Pith. sign in

REVIEW 3 major objections 5 minor 32 references

Neural Network Certification Informed Power System Transient Stability Preventive Control with Renewable Energy

T0 review · 3 major / 5 minor · reviewed 2026-08-12 · deepseek-v4-flash

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

Pith's one-line read Embedding a deep-belief-network surrogate in optimal power flow and certifying it with α,β-CROWN yields safety-verified preventive control with a definitive verdict in 99.6% of tested cases at negligible extra cost.

desk verdict The integration is real and the numbers are plausible, but the paper's central claim that the safety margin λ is certified does not survive close reading: the verifier checks the binary classifier, not the margin. read the letter →

arxiv 2411.08329 v1 pith:JUDSSO4L submitted 2024-11-13 eess.SY cs.SY

Tong Su , Junbo Zhao This is my paper · ORCID
classification eess.SYcs.SY
keywords transientstabilitypreventivecontrolneuralnetworkcertificationdeepbeliefalpha-beta-CROWNTSC-OPFrenewableenergyuncertaintyrobustness
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

Preventive control for transient stability normally relies on time-domain simulation, which is too slow to certify against every possible measurement error, renewable fluctuation, or load change. The authors replace the simulation with a deep belief network (DBN) embedded inside the optimal-power-flow solver, then certify the proposed dispatch with the α,β-CROWN verifier, which proves that the classifier's stable label holds for every operating point in an uncertainty box. If a candidate strategy fails certification, the transient stability margin $\lambda$ is raised by bisection until the strategy passes. On a modified western South Carolina 500-bus system, the method obtains a definitive safety verdict in 99.6% of tested cases and produces a verified preventive strategy with only a $22 increase in operating cost. The point is that certification, not sampling, is what closes the safety gap for surrogate-based transient stability control.

What carries the argument

The load-bearing object is the α,β-CROWN verifier applied to a ReLU-activated DBN classifier: a bound-propagation and branch-and-bound method that computes a certified lower bound on the network output over an input uncertainty set $C = \{x : \|x-\hat{x}\|_\infty \leq \epsilon\}$. It carries the argument by converting the question "could any perturbation inside $C$ flip the stable label?" into a provable optimization bound. The DBN estimation network supplies the TSI constraint and its gradients inside the primal-dual interior-point method, while the DBN classification network is what gets certified. The bisection on $\lambda$ ties the two together: it raises the transient stability margin until the classifier's certified lower bound is positive, making safety a constraint of the optimization rather than a post-hoc sample check.

What would settle it

Take a strategy that the method certifies as safe and densely sample or adversarially search the corresponding uncertainty box; run full time-domain simulation at each point and compute the true TSI. Finding even one point inside the box whose true rotor-angle swing gives TSI below the stability threshold would show that the certification holds for the surrogate, not for the physical system.

Watch

Extended reading notes

Core claim

The paper's central discovery is that neural-network certification can be made load-bearing for transient stability preventive control. A DBN with ReLU activations outputs TSI; a second DBN classifier is verified over the $\ell_\infty$ ball $C$ around the forecast/control inputs using three cascaded checks: PGD attack, α-CROWN incomplete bound propagation, and β-CROWN complete branch-and-bound. If the lower bound of the classifier output on $C$ stays positive, every point in $C$ is classified stable. When certification fails, bisection on $\lambda$ raises the required TSI margin until TSC-OPF finds a strategy that certifies. In the numerical study, 63.4% of random strategies were found unsafe by PGD, 29.7% certified safe by α-CROWN, 6.5% certified safe by β-CROWN, and 0.4% remained unknown; the verified preventive strategy shifted 25.53 MW between two synchronous generators and increased cost from $19,035.6 to $19,057.8.

Load-bearing premise

The argument assumes that certifying the DBN classifier over the uncertainty box is the same as certifying the physical power system, even though the DBN is only an approximation (99.15% classification accuracy and mean absolute TSI error of 1.86) with no formal error bound linking its output to the true TSI.

