arxiv: 2605.05105 · v1 · submitted 2026-05-06 · 📡 eess.SY · cs.SY· math.DS

Recognition: unknown

Minimizing the Expected Cost of Synchronization in Lossless Power Networks

Gerald Ogbonna , David Bindel , Lindsay C. Anderson

Authors on Pith no claims yet

Pith reviewed 2026-05-08 16:00 UTC · model grok-4.3

classification 📡 eess.SY cs.SYmath.DS

keywords power networkssynchronizationconvex optimizationgraph Laplacianphase cohesiontransientslinear matrix inequalitynetwork modification

0 comments

The pith

Optimally modifying the connections in lossless power networks minimizes the expected cost of synchronization transients through a convex optimization that incorporates a linear matrix inequality for phase cohesion.

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

Power networks with stochastic renewable generation experience costly transients during synchronization. The work models the network via graph Laplacian matrices and poses the minimization of expected synchronization cost as an optimization over possible network modifications. Under certain assumptions this program is convex, and the authors derive a linear matrix inequality whose feasibility certifies the existence and uniqueness of phase-cohesive steady-state angles. This LMI is included directly as a convex constraint, enabling the identification of critical links whose adjustment reduces transients, as demonstrated by simulations on the IEEE 30-bus test system.

Core claim

In lossless power networks the expected cost of synchronization can be minimized by solving a convex optimization problem over graph Laplacian matrices; feasibility of a derived linear matrix inequality guarantees the existence and uniqueness of phase-cohesive steady-state angles and can be imposed directly as a constraint, allowing the method to identify critical links whose modification reduces transients and improves performance metrics.

What carries the argument

Graph Laplacian matrices representing the network topology, optimized to minimize synchronization cost, with the feasibility of a linear matrix inequality acting as the convex certificate for the existence of unique phase-cohesive equilibria.

If this is right

Critical links can be identified for targeted modifications that reduce synchronization costs.
Dynamic simulations on standard test systems exhibit significant transient reduction and gains across multiple performance metrics.
The sparsity-optimality trade-off can be navigated by applying a reweighted l1 heuristic to the convex program.

Where Pith is reading between the lines

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

The geometric interpretations of the optimization may link to other problems of designing or controlling networked dynamical systems.
Adapting the linear matrix inequality for small line losses could extend the approach to more realistic lossy networks.
The framework might support online re-optimization when renewable output statistics change.

Load-bearing premise

The network is lossless and satisfies the assumptions required for the optimization problem to be convex.

What would settle it

A lossless network in which the linear matrix inequality is feasible yet multiple distinct phase-cohesive steady-state angle solutions exist, or dynamic simulations of the optimized network show no reduction in transients.

Figures

Figures reproduced from arXiv: 2605.05105 by David Bindel, Gerald Ogbonna, Lindsay C. Anderson.

**Figure 1.** Figure 1: Kron reduction on G with vertex set V = {1, . . . , 7} and boundary nodes VG = {1, 3, 7}. Suppose the edge weights aij ≥ 0 on the graph represent the conductances (i.e., admittance for power networks) of the branches {i, j} ∈ E. The vector of node voltages v and current injections J are related by Lv = J where L, the conductance matrix of the network, is the graph Laplacian matrix. The effective resistance… view at source ↗

**Figure 2.** Figure 2: Second-order oscillators coupled through the Network. view at source ↗

**Figure 3.** Figure 3: Spectral transformation of feedback loop. view at source ↗

**Figure 4.** Figure 4: The closed loop transfer function from disturbance view at source ↗

**Figure 5.** Figure 5: Weighted undirected Path graph on 3 nodes, with boundary nodes VG = {1, 2} and VL = {3}. The constant matrix C = blkdiag(Π2, 0) and for a regularization parameter β = 1, L(x)reg,β = L(x) + 1 3 11⊺ view at source ↗

