The flying-wing aircraft has excellent aerodynamic efficiency and stealth performance.However,due to the lack of tails,the flying-wing aircraft has a serious attitude control problem.In this paper,the effective flow c...The flying-wing aircraft has excellent aerodynamic efficiency and stealth performance.However,due to the lack of tails,the flying-wing aircraft has a serious attitude control problem.In this paper,the effective flow control strategy of three-axis control is proposed by using continuous jets for a flapless flying-wing aircraft.The wind tunnel test of two kinds of flying-wing models,namely one flow control model and one mechanical control model,is conducted,and the control effect is analyzed and compared.By simultaneous blowing of the circulation control actuators inboard and differential blowing of the circulation control actuators outboard,the pitch and roll controls are achieved,respectively.It also has an effective control effect at very large angles of attack where the conventional control surface fails.A linear relationship is found between the increment of the controlled aerodynamic force/moment coefficient and the momentum coefficient for circulation control actuators.Moreover,to resolve the difficulty in yaw control,a novel wingtip jet is proposed based on the concept of the all-moving tip and compared with apex jet and circulation control jet.It is found that the wingtip jet is the most efficient actuator,followed by the simultaneous-blowing circulation control jet.Therefore,based on the research above,two optimized fluidic control configurations are proposed.One employs circulation control jet and wingtip jet,and the other is completely dependent on circulation control jet.Finally,the flow control mechanism of circulation control is discussed.Circulation control significantly accelerates the flow on the upper surface of the airfoil in attached flow and reduces the flow separation region in separated flow,leading to aerodynamic performance improvement.These results provide an important theoretic basis for the flapless flight control of flying-wing aircraft.展开更多
Modern intelligent systems,such as autonomous vehicles and face recognition,must continuously adapt to new scenarios while preserving their ability to handle previously encountered situations.However,when neural netwo...Modern intelligent systems,such as autonomous vehicles and face recognition,must continuously adapt to new scenarios while preserving their ability to handle previously encountered situations.However,when neural networks learn new classes sequentially,they suffer from catastrophic forgetting—the tendency to lose knowledge of earlier classes.This challenge,which lies at the core of class-incremental learning,severely limits the deployment of continual learning systems in real-world applications with streaming data.Existing approaches,including rehearsalbased methods and knowledge distillation techniques,have attempted to address this issue but often struggle to effectively preserve decision boundaries and discriminative features under limited memory constraints.To overcome these limitations,we propose a support vector-guided framework for class-incremental learning.The framework integrates an enhanced feature extractor with a Support Vector Machine classifier,which generates boundary-critical support vectors to guide both replay and distillation.Building on this architecture,we design a joint feature retention strategy that combines boundary proximity with feature diversity,and a Support Vector Distillation Loss that enforces dual alignment in decision and semantic spaces.In addition,triple attention modules are incorporated into the feature extractor to enhance representation power.Extensive experiments on CIFAR-100 and Tiny-ImageNet demonstrate effective improvements.On CIFAR-100 and Tiny-ImageNet with 5 tasks,our method achieves 71.68%and 58.61%average accuracy,outperforming strong baselines by 3.34%and 2.05%.These advantages are consistently observed across different task splits,highlighting the robustness and generalization of the proposed approach.Beyond benchmark evaluations,the framework also shows potential in few-shot and resource-constrained applications such as edge computing and mobile robotics.展开更多
The development of clinical candidates that modify the natural progression of sporadic Parkinson's disease and related synucleinopathies is a praiseworthy endeavor,but extremely challenging.Therapeutic candidates ...The development of clinical candidates that modify the natural progression of sporadic Parkinson's disease and related synucleinopathies is a praiseworthy endeavor,but extremely challenging.Therapeutic candidates that were successful in preclinical Parkinson's disease animal models have repeatedly failed when tested in clinical trials.While these failures have many possible explanations,it is perhaps time to recognize that the problem lies with the animal models rather than the putative candidate.In other words,the lack of adequate animal models of Parkinson's disease currently represents the main barrier to preclinical identification of potential disease-modifying therapies likely to succeed in clinical trials.However,this barrier may be overcome by the recent introduction of novel generations of viral vectors coding for different forms of alpha-synuclein species and related genes.Although still facing several limitations,these models have managed to mimic the known neuropathological hallmarks of Parkinson's disease with unprecedented accuracy,delineating a more optimistic scenario for the near future.展开更多
Fluidic Thrust Vectoring(FTV)is used for the yaw attitude control of tailless flying wing,which can significantly improve stealth performance,maneuverability and lateral/heading maneuverability.The FTV control scheme ...Fluidic Thrust Vectoring(FTV)is used for the yaw attitude control of tailless flying wing,which can significantly improve stealth performance,maneuverability and lateral/heading maneuverability.The FTV control scheme of co-directional secondary flow was designed based on a 30 kgf thrust turbojet engine,an equivalent rudder deflection control variable of Mass Flow Combination(MFC)was proposed,and a control model was established to form a FTV control system scheme,which was integrated with the flight control system of a 100 kg tailless flying wing with medium aspect ratio to achieve closed-loop control of the yaw attitude based on FTV.The heading stability augmentation and maneuvering control characteristics and time response characteristics of tailless flying wing by FTV were quantitatively studied through virtual flight test in a wind tunnel at a wind speed of 35 m/s.The results show that the control strategy based on MFC achieves bidirectional continuous and stable control of thrust vector angle in a range of±11°,and the thrust vector angle varies monotonically with MFC;the co-directional FTV realizes bidirectional continuous and stable control of the yaw attitude of tailless flying wing,without longitudinal/lateral coupling moment.The increment of the maximum yawing moment coefficient is 0.0029,the maximum yaw rate is 7.55(°)/s,and the response time of the yaw rate of the vectoring nozzle actuated by the secondary flow is about 0.06 s,which satisfies the heading stability augmentation and maneuvering control response requirements of the aircraft with statically unstable heading,and provides new control means for the heading rudderless attitude control of tailless flying wing.展开更多
