针对现有SLAM算法在渲染真实感、内存占用和复杂场景适应性方面的不足,提出了一种基于3D Gaussians Splatting的密集SLAM算法——TIGO-SLAM(tensor illumination and Gaussian optimization for indoor SLAM)。该算法集成了基于神经网...针对现有SLAM算法在渲染真实感、内存占用和复杂场景适应性方面的不足,提出了一种基于3D Gaussians Splatting的密集SLAM算法——TIGO-SLAM(tensor illumination and Gaussian optimization for indoor SLAM)。该算法集成了基于神经网络的张量光照模型、改进的高斯遮罩算法以及网格化神经场的几何和颜色属性表示,具体创新包括:a)基于神经网络的张量光照模型,增强镜面反射与漫反射效果,从而提升了渲染真实感;b)通过冗余高斯剔除机制改进高斯遮罩算法,有效降低了内存消耗并提高了实时性;c)结合网格化神经场的几何与颜色属性表示,采用优化的码本存储方式,显著提高了渲染性能和场景重建精度。实验结果表明,TIGO-SLAM在室内场景渲染、内存优化和复杂场景适应性方面均有显著提升,特别是在动态室内环境中的渲染和重建效果表现突出,为SLAM技术在资源受限设备上的应用提供了新的可能。展开更多
Advances in software and hardware technologies have facilitated the production of quadrotor unmanned aerial vehicles(UAVs).Nowadays,people actively use quadrotor UAVs in essential missions such as search and rescue,co...Advances in software and hardware technologies have facilitated the production of quadrotor unmanned aerial vehicles(UAVs).Nowadays,people actively use quadrotor UAVs in essential missions such as search and rescue,counter-terrorism,firefighting,surveillance,and cargo transportation.While performing these tasks,quadrotors must operate in noisy environments.Therefore,a robust controller design that can control the altitude and attitude of the quadrotor in noisy environments is of great importance.Many researchers have focused only on white Gaussian noise in their studies,whereas researchers need to consider the effects of all colored noises during the operation of the quadrotor.This study aims to design a robust controller that is resistant to all colored noises.Firstly,a nonlinear quadrotormodel was created with MATLAB.Then,a backstepping controller resistant to colored noises was designed.Thedesigned backstepping controller was tested under Gaussian white,pink,brown,blue,and purple noises.PID and Lyapunov-based controller designswere also carried out,and their time responses(rise time,overshoot,settling time)were compared with those of the backstepping controller.In the simulations,time was in seconds,altitude was in meters,and roll,pitch,and yaw references were in radians.Rise and settling time values were in seconds,and overshoot value was in percent.When the obtained values are examined,simulations prove that the proposed backstepping controller has the least overshoot and the shortest settling time under all noise types.展开更多
In engineering practice,it is often necessary to determine functional relationships between dependent and independent variables.These relationships can be highly nonlinear,and classical regression approaches cannot al...In engineering practice,it is often necessary to determine functional relationships between dependent and independent variables.These relationships can be highly nonlinear,and classical regression approaches cannot always provide sufficiently reliable solutions.Nevertheless,Machine Learning(ML)techniques,which offer advanced regression tools to address complicated engineering issues,have been developed and widely explored.This study investigates the selected ML techniques to evaluate their suitability for application in the hot deformation behavior of metallic materials.The ML-based regression methods of Artificial Neural Networks(ANNs),Support Vector Machine(SVM),Decision Tree Regression(DTR),and Gaussian Process Regression(GPR)are applied to mathematically describe hot flow stress curve datasets acquired experimentally for a medium-carbon steel.Although the GPR method has not been used for such a regression task before,the results showed that its performance is the most favorable and practically unrivaled;neither the ANN method nor the other studied ML techniques provide such precise results of the solved regression analysis.展开更多
To further understand the performance of the energy harvesters under the influence of the wind force and the random excitation,this paper investigates the stochastic response of the bio-inspired energy harvesters subj...To further understand the performance of the energy harvesters under the influence of the wind force and the random excitation,this paper investigates the stochastic response of the bio-inspired energy harvesters subjected to Gaussian white noise and galloping excitation,simulating the flapping pattern of a seagull and its interaction with wind force.The equivalent linearization method is utilized to convert the original nonlinear model into the Itôstochastic differential equation by minimizing the mean squared error.Then,the second-order steady-state moments about the displacement,velocity,and voltage are derived by combining the moment analysis theory.The theoretical results are simulated numerically to analyze the stochastic response performance under different noise intensities,wind speeds,stiffness coefficients,and electromechanical coupling coefficients,time domain analysis is also conducted to study the performance of the harvester with different parameters.The results reveal that the mean square displacement and voltage increase with increasing the noise intensity and wind speed,larger absolute values of stiffness coefficient correspond to smaller mean square displacement and voltage,and larger electromechanical coupling coefficients can enhance the mean square voltage.Finally,the influence of wind speed and electromechanical coupling coefficient on the stationary probability density function(SPDF)is investigated,revealing the existence of a bimodal distribution under varying environmental conditions.展开更多