Editorial extensions

If this is right

  • Operators can deploy a DBN-based TSC-OPF online: DBN inference is under 0.01 s and the full PGD/α-CROWN/β-CROWN verification averages 0.196 s per strategy.
  • A strategy that passes certification is guaranteed, within the surrogate model, to have no misclassified point in the uncertainty box, so safety no longer depends on how many Monte Carlo samples happen to be drawn.
  • The bisection on the transient stability margin is a direct economic-security dial: raising it until certification passes increased cost by only $22 on the 500-bus case while excluding adversarial operating points.
  • The verification cascade raises the definitive-verdict rate from 63.4% with PGD alone to 99.6%, leaving only 0.4% of strategies unknown.
  • Because the PGD, α-CROWN, and β-CROWN checks are decoupled, certification can be parallelized across servers, supporting the paper's scalability claim.

Reading between the lines

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

  • Editorial inference: the same certification loop transfers to other surrogate-constrained OPF problems, such as voltage stability, frequency constraints, or small-signal stability, wherever the surrogate is a ReLU network and the uncertainty is bounded.
  • Editorial inference: the 0.4% unknown rate is not a hard ceiling; β-CROWN's branching can in principle be extended until all unstable ReLUs are split, at the price of verification time, so the method offers a tunable completeness-versus-compute trade.
  • Editorial inference: because the certified lower bound is a differentiable function of the network parameters and the input box, one could optimize the safety margin directly inside the OPF rather than by outer bisection, potentially reducing the number of TSC-OPF re-solves.
  • Editorial inference: using correlated, forecast-error-shaped uncertainty regions instead of a box could shrink the certified set and lower the cost of verified dispatch, since the current $\ell_\infty$ ball treats each input's deviation independently.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 5 minor

Summary. The paper proposes a preventive control method for power system transient stability that replaces time-domain simulations with a deep belief network (DBN) surrogate inside a TSC-OPF solved by a primal-dual interior-point method, and then uses the neural-network verifier α,β-CROWN to certify robustness of the resulting dispatch against input uncertainties such as measurement errors and RES/load forecast errors. The authors claim that the certification results enable adjustment of a transient stability safety margin λ, balancing security and economics, and demonstrate the approach on a modified South Carolina 500-bus system, reporting a small cost increase ($22.2) for a 'safety-verified' preventive control strategy.

Significance. If the central claim were fully supported, the paper would make a meaningful contribution by coupling formal neural-network robustness certification with TSC-OPF on a large-scale test system. The paper has clear strengths: it uses a complete verifier (β-CROWN with branch-and-bound) in addition to incomplete bounds, provides computational timing numbers, and anchors the demonstration with Monte Carlo checks on the physical model. However, the current gap between the property actually certified (binary DBN-C classification over an input box) and the property claimed in the TSC-OPF (DBN-E TSI margin λ) is a load-bearing correctness issue. The work is therefore promising but needs substantial revision before the safety-margin claims can be accepted.

