Stall and surge represent two prevalent forms of aerodynamic instability in engine.Stall is characterized by the circumferential propagation of rotating stall cells,while surge manifests as large-amplitude axial oscillations in pres-sure and mass flow.Failure to promptly mitigate such instabilities once initiated can lead to a rapid degradation in en-gine performance and,in severe cases,catastrophic mechanical damage.Current stall/surge detection or prevention systems deployed on component test facilities typically rely on feature vectors extracted from time-and frequency-domain signal analyses for instability detection.However,these conventional approaches often exhibit limited respon-siveness and insufficient sensitivity to incipient stall events.In light of recent advances in active engine control method-ologies,there is a growing need for high-fidelity,real-time monitoring of the initial perturbation waves associated with aerodynamic instability.This demand necessitates stall detection algorithms with enhanced speed and accuracy in identifying the onset of instability.To address this challenge,this paper presents a stall identification algorithm based on cross-correlation of power spectral densities.By computing the correlation coefficient between signals acquired from two spatially separated sensor channels,the proposed method enables timely detection of stall or surge inception and triggers an early warning.The algorithm effectively advances the lead time for stall prediction,thereby establishing a critical technical foundation for the implementation of active instability control strategies in engine.
To address the insufficient exploration of subtask coupling relationships and limited dynamic adaptability in existing carrier-based aircraft support operation scheduling research,this study investigates a multi-stage scheduling problem for carrier-based aircraft support operations.Firstly,by modeling both support station allocation and aircraft servicing sequence determination as a multi-agent Markov decision process,this paper establishes a mathematical characterization of the sequential coupling relationships between subtasks in support operation scheduling.Subse-quently,an independent Deep Q-Network(DQN)based multi-agent collaborative decision-making framework is pro-posed,incorporating a distributed training-execution mechanism that specially includes a support station allocation module,an aircraft servicing sequence decision module,and a multi-agent collaborative scheduling module.Further-more,a collaborative scheduling algorithm based on the multi-stage sequential decision-making mechanism is devel-oped to solve the proposed model.Finally,simulation results demonstrate that the proposed algorithm achieves a 27.08%and 14.19%improvement in average reward,and a 56.44%and 45.43%improvement in reward standard deviation,over the Dueling DQN and N-step DQN methods,respectively,verifying the effectiveness of the multi-stage collaborative decision-making mechanism in addressing complex scheduling problems.
The influence of the interstage sealing air-entraining jet structure at different positions and geometric struc-tures on the aerodynamic performance of the compressor cascade is deeply explored.The effect and mechanism of the interstage sealing air-entraining jet on the vortex structure in the corner region of the compressor cascade under the conditions of variable incidence angle are analyzed.It is found that with the movement of the jet port to the trailing edge and the increase of the width and height of the jet port,the suppression effect of the air-entraining jet on the total pressure loss of the cascade increases first and then decreases.When the jet port is located at the starting of corner separation,the height is 20%blade height and the width is 8%chord length,the improvement effect is the best,and the total pressure loss can be reduced by 8.5%compared with the prototype.In addition,the arrangement of the air-entraining jet structure under different incidence angle conditions reduces the influence range of the passage vortex pitchwise direction,but leads to the increase of the separation vortex height in different degrees.With the increase of the incidence angle,the suppression of the separation vortex in the pitchwise direction by the air-entraining jet struc-ture increases first and then decreases,and the increase effect of the flow loss near the mid-span gradually weakens.For this reason,the improvement effect of the air-entrining jet structure on the cascade performance is first enhanced and then weakened.The best improvement effect is shown at 0°,and significantly improve the aerodynamic perfor-mance of its high-load compressor cascade.
Diagnosis and early warning of compressor instability in aero engines are among the current research hotspots and challenges in the field of aero engines. To address the issues, such as capturing non-integer order frequency components in dynamic pressure signals, limited feature dimensionality, complex stall mechanisms, and the difficulty in quantifying evolution trajectories of compressor instability, a compressor aerodynamic instability early warning method is proposed based on phase-locked averaging filter and moment function neural network, using a multi-stage high-speed compressor as the research object. The method first employs phase-locked averaging filter to extract non-integer order frequency disturbance features under high-load conditions. Subsequently, an instability early warning model based on moment function neural network is constructed, which utilizes moment functions to capture global statistical features of instability and local detail features of asymmetric separation and intermittent pulses in early-stage weak signals. Next, Box-Cox transformation is introduced to eliminate heterogeneity among higher-order moment features, and multi-layer perceptron network layers are adopted to achieve early warning and diagnosis of compressor instability. Finally, the effectiveness of the proposed method is validated through test data at different rotational speeds. Results demonstrate that the method accurately characterizes the evolutionary laws of higher-order moment feature spaces, enabling efficient visualization, identification, and separation of instability precursors and steady-state data. Compared with the traditional surge detection method on the compressor rig, it can provide instability warning up to 4.8 s in advance.
