Pith. sign in

REVIEW 2 major objections 5 minor 32 references

Optimizing Flexibility in Power Systems by Maximizing the Region of Manageable Uncertainties

T0 review · 2 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 The paper establishes that a power grid's flexibility—the range of uncertain loads and generations it can safely absorb—can be maximized by solving an existence-constrained semi-infinite optimization problem that returns rigorous bounds…

desk verdict A credible extension of semi-infinite optimization to power-system flexibility maximization, with two caveats: an unverified existence assumption and 5% solver gaps that weaken the claim of rigorous global bounds. read the letter →

arxiv 2411.18178 v3 pith:VE2LTRGX submitted 2024-11-27 math.OC

Aron Zingler , Stephane Fliscounakis , Patrick Panciatici , Alexander Mitsos This is my paper · ORCID
classification math.OC MSC 90C3490C1190C26
keywords flexibilityanalysissemi-infiniteprogrammingoperationunderuncertaintyoptimalpowerflowDCapproximationmixed-integerworst-casetransfercapacity
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

The paper defines a power grid's flexibility as the size of the region of uncertain load and generation values that can be managed, meaning that for every value in the region there is a control action keeping the grid within limits. It then formulates the task of maximizing this region as an existence-constrained semi-infinite optimization problem, where one decision (the preventive generator set-points) is made before the uncertainty is known and a second decision (phase-shifter settings and bus merges) is made after. A specialized discretization algorithm solves this problem for a DC-flow grid model with mixed-integer controls, returning a lower and an upper bound on the flexibility index and the preventive actions that achieve the larger region. If the method works as claimed, an operator could use it to answer concrete questions such as how much the injections in one region can deviate before a line overloads, or what additional power transfer capacity between two regions can be guaranteed in the worst case.

What carries the argument

The load-bearing object is the flexibility index $\delta$, which scales a user-chosen parametric uncertainty region $T(\delta,x)$; an existence-constrained semi-infinite program is an optimization problem with a constraint that must hold for every uncertainty value and, for each value, asserts the existence of a control. The paper's Proposition 1 is the identity that carries the argument: the existential constraint over the variable region $T(\delta,x)$ is replaced by a min-of-two-objectives constraint over the fixed host set $\bar Y$, making the problem solvable by standard discretization. At each iteration the algorithm solves a relaxed mixed-integer linear master problem and a maxmin worst-case problem that adds the most violating uncertainty point to the discretization; two specializations drop redundant discretization points and transform retained points to the boundary of the shrinking hyperbox. The numerical engine is the uniqueness and continuity of the implicit grid-state function $s(x,y,z)$ together with the load-distribution model, which fixes the generator control offsets uniquely once the uncertainty and set-points are given.

What would settle it

Take a returned solution $(\delta,x)$ and evaluate the worst-case generation problem over the full host set; if any point $y$ inside $T(\delta,x)$ has no feasible control $z$ satisfying $g(x,y,z)\le0$, the lower-bound certificate is wrong. A simpler direct check on the motivating example is to enumerate the polyhedron of manageable injections shown in the paper and verify that the returned hyperbox is truly contained in it.

Watch

Extended reading notes

Core claim

The central discovery is that the maximal manageable uncertainty region can be optimized, not merely evaluated, by turning the problem into an existence-constrained semi-infinite program and solving it with an adaptive discretization scheme. The key step is a reformulation: when the uncertainty region has the form $T(x,\delta)=\{y\in\bar Y:h(x,y)\le\delta\}$ with continuous $h$, the constraint that every $y$ in $T(x,\delta)$ admits a feasible control is equivalent, up to an arbitrarily small objective error, to requiring for all $y$ in the fixed host set $\bar Y$ that $\min_{z\in Z(x,y)}\min\{\alpha(\delta-h(x,y)),g(x,y,z)\}\le0$. This removes the dependence of the uncertainty set on the preventive actions and lets the algorithm iterate between a relaxed master problem and a worst-case generation problem. For the power-system model, which uses the DC flow approximation, piecewise-linear phase shifters, bus merging, and a unique implicit load-distribution response, each subproblem is a mixed-integer linear program. The result is a rigorous interval for the flexibility index and guaranteed preventive actions for the returned uncertainty region.