The advantages of genome selection(GS) in animal and plant breeding are self-evident.Traditional parametric models have disadvantage in better fit the increasingly large sequencing data and capture complex effects acc...The advantages of genome selection(GS) in animal and plant breeding are self-evident.Traditional parametric models have disadvantage in better fit the increasingly large sequencing data and capture complex effects accurately.Machine learning models have demonstrated remarkable potential in addressing these challenges.In this study,we introduced the concept of mixed kernel functions to explore the performance of support vector machine regression(SVR) in GS.Six single kernel functions(SVR_L,SVR_C,SVR_G,SVR_P,SVR_S,SVR_L) and four mixed kernel functions(SVR_GS,SVR_GP,SVR_LS,SVR_LP) were used to predict genome breeding values.The prediction accuracy,mean squared error(MSE) and mean absolute error(MAE) were used as evaluation indicators to compare with two traditional parametric models(GBLUP,BayesB) and two popular machine learning models(RF,KcRR).The results indicate that in most cases,the performance of the mixed kernel function model significantly outperforms that of GBLUP,BayesB and single kernel function.For instance,for T1 in the pig dataset,the predictive accuracy of SVR_GS is improved by 10% compared to GBLUP,and by approximately 4.4 and 18.6% compared to SVR_G and SVR_S respectively.For E1 in the wheat dataset,SVR_GS achieves 13.3% higher prediction accuracy than GBLUP.Among single kernel functions,the Laplacian and Gaussian kernel functions yield similar results,with the Gaussian kernel function performing better.The mixed kernel function notably reduces the MSE and MAE when compared to all single kernel functions.Furthermore,regarding runtime,SVR_GS and SVR_GP mixed kernel functions run approximately three times faster than GBLUP in the pig dataset,with only a slight increase in runtime compared to the single kernel function model.In summary,the mixed kernel function model of SVR demonstrates speed and accuracy competitiveness,and the model such as SVR_GS has important application potential for GS.展开更多
Parkinson’s disease(PD)is a debilitating neurological disorder affecting over 10 million people worldwide.PD classification models using voice signals as input are common in the literature.It is believed that using d...Parkinson’s disease(PD)is a debilitating neurological disorder affecting over 10 million people worldwide.PD classification models using voice signals as input are common in the literature.It is believed that using deep learning algorithms further enhances performance;nevertheless,it is challenging due to the nature of small-scale and imbalanced PD datasets.This paper proposed a convolutional neural network-based deep support vector machine(CNN-DSVM)to automate the feature extraction process using CNN and extend the conventional SVM to a DSVM for better classification performance in small-scale PD datasets.A customized kernel function reduces the impact of biased classification towards the majority class(healthy candidates in our consideration).An improved generative adversarial network(IGAN)was designed to generate additional training data to enhance the model’s performance.For performance evaluation,the proposed algorithm achieves a sensitivity of 97.6%and a specificity of 97.3%.The performance comparison is evaluated from five perspectives,including comparisons with different data generation algorithms,feature extraction techniques,kernel functions,and existing works.Results reveal the effectiveness of the IGAN algorithm,which improves the sensitivity and specificity by 4.05%–4.72%and 4.96%–5.86%,respectively;and the effectiveness of the CNN-DSVM algorithm,which improves the sensitivity by 1.24%–57.4%and specificity by 1.04%–163%and reduces biased detection towards the majority class.The ablation experiments confirm the effectiveness of individual components.Two future research directions have also been suggested.展开更多
Due to elimination of horizontal and vertical tails,flying wing aircraft has poor longitudinal and directional dynamic characteristics.In addition,flying wing aircraft uses drag rudders for yaw control,which tends to ...Due to elimination of horizontal and vertical tails,flying wing aircraft has poor longitudinal and directional dynamic characteristics.In addition,flying wing aircraft uses drag rudders for yaw control,which tends to generate strong three-axis control coupling.To overcome these problems,a flight control law design method that couples the longitudinal axis with the lateraldirectional axes is proposed.First,the three-axis coupled control augmentation structure is specified.In the structure,a‘‘soft/hard"cross-connection method is developed for three-axis dynamic decoupling and longitudinal control response decoupling from the drag rudders;maneuvering turn angular rate estimation and subtraction are used in the yaw axis to improve the directional damping.Besides,feedforward control is adopted to improve the maneuverability and control decoupling performance.Then,detailed design methods for feedback and feedforward control parameters are established using eigenstructure assignment and model following technique.Finally,the proposed design method is evaluated and compared with conventional method by numeric simulations.The influences of control derivatives variation of drag rudders on the method are also analyzed.It is demonstrated that the method can effectively improve the dynamic characteristics of flying wing aircraft,especially the directional damping characteristics,and decouple the longitudinal responses from the drag rudders.展开更多
In order to measure three-axis intersection error, two crosshair targets were fixed in the inner axis frame of a three-axis turntable. Also a theodolite was used to point its telescope to the targets and to measure th...In order to measure three-axis intersection error, two crosshair targets were fixed in the inner axis frame of a three-axis turntable. Also a theodolite was used to point its telescope to the targets and to measure the horizontal angles when three axes were on equi-spaced angle positions. The calculation equations of the axis intersection were deduced from the mounting position of the theodolite, positions of two targets, angular positions of three axes, and the measured horizontal angles with the theodolite. Finally, a practical measurement is carried out on a horizontal three-axis turntable and error analysis is conducted.展开更多