The generation and reconnection of magneticflux ropes in a plasma irradiated by two Laguerre–Gaussian laser pulses with different frequen-cies and opposite topological charges are investigated numerically by particle-...The generation and reconnection of magneticflux ropes in a plasma irradiated by two Laguerre–Gaussian laser pulses with different frequen-cies and opposite topological charges are investigated numerically by particle-in-cell simulations.It is shown that twisted plasma currents and hence magneticflux ropes can be effectively generated as long as the laser frequency difference matches the electron plasma frequency.More importantly,subsequent reconnection of magneticflux ropes can occur.Typical signatures of magnetic reconnection,such as magnetic island formation and plasma heating,are identified in the reconnection of magneticflux ropes.Notably,it is found that a strong axial magneticfield can be generated on the axis,owing to the azimuthal current induced during the reconnection of the ropes.This indicates that in the reconnection of magneticflux ropes,the energy can be transferred not only from the magneticfield to the plasma but also from the plasma current back to the magneticfield.This work opens a new avenue to the study of magneticflux ropes,which helps in understanding magnetic topology changes,and resultant magnetic energy dissipation,plasma heating,and particle acceleration found in solarflares,and magnetic confinement fusion devices.展开更多
Topological band theory has been studied for free fermions for decades,and one of the most profound physical results is the bulk-boundary correspondence.Recently a focus in topological physics is extending topological...Topological band theory has been studied for free fermions for decades,and one of the most profound physical results is the bulk-boundary correspondence.Recently a focus in topological physics is extending topological classification to mixed states.Here,we focus on Gaussian mixed states for which the modular Hamiltonians of the density matrix are quadratic free fermion models with U(1)symmetry and can be classified by topological invariants.The bulk-boundary correspondence is then manifested as stable gapless modes of the modular Hamiltonian and degenerate spectrum of the density matrix.In this article,we show that these gapless modes can be detected by the full counting statistics,mathematically described by a function introduced as F(θ).A divergent derivative atθ=πcan be used to probe the gapless modes in the modular Hamiltonian.Based on this,a topological indicator,whose quantization to unity senses topologically nontrivial mixed states,is introduced.We present the physical intuition of these results and also demonstrate these results with concrete models in both one-and two-dimensions.Our results pave the way for revealing the physical significance of topology in mixed states.展开更多
Dear Editor,This letter presents a joint probabilistic scheduling and resource allocation method(PSRA) for 5G-based wireless networked control systems(WNCSs). As a control-aware optimization method, PSRA minimizes the...Dear Editor,This letter presents a joint probabilistic scheduling and resource allocation method(PSRA) for 5G-based wireless networked control systems(WNCSs). As a control-aware optimization method, PSRA minimizes the linear quadratic Gaussian(LQG) control cost of WNCSs by optimizing the activation probability of subsystems, the number of uplink repetitions, and the durations of uplink and downlink phases. Simulation results show that PSRA achieves smaller LQG control costs than existing works.展开更多
Energy resolution calibration is crucial for gamma-ray spectral analysis,as measured using a scintillation detector.A locally constrained regularization method was proposed to determine the resolution calibration para...Energy resolution calibration is crucial for gamma-ray spectral analysis,as measured using a scintillation detector.A locally constrained regularization method was proposed to determine the resolution calibration parameters.First,a Monte Carlo simulation model consistent with an actual measurement system was constructed to obtain the energy deposition distribution in the scintillation crystal.Subsequently,the regularization objective function is established based on weighted least squares and additional constraints.Additional constraints were designed using a special weighting scheme based on the incident gamma-ray energies.Subsequently,an intelligent algorithm was introduced to search for the optimal resolution calibration parameters by minimizing the objective function.The most appropriate regularization parameter was determined through mathematical experiments.When the regularization parameter was 30,the calibrated results exhibited the minimum RMSE.Simulations and test pit experiments were conducted to verify the performance of the proposed method.The simulation results demonstrate that the proposed algorithm can determine resolution calibration parameters more accurately than the traditional weighted least squares,and the test pit experimental results show that the R-squares between the calibrated and measured spectra are larger than 0.99.The accurate resolution calibration parameters determined by the proposed method lay the foundation for gamma-ray spectral processing and simulation benchmarking.展开更多