major comments (3)
  1. [Section III-F, Algorithms 2 and 3, Eq. (10)] The verification performed by Algorithm 2 certifies the DBN-C binary classifier over the input set C (i.e., f(x) > 0), whereas the TSC-OPF constraint (10) uses a different model, DBN-E, with the safety margin λ. The paper never establishes that a certified stable classification of DBN-C over C implies DBN-E(x) > λ for all x in C. Consequently, the λ values reported in Table II (e.g., λ = 63.28125) are not the certified margin; they are only the surrogate constraint margin in the OPF. This directly contradicts the abstract's claim that 'the yielded certification results allow us to further adjust the transient stability safety margin.' To support this claim, the verifier must be applied to DBN-E with the threshold λ (e.g., verify f_E(x) > λ for all x in C), or the paper must clearly restate its contribution as certifying only binary transient stability classification.
  2. [Section IV-A, Table I] The statement that 'α, β-CROWN can obtain the robustness verification result for the preventive control strategy ex in 99.6% of cases, and this result is complete' is misleading. Only the 6.5% of cases labeled 'safe-complete' are verified by the complete β-CROWN verifier; the 29.7% 'safe-incomplete' cases rely on α-CROWN, which is an incomplete verifier, and the 63.4% 'unsafe-PGD' cases are actually adversarial examples found by PGD, not complete verification results. The 99.6% figure aggregates these incomparable categories. The paper should either separate the statistics by verification type or rephrase the completeness claim so that it does not suggest that 99.6% of cases were completely verified.
  3. [Section IV-B, final MCS (Fig. 13)] The paper states that 'Undoubtedly, the verified ex can ensure that the system maintains transient stability.' This is stronger than what the evidence supports. The formal certification applies to the DBN-C classifier, not to the physical TSI; the reported DBN classification accuracy (99.15%) and MAE of TSI (1.86) mean that the certified property does not formally bound the true physical TSI. The Monte Carlo sampling of points in C is finite and does not close this gap. The assertion should be weakened to 'the surrogate classifier is certified over C, and sampled physical trajectories were stable' unless a formal relationship between DBN-C certification and physical stability is provided.
minor comments (5)
  1. [Table II] The title of Table II is a copy of Table I's title ('Statistical Probabilities of NN Robustness Verification Results for Random Scenarios'); it should describe the bisection iterations of the TSC-OPF, e.g., 'Iterative TSC-OPF results with robustness verification.'
  2. [Section II-C, Eq. (4)] Equation (4) is typeset incorrectly: the line flow limit and voltage magnitude constraints appear to be concatenated without proper separators, making the constraint set ambiguous.
  3. [Algorithm 2 and Section III-F] Algorithm 2 returns 'unknow' instead of 'unknown', and the text in Section III-F says 'Return directed' instead of 'Return directly.'
  4. [Section IV-A] The DBN architecture, hyperparameters, and the train/validation split for DBN-C and DBN-E are not given in this paper; the reader is referred to [31], but the exact settings used for the 500-bus case study should be reported for reproducibility.
  5. [Section IV-B] The perturbation ranges used for the preventive control case (20% for IBRs, 8% for SGs, 10% for loads) are not justified; the paper should explain how these values are derived from realistic measurement and forecast error characteristics or state that they are chosen for demonstration.

Circularity Check

0 steps flagged · score 0.0 of 10

No circular derivation: the reported certified-margin gap is a soundness issue, not a self-referential reduction.

full rationale

Walking the derivation chain, no step reduces by construction to its own input. The TSC-OPF constraint (10) uses the DBN-E estimate gTSI > lambda, while the verifier in Algorithm 2 operates on DBN-C, the binary classifier. These are explicitly different models in Algorithm 1 ('DBN-C outputs the 0/1 classification labels for TSI, and DBN-E outputs the TSI'). The paper's Section IV-A even concedes that safe-complete verification stops 'until the lower bound is greater than 0. Therefore, its final margin is a small positive value,' yet Table II reports the final lambda as 63.28125 with TSI equal to 63.28125. That means the certified property is binary stability (TSI>0), not the lambda=63.28 margin claimed in the preventive-control narrative. This is a real and load-bearing overclaim about what is certified, but it is a correctness/soundness gap, not circularity: lambda is chosen by bisection over OPF feasibility and DBN-C verification, and beta-CROWN's lower bound is not defined as lambda, nor is lambda fitted to that bound. No equation in the paper is equivalent to a fitted value by construction, and no uniqueness or ansatz result is imported from the authors' prior work. The self-citations [27] and [31] supply DBN architecture and training details, but the certification mechanism is the external alpha,beta-CROWN literature ([18]-[20]) and is separately anchored by time-domain-simulation Monte Carlo checks. Accordingly, the honest circularity finding is 0: the derivation is not circular, even though the claimed safety margin is not actually certified.

Assumptions & free parameters 3 free parameters · 6 assumptions · 0 invented entities

The central claim rests on the surrogate being a faithful stand-in for the physical transient stability check, on the box uncertainty model, and on the representativeness of the test system. No new physical entities are introduced; the DBN and verifier are computational tools. The free parameters are human-chosen bounds and hyperparameters that shape the numerical demonstration.