The most challenging problem of navigation in three-axis stabilized geostationary satellite is accurate calculation of misalignment angles, deduced by orbit measurement error, attitude measurement error, thermal elast...The most challenging problem of navigation in three-axis stabilized geostationary satellite is accurate calculation of misalignment angles, deduced by orbit measurement error, attitude measurement error, thermal elastic deformation, time synchronization error, and so on. Before the satellite is launched, the misalignment model must be established and validated. But there were no observation data, which is a non-negligible risk of yielding the greatest returns on investment. On the basis of misalignment modeling using landmarks and stars, which is not available between different organizations and is developed by ourselves, experimental data are constructed to validate the navigation processing flow as well as misalignment calculation accuracy. In the condition of using landmarks, the maximum misalignment calculation errors of roll, pitch, and yaw axis are 2, 2, and 104 micro radians, respectively, without considering the accuracy of image edge detection. While in the condition of using stars, the maximum errors of roll, pitch, and yaw axis are 1, 1, and 3 micro radians, respectively, without considering the accuracy of star center extraction. Results are rather encouraging, which pave the way for high-accuracy image navigation of three-axis stabilized geostationary satellite. The misalignment modeling as well as calculation method has been used in the new generation of geostationary meteorological satellite in China, FY-4 series, the first satellite of which was launched at the end of 2016.展开更多
Angular velocity stabilization control and attitude stabilization control for an underactuated spacecraft using only two single gimbal control moment gyros （SGCMGs） as actuators is investigated. First of all, the dy...Angular velocity stabilization control and attitude stabilization control for an underactuated spacecraft using only two single gimbal control moment gyros （SGCMGs） as actuators is investigated. First of all, the dynamic model of the underactuated spacecraft is established and the singularity of different configurations with the two SGCMGs is analyzed. Under the assumption that the gimbal axes of the two SGCMGs are installed in any direction, and that the total system angular momentum is not zero, a state feedback control law via Lyapunov method is designed to globally asymptotically stabilize the angular velocity of spacecraft. Under the assumption that the gimbal axes of the two SGCMGs are coaxially installed along anyone of the three principal axes of spacecraft inertia, and that the total system angular momentum is zero, a discontinuous state feedback control law is designed to stabilize three-axis attitude of spacecraft with respect to the inertial frame. Furthermore, the singularity escape of SGCMGs for the above two control problems is also studied. Simulation results demonstrate the validity of the control laws.展开更多
A novel hybrid robust three-axis attitude control approach,namely HRTAC,is considered along with the well-known developments in the area of space systems,since there is a consensus among the related experts that the n...A novel hybrid robust three-axis attitude control approach,namely HRTAC,is considered along with the well-known developments in the area of space systems,since there is a consensus among the related experts that the new insights may be taken into account as decision points to outperform the available materials.It is to note that the traditional control approaches may generally be upgraded,as long as a number of modifications are made with respect to state-of-the-art,in order to propose high-precision outcomes.Regarding the investigated issues,the robust sliding mode finite-time control approach is first designed to handle three-axis angular rates in the inner control loop,which consists of the pulse width pulse frequency modulations in line with the control allocation scheme and the system dynamics.The main subject to employ these modulations that is realizing in association with the control allocation scheme is to be able to handle a class of overactuated systems,in particular.The proportional derivative based linear quadratic regulator approach is then designed to handle three-axis rotational angles in the outer control loop,which consists of the system kinematics that is correspondingly concentrated to deal with the quaternion based model.The utilization of the linear and its nonlinear terms,simultaneously,are taken into real consideration as the research motivation,while the performance results are of the significance as the improved version in comparison with the recent investigated outcomes.Subsequently,there is a stability analysis to verify and guarantee the closed loop system performance in coping with the whole of nominal referenced commands.At the end,the effectiveness of the approach considered here is highlighted in line with a number of potential recent benchmarks.展开更多
For the petroleum industry, to reduce the risk of a gas explosion in dangerous working areas, the use of explosion-proof equipment such as air-driven devices which are free from explosions becomes essential. Moreover,...For the petroleum industry, to reduce the risk of a gas explosion in dangerous working areas, the use of explosion-proof equipment such as air-driven devices which are free from explosions becomes essential. Moreover, for the purpose of saving manpower, a remote operation using a robot via a visual monitoring system and a network is used. However, to overcome the drawback of costly manpower and to improve safety in explosion-prone zones, a three-axis robot using a remote network control system is proposed. In this paper, the three-axis robot can be monitored online via the USB protocol. Furthermore, it also can be remotely manipulated via the TCP/IP protocol by clicking the command of the VB interface on the client pc. Consequently, the remote-control three-axis robot can not only work for people in severe and dangerous circumstances but also can reduce the cost of manpower.展开更多
In order to meet tracking performance index of three-axis hydraulic simulator, based on classical quantitative feedback theory （QFT）, an improved QFT technique is used to synthesize controller of low gain and bandwi...In order to meet tracking performance index of three-axis hydraulic simulator, based on classical quantitative feedback theory （QFT）, an improved QFT technique is used to synthesize controller of low gain and bandwidth. By choosing a special nominal plant, the improved method assigns relative magnitude and phase tracking error between system uncertainty and nominal control plant. Relative tracking error induced by system uncertainty is transformed into sensitivity problem and relative tracking error induced by nominal plant forms into a region on Nichols chart. The two constraints further form into a combined bound which is fit for magnitude and phase loop shaping. Because of leaving out pre-filter of classical QFT controller structure, tracking performance is enhanced greatly. Furthermore, a cascaded two-loop control strategy is proposed to heighten control effect. The improved technique＇s efficacy is validated by simulation and experiment results.展开更多
