The transport properties of liquid mixtures confined within porous media can change significantly from those observed for bulk mixtures due to changes in the liquid structuring within the pore space.Here,pulsed field ...The transport properties of liquid mixtures confined within porous media can change significantly from those observed for bulk mixtures due to changes in the liquid structuring within the pore space.Here,pulsed field gradient NMR was used to measure the diffusion coefficient of ethanol in ethanol-water liquid mixtures confined within silicas with pore diameters of 6 nm and 3 nm as a function of composition.For liquids imbibed within the 6 nm pores,the composition dependence of the ethanol diffusion coefficient closely followed that of the bulk liquid mixture and the absolute diffusion coefficients were reduced by a tortuosity factor of 3,with a minor contribution due to liquid-surface interactions.For liquids imbibed within the 3 nm pores,the diffusion coefficient of ethanol decreased as the composition of ethanol within the pore space increased,and for single-component ethanol imbibition the effective tortuosity was 63.Fast field cycling NMR experiments showed that the diffusion behaviour was not controlled by an increase in ethanol adsorption strength.A geometric analysis of the pore space was consistent with a highly confined system in which most molecules interacted with the pore walls.Under such confinement,the liquid structuring within the bulk pore space did not reflect that of the bulk liquid mixtures,and the observed decrease in diffusion coefficient as ethanol composition increased was consistent with an increase in confinement due to the larger size of the ethanol molecule.展开更多
Nanoconfinement is a promising approach to simultaneously enhance the thermodynamics,kinetics,and cycling stability of hydrogen storage materials.The introduction of supporting scaffolds usually causes a reduction in ...Nanoconfinement is a promising approach to simultaneously enhance the thermodynamics,kinetics,and cycling stability of hydrogen storage materials.The introduction of supporting scaffolds usually causes a reduction in the total hydrogen storage capacity due to“dead weight.”Here,we synthesize an optimized N-doped porous carbon(rN-pC)without heavy metal as supporting scaffold to confine Mg/MgH_(2) nanoparticles(Mg/MgH_(2)@rN-pC).rN-pC with 60 wt%loading capacity of Mg(denoted as 60 Mg@rN-pC)can adsorb and desorb 0.62 wt%H_(2) on the rN-pC scaffold.The nanoconfined MgH_(2) can be chemically dehydrided at 175℃,providing~3.59 wt%H_(2) with fast kinetics(fully dehydrogenated at 300℃ within 15 min).This study presents the first realization of nanoconfined Mg-based system with adsorption-active scaffolds.Besides,the nanoconfined MgH_(2) formation enthalpy is reduced to~68 kJ mol^(−1) H_(2) from~75 kJ mol^(−1) H_(2) for pure MgH_(2).The composite can be also compressed to nanostructured pellets,with volumetric H_(2) density reaching 33.4 g L^(−1) after 500 MPa compression pressure,which surpasses the 24 g L^(−1) volumetric capacity of 350 bar compressed H_(2).Our approach can be implemented to the design of hybrid H_(2) storage materials with enhanced capacity and desorption rate.展开更多
Heart disease remains a leading cause of mortality worldwide,emphasizing the urgent need for reliable and interpretable predictive models to support early diagnosis and timely intervention.However,existing Deep Learni...Heart disease remains a leading cause of mortality worldwide,emphasizing the urgent need for reliable and interpretable predictive models to support early diagnosis and timely intervention.However,existing Deep Learning(DL)approaches often face several limitations,including inefficient feature extraction,class imbalance,suboptimal classification performance,and limited interpretability,which collectively hinder their deployment in clinical settings.To address these challenges,we propose a novel DL framework for heart disease prediction that integrates a comprehensive preprocessing pipeline with an advanced classification architecture.The preprocessing stage involves label encoding and feature scaling.To address the issue of class imbalance inherent in the personal key indicators of the heart disease dataset,the localized random affine shadowsampling technique is employed,which enhances minority class representation while minimizing overfitting.At the core of the framework lies the Deep Residual Network(DeepResNet),which