Coal is a crucial fossil energy in today’s society,and the detection of sulfir(S) and nitrogen(N)in coal is essential for the evaluation of coal quality.Therefore,an efficient method is needed to quantitatively analy...Coal is a crucial fossil energy in today’s society,and the detection of sulfir(S) and nitrogen(N)in coal is essential for the evaluation of coal quality.Therefore,an efficient method is needed to quantitatively analyze N and S content in coal,to achieve the purpose of clean utilization of coal.This study applied laser-induced breakdown spectroscopy(LIBS) to test coal quality,and combined two variable selection algorithms,competitive adaptive reweighted sampling(CARS) and the successive projections algorithm(SPA),to establish the corresponding partial least square(PLS) model.The results of the experiment were as follows.The PLS modeled with the full spectrum of 27,620 variables has poor accuracy,the coefficient of determination of the test set(R^2 P) and root mean square error of the test set(RMSEP) of nitrogen were 0.5172 and 0.2263,respectively,and those of sulfur were0.5784 and 0.5811,respectively.The CARS-PLS screened 37 and 25 variables respectively in the detection of N and S elements,but the prediction ability of the model did not improve significantly.SPA-PLS finally screened 14 and 11 variables respectively through successive projections,and obtained the best prediction effect among the three methods.The R^2 P and RMSEP of nitrogen were0.9873 and 0.0208,respectively,and those of sulfur were 0.9451 and 0.2082,respectively.In general,the predictive results of the two elements increased by about 90% for RMSEP and 60% for R2 P compared with PLS.The results show that LIBS combined with SPA-PLS has good potential for detecting N and S content in coal,and is a very promising technology for industrial application.展开更多
Transmission line(TL)Parameter Identification(PI)method plays an essential role in the transmission system.The existing PI methods usually have two limitations:(1)These methods only model for single TL,and can not con...Transmission line(TL)Parameter Identification(PI)method plays an essential role in the transmission system.The existing PI methods usually have two limitations:(1)These methods only model for single TL,and can not consider the topology connection of multiple branches for simultaneous identification.(2)Transient bad data is ignored by methods,and the random selection of terminal section data may cause the distortion of PI and have serious consequences.Therefore,a multi-task PI model considering multiple TLs’spatial constraints and massive electrical section data is proposed in this paper.The Graph Attention Network module is used to draw a single TL into a node and calculate its influence coefficient in the transmission network.Multi-Task strategy of Hard Parameter Sharing is used to identify the conductance ofmultiple branches simultaneously.Experiments show that themethod has good accuracy and robustness.Due to the consideration of spatial constraints,the method can also obtain more accurate conductance values under different training and testing conditions.展开更多
Single crystals of 4SC(NH2)2–Ni1-xCux Cl2(x = 0.03)(Cu-DTN) containing spin S = 1/2 Cu2+and S = 1 Ni2+cations are synthesized by slow evaporation methods. Structural characterization demonstrates that the Cu-DTN is o...Single crystals of 4SC(NH2)2–Ni1-xCux Cl2(x = 0.03)(Cu-DTN) containing spin S = 1/2 Cu2+and S = 1 Ni2+cations are synthesized by slow evaporation methods. Structural characterization demonstrates that the Cu-DTN is of a tetrahedral structure with lattice parameter c being 9.0995 ?A, which is 1.32% expansion compared with that of parent material DTN due to the larger radius of the Cu ion. Direct current(DC) susceptibility measurements show that both the antiferromagnetic exchange interaction at low temperature and the large anisotropy of susceptibilities are suppressed after doping the Cu ion, which could be related to the structural distortion and the increase of the super-exchange paths in Cu-DTN.展开更多
Detailed density functional theory(DFT)calculations of the structural,mechanical,thermodynamic,and electronicproperties of crystalline CaF2 with five different structures in the pressure range of 0 GPa–150 GPa are pe...Detailed density functional theory(DFT)calculations of the structural,mechanical,thermodynamic,and electronicproperties of crystalline CaF2 with five different structures in the pressure range of 0 GPa–150 GPa are performed byboth GGA(generalized gradient approximation)-PBE(Perdew–Burke–Ernzerhof)and LDA(local density approximation)-CAPZ(Cambridge Serial Total Energy Package).It is found that the enthalpy differences imply that the fluorite phase→PbCl2-type phase→Ni2In-type phase transition in CaF2 occurs at PGGA1=8.0 GPa,PGGA2=111.4 GPa by usingthe XC of GGA,and PLDA1=4.5 GPa,PLDA2=101.7 GPa by LDA,respectively,which is consistent with previousexperiments and theoretical conclusions.Moreover,the enthalpy differences between PbCl2-type and Ni2In-type phases inone molecular formula become very small at the pressure of about 100 GPa,indicating the possibility of coexistence of twophase at high pressures.This may be the reason why the transition pressure of the second phase transition in other reportsis so huge(68 GPa–278 GPa).The volume changed in the second phase transition are also consistent with the enthalpydifference result.Besides,the pressure dependence of mechanical and thermodynamic properties of CaF2 is studied.Itis found that the high-pressure phase of Ni2In-type structure has better stiffness in CaF2 crystal,and the hardness of thematerial has hardly changed in the second phase transition.Finally,the electronic structure of CaF2 is also analyzed withthe change of pressure.By analyzing the band gap and density of states,the large band gap indicates the CaF2 crystal isalways an insulator at 0 GPa–150 GPa.展开更多
