Fragments,as the primary damage elements of blast-fragmentation warhead,have their damage lethality which is quantitatively evaluated through precise testing of parameters such as fragment velocity,spatial distribution and mass characteristics.This paper systematically reviews the latest advancements in the parameter testing technologies for the fragment fields of blast-fragmentation warheads,focusing on comparative analysis under two typical conditions of static and dynamic detonations.In the context of static detonation testing,the principles and features of the contact-type technologies such as net targets,the sectional optoelectronic technologies of light curtains and radar and the 3D reconstruction technologies like high-speed stereovision are compared in detail,and their technological improvements and development trends are elaborated.In the context of dynamic detonation testing,the research achievements in testing methods and simulation modeling at home and abroad are reviewed,and the unique challenges such as detonation point control and spatiotemporal synchronizationunder dynamic detonation conditions as well as the corresponding solutions are thoroughly analyzed.Furthermore,this paper also explores the applications and enabling potential of intelligent algorithms represented by machine learning (particularly deep learning) in the aspects fragment target recognition,trajectory tracking,data fusion,and dynamic explosion parameter prediction.Finally,it offers prospects for the future development trends in fragment field parameter testing technologies,and proposes the need to prioritize high-precision dynamic detonation fragment parameter testing technologies,enhance the 3D reconstruction capabilities and strengthen the integration of machine learning in testing methodologies,thereby supporting the optimized design and damage effectiveness evaluation of blast-fragmentation warheads.
For the flight control of quadrotor unmanned aerial vehicles (UAVs) affected by multi-source disturbance,a self-tuning sliding mode flight control algorithm based on multi-source disturbance compensation and hierarchical sliding mode control is proposed.The quadrotor UAV system is divided into a two-degrees-of-freedom dual-input fully-actuated subsystem and a four-degrees-of-freedom dual-input underactuated subsystem,based on which the sliding variable of the fully-actuated subsystem is defined.A sliding variable construction method for hierarchical sliding mode controller is proposed,and the sliding variable of the underactuated subsystem is designed by adopting this method.A generalized Super-Twisting disturbance approximator is designed to observe the multi-source disturbance,and a fuzzy compensator is designed to compensate for the approximation error of the disturbance approximator.A reaching gain self-tuning sliding mode controller is designed by utilizing the combined power reaching law,and the disturbance approximator with fuzzy compensator is used to suppress the influence of the multi-source disturbance.The roles of the core modules in the proposed control algorithm are analyzed,and the proposed control algorithm is compared with the existing control algorithms.The comparison results show that the proposed control algorithm has good multi-source disturbance rejection ability,and can also provide superior dynamic control performance.
A low-sidelobe array pattern design method utilizing fmincon is proposed based on MATLAB.The array beam optimization problem is transformed into a nonlinear constrained optimization problem with continuous variables by constructing a hybrid penalty cost function with the objective of minimizing the sidelobe level in the target region and utilizing the nonlinear constraints for controlling the gain fluctuation of mainlobe.Numerical results demonstrate that The designed array can achieve a suppression of 15-20dB sidelobe level within specified spatial regions while ensuring that the gain fluctuation of mainlobe is less than 0.3dB.And the optimization process exhibits rapid convergence and excellent stability.The proposed method gives consideration to optimization precision and computational efficiency.It provides a high-performance and high-reliability solution for antenna array designs requiring the mainlobe shaping and sidelobe suppression in specific region,has significant engineering practical value.
The shaped charge is widely used in the field of unexploded ordnance destruction disposal due to its high energy utilization rate.For the safe and effective disposal of unexploded ordnance in water,this paper investigates the factors influencing on the underwater initiation capability of jets.Numerical simulation methods are employed to characterize the jet formation process of two charge structures in water.Focusing on the flared liner shaped charge structure,the effects of the charge structure,the length of the combined cavity structure at the head,and the water depth on the initiation capability of the shaped charge jet are analyzed.The results indicate that the flared liner charge structure affects the jet’s initiation capability in the following order of significance:liner height,liner thickness,explosive column height,and liner curvature.The combined cavity structure significantly reduces the energy loss of the jet by 74.8%.The Random Forest (RF) model effectively predicts the nonlinear variation characteristics of the contact velocity between jet and shelled charges in water with the increase in water depth.
To address the challenge that the frequency-agile radars in the stable tracking phase in modern complex electromagnetic environments struggle have difficulty in balancing both coherent integration gain and anti-jamming capability under blanket jamming,this paper proposes a joint control method for frequency and dwell time based on proximal policy optimization (PPO).Firstly,the trade-off relationship between radar coherent integration gain and intercept risk is analyzed.The dynamic combined jamming behavior of jammers is modeled,and a Markov decision process model for radar anti-jamming,incorporating signal-level feature feedback,is constructed.Secondly,the PPO algorithm is introduced.A compound reward function that takes into account both detection performance and survivability is designed,enabling the radar agent to autonomously learn the optimal strategy in unknown dynamical adversarial environments.This strategy can adjust the operating frequency and dwell time in real-time based on environmental feedback to maximize the signal-to-interference-plus-noise ratio.Simulated results demonstrate that,compared to traditional strategies,the proposed strategy can effectively evade the combined blanket jamming.It significantly reduces the probability of being jammed while substantially improving the output SINR,exhibiting strong environmental adaptability and robustness.