**Figure 6.** Figure 6: Level sets of the objective function f(x) = trace(C⊺L −1 reg,β), the scaled-simplex Xα, and Xsync for α = 10 and γ = π/4 rad. (a) Feasible case (trivial solution): ψ = 2.0 p.u. (b) Feasible case (non-trivial solution): ψ = 3.0 p.u. (c) Infeasible case: ψ = 4.0 p.u. Proof. Using the definition of effective resistance, r eff ij (G(x)) = (ei − ej ) ⊺L(x) † (ei − ej ) = (ei − ej ) ⊺V Λ †V ⊺ (ei − ej ), where Λ… view at source ↗

**Figure 7.** Figure 7: shows that over the range of considered values of α, the optimal method (red line) clearly has a consistently better (lower) objective value. The proportional strategy, which simply continues the trend of how the edge weights have been previously allocated, has the worst performance overall for this Network. For instance, to reduce the size of the Gred to approximately 4.0, we only require α ≈ 15 via the o… view at source ↗

**Figure 8.** Figure 8: Optimal solution x ⋆ = arg min P1 for α = 50 on the IEEE 30-bus network. (a) Original network. Rtot(Gred) = 5.46. (b) The network postoptimization with optimal value Rtot(Gred(x ⋆)) = 2.91 and ∥x ⋆∥0 = 13. (a) (b) (c) view at source ↗

**Figure 9.** Figure 9: Comparing synchronization costs ∥ω∥2 for α = 50 allocated using strategies (1) - (3) to the optimal (4) for 500 samples of disturbance u0 ∼ N (0, I). (a) Optimal vs. Proportional, with a 15.79% average improvement. (b) Optimal vs. Uniform, with a 11.39% average improvement. (c) Optimal vs. Random Uniform, with a 14.64% average improvement. with parameters ψ = 0.45 p.u. and γ = π/4 rad. The optimal value of… view at source ↗

**Figure 10.** Figure 10: Optimal solution for α = 50 with γ-cohesiveness constraint for γ = π/4 rad and ψ = 0.45 p.u., ∥x ⋆∥0 = 12, and optimal value Rtot(Gred(x ⋆)) = 3.511. For the same set of random disturbances from Section V-B, the rewired network exhibits significantly improved transient response as shown view at source ↗

**Figure 11.** Figure 11: Optimal network for α = 0 with γ-cohesiveness constraint for γ = π/4 rad and ψ = 0.45 p.u., relaxing the non-negativity constraint on the design vector. (a) Optimal graph with Rtot(Gred(x ⋆)) = 1.8954 and ∥x ⋆∥0 = 41. (b) Phase and frequency trajectories on the original network for some realization of u0 (c) Phase and frequency trajectories on optimally rewired network for same u0. Using view at source ↗

**Figure 12.** Figure 12: Comparing ∥ω∥2 on the original network to the network with optimally reallocated total edge weights for the same samples of u0 ∼ N (0, I). The resulting average reduction in ∥ω∥2 is 36.19% view at source ↗

**Figure 14.** Figure 14: Reweighted ℓ1 solutions with and without γ-cohesiveness constraints for α = 50. (a) Solutions without γ-cohesiveness constraint. (b) Solutions with γ-cohesiveness constraint (γ = π/4, ψ = 0.45 p.u.). work is a convex problem and provide semidefinite programs for solving it. We further provide a sufficient condition, in terms of the algebraic connectivity λ2, for the existence and uniqueness of a synchroni… view at source ↗

read the original abstract

The reliable operation of large-scale electric power networks is increasingly challenging, particularly with the integration of stochastic renewable generation. In this work, we address the problem of minimizing network transients by optimally modifying the underlying network. We formulate the problem in terms of graph Laplacian matrices and show that, under certain assumptions, the problem is convex. We derive a linear matrix inequality whose feasibility guarantees the existence and uniqueness of phase cohesive steady-state angles; this condition can be directly incorporated as a convex constraint in the optimization framework and we provide several geometric interpretations of the optimization problem. The proposed method is validated on the IEEE 30-bus test system, where results demonstrate that our approach effectively identifies critical links on the network. Dynamic simulations show a significant reduction in network transients and overall improvements across several performance metrics. We explore the sparsity-optimality trade-off using a reweighted $\ell_1$ heuristic.