To dates,most ship detection approaches for single-pol synthetic aperture radar(SAR) imagery try to ensure a constant false-alarm rate(CFAR).A high performance ship detector relies on two key components:an accurate es...To dates,most ship detection approaches for single-pol synthetic aperture radar(SAR) imagery try to ensure a constant false-alarm rate(CFAR).A high performance ship detector relies on two key components:an accurate estimation to a sea surface distribution and a fine designed CFAR algorithm.First,a novel nonparametric sea surface distribution estimation method is developed based on n-order Bézier curve.To estimate the sea surface distribution using n-order Bézier curve,an explicit analytical solution is derived based on a least square optimization,and the optimal selection also is presented to two essential parameters,the order n of Bézier curve and the number m of sample points.Next,to validate the ship detection performance of the estimated sea surface distribution,the estimated sea surface distribution by n-order Bézier curve is combined with a cell averaging CFAR(CA-CFAR).To eliminate the possible interfering ship targets in background window,an improved automatic censoring method is applied.Comprehensive experiments prove that in terms of sea surface estimation performance,the proposed method is as good as a traditional nonparametric Parzen window kernel method,and in most cases,outperforms two widely used parametric methods,K and G0 models.In terms of computation speed,a major advantage of the proposed estimation method is the time consuming only depended on the number m of sample points while independent of imagery size,which makes it can achieve a significant speed improvement to the Parzen window kernel method,and in some cases,it is even faster than two parametric methods.In terms of ship detection performance,the experiments show that the ship detector which constructed by the proposed sea surface distribution model and the given CA-CFAR algorithm has wide adaptability to different SAR sensors,resolutions and sea surface homogeneities and obtains a leading performance on the test dataset.展开更多
Overcharging is an important safety issue in the charging process of electric vehicle power batteries,and can easily lead to accelerated battery aging and serious safety accidents.It is necessary to accurately predict...Overcharging is an important safety issue in the charging process of electric vehicle power batteries,and can easily lead to accelerated battery aging and serious safety accidents.It is necessary to accurately predict the vehicle’s charging time to effectively prevent the battery from overcharging.Due to the complex structure of the battery pack and various charging modes,the traditional charging time prediction method often encounters modeling difficulties and low accuracy.In response to the above problems,data drivers and machine learning theories are applied.On the basis of fully considering the different electric vehicle battery management system(BMS)charging modes,a charging time prediction method with charging mode recognition is proposed.First,an intelligent algorithm based on dynamic weighted density peak clustering(DWDPC)and random forest fusion is proposed to classify vehicle charging modes.Then,on the basis of an improved simplified particle swarm optimization(ISPSO)algorithm,a high-performance charging time prediction method is constructed by fully integrating long short-term memory(LSTM)and a strong tracking filter.Finally,the data run by the actual engineering system are verified for the proposed charging time prediction algorithm.Experimental results show that the new method can effectively distinguish the charging modes of different vehicles,identify the charging characteristics of different electric vehicles,and achieve high prediction accuracy.展开更多
Waste collection is an important part of waste management system.Transportation costs and carbon emissions can be greatly reduced by proper vehicle routing.Meanwhile,each vehicle can work again after achieving its cap...Waste collection is an important part of waste management system.Transportation costs and carbon emissions can be greatly reduced by proper vehicle routing.Meanwhile,each vehicle can work again after achieving its capacity limit and unloading the waste.For this,an energy-efficient multi-trip vehicle routing model is established for municipal solid waste collection,which incorporates practical factors like the limited capacity,maximum working hours,and multiple trips of each vehicle.Considering both economy and environment,fixed costs,fuel costs,and carbon emission costs are minimized together.To solve the formulated model effectively,contribution-based adaptive particle swarm optimization is proposed.Four strategies named greedy learning,multi-operator learning,exploring learning,and exploiting learning are specifically designed with