With expeditious advancements in AI-driven facial manipulation techniques,particularly deepfake technology,there is growing concern over its potential misuse.Deepfakes pose a significant threat to society,partic-ularl...With expeditious advancements in AI-driven facial manipulation techniques,particularly deepfake technology,there is growing concern over its potential misuse.Deepfakes pose a significant threat to society,partic-ularly by infringing on individuals’privacy.Amid significant endeavors to fabricate systems for identifying deepfake fabrications,existing methodologies often face hurdles in adjusting to innovative forgery techniques and demonstrate increased vulnerability to image and video clarity variations,thereby hindering their broad applicability to images and videos produced by unfamiliar technologies.In this manuscript,we endorse resilient training tactics to amplify generalization capabilities.In adversarial training,models are trained using deliberately crafted samples to deceive classification systems,thereby significantly enhancing their generalization ability.In response to this challenge,we propose an innovative hybrid adversarial training framework integrating Virtual Adversarial Training(VAT)with Two-Generated Blurred Adversarial Training.This combined framework bolsters the model’s resilience in detecting deepfakes made using unfamiliar deep learning technologies.Through such adversarial training,models are prompted to acquire more versatile attributes.Through experimental studies,we demonstrate that our model achieves higher accuracy than existing models.展开更多
We apply methods of algebraic integral geometry to prove a special case of the Gaussian kinematic formula of Adler-Taylor.The idea,suggested already by Adler and Taylor,is to view the GKF as the limit of spherical kin...We apply methods of algebraic integral geometry to prove a special case of the Gaussian kinematic formula of Adler-Taylor.The idea,suggested already by Adler and Taylor,is to view the GKF as the limit of spherical kinematic formulas for spheres of large dimension N and curvature1/N.展开更多
Transpiration cooling is crucial for the performance of aerospace engine components,relying heavily on the processing quality and accuracy of microchannels.Laser powder bed fusion(LPBF)offers the potential for integra...Transpiration cooling is crucial for the performance of aerospace engine components,relying heavily on the processing quality and accuracy of microchannels.Laser powder bed fusion(LPBF)offers the potential for integrated manufacturing of complex parts and precise microchannel fabrication,essential for engine cooling applications.However,optimizing LPBF’s extensive process parameters to control processing quality and microchannel accuracy effectively remains a significant challenge,especially given the time-consuming and labor-intensive nature of handling numerous variables and the need for thorough data analysis and correlation discovery.This study introduced a combined methodology of high-throughput experiments and Gaussian process algorithms to optimize the processing quality and accuracy of nickel-based high-temperature alloy with microchannel structures.250 parameter combinations,including laser power,scanning speed,channel diameter,and spot compensation,were designed across ten high-throughput specimens.This setup allowed for rapid and efficient evaluation of processing quality and microchannel accuracy.Employing Bayesian optimization,the Gaussian process model accurately predicted processing outcomes over a broad parameter range.The correlation between various processing parameters,processing quality and accuracy was revealed,and various optimized process combinations were summarized.Verification through computed Tomography testing of the specimens confirmed the effectiveness and precision of this approach.The approach introduced in this research provides a way for quickly and efficiently optimizing the process parameters and establishing process-property relationships for LPBF,which has broad application value.展开更多