free parameters (3)
  • DBN architecture and hyperparameters = Not specified in this paper (see [31])
    Number of layers, neurons per layer, and learning hyperparameters are chosen by the authors; the paper only says 'set the number of layers, the number of neurons per layer, the activation function ReLU, and other hyperparameters' in Algorithm 1.
  • Perturbation bounds for the case study = IBR 20%, SG 8%, load 10%
    The maximum perturbation range used to construct C is chosen by the authors in Section IV-B; it is an input to the verification and determines the verified result.
  • Bisection parameters for λ = λ_left=0, λ_right=90, Δλ=2, ζ=1
    Algorithm 3 sets these values; they control the search for the safety margin.
assumptions (6)
  • domain assumption The DBN surrogate accurately represents the true transient stability index over the relevant input space.
    The entire certification and optimization rely on the DBN's TSI estimate replacing the time-domain simulation. The paper reports 99.15% classification accuracy and mean absolute error 1.86, but does not provide a formal bound on the approximation error.
  • domain assumption The ℓ∞ ball C with element-wise bounds captures all relevant uncertainties (measurement errors, RES fluctuations, load fluctuations).
    Section III-F and Algorithm 2 construct C as an ℓ∞ ball around the forecast; this models uncertainty but may not cover correlated or non-box-shaped disturbances.
  • domain assumption The synthetic South Carolina 500-bus system and the selected fault contingency are representative of real operating conditions.
    The test system is a modified synthetic grid; the specific contingency is not described. Results may not generalize to other systems or contingencies.
  • domain assumption The primal-dual interior-point method converges to a feasible optimum of the nonconvex TSC-OPF problem.
    The paper states the DBN-embedded problem is nonlinear and nonconvex and solves it with PDIPM, but does not guarantee global optimality or even convergence from arbitrary starts.
  • standard math The α,β-CROWN verifier soundly certifies the DBN classifier.
    The soundness of α,β-CROWN is established in prior literature [19][20]; the paper relies on this without reproving it.
  • domain assumption A TSI value greater than λ correctly indicates transient stability in the physical system.
    The TSI definition from Eq. (9) is a standard domain criterion; it is used in the DBN surrogate and in the Monte Carlo validation.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Neural Network Certification Informed Power System Transient Stability Preventive Control with Renewable Energy." pith.science (2026). https://pith.science/paper/JUDSSO4L

@misc{pith2026241108329,
  author       = {Pith},
  title        = {Pith review of: Neural Network Certification Informed Power System Transient Stability Preventive Control with Renewable Energy},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/JUDSSO4L}},
  note         = {Machine review of arXiv:2411.08329}
}
abstract

Existing machine learning-based surrogate modeling methods for transient stability constrained-optimal power flow (TSC-OPF) lack certifications in the presence of unseen disturbances or uncertainties. This may lead to divergence of TSC-OPF or insecure control strategies. This paper proposes a neural network certification-informed power system transient stability preventive control method considering the impacts of various uncertainty resources, such as errors from measurements, fluctuations in renewable energy sources (RESs) and loads, etc. A deep belief network (DBN) is trained to estimate the transient stability, replacing the time-consuming time-domain simulation-based calculations. Then, DBN is embedded into the iterations of the primal-dual interior-point method to solve TSC-OPF. To guarantee the robustness of the solutions, the neural network verifier $\alpha, \beta$-CROWN to deal with uncertainties from RESs and loads is proposed. The yielded certification results allow us to further adjust the transient stability safety margin under the iterated TSC-OPF solution process, balancing system security and economics. Numerical results on a modified western South Carolina 500-bus system demonstrate that the proposed method can efficiently and quickly obtain the safety-verified preventive control strategy through RES curtailment and generator dispatch with only a slight increase in cost.

Figures

Figures reproduced from arXiv: 2411.08329 by the authors.

