pith. sign in

arxiv: 2604.07106 · v1 · submitted 2026-04-08 · 📡 eess.SY · cs.SY

Decision-focused Conservation Voltage Reduction to Consider the Cascading Impact of Forecast Errors

Qintao Du , Ran Li , Weiyi Lv , Huan Zhou , Moduo Yu , Jianzhe Liu This is my paper

Pith reviewed 2026-05-10 17:58 UTC · model grok-4.3

classification 📡 eess.SY cs.SY
keywords conservation voltage reductionvolt-var controlbi-level optimizationdecision-focused forecastingmulti-stage controlforecast error propagationdistribution system optimization
0
0 comments X

The pith

Bi-level forecasting framework accounts for cascading forecast errors to improve multi-stage conservation voltage reduction.

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

The paper aims to establish that conventional sequential forecasting for conservation voltage reduction ignores how prediction errors in one stage affect later control decisions, leading to suboptimal schedules for slow-acting devices and poorer coordination with fast-acting ones. By training forecasting models inside a bi-level structure that includes the full multi-stage Volt-Var Control optimization, the new framework learns trade-offs across day-ahead, intra-day, and real-time horizons directly. This matters for distribution systems because it can raise energy savings while preserving voltage security, rather than forcing an explicit compromise. The approach is made tractable by a modified sensitivity-driven integer L-shaped solver that feeds both numerical sensitivities for discrete variables and analytical duals for continuous parameters back into training.

Core claim

The central claim is that embedding the downstream multi-stage VVC optimization inside upstream forecasting model training via a bi-level multi-timescale forecasting (Bi-MTF) framework allows the models to internalize the cascading impact of forecast errors; the resulting day-ahead schedules for OLTCs and CBs then support more effective real-time coordination with SVGs and PV inverters, producing a better Pareto front between voltage security and CVR efficiency than MSE-trained sequential methods.

What carries the argument

The bi-level multi-timescale forecasting (Bi-MTF) framework that integrates downstream multi-stage VVC optimization into upstream forecast training, solved by a modified sensitivity-driven integer L-shaped method with hybrid gradient feedback combining numerical sensitivity analysis for discrete variables and analytical dual information for continuous parameters.

If this is right

  • Energy savings increase from 2.74% to 3.41% as fast-acting device capacity grows, versus only 1.50% to 1.76% for conventional methods.
  • Day-ahead schedules for slow-acting devices become more compatible with real-time fast-acting adjustments, reducing the security-efficiency trade-off.
  • The Pareto front between real-time voltage security and CVR efficiency improves across multiple timescales.
  • The hybrid gradient feedback mechanism keeps the bi-level problem computationally tractable on distribution networks.

Where Pith is reading between the lines

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

  • The same decision-focused training structure could be applied to other hierarchical control problems in power systems where forecast errors propagate through multiple decision stages.
  • End-to-end integration of optimization into forecasting may allow operators to operate closer to voltage limits without added risk when uncertainty is high.
  • Extending the solver to stochastic or robust variants could further reduce the need for conservative margins in networks with high renewable penetration.

Load-bearing premise

The modified sensitivity-driven integer L-shaped method can reliably solve the bi-level problem to optimality or near-optimality and that the single modified IEEE 33-bus test case with its chosen device capacities and error distributions is representative of real distribution systems.

What would settle it

Applying the Bi-MTF method to a larger real distribution network or different error distributions and finding that energy savings no longer exceed those of conventional MSE-based sequential methods by a comparable margin.

Figures

Figures reproduced from arXiv: 2604.07106 by Huan Zhou, Jianzhe Liu, Moduo Yu, Qintao Du, Ran Li, Weiyi Lv.