To address the early separation problem in the Menter Shear-Stress Transport(SST)turbulence model,a correction for the Turbulent Kinetic Energy(TKE)production term,P_(k),is introduced to account for the effect of the ...To address the early separation problem in the Menter Shear-Stress Transport(SST)turbulence model,a correction for the Turbulent Kinetic Energy(TKE)production term,P_(k),is introduced to account for the effect of the Adverse Pressure Gradient(APG).The correction is determined based on the distribution of Pkin the APG region before separation.When the friction coefficient C_(f) is decomposed,its direct dependence on Pkis clearly observed.However,with the introduction of Bradshaw’s assumption,Pkin the SST turbulence model is over-suppressed,resulting in a lower inner peak or no significant inner peak distribution at all.To address this problem,this paper proposes a Gaussian function,HGauss,which corrects the numerical values of P_(k) involved in the calculation of the Menter SST model by focusing on the inner peak region of P_(k).The modified SST model is then applied to four cases with APGs.The modification leads to an increase in the wall friction coefficient C_(f)in the APG region and causes a downstream shift in the separation location,improving the model’s consistency with high-accuracy data and experimental results.It is demonstrated that this correction can improve the early separation problem in the Menter SST turbulence model.展开更多
Lithium-ion batteries(LIBs)have been widely used in mobile energy storage systems because of their high energy density,long life,and strong environmental adaptability.Accurately estimating the state of health(SOH)for ...Lithium-ion batteries(LIBs)have been widely used in mobile energy storage systems because of their high energy density,long life,and strong environmental adaptability.Accurately estimating the state of health(SOH)for LIBs is promising and has been extensively studied for many years.However,the current prediction methods are susceptible to noise interference,and the estimation accuracy has room for improvement.Motivated by this,this paper proposes a novel battery SOH estimation method,the Beluga Whale Optimization(BWO)and Noise-Input Gaussian Process(NIGP)Stacked Model(BGNSM).This method integrates the BWO-optimized Gaussian Process Regression(GPR)with the NIGP.It combines their predictions using a stacked GPR model which reduces the problem of large input data noise and improves the prediction accuracy.The experimental results show that the BGNSM method has good accuracy,generalization ability,and robustness,and performs well in small sample situations.The Root Mean Square Error(RMSE)and Mean Absolute Error(MAE)are as low as 0.218%and 0.164%,respectively,which is close to 0.At the same time,R-Square(R^(2))is as high as 0.9948,which is close to 1,indicating that the estimated results in this paper are highly consistent with the actual results.展开更多
The elastic properties of membranes are typically characterized by a few phenomenological parameters,including bending and Gaussian curvature moduli measuring the membrane rigidity against its deformation and topologi...The elastic properties of membranes are typically characterized by a few phenomenological parameters,including bending and Gaussian curvature moduli measuring the membrane rigidity against its deformation and topological change,as well as spontaneous curvature arising from the asymmetry between the two leaflets in the lipid bilayers.Though tether-based and fluctuationbased experiments are commonly utilized to measure the bending modulus,measuring the Gaussian curvature modulus and the spontaneous curvature of the membrane is considered to be much more difficult.In this paper,we study the buckling process of a circular membrane with nonzero spontaneous curvature under compressive stresses.It is found that when the stress exceeds a critical value,the circular membrane will transform from a spherical cap to a buckled shape,with its buckling degree enhanced with the increase of stress until its base is constricted to almost zero.As the stress-strain relationship of the buckled membrane strongly depends on the Gaussian curvature modulus and the spontaneous curvature,we therefore propose a method to determine the Gaussian curvature modulus and the spontaneous curvature simultaneously by measuring its stress-strain relationship during a buckling process.展开更多
Adversarial attacks pose a significant threat to artificial intelligence systems by exposing them to vulnerabilities in deep learning models.Existing defense mechanisms often suffer drawbacks,such as the need for mode...Adversarial attacks pose a significant threat to artificial intelligence systems by exposing them to vulnerabilities in deep learning models.Existing defense mechanisms often suffer drawbacks,such as the need for model retraining,significant inference time overhead,and limited effectiveness against specific attack types.Achieving perfect defense against adversarial attacks remains elusive,emphasizing the importance of mitigation strategies.In this study,we propose a defense mechanism that applies random cropping and Gaussian filtering to input images to mitigate the impact of adversarial attacks.First,the image was randomly cropped to vary its dimensions and then placed at the center of a fixed 299299 space,with the remaining areas filled with zero padding.Subsequently,Gaussian×filtering with a 77 kernel and a standard deviation of two was applied using a convolution operation.Finally,the×smoothed image was fed into the classification model.The proposed defense method consistently appeared in the upperright region across all attack scenarios,demonstrating its ability to preserve classification performance on clean images while significantly mitigating adversarial attacks.This visualization confirms that the proposed method is effective and reliable for defending against adversarial perturbations.Moreover,the proposed method incurs minimal computational overhead,making it suitable for real-time applications.Furthermore,owing to its model-agnostic nature,the proposed method can be easily incorporated into various neural network architectures,serving as a fundamental module for adversarial defense strategies.展开更多