Load-bearing premise

Everything rests on the assumption that, for every preventive action and every uncertainty in the host set, the grid state is uniquely determined and at least one feasible control action exists; the paper assumes this rather than proving it for all iterates.

Editorial extensions

If this is right

  • With the scaled-hyperbox parameterization, the returned $\delta$ is a guaranteed inner approximation: every uncertainty inside the box $[y_0-\Delta^-\delta,\,y_0+\Delta^+\delta]$ can be managed, and no larger scaled box can be certified.
  • With the transfer parameterization, the result guarantees that every additional power transfer between two regions up to the returned capacity is safe, not merely that the maximal transfer is safe.
  • For fixed preventive actions, the auxiliary flexibility-index problem can be solved on its own, giving a flexibility measure for an already chosen operating point.
  • The method can in principle be adapted to any continuous parameterization $h$ of the uncertainty region, which the medium-scale experiments exploit by comparing the hyperbox and transfer formulations.
  • On the medium-scale French-network instance, the transfer formulation returns approximately $[4530,4760]$ MW of guaranteed France-to-Spain additional transfer capacity in about 300 seconds; the hyperbox formulation returns $\delta\in[0.546,0.573]$ in about 83 minutes.

Reading between the lines

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

  • The strong dependence of run time on the scaling parameter $\alpha$, which the paper documents but does not resolve, suggests an adaptive $\alpha$-update rule as a natural extension; the authors leave the choice of a systematic value open.
  • Because Proposition 1 only needs $h$ to be continuous, the same discretization could be tested on rotated ellipsoids, zonotopes, or other convex bodies, giving less conservative approximations of the manageable region than an axis-aligned hyperbox.
  • The paper's stated assumption that grid states are uniquely determined by $(x,y,z)$ could be checked numerically on returned iterates by computing the implicit function residual; if residuals are not small, the certified bounds would not be valid for the actual model.
  • A direct practical extension is to treat $\alpha$ and the integrality tolerance as tunable hyperparameters on a benchmark set, since the observed time-outs indicate that solver settings materially change which instances can be solved.
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

2 major / 5 minor

Summary. The manuscript proposes a method for maximizing a scalar flexibility index δ that measures the size of a parametrized uncertainty region T(δ, x) that a DC-power-flow grid can manage in the worst case through preventive actions x and corrective controls z, including mixed-integer controls such as bus merging. The problem is formulated as an existence-constrained semi-infinite optimization problem, and the authors prove an exact reformulation (Proposition 1) that replaces the x-dependent uncertainty set with a fixed host set, then specialize the discretization algorithm of Djelassi and Mitsos (2021). The method is tested on a two-generator motivating example, a modified 30-bus IEEE system, and a 6,800-node RTE-based instance, for two parameterizations: a scaled hyperbox of uncertain injections and an additional net power-transfer capacity between two regions. The reported results are bracketed intervals, e.g., δ ∈ [0.925, 0.954] and δ ∈ [0.590, 0.614] on the small instance, and δ ∈ [4530, 4760] MW and δ ∈ [0.546, 0.573] on the medium instance.

Significance. If the algorithmic guarantees are valid, the paper represents a useful methodological contribution: it connects the process-systems flexibility index to existence-constrained semi-infinite optimization in a power-system setting, accommodates mixed-integer controls, and demonstrates computational tractability on a medium-scale DC-flow instance. The reformulation in Proposition 1 is a helpful device, and the transformation-based specialization of Section 6.3.2 is a practical enhancement. The numerical section is transparent about wall-clock variability and timeouts, and the medium-scale instance data are made available. The central caveat is that the rigorous validity of the reported bounds depends on a nonempty-control feasibility condition that is not verified for all reported instances; this is the main issue that must be resolved before the paper's claims are fully supported.