Extreme thermal loads induced by shock/shock interaction seriously impair the performance metrics and structural safety of hypersonic vehicles,among which Type Ⅳ shock/shock interaction imposes the most significant im-pact.Using numerical simulation methods,the control effects and mechanisms of steady/oscillating jets in active jet systems on Type Ⅳ shock/shock interaction were systematically investigated.The results demonstrate that steady jets can remarkably improve the flow field characteristics of Type Ⅳ shock/shock interaction:when the jet Pressure Ratio(PR)is 5,compared with the uncontrolled flow field,the wall drag and maximum heat flux of the blunt body are re-duced by 40.7%and 40%,respectively.The core control mechanism is that the shock interaction point moves for-ward,the shock interaction type transitions from Type Ⅳ to a Type Ⅲ-like pattern,and the flow field structure changes from two triple-shock points to a single triple-shock point,eliminating the impact of the originally alternating expansion and compression waves on the wall.Further studies indicate that the drag and heat reduction effects of steady jets are strengthened with the increase of PR;however,a higher PR will aggravate the flow field instability.In contrast,oscil-lating jets arranged at the same position exhibit weaker flow control performance than steady jets under the same PR,which is due to internal pressure attenuation and the unidirectional oscillation effect.To address the bottleneck of high computational cost in traditional Computational Fluid Dynamics(CFD)simulations and efficiently determine the optimal PR under given incoming flow conditions,a Deep Neural Network(DNN)model based on McCulloch-Pitts(M-P)neurons was established for rapid flow field reconstruction.The results show that the flow field prediction speed of this method is increased by four orders of magnitude compared with CFD simulations,and the prediction accuracy ex-ceeds 0.99 when PR<18.This work provides an efficient approach for the parameter optimization of active flow con-trol targeting Type Ⅳ shock/shock interactions.
To meet the requirements of a complete flight mission profile for the flying-wing Unmanned Aerial Vehicles(UAVs),autonomous and reliable attitude control under wide flying envelope plays a crucial role.How to overcome the influence of flying-wing UAVs'insufficient longitudinal maneuverability,weak directional stability,as well as the combined effects of strong nonlinearity under wide operating conditions and external disturbances,to achieve precise and stable control,is a significant challenge.This paper focuses on the study of autonomous and reliable attitude con-trol of flying-wing UAVs under wide flying envelope.Firstly,to overcome the effects of strong nonlinearity and model deviations of the aircraft,an Incremental Nonlinear Dynamic Inversion(INDI)-based attitude control method is de-signed.Combined with the dynamical characteristics of coordinated turns for vehicles,reasonable calculation of pseudo commands is achieved.Based on the proposed method,extensive simulation tests for longitudinal,lateral-direction channels and crosswind disturbances environments were carried out,to verify the effectiveness and limitations of the method.Secondly,to address the differential calculation problem in INDI,a Tracking Differentiator(TD)-based pseudo-command optimization method was designed.Finally,simulation and actual flight experiments of the attitude controller based on combining INDI and TD were conducted,and the advantages of the designed method were validated through tests in noise environments.
The quaternion-based classical Terminal Sliding Mode(TSM)control methods tend to induce attitude un-winding problem when directly applied to spacecraft attitude control mission.Moreover,control practice must account for constraints such as actuator output limits and unknown external disturbances.To address these issues,this paper proposes a TSM control method that integrates both anti-unwinding and anti-windup capabilities.The system's kine-matic and dynamic models are established based on the unit error quaternion.A new terminal sliding surface is de-signed,and the finite-time stability and unwinding resistance of the sliding mode are proven using an asymmetric Ly-apunov function.Further,a control law with dynamic parameters and an auxiliary system is constructed to ensure finite-time convergence of system states and maintain global unwinding resistance.A nonlinear Disturbance OBserver(DOB)is introduced for dynamic compensation of lumped disturbances,effectively enhancing system robustness.Simulation results show that the method completes the attitude pointing process within 30 s,avoids unwinding,en-sures the control torque meets the 0.1 N·m amplitude constraint,and achieves significantly higher control accuracy than traditional methods,providing an innovative solution for high-precision spacecraft attitude control.