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 casts synchronization-cost minimization in lossless grids as a convex program by folding a phase-cohesiveness LMI into the graph-Laplacian design, then tests it on the IEEE 30-bus case with a reweighted-l1 sparsity heuristic.

read the letter

The core contribution is showing that, for lossless networks, you can minimize an expected transient cost while enforcing phase cohesiveness via a single LMI that stays convex when you optimize edge weights. They also give geometric readings of the problem and use the reweighted-l1 trick to trade off sparsity against performance. On the IEEE 30-bus system the method picks out a few critical links and the dynamic simulations show visibly smaller angle swings and better aggregate metrics. That is useful incremental work for the power-systems control crowd. The LMI is derived from standard swing-equation analysis and appears to be a sufficient condition rather than a tight necessary one, so the optimizer could in principle accept networks where multiple cohesive equilibria exist; the paper does not appear to stress-test that edge case after topology changes. The lossless assumption is stated up front but limits direct applicability once line resistances matter. Validation is limited to one standard test system with no Monte-Carlo sweep over renewable scenarios or head-to-head comparison against other topology-design heuristics. The math and citations look standard rather than circular. This is the sort of paper a control-oriented power-systems group would want to see in the literature; it is not foundational but it is cleanly executed and reproducible enough to be worth refereeing. I would bring it to a reading group for the LMI construction and the sparsity results, though I would not cite it in my own work unless I needed exactly this formulation.

Referee Report

1 major / 3 minor

Summary. The paper addresses minimizing expected synchronization cost (transients) in lossless power networks by optimizing modifications to the underlying graph Laplacian. Under unspecified assumptions, the formulation is claimed to be convex. A linear matrix inequality (LMI) is derived whose feasibility certifies existence and uniqueness of phase-cohesive steady-state angles; this LMI is incorporated directly as a convex constraint. Geometric interpretations are provided, the method is validated on the IEEE 30-bus system (showing reduced transients and improved metrics), and a reweighted ℓ1 heuristic is used to explore the sparsity-optimality trade-off.

Significance. If the convexity claim and the LMI-based guarantee hold rigorously, the work provides a tractable optimization-based approach to network redesign for improved synchronization under stochastic renewables, with direct applicability to power-system planning. The explicit incorporation of steady-state guarantees via LMI and the empirical validation on a standard test case are positive features; the sparsity exploration adds practical insight.

major comments (1)

[LMI derivation and convexity section (around the statement that feasibility guarantees existence and uniqueness)] The central claim that the derived LMI guarantees both existence and uniqueness of phase-cohesive angles (and remains valid after Laplacian modification) requires explicit verification. Standard power-flow analysis shows that uniqueness typically requires strict diagonal dominance or positive-definiteness of the reduced Jacobian; an LMI relaxation may be only sufficient for existence and could admit networks with multiple or non-cohesive solutions once edge weights are optimized. This directly affects the validity of using the LMI as a constraint in the optimization.

minor comments (3)

[Introduction and problem formulation] The assumptions under which convexity holds are stated as 'certain assumptions' in the abstract and introduction but are not listed explicitly with the problem formulation; a dedicated assumptions subsection or theorem statement would improve clarity.
[Numerical results and validation] The IEEE 30-bus validation reports reduced transients but lacks quantitative error bars, comparison to baseline methods (e.g., random or degree-based edge addition), or sensitivity analysis to the LMI parameters.
[Problem formulation] Notation for the graph Laplacian modifications and the expected-cost objective should be introduced with a clear table of symbols to avoid ambiguity when reading the optimization problem.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for the careful and constructive review of our manuscript. We address the major comment below.

read point-by-point responses

Referee: [LMI derivation and convexity section (around the statement that feasibility guarantees existence and uniqueness)] The central claim that the derived LMI guarantees both existence and uniqueness of phase-cohesive angles (and remains valid after Laplacian modification) requires explicit verification. Standard power-flow analysis shows that uniqueness typically requires strict diagonal dominance or positive-definiteness of the reduced Jacobian; an LMI relaxation may be only sufficient for existence and could admit networks with multiple or non-cohesive solutions once edge weights are optimized. This directly affects the validity of using the LMI as a constraint in the optimization.