Figure 10
Figure 10. Figure 10: Nodal voltage topology corresponding to Point B (abundant SVG capacity, 𝑄 𝑆𝑉𝐺,𝑚𝑎𝑥 = 0.6MVar). (a) The BASE method results in a con￾servative high-voltage profile (red/orange nodes). (b) The proposed BI-MTF method effectively lowers the global voltage profile [PITH_FULL_IMAGE:figures/full_fig_p011_10.png] view at source ↗
read the original abstract

Conservation Voltage Reduction (CVR) relies on the effective coordination of slow-acting devices, such as OLTCs and CBs, and fast-acting devices, such as SVGs and PV inverters, typically implemented through a hierarchical multi-stage Volt-Var Control (VVC) spanning day-ahead scheduling, intra-day dispatch, and real-time control. However, existing sequential methods fail to account for the cas-cading impact of forecast errors on multi-stage decision-making. This oversight results in suboptimal day-ahead schedules for OLTCs and CBs that hinder the ef-fective coordination with fast-acting SVGs and inverters, inevitably driving a trade-off between real-time voltage security and CVR efficiency. To improve the Pareto front of this trade-off, this paper proposes a novel bi-level multi-timescale forecasting (Bi-MTF) framework for multi-stage VVC optimization. By integrating the downstream multi-stage VVC optimization into the upstream forecasting mod-els training, the decision-focused forecasting models are able to learn the trade-offs across temporal horizons. To solve the computationally challenging bi-level for-mulation, a modified sensitivity-driven integer L-shaped method is developed. It utilizes a hybrid gradient feedback mechanism that integrates numerical sensitivity analysis for discrete variables with analytical dual information for continuous fore-cast parameters to ensure tractability. Numerical results on a modified IEEE 33-bus system demonstrate that the proposed approach yields superior energy savings and operational safety compared to conventional MSE-based sequential paradigms. Specifically, as the capacity of fast-acting devices increases, the energy savings of the proposed method rise from 2.74% to 3.41%, which is far superior to the 1.50% to 1.76% achieved by conventional MSE-based sequential paradigms.

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.

Referee Report

3 major / 2 minor

Summary. The manuscript proposes a bi-level multi-timescale forecasting (Bi-MTF) framework for multi-stage Volt-Var Control in Conservation Voltage Reduction. It integrates downstream multi-stage VVC optimization into upstream forecasting model training via decision-focused learning to account for cascading forecast errors on day-ahead OLTC/CB schedules and their coordination with fast-acting SVGs/PV inverters. A modified sensitivity-driven integer L-shaped method with hybrid gradient feedback (numerical sensitivities for discrete variables, analytical duals for continuous parameters) is developed to solve the bi-level problem, and experiments on a modified IEEE 33-bus system report superior energy savings (2.74% to 3.41% as fast-device capacity grows) and operational safety versus MSE-based sequential baselines (1.50% to 1.76%).

Significance. If the solver reliably attains near-optimal solutions, the decision-focused approach offers a meaningful advance by directly optimizing the multi-stage operational objective rather than an intermediate loss, potentially tightening the efficiency-security trade-off in hierarchical VVC. The hybrid feedback mechanism for mixed discrete-continuous bi-level problems is a technical contribution, and the concrete percentage gains on a standard test system provide a falsifiable benchmark.

major comments (3)
  1. [Numerical Results] Numerical Results section: The headline energy savings (2.74–3.41 % versus 1.50–1.76 % for the MSE baseline) are reported without optimality gaps, iteration limits, duality gaps, or sub-optimality measures from the modified sensitivity-driven integer L-shaped solver. Because both the proposed Bi-MTF model and the sequential baseline are solved with the same method, the absence of these diagnostics prevents confirming that the Pareto-front improvement is attributable to decision-focused training rather than solver artifacts or inconsistent convergence.
  2. [Method] Method section (bi-level formulation and solver): The hybrid gradient feedback is claimed to ensure tractability by mixing numerical sensitivity analysis for discrete OLTC/CB variables with analytical dual information for continuous forecast parameters, yet no error analysis, approximation bounds, or small-instance exact-solver benchmark is provided. This is load-bearing for the central claim that the bi-level problem is solved reliably enough to support the reported operational gains.
  3. [Experimental Setup] Experimental Setup: Results are shown only for a single modified IEEE 33-bus system with chosen device capacities and error distributions; no ablation on the sensitivity mechanism, no sensitivity to forecast error magnitude, and no error bars or multiple-run statistics accompany the energy-savings figures. These omissions weaken the robustness of the claim that the approach yields superior performance as fast-acting device capacity increases.
minor comments (2)
  1. [Abstract] Abstract contains hyphenation artifacts ('cas-cading', 'mod-el', 'fore-cast') that should be corrected.
  2. Notation for the bi-level variables and the hybrid gradient terms could be clarified with an explicit table of symbols to aid readability.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for the constructive and detailed comments, which help improve the clarity and rigor of our work. We address each major comment point by point below, indicating where revisions will be made to the manuscript.

