Aerospace Science and Technology最新文献

{"title":"Multispectral tomographic absorption spectroscopy for a supersonic rocket engine model using region-specific absorption line selection and deep learning correction","authors":"Dingfeng Shi ,&nbsp;Zhenyu Xu ,&nbsp;Cheng Du ,&nbsp;Zhouyang Cong ,&nbsp;Zhongqiang Yu ,&nbsp;Weiqi Xing ,&nbsp;Wei Nie ,&nbsp;Mingdi Bao ,&nbsp;Kin-Pang Cheong ,&nbsp;An Huang ,&nbsp;Chaozong Wang ,&nbsp;Ruifeng Kan","doi":"10.1016/j.ast.2026.111781","DOIUrl":"10.1016/j.ast.2026.111781","url":null,"abstract":"<div><div>The tomographic measurement of combustion flow fields in aerospace propulsion systems remains a significant challenge due to the large temperature variations, steep spatial gradients, and the ill-posed nature of inversion problem under noisy and sparse detection conditions. While multispectral information provides an extended temperature sensitivity, inversion errors can weaken the self-consistency among multiple absorption lines and amplify the reconstruction uncertainties. To address these challenges, a highly precise linear multispectral tomographic absorption spectroscopy was developed for the reconstruction of non-uniform combustion fields over a wide temperature range, extending from room temperature to high temperature. A three-step temperature reconstruction strategy was proposed, in which four absorption lines were selectively chosen to ensure optimal sensitivity and accuracy across different temperature regions. In addition, a deep neural network was developed to identify local reconstruction distortions and provide a correct line selection. Numerical simulations on synthetic phantoms and fundamental experiments on a Mckenna burner demonstrated significant improvement in reconstruction accuracy compared with conventional fixed absorption line combination methods. Finally, the proposed approach was applied on a rocket engine model, performing tomographic measurements at ten cross-sectional planes downstream of the supersonic jet flame. The reconstructed quasi-3D temperature distributions captured the transition of the jet flame from supersonic to subsonic state. This study establishes a reliable framework for wide-range temperature tomography and confirm the robustness and applicability in extreme aerospace combustion diagnostics.</div></div>","PeriodicalId":50955,"journal":{"name":"Aerospace Science and Technology","volume":"173 ","pages":"Article 111781"},"PeriodicalIF":5.8,"publicationDate":"2026-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146072615","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"ATS-Kriging: An adaptive two-stage Kriging sequential modeling for performance prediction of aero-engine structure","authors":"Cheng Yan ,&nbsp;Ben Pei ,&nbsp;Kai-Cheng Liu ,&nbsp;Lan Xiang ,&nbsp;Yu-Chen Liao ,&nbsp;Shu-Qing Tian ,&nbsp;Da Teng","doi":"10.1016/j.ast.2026.111798","DOIUrl":"10.1016/j.ast.2026.111798","url":null,"abstract":"<div><div>To address the challenges of limited prediction accuracy and low modeling efficiency of surrogate models in high-dimensional nonlinear design spaces, this study proposes an adaptive two-stage Kriging sequential modeling method (ATS-Kriging). Based on the two-stage Kriging (TS-Kriging) framework, the proposed method integrates an adaptive sampling strategy to significantly enhance modeling precision. The first stage of TS-Kriging assigns differential weight coefficients to individual sample points based on an analysis of their prediction errors. The second stage mitigates the adverse influence of anomalous samples by reducing the weight assigned to those with high errors. Furthermore, an adaptive sampling strategy is introduced to augment this framework. This strategy dynamically adjusts the weight factors of the sampling function to strike an optimal balance between local exploitation and global exploration. This facilitates a more comprehensive capture of the design space's key features and promotes intensified sampling in regions with high potential, thereby improving the surrogate model's fit in critical areas.​ To validate the modeling performance of ATS-Kriging, two classical numerical examples and two typical aero-engine engineering cases are employed for comparative analysis in terms of modeling accuracy and efficiency. The results show that ATS-Kriging exhibits superior overall modeling performance in both types of tests. In the numerical examples, compared to other methods, the prediction root mean square error of ATS-Kriging is reduced by at least approximately 9.6%, and the number of samples required to achieve the same accuracy is reduced by at least approximately 3.7%. In the engineering cases, its prediction error is reduced by at least approximately 4.2%, and modeling efficiency is improved by at least approximately 6%. The results indicate that ATS-Kriging can significantly enhance both modeling accuracy and efficiency. This method provides a new viable approach for efficient and high-accuracy approximate modeling in high-dimensional nonlinear optimization problems.