{"id":"f6bb7af5-dc70-46a6-b95f-4e2e39253351","arxiv_id":"2502.07910","paper_version":3,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":4,"one_line_summary":"Uniform suction and blowing eliminate separation on a NACA4412 wing at 11 degrees angle of attack and increase lift up to 11.5 percent, but drag increases enough to reduce aerodynamic efficiency.","lead":"High-fidelity simulations of a NACA4412 wing at 11 degrees angle of attack show that steady suction and blowing can eliminate flow separation and raise lift by up to 11 percent, but the drag penalty means overall lift-to-drag ratio does not improve. The study provides detailed boundary-layer and spectral data for flow-control design, though its conclusion about periodic control is at odds with its own appendix.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The abstract's claim that periodic control 'neither enhanced separation delay nor improved efficiency' is contradicted by the paper's own Table A.4, where every periodic case reduces ℓsep from 0.14 to 0.05-0.06.","rationale":"The reader's verdict is CONDITIONAL, and I agree with that disposition, but my main concern differs from the reader's stated weakest assumption. The weakest assumption in the reader's review is statistical convergence; that is a real concern for the small L/D differences. However, the more load-bearing issue is the abstract's periodic-control claim, because it is a central claim and it is directly falsified by the authors' own appendix. Table A.4 reports ℓsep = 0.05-0.06 for all nine periodic cases versus 0.14 for the uncontrolled case, a consistent 57-64% separation-delay improvement. The periodic simulations used only ~1.6 flow-over times, but the uniformity of the result across very different frequencies and intensities makes an averaging artifact unlikely to reverse all nine cases. The efficiency part of the abstract claim is supported: no periodic case improves L/D. Therefore the paper's central conclusion about periodic control must be corrected, but the steady-control results (Cases A-E) are not undermined. This is why I recommend keeping CONDITIONAL rather than rejecting. I credit the paper's substantial LES dataset, AMR, and the consistent separation-delay and lift trends for steady control. The contradiction is in a summary sentence, not in the underlying data, so it can be fixed by rewording. I mark agreement as partial because the reader flagged the abstract-appendix discrepancy in the strongest_claim but selected statistical convergence as the weakest assumption.","tokens_in":25528,"tokens_out":4473,"duration_ms":39369,"concrete_test":"Compile the ℓsep values from Table A.4 and compute each periodic case's change relative to Ref. If, as reported, all nine cases show ℓsep < 0.14 (they are 0.05-0.06), the abstract's 'neither enhanced separation delay' is falsified under the natural uncontrolled baseline. To rule out the 1.6-flow-over-time averaging artifact, rerun one representative periodic case (e.g., Case 1, F*=1.0, ψ=0.1%) for the same 2 flow-over times used for Cases A-E and recompute ℓsep; if it remains below 0.14, the contradiction stands.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The paper's central claim includes the abstract statement that periodic control 'neither enhanced separation delay nor improved efficiency.' This is contradicted by the authors' own appendix data. In Table A.4, all nine periodic-control cases reduce the separation length from ℓsep = 0.14 in the uncontrolled case to ℓsep = 0.05 or 0.06, i.e., a 57-64% reduction. Thus, relative to the uncontrolled reference, periodic control clearly delays separation. The efficiency clause is supported (every periodic case has L/D below 24.88), so the false part is specifically the separation-delay assertion. The conclusion's wording ('neither further delayed separation') is narrower and can be read as comparing to the steady cases; the abstract's 'neither enhanced separation delay' cannot be read that way without an explicit qualifier. Because this sentence is in the abstract and is repeated in the summary, it is a load-bearing part of the claimed findings, not a typo isolated to one table. The fix is localized but mandatory: amend the abstract/summary to state that periodic control delayed separation but did not improve L/D, or specify the comparison baseline.