read point-by-point responses

  1. Referee: [Numerical Results] Numerical Results section: The headline energy savings (2.74–3.41 % versus 1.50–1.76 % for the MSE baseline) are reported without optimality gaps, iteration limits, duality gaps, or sub-optimality measures from the modified sensitivity-driven integer L-shaped solver. Because both the proposed Bi-MTF model and the sequential baseline are solved with the same method, the absence of these diagnostics prevents confirming that the Pareto-front improvement is attributable to decision-focused training rather than solver artifacts or inconsistent convergence.

    Authors: We agree that solver diagnostics are important for validating the results. In the revised manuscript, we will add tables reporting optimality gaps, iteration counts, and duality information for both the Bi-MTF and MSE baselines under the tested device capacities. This will confirm consistent solver behavior and attribute the energy-savings gains to the decision-focused training. revision: yes

  2. Referee: [Method] Method section (bi-level formulation and solver): The hybrid gradient feedback is claimed to ensure tractability by mixing numerical sensitivity analysis for discrete OLTC/CB variables with analytical dual information for continuous forecast parameters, yet no error analysis, approximation bounds, or small-instance exact-solver benchmark is provided. This is load-bearing for the central claim that the bi-level problem is solved reliably enough to support the reported operational gains.

    Authors: The hybrid feedback is an approximation chosen for computational tractability in the mixed discrete-continuous setting. We will revise the Method section to include a discussion of the approximation error, along with results from small-scale instances solved exactly (via enumeration or commercial solvers) to quantify any sub-optimality and support the reliability of the reported gains. revision: yes

  3. Referee: [Experimental Setup] Experimental Setup: Results are shown only for a single modified IEEE 33-bus system with chosen device capacities and error distributions; no ablation on the sensitivity mechanism, no sensitivity to forecast error magnitude, and no error bars or multiple-run statistics accompany the energy-savings figures. These omissions weaken the robustness of the claim that the approach yields superior performance as fast-acting device capacity increases.

    Authors: We acknowledge the value of additional robustness checks. The revised manuscript will include multiple independent runs with error bars, sensitivity analysis varying forecast error magnitudes, and an ablation study isolating the sensitivity mechanism. While the IEEE 33-bus system is a standard benchmark and the capacity-sweep trends are consistent, we note that broader validation across additional networks would further strengthen generalizability but is beyond the current scope. revision: partial

Circularity Check

0 steps flagged

No significant circularity: decision-focused training and solver remain independent of claimed outputs

full rationale

The paper's central derivation integrates downstream multi-stage VVC optimization directly into upstream forecasting training via a bi-level formulation, which is the explicit goal of decision-focused learning rather than a proxy fit that reduces to its inputs by construction. No equations, self-citations, or uniqueness theorems are invoked that would make the reported energy savings (2.74–3.41 %) or Pareto improvements equivalent to the training data or solver parameters. The modified sensitivity-driven integer L-shaped method is presented as a novel hybrid gradient mechanism for tractability, without load-bearing reliance on prior author work as an external fact. Numerical results on the modified IEEE 33-bus system are offered as empirical evidence, not as a renaming of known patterns or a self-definitional loop. The derivation chain is therefore self-contained against external benchmarks.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