</div></div>","PeriodicalId":50955,"journal":{"name":"Aerospace Science and Technology","volume":"173 ","pages":"Article 111798"},"PeriodicalIF":5.8,"publicationDate":"2026-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146071752","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"High-fidelity three-dimensional aerodynamic flow prediction on wings with physics-constrained dual-parallel attention UNet++","authors":"Rongfeng Cui (崔榕峰) ,&nbsp;Qiao Zhang (张巧) ,&nbsp;Weiwei Zhang (张伟伟) ,&nbsp;Wenbo Lu (鲁文博) ,&nbsp;Liangjie Gao (高亮杰)","doi":"10.1016/j.ast.2026.111846","DOIUrl":"10.1016/j.ast.2026.111846","url":null,"abstract":"<div><div>Accurate flow field data provide a robust foundation for analyzing concentrated force distribution and implementing flow control strategies. Nevertheless, current deep neural network methods exhibit limitations in accuracy when applied to reconstruct three-dimensional wing flow fields. To address this challenge, we propose an intelligent flow field reconstruction technique termed physics-constrained Dual-Parallel Attention UNet++ (DPAtt-UNet++). This method utilizes the Unet++ neural network architecture as its backbone, integrating a dual-parallel attention mechanism and nested network structure. Furthermore, a physics-constrained hierarchical loss function is introduced, incorporating the residuals of the governing Navier-Stokes equations as soft constraints to enforce physical consistency during training. Comprehensive evaluations demonstrate that the proposed DPAtt-UNet++ outperforms not only the baseline U-Net by approximately 10% in reconstruction accuracy, but also shows clear improvements over both standard UNet++ and a non-physics-constrained DPAtt-UNet++, validating the effectiveness of the integrated attention mechanism and physical constraints. Tests on wings constructed from different airfoil profiles confirm robust generalization capability across varying flow conditions and geometric shapes. Moreover, the method achieves approximately 2–3 orders of magnitude faster reconstruction speed compared to the Computational Fluid Dynamics (CFD) method in the online prediction phase. These results demonstrate the method can accurately and efficiently reconstruct flow fields for different geometries under various flow conditions.</div></div>","PeriodicalId":50955,"journal":{"name":"Aerospace Science and Technology","volume":"173 ","pages":"Article 111846"},"PeriodicalIF":5.8,"publicationDate":"2026-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146109798","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Off-design adjustment strategies and internal flow mechanisms of novel locally adjustable-thickness turbines in variable cycle engine system","authors":"Mai Li ,&nbsp;Jun Liu ,&nbsp;Wenying Ju ,&nbsp;Hang Yuan ,&nbsp;Pei Wang ,&nbsp;Pengfei Wang ,&nbsp;Xingen Lu","doi":"10.1016/j.ast.2026.111840","DOIUrl":"10.1016/j.ast.2026.111840","url":null,"abstract":"<div><div>Variable Geometry Turbines (VGTs) serve as core components of variable cycle engines. Conventional VGTs employ endwall-mounted adjustment mechanisms to achieve variable vane angles for off-design operation. However, under significant off-design conditions, strong separation induced by high-incidence inflow and enhanced endwall leakage flows due to excessive clearances lead to drastic loss increases. This study innovatively proposes a novel turbine blade adjustment scheme enabling localized thickness and throat area modulation through partial suction-side rotation. Using the high-load turbine cascade IET-LPTA as the research object, high-fidelity numerical methods validated by experimental data were employed. These methods were used to systematically investigate internal flow mechanisms under varying suction-side rotation angles (<em>θ</em>) and pivot positions. Detailed analyses of boundary layer, wake, and near-wall parameter variations were conducted, with flow regulation effectiveness and loss characteristics summarized. Results demonstrate that increased <em>θ</em> enhances blade loading, advances boundary layer separation-transition-reattachment events, enlarges separation bubbles, and significantly increases wake and turbulent dissipation losses. Pivot position critically influences separation bubble structure and location. Forward pivots (e.g., 0.6<em>C_x</em>) induce earlier separation and reorganize near-wall low-energy fluid into ‘separation-reattachment-secondary separation-transition-secondary reattachment’ patterns, forming bimodal bubble structures with reduced losses. Both <em>θ</em> and pivot adjustments enable flow regulation, exhibiting nonlinear coupled effects on losses. For flow variations ≤1.5%, <em>θ</em> adjustment optimizes loss performance; beyond this threshold, pivot adjustment proves superior. Mechanistically, the flow‑capacity variation is dominated by the effective throat‑area change, while the separation characteristics are governed by the redistribution of suction‑side adverse pressure gradient and the resulting boundary‑layer response. This research provides theoretical foundations and design guidelines for high-performance VGT development.