Accurately estimating the State of Health(SOH)and Remaining Useful Life(RUL)of lithium-ion batteries(LIBs)is crucial for the continuous and stable operation of battery management systems.However,due to the complex int...Accurately estimating the State of Health(SOH)and Remaining Useful Life(RUL)of lithium-ion batteries(LIBs)is crucial for the continuous and stable operation of battery management systems.However,due to the complex internal chemical systems of LIBs and the nonlinear degradation of their performance,direct measurement of SOH and RUL is challenging.To address these issues,the Twin Support Vector Machine(TWSVM)method is proposed to predict SOH and RUL.Initially,the constant current charging time of the lithium battery is extracted as a health indicator(HI),decomposed using Variational Modal Decomposition(VMD),and feature correlations are computed using Importance of Random Forest Features(RF)to maximize the extraction of critical factors influencing battery performance degradation.Furthermore,to enhance the global search capability of the Convolution Optimization Algorithm(COA),improvements are made using Good Point Set theory and the Differential Evolution method.The Improved Convolution Optimization Algorithm(ICOA)is employed to optimize TWSVM parameters for constructing SOH and RUL prediction models.Finally,the proposed models are validated using NASA and CALCE lithium-ion battery datasets.Experimental results demonstrate that the proposed models achieve an RMSE not exceeding 0.007 and an MAPE not exceeding 0.0082 for SOH and RUL prediction,with a relative error in RUL prediction within the range of[-1.8%,2%].Compared to other models,the proposed model not only exhibits superior fitting capability but also demonstrates robust performance.展开更多
Thrust-vectoring capability has become a critical feature for propulsion systems as space missions move from static to dynamic.Thrust-vectoring is a well-developed area of rocket engine science.For electric propulsion...Thrust-vectoring capability has become a critical feature for propulsion systems as space missions move from static to dynamic.Thrust-vectoring is a well-developed area of rocket engine science.For electric propulsion,however,it is an evolving field that has taken a new leap forward in recent years.A review and analysis of thrust-vectoring schemes for electric propulsion systems have been conducted.The scope of this review includes thrust-vectoring schemes that can be implemented for electrostatic,electromagnetic,and beam-driven thrusters.A classification of electric propulsion schemes that provide thrust-vectoring capability is developed.More attention is given to schemes implemented in laboratory prototypes and flight models.The final part is devoted to a discussion on the suitability of different electric propulsion systems with thrust-vectoring capability for modern space mission operations.The thrust-vectoring capability of electric propulsion is necessary for inner and outer space satellites,which are at a disadvantage with conventional unidirectional propulsion systems due to their limited maneuverability.展开更多
The three-axis servo system with the core of gyro stabilization is the foundation to realize its function, and a key technology of the seeker devolopment. In order to reduce the costs, improve the efficiency of resear...The three-axis servo system with the core of gyro stabilization is the foundation to realize its function, and a key technology of the seeker devolopment. In order to reduce the costs, improve the efficiency of research and devolopment, a new method that instead of physical prototype by virtual prototype was proposed. Adams and MATLAB/simulink are used to establish the mechanical dynamics model and controller model of the three-axis servo system. The simulation data which was processed and analyzed is compared with test data, to determine the control parameters of the virtual prototype and improve the accuracy of the model, and test the multiple condition simulation,which can provide a reference for practical production.The simulation results verify the feasibility of the models.展开更多
Vector winds play a crucial role in weather and climate,as well as the effective utilization of wind energy resources.However,limited research has been conducted on treating the wind field as a vector field in the eva...Vector winds play a crucial role in weather and climate,as well as the effective utilization of wind energy resources.However,limited research has been conducted on treating the wind field as a vector field in the evaluation of numerical weather prediction models.In this study,the authors treat vector winds as a whole by employing a vector field evaluation method,and evaluate the mesoscale model of the China Meteorological Administration(CMA-MESO)and ECMWF forecast,with reference to ERA5 reanalysis,in terms of multiple aspects of vector winds over eastern China in 2022.The results show that the ECMWF forecast is superior to CMA-MESO in predicting the spatial distribution and intensity of 10-m vector winds.Both models overestimate the wind speed in East China,and CMA-MESO overestimates the wind speed to a greater extent.The forecasting skill of the vector wind field in both models decreases with increasing lead time.The forecasting skill of CMA-MESO fluctuates more and decreases faster than that of the ECMWF forecast.There is a significant negative correlation between the model vector wind forecasting skill and terrain height.This study provides a scientific evaluation of the local application of vector wind forecasts of the CMA-MESO model and ECMWF forecast.展开更多
Open networks and heterogeneous services in the Internet of Vehicles(IoV)can lead to security and privacy challenges.One key requirement for such systems is the preservation of user privacy,ensuring a seamless experie...Open networks and heterogeneous services in the Internet of Vehicles(IoV)can lead to security and privacy challenges.One key requirement for such systems is the preservation of user privacy,ensuring a seamless experience in driving,navigation,and communication.These privacy needs are influenced by various factors,such as data collected at different intervals,trip durations,and user interactions.To address this,the paper proposes a Support Vector Machine(SVM)model designed to process large amounts of aggregated data and recommend privacy preserving measures.The model analyzes data based on user demands and interactions with service providers or neighboring infrastructure.It aims to minimize privacy risks while ensuring service continuity and sustainability.The SVMmodel helps validate the system’s reliability by creating a hyperplane that distinguishes between maximum and minimum privacy recommendations.The results demonstrate the effectiveness of the proposed SVM model in enhancing both privacy and service performance.展开更多