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper presents wall-resolved large-eddy simulations of a NACA4412 wing section at AoA = 11° and Re_c = 200,000, with spanwise width Lz = 0.6c, using the spectral-element solver Nek5000 with AMR and roughly 3.9×10^8 grid points. The authors compare five steady-control configurations (uniform suction on the suction side, uniform blowing on the pressure side, and combinations at three intensities) and, in an appendix, nine periodic-control cases. The reported findings are that combined suction and blowing can eliminate separation (ℓsep = 0 for Cases B and C), increase lift by up to 11.45%, but generally increase drag so that only pressure-side blowing (Case E) modestly improves L/D. The abstract further claims that periodic control 'neither enhanced separation delay nor improved efficiency.' The manuscript includes analysis of boundary-layer integral quantities, Reynolds stresses, and power-spectral densities.","tokens_in":25732,"tokens_out":6431,"duration_ms":54096,"significance":"If the results are robust, the study is a valuable contribution: it provides a high-resolution LES dataset for a separated wing section with a wide span, it demonstrates that a RANS-optimized control configuration (Case A) does not improve L/D in LES, and it documents nontrivial modifications of TBL statistics and spectra under control. The explicit comparison of steady versus periodic control at a single AoA is also useful. However, the central periodic-control claim is directly contradicted by the paper's own appendix table, and the aerodynamic-efficiency conclusions rest on force averages over only about two flow-over times without uncertainty estimates. Both issues are load-bearing and require correction before the findings can be accepted as stated. The novelty claim of being the first high-resolution LES study of separation control is too broad and should be softened.","major_comments":[{"comment":"The abstract states that periodic control 'neither enhanced separation delay nor improved efficiency.' This is contradicted by the manuscript's own Table A.4: every one of the nine periodic-control cases reduces ℓsep from 0.14 to 0.04–0.06, a 57–71% reduction relative to the uncontrolled reference. The summary's wording ('neither further delayed separation') is ambiguous but the abstract's claim is false if the comparison baseline is the uncontrolled case. The authors must revise the abstract and summary to say either that periodic control delayed separation but did not improve L/D, or explicitly state that periodic control did not outperform the steady cases. This is not a cosmetic issue, as the sentence is a headline finding.","section":"Abstract; §4 (Summary and conclusions); Appendix A, Table A.4"},{"comment":"The aerodynamic-efficiency conclusions lack uncertainty quantification. Section 2.2 states that 'for each case, simulations were run for at least 2 flow-over times' and that this is 'equivalent to ≈12 flow-over times' for a domain with Lz=0.1c, but no derivation or convergence diagnostics are provided. Table 2 reports L/D changes as small as −2.06% (Case A) and +0.79% (Case E). With only two flow-over times of averaging in a separated flow with low-frequency unsteadiness, these differences may be within the statistical uncertainty of the force coefficients. The authors should provide running-time averages or block-averaged uncertainties for Cl, Cd, and L/D for the main cases, and should justify the spanwise-width/time equivalence. The near-identical values reported for Cases 3 and 4 in Table A.4, which differ only in frequency, further suggest that the averaging period is too short to distinguish configurations.","section":"§2.2 (statistical convergence) and Table 2"},{"comment":"There is an internal inconsistency in the discussion of the local force distributions. The text states: 'as the only case that improves aerodynamic efficiency, Case D shows a clear reduction in Γd, particularly by decreasing Γd,p in the region x/c = 0.1 to 0.3.' This contradicts Table 2, which shows Case D reduces L/D by 4.21% and Case E is the only case that improves L/D (+0.79%). In addition, the preceding paragraph attributes a '1.1% increase in Cd,p' to Case D, whereas Table 2 lists Cd,p = +3.57% for Case D and +1.10% for Case E. These errors indicate a systematic mislabeling of Cases D and E in this section; the local-force analysis and its interpretation must be corrected.","section":"§3.1 and Fig. 5"}],"minor_comments":[{"comment":"The spanwise width is written as 'Lz = 0.6' without units; it should be 'Lz = 0.6c'.","section":"Abstract"},{"comment":"The caption contains typos: 'Darg-force distributions' and 'cricle and dimonand markers' should be 'drag-force distributions' and 'circle and diamond markers'.","section":"Fig. 5 caption"},{"comment":"The