</div></div>","PeriodicalId":50955,"journal":{"name":"Aerospace Science and Technology","volume":"173 ","pages":"Article 111840"},"PeriodicalIF":5.8,"publicationDate":"2026-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146109797","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A general digital twin modeling framework for full-field structural strength assessment and prediction","authors":"Kuo Tian ,&nbsp;Ziyu Xu ,&nbsp;Zhiyong Sun ,&nbsp;Xuanwei Hu ,&nbsp;Peng Zhang ,&nbsp;Zhiyong Zhou","doi":"10.1016/j.ast.2026.111766","DOIUrl":"10.1016/j.ast.2026.111766","url":null,"abstract":"<div><div>A general digital twin modeling framework that can achieve full-field structural strength assessment and prediction is proposed in this paper, which mainly comprises the physical layer, the simulation layer, the data layer, the model layer, and the service layer. In the framework, two key technologies are proposed. Firstly, to standardize the multi-source heterogeneous simulation data and sensor data, the load-response-coordinate data association model is created, laying the foundation for digital twin modeling. Secondly, the hierarchical multi-source data fusion method is proposed to achieve the fusion of simulation data and sensor data, thereby establishing a digital twin model and realizing high overall precision as well as accurate interpolation at sensor local positions. To demonstrate the effectiveness of the proposed digital twin modeling framework, two experimental studies on the open-hole plate and the hierarchical stiffened plate are carried out. Results indicate that the proposed framework can establish a high-precision digital twin model that ensures real-time monitoring and obtains better prediction and decision-making for the loading control, and provides an intelligent solution for full-field structural strength assessment and prediction.</div></div>","PeriodicalId":50955,"journal":{"name":"Aerospace Science and Technology","volume":"173 ","pages":"Article 111766"},"PeriodicalIF":5.8,"publicationDate":"2026-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146047961","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Dynamics of vibration mode competition between flutter and forced vibration in thin-walled scramjet isolators under fluctuating backpressure","authors":"Ruoshuai Zhao ,&nbsp;Xianzong Meng ,&nbsp;Zebin Zhang ,&nbsp;Zhengyin Ye","doi":"10.1016/j.ast.2026.111747","DOIUrl":"10.1016/j.ast.2026.111747","url":null,"abstract":"<div><div>Backpressure fluctuations in thin-walled scramjet isolators can induce vibration mode competition between the intrinsic flutter of elastic panels and forced vibration, yet the underlying dynamics of this interaction remain insufficiently understood. This study employed a two-way coupled CFD/CSD framework to investigate the interplay and competition between the panel’s intrinsic flutter and the forced vibration induced by periodic backpressure fluctuations over a broad range of fluctuation frequencies (<span><math><msub><mi>f</mi><mi>b</mi></msub></math></span>) and amplitudes (<span><math><mrow><msub><mi>A</mi><mi>b</mi></msub><mrow><mo>)</mo></mrow></mrow></math></span>. Under steady backpressure, the elastic panel exhibits flutter-dominated limit cycle oscillations. When fluctuating backpressure is introduced, periodic excitation superimposes on the intrinsic flutter, giving rise to four distinct dynamic regimes: (i) flutter-dominated regime at low <span><math><msub><mi>f</mi><mi>b</mi></msub></math></span>; (ii) forced-vibration regime at mid <span><math><msub><mi>f</mi><mi>b</mi></msub></math></span> with resonance near panel’s natural frequency; (iii) a transitional regime with a coupled frequency; and (iv) flutter re-emergence at high <span><math><msub><mi>f</mi><mi>b</mi></msub></math></span>. At higher <span><math><msub><mi>A</mi><mi>b</mi></msub></math></span>, only the flutter-dominated and forced-vibration regimes occur, separated by clear <span><math><msub><mi>f</mi><mi>b</mi></msub></math></span> boundaries. With increasing <span><math><msub><mi>f</mi><mi>b</mi></msub></math></span>, the panel starts in flutter dominance with near constant oscillation amplitudes, transitions to forced vibration near resonance with rapid amplitude growth, and returns to flutter dominance beyond resonance. Reducing <span><math><msub><mi>A</mi><mi>b</mi></msub></math></span> lowers the direct energy input from backpressure fluctuations and promotes the emergence of the transitional regime, which occurs more frequently and persist longer, blurring the regime boundaries between flutter and forced vibration. As a result, vibration mode competition fundamentally governs the structural response, which in turn controls shock-train motion, separation behavior, and overall isolator performance. Regime transitions cause abrupt changes in mean values and oscillation amplitudes of these flow features, with resonance exerting the strongest influence by driving extrema in both metrics. Away from resonance, flutter- or transitional-dominated regimes exhibit weaker sensitivity to <span><math><msub><mi>f</mi><mi>b</mi></msub></math></span>, although large <span><math><msub><mi>A</mi><mi>b</mi></msub></math></span> can still generate significant unsteadiness.