Authors: We appreciate the referee pointing out the need for explicit verification of the uniqueness claim. In the manuscript, the LMI is derived directly from the requirement that the reduced Jacobian (for the lossless case) remains positive definite at the phase-cohesive equilibrium; positive-definiteness of this Jacobian is a standard sufficient condition for local uniqueness of the power-flow solution. Because the LMI is linear in the Laplacian entries, any feasible modification preserves the positive-definiteness property and therefore the uniqueness guarantee. We will add a dedicated appendix that explicitly connects the LMI to the Jacobian positive-definiteness condition and recalls the relevant power-flow uniqueness theorem, thereby making the argument self-contained. revision: yes

Circularity Check

0 steps flagged

No circularity; LMI and convexity derived from standard graph-theoretic analysis

full rationale

The paper formulates the synchronization cost minimization using graph Laplacians for lossless networks, derives an LMI feasibility condition from the power-flow equations to certify existence and uniqueness of phase-cohesive equilibria, and treats this LMI as a convex constraint in the optimizer. No step reduces by construction to a fitted parameter renamed as prediction, a self-definitional loop, or a load-bearing self-citation whose prior result itself depends on the current claim. The derivation relies on standard properties of symmetric Laplacians and Lyapunov-like stability conditions for swing dynamics, which are independent of the optimization outcome. Empirical validation on the IEEE 30-bus system is post-derivation and does not feed back into the claimed convexity or LMI guarantee. The central result therefore remains self-contained against external benchmarks.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

Abstract-only review; full derivation details unavailable. The formulation rests on the lossless-network assumption and unspecified conditions for convexity.

axioms (2)

domain assumption Network is lossless
Stated in title and abstract as the setting for the Laplacian formulation.
ad hoc to paper Problem is convex under certain assumptions
Abstract invokes this to claim tractability but does not list the assumptions.

pith-pipeline@v0.9.0 · 5454 in / 1385 out tokens · 40555 ms · 2026-05-08T16:00:14.063582+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

40 extracted references · 22 canonical work pages

[1]

Global Analysis of Synchronization Performance for Power Systems: Bridging the Theory-Practice Gap,

F. Paganini and E. Mallada, “Global Analysis of Synchronization Performance for Power Systems: Bridging the Theory-Practice Gap,” IEEE Transactions on Automatic Control, vol. 65, no. 7, pp. 3007–3022, Jul. 2020. [Online]. Available: https://ieeexplore.ieee.org/ document/8844770/

work page arXiv 2020
[2]

Chiang,Direct methods for stability analysis of electric power systems: theoretical foundation, BCU methodologies, and applications

H.-D. Chiang,Direct methods for stability analysis of electric power systems: theoretical foundation, BCU methodologies, and applications. Hoboken, N.J: Wiley, 2011

2011
[3]

Model Predictive Control for Power System Frequency Control Taking into Account Imbalance Uncertainty,

A. M. Ersdal, D. Fabozzi, L. Imsland, and N. F. Thornhill, “Model Predictive Control for Power System Frequency Control Taking into Account Imbalance Uncertainty,”IFAC Proceedings Volumes, vol. 47, no. 3, pp. 981–986, 2014. [Online]. Available: https://linkinghub.elsevier.com/retrieve/pii/S1474667016417428

2014
[4]

Global performance metrics for synchronization of heterogeneously rated power systems: The role of machine models and inertia,

F. Paganini and E. Mallada, “Global performance metrics for synchronization of heterogeneously rated power systems: The role of machine models and inertia,” in2017 55th Annual Allerton Conference on Communication, Control, and Computing (Allerton). Monticello, IL, USA: IEEE, Oct. 2017, pp. 324–331. [Online]. Available: http://ieeexplore.ieee.org/document/8262755/

work page arXiv 2017
[5]

