Typhoon Bebinca in 2024 experienced a nearshore outbreak(a rapid intensification(RI)near the coast),making accurate forecasting of unpredictable tracks and intensities highly challenging.The AI model is superior to th...Typhoon Bebinca in 2024 experienced a nearshore outbreak(a rapid intensification(RI)near the coast),making accurate forecasting of unpredictable tracks and intensities highly challenging.The AI model is superior to the numerical model for typhoon track prediction but performs worse for intensity forecasting.Vortex initialization is an effective approach to further improve numerical prediction via cycle assimilation,accounting for multiple relocating TC centers and adjusting the typhoon initial structure.In addition,by integrating numerical runs with an AI weather model through real-time dynamic weight correction of the forecast,the predictive skill is further improved.For example,it can reduce the deviation of 72-h track forecasting by 25%compared with the numerical model and decrease the intensity deviation by 2%and 56%relative to the numerical run and AI forecasts,respectively.On the basis of the best-performing forecasting,the inner-core convective burst(CB)characteristics are illuminated.The attributions of the nearshore outbreak and RI of Typhoon Bebinca are examined.From the viewpoint of bottom-up convection growth,the CB is associated with the energy supply from the high-boundarylayer CAPE,the following upward-developing secondary circulation,and accompanying latent heat release of hydrometeors.The contracted radius of maximum winds(RMW)and increased inertial stability within the inner core region effectively prevent the escape of the high-energy atmosphere and favor rapid intensification and maintenance of the offshore burst of a typhoon.The intensifying secondary circulation further promotes the primary circulation of the TC and RI processes through the gradient wind balance.展开更多
The operational Tropical Regional Atmospheric Model System(TRAMS)often underestimates initial typhoon intensity when using the global analysis field as the initial condition.The TRAMS tropical cyclone(TC)initializatio...The operational Tropical Regional Atmospheric Model System(TRAMS)often underestimates initial typhoon intensity when using the global analysis field as the initial condition.The TRAMS tropical cyclone(TC)initialization scheme,developed based on the incremental analysis updates(IAU)technique,effectively reduces initial bias.However,the original IAU-based TC initialization scheme only adjusts the wind field at the analysis moment,with other variables adjusted implicitly under the model's constraints according to a gradually inserted wind increment(named“univariate adjustment scheme”hereafter).The univariate adjustment scheme requires approximately 3 h to reach a dynamic equilibrium state,which constrains the assimilation of hourly TC observations and causes excessive dissipation of meaningful short-wave information in adjustment increments.To address this limitation,this study develops a multivariate adjustment IAU-based TC initialization scheme that incorporates gradient wind balance and hydrostatic balance as its largescale constraints.Numerical experiments with TC Hato(2017)demonstrate that the multivariate adjustment scheme reduces the IAU relaxation time to 1 h while marginally improving forecast skill.These findings are consistently replicated across 12 additional TC cases.The development of the IAU-based multivariate adjustment initialization scheme establishes a foundation for 4-D initialization using hourly TC observations.展开更多
Recent advancements in computational and database technologies have led to the exponential growth of large-scale medical datasets,significantly increasing data complexity and dimensionality in medical diagnostics.Effi...Recent advancements in computational and database technologies have led to the exponential growth of large-scale medical datasets,significantly increasing data complexity and dimensionality in medical diagnostics.Efficient feature selection methods are critical for improving diagnostic accuracy,reducing computational costs,and enhancing the interpretability of predictive models.Particle Swarm Optimization(PSO),a widely used metaheuristic inspired by swarm intelligence,has shown considerable promise in feature selection tasks.However,conventional PSO often suffers from premature convergence and limited exploration capabilities,particularly in high-dimensional spaces.To overcome these limitations,this study proposes an enhanced PSO framework incorporating Orthogonal Initializa-tion and a Crossover Operator(OrPSOC).Orthogonal Initialization ensures a diverse and uniformly distributed initial particle population,substantially improving the algorithm’s exploration capability.The Crossover Operator,inspired by genetic algorithms,introduces additional diversity during the search process,effectively mitigating premature convergence and enhancing global search performance.The effectiveness of OrPSOC was rigorously evaluated on three benchmark medical datasets—Colon,Leukemia,and Prostate Tumor.Comparative analyses were conducted against traditional filter-based methods,including Fast Clustering-Based Feature Selection Technique(Fast-C),Minimum Redundancy Maximum Relevance(MinRedMaxRel),and Five-Way Joint Mutual Information(FJMI),as well as prominent metaheuristic algorithms such as standard PSO,Ant Colony Optimization(ACO),Comprehensive Learning Gravitational Search Algorithm(CLGSA),and Fuzzy-Based CLGSA(FCLGSA).Experimental results demonstrated that OrPSOC consistently outperformed these existing methods in terms of classification accuracy,computational efficiency,and result stability,achieving significant improvements even with fewer selected features.Additionally,a sensitivity analysis of the crossover parameter provided valuable insights into parameter tuning and its impact on model performance.These findings highlight the superiority and robustness of the proposed OrPSOC approach for feature selection in medical diagnostic applications and underscore its potential for broader adoption in various high-dimensional,data-driven fields.展开更多