The framework rests on standard power-system modeling assumptions (balanced three-phase flow, linearizable device models) and optimization assumptions (convex relaxations or exactness of the lower-level problem) that are typical in the VVC literature; no new invented entities are introduced.

axioms (2)
  • domain assumption The lower-level multi-stage VVC problem admits a tractable convex or mixed-integer convex reformulation whose dual information can be used for gradient feedback.
    Invoked when the hybrid sensitivity-driven L-shaped method is introduced to solve the bi-level program.
  • domain assumption Forecast errors are independent across time scales and can be represented by a finite set of scenarios whose probabilities are known.
    Required for the multi-timescale forecasting model to propagate errors through the day-ahead, intra-day, and real-time stages.

pith-pipeline@v0.9.0 · 5629 in / 1668 out tokens · 37681 ms · 2026-05-10T17:58:23.383107+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

33 extracted references · 33 canonical work pages

  1. [1]

    Review on implementation and assessment of conservation voltage reduction

    Ma Wang Z, Wang J. Review on implementation and assessment of conservation voltage reduction. IEEE Trans Power Syst 2014;29(3): 1306-1315. https://doi.org/10.1109/TPWRS.2013. 2288518. https:// ieeexplore.ieee.org/document/6670787

    work page doi:10.1109/tpwrs.2013 2014

  2. [2]

    CVR and loss optimization through active voltage management: A tradeoff analysis

    Gharavi H, Ochoa LF, Liu X, Paterson G, Ingham B, McLoone S. CVR and loss optimization through active voltage management: A tradeoff analysis. IEEE Trans Power Deliv 2021;36(6):3466-3476. https://doi.org/10.1109/TPWRD.2020.3043245. https://ieeexplore. ieee.org/document/9286567

    work page doi:10.1109/tpwrd.2020.3043245 2021

  3. [3]

    Standard language protocols for photovoltaics and storage grid integration

    Seal B. Standard language protocols for photovoltaics and storage grid integration. Palo Alto, CA, USA: Electric Power Research Institute

    work page

  4. [4]

    Available at: https://www.tdworld

    Technical Report 1020906. Available at: https://www.tdworld. com/grid-innovations/generation-and-renewables/article/20963273/ standard-language-protocols-for-photovoltaics-and-storage-grid- integration

    work page arXiv

  5. [5]

    A novel stochastic V olt/V AR/pressure optimization-based conservation voltage reduction technique in integrated electricity and natural gas systems with PV smart inverters

    Karamdel S, Liang X, Faried SO, Cheruvallath J. A novel stochastic V olt/V AR/pressure optimization-based conservation voltage reduction technique in integrated electricity and natural gas systems with PV smart inverters. IEEE Trans Ind Appl 2025;61(2):2271-2286. https://doi.org/10.1109/TIA.2025.3532570. https://ieeexplore.ieee. org/document/10848276

    work page doi:10.1109/tia.2025.3532570 2025

  6. [6]

    A multi-mode data-driven V olt/Var control strategy with conservation voltage reduction in active distribution networks

    Sun X, Qiu J, Tao Y , Ma Y , Zhao J. A multi-mode data-driven V olt/Var control strategy with conservation voltage reduction in active distribution networks. IEEE Trans Sustain Energy 2022;13(2):1073-

    work page 2022

  7. [7]

    https://ieeexplore

    https://doi.org/10.1109/TSTE.2022.3149267. https://ieeexplore. ieee.org/document/9707859

    work page doi:10.1109/tste.2022.3149267 2022

  8. [8]

    Two-stage V olt/Var control in active distribution networks with multi-agent deep reinforcement learning method

    Sun X, Qiu J. Two-stage V olt/Var control in active distribution networks with multi-agent deep reinforcement learning method. IEEE Trans Smart Grid 2021;12(4):2903-2912. https://doi.org/10.1109/ TSG.2021.3052998. https://ieeexplore.ieee. org/document/9328796

    work page arXiv 2021

  9. [9]