employs hierarchical residual transformations to facilitate efficient feature extraction and capture complex,non-linear relationships in the data.Experimental results demonstrate that the proposed model significantly outperforms existing techniques,achieving improvements of 3.26%in accuracy,3.16%in area under the receiver operating characteristics,1.09%in recall,and 1.07%in F1-score.Furthermore,robustness is validated using 10-fold crossvalidation,confirming the model’s generalizability across diverse data distributions.Moreover,model interpretability is ensured through the integration of Shapley additive explanations and local interpretable model-agnostic explanations,offering valuable insights into the contribution of individual features to model predictions.Overall,the proposed DL framework presents a robust,interpretable,and clinically applicable solution for heart disease prediction.展开更多
The concept of hybrid ships has gained significant attention in recent years,as they offer an effective means of enhancing energy utilization and reducing environmental pollution.However,the navigational environments ...The concept of hybrid ships has gained significant attention in recent years,as they offer an effective means of enhancing energy utilization and reducing environmental pollution.However,the navigational environments of ships are often subject to changes,which in turn affect their energy efficiency in a complex manner.It is therefore evident that enhancing the energy efficiency of hybrid ships is a worthwhile goal.In this study,we take a diesel-electric hybrid ship navigating in inland waterways as the research object,and propose a hierarchical optimization method for ship energy efficiency.The upper-layer control establishes a predictive model for propulsion motor speed and fuel consumption through multivariate time series predictions,and employs the model predictive control(MPC)method to optimize the propulsion motor speed.The lower-layer control utilizes an equivalent fuel consumption minimization method,which is based on improving the equivalence factor.This involves combining the variation of the supercapacitor’s state of charge(SOC)with the propulsion motor speed obtained from the MPC optimization in the upper-layer control.Furthermore,a proportional integral(PI)controller is used to adjust the equivalence factor,in order to adapt the equivalent fuel consumption minimization method to the working conditions.Our results demonstrate that the proposed hierarchical optimization method can reduce the energy efficiency operating indicator(EEOI)by approximately 11.54%and the fuel consumption by approximately 9.47%in comparison to the pre-optimization scenario.展开更多
文摘The transport properties of liquid mixtures confined within porous media can change significantly from those observed for bulk mixtures due to changes in the liquid structuring within the pore space.Here,pulsed field gradient NMR was used to measure the diffusion coefficient of ethanol in ethanol-water liquid mixtures confined within silicas with pore diameters of 6 nm and 3 nm as a function of composition.For liquids imbibed within the 6 nm pores,the composition dependence of the ethanol diffusion coefficient closely followed that of the bulk liquid mixture and the absolute diffusion coefficients were reduced by a tortuosity factor of 3,with a minor contribution due to liquid-surface interactions.For liquids imbibed within the 3 nm pores,the diffusion coefficient of ethanol decreased as the composition of ethanol within the pore space increased,and for single-component ethanol imbibition the effective tortuosity was 63.Fast field cycling NMR experiments showed that the diffusion behaviour was not controlled by an increase in ethanol adsorption strength.A geometric analysis of the pore space was consistent with a highly confined system in which most molecules interacted with the pore walls.Under such confinement,the liquid structuring within the bulk pore space did not reflect that of the bulk liquid mixtures,and the observed decrease in diffusion coefficient as ethanol composition increased was consistent with an increase in confinement due to the larger size of the ethanol molecule.
基金supported by the National Key R&D Program of China(2022YFB3803700)National Natural Science Foundation of China(52171186)+1 种基金Young Elite Scientists Sponsorship Program by CAST(2023QNRC001)support from“Zhiyuan Honor Program”for doctoral students,Shanghai Jiao Tong University.