Graph Laplacian Spectrum and Primary Frequency Regulation,

L. Guo, C. Zhao, and S. H. Low, “Graph Laplacian Spectrum and Primary Frequency Regulation,” in2018 IEEE Conference on Decision and Control (CDC). Miami Beach, FL: IEEE, Dec. 2018, pp. 158–165. [Online]. Available: https://ieeexplore.ieee.org/document/8619252/

work page arXiv 2018
[6]

Improving damping of power networks: Power scheduling and impedance adaptation,

E. Mallada and A. Tang, “Improving damping of power networks: Power scheduling and impedance adaptation,” inIEEE Conference on Decision and Control and European Control Conference. Orlando, FL, USA: IEEE, Dec. 2011, pp. 7729–7734. [Online]. Available: http://ieeexplore.ieee.org/document/6161287/

work page arXiv 2011
[7]

A fundamental study of inter-area oscillations in power systems,

M. Klein, G. Rogers, and P. Kundur, “A fundamental study of inter-area oscillations in power systems,”IEEE Transactions on Power Systems, vol. 6, no. 3, pp. 914–921, Aug. 1991. [Online]. Available: http://ieeexplore.ieee.org/document/119229/

1991
[8]

A critical review of cascading failure analysis and modeling of power system,

H. Guo, C. Zheng, H. H.-C. Iu, and T. Fernando, “A critical review of cascading failure analysis and modeling of power system,” Renewable and Sustainable Energy Reviews, vol. 80, pp. 9–22, Dec
[9]

Available: https://linkinghub.elsevier.com/retrieve/pii/ S1364032117308432

[Online]. Available: https://linkinghub.elsevier.com/retrieve/pii/ S1364032117308432
[10]

Improving robustness of complex networks via the effective graph resistance,

X. Wang, E. Pournaras, R. E. Kooij, and P. Van Mieghem, “Improving robustness of complex networks via the effective graph resistance,”The European Physical Journal B, vol. 87, no. 9, p. 221, Sep. 2014. [Online]. Available: http://link.springer.com/10.1140/epjb/e2014-50276-0

work page doi:10.1140/epjb/e2014-50276-0 2014
[11]

Designing Robust Networks of Coupled Phase-Oscillators with Applications to the High V oltage Electric Grid,

S. V . Nagpal, G. G. Nair, F. Parise, and C. L. Anderson, “Designing Robust Networks of Coupled Phase-Oscillators with Applications to the High V oltage Electric Grid,”IEEE Transactions on Control of Network Systems, pp. 1–12, 2022, conference Name: IEEE Transactions on Control of Network Systems

2022
[12]

Robust Graph Topologies for Networked Systems,

W. Abbas and M. Egerstedt, “Robust Graph Topologies for Networked Systems,”IFAC Proceedings Volumes, vol. 45, no. 26, pp. 85–90, Sep. 2012. [Online]. Available: https://linkinghub.elsevier.com/retrieve/ pii/S147466701534814X 16

2012
[13]

Graph Vulnerability and Robustness: A Survey,

S. Freitas, D. Yang, S. Kumar, H. Tong, and D. H. Chau, “Graph Vulnerability and Robustness: A Survey,”IEEE Transactions on Knowledge and Data Engineering, pp. 1–1, 2022. [Online]. Available: https://ieeexplore.ieee.org/document/9745316/

work page arXiv 2022
[14]

Schur-convex robustness measures in dynamical networks,

M. Siami and N. Motee, “Schur-convex robustness measures in dynamical networks,” in2014 American Control Conference. Portland, OR, USA: IEEE, Jun. 2014, pp. 5198–5203. [Online]. Available: http://ieeexplore.ieee.org/document/6859345/

work page arXiv 2014
[15]

Global Robustness vs. Local Vulnerabilities in Complex Synchronous Networks,

M. Tyloo and P. Jacquod, “Global Robustness vs. Local Vulnerabilities in Complex Synchronous Networks,”Physical Review E, vol. 100, no. 3, p. 032303, Sep. 2019, arXiv:1905.03582 [cond-mat, physics:nlin, physics:physics]. [Online]. Available: http://arxiv.org/abs/1905.03582

work page arXiv 2019
[16]