major comments (2)
  1. [Section 4 and Section 5.3.2, Eq. (9)] The global-validity guarantee stated in the abstract and Section 6.2 is conditional on the assumption, made in Section 4, that Z(x, y) is nonempty on X × Ybar. In the concrete DC model, this reduces to the load-distribution feasibility condition (9). The paper neither includes (9) as a constraint in the optimization problems (13) and (14) nor verifies it over the full host set for the transfer-capacity parametrization of Section 7.2. For the hyperbox parametrization, Section 8.1 chooses δ_UB to guarantee (9), and Section 8.2 states that generator bounds were increased 'to avoid a violation of (9)', but no such verification is reported for the medium-scale instance of Section 8.3. If (9) fails for some y ∈ Ybar, the expression min_{z ∈ Z(x,y)} ... is not well defined, the worst-case generation subproblem (12f) can encounter infeasible inner problems, and the intervals δ ∈ [4530, 4760] MW and δ ∈ [0.546, 0.573] are not rigorously justified. Please either impose (9) explicitly, verify it for all (x, y) encountered, or restrict Ybar and state the resulting limitation.
  2. [Section 6.1, Eq. (12g)] The displayed equivalence '0 ≥ sup_y min_z min{α(δ−h), g} ⇔ ∀y∈Ybar [∃z∈Z(x,y): g(x,y,z) ≤ 0]' is not correct. The right-hand side should be '∀y∈Ybar [∃z∈Z(x,y): min{α(δ−h(x,y)), g(x,y,z)} ≤ 0]' (equivalently, with the min expression inside the quantifier). As printed, the proof asserts a strictly stronger statement, since it would require g ≤ 0 for every y in the host set, which is not equivalent to the original constraint. The proposition statement is correct, so this is a fixable proof error, but it must be corrected before the central reformulation can be considered proved.
minor comments (5)
  1. [Section 6.1, Proposition 1 statement] The displayed reformulation in the proposition statement is missing the '≤ 0' after the min expression; compare with equation (12e).
  2. [Section 6.3.2, Eq. (14)] The transformation divides by h(y_d); the case h(y_d) = 0 (for example y_d = y0) is not discussed, and a non-zero assumption should be stated explicitly.
  3. [Section 8.1 and Fig. 5] The reported δ ∈ [-1.857, -1.857] is the algorithm's bracket; it would be clearer to state explicitly that this is a computed interval up to the chosen tolerances, and to explain the negative sign convention (the objective is -δ).
  4. [Section 7.2, Eq. (19)] The role of the αh(x, y) term inside the constraint g is not immediately clear; a sentence explaining that this encodes the min-form of the Proposition 1 reformulation would improve readability.
  5. [Tables 2 and 3] The paper reports ranges over three repetitions but does not state whether the repetitions differ in random seeds or solver settings; adding this information would help readers interpret the variability.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the flexibility bounds are produced by solving the stated optimization, and the cited prior algorithms are independent support rather than inputs that force the results.

full rationale

The paper's central derivation is self-contained. The flexibility index δ is the decision variable of the optimization, not a fitted parameter; the reported intervals [δ_LB, δ_UB] are computed by lower- and upper-bounding MILP subproblems. Proposition 1 is proved in Eqs. (12a)-(12g), including the exactness argument (δ−ε), so the reformulation to a fixed uncertainty set does not assume the conclusion. The discretization algorithm and grid model are taken from the authors' prior work (Djelassi and Mitsos 2021; Djelassi et al. 2018), but those citations supply an independently published algorithmic framework and model equations; they are not used to forbid alternatives or to define the present result into existence. No quantity is fitted and then renamed as a prediction, and no ansatz is smuggled in through citation: the PST piecewise-linear model and load-distribution equation (8) are explicit modeling assumptions. The main caveat is validity, not circularity: Section 4 assumes Z(x,y) nonempty and states unique on X×Ybar, and Section 5 notes this requires (9); the paper verifies (9) for the hyperbox parameterization via δ_UB but not for the transfer-capacity case. If (9) fails for some y, the inner control offsets do not exist and the claimed bounds would not be valid for those points. That is an unverified assumption affecting correctness, not a case where the derivation reduces to its own inputs.