</div></div>","PeriodicalId":50955,"journal":{"name":"Aerospace Science and Technology","volume":"173 ","pages":"Article 111747"},"PeriodicalIF":5.8,"publicationDate":"2026-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146014252","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Impact localization in aircraft composite structures using GWO-VMD and adaptive correlation thresholding","authors":"Minghua Wang ,&nbsp;Yi Zhang ,&nbsp;Di Wu ,&nbsp;Yue Wang ,&nbsp;Xinlin Qing ,&nbsp;Yishou Wang","doi":"10.1016/j.ast.2026.111777","DOIUrl":"10.1016/j.ast.2026.111777","url":null,"abstract":"<div><div>Impact localization based on reference databases (RD) is widely used for its simplicity and low dependence on structural modeling. However, it is difficult to obtain reliable impact signal features under nonlinear wave propagation effects in complex aircraft structures. To overcome these limitations, this paper proposes an impact localization method integrating Grey Wolf Optimizer (GWO) and Variational Mode Decomposition (VMD) for feature construction, combined with an adaptive correlation threshold correction strategy. A GWO-VMD adaptive signal decomposition model is developed to optimize VMD parameters, improving the representativeness of modal components for impact features. An energy-threshold-based mode selection mechanism automatically constructs a highly correlated reconstructed reference database, ensuring reliable mapping between features and impact responses. Additionally, an impact localization framework based on an information fusion matrix and dynamic correlation threshold correction is designed to suppress interference from redundant matching results. Experimental validation on a composite stiffened panel and a full-scale composite wing confirms the effectiveness of the proposed method.</div></div>","PeriodicalId":50955,"journal":{"name":"Aerospace Science and Technology","volume":"173 ","pages":"Article 111777"},"PeriodicalIF":5.8,"publicationDate":"2026-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146048184","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Exploring the feasibility of parachute recovery systems for catastrophic failures in passenger aircraft","authors":"Samandar Khan Afridi ,&nbsp;Shakoor Akhtar ,&nbsp;Talha Zafar Khan ,&nbsp;Mohsin Ali Koondhar ,&nbsp;Ibrahim Mahariq ,&nbsp;Ezzeddine Touti","doi":"10.1016/j.ast.2026.111866","DOIUrl":"10.1016/j.ast.2026.111866","url":null,"abstract":"<div><div>This paper investigates the feasibility and effectiveness of a Parachute Recovery System (PRS) as an advanced airframe safety enhancement for commercial aircraft, focusing on the Boeing 777. A mixed-methods approach integrates analytical, order-of-magnitude aerodynamic modeling, expert qualitative assessments, and real-world case studies. Simulation results indicate a generated drag force of approximately 26.4 kN, demonstrating the PRS capability to achieve limited drag contribution under idealized descent assumptions during in-flight emergencies. Evidence from Cirrus SR20 and SR22 aircraft further validates PRS performance, achieving safe, low-impact landings with high survivability rates. However, large-scale implementation poses considerable engineering, financial, and regulatory challenges, including structural reinforcement, deployment reliability, and certification complexity. The cost-benefit analysis suggests that although initial and maintenance costs are significant, they may be offset by long-term safety improvements and reduced insurance liabilities. The study recommends the integration of lightweight composite materials, multi-stage deployment systems, comprehensive testing, and specialized pilot training, alongside collaboration among manufacturers, airlines, and regulatory authorities to streamline certification and operational adoption. The findings highlight the potential of PRS to advance next-generation aviation safety, enhance passenger survivability, and establish new benchmarks in commercial aircraft design and emergency recovery systems.