文摘针对现有SLAM算法在渲染真实感、内存占用和复杂场景适应性方面的不足,提出了一种基于3D Gaussians Splatting的密集SLAM算法——TIGO-SLAM(tensor illumination and Gaussian optimization for indoor SLAM)。该算法集成了基于神经网络的张量光照模型、改进的高斯遮罩算法以及网格化神经场的几何和颜色属性表示,具体创新包括:a)基于神经网络的张量光照模型,增强镜面反射与漫反射效果,从而提升了渲染真实感;b)通过冗余高斯剔除机制改进高斯遮罩算法,有效降低了内存消耗并提高了实时性;c)结合网格化神经场的几何与颜色属性表示,采用优化的码本存储方式,显著提高了渲染性能和场景重建精度。实验结果表明,TIGO-SLAM在室内场景渲染、内存优化和复杂场景适应性方面均有显著提升,特别是在动态室内环境中的渲染和重建效果表现突出,为SLAM技术在资源受限设备上的应用提供了新的可能。
文摘Advances in software and hardware technologies have facilitated the production of quadrotor unmanned aerial vehicles(UAVs).Nowadays,people actively use quadrotor UAVs in essential missions such as search and rescue,counter-terrorism,firefighting,surveillance,and cargo transportation.While performing these tasks,quadrotors must operate in noisy environments.Therefore,a robust controller design that can control the altitude and attitude of the quadrotor in noisy environments is of great importance.Many researchers have focused only on white Gaussian noise in their studies,whereas researchers need to consider the effects of all colored noises during the operation of the quadrotor.This study aims to design a robust controller that is resistant to all colored noises.Firstly,a nonlinear quadrotormodel was created with MATLAB.Then,a backstepping controller resistant to colored noises was designed.Thedesigned backstepping controller was tested under Gaussian white,pink,brown,blue,and purple noises.PID and Lyapunov-based controller designswere also carried out,and their time responses(rise time,overshoot,settling time)were compared with those of the backstepping controller.In the simulations,time was in seconds,altitude was in meters,and roll,pitch,and yaw references were in radians.Rise and settling time values were in seconds,and overshoot value was in percent.When the obtained values are examined,simulations prove that the proposed backstepping controller has the least overshoot and the shortest settling time under all noise types.
基金supported by the SP2024/089 Project by the Faculty of Materials Science and Technology,VˇSB-Technical University of Ostrava.
文摘In engineering practice,it is often necessary to determine functional relationships between dependent and independent variables.These relationships can be highly nonlinear,and classical regression approaches cannot always provide sufficiently reliable solutions.Nevertheless,Machine Learning(ML)techniques,which offer advanced regression tools to address complicated engineering issues,have been developed and widely explored.This study investigates the selected ML techniques to evaluate their suitability for application in the hot deformation behavior of metallic materials.The ML-based regression methods of Artificial Neural Networks(ANNs),Support Vector Machine(SVM),Decision Tree Regression(DTR),and Gaussian Process Regression(GPR)are applied to mathematically describe hot flow stress curve datasets acquired experimentally for a medium-carbon steel.Although the GPR method has not been used for such a regression task before,the results showed that its performance is the most favorable and practically unrivaled;neither the ANN method nor the other studied ML techniques provide such precise results of the solved regression analysis.
文摘To further understand the performance of the energy harvesters under the influence of the wind force and the random excitation,this paper investigates the stochastic response of the bio-inspired energy harvesters subjected to Gaussian white noise and galloping excitation,simulating the flapping pattern of a seagull and its interaction with wind force.The equivalent linearization method is utilized to convert the original nonlinear model into the Itôstochastic differential equation by minimizing the mean squared error.Then,the second-order steady-state moments about the displacement,velocity,and voltage are derived by combining the moment analysis theory.The theoretical results are simulated numerically to analyze the stochastic response performance under different noise intensities,wind speeds,stiffness coefficients,and electromechanical coupling coefficients,time domain analysis is also conducted to study the performance of the harvester with different parameters.The results reveal that the mean square displacement and voltage increase with increasing the noise intensity and wind speed,larger absolute values of stiffness coefficient correspond to smaller mean square displacement and voltage,and larger electromechanical coupling coefficients can enhance the mean square voltage.Finally,the influence of wind speed and electromechanical coupling coefficient on the stationary probability density function(SPDF)is investigated,revealing the existence of a bimodal distribution under varying environmental conditions.