Figure 1
Figure 1. Incomplete verification. (a): f ≥ 0, xe is verified. (b): f < 0, the verification result is unknown. A commonly used incomplete verifier is the linear pro￾gramming (LP) verifier, which relaxes non-convex ReLU constraints with linear constraints and transforms (12) into a LP problem. For ReLU(z (i) j ) := max(0, z (i) j ) and its intermediate layer bounds l (i) j ≤ z (i) j ≤ u (i) j , each ReLU can be relaxed by the … view at source ↗
Figure 2
Figure 2. Relaxations of a ReLU. (a)(b) No relaxation when [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. BaB-based NN splitting and verification. [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
Figures from the paper (9 more)
Figure 4
Figure 4. Figure 4: Proposed α, β-CROWN-based transient stability preventive control framework. is safe, reduce λ (with a minimum of 0); otherwise, increase λ until the verification result is safe, resulting in the final safety-verified xe. When TSC-OPF does not converge and the NN verifi…
Figure 5
Figure 5. Figure 5: One-line diagram of the South Carolina 500-bus system. [PITH_FULL_IMAGE:figures/full_fig_p006_5.png]
Figure 7
Figure 7. Figure 7: Calculation of maximum possible perturbation and safety margin. [PITH_FULL_IMAGE:figures/full_fig_p007_7.png]
Figure 6
Figure 6. Figure 6: Comparison of NN verification results with different perturbations. [PITH_FULL_IMAGE:figures/full_fig_p007_6.png]
Figure 8
Figure 8. Figure 8: Comparison of active power generation of IBRs and SGs before and [PITH_FULL_IMAGE:figures/full_fig_p007_8.png]
Figure 9
Figure 9. Figure 9: Rotor angle trajectories before and after preventive control without [PITH_FULL_IMAGE:figures/full_fig_p008_9.png]
Figure 11
Figure 11. Figure 11: Comparison of active power generation of IBRs and SGs before and [PITH_FULL_IMAGE:figures/full_fig_p008_11.png]
Figure 12
Figure 12. Figure 12: Rotor angle trajectories before and after preventive control with [PITH_FULL_IMAGE:figures/full_fig_p008_12.png]
Figure 13
Figure 13. Figure 13: Rotor angle trajectories of points in C with robustness verification. V. CONCLUSION This paper proposes an NN robustness verification-based power system transient stability preventive control method considering errors from various measurement devices, fluc￾tuations in…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

32 extracted references · 28 canonical work pages

  1. [1]

    A review of deep learning for renewable energy forecasting,

    H. Wang, Z. Lei, X. Zhang, B. Zhou, and J. Peng, “A review of deep learning for renewable energy forecasting,” Energy Convers. Manage. , vol. 198, 2019, Art. no. 111799

    work page 2019

  2. [2]

    A review of deep learning methods applied on load forecasting,

    A. Almalaq and G. Edwards, “A review of deep learning methods applied on load forecasting,” in Proc. IEEE Int. Conf. Mach. Learn. Appl., 2017, pp. 511–516

    work page 2017

  3. [3]

    A review of machine learning approaches to power system security and stability,

    O. A. Alimi, K. Ouahada, and A. M. Abu-Mahfouz, “A review of machine learning approaches to power system security and stability,” IEEE Access, vol. 8, pp. 113 512–113 531, 2020

    work page 2020

  4. [4]

    Intriguing properties of neural networks,

    C. Szegedy, W. Zaremba, I. Sutskever, J. Bruna, D. Erhan, I. Goodfellow, and R. Fergus, “Intriguing properties of neural networks,” in Proc. Int. Conf. Learn. Represent. , 2013

    work page 2013

  5. [5]

    Wild patterns: Ten years after the rise of adversarial machine learning,

    B. Biggio and F. Roli, “Wild patterns: Ten years after the rise of adversarial machine learning,” in Proc. ACM SIGSAC Conf. Comput. Commun. Secur., 2018, pp. 2154–2156

    work page 2018

  6. [6]

    The robustness of deep networks: A geometrical perspective,

    A. Fawzi, S.-M. Moosavi-Dezfooli, and P. Frossard, “The robustness of deep networks: A geometrical perspective,” IEEE Signal Process. Mag., vol. 34, no. 6, pp. 50–62, 2017

    work page 2017

  7. [7]

    Standard IEEE C37.118.1-2011, Dec

    IEEE Standard for Synchrophasor Measurements for Power Systems . Standard IEEE C37.118.1-2011, Dec. 2011

    work page 2011

  8. [8]

    Impact of measurement errors on synchrophasor applications,

    J. Zhao, J. Tan, L. Wu, L. Zhan, W. Yao, Y . Liu, J. R. Gracia, and P. D. Ewing, “Impact of measurement errors on synchrophasor applications,” in Proc. IEEE Power Energy Soc. Gen. Meeting , 2017, pp. 1–5

    work page 2017

Show all 32 references
  1. [9]