Assumptions & free parameters 3 free parameters · 5 assumptions · 0 invented entities

The formulation introduces no new physical entities. Imported assumptions are the DC flow approximation, uniqueness of the implicit grid state with nonempty control sets, uniqueness of the load-distribution solution under condition (9), and global solution of the MILP subproblems. Free parameters are algorithmic tuning choices (alpha, tolerances, restriction values) that influence runtimes and interval tightness but not the validity of the reported bounds.

free parameters (3)
  • Alpha scaling constant in the exact reformulation = alpha' = 0.5 (small instance), alpha' = 10 (medium instance), alpha = alpha' / delta_norm
    Introduced in Proposition 1 to weight the two objectives in worst-case generation (12f). Runtime varies by more than 2x across alpha values (Tables 2-3) and no automatic selection rule is given (Section 8.2.1).
  • Initial restriction epsilon_R^0 = 0.05 for the main algorithm, 0.005 for the auxiliary problem
    Parameter of the restriction-of-right-hand-side method (Mitsos 2011); set by hand under the stated tolerances (Section 8).
  • Relative optimality tolerance = 0.05 overall, 0.025 auxiliary
    Termination tolerance for the MILP solver; the reported delta intervals are valid only within this tolerance (Section 8).
assumptions (5)
  • domain assumption DC power flow approximation is an adequate model of grid operation for this analysis (affine line flows, Eq. (2))
    Used in all numerical claims; the paper states in the conclusion that nonlinear AC behavior would be missed.
  • domain assumption Grid state is uniquely given by a continuous implicit function s(x,y,z) and Z(x,y) is nonempty on X times Ybar
    Stated in Section 4; required for the ESIP formulation and for Proposition 1; not proven for all algorithm iterates.
  • domain assumption The load-distribution control via the mid-function (8) has a unique solution whenever condition (9) holds
    Assumed in Section 5.3.2; degeneracies of the mid function could admit multiple or no solutions.
  • domain assumption MILP subproblems are solved to global optimality within stated relative gaps by Gurobi 10.0.2
    Global subproblem solutions are required for the convergence proof (Sections 6.1-6.2); the implementation uses a 5 percent relative gap (Section 8).
  • standard math Convergence and finite termination of the adaptive discretization algorithm (Blankenship-Falk 1976; Mitsos 2011)
    Inherited from published results and relied on for the claimed rigorous bounds.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Optimizing Flexibility in Power Systems by Maximizing the Region of Manageable Uncertainties." pith.science (2026). https://pith.science/paper/VE2LTRGX

@misc{pith2026241118178,
  author       = {Pith},
  title        = {Pith review of: Optimizing Flexibility in Power Systems by Maximizing the Region of Manageable Uncertainties},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/VE2LTRGX}},
  note         = {Machine review of arXiv:2411.18178}
}
read the original abstract

Motivated by the increasing need to hedge against load and generation uncertainty in the operation of power grids, we propose flexibility maximization during operation. We consider flexibility explicitly as the amount of uncertainty that can be handled while still ensuring nominal grid operation in the worst-case. We apply the proposed flexibility optimization in the context of a DC flow approximation. By using a corresponding parameterization, we can find the maximal range of uncertainty and a range for the manageable power transfer between two parts of a network subject to uncertainty. We formulate the corresponding optimization problem as an (existence-constrained) semi-infinite optimization problem and specialize an existing algorithm for its solution.

Figures

Figures reproduced from arXiv: 2411.18178 by the authors.

Figure 1
Figure 1. Motivating power grid instance with two generators and two uncertain [PITH_FULL_IMAGE:figures/full_fig_p004_1.png] view at source ↗
Figure 2
Figure 2. Illustration of the behavior of phase shifters. [PITH_FULL_IMAGE:figures/full_fig_p007_2.png] view at source ↗
Figure 3
Figure 3. Illustration of the behavior of generator [PITH_FULL_IMAGE:figures/full_fig_p008_3.png] view at source ↗
Figures from the paper (3 more)
Figure 4
Figure 4. Figure 4: Illustration of the behavior of the algorithm with and without transfor [PITH_FULL_IMAGE:figures/full_fig_p012_4.png]
Figure 5
Figure 5. Figure 5: Visualization of the result from the motivating example for the final pre [PITH_FULL_IMAGE:figures/full_fig_p015_5.png]
Figure 6
Figure 6. Figure 6: Visualization of the small grid instance with two regions marked. [PITH_FULL_IMAGE:figures/full_fig_p016_6.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

32 extracted references · 22 canonical work pages

  1. [1]