Resistance distance

D. J. Klein and M. Randi, “Resistance distance.”
[17]

Minimizing Effective Resistance of a Graph , Volume =

A. Ghosh, S. Boyd, and A. Saberi, “Minimizing Effective Resistance of a Graph,”SIAM Review, vol. 50, no. 1, pp. 37–66, Jan. 2008. [Online]. Available: http://epubs.siam.org/doi/10.1137/050645452

work page doi:10.1137/050645452 2008
[18]

Tractable approximation algorithms for the NP-hard problem of growing linear consensus networks,

M. Siami and N. Motee, “Tractable approximation algorithms for the NP-hard problem of growing linear consensus networks,” in2016 American Control Conference (ACC). Boston, MA, USA: IEEE, Jul. 2016, pp. 6429–6434. [Online]. Available: http: //ieeexplore.ieee.org/document/7526681/

work page arXiv 2016
[19]

An Effective Resistance based Genetic Algorithm for Community Detection:,

C. Pizzuti and A. Socievole, “An Effective Resistance based Genetic Algorithm for Community Detection:,” inProceedings of the 13th International Joint Conference on Computational Intelligence. Valletta, Malta: SCITEPRESS - Science and Technology Publications, 2021, pp. 28–36. [Online]. Available: https://www.scitepress.org/DigitalLibrary/ Link.aspx?doi=10...

work page doi:10.5220/0010644300003063 2021
[20]

Optimal network design for synchronization of coupled oscillators,

M. Fazlyab, F. D ¨orfler, and V . M. Preciado, “Optimal network design for synchronization of coupled oscillators,” Jun. 2017, arXiv:1503.07254 [math]. [Online]. Available: http://arxiv.org/abs/1503.07254

work page arXiv 2017
[21]

Optimal topology design for dynamic networks,

R. Dai and M. Mesbahi, “Optimal topology design for dynamic networks,” inIEEE Conference on Decision and Control and European Control Conference. Orlando, FL, USA: IEEE, Dec. 2011, pp. 1280–

2011
[22]

Available: http://ieeexplore.ieee.org/document/6161026/

[Online]. Available: http://ieeexplore.ieee.org/document/6161026/

work page arXiv
[23]

The Fastest Mixing Markov Process on a Graph and a Connection to a Maximum Variance Unfolding Problem,

J. Sun, S. Boyd, L. Xiao, and P. Diaconis, “The Fastest Mixing Markov Process on a Graph and a Connection to a Maximum Variance Unfolding Problem,”SIAM Review, vol. 48, no. 4, pp. 681–699, Jan. 2006. [Online]. Available: http://epubs.siam.org/doi/10.1137/S0036144504443821

work page doi:10.1137/s0036144504443821 2006
[24]

Flexible AC transmission,

N. Hingorani, “Flexible AC transmission,”IEEE Spectrum, vol. 30, no. 4, pp. 40–45, Apr. 1993. [Online]. Available: http://ieeexplore.ieee. org/document/206621/

1993
[25]

Kron,Tensor Analysis of Networks, ser

G. Kron,Tensor Analysis of Networks, ser. General Electric series. J. Wiley & Sons, Incorporated, 1939. [Online]. Available: https://books.google.com/books?id=MMlungEACAAJ

1939
[26]

Completely positive linear maps on complex matrices,

R. L. Smith, “Some interlacing properties of the Schur complement of a Hermitian matrix,”Linear Algebra and its Applications, vol. 177, pp. 137–144, Dec. 1992. [Online]. Available: https: //linkinghub.elsevier.com/retrieve/pii/002437959290321Z

work page arXiv 1992
[27]

Graph Spectra and Modal Dynamics of Oscillatory Net- works

B. Ayazifar, “Graph Spectra and Modal Dynamics of Oscillatory Net- works.”
[28]

Synchronization of Power Networks: Network Reduction and Effective Resistance,

F. Dorfler and F. Bullo, “Synchronization of Power Networks: Network Reduction and Effective Resistance,”IFAC Proceedings Volumes, vol. 43, no. 19, pp. 197–202, 2010. [Online]. Available: https://linkinghub.elsevier.com/retrieve/pii/S1474667016309673