    A review of false data injection attacks against modern power systems,

    G. Liang, J. Zhao, F. Luo, S. R. Weller, and Z. Y . Dong, “A review of false data injection attacks against modern power systems,” IEEE Trans. Smart Grid, vol. 8, no. 4, pp. 1630–1638, Jul. 2017

    2017

  2. [10]

    Detection of faults and attacks including false data injection attack in smart grid using kalman filter,

    K. Manandhar, X. Cao, F. Hu, and Y . Liu, “Detection of faults and attacks including false data injection attack in smart grid using kalman filter,” IEEE Trans. Control. Netw. Syst. , vol. 1, no. 4, pp. 370–379, 2014

    2014

  3. [11]

    Adversarial examples: Attacks and defenses for deep learning,

    X. Yuan, P. He, Q. Zhu, and X. Li, “Adversarial examples: Attacks and defenses for deep learning,” IEEE Trans. Neural Netw. Learn. Syst., vol. 30, no. 9, pp. 2805–2824, Sep. 2019

    2019

  4. [12]

    Robustness verification for machine-learning-based power system dynamic security assessment models under adversarial examples,

    C. Ren and Y . Xu, “Robustness verification for machine-learning-based power system dynamic security assessment models under adversarial examples,” IEEE Trans. Control Netw. Syst. , vol. 9, no. 4, pp. 1645– 1654, Dec. 2022

    2022

  5. [13]

    Joint adversarial example and false data injection attacks for state estimation in power systems,

    J. Tian, B. Wang, Z. Wang, K. Cao, J. Li, and M. Ozay, “Joint adversarial example and false data injection attacks for state estimation in power systems,” IEEE Trans. Cybern., vol. 52, no. 12, pp. 13 699–13 713, Dec. 2022

    2022

  6. [14]

    Adversarial attacks and defense for cnn based power quality recognition in smart grid,

    J. Tian, B. Wang, J. Li, and Z. Wang, “Adversarial attacks and defense for cnn based power quality recognition in smart grid,” IEEE Trans. Netw. Sci. Eng., vol. 9, no. 2, pp. 807–819, Mar./Apr. 2022

    2022

  7. [15]

    A universal defense strategy for data-driven power system stability assessment models under adversarial examples,

    C. Ren and Y . Xu, “A universal defense strategy for data-driven power system stability assessment models under adversarial examples,” IEEE Internet Things J. , May 2023

    2023

  8. [16]

    Vulnerability analysis, robustness verification, and mitigation strategy for machine learning-based power system stability assessment model under adversar- ial examples,

    C. Ren, X. Du, Y . Xu, Q. Song, Y . Liu, and R. Tan, “Vulnerability analysis, robustness verification, and mitigation strategy for machine learning-based power system stability assessment model under adversar- ial examples,” IEEE Trans. Smart Grid , vol. 13, no. 2, pp. 1622–16...

    2022

  9. [17]

    Verification of neural network behaviour: Formal guarantees for power system applications,

    A. Venzke and S. Chatzivasileiadis, “Verification of neural network behaviour: Formal guarantees for power system applications,” IEEE Trans. Smart Grid , vol. 12, no. 1, pp. 383–397, Jan. 2021

    2021

  10. [18]

    Ef- ficient neural network robustness certification with general activation functions,

    H. Zhang, T.-W. Weng, P.-Y . Chen, C.-J. Hsieh, and L. Daniel, “Ef- ficient neural network robustness certification with general activation functions,” in Proc. Adv. Neural Inf. Process. Syst. , vol. 31, 2018, p. 4939–4948

    2018

  11. [19]

    Fast and complete: Enabling complete neural network verification with rapid and massively parallel incomplete verifiers,

    K. Xu, H. Zhang, S. Wang, Y . Wang, S. Jana, X. Lin, and C.-J. Hsieh, “Fast and complete: Enabling complete neural network verification with rapid and massively parallel incomplete verifiers,” in Proc. Int. Conf. Learn. Represent., 2021, p. 1–15

    2021

  12. [20]

    Beta-CROWN: Efficient bound propagation with per-neuron split constraints for neural network robustness verification,

    S. Wang, H. Zhang, K. Xu, X. Lin, S. Jana, C.-J. Hsieh, and J. Z. Kolter, “Beta-CROWN: Efficient bound propagation with per-neuron split constraints for neural network robustness verification,” in Proc. Adv. Neural Inf. Process. Syst. , vol. 34, 2021, pp. 29 909–29 921