    Inverter-less hybrid voltage/V ar control for distribution circuits with photovoltaic generators

    Wang Z, Chen H, Wang J, Begovic M. Inverter-less hybrid voltage/V ar control for distribution circuits with photovoltaic generators. IEEE Trans Smart Grid 2014;5(6):2718-2728. https://doi.org/10.1109/TSG.2014.2324569. https://ieeexplore.ieee. org/document/6895183

    work page doi:10.1109/tsg.2014.2324569 2014

  10. [10]

    Multi-timescale coordinated voltage/var control of high renewable-penetrated distribution systems

    Xu Y , Dong ZY , Zhang R, Hill DJ. Multi-timescale coordinated voltage/var control of high renewable-penetrated distribution systems. IEEE Trans Power Syst 2017;32(6):4398-4408. https://doi.org/10. 1109/TPWRS.2017.2669343. https://ieeexplore. ieee.org/document/ 7857812

    work page arXiv 2017

  11. [12]

    Multistage multiobjective V olt/V AR control for smart grid-enabled CVR with solar PV penetration

    Singh S, Pamshetti VB, Thakur AK, Singh SP. Multistage multiobjective V olt/V AR control for smart grid-enabled CVR with solar PV penetration. IEEE Syst J 2021;15(2):2767-2778. https://doi. org/10.1109/JSYST.2020.2992985. https://ieeexplore.ieee.org/ document/9099637

    work page doi:10.1109/jsyst.2020.2992985 2021

  12. [13]

    Hierarchically-coordinated voltage/var control of distribution networks using PV inverters

    Zhang C, Xu Y . Hierarchically-coordinated voltage/var control of distribution networks using PV inverters. IEEE Trans Smart Grid 2020;11(4):2942-2953. https://doi.org/10.1109/TSG.2020.2968394. https://ieeexplore.ieee.org/document/8964320

    work page doi:10.1109/tsg.2020.2968394 2020

  13. [14]

    Stochastic distribution system operation considering voltage regulation risks in the presence of PV generation

    Agalgaonkar YP, Pal BC, Jabr RA. Stochastic distribution system operation considering voltage regulation risks in the presence of PV generation. IEEE Trans Sustain Energy 2015;6(4):1315-1324. https://doi.org/10.1109/TSTE.2015.2433794. https://ieeexplore.ieee. org/document/7122359

    work page doi:10.1109/tste.2015.2433794 2015

  14. [15]

    Dynamic Graph-Based Anomaly Detection in the Electrical Grid

    Guo Y , Zhang Q, Wang Z. Cooperative peak shaving and voltage regulation in unbalanced distribution feeders. IEEE Trans Power Syst 2021;36(6):5235-5244. https://doi.org/10.1109/TPWRS.2021. 3069781. https://ieeexplore.ieee.org/document/9390352

    work page doi:10.1109/tpwrs.2021 2021

  15. [16]

    A multi-objective minimax regret robust V Ar planning model

    López J, Pozo D, Contreras J, Mantovani JRS. A multi-objective minimax regret robust V Ar planning model. IEEE Trans Power Syst 2017;32(3):1761-1771. https://doi.org/10.1109/TPWRS.2016. 2613544. https://ieeexplore.ieee.org/document/7577885

    work page doi:10.1109/tpwrs.2016 2017

  16. [17]

    Distributed adaptive robust voltage/var control with network partition in active distribution networks

    Li P , Zhang C, Wu Z, Xu Y , Hu M, Dong Z. Distributed adaptive robust voltage/var control with network partition in active distribution networks. IEEE Trans Smart Grid 2020;11(3):2245-2256. https://doi. org/10.1109/TSG.2019.2950120. https://ieeexplore. ieee.org/ document/8887238

    work page doi:10.1109/tsg.2019.2950120 2020

  17. [18]