To address the strong nonlinearity, multi-actuator coupling, and stringent safety and robustness requirements of combined power engines during modal transition, an intelligent robust control method for the modal transition process is investigated. Focusing on the coordinated satisfaction of thrust tracking performance and safety constraints, an intelligent control framework based on deep reinforcement learning is established, in which an adversarial training mechanism is introduced to enhance robustness against observation disturbances and uncertainties. Based on a component-level engine simulation modal, multi-input multi-output control strategies and stage-wise training environments are designed for different modal transition processes, enabling adaptive policy learning and robustness improvement through adversarial reinforcement learning. In addition, to cope with engine parameter variations, a distributed control architecture based on multi-agent reinforcement learning is developed, and controller training is carried out under a centralized training and distributed execution scheme. Furthermore, real-time performance is validated on a hardware-in-the-loop platform, showing that the control cycle meets millisecond-level real-time requirements. Simulation results demonstrate that, under typical modal transition conditions, the proposed control method achieves steady-state thrust tracking errors within 1%, while exhibiting superior performance in thrust fluctuation amplitudes during modal transition compared with conventional control approaches. Under observation disturbances and parameter deviations, safety constraints are consistently satisfied without violation. The results indicate that the proposed intelligent robust control method effectively improves control accuracy, safety, and robustness during modal transition of combined power engines, providing a feasible solution for intelligent control of wide-speed-range combined power propulsion systems.
The thin-walled structure with lattice and stiffeners is a typical hybrid structure and effectively combines the load-bearing merits of lightweight lattices and thin-walled stiffened configurations, while demonstrating significant multifunctional potential that provides novel technical solutions for aerospace structural lightweighting. The increasing maturity of metal additive manufacturing technologies has laid a reliable foundation for the practical application of lattice structures. To facilitate the implementation of lattice structures in aerospace engineering, this paper focuses on thin-walled structures with lattice and stiffeners, primarily from the perspective of load-bearing structural design and its practical applications. Building upon the joint team's recent exploratory applications, this paper systematically outlines four critical aspects: fundamental characteristics of lattice structures, macroscale mechanical analysis methodologies, design of the lattice Representative Volume Element (RVE), eptimization design methodologies for thin-walled load-bearing structures with lattice and stiffeners. These systematic analyses aim to establish comprehensive reference guidelines for engineering designers. Furthermore, based on challenges encountered in aerospace, aviation, and aero-engine applications, this paper identifies priority research domains requiring urgent attention and critical technologies demanding breakthroughs in the hybrid structure design, offering valuable insights for researchers in related fields.
To address the issue of insufficient bending energy absorption in Carbon Fiber Reinforced Polymer (CFRP) C-frames for civil aircraft fuselages, this study reveals the failure mechanisms and energy dissipation characteristics through quasi-static four-point bending numerical simulations and experimental benchmarking. Furthermore, the titanium alloy local reinforcement design method is proposed based on the failure mode control. The results indicate that the bending failure of the CFRP C-frame originates from the coupling effect between upper flange buckling and web bulging, which induces high interlayer stress and initial delamination at the upper corner. Consequently, the load-bearing capacity plummets to 12% of the peak load after failure, significantly constraining energy absorption. Energy dissipation exhibits significant regional heterogeneity: the web acts as the core energy absorption zone (accounting for 50.2%), followed by the upper flange (24.1%), while the upper corner, serving as the failure initiation point, contributes only 13.8%. Among the titanium alloy local reinforcement strategies, the upper corner local reinforcement (UC configuration) yields optimal performance. By leveraging the plastic deformation of titanium alloy to effectively delay initial failure, this configuration achieves a 26.1% increase in total energy absorption and a 22.3% increase in specific energy absorption, with a structural weight increase of only 3.4%. The titanium alloy local reinforcement design achieves the best balance between lightweighting and crashworthiness, providing a theoretical basis and engineering guidance for the crashworthiness design of civil aircraft fuselage structures.