The Macao Science Satellite-1(known as MSS-1)is the first scientific exploration satellite that was designed to measure the Earth's low latitude magnetic field at high resolution and with high precision by collect...The Macao Science Satellite-1(known as MSS-1)is the first scientific exploration satellite that was designed to measure the Earth's low latitude magnetic field at high resolution and with high precision by collecting data in a near-equatorial orbit.Magnetic field data from MSS-1's onboard Vector Fluxgate Magnetometer(VFM),collected at a sample rate of 50 Hz,allows us to detect and investigate sources of magnetic data contamination,from DC to relevant Nyquist frequency.Here we report two types of artificial disturbances in the VFM data.One is V-shaped events concentrated at night,with frequencies sweeping from the Nyquist frequency down to zero and back up.The other is 5-Hz events(ones that exhibit a distinct 5 Hz spectrum peak);these events are always accompanied by intervals of spiky signals,and are clearly related to the attitude control of the satellite.Our analyses show that VFM noise levels in daytime are systematically lower than in nighttime.The daily average noise levels exhibit a period of about 52 days.The V-shaped events are strongly correlated with higher VFM noise levels.展开更多
文摘The flying-wing aircraft has excellent aerodynamic efficiency and stealth performance.However,due to the lack of tails,the flying-wing aircraft has a serious attitude control problem.In this paper,the effective flow control strategy of three-axis control is proposed by using continuous jets for a flapless flying-wing aircraft.The wind tunnel test of two kinds of flying-wing models,namely one flow control model and one mechanical control model,is conducted,and the control effect is analyzed and compared.By simultaneous blowing of the circulation control actuators inboard and differential blowing of the circulation control actuators outboard,the pitch and roll controls are achieved,respectively.It also has an effective control effect at very large angles of attack where the conventional control surface fails.A linear relationship is found between the increment of the controlled aerodynamic force/moment coefficient and the momentum coefficient for circulation control actuators.Moreover,to resolve the difficulty in yaw control,a novel wingtip jet is proposed based on the concept of the all-moving tip and compared with apex jet and circulation control jet.It is found that the wingtip jet is the most efficient actuator,followed by the simultaneous-blowing circulation control jet.Therefore,based on the research above,two optimized fluidic control configurations are proposed.One employs circulation control jet and wingtip jet,and the other is completely dependent on circulation control jet.Finally,the flow control mechanism of circulation control is discussed.Circulation control significantly accelerates the flow on the upper surface of the airfoil in attached flow and reduces the flow separation region in separated flow,leading to aerodynamic performance improvement.These results provide an important theoretic basis for the flapless flight control of flying-wing aircraft.
基金supported by the Gansu Provincial Natural Science Foundation(grant number 25JRRA074)the Gansu Provincial Key R&D Science and Technology Program(grant number 24YFGA060)the National Natural Science Foundation of China(grant number 62161019).
文摘Modern intelligent systems,such as autonomous vehicles and face recognition,must continuously adapt to new scenarios while preserving their ability to handle previously encountered situations.However,when neural networks learn new classes sequentially,they suffer from catastrophic forgetting—the tendency to lose knowledge of earlier classes.This challenge,which lies at the core of class-incremental learning,severely limits the deployment of continual learning systems in real-world applications with streaming data.Existing approaches,including rehearsalbased methods and knowledge distillation techniques,have attempted to address this issue but often struggle to effectively preserve decision boundaries and discriminative features under limited memory constraints.To overcome these limitations,we propose a support vector-guided framework for class-incremental learning.The framework integrates an enhanced feature extractor with a Support Vector Machine classifier,which generates boundary-critical support vectors to guide both replay and distillation.Building on this architecture,we design a joint feature retention strategy that combines boundary proximity with feature diversity,and a Support Vector Distillation Loss that enforces dual alignment in decision and semantic spaces.In addition,triple attention modules are incorporated into the feature extractor to enhance representation power.Extensive experiments on CIFAR-100 and Tiny-ImageNet demonstrate effective improvements.On CIFAR-100 and Tiny-ImageNet with 5 tasks,our method achieves 71.68%and 58.61%average accuracy,outperforming strong baselines by 3.34%and 2.05%.These advantages are consistently observed across different task splits,highlighting the robustness and generalization of the proposed approach.Beyond benchmark evaluations,the framework also shows potential in few-shot and resource-constrained applications such as edge computing and mobile robotics.
基金supported by grants PID2020-120308RB-I00 and PID2023-147802OB-I00 funded by MICIU/AEI/10.13039/501100011033FEDER,UE,by Aligning Science Across Parkinson’s(ref.ASAP-020505)through the Michael J.Fox Foundation for Parkinson’s Research+1 种基金by CiberNed Intramural Collaborative Projects(ref.PI2020/09)by the Spanish Fundación Mutua Madrile?a de Investigación Médica(to JLL)。
文摘The development of clinical candidates that modify the natural progression of sporadic Parkinson's disease and related synucleinopathies is a praiseworthy endeavor,but extremely challenging.Therapeutic candidates that were successful in preclinical Parkinson's disease animal models have repeatedly failed when tested in clinical trials.While these failures have many possible explanations,it is perhaps time to recognize that the problem lies with the animal models rather than the putative candidate.In other words,the lack of adequate animal models of Parkinson's disease currently represents the main barrier to preclinical identification of potential disease-modifying therapies likely to succeed in clinical trials.However,this barrier may be overcome by the recent introduction of novel generations of viral vectors coding for different forms of alpha-synuclein species and related genes.Although still facing several limitations,these models have managed to mimic the known neuropathological hallmarks of Parkinson's disease with unprecedented accuracy,delineating a more optimistic scenario for the near future.