    Journal of Modern Power Systems and Clean Energy 7(5):987--1007, doi:https://doi.org/10.1007/s40565-019-0527-4

    Akrami A, Doostizadeh M, Aminifar F (2019) Power system flexibility: an overview of emergence to evolution. Journal of Modern Power Systems and Clean Energy 7(5):987--1007, doi:https://doi.org/10.1007/s40565-019-0527-4

    work page doi:10.1007/s40565-019-0527-4 2019

  2. [2]

    Energy Reports 6:101--106, doi:https://doi.org/10.1016/j.egyr.2019.11.048

    Babatunde OM, Munda JL, Hamam Y (2020) Power system flexibility: A review. Energy Reports 6:101--106, doi:https://doi.org/10.1016/j.egyr.2019.11.048

    work page doi:10.1016/j.egyr.2019.11.048 2020

  3. [3]

    Journal of Optimization Theory and Applications 19(2):261--281, doi:https://doi.org/10.1007/bf00934096

    Blankenship JW, Falk JE (1976) Infinitely constrained optimization problems. Journal of Optimization Theory and Applications 19(2):261--281, doi:https://doi.org/10.1007/bf00934096

    work page doi:10.1007/bf00934096 1976

  4. [4]

    In: 2015 IEEE Eindhoven PowerTech, IEEE, IEEE, pp 1--6, doi:https://doi.org/10.1109/ptc.2015.7232514

    Bucher MA, Delikaraoglou S, Heussen K, Pinson P, Andersson G (2015) On quantification of flexibility in power systems. In: 2015 IEEE Eindhoven PowerTech, IEEE, IEEE, pp 1--6, doi:https://doi.org/10.1109/ptc.2015.7232514

    work page arXiv 2015

  5. [5]

    In: 2021 IEEE Madrid PowerTech, IEEE, pp 1--6, doi:https://doi.org/10.1109/powertech46648.2021.9494935

    Capitanescu F (2021) Power system flexibility region under uncertainty with respect to congestion and voltage constraints. In: 2021 IEEE Madrid PowerTech, IEEE, pp 1--6, doi:https://doi.org/10.1109/powertech46648.2021.9494935

    arXiv 2021

  6. [6]

    Dissertation, Rheinisch-Westfälische Technische Hochschule Aachen, doi:https://doi.org/10.18154/RWTH-2020-09163

    Djelassi H (2020) D iscretization-based algorithms for the global solution of hierarchical programs. Dissertation, Rheinisch-Westfälische Technische Hochschule Aachen, doi:https://doi.org/10.18154/RWTH-2020-09163

    work page doi:10.18154/rwth-2020-09163 2020

  7. [7]

    Journal of Global Optimization 68(2):227--253, doi:https://doi.org/10.1007/s10898-016-0476-7

    Djelassi H, Mitsos A (2017) A hybrid discretization algorithm with guaranteed feasibility for the global solution of semi-infinite programs. Journal of Global Optimization 68(2):227--253, doi:https://doi.org/10.1007/s10898-016-0476-7

    work page doi:10.1007/s10898-016-0476-7 2017

  8. [8]

    Journal of Optimization Theory and Applications 188(3):863--881, doi:https://doi.org/10.1007/s10957-021-01813-2

    Djelassi H, Mitsos A (2021) Global solution of semi-infinite programs with existence constraints. Journal of Optimization Theory and Applications 188(3):863--881, doi:https://doi.org/10.1007/s10957-021-01813-2

    work page doi:10.1007/s10957-021-01813-2 2021

Show all 32 references
  1. [9]

    In: 2018 Power Systems Computation Conference (PSCC), IEEE, pp 1--7, doi:https://doi.org/10.23919/pscc.2018.8444136

    Djelassi H, Fliscounakis S, Mitsos A, Panciatici P (2018) Hierarchical programming for worst-case analysis of power grids. In: 2018 Power Systems Computation Conference (PSCC), IEEE, pp 1--7, doi:https://doi.org/10.23919/pscc.2018.8444136

    2018

  2. [10]

    Journal of Global Optimization 75(2):341--392, doi:https://doi.org/10.1007/s10898-019-00764-3