Physics-informed neural networks(PINNs)have shown considerable promise for performing numerical simulations in fluid mechanics.They provide mesh-free,end-to-end approaches by embedding physical laws into their loss fu...Physics-informed neural networks(PINNs)have shown considerable promise for performing numerical simulations in fluid mechanics.They provide mesh-free,end-to-end approaches by embedding physical laws into their loss functions.However,when addressing complex flow problems,PINNs still face some challenges such as activation saturation and vanishing gradients in deep network training,leading to slow convergence and insufficient prediction accuracy.We present physics-informed neural networks incorporating lattice Boltzmann method optimized by tanh robust weight initialization(T-PINN-LBM)to address these challenges.This approach fuses the mesoscopic lattice Boltzmann model with the automatic differentiation framework of PINNs.It also implements a tanh robust weight initialization method derived from fixed point analysis.This model effectively mitigates activation and gradient decay in deep networks,improving convergence speed and data efficiency in multiscale flow simulations.We validate the effectiveness of the model on the classical arithmetic example of lid-driven cavity flow.Compared to the traditional Xavier initialized PINN and PINN-LBM,T-PINNLBM reduces the mean absolute error(MAE)by one order of magnitude at the same network depth and maintains stable convergence in deeper networks.The results demonstrate that this model can accurately capture complex flow structures without prior data,providing a new feasible pathway for data-free driven fluid simulation.展开更多
This paper presents an analysis of an equilateral triangular array formation initialization for space-based gravitational wave observatory(GWO)near Lagrange points in the circular-restricted three-body problem.A stabl...This paper presents an analysis of an equilateral triangular array formation initialization for space-based gravitational wave observatory(GWO)near Lagrange points in the circular-restricted three-body problem.A stable configuration is essential for the continuous observation of gravitational waves(GWs).However,the motion near the collinear libration points is highly unstable.This problem is examined by output regulation theory.Using the tracking aspect,the equilateral triangular array formation is established in two periods and the fuel consumption is calculated.Furthermore,the natural evolution of the formation without control input is analyzed,and the effective stability duration is quantified to determine the timing of control interventions.Finally,to observe the GWs in same direction with different frequency bands,scale reconfiguration is employed.展开更多
基金supported by the Strategic Priority Research Program of the Chinese Academy of Sciences (Grant No.XDB0760300)the National Natural Science Foundation of China (Grant Nos.42175010 and 41875079)。
文摘Typhoon Bebinca in 2024 experienced a nearshore outbreak(a rapid intensification(RI)near the coast),making accurate forecasting of unpredictable tracks and intensities highly challenging.The AI model is superior to the numerical model for typhoon track prediction but performs worse for intensity forecasting.Vortex initialization is an effective approach to further improve numerical prediction via cycle assimilation,accounting for multiple relocating TC centers and adjusting the typhoon initial structure.In addition,by integrating numerical runs with an AI weather model through real-time dynamic weight correction of the forecast,the predictive skill is further improved.For example,it can reduce the deviation of 72-h track forecasting by 25%compared with the numerical model and decrease the intensity deviation by 2%and 56%relative to the numerical run and AI forecasts,respectively.On the basis of the best-performing forecasting,the inner-core convective burst(CB)characteristics are illuminated.The attributions of the nearshore outbreak and RI of Typhoon Bebinca are examined.From the viewpoint of bottom-up convection growth,the CB is associated with the energy supply from the high-boundarylayer CAPE,the following upward-developing secondary circulation,and accompanying latent heat release of hydrometeors.The contracted radius of maximum winds(RMW)and increased inertial stability within the inner core region effectively prevent the escape of the high-energy atmosphere and favor rapid intensification and maintenance of the offshore burst of a typhoon.The intensifying secondary circulation further promotes the primary circulation of the TC and RI processes through the gradient wind balance.