基金supported by the National Natural Science Foundation of China(Grant Nos.12375236 and 12135009)the Strategic Priority Research Program of the Chinese Academy of Sciences(Grant Nos.XDA25050100 and XDA25010100).
文摘The generation and reconnection of magneticflux ropes in a plasma irradiated by two Laguerre–Gaussian laser pulses with different frequen-cies and opposite topological charges are investigated numerically by particle-in-cell simulations.It is shown that twisted plasma currents and hence magneticflux ropes can be effectively generated as long as the laser frequency difference matches the electron plasma frequency.More importantly,subsequent reconnection of magneticflux ropes can occur.Typical signatures of magnetic reconnection,such as magnetic island formation and plasma heating,are identified in the reconnection of magneticflux ropes.Notably,it is found that a strong axial magneticfield can be generated on the axis,owing to the azimuthal current induced during the reconnection of the ropes.This indicates that in the reconnection of magneticflux ropes,the energy can be transferred not only from the magneticfield to the plasma but also from the plasma current back to the magneticfield.This work opens a new avenue to the study of magneticflux ropes,which helps in understanding magnetic topology changes,and resultant magnetic energy dissipation,plasma heating,and particle acceleration found in solarflares,and magnetic confinement fusion devices.
基金supported by the National Key R&D Program of China(Grant No.2023YFA1406702)the Innovation Program for Quantum Science and Technology 2021ZD0302005+1 种基金the XPLORER Prizepartly supported by the Start-up Research Fund of Southeast University(RF1028624190)。
文摘Topological band theory has been studied for free fermions for decades,and one of the most profound physical results is the bulk-boundary correspondence.Recently a focus in topological physics is extending topological classification to mixed states.Here,we focus on Gaussian mixed states for which the modular Hamiltonians of the density matrix are quadratic free fermion models with U(1)symmetry and can be classified by topological invariants.The bulk-boundary correspondence is then manifested as stable gapless modes of the modular Hamiltonian and degenerate spectrum of the density matrix.In this article,we show that these gapless modes can be detected by the full counting statistics,mathematically described by a function introduced as F(θ).A divergent derivative atθ=πcan be used to probe the gapless modes in the modular Hamiltonian.Based on this,a topological indicator,whose quantization to unity senses topologically nontrivial mixed states,is introduced.We present the physical intuition of these results and also demonstrate these results with concrete models in both one-and two-dimensions.Our results pave the way for revealing the physical significance of topology in mixed states.
基金supported by the Liaoning Revitalization Talents Program(XLYC2203148)
文摘Dear Editor,This letter presents a joint probabilistic scheduling and resource allocation method(PSRA) for 5G-based wireless networked control systems(WNCSs). As a control-aware optimization method, PSRA minimizes the linear quadratic Gaussian(LQG) control cost of WNCSs by optimizing the activation probability of subsystems, the number of uplink repetitions, and the durations of uplink and downlink phases. Simulation results show that PSRA achieves smaller LQG control costs than existing works.
基金supported by the National Natural Science Foundation of China(No.41804141)。
文摘Energy resolution calibration is crucial for gamma-ray spectral analysis,as measured using a scintillation detector.A locally constrained regularization method was proposed to determine the resolution calibration parameters.First,a Monte Carlo simulation model consistent with an actual measurement system was constructed to obtain the energy deposition distribution in the scintillation crystal.Subsequently,the regularization objective function is established based on weighted least squares and additional constraints.Additional constraints were designed using a special weighting scheme based on the incident gamma-ray energies.Subsequently,an intelligent algorithm was introduced to search for the optimal resolution calibration parameters by minimizing the objective function.The most appropriate regularization parameter was determined through mathematical experiments.When the regularization parameter was 30,the calibrated results exhibited the minimum RMSE.Simulations and test pit experiments were conducted to verify the performance of the proposed method.The simulation results demonstrate that the proposed algorithm can determine resolution calibration parameters more accurately than the traditional weighted least squares,and the test pit experimental results show that the R-squares between the calibrated and measured spectra are larger than 0.99.The accurate resolution calibration parameters determined by the proposed method lay the foundation for gamma-ray spectral processing and simulation benchmarking.