their own searching priorities.By assessing the contribution of each learning strategy during the process of evolution,an appropriate one is selected and assigned to each individual adaptively to improve the searching efficiency of the algorithm.Moreover,an improved local search operator is performed on the trips with the largest number of waste sites so that both the exploiting ability and the convergence accuracy of the algorithm are improved.Performance of the proposed algorithm is tested on ten waste collection instances,which include one real-world case derived from the Green Ring Company of Jiangbei New District,Nanjing,China,and nine synthetic instances with increasing scales generated from the commonly-used capacitated vehicle routing problem benchmark datasets.Comparisons with five state-of-the-art algorithms show that the proposed algorithm can obtain a solution with a higher accuracy for the constructed model.展开更多
Despite the demonstrated success of numerous correlation filter(CF)based tracking approaches,their assumption of circulant structure of samples introduces significant redundancy to learn an effective classifier.In thi...Despite the demonstrated success of numerous correlation filter(CF)based tracking approaches,their assumption of circulant structure of samples introduces significant redundancy to learn an effective classifier.In this paper,we develop a fast manifold regularized context-aware correlation tracking algorithm that mines the local manifold structure information of different types of samples.First,different from the traditional CF based tracking that only uses one base sample,we employ a set of contextual samples near to the base sample,and impose a manifold structure assumption on them.Afterwards,to take into account the manifold structure among these samples,we introduce a linear graph Laplacian regularized term into the objective of CF learning.Fortunately,the optimization can be efficiently solved in a closed form with fast Fourier transforms(FFTs),which contributes to a highly efficient implementation.Extensive evaluations on the OTB100 and VOT2016 datasets demonstrate that the proposed tracker performs favorably against several state-of-the-art algorithms in terms of accuracy and robustness.Especially,our tracker is able to run in real-time with 28 fps on a single CPU.展开更多
A constrained multi-objective optimization model for the low-carbon vehicle routing problem(VRP)is established.A carbon emission measurement method considering various practical factors is introduced.It minimizes both...A constrained multi-objective optimization model for the low-carbon vehicle routing problem(VRP)is established.A carbon emission measurement method considering various practical factors is introduced.It minimizes both the total carbon emissions and the longest time consumed by the sub-tours,subject to the limited number of available vehicles.According to the characteristics of the model,a region enhanced discrete multi-objective fireworks algorithm is proposed.A partial mapping explosion operator,a hybrid mutation for adjusting the sub-tours,and an objective-driven extending search are designed,which aim to improve the convergence,diversity,and spread of the non-dominated solutions produced by the algorithm,respectively.Nine low-carbon VRP instances with different scales are used to verify the effectiveness of the new strategies.Furthermore,comparison results with four state-of-the-art algorithms indicate that the proposed algorithm has better performance of convergence and distribution on the low-carbon VRP.It provides a promising scalability to the problem size.展开更多
基金the Jiangsu Government Scholarship for Overseas Studies (JS-2019-031)the Startup Foundation for Introducing Talent of NUIST (2243141701023)。
文摘Coal is a crucial fossil energy in today’s society,and the detection of sulfir(S) and nitrogen(N)in coal is essential for the evaluation of coal quality.Therefore,an efficient method is needed to quantitatively analyze N and S content in coal,to achieve the purpose of clean utilization of coal.This study applied laser-induced breakdown spectroscopy(LIBS) to test coal quality,and combined two variable selection algorithms,competitive adaptive reweighted sampling(CARS) and the successive projections algorithm(SPA),to establish the corresponding partial least square(PLS) model.The results of the experiment were as follows.The PLS modeled with the full spectrum of 27,620 variables has poor accuracy,the coefficient of determination of the test set(R^2 P) and root mean square error of the test set(RMSEP) of nitrogen were 0.5172 and 0.2263,respectively,and those of sulfur were0.5784 and 0.5811,respectively.The CARS-PLS screened 37 and 25 variables respectively in the detection of N and S elements,but the prediction ability of the model did not improve significantly.SPA-PLS finally screened 14 and 11 variables respectively through successive projections,and obtained the best prediction effect among the three methods.The R^2 P and RMSEP of nitrogen were0.9873 and 0.0208,respectively,and those of sulfur were 0.9451 and 0.2082,respectively.In general,the predictive results of the two elements increased by about 90% for RMSEP and 60% for R2 P compared with PLS.The results show that LIBS combined with SPA-PLS has good potential for detecting N and S content in coal,and is a very promising technology for industrial application.