To address the issue of aeroelastic instability in fuze turbine generators under wide-Mach-number and high-rotational-speed conditions,the deformation characteristics of blades and their impact on the aerodynamic torque are investigated using a sequential fluid-structure coupling method.The decoupling analysis of the loads reveals that the predominant cause of blade deformation is the high-speed centrifugal load,accounting for over 98% of the total deformation,rather than the aerodynamic load.The research demonstrates that this centrifugal-induced deformation leads to the significant geometric distortion of the blades,compromising the original aerodynamic profile.Consequently,the driving torque of turbine generator declines gradiently as the rotational speed increases.The torque is reduced by approximately 26.6% at the maximum rotational speed.The results reveal the underlying mechanism of the flexible blade deformation affecting the performance of fuze turbine generator,providing a theoretical foundation for the robust design of fuze power supplies.
The performance of signal acquisition in satellite navigation receivers is significantly influenced by the setting of detection thresholds.The existing methods are difficult to to adapt to the increasingly complex electromagnetic environments in urban areas,where they are prone to generate false alarms under multiple interference conditions,severely affecting acquisition reliability.To address this issue,a constant false alarm rate (CFAR) detection technique is incorporated into the acquisition of BeiDou B1C signals.An algorithm that dynamically adjusts the decision threshold based on the estimated number of interference targets is proposed.This algorithm adaptively determines the decision threshold of signal acquisition according to the real-time estimated number of interference targets in the environment,thereby enhancing the performance of receiver in multi-interference environment.Simulated results demonstrate that the proposed algorithm significantly improves the robustness and acquisition sensitivity of the receiver in complex interference environments,thereby increasing the operational reliability of military equipment in urban warfare.
Manual assembly of aero-engine casings with dissimilar flanges suffers from low efficiency,poor mating accuracy and high safety risk.To overcome these problems,this paper proposes a compliant control scheme for a six-degree-of-freedom Stewart platform that learns to insert a flange test specimen via reinforcement learning.Firstly,the kinematic and dynamic models of the Stewart platform are established,and its ability to adjust posture under complex constraints is analyzed.Secondly,a data-driven compliant strategy that fuses admittance control with a reinforcement-learning algorithm is designed to realize high-precision mating and gentle insertion.Finally,a co-simulation framework combining Webots and PyCharm is built to verify the effectiveness and robustness of the proposed method under various initial misalignments and disturbances.Results show that,compared with traditional fixed parameter admittance control methods,the proposed method reduces the maximum contact force of the platform by 30.4% and improves the assembly efficiency by about 90% during the assembly process,and achieves a steady-state position error of only 7.5mm.This indicates that the proposed method significantly improves work efficiency and accuracy while ensuring assembly safety,and has good engineering promotion value.
The evaluation of aviation weaponry’s combat capability is not merely a technical matter,but also a strategic one.It directly determines whether the equipment can fulfill its intended operational roles on the battlefield and whether future warfare patterns can be accurately predicted.To address the inherent limitations of traditional evaluation methods—including such as over-reliance on expert experience,insufficient persuasiveness of evaluation models,and difficulties in indicator acquisition,this paper proposes a combat capability evaluation approach for aviation weaponry based on Uncertainty Structural Equation Modeling (USEM).Specifically,Triangular Fuzzy Number (TFN) and Deep Neural Networks (DNN) are both integrated to process collected simulation data,aiming to mitigate the drawbacks of limited data availability and significant subjectivity in expert-derived data,thereby enhancing modeling accuracy.The combat capability indicator values processed via the triangular fuzzy number algorithm are converted into SPSS-compatible files,and imported into AMOS for analysis,enabling the establishment of correlations between the combat capability index values and the structural equation model.The results demonstrate that the CN value ranges from 0 to 3,and the RMSEA value is less than 0.08,The CFI and TLI values are both greater than 0.9,all indicating excellent model fit.Furthermore,the model’s estimated values align with the actual missile combat capability indicators,verifying its validity for evaluating the combat capability of aviation weapons and equipment.
To address the threats posed to the unmanned aerial vehicles (UAVs) by flame spread and smoke dispersion in forest fires,this paper proposes a real-time method for generating a regular virtual tube in a dynamic environment with coupled fire and smoke to support safety-critical UAV flight.Firstly,a dynamic environment prediction model is developed based on the wildfire spread dynamics and a smoke advection-diffusion mechanism.A dual-metric mechanism is then proposed to simultaneously characterize the real-time positional risk and spatiotemporal predictive risk,resulting in a unified quantified cost field for both static and dynamic obstacles and a spatiotemporal constraint window.Next,a risk-aware tube-generation framework is built to search a spatiotemporally feasible path by using a risk-aware A* algorithm and generate the regular virtual tube by using an adaptive safe virtual-tube generation algorithm.Simulated results show that the proposed method can regenerate the virtual tube in real time as the fire conditions evolve.Compared with the existing methods,the proposed method increases the minimum safe distance by 43.2%,reduces the maximum cost by 28.7%,and decreases the replanning time by 82.9%,significantly improving both the trajectory safety and flight safety of UAVs in multi-source high-risk environments.