基金supported by King Saud University,Riyadh,Saudi Arabia,through the Researchers Supporting Project under Grant RSP2025R493。
文摘With expeditious advancements in AI-driven facial manipulation techniques,particularly deepfake technology,there is growing concern over its potential misuse.Deepfakes pose a significant threat to society,partic-ularly by infringing on individuals’privacy.Amid significant endeavors to fabricate systems for identifying deepfake fabrications,existing methodologies often face hurdles in adjusting to innovative forgery techniques and demonstrate increased vulnerability to image and video clarity variations,thereby hindering their broad applicability to images and videos produced by unfamiliar technologies.In this manuscript,we endorse resilient training tactics to amplify generalization capabilities.In adversarial training,models are trained using deliberately crafted samples to deceive classification systems,thereby significantly enhancing their generalization ability.In response to this challenge,we propose an innovative hybrid adversarial training framework integrating Virtual Adversarial Training(VAT)with Two-Generated Blurred Adversarial Training.This combined framework bolsters the model’s resilience in detecting deepfakes made using unfamiliar deep learning technologies.Through such adversarial training,models are prompted to acquire more versatile attributes.Through experimental studies,we demonstrate that our model achieves higher accuracy than existing models.
文摘We apply methods of algebraic integral geometry to prove a special case of the Gaussian kinematic formula of Adler-Taylor.The idea,suggested already by Adler and Taylor,is to view the GKF as the limit of spherical kinematic formulas for spheres of large dimension N and curvature1/N.
基金project supported by the National Natural Science Foundation of China(Grant Nos.52225503 and 52405380)National Key Research and Development Program(Grant Nos.2023YFB4603303 and 2023YFB4603304)+4 种基金Key Research and Development Program of Jiangsu Province(Grant Nos.BE2022069 and BE2022069-3)National Natural Science Foundation of China for Creative Research Groups(Grant No.51921003)The 15th Batch of“Six Talents Peaks”Innovative Talents Team Program of Jiangsu province(Grant Nos.TD-GDZB-001)Shanghai Aerospace Science and Technology Innovation Fund Project(Grant No.SAST2023-066)The Fundamental Research Funds for the Central Universities(Grant Nos.NS2023035 and NP2024128)。
文摘Transpiration cooling is crucial for the performance of aerospace engine components,relying heavily on the processing quality and accuracy of microchannels.Laser powder bed fusion(LPBF)offers the potential for integrated manufacturing of complex parts and precise microchannel fabrication,essential for engine cooling applications.However,optimizing LPBF’s extensive process parameters to control processing quality and microchannel accuracy effectively remains a significant challenge,especially given the time-consuming and labor-intensive nature of handling numerous variables and the need for thorough data analysis and correlation discovery.This study introduced a combined methodology of high-throughput experiments and Gaussian process algorithms to optimize the processing quality and accuracy of nickel-based high-temperature alloy with microchannel structures.250 parameter combinations,including laser power,scanning speed,channel diameter,and spot compensation,were designed across ten high-throughput specimens.This setup allowed for rapid and efficient evaluation of processing quality and microchannel accuracy.Employing Bayesian optimization,the Gaussian process model accurately predicted processing outcomes over a broad parameter range.The correlation between various processing parameters,processing quality and accuracy was revealed,and various optimized process combinations were summarized.Verification through computed Tomography testing of the specimens confirmed the effectiveness and precision of this approach.The approach introduced in this research provides a way for quickly and efficiently optimizing the process parameters and establishing process-property relationships for LPBF,which has broad application value.
基金supported by the National Natural Science Foundation of China(No.92252201)。
文摘To address the early separation problem in the Menter Shear-Stress Transport(SST)turbulence model,a correction for the Turbulent Kinetic Energy(TKE)production term,P_(k),is introduced to account for the effect of the Adverse Pressure Gradient(APG).The correction is determined based on the distribution of Pkin the APG region before separation.When the friction coefficient C_(f) is decomposed,its direct dependence on Pkis clearly observed.However,with the introduction of Bradshaw’s assumption,Pkin the SST turbulence model is over-suppressed,resulting in a lower inner peak or no significant inner peak distribution at all.To address this problem,this paper proposes a Gaussian function,HGauss,which corrects the numerical values of P_(k) involved in the calculation of the Menter SST model by focusing on the inner peak region of P_(k).The modified SST model is then applied to four cases with APGs.The modification leads to an increase in the wall friction coefficient C_(f)in the APG region and causes a downstream shift in the separation location,improving the model’s consistency with high-accuracy data and experimental results.It is demonstrated that this correction can improve the early separation problem in the Menter SST turbulence model.