phrase 'flavorable-pressure-gradient conditions' should be 'favorable-pressure-gradient conditions'.","section":"§3.2"},{"comment":"The word 'siginificantly' should be 'significantly'.","section":"§3.4"},{"comment":"The heading 'Competing interesting' should be 'Competing interests'.","section":"Declarations"},{"comment":"The reference to Prandtl contains an encoding artifact ('Uø ber'); it should be 'Über'.","section":"References"},{"comment":"The claim 'to the authors' best knowledge, this is the first numerical study utilizing high-resolution LESs to provide comprehensive assessments on separation control' is overly broad; earlier LES-based active-flow-control studies exist. Recommend making the novelty claim more specific, e.g., first high-resolution LES assessment of these steady/periodic uniform blowing and suction configurations on a wing section at this angle of attack.","section":"Abstract and §1"}],"recommendation":"major_revision","confidential_remarks":"The paper has a solid computational core and the steady-control results are plausible, but the abstract's periodic-control claim is demonstrably contradicted by the appendix data, and the L/D conclusions rest on very short averaging without error bars. The Case D/Case E mislabeling in §3.1 is a clear textual error that also affects the interpretation of the local-force analysis. These are fixable in revision, but the revisions are substantive rather than cosmetic. I would also recommend the editor ask the authors to provide convergence evidence (e.g., time histories of Cl and Cd) and to moderate the 'first' novelty claim."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The paper delivers a genuinely new dataset: high-resolution LES of a NACA4412 at AoA=11 deg, Re=200k, spanwise width 0.6c, comparing combined blowing/suction, individual controls, and periodic excitation, with boundary-layer statistics and spectra. The steady-control results are plausible and internally consistent: separation is reduced or eliminated, lift rises with control intensity, drag rises, and L/D mostly falls. Case E (blowing on pressure side only) is the exception, improving L/D marginally. That is the paper's core contribution and I think it holds.\n\nThe soft spots are real, though localized. The abstract's claim that periodic control 'neither enhanced separation delay nor improved efficiency' is contradicted by the authors' own Table A.4: every periodic case cuts separation length from 0.14 to 0.05-0.06, a 57-64% reduction. The conclusion's phrase 'neither further delayed separation' can be read as comparison to the steady cases, but the abstract cannot. That needs fixing before publication. There is also a clear slip in Sec. 3.1: the text says 'as the only case that improves aerodynamic efficiency, Case D...' but Table 2 shows Case E is the only one with positive L/D change. Minor typo, but confusing.\n\nThe bigger statistical concern: force averages are based on at least two flow-over times, which the authors argue is equivalent to about twelve flow-over times on a narrower span. But no error bars or convergence diagnostics are shown, and the L/D differences the paper discusses are small (0.8% to 11.5%). Without uncertainty estimates, the ranking of configurations, especially whether Case E really improves efficiency, is not firmly established. The authors should either provide convergence data or soften the claims. Data availability is also promised only after publication.\n\nNone of this destroys the central steady-control finding. The paper is a serious piece of simulation work with a useful dataset, and it deserves a proper review. The referee should insist on the abstract correction, the typo, and some kind of statistical support for the efficiency comparisons. I'd take it to the next reading group.","headline":"Good LES dataset with a real abstract contradiction and thin statistical basis for the efficiency ranking.","tokens_in":26332,"tokens_out":2672,"would_cite":true,"duration_ms":24101,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"At 11 degrees angle of attack, steady suction on the suction side of a NACA4412 wing can eliminate flow separation, raising lift by up to 11 percent, but the accompanying drag increase reduces the lift-to-drag ratio; only pressure-side…","keywords":["flow separation control","NACA4412 wing","large-eddy simulation","uniform blowing and suction","periodic excitation","lift-to-drag ratio","turbulent boundary layer","adverse pressure gradient"],"falsifier":"Repeat the five control cases and the uncontrolled baseline in the same setup