Gas path fault diagnosis is an essential component of aero-engine health management systems, with fault feature extraction being its key aspect. In recent years, with the advancement of deep learning techniques, gas path fault feature extraction methods based on graph neural networks have attracted considerable attention. However, conventional graph neural networks can only capture linear weighted aggregation relationships among nodes while neglecting the nonlinear coupling relationships prevalent in engine gas path systems, resulting in insufficient cross-condition diagnostic accuracy and interpretability. To address this issue, a gas path fault diagnosis method based on nonlinear correlation mining is proposed. This method achieves interpretable extraction of gas path fault features through an original nonlinear correlation mining layer and an improved graph convolutional layer. The performance of the proposed method was validated using full-lifecycle simulation data encompassing 500 flight sorties. The proposed method achieved zero false alarms throughout the entire operational period, with a detection rate of 88.99% and an isolation rate of 98.61%, significantly outperforming comparative methods based on convolutional, graph convolutional, and graph attention networks in terms of convergence and diagnostic accuracy.
To address the requirements of modern aircraft for high maneuverability and strong countermeasure capa-bility in complex tactical environments,and resolve the drawbacks of low accuracy and slow response in conventional indirect thrust control of engines,this paper proposes a direct thrust control method for multi-companion vectoring en-gines.First,a nonlinear component-level model consisting of one main engine and two companion engines is estab-lished,with refined modeling conducted for key components such as the bleed air system.To achieve accurate esti-mation of engine thrust,an on-board adaptive model based on the Unscented Kalman Filter(UKF)is designed.This model incorporates a steady/dynamic discrimination logic to on-line identify the engine's performance degradation pa-rameters,thereby effectively suppressing the interference of flight dynamics on health assessment.On this basis,a data-driven Model-Free Adaptive Control(MFAC)strategy is further proposed,and a Multi-Input Multi-Output(MIMO)direct thrust controller is constructed to realize decoupled,rapid,and precise control of the thrust of the main and com-panion engines.Simulation results demonstrate that the designed on-board adaptive model can accurately track the actual state of the engine,and its thrust estimation results are highly consistent with the true values;the direct thrust controller responds rapidly,achieving stable and precise tracking of the commanded thrust under the conditions that the settling time of the main engine is less than 1 s and the overshoot is below 4%.This study provides an effective so-lution for the control system design of novel combined vectored-thrust engines and verifies the feasibility and potential of this scheme in improving engine control performance.
The technology of material transportation between multiple aircraft via a connecting flying boom has broad application scenarios, such as material transfer between two aircraft, emergency repair of faulty aircraft in flight, and recovery of small unmanned aerial vehicles. The excellent aerodynamic and control characteristics of the flying boom are essential prerequisites for achieving high-precision docking and stable transportation. Focusing on a slender flying boom with control surfaces, its aerodynamic and flight control characteristics are investigated systematically through a combined approach of wind tunnel testing and numerical simulation. First, aerodynamic parameters and control surface efficiency data of the flying boom under various conditions were obtained through wind tunnel force measurement tests. Subsequently, a dynamic model of the flying boom was established based on the floating frame of reference theory to evaluate its dynamic stability, and a flight control law for the flying boom was designed on this basis. Then, open- and closed-loop control characteristics of the flying boom were verified through wind tunnel flight tests. Finally, numerical simulations were employed to analyze the influence of Reynolds number and Mach number on the aerodynamic characteristics of the flying boom under typical operating conditions. The evaluation results confirm the feasibility of applying the wind tunnel test conclusions in engineering applications.