基金supported by the National Natural Science Foundation of China(Project No.62273176)“Joint Laboratory Project of Intelligent Power and Control Applications”(Project No.1003-KFA24090)the National Key Research and Development Program of China(Project No.2024YFB3311401).
文摘Lithium-ion batteries(LIBs)have been widely used in mobile energy storage systems because of their high energy density,long life,and strong environmental adaptability.Accurately estimating the state of health(SOH)for LIBs is promising and has been extensively studied for many years.However,the current prediction methods are susceptible to noise interference,and the estimation accuracy has room for improvement.Motivated by this,this paper proposes a novel battery SOH estimation method,the Beluga Whale Optimization(BWO)and Noise-Input Gaussian Process(NIGP)Stacked Model(BGNSM).This method integrates the BWO-optimized Gaussian Process Regression(GPR)with the NIGP.It combines their predictions using a stacked GPR model which reduces the problem of large input data noise and improves the prediction accuracy.The experimental results show that the BGNSM method has good accuracy,generalization ability,and robustness,and performs well in small sample situations.The Root Mean Square Error(RMSE)and Mean Absolute Error(MAE)are as low as 0.218%and 0.164%,respectively,which is close to 0.At the same time,R-Square(R^(2))is as high as 0.9948,which is close to 1,indicating that the estimated results in this paper are highly consistent with the actual results.
基金the financial support from the National Natural Science Foundation of China under Grant Nos.12174323 and 12474199Fundamental Research Funds for Central Universities of China under Grant No.20720240144(RM)111 project B16029。
文摘The elastic properties of membranes are typically characterized by a few phenomenological parameters,including bending and Gaussian curvature moduli measuring the membrane rigidity against its deformation and topological change,as well as spontaneous curvature arising from the asymmetry between the two leaflets in the lipid bilayers.Though tether-based and fluctuationbased experiments are commonly utilized to measure the bending modulus,measuring the Gaussian curvature modulus and the spontaneous curvature of the membrane is considered to be much more difficult.In this paper,we study the buckling process of a circular membrane with nonzero spontaneous curvature under compressive stresses.It is found that when the stress exceeds a critical value,the circular membrane will transform from a spherical cap to a buckled shape,with its buckling degree enhanced with the increase of stress until its base is constricted to almost zero.As the stress-strain relationship of the buckled membrane strongly depends on the Gaussian curvature modulus and the spontaneous curvature,we therefore propose a method to determine the Gaussian curvature modulus and the spontaneous curvature simultaneously by measuring its stress-strain relationship during a buckling process.
基金supported by the Glocal University 30 Project Fund of Gyeongsang National University in 2025.
文摘Adversarial attacks pose a significant threat to artificial intelligence systems by exposing them to vulnerabilities in deep learning models.Existing defense mechanisms often suffer drawbacks,such as the need for model retraining,significant inference time overhead,and limited effectiveness against specific attack types.Achieving perfect defense against adversarial attacks remains elusive,emphasizing the importance of mitigation strategies.In this study,we propose a defense mechanism that applies random cropping and Gaussian filtering to input images to mitigate the impact of adversarial attacks.First,the image was randomly cropped to vary its dimensions and then placed at the center of a fixed 299299 space,with the remaining areas filled with zero padding.Subsequently,Gaussian×filtering with a 77 kernel and a standard deviation of two was applied using a convolution operation.Finally,the×smoothed image was fed into the classification model.The proposed defense method consistently appeared in the upperright region across all attack scenarios,demonstrating its ability to preserve classification performance on clean images while significantly mitigating adversarial attacks.This visualization confirms that the proposed method is effective and reliable for defending against adversarial perturbations.Moreover,the proposed method incurs minimal computational overhead,making it suitable for real-time applications.Furthermore,owing to its model-agnostic nature,the proposed method can be easily incorporated into various neural network architectures,serving as a fundamental module for adversarial defense strategies.
