    R. Kaye and F. F. Wu, “Analysis of linearized decoupled power flow approximations for steady-state security assessment,”IEEE Transations on Circuits and Systems, vol. 31, no. 7, pp. 623–636, July 1984

    1984

  4. [12]

    On the existence and linear approximation of the power flow solution in power distribution networks,

    S. Bolognani and S. Zampieri, “On the existence and linear approximation of the power flow solution in power distribution networks,”IEEE Transactions on Power Systems, vol. 31, no. 1, pp. 163–172, 2016

    2016

  5. [13]

    Error bounds on the DC power flow approximation: A convex relaxation approach,

    K. Dvijotham and D. K. Molzahn, “Error bounds on the DC power flow approximation: A convex relaxation approach,” in55th IEEE Conference on Decision and Control (CDC), 2016, pp. 2411–2418

    2016

  6. [14]

    Solutions of DC OPF are never AC feasible,

    K. Baker, “Solutions of DC OPF are never AC feasible,” inACM e- Energy, 2021

    2021

  7. [15]

    Learning optimal solutions for extremely fast ACOPF,

    A. Zamzam and K. Baker, “Learning optimal solutions for extremely fast ACOPF,” inIEEE SmartGridComm, 2020

    2020

  8. [16]

    A linear outer approxima- tion of line losses for DC-based optimal power flow problems,

    H. Zhao, M. Tanneau, and P. Van Hentenryck, “A linear outer approxima- tion of line losses for DC-based optimal power flow problems,”Electric Power Systems Research, vol. 212, p. 108272, 2022, presented at the 22nd Power Systems Computation Conference (PSCC)

    2022

  9. [17]

    Continuously differentiable analytical models for implicit control within power flow,

    A. Agarwal, A. Pandey, M. Jereminov, and L. Pileggi, “Continuously differentiable analytical models for implicit control within power flow,” arXiv:1811.02000, Nov. 2018

    2018 arXiv

  10. [18]

    AC feasibility restoration in market clearing: Problem formulation and improvement,

    X. Fang, Z. Yang, J. Yu, and Y . Wang, “AC feasibility restoration in market clearing: Problem formulation and improvement,”IEEE Transactions on Industrial Informatics, vol. 18, no. 11, 2022

    2022

  11. [19]

    AC power flow feasibility restoration via a state estimation-based post-processing algorithm,

    B. Taheri and D. K. Molzahn, “AC power flow feasibility restoration via a state estimation-based post-processing algorithm,”Electric Power Systems Research, vol. 235, no. 110642, Oct. 2024, presented at the23rd Power Systems Computation Conference (PSCC)

    2024

  12. [20]

    Approximating line losses and apparent power in AC power flow linearizations,

    C. Coffrin, P. Van Hentenryck, and R. Bent, “Approximating line losses and apparent power in AC power flow linearizations,” inIEEE Power and Energy Society General Meeting, 2012

    2012

  13. [21]

    A. J. Wood and B. F. Wollenberg,Power Generation, Operation, and Control, 2nd ed. New York: Wiley, 1996

    1996

  14. [22]

    Iterative DCOPF model using distributed slack bus,

    V . N. Bharatwaj, A. R. Abhyankar, and P. R. Bijwe, “Iterative DCOPF model using distributed slack bus,” inIEEE Power and Energy Society General Meeting, 2012

    2012

  15. [23]

    On PV–PQ bus type switching logic in power flow computation,

    J. Zhao, H. Chiang, P. Ju, and H. Li, “On PV–PQ bus type switching logic in power flow computation,” in16th Power Systems Computation Conference (PSCC), Glasgow, Scotland, UK, 2008

    2008

  16. [24]

    On the accuracy of power flow and load margin calculation caused by incorrect logical PV/PQ switching: Analytics and improved methods,

    L. Zeng, H.-D. Chiang, L. S. Neves, and L. F. C. Alberto, “On the accuracy of power flow and load margin calculation caused by incorrect logical PV/PQ switching: Analytics and improved methods,”International Journal of Electrical Power & Energy Systems, vol. 147, 2023

    2023

  17. [25]

    Matpower: Steady-state operations, planning, and analysis tools for power systems research and education,

    R. D. Zimmerman, C. E. Murillo-S ´anchez, and R. J. Thomas, “Matpower: Steady-state operations, planning, and analysis tools for power systems research and education,”IEEE Transactions on Power Systems, vol. 26, no. 1, pp. 12–19, 2011

    2011

  18. [26]

    PowerModels.jl: An open-source framework for exploring power flow formulations,

    C. Coffrin, R. Bent, K. Sundar, Y . Ng, and H. Wang, “PowerModels.jl: An open-source framework for exploring power flow formulations,” in 20th Power Systems Computation Conference (PSCC), 2018

    2018

  19. [27]

    pandapower—An open-source Python tool for convenient modeling, analysis, and optimization of electric power systems,

    L. Thurner, A. Scheidler, F. Sch ¨afer, J.-H. Menke, J. Dollichon, F. Meier, and J. M. Myrzik, “pandapower—An open-source Python tool for convenient modeling, analysis, and optimization of electric power systems,”IEEE Transactions on Power Systems, vol. 33, p. 6510, 2018. 24th...

    2018

Pith tools

Reviewed August 3, 2026 · model on record in the stance chip above.