Velocity field measurement techniques based on tracer particle imaging are widely employed in flow diagnostics. However, their application in complex flow environments is inevitably challenged by aero-optical aberrations. For instance, in combustion or supersonic/hypersonic environments, refractive index fields with drastic spatial variations induce severe aero optical effects, causing significant degradation of tracer particle images and compromising the accuracy of velocity measurements. To mitigate this issue, a novel technical approach for the non-blind restoration of degraded images of tracer particles via Singular Value Decomposition (SVD) of the Point Spread Function (PSF) is proposed. This method first performs SVD on PSFs obtained at multiple calibration positions within the field of view and subsequently reconstructs the full-field PSF distribution efficiently using a limited number of spatial modes and their corresponding spatial weighting coefficients. Furthermore, a Total Variation (TV) regularized Richardson-Lucy iterative deconvolution algorithm is employed to perform non-blind restoration on the degraded particle images using the reconstructed local PSFs. Through various numerical simulations, the influence of the number of SVD modes and image noise levels on the restoration performance is systematically investigated, and the restoration performance is compared with classical restoration algorithms. The results show that the proposed method works effectively in processing the sparse particle fields. It effectively recovers particle morphology blurred by aero-optical effects and substantially enhances the Peak Signal-to-Noise Ratio (PSNR) and Structural Similarity Index Measure (SSIM) of the images, offering a new perspective for the application of tracer-based velocimetry in harsh optical environments.
The time-varying characteristics of Low Earth Orbit (LEO) satellite networks and the unbalanced distribution of ground stations pose severe challenges to network robustness optimization and efficient traffic management. To address these issues, this paper proposes a LEO satellite key node evaluation algorithm fusing multi-dimensional spatiotemporal features, aiming to accurately identify the key nodes that maintain efficient communication between ground stations. The algorithm constructs a time-varying topological graph based on the two-layer interaction between inter-satellite and satellite-ground networks, designs a multi-dimensional node feature system from the perspectives of local structural attributes and global dependency relationships, and establishes a multi-dimensional spatiotemporal feature extraction model by integrating Multi-Layer Graph Convolutional Networks (MLGCNs) and Long Short-Term Memory (LSTM) networks. This model captures the spatiotemporal evolution law of the network and completes the node importance evaluation. Simulation results show that the proposed algorithm has significant advantages in both the accuracy of evaluation results and temporal stability; implementing a traffic diversion strategy based on the key nodes identified by the algorithm can effectively alleviate network congestion in high-load scenarios, providing a new research idea for the load optimization strategy of satellite networks.
With the growing engineering demand for reusable liquid rocket engines,ensuring stable and safe opera-tion under extreme conditions-high heat flux,high chamber pressure,and violent combustion-has become increasingly challenging.Plume spectroscopic diagnostics,featuring non-contact measurement,high sensitivity,and multi-parameter sensing capability,has emerged as an important technical route for engine health monitoring and fault iden-tification.We first analyze the key challenges in liquid rocket engine development and the diagnostic requirements for fault monitoring.The diagnostic mechanism based on atomic emission spectroscopy is then systematically elaborated,followed by a review of domestic and international research progress and the current state of the art in plume spectros-copy.Next,the core enabling technologies are summarized,including the construction of a spectrum-material-fault-mode database,controlled metal-impurity doping combustion tests,quantitative inversion of alloy species concentra-tions in the plume,flight-environment spectral diagnostics,and spectral-line interference and mitigation strategies for LOX/kerosene engines.The applicability,accuracy,and engineering feasibility of three representative measurement techniques-Fabry-Pérot interferometry,Fourier-transform infrared spectroscopy,and laser-induced breakdown spectroscopy-are further evaluated and compared for plume spectral acquisition.Finally,future trends are discussed,with emphasis on multimodal data fusion,artificial-intelligence-enabled analysis,on-chip spectroscopy combined with edge computing,and extensions toward the near-/mid-/far-infrared and terahertz bands,highlighting the broad pros-pects of plume spectroscopy for intelligent operation and predictive maintenance of liquid rocket engines.
To enable efficient, high-fidelity construction of white-box reduced-order aerodynamic models, a frequency-domain unsteady aerodynamic modeling approach is proposed based on Sparse Identification of Nonlinear Dynamics (SINDy). The proposed method uses simulation data of harmonic aircraft motions at representative amplitudes and frequencies, constructs a candidate function library guided by classical algebraic aerodynamic model architectures, and applies sparse regression to select optimal terms and identify parameters-thereby automatically yielding sparse, highly interpretable reduced-order aerodynamic models. Leveraging both classical algebraic model structures and Theodorsen's unsteady aerodynamic theory, we formulate a globally sampled unified model (SINDyA) and a parameter-varying local model (SINDyB). The approach is validated on two canonical problems-transonic pitch oscillations of the NACA64A010 airfoil and of the CHN-T1 aircraft-using lift and pitching-moment coefficients as modeling targets. Results indicate that the identified models require only a small number of dominant terms to capture the key nonlinear and hysteretic features of the unsteady aerodynamics; the SINDyB model, which performs local interpolation of coefficients, achieves higher prediction accuracy. Because the pitching-moment coefficient exhibits stronger nonlinearity, its prediction proves markedly more challenging than that of lift. The models predict aerodynamic responses accurately under small-amplitude excitations, while performance degrades for large-amplitude, high-frequency cases. These findings demonstrate the promise of symbolic machine-learning methods for constructing high-accuracy, interpretable unsteady aerodynamic models and highlight their potential for engineering application.