</div></div>","PeriodicalId":50955,"journal":{"name":"Aerospace Science and Technology","volume":"173 ","pages":"Article 111866"},"PeriodicalIF":5.8,"publicationDate":"2026-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146134792","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Design and implementation of real-time dihedral angle control for enhanced flight stability of quadrotor UAV","authors":"Nihat Çabuk","doi":"10.1016/j.ast.2026.111844","DOIUrl":"10.1016/j.ast.2026.111844","url":null,"abstract":"<div><div>This study presents a real-time geometry-aware control strategy and its experimental validation for quadrotor UAVs equipped with actively adjustable dihedral angles. By integrating a cascaded PID control architecture with dynamic dihedral modulation, the system adapts its aerodynamic configuration during flight to enhance stability and responsiveness. Unlike conventional fixed-geometry multirotors, this platform enables geometric tuning in flight via a centralized actuation mechanism. Nonlinear simulations and autonomous flight tests were conducted for five different dihedral configurations (<span><math><mrow><mi>γ</mi><mo>=</mo><mo>−</mo><msup><mn>7</mn><mo>∘</mo></msup><mo>,</mo><mo>−</mo><mn>3</mn><mo>.</mo><msup><mn>5</mn><mo>∘</mo></msup><mo>,</mo><msup><mn>0</mn><mo>∘</mo></msup><mo>,</mo><mo>+</mo><mn>3</mn><mo>.</mo><msup><mn>5</mn><mo>∘</mo></msup><mo>,</mo><mo>+</mo><msup><mn>7</mn><mo>∘</mo></msup></mrow></math></span>) under identical flight scenarios, including takeoff, hover and landing. Performance metrics such as altitude accuracy, attitude stability (roll, pitch, yaw), and structural vibration levels were analyzed. The findings validate the feasibility and effectiveness of geometry-aware control for multirotor systems. In addition, this work introduces a novel class of UAVs capable of real-time structural reconfiguration, enabling adaptation to changing flight conditions, payload variations, or mission profiles.</div></div>","PeriodicalId":50955,"journal":{"name":"Aerospace Science and Technology","volume":"173 ","pages":"Article 111844"},"PeriodicalIF":5.8,"publicationDate":"2026-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146134808","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Correlation between aerodynamic forces and vortex dynamics of a NACA0012 wing section in compressible dynamic stall via IDDES","authors":"Dechuan Ma,&nbsp;Gaohua Li,&nbsp;Jiahao Liu,&nbsp;Can Liu,&nbsp;Fuxin Wang","doi":"10.1016/j.ast.2026.111843","DOIUrl":"10.1016/j.ast.2026.111843","url":null,"abstract":"<div><div>Dynamic stall on helicopter retreating blades involves complex flow phenomena, particularly with the emergence of local supersonic regions and shock waves. This study investigates the force generation mechanisms of a pitching NACA0012 wing section in dynamic stall using improved delayed detached eddy simulations (IDDES). At a moderate Reynolds number of <span><math><mrow><mi>R</mi><msub><mi>e</mi><mi>c</mi></msub><mo>=</mo><mn>500</mn><mo>,</mo><mn>000</mn></mrow></math></span> and reduced frequencies <span><math><mrow><mi>k</mi><mo>=</mo><mn>0.15</mn></mrow></math></span> and 0.25, the compressibility effects are examined by varying the freestream Mach number (<span><math><mrow><msub><mi>M</mi><mi>∞</mi></msub><mo>=</mo><mn>0.1</mn></mrow></math></span>, 0.3, and 0.5). An extended force partitioning method (E-FPM) is proposed to establish a direct linkage between flow fields and aerodynamic forces in compressible flows. In all cases, the majority of force production is attributed to the second Galilean invariant of the velocity gradient tensor, while the remainder arises from nonzero velocity divergence and density fluctuations due to compressibility. Prior to stall onset, leading-edge suction dominates lift and drag production, and turbulent separation vortices (TSVs) also have a positive contribution. As <em>M</em>_∞ increases, the leading-edge stagnation point moves upstream. The insufficient flow acceleration reduces fluid stretching and strain around the high-curvature leading edge, causing a loss in lift when <em>M</em>_∞ reaches 0.5. Upon stall onset, the dynamic stall vortex (DSV) becomes the main force contributor. At higher <em>M</em>_∞, the DSV forms earlier due to advanced stall onset, which leads to earlier drag divergence and increased drag. However, the DSV also sheds earlier and weakens with enhanced compressibility. The reduced vorticity and increased density fluctuations within the vortex core region of the DSV result in lower peak lift and drag. With the DSV shedding, its positive contribution from the vortex core region diminishes without vorticity feed. The negative contribution from the vortex-induced stretching and strain becomes dominant and leads to lift stall. This work provides new insights into compressible dynamic stall physics and demonstrates the E-FPM’s effectiveness in identifying the physical origins of aerodynamic forces in such compressible, vortex-dominated flows.</div></div>","PeriodicalId":50955,"journal":{"name":"Aerospace Science and Technology","volume":"173 ","pages":"Article 111843"},"PeriodicalIF":5.8,"publicationDate":"2026-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146134810","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}