    Tractable data enriched distributionally robust chance-constrained conservation voltage reduction

    Zhang Q, Bu F, Guo Y , Wang Z. Tractable data enriched distributionally robust chance-constrained conservation voltage reduction. IEEE Trans Power Syst 2024;39(1):821-835. https://doi.org/10.1109/TPWRS.2023.3244895. https://ieeexplore. ieee.org/document/10044219

    work page doi:10.1109/tpwrs.2023.3244895 2024

  18. [19]

    Energy storage planning in active distribution grids: A chance-constrained optimization with non- parametric probability functions

    Hejazi H, Mohsenian-Rad H. Energy storage planning in active distribution grids: A chance-constrained optimization with non- parametric probability functions. IEEE Trans Smart Grid 2018;9(3):1972-1985. https://doi.org/10.1109/TSG.2016.2604286. https://ieeexplore.ieee.org/document/7556322. Q. Du, R. Li, Y. Chen, H. Zhang Applied energy 16

    work page doi:10.1109/tsg.2016.2604286 2018

  19. [20]

    MPC-based coordinated voltage control in active distribution networks incorporating CVR and DR

    Dutta A, Ganguly S, Kumar C. MPC-based coordinated voltage control in active distribution networks incorporating CVR and DR. IEEE Trans Ind Appl 2022;58(4):4309-4318. https://doi.org/10.1109/ TIA.2022.3163108. https://ieeexplore.ieee.org/document/9744565

    work page arXiv 2022

  20. [21]

    V oltage regulation and loss minimization of active distribution networks with uncertainties using chance- constrained model predictive control

    Dar MR, Ganguly S. V oltage regulation and loss minimization of active distribution networks with uncertainties using chance- constrained model predictive control. IEEE Trans Power Syst 2025;40(3):2737-2749. https://doi.org/10.1109/TPWRS.2024. 3504532. https://ieeexplore.ieee.org/document/10762839

    work page doi:10.1109/tpwrs.2024 2025

  21. [22]

    Distributed coordinated voltage control for distribution networks with DG and OLTC based on MPC and gradient projection

    Jiao W, Chen J, Wu Q, Li C, Zhou B, Huang S. Distributed coordinated voltage control for distribution networks with DG and OLTC based on MPC and gradient projection. IEEE Trans Power Syst 2022;37(1):680-690. https://doi.org/10.1109/TPWRS.2021.3095523. https://ieeexplore.ieee.org/document/9478225

    work page doi:10.1109/tpwrs.2021.3095523 2022

  22. [23]

    V oltage regulation enhanced hierarchical coordinated V olt/Var and V olt/Watt control for active distribution networks with soft open points

    Hou Q, Dai N, Huang Y . V oltage regulation enhanced hierarchical coordinated V olt/Var and V olt/Watt control for active distribution networks with soft open points. IEEE Trans Sustain Energy 2024;15(3):2021-2037. https://doi.org/10.1109/TSTE.2024.3394049. https://ieeexplore.ieee.org/document/10508977

    work page doi:10.1109/tste.2024.3394049 2024

  23. [24]

    Multistage coordination V olt/V AR control with CVR in active distribution network in presence of inverter-based DG units and soft open points

    Pamshetti VB, Singh S, Thakur AK, Singh SP. Multistage coordination V olt/V AR control with CVR in active distribution network in presence of inverter-based DG units and soft open points. IEEE Trans Ind Appl 2021;57(3):2035-2047. https://doi.org/10.1109/ TIA.2021.3063667. https://ieeexplore.ieee.org/document/9369032

    work page arXiv 2021

  24. [25]

    Integrated CVR and demand response framework for advanced distribution management systems

    Hossan MS, Chowdhury B. Integrated CVR and demand response framework for advanced distribution management systems. IEEE Trans Sustain Energy 2020;11(1):534-544. https://doi.org/10.1109/ TSTE.2019.2897333. https://ieeexplore.ieee.org/document/8633932

    work page arXiv 2020

  25. [26]