基金supported by the National University of Defense Technology(NUDT)Research Initiation Funding for High-Level Scientific and Technological Innovative Talents(202402-YJRC-LJ-001)the National Natural Science Foundation of China(Grant No.U2142213)+1 种基金the Basic and Applied Basic Research Foundation of Guangdong Province(Grants 2025A1515011835,2022A1515011870)the National Natural Science Foundation of China(Grant No.42305167)。
文摘The operational Tropical Regional Atmospheric Model System(TRAMS)often underestimates initial typhoon intensity when using the global analysis field as the initial condition.The TRAMS tropical cyclone(TC)initialization scheme,developed based on the incremental analysis updates(IAU)technique,effectively reduces initial bias.However,the original IAU-based TC initialization scheme only adjusts the wind field at the analysis moment,with other variables adjusted implicitly under the model's constraints according to a gradually inserted wind increment(named“univariate adjustment scheme”hereafter).The univariate adjustment scheme requires approximately 3 h to reach a dynamic equilibrium state,which constrains the assimilation of hourly TC observations and causes excessive dissipation of meaningful short-wave information in adjustment increments.To address this limitation,this study develops a multivariate adjustment IAU-based TC initialization scheme that incorporates gradient wind balance and hydrostatic balance as its largescale constraints.Numerical experiments with TC Hato(2017)demonstrate that the multivariate adjustment scheme reduces the IAU relaxation time to 1 h while marginally improving forecast skill.These findings are consistently replicated across 12 additional TC cases.The development of the IAU-based multivariate adjustment initialization scheme establishes a foundation for 4-D initialization using hourly TC observations.
文摘Recent advancements in computational and database technologies have led to the exponential growth of large-scale medical datasets,significantly increasing data complexity and dimensionality in medical diagnostics.Efficient feature selection methods are critical for improving diagnostic accuracy,reducing computational costs,and enhancing the interpretability of predictive models.Particle Swarm Optimization(PSO),a widely used metaheuristic inspired by swarm intelligence,has shown considerable promise in feature selection tasks.However,conventional PSO often suffers from premature convergence and limited exploration capabilities,particularly in high-dimensional spaces.To overcome these limitations,this study proposes an enhanced PSO framework incorporating Orthogonal Initializa-tion and a Crossover Operator(OrPSOC).Orthogonal Initialization ensures a diverse and uniformly distributed initial particle population,substantially improving the algorithm’s exploration capability.The Crossover Operator,inspired by genetic algorithms,introduces additional diversity during the search process,effectively mitigating premature convergence and enhancing global search performance.The effectiveness of OrPSOC was rigorously evaluated on three benchmark medical datasets—Colon,Leukemia,and Prostate Tumor.Comparative analyses were conducted against traditional filter-based methods,including Fast Clustering-Based Feature Selection Technique(Fast-C),Minimum Redundancy Maximum Relevance(MinRedMaxRel),and Five-Way Joint Mutual Information(FJMI),as well as prominent metaheuristic algorithms such as standard PSO,Ant Colony Optimization(ACO),Comprehensive Learning Gravitational Search Algorithm(CLGSA),and Fuzzy-Based CLGSA(FCLGSA).Experimental results demonstrated that OrPSOC consistently outperformed these existing methods in terms of classification accuracy,computational efficiency,and result stability,achieving significant improvements even with fewer selected features.Additionally,a sensitivity analysis of the crossover parameter provided valuable insights into parameter tuning and its impact on model performance.These findings highlight the superiority and robustness of the proposed OrPSOC approach for feature selection in medical diagnostic applications and underscore its potential for broader adoption in various high-dimensional,data-driven fields.
基金supported by the National Natural Science Foundation of China(Grant Nos.12474453,12174085,and 12404530).
文摘Physics-informed neural networks(PINNs)have shown considerable promise for performing numerical simulations in fluid mechanics.They provide mesh-free,end-to-end approaches by embedding physical laws into their loss functions.However,when addressing complex flow problems,PINNs still face some challenges such as activation saturation and vanishing gradients in deep network training,leading to slow convergence and insufficient prediction accuracy.We present physics-informed neural networks incorporating lattice Boltzmann method optimized by tanh robust weight initialization(T-PINN-LBM)to address these challenges.This approach fuses the mesoscopic lattice Boltzmann model with the automatic differentiation framework of PINNs.It also implements a tanh robust weight initialization method derived from fixed point analysis.This model effectively mitigates activation and gradient decay in deep networks,improving convergence speed and data efficiency in multiscale flow simulations.We validate the effectiveness of the model on the classical arithmetic example of lid-driven cavity flow.Compared to the traditional Xavier initialized PINN and PINN-LBM,T-PINNLBM reduces the mean absolute error(MAE)by one order of magnitude at the same network depth and maintains stable convergence in deeper networks.The results demonstrate that this model can accurately capture complex flow structures without prior data,providing a new feasible pathway for data-free driven fluid simulation.
基金supported by the China Scholarship Council(CSC)(No.202206290131)。
文摘This paper presents an analysis of an equilateral triangular array formation initialization for space-based gravitational wave observatory(GWO)near Lagrange points in the circular-restricted three-body problem.A stable configuration is essential for the continuous observation of gravitational waves(GWs).However,the motion near the collinear libration points is highly unstable.This problem is examined by output regulation theory.Using the tracking aspect,the equilateral triangular array formation is established in two periods and the fuel consumption is calculated.Furthermore,the natural evolution of the formation without control input is analyzed,and the effective stability duration is quantified to determine the timing of control interventions.Finally,to observe the GWs in same direction with different frequency bands,scale reconfiguration is employed.