Aircraft stall,characterized by abrupt lift reduction due to severe aerodynamic flow separation on the wing surface,represents a safety critical phenomenon that can trigger problems such as stall roll and stall spin,posing sig-nificant threats to flight safety.Establishing high-precision dynamic models and conducting stall characteristic analysis are pivotal technical measures for preventing stall risks and enabling effective stall recovery.To address the trade-off between model complexity and accuracy in unsteady aerodynamic force modeling,a hybrid modeling approach inte-grating Particle Swarm Optimization and Extreme Learning Machine(PSO-ELM)was developed.By fusing multi-source wind tunnel test data,this method constructs an unsteady coupled aerodynamic force model with high predic-tive accuracy,low computational complexity,and robust adaptability to diverse operating conditions,thereby enhanc-ing the reliability of stall characteristic analysis and boundary computation.To overcome the limitations of conventional bifurcation analysis algorithms,including initial-value dependency and fixed-step rigidity that hinder thorough explora-tion of solution spaces,a modified bifurcation analysis algorithm incorporating"random state-point generation"and"adaptive step-size adjustment"was proposed.Combined with saddle-node manifold theory,this approach enables global analysis of longitudinal nonlinear dynamic characteristics in aircraft.Using a statically unstable aircraft's longitu-dinal model as a case study,simulations of both open-loop and backstepping control-integrated closed-loop systems were conducted.The focus was on analyzing global longitudinal dynamic behavior and solving stall boundaries,pro-viding dynamic support for mitigating unsteady hysteresis effects,expanding stable flight envelopes,and suppressing unintended stalls.
In desert environments,the dust cloud induced by rotor downwash(the"brownout"phenomenon)con-sists of an enormous number of sand and dust particles,which causes a sharp drop in visibility around the aircraft,posing a serious flight safety hazard.In existing relevant numerical simulation studies,the method of tracking real par-ticles incurs extremely high computational costs,while simulations with reduced particle counts can only achieve quali-tative analysis,failing to meet the requirements of quantitative prediction.Based on the Coarse-Grained Discrete Ele-ment Method(CG-DEM),four-way coupling is adopted to account for inter-particle collisions and the feedback effect of particles on the flow field.By tracking Coarse-Grained(CG)particles,simulation results that quantitatively agree with the actual development of rotor-induced dust cloud can be obtained.To consider the influence of turbulence on particle transport in rotor-induced dust cloud,a particle drag force model incorporating turbulent effect is employed.The results show that the CG-DEM can effectively realize the quantitative simulation of rotor-induced dust cloud.For the case in this study,the use of the drag force model considering turbulent effect improves the prediction accuracy.In the steady state,the number of airborne particles increases by approximately 27%compared with the traditional drag model,and the error between the predicted sand transport rate and the experimental measurement is less than 5%.
High-precision aeroengine digital models require well-matched component characteristics data. An improved component characteristics modification method is proposed to address the non-unique results and non-smooth curved surfaces from the traditional high-to-low method. Preliminary scaling modification is carried out through reference points to ensure high accuracy at design point and broad coverage across all operation points. Fitness function constructed from target parameters is optimized by using particle swarm optimization algorithm to determine optimal component correction coefficients for all operation points. Component characteristic maps are expanded by adding characteristic lines corresponding to these conditions. Both the expanded and original lines are scaled using the obtained coefficients, resulting in a fully-matched and continuously smooth characteristic curved surface. A comparative simulation between the improved method and traditional method was conducted on a two-spool mixed-flow engine. The simulation results demonstrate that all the relative errors of the target parameters corrected by the improved method are less than 1%, outperforming the tradition method in overall accuracy while successfully avoiding non-smooth correction results.