    The Concrete distribution: a continuous relaxation of discrete random variables

    Maddison CJ, Mnih A, Teh YW. The Concrete distribution: a continuous relaxation of discrete random variables. In: Proc Int Conf Learn Representations (ICLR); 2017. Available at: https://openreview. net/forum?id=S1jE5L5gl

    work page 2017

  26. [27]

    A bilevel framework for decision- making under uncertainty with contextual information

    Muñoz MA, Pineda S, Morales JM. A bilevel framework for decision- making under uncertainty with contextual information. Omega 2022; 108:102575. https://doi.org/10.1016/j.omega.2021.102575. https:// www.sciencedirect.com/science/article/pii/S0305048321001845

    work page doi:10.1016/j.omega.2021.102575 2022

  27. [28]

    Decision-focused learning for power system decision-making under uncertainty

    Zhang H, Li R, Du Q, Tao J, et al. Decision-focused learning for power system decision-making under uncertainty. IEEE Trans Power Syst 2026;41(1):307-323. https://doi.org/10.1109/TPWRS.2025. 3597806. https://ieeexplore.ieee.org/abstract/document/11122623

    work page doi:10.1109/tpwrs.2025 2026

  28. [29]

    Distributionally robust optimization of photovoltaic power with lifted linear decision rule for distribution system voltage regulation

    Xu X, Gao Y , Wang H, Yan Z, Shahidehpour M, Tan Z. Distributionally robust optimization of photovoltaic power with lifted linear decision rule for distribution system voltage regulation. IEEE Trans Sustain Energy 2024;15(2):758-772. https://doi.org/10.1109/ TSTE.2023.3303417. https://ieeexplore.ieee.org/document/10214028

    work page arXiv 2024

  29. [30]

    Finite difference weights, spectral differentiation, and superconvergence

    Sadiq B, Viswanath D. Finite difference weights, spectral differentiation, and superconvergence. Math Comp 2014;83(289): 2403-2427. https://doi.org/10.1090/S0025-5718-2014-02798-1. https://www.ams.org/journals/mcom/2014-83-289/S0025-5718-2014- 02798-1

    work page doi:10.1090/s0025-5718-2014-02798-1 2014

  30. [31]

    Zero duality gap in optimal power flow problem

    Lavaei J, Low SH. Zero duality gap in optimal power flow problem. IEEE Trans Power Syst 2012;27(1):92-107. https://doi.org/10.1109/ TPWRS.2011.2160974. https://ieeexplore.ieee.org/document/ 5971792

    work page arXiv 2012

  31. [32]

    Network reconfiguration in distribution systems for loss reduction and load balancing

    Baran ME, Wu FF. Network reconfiguration in distribution systems for loss reduction and load balancing. IEEE Trans Power Deliv 1989;4(2):1401-1407. https://doi.org/10.1109/61.25627

    work page doi:10.1109/61.25627 1989

  32. [33]

    The National Solar Radiation Data Base (NSRDB)

    Sengupta M, Xie Y , Lopez A, Habte A, Maclaurin G, Shelby J. The National Solar Radiation Data Base (NSRDB). Renew Sustain Energy Rev 2018;89:51-60. https://doi.org/10.1016/j.rser.2018.03.003. https://www.sciencedirect.com/science/article/pii/S136403211830087 X. https://ieeexplore.ieee.org/document/25627

    work page doi:10.1016/j.rser.2018.03.003 2018

  33. [34]

    Bi-level V olt-V AR optimization to coordinate smart inverters with voltage control devices

    Jha RR, Dubey A, Liu C-C, Schneider KP. Bi-level V olt-V AR optimization to coordinate smart inverters with voltage control devices. IEEE Trans Power Syst 2019;34(3):1801-1813. https://doi. org/10.1109/TPWRS.2018.2890613. https://ieeexplore.ieee.org/ document/8598813

    work page doi:10.1109/tpwrs.2018.2890613 2019