but average over at least ten flow-over times, or run several independent realizations and compute confidence intervals for Cl, Cd, and L/D; if the reported differences of +0.79 to -11.5 percent in lift-to-drag ratio fall inside the uncertainty band, or the rank ordering changes, the conclusion that only pressure-side blowing improves efficiency would be overturned.","tokens_in":25295,"feed_emoji":"🛫","tokens_out":6430,"duration_ms":55344,"temperature":0.7,"pith_summary":"This paper asks whether steady or periodic wall blowing and suction can reattach the separated flow over a NACA4412 wing at 11 degrees angle of attack and chord Reynolds number 200,000, and at what cost in lift and drag. Using wall-resolved large-eddy simulations with a wide spanwise domain, it finds that steady suction on the suction side combined with blowing on the pressure side can shrink or eliminate the separated region and raise lift by up to about 11 percent, but the accompanying drag rise lowers the lift-to-drag ratio. Suction alone gives the same separation delay as the combined case with a smaller lift gain, while pressure-side blowing alone is the only tested steady case that slightly improves aerodynamic efficiency. The authors conclude that periodic control, tested near the trailing-edge separation point, neither further delays separation nor improves efficiency. The study matters because it tests control ideas in three-dimensional resolved turbulence rather than two-dimensional models, and it separates the lift benefit from the efficiency penalty.","feed_headline":"Steady suction reattaches wing flow but costs efficiency","feed_subtitle":"High-fidelity simulation at 11 degrees finds lift up 11 percent while lift-to-drag ratio drops.","key_machinery":"The load-bearing machinery is a wall-resolved large-eddy simulation of a NACA4412 section with spanwise width L_z = 0.6c, discretized with a spectral-element method and adaptive mesh refinement, resolving about 90 percent of the dissipation. Control is injected as a steady wall-normal velocity psi between 0.25 and 1.0 percent of U_infinity over 0.25 <= x/c <= 0.86 on the suction and/or pressure side, with its momentum input quantified by the momentum coefficient C_mu = rho $psi^{2}$ ell_ctrl / (0.5 rho $U_infinity^{2}$). The primary outcome variable is the separation length ell_sep = (x_TE - x_sep)/c, computed from the sign change of the skin-friction coefficient. The argument works by showing that C_mu scales the separation delay and lift rise, while decomposition of total drag into skin-friction and pressure components explains why aerodynamic efficiency falls even when lift rises.","core_discovery":"At AoA = 11 degrees and Re_c = 200,000, steady uniform suction applied to the suction side from x/c = 0.25 to 0.86 is the dominant mechanism for separation delay: suction alone at psi = 0.25 percent U_infinity reduces the separation length from 0.14 to 0.02, matching the combined suction-and-blowing case, and combined control at psi = 0.50 and 1.00 percent U_infinity eliminates separation entirely. Lift tracks the momentum coefficient, rising by 3.2, 5.9, and 11.45 percent for Cases A, B, and C, but total drag rises by 5.4, 11.2, and 25.9 percent, so the lift-to-drag ratio falls by 2.1, 4.7, and 11.5 percent. Only pressure-side blowing alone (Case E) improves aerodynamic efficiency, by 0.79 percent, through a 12.5 percent reduction in skin-friction drag. The paper interprets this as a structural trade-off: control adds near-wall momentum to delay separation and increase lift, and that same momentum addition costs drag. It also reports that periodic control at several frequencies and momentum coefficients did not beat the steady configurations in separation delay or efficiency.","pith_inferences":["A natural extension, suggested by the paper's own discussion, is that at higher angles of attack where the separated region is much longer, the same steady suction may shift the balance and yield a net gain in lift-to-drag ratio; the paper does not test that regime.","Because suction alone matches the combined case in separation delay, optimizing the suction distribution or moving it farther upstream may recover part of the drag penalty; this is a testable extension of the reported data.","The abstract's statement that periodic control 'neither enhanced separation delay nor improved efficiency' should be read as no improvement over the steady combined cases: the appendix tables show every periodic case does reduce the separation length relative to the uncontrolled wing, and the small efficiency differences