文摘Fluidic Thrust Vectoring(FTV)is used for the yaw attitude control of tailless flying wing,which can significantly improve stealth performance,maneuverability and lateral/heading maneuverability.The FTV control scheme of co-directional secondary flow was designed based on a 30 kgf thrust turbojet engine,an equivalent rudder deflection control variable of Mass Flow Combination(MFC)was proposed,and a control model was established to form a FTV control system scheme,which was integrated with the flight control system of a 100 kg tailless flying wing with medium aspect ratio to achieve closed-loop control of the yaw attitude based on FTV.The heading stability augmentation and maneuvering control characteristics and time response characteristics of tailless flying wing by FTV were quantitatively studied through virtual flight test in a wind tunnel at a wind speed of 35 m/s.The results show that the control strategy based on MFC achieves bidirectional continuous and stable control of thrust vector angle in a range of±11°,and the thrust vector angle varies monotonically with MFC;the co-directional FTV realizes bidirectional continuous and stable control of the yaw attitude of tailless flying wing,without longitudinal/lateral coupling moment.The increment of the maximum yawing moment coefficient is 0.0029,the maximum yaw rate is 7.55(°)/s,and the response time of the yaw rate of the vectoring nozzle actuated by the secondary flow is about 0.06 s,which satisfies the heading stability augmentation and maneuvering control response requirements of the aircraft with statically unstable heading,and provides new control means for the heading rudderless attitude control of tailless flying wing.
基金supported by the China Agriculture Research System of MOF and MARAthe National Natural Science Foundation of China (31872337 and 31501919)the Agricultural Science and Technology Innovation Project,China (ASTIP-IAS02)。
文摘The advantages of genome selection(GS) in animal and plant breeding are self-evident.Traditional parametric models have disadvantage in better fit the increasingly large sequencing data and capture complex effects accurately.Machine learning models have demonstrated remarkable potential in addressing these challenges.In this study,we introduced the concept of mixed kernel functions to explore the performance of support vector machine regression(SVR) in GS.Six single kernel functions(SVR_L,SVR_C,SVR_G,SVR_P,SVR_S,SVR_L) and four mixed kernel functions(SVR_GS,SVR_GP,SVR_LS,SVR_LP) were used to predict genome breeding values.The prediction accuracy,mean squared error(MSE) and mean absolute error(MAE) were used as evaluation indicators to compare with two traditional parametric models(GBLUP,BayesB) and two popular machine learning models(RF,KcRR).The results indicate that in most cases,the performance of the mixed kernel function model significantly outperforms that of GBLUP,BayesB and single kernel function.For instance,for T1 in the pig dataset,the predictive accuracy of SVR_GS is improved by 10% compared to GBLUP,and by approximately 4.4 and 18.6% compared to SVR_G and SVR_S respectively.For E1 in the wheat dataset,SVR_GS achieves 13.3% higher prediction accuracy than GBLUP.Among single kernel functions,the Laplacian and Gaussian kernel functions yield similar results,with the Gaussian kernel function performing better.The mixed kernel function notably reduces the MSE and MAE when compared to all single kernel functions.Furthermore,regarding runtime,SVR_GS and SVR_GP mixed kernel functions run approximately three times faster than GBLUP in the pig dataset,with only a slight increase in runtime compared to the single kernel function model.In summary,the mixed kernel function model of SVR demonstrates speed and accuracy competitiveness,and the model such as SVR_GS has important application potential for GS.
基金The work described in this paper was fully supported by a grant from Hong Kong Metropolitan University(RIF/2021/05).
文摘Parkinson’s disease(PD)is a debilitating neurological disorder affecting over 10 million people worldwide.PD classification models using voice signals as input are common in the literature.It is believed that using deep learning algorithms further enhances performance;nevertheless,it is challenging due to the nature of small-scale and imbalanced PD datasets.This paper proposed a convolutional neural network-based deep support vector machine(CNN-DSVM)to automate the feature extraction process using CNN and extend the conventional SVM to a DSVM for better classification performance in small-scale PD datasets.A customized kernel function reduces the impact of biased classification towards the majority class(healthy candidates in our consideration).An improved generative adversarial network(IGAN)was designed to generate additional training data to enhance the model’s performance.For performance evaluation,the proposed algorithm achieves a sensitivity of 97.6%and a specificity of 97.3%.The performance comparison is evaluated from five perspectives,including comparisons with different data generation algorithms,feature extraction techniques,kernel functions,and existing works.Results reveal the effectiveness of the IGAN algorithm,which improves the sensitivity and specificity by 4.05%–4.72%and 4.96%–5.86%,respectively;and the effectiveness of the CNN-DSVM algorithm,which improves the sensitivity by 1.24%–57.4%and specificity by 1.04%–163%and reduces biased detection towards the majority class.The ablation experiments confirm the effectiveness of individual components.Two future research directions have also been suggested.
基金supported by the Fundamental Research Funds for the Central Universities of China(No.:YWF-19-BJ-J-322)。
文摘Due to elimination of horizontal and vertical tails,flying wing aircraft has poor longitudinal and directional dynamic characteristics.In addition,flying wing aircraft uses drag rudders for yaw control,which tends to generate strong three-axis control coupling.To overcome these problems,a flight control law design method that couples the longitudinal axis with the lateraldirectional axes is proposed.First,the three-axis coupled control augmentation structure is specified.In the structure,a‘‘soft/hard"cross-connection method is developed for three-axis dynamic decoupling and longitudinal control response decoupling from the drag rudders;maneuvering turn angular rate estimation and subtraction are used in the yaw axis to improve the directional damping.Besides,feedforward control is adopted to improve the maneuverability and control decoupling performance.Then,detailed design methods for feedback and feedforward control parameters are established using eigenstructure assignment and model following technique.Finally,the proposed design method is evaluated and compared with conventional method by numeric simulations.The influences of control derivatives variation of drag rudders on the method are also analyzed.It is demonstrated that the method can effectively improve the dynamic characteristics of flying wing aircraft,especially the directional damping characteristics,and decouple the longitudinal responses from the drag rudders.