基金supported by the National Natural Science Foundation of PR China(42075130)the Postgraduate Research and Innovation Project of Jiangsu Province(1534052101133).
文摘Transmission line(TL)Parameter Identification(PI)method plays an essential role in the transmission system.The existing PI methods usually have two limitations:(1)These methods only model for single TL,and can not consider the topology connection of multiple branches for simultaneous identification.(2)Transient bad data is ignored by methods,and the random selection of terminal section data may cause the distortion of PI and have serious consequences.Therefore,a multi-task PI model considering multiple TLs’spatial constraints and massive electrical section data is proposed in this paper.The Graph Attention Network module is used to draw a single TL into a node and calculate its influence coefficient in the transmission network.Multi-Task strategy of Hard Parameter Sharing is used to identify the conductance ofmultiple branches simultaneously.Experiments show that themethod has good accuracy and robustness.Due to the consideration of spatial constraints,the method can also obtain more accurate conductance values under different training and testing conditions.
基金Project supported by the National Natural Science Foundation of China(Grant Nos.11404316 and 11304159)the Natural Science Foundation of Jiangsu Province+2 种基金Chinathe Science Foundation of Nanjing University of Posts and TelecommunicationsChina(Grant Nos.BK20140863 and NY213075)
文摘Single crystals of 4SC(NH2)2–Ni1-xCux Cl2(x = 0.03)(Cu-DTN) containing spin S = 1/2 Cu2+and S = 1 Ni2+cations are synthesized by slow evaporation methods. Structural characterization demonstrates that the Cu-DTN is of a tetrahedral structure with lattice parameter c being 9.0995 ?A, which is 1.32% expansion compared with that of parent material DTN due to the larger radius of the Cu ion. Direct current(DC) susceptibility measurements show that both the antiferromagnetic exchange interaction at low temperature and the large anisotropy of susceptibilities are suppressed after doping the Cu ion, which could be related to the structural distortion and the increase of the super-exchange paths in Cu-DTN.
基金Project supported by the National Natural Science Foundation of China(Grant No.61971229).
文摘Detailed density functional theory(DFT)calculations of the structural,mechanical,thermodynamic,and electronicproperties of crystalline CaF2 with five different structures in the pressure range of 0 GPa–150 GPa are performed byboth GGA(generalized gradient approximation)-PBE(Perdew–Burke–Ernzerhof)and LDA(local density approximation)-CAPZ(Cambridge Serial Total Energy Package).It is found that the enthalpy differences imply that the fluorite phase→PbCl2-type phase→Ni2In-type phase transition in CaF2 occurs at PGGA1=8.0 GPa,PGGA2=111.4 GPa by usingthe XC of GGA,and PLDA1=4.5 GPa,PLDA2=101.7 GPa by LDA,respectively,which is consistent with previousexperiments and theoretical conclusions.Moreover,the enthalpy differences between PbCl2-type and Ni2In-type phases inone molecular formula become very small at the pressure of about 100 GPa,indicating the possibility of coexistence of twophase at high pressures.This may be the reason why the transition pressure of the second phase transition in other reportsis so huge(68 GPa–278 GPa).The volume changed in the second phase transition are also consistent with the enthalpydifference result.Besides,the pressure dependence of mechanical and thermodynamic properties of CaF2 is studied.Itis found that the high-pressure phase of Ni2In-type structure has better stiffness in CaF2 crystal,and the hardness of thematerial has hardly changed in the second phase transition.Finally,the electronic structure of CaF2 is also analyzed withthe change of pressure.By analyzing the band gap and density of states,the large band gap indicates the CaF2 crystal isalways an insulator at 0 GPa–150 GPa.
基金The National Natural Science Foundation of China under contract No.61471024the National Marine Technology Program for Public Welfare under contract No.201505002-1the Beijing Higher Education Young Elite Teacher Project under contract No.YETP0514
文摘To dates,most ship detection approaches for single-pol synthetic aperture radar(SAR) imagery try to ensure a constant false-alarm rate(CFAR).A high performance ship detector relies on two key components:an accurate estimation to a sea surface distribution and a fine designed CFAR algorithm.First,a novel nonparametric sea surface distribution estimation method is developed based on n-order Bézier curve.To estimate the sea surface distribution using n-order Bézier curve,an explicit analytical solution is derived based on a least square optimization,and the optimal selection also is presented to two essential parameters,the order n of Bézier curve and the number m of sample points.Next,to validate the ship detection performance of the estimated sea surface distribution,the estimated sea surface distribution by n-order Bézier curve is combined with a cell averaging CFAR(CA-CFAR).To eliminate the possible interfering ship targets in background window,an improved automatic censoring method is applied.Comprehensive experiments prove that in terms of sea surface estimation performance,the proposed method is as good as a traditional nonparametric Parzen window kernel method,and in most cases,outperforms two widely used parametric methods,K and G0 models.In terms of computation speed,a major advantage of the proposed estimation method is the time consuming only depended on the number m of sample points while independent of imagery size,which makes it can achieve a significant speed improvement to the Parzen window kernel method,and in some cases,it is even faster than two parametric methods.In terms of ship detection performance,the experiments show that the ship detector which constructed by the proposed sea surface distribution model and the given CA-CFAR algorithm has wide adaptability to different SAR sensors,resolutions and sea surface homogeneities and obtains a leading performance on the test dataset.