would need longer averaging to rank reliably.","A closed-loop or learning-based controller that modulates suction and blowing in time, rather than using steady or single-frequency forcing, is a plausible route to recovering some efficiency while keeping the separation delay; the paper names this direction but does not pursue it."],"forward_implications":["If the force results hold, steady suction over the suction side from quarter-chord to near the trailing edge can restore attached flow on a NACA4412 at AoA = 11 degrees and Re_c = 200,000.","Combined suction and blowing provides a controllable lift increase of up to about 11 percent, but at psi = 1.0 percent U_infinity the drag rise costs roughly 11.5 percent of the lift-to-drag ratio.","Pressure-side blowing alone is the only tested steady case with a positive, though marginal, efficiency change, obtained by lowering skin-friction drag rather than pressure drag.","Suction on the suction side dominates separation delay, which suggests that control placement and momentum budget matter more than the total momentum coefficient alone.","The conclusions are specific to one airfoil, one angle of attack, and one Reynolds number; control strategies that work for attached boundary layers at lower angles of attack do not transfer directly to separated conditions."],"supporting_citations":[{"why":"Supplies the spectral-element solver used to run the large-eddy simulations.","marker":"Fischer et al. (2008)"},{"why":"Provides the implicit relaxation-filter subgrid-scale model that accounts for unresolved dissipation and was validated in wing simulations.","marker":"Negi et al. (2018)"},{"why":"Supplies the volume-forcing tripping method and parameters used at x/c = 0.1 on both sides of the wing.","marker":"Hosseini et al. (2016)"},{"why":"Identifies the optimized suction-and-blowing configuration replicated here as Case A, originally obtained with two-dimensional RANS at higher Reynolds number.","marker":"Mallor et al. (2024a)"},{"why":"Provides the lower-angle-of-attack uniform blowing and suction reference data and the control-area choice used for comparison.","marker":"Atzori et al. (2020)"},{"why":"Validates the simulation approach for wing sections and supports the resolution and spanwise-width criteria used in the present setup.","marker":"Vinuesa et al. (2018)"},{"why":"Documents the high-angle-of-attack database that establishes the uncontrolled separation length and the requirement of at least L_z = 0.4c spanwise width.","marker":"Mallor et al. (2024b)"},{"why":"Provides the periodic-excitation mechanism and the optimal-frequency reasoning used to choose the periodic control frequencies.","marker":"Greenblatt and Wygnanski (2000)"},{"why":"Supplies the vortex-shedding frequency behavior used to interpret the wake power-spectral-density results.","marker":"Yarusevych et al. (2009)"}],"fun_headline_variants":["Suction delays wing separation but cuts efficiency","Steady suction boosts lift 11% but drag rises 26%","Periodic control fails to match steady suction","Wing suction: separation delayed, efficiency lost"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The central claim rests on treating lift and drag coefficients averaged over about two flow-over times as statistically converged, without error bars, so the small efficiency differences between configurations are taken as real.","fun_headline_variants_meta":{"raw":{"variants":["Suction delays wing separation but cuts efficiency","Steady suction boosts lift 11% but drag rises 26%","Periodic control fails to match steady suction","Wing suction: separation delayed, efficiency lost"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000334,"raw_usage":{"total_tokens":1918,"prompt_tokens":1072,"completion_tokens":846,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":688,"completion_tokens_details":{"reasoning_tokens":784}},"tokens_in":688,"tokens_out":846,"duration_ms":8929,"temperature":1.0,"reasoning_tokens":784,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-08T11:27:08.063479+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Repeat the five control cases and the uncontrolled baseline in the same setup but average over at least ten flow-over times, or run several independent realizations and compute confidence intervals for Cl, Cd, and L/D; if the reported differences of +0.79 to -11.5 percent in lift-to-drag ratio fall inside the uncertainty band, or the rank ordering changes, the conclusion that only pressure-side blowing improves efficiency would be overturned.","supporting_citations":[],"review_version":1}