    2021

  13. [21]

    Neural network verification with branch-and-bound for general nonlinearities,

    Z. Shi, Q. Jin, Z. Kolter, S. Jana, C.-J. Hsieh, and H. Zhang, “Neural network verification with branch-and-bound for general nonlinearities,” arXiv preprint arXiv:2405.21063 , 2024

    2024 arXiv

  14. [22]

    Automatic perturbation analysis for scalable certified robustness and beyond,

    K. Xu, Z. Shi, H. Zhang, Y . Wang, K.-W. Chang, M. Huang, B. Kailkhura, X. Lin, and C.-J. Hsieh, “Automatic perturbation analysis for scalable certified robustness and beyond,” in Proc. Adv. Neural Inf. Process. Syst., vol. 33, 2020, pp. 1129–1141

    2020

  15. [23]

    Robustly coordinated generation dis- patch and load shedding for power systems against transient instability under uncertain wind power,

    H. Yuan, Y . Xu, and C. Zhang, “Robustly coordinated generation dis- patch and load shedding for power systems against transient instability under uncertain wind power,” IEEE Trans. Power Syst. , vol. 37, no. 2, pp. 1032–1043, Mar. 2022

    2022

  16. [24]

    Real-time prediction and control of transient stability using transient energy function,

    P. Bhui and N. Senroy, “Real-time prediction and control of transient stability using transient energy function,” IEEE Trans. Power Syst. , vol. 32, no. 2, pp. 923–934, Mar. 2017

    2017

  17. [25]

    Preventive-corrective coordinated transient stability dispatch of power systems with uncertain wind power,

    H. Yuan and Y . Xu, “Preventive-corrective coordinated transient stability dispatch of power systems with uncertain wind power,” IEEE Trans. Power Syst., vol. 35, no. 5, pp. 3616–3626, Sep. 2020

    2020

  18. [26]

    Dynamic security-constrained rescheduling of power systems using trajectory sensitivities,

    T. B. Nguyen and M. Pai, “Dynamic security-constrained rescheduling of power systems using trajectory sensitivities,”IEEE Trans. Power Syst., vol. 18, no. 2, pp. 848–854, May 2003

    2003

  19. [27]

    Deep sigma point processes-assisted chance-constrained power system transient stability preventive control,

    T. Su, J. Zhao, and X. Chen, “Deep sigma point processes-assisted chance-constrained power system transient stability preventive control,” IEEE Trans. Power Syst. , vol. 39, no. 1, pp. 1965–1978, Jan. 2024

    1965

  20. [28]

    MAT- POWER: 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, “MAT- POWER: Steady-state operations, planning, and analysis tools for power systems research and education,” IEEE Trans. Power Syst., vol. 26, no. 1, pp. 12–19, Feb. 2011

    2011

  21. [29]

    Kundur, Power System Stability and Control

    P. Kundur, Power System Stability and Control . New York, NY , USA: McGraw-Hill, 1994

    1994

  22. [30]

    An online power system stability monitoring system using convolutional neural networks,

    A. Gupta, G. Gurrala, and P. Sastry, “An online power system stability monitoring system using convolutional neural networks,” IEEE Trans. Power Syst., vol. 34, no. 2, pp. 864–872, Mar. 2019

    2019

  23. [31]

    Deep belief network enabled surrogate modeling for fast preventive control of power system transient stability,

    T. Su, Y . Liu, J. Zhao, and J. Liu, “Deep belief network enabled surrogate modeling for fast preventive control of power system transient stability,” IEEE Trans. Ind. Informat. , vol. 18, no. 1, pp. 315–326, Jan. 2022

    2022

  24. [32]

    Creation of synthetic electric grid models for transient stability studies,

    T. Xu, A. B. Birchfield, K. S. Shetye, and T. J. Overbye, “Creation of synthetic electric grid models for transient stability studies,” in Proc. Bulk Power Syst. Dynamics Control Symp. , Sep. 2017, pp. 1–6

    2017

Pith tools

Reviewed August 12, 2026 · model on record in the stance chip above.