    Djelassi H, Glass M, Mitsos A (2019) Discretization-based algorithms for generalized semi-infinite and bilevel programs with coupling equality constraints. Journal of Global Optimization 75(2):341--392, doi:https://doi.org/10.1007/s10898-019-00764-3

    2019 doi

  3. [11]

    Naval Research Logistics Quarterly 24(3):441--450, doi:https://doi.org/10.1002/nav.3800240307

    Falk JE, Hoffman K (1977) A nonconvex max‐min problem. Naval Research Logistics Quarterly 24(3):441--450, doi:https://doi.org/10.1002/nav.3800240307

    1977 doi

  4. [12]

    In: 2019 IEEE Power & Energy Society General Meeting (PESGM), IEEE, pp 1--5, doi:https://doi.org/10.1109/pesgm40551.2019.8974032

    G \'o mez JD, Borda CE (2019) Operational flexibility in power systems—a geometrical view. In: 2019 IEEE Power & Energy Society General Meeting (PESGM), IEEE, pp 1--5, doi:https://doi.org/10.1109/pesgm40551.2019.8974032

    2019

  5. [13]

    Computers & Chemical Engineering 11(6):675--693, doi:https://doi.org/10.1016/0098-1354(87)87011-4

    Grossmann IE, Floudas CA (1987) Active constraint strategy for flexibility analysis in chemical processes. Computers & Chemical Engineering 11(6):675--693, doi:https://doi.org/10.1016/0098-1354(87)87011-4

    1987 doi

  6. [14]

    Computers & Chemical Engineering 7(4):439--462, doi:https://doi.org/10.1016/0098-1354(83)80022-2

    Grossmann IE, Halemane KP, Swaney RE (1983) Optimization strategies for flexible chemical processes. Computers & Chemical Engineering 7(4):439--462, doi:https://doi.org/10.1016/0098-1354(83)80022-2

    1983 doi

  7. [15]

    OECD, doi:https://doi.org/10.1787/9789264111394-en, ://www.oecd-ilibrary.org/content/publication/9789264111394-en

    International Energy Agency (2011) Harnessing Variable Renewables: A Guide to the Balancing Challenge. OECD, doi:https://doi.org/10.1787/9789264111394-en, ://www.oecd-ilibrary.org/content/publication/9789264111394-en

    2011

  8. [16]

    IEEE Transactions on Power Systems 36(6):4953--4966, doi:https://doi.org/10.1109/tpwrs.2021.3075925

    Lee D, Turitsyn K, Molzahn DK, Roald LA (2021) Robust ac optimal power flow with robust convex restriction. IEEE Transactions on Power Systems 36(6):4953--4966, doi:https://doi.org/10.1109/tpwrs.2021.3075925

    2021

  9. [17]

    In: 2013 IEEE power & energy society general meeting, IEEE, pp 1--11, doi:https://doi.org/10.1109/tste.2012.2212471

    Ma J, Silva V, Belhomme R, Kirschen DS, Ochoa LF (2013) Evaluating and planning flexibility in sustainable power systems. In: 2013 IEEE power & energy society general meeting, IEEE, pp 1--11, doi:https://doi.org/10.1109/tste.2012.2212471

    2013

  10. [18]

    In: 2011 IEEE power and energy society general meeting, IEEE, pp 1--8, doi:https://doi.org/10.1109/pes.2011.6039617

    Menemenlis N, Huneault M, Robitaille A (2011) Thoughts on power system flexibility quantification for the short-term horizon. In: 2011 IEEE power and energy society general meeting, IEEE, pp 1--8, doi:https://doi.org/10.1109/pes.2011.6039617

    2011

  11. [19]

    Optimization 60(10-11):1291--1308, doi:https://doi.org/10.1080/02331934.2010.527970

    Mitsos A (2011) Global optimization of semi-infinite programs via restriction of the right-hand side. Optimization 60(10-11):1291--1308, doi:https://doi.org/10.1080/02331934.2010.527970

    2011

  12. [20]

    Journal of Global Optimization 61(1):1--17, doi:https://doi.org/10.1007/s10898-014-0146-6

    Mitsos A, Tsoukalas A (2015) Global optimization of generalized semi-infinite programs via restriction of the right hand side. Journal of Global Optimization 61(1):1--17, doi:https://doi.org/10.1007/s10898-014-0146-6