文摘In order to measure three-axis intersection error, two crosshair targets were fixed in the inner axis frame of a three-axis turntable. Also a theodolite was used to point its telescope to the targets and to measure the horizontal angles when three axes were on equi-spaced angle positions. The calculation equations of the axis intersection were deduced from the mounting position of the theodolite, positions of two targets, angular positions of three axes, and the measured horizontal angles with the theodolite. Finally, a practical measurement is carried out on a horizontal three-axis turntable and error analysis is conducted.
文摘The most challenging problem of navigation in three-axis stabilized geostationary satellite is accurate calculation of misalignment angles, deduced by orbit measurement error, attitude measurement error, thermal elastic deformation, time synchronization error, and so on. Before the satellite is launched, the misalignment model must be established and validated. But there were no observation data, which is a non-negligible risk of yielding the greatest returns on investment. On the basis of misalignment modeling using landmarks and stars, which is not available between different organizations and is developed by ourselves, experimental data are constructed to validate the navigation processing flow as well as misalignment calculation accuracy. In the condition of using landmarks, the maximum misalignment calculation errors of roll, pitch, and yaw axis are 2, 2, and 104 micro radians, respectively, without considering the accuracy of image edge detection. While in the condition of using stars, the maximum errors of roll, pitch, and yaw axis are 1, 1, and 3 micro radians, respectively, without considering the accuracy of star center extraction. Results are rather encouraging, which pave the way for high-accuracy image navigation of three-axis stabilized geostationary satellite. The misalignment modeling as well as calculation method has been used in the new generation of geostationary meteorological satellite in China, FY-4 series, the first satellite of which was launched at the end of 2016.
文摘Angular velocity stabilization control and attitude stabilization control for an underactuated spacecraft using only two single gimbal control moment gyros （SGCMGs） as actuators is investigated. First of all, the dynamic model of the underactuated spacecraft is established and the singularity of different configurations with the two SGCMGs is analyzed. Under the assumption that the gimbal axes of the two SGCMGs are installed in any direction, and that the total system angular momentum is not zero, a state feedback control law via Lyapunov method is designed to globally asymptotically stabilize the angular velocity of spacecraft. Under the assumption that the gimbal axes of the two SGCMGs are coaxially installed along anyone of the three principal axes of spacecraft inertia, and that the total system angular momentum is zero, a discontinuous state feedback control law is designed to stabilize three-axis attitude of spacecraft with respect to the inertial frame. Furthermore, the singularity escape of SGCMGs for the above two control problems is also studied. Simulation results demonstrate the validity of the control laws.
文摘A novel hybrid robust three-axis attitude control approach,namely HRTAC,is considered along with the well-known developments in the area of space systems,since there is a consensus among the related experts that the new insights may be taken into account as decision points to outperform the available materials.It is to note that the traditional control approaches may generally be upgraded,as long as a number of modifications are made with respect to state-of-the-art,in order to propose high-precision outcomes.Regarding the investigated issues,the robust sliding mode finite-time control approach is first designed to handle three-axis angular rates in the inner control loop,which consists of the pulse width pulse frequency modulations in line with the control allocation scheme and the system dynamics.The main subject to employ these modulations that is realizing in association with the control allocation scheme is to be able to handle a class of overactuated systems,in particular.The proportional derivative based linear quadratic regulator approach is then designed to handle three-axis rotational angles in the outer control loop,which consists of the system kinematics that is correspondingly concentrated to deal with the quaternion based model.The utilization of the linear and its nonlinear terms,simultaneously,are taken into real consideration as the research motivation,while the performance results are of the significance as the improved version in comparison with the recent investigated outcomes.Subsequently,there is a stability analysis to verify and guarantee the closed loop system performance in coping with the whole of nominal referenced commands.At the end,the effectiveness of the approach considered here is highlighted in line with a number of potential recent benchmarks.
文摘For the petroleum industry, to reduce the risk of a gas explosion in dangerous working areas, the use of explosion-proof equipment such as air-driven devices which are free from explosions becomes essential. Moreover, for the purpose of saving manpower, a remote operation using a robot via a visual monitoring system and a network is used. However, to overcome the drawback of costly manpower and to improve safety in explosion-prone zones, a three-axis robot using a remote network control system is proposed. In this paper, the three-axis robot can be monitored online via the USB protocol. Furthermore, it also can be remotely manipulated via the TCP/IP protocol by clicking the command of the VB interface on the client pc. Consequently, the remote-control three-axis robot can not only work for people in severe and dangerous circumstances but also can reduce the cost of manpower.
文摘In order to meet tracking performance index of three-axis hydraulic simulator, based on classical quantitative feedback theory （QFT）, an improved QFT technique is used to synthesize controller of low gain and bandwidth. By choosing a special nominal plant, the improved method assigns relative magnitude and phase tracking error between system uncertainty and nominal control plant. Relative tracking error induced by system uncertainty is transformed into sensitivity problem and relative tracking error induced by nominal plant forms into a region on Nichols chart. The two constraints further form into a combined bound which is fit for magnitude and phase loop shaping. Because of leaving out pre-filter of classical QFT controller structure, tracking performance is enhanced greatly. Furthermore, a cascaded two-loop control strategy is proposed to heighten control effect. The improved technique＇s efficacy is validated by simulation and experiment results.