2010
[29]

S. P. Boyd and L. Vandenberghe,Convex optimization, version 29 ed. Cambridge New York Melbourne New Delhi Singapore: Cambridge University Press, 2023

2023
[30]

S. P. Boyd, L. El Ghaoui, E. Feron, and V . Balakrishnan,Linear matrix inequalities in system and control theory, ser. SIAM studies in applied mathematics. Philadelphia, Pa: SIAM, Society for Industrial and Applied Mathematics, 1994, no. 15

1994
[31]

Electrical Networks and Algebraic Graph Theory: Models, Properties, and Applications,

F. Dorfler, J. W. Simpson-Porco, and F. Bullo, “Electrical Networks and Algebraic Graph Theory: Models, Properties, and Applications,” Proceedings of the IEEE, vol. 106, no. 5, pp. 977–1005, May 2018. [Online]. Available: https://ieeexplore.ieee.org/document/8347206/

work page arXiv 2018
[32]

Cardinality Minimization, Constraints, and Regularization: A Survey,

A. M. Tillmann, D. Bienstock, A. Lodi, and A. Schwartz, “Cardinality Minimization, Constraints, and Regularization: A Survey,” Aug. 2022, arXiv:2106.09606 [math]. [Online]. Available: http: //arxiv.org/abs/2106.09606

work page arXiv 2022
[33]

Rank minimization and applications in system theory,

M. Fazel, H. Hindi, and S. Boyd, “Rank minimization and applications in system theory,” inProceedings of the 2004 American Control Conference. Boston, MA, USA: IEEE, 2004. [Online]. Available: https://ieeexplore.ieee.org/document/1384521/

work page arXiv 2004
[34]

Enhancing Sparsity by Reweighted l 1 Minimization,

E. J. Cand `es, M. B. Wakin, and S. P. Boyd, “Enhancing Sparsity by Reweighted l 1 Minimization,”Journal of Fourier Analysis and Applications, vol. 14, no. 5-6, pp. 877–905, Dec. 2008. [Online]. Available: http://link.springer.com/10.1007/s00041-008-9045-x

work page doi:10.1007/s00041-008-9045-x 2008
[35]

Majorization-Minimization Algorithms in Signal Processing, Communications, and Machine Learning,

Y . Sun, P. Babu, and D. P. Palomar, “Majorization-Minimization Algorithms in Signal Processing, Communications, and Machine Learning,”IEEE Transactions on Signal Processing, vol. 65, no. 3, pp. 794–816, Feb. 2017. [Online]. Available: http://ieeexplore.ieee.org/ document/7547360/

work page arXiv 2017
[36]

MATPOWER: Steady-State Operations, Planning, and Analysis Tools for Power Systems Research and Education,

R. D. Zimmerman, C. E. Murillo-Sanchez, 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, Feb. 2011. [Online]. Available: http://ieeexplore.ieee.org/document/5491276/

work page arXiv 2011
[37]

CVXPY: A Python-embedded modeling lan- guage for convex optimization,

S. Diamond and S. Boyd, “CVXPY: A Python-embedded modeling lan- guage for convex optimization,”Journal of Machine Learning Research, vol. 17, no. 83, pp. 1–5, 2016

2016
[38]

ApS,The MOSEK Python Fusion API manual

M. ApS,The MOSEK Python Fusion API manual. Version 11.0.,
[39]

Available: https://docs.mosek.com/latest/pythonfusion/ index.html

[Online]. Available: https://docs.mosek.com/latest/pythonfusion/ index.html
[40]

Synchronization in complex oscillator networks and smart grids,

F. D ¨orfler, M. Chertkov, and F. Bullo, “Synchronization in complex oscillator networks and smart grids,”Proceedings of the National Academy of Sciences, vol. 110, no. 6, pp. 2005–2010, Feb. 2013. [Online]. Available: https://pnas.org/doi/full/10.1073/pnas.1212134110 PLACE PHOTO HERE Gerald Ogbonnais a PhD candidate in the Sys- tems Engineering departmen...

work page doi:10.1073/pnas.1212134110 2005