    2015 doi

  13. [21]

    IEEE Transactions on Sustainable Energy 6(3):800--809, doi:https://doi.org/10.1109/tste.2015.2410760

    Nosair H, Bouffard F (2015) Flexibility envelopes for power system operational planning. IEEE Transactions on Sustainable Energy 6(3):800--809, doi:https://doi.org/10.1109/tste.2015.2410760

    2015

  14. [22]

    European Journal of Operational Research 119(2):301--313, doi:https://doi.org/10.1016/s0377-2217(99)00132-0

    Still G (1999) Generalized semi-infinite programming: Theory and methods. European Journal of Operational Research 119(2):301--313, doi:https://doi.org/10.1016/s0377-2217(99)00132-0

    1999 doi

  15. [23]

    European Transactions on Electrical Power 22(6):846--865, doi:https://doi.org/10.1002/etep.617

    Wei-qing S, Cheng-min W, Yan Z (2012) Flexibility evaluation and flexible comprehensive optimization in power systems. European Transactions on Electrical Power 22(6):846--865, doi:https://doi.org/10.1002/etep.617

    2012 doi

  16. [24]

    Computers & Chemical Engineering 166:107969, doi:https://doi.org/10.1016/j.compchemeng.2022.107969

    Zhao F, Ochoa MP, Grossmann IE, Garc \' a-Mu \ n oz S, Stamatis SD (2022) Novel formulations of flexibility index and design centering for design space definition. Computers & Chemical Engineering 166:107969, doi:https://doi.org/10.1016/j.compchemeng.2022.107969

    2022

  17. [25]

    IEEE Transactions on Power Systems 31(1):339--347, doi:https://doi.org/10.1109/TPWRS.2015.2390038

    Zhao J, Zheng T, Litvinov E (2015) A unified framework for defining and measuring flexibility in power system. IEEE Transactions on Power Systems 31(1):339--347, doi:https://doi.org/10.1109/TPWRS.2015.2390038

    2015

  18. [26]

    ://publications.rwth-aachen.de/record/828516

    Zingler A, Fliscounakis S (2021) M edium-scale power grid instance with 6800 nodes for DC -flow approximation. ://publications.rwth-aachen.de/record/828516

    2021

  19. [27]

    , " * write output.state after.block = add.period write newline

    ENTRY address archive author booktitle chapter doi edition editor eid eprint howpublished institution journal key month note number organization pages publisher school series title type url volume year label extra.label sort.label short.list INTEGERS output.state before.all mi...

  20. [28]

    write newline

    " write newline "" before.all 'output.state := FUNCTION add.period duplicate empty 'skip "." * add.blank if FUNCTION if.digit duplicate "0" = swap duplicate "1" = swap duplicate "2" = swap duplicate "3" = swap duplicate "4" = swap duplicate "5" = swap duplicate "6" = swap dupl...

  21. [29]

    , " * write output.state after.block = add.period write newline

    ENTRY address author booktitle chapter doi edition editor eid howpublished institution journal key month note number organization pages publisher school series title type url volume year label INTEGERS output.state before.all mid.sentence after.sentence after.block FUNCTION in...

  22. [30]

    write newline

    " write newline "" before.all 'output.state := FUNCTION if.digit duplicate "0" = swap duplicate "1" = swap duplicate "2" = swap duplicate "3" = swap duplicate "4" = swap duplicate "5" = swap duplicate "6" = swap duplicate "7" = swap duplicate "8" = swap "9" = or or or or or or...

  23. [31]

    , " * write output.state after.block = add.period write newline

    ENTRY address author booktitle chapter doi edition editor eid howpublished institution journal key month note number organization pages publisher school series title type url volume year label INTEGERS output.state before.all mid.sentence after.sentence after.block FUNCTION in...

  24. [32]

    write newline

    " write newline "" before.all 'output.state := FUNCTION if.digit duplicate "0" = swap duplicate "1" = swap duplicate "2" = swap duplicate "3" = swap duplicate "4" = swap duplicate "5" = swap duplicate "6" = swap duplicate "7" = swap duplicate "8" = swap "9" = or or or or or or...

Pith tools

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