基金funded by the Pyramid Talent Training Project of Beijing University of Civil Engineering and Architecture under Grant GJZJ20220802。
文摘Accurately estimating the State of Health(SOH)and Remaining Useful Life(RUL)of lithium-ion batteries(LIBs)is crucial for the continuous and stable operation of battery management systems.However,due to the complex internal chemical systems of LIBs and the nonlinear degradation of their performance,direct measurement of SOH and RUL is challenging.To address these issues,the Twin Support Vector Machine(TWSVM)method is proposed to predict SOH and RUL.Initially,the constant current charging time of the lithium battery is extracted as a health indicator(HI),decomposed using Variational Modal Decomposition(VMD),and feature correlations are computed using Importance of Random Forest Features(RF)to maximize the extraction of critical factors influencing battery performance degradation.Furthermore,to enhance the global search capability of the Convolution Optimization Algorithm(COA),improvements are made using Good Point Set theory and the Differential Evolution method.The Improved Convolution Optimization Algorithm(ICOA)is employed to optimize TWSVM parameters for constructing SOH and RUL prediction models.Finally,the proposed models are validated using NASA and CALCE lithium-ion battery datasets.Experimental results demonstrate that the proposed models achieve an RMSE not exceeding 0.007 and an MAPE not exceeding 0.0082 for SOH and RUL prediction,with a relative error in RUL prediction within the range of[-1.8%,2%].Compared to other models,the proposed model not only exhibits superior fitting capability but also demonstrates robust performance.
基金performed at large-scale research facility"Beam-M"of Bauman Moscow State Technical University following the government task by the Ministry of Science and Higher Education of the Russian Federation(No.FSFN-2024-0007).
文摘Thrust-vectoring capability has become a critical feature for propulsion systems as space missions move from static to dynamic.Thrust-vectoring is a well-developed area of rocket engine science.For electric propulsion,however,it is an evolving field that has taken a new leap forward in recent years.A review and analysis of thrust-vectoring schemes for electric propulsion systems have been conducted.The scope of this review includes thrust-vectoring schemes that can be implemented for electrostatic,electromagnetic,and beam-driven thrusters.A classification of electric propulsion schemes that provide thrust-vectoring capability is developed.More attention is given to schemes implemented in laboratory prototypes and flight models.The final part is devoted to a discussion on the suitability of different electric propulsion systems with thrust-vectoring capability for modern space mission operations.The thrust-vectoring capability of electric propulsion is necessary for inner and outer space satellites,which are at a disadvantage with conventional unidirectional propulsion systems due to their limited maneuverability.
文摘The three-axis servo system with the core of gyro stabilization is the foundation to realize its function, and a key technology of the seeker devolopment. In order to reduce the costs, improve the efficiency of research and devolopment, a new method that instead of physical prototype by virtual prototype was proposed. Adams and MATLAB/simulink are used to establish the mechanical dynamics model and controller model of the three-axis servo system. The simulation data which was processed and analyzed is compared with test data, to determine the control parameters of the virtual prototype and improve the accuracy of the model, and test the multiple condition simulation,which can provide a reference for practical production.The simulation results verify the feasibility of the models.
基金primarily supported by the National Key R&D Program of China[grant number 2021YFC3000904]the Jiangsu Provincial Key Technology R&D Program[grant number BE2022851]National Natural Science Foundation of China[grant number 42405035]。
文摘Vector winds play a crucial role in weather and climate,as well as the effective utilization of wind energy resources.However,limited research has been conducted on treating the wind field as a vector field in the evaluation of numerical weather prediction models.In this study,the authors treat vector winds as a whole by employing a vector field evaluation method,and evaluate the mesoscale model of the China Meteorological Administration(CMA-MESO)and ECMWF forecast,with reference to ERA5 reanalysis,in terms of multiple aspects of vector winds over eastern China in 2022.The results show that the ECMWF forecast is superior to CMA-MESO in predicting the spatial distribution and intensity of 10-m vector winds.Both models overestimate the wind speed in East China,and CMA-MESO overestimates the wind speed to a greater extent.The forecasting skill of the vector wind field in both models decreases with increasing lead time.The forecasting skill of CMA-MESO fluctuates more and decreases faster than that of the ECMWF forecast.There is a significant negative correlation between the model vector wind forecasting skill and terrain height.This study provides a scientific evaluation of the local application of vector wind forecasts of the CMA-MESO model and ECMWF forecast.
基金supported by the Deanship of Graduate Studies and Scientific Research at University of Bisha for funding this research through the promising program under grant number(UB-Promising-33-1445).
文摘Open networks and heterogeneous services in the Internet of Vehicles(IoV)can lead to security and privacy challenges.One key requirement for such systems is the preservation of user privacy,ensuring a seamless experience in driving,navigation,and communication.These privacy needs are influenced by various factors,such as data collected at different intervals,trip durations,and user interactions.To address this,the paper proposes a Support Vector Machine(SVM)model designed to process large amounts of aggregated data and recommend privacy preserving measures.The model analyzes data based on user demands and interactions with service providers or neighboring infrastructure.It aims to minimize privacy risks while ensuring service continuity and sustainability.The SVMmodel helps validate the system’s reliability by creating a hyperplane that distinguishes between maximum and minimum privacy recommendations.The results demonstrate the effectiveness of the proposed SVM model in enhancing both privacy and service performance.
基金supported by the National Key R&D Program of China(Grant2022YFF0503700)the National Natural Science Foundation of China(42474200 and 42174186)。
文摘The Macao Science Satellite-1(known as MSS-1)is the first scientific exploration satellite that was designed to measure the Earth's low latitude magnetic field at high resolution and with high precision by collecting data in a near-equatorial orbit.Magnetic field data from MSS-1's onboard Vector Fluxgate Magnetometer(VFM),collected at a sample rate of 50 Hz,allows us to detect and investigate sources of magnetic data contamination,from DC to relevant Nyquist frequency.Here we report two types of artificial disturbances in the VFM data.One is V-shaped events concentrated at night,with frequencies sweeping from the Nyquist frequency down to zero and back up.The other is 5-Hz events(ones that exhibit a distinct 5 Hz spectrum peak);these events are always accompanied by intervals of spiky signals,and are clearly related to the attitude control of the satellite.Our analyses show that VFM noise levels in daytime are systematically lower than in nighttime.The daily average noise levels exhibit a period of about 52 days.The V-shaped events are strongly correlated with higher VFM noise levels.