文摘Nanoconfinement is a promising approach to simultaneously enhance the thermodynamics,kinetics,and cycling stability of hydrogen storage materials.The introduction of supporting scaffolds usually causes a reduction in the total hydrogen storage capacity due to“dead weight.”Here,we synthesize an optimized N-doped porous carbon(rN-pC)without heavy metal as supporting scaffold to confine Mg/MgH_(2) nanoparticles(Mg/MgH_(2)@rN-pC).rN-pC with 60 wt%loading capacity of Mg(denoted as 60 Mg@rN-pC)can adsorb and desorb 0.62 wt%H_(2) on the rN-pC scaffold.The nanoconfined MgH_(2) can be chemically dehydrided at 175℃,providing~3.59 wt%H_(2) with fast kinetics(fully dehydrogenated at 300℃ within 15 min).This study presents the first realization of nanoconfined Mg-based system with adsorption-active scaffolds.Besides,the nanoconfined MgH_(2) formation enthalpy is reduced to~68 kJ mol^(−1) H_(2) from~75 kJ mol^(−1) H_(2) for pure MgH_(2).The composite can be also compressed to nanostructured pellets,with volumetric H_(2) density reaching 33.4 g L^(−1) after 500 MPa compression pressure,which surpasses the 24 g L^(−1) volumetric capacity of 350 bar compressed H_(2).Our approach can be implemented to the design of hybrid H_(2) storage materials with enhanced capacity and desorption rate.
基金funded by Ongoing Research Funding Program for Project number(ORF-2025-648),King Saud University,Riyadh,Saudi Arabia.
文摘Heart disease remains a leading cause of mortality worldwide,emphasizing the urgent need for reliable and interpretable predictive models to support early diagnosis and timely intervention.However,existing Deep Learning(DL)approaches often face several limitations,including inefficient feature extraction,class imbalance,suboptimal classification performance,and limited interpretability,which collectively hinder their deployment in clinical settings.To address these challenges,we propose a novel DL framework for heart disease prediction that integrates a comprehensive preprocessing pipeline with an advanced classification architecture.The preprocessing stage involves label encoding and feature scaling.To address the issue of class imbalance inherent in the personal key indicators of the heart disease dataset,the localized random affine shadowsampling technique is employed,which enhances minority class representation while minimizing overfitting.At the core of the framework lies the Deep Residual Network(DeepResNet),which employs hierarchical residual transformations to facilitate efficient feature extraction and capture complex,non-linear relationships in the data.Experimental results demonstrate that the proposed model significantly outperforms existing techniques,achieving improvements of 3.26%in accuracy,3.16%in area under the receiver operating characteristics,1.09%in recall,and 1.07%in F1-score.Furthermore,robustness is validated using 10-fold crossvalidation,confirming the model’s generalizability across diverse data distributions.Moreover,model interpretability is ensured through the integration of Shapley additive explanations and local interpretable model-agnostic explanations,offering valuable insights into the contribution of individual features to model predictions.Overall,the proposed DL framework presents a robust,interpretable,and clinically applicable solution for heart disease prediction.
基金supported by the National Natural Science Foundation of China(No.52571367)and the Commissions Project of China(No.CBG4N21).
文摘The concept of hybrid ships has gained significant attention in recent years,as they offer an effective means of enhancing energy utilization and reducing environmental pollution.However,the navigational environments of ships are often subject to changes,which in turn affect their energy efficiency in a complex manner.It is therefore evident that enhancing the energy efficiency of hybrid ships is a worthwhile goal.In this study,we take a diesel-electric hybrid ship navigating in inland waterways as the research object,and propose a hierarchical optimization method for ship energy efficiency.The upper-layer control establishes a predictive model for propulsion motor speed and fuel consumption through multivariate time series predictions,and employs the model predictive control(MPC)method to optimize the propulsion motor speed.The lower-layer control utilizes an equivalent fuel consumption minimization method,which is based on improving the equivalence factor.This involves combining the variation of the supercapacitor’s state of charge(SOC)with the propulsion motor speed obtained from the MPC optimization in the upper-layer control.Furthermore,a proportional integral(PI)controller is used to adjust the equivalence factor,in order to adapt the equivalent fuel consumption minimization method to the working conditions.Our results demonstrate that the proposed hierarchical optimization method can reduce the energy efficiency operating indicator(EEOI)by approximately 11.54%and the fuel consumption by approximately 9.47%in comparison to the pre-optimization scenario.