文摘Overcharging is an important safety issue in the charging process of electric vehicle power batteries,and can easily lead to accelerated battery aging and serious safety accidents.It is necessary to accurately predict the vehicle’s charging time to effectively prevent the battery from overcharging.Due to the complex structure of the battery pack and various charging modes,the traditional charging time prediction method often encounters modeling difficulties and low accuracy.In response to the above problems,data drivers and machine learning theories are applied.On the basis of fully considering the different electric vehicle battery management system(BMS)charging modes,a charging time prediction method with charging mode recognition is proposed.First,an intelligent algorithm based on dynamic weighted density peak clustering(DWDPC)and random forest fusion is proposed to classify vehicle charging modes.Then,on the basis of an improved simplified particle swarm optimization(ISPSO)algorithm,a high-performance charging time prediction method is constructed by fully integrating long short-term memory(LSTM)and a strong tracking filter.Finally,the data run by the actual engineering system are verified for the proposed charging time prediction algorithm.Experimental results show that the new method can effectively distinguish the charging modes of different vehicles,identify the charging characteristics of different electric vehicles,and achieve high prediction accuracy.
基金This work was supported by the Guangdong Provincial Key Laboratory(No.2020B121201001)National Natural Science Foundation of China(NSFC)(Nos.61502239 and 62002148)+1 种基金Natural Science Foundation of Jiangsu Province of China(No.BK20150924)Shenzhen Science and Technology Program(No.KQTD2016112514355531).
文摘Waste collection is an important part of waste management system.Transportation costs and carbon emissions can be greatly reduced by proper vehicle routing.Meanwhile,each vehicle can work again after achieving its capacity limit and unloading the waste.For this,an energy-efficient multi-trip vehicle routing model is established for municipal solid waste collection,which incorporates practical factors like the limited capacity,maximum working hours,and multiple trips of each vehicle.Considering both economy and environment,fixed costs,fuel costs,and carbon emission costs are minimized together.To solve the formulated model effectively,contribution-based adaptive particle swarm optimization is proposed.Four strategies named greedy learning,multi-operator learning,exploring learning,and exploiting learning are specifically designed with their own searching priorities.By assessing the contribution of each learning strategy during the process of evolution,an appropriate one is selected and assigned to each individual adaptively to improve the searching efficiency of the algorithm.Moreover,an improved local search operator is performed on the trips with the largest number of waste sites so that both the exploiting ability and the convergence accuracy of the algorithm are improved.Performance of the proposed algorithm is tested on ten waste collection instances,which include one real-world case derived from the Green Ring Company of Jiangbei New District,Nanjing,China,and nine synthetic instances with increasing scales generated from the commonly-used capacitated vehicle routing problem benchmark datasets.Comparisons with five state-of-the-art algorithms show that the proposed algorithm can obtain a solution with a higher accuracy for the constructed model.
基金NSF of Jiangsu province(BK20170040)the National Natural Science Foundation of China(Grant Nos.61872189,61876088,61605083)+1 种基金the NSF of Jiangsu Higher Education Institutions of China(16KJB510023)Postgraduate Research&Practice Innovation Program of Jiangsu Province(KYCX17_0903).
文摘Despite the demonstrated success of numerous correlation filter(CF)based tracking approaches,their assumption of circulant structure of samples introduces significant redundancy to learn an effective classifier.In this paper,we develop a fast manifold regularized context-aware correlation tracking algorithm that mines the local manifold structure information of different types of samples.First,different from the traditional CF based tracking that only uses one base sample,we employ a set of contextual samples near to the base sample,and impose a manifold structure assumption on them.Afterwards,to take into account the manifold structure among these samples,we introduce a linear graph Laplacian regularized term into the objective of CF learning.Fortunately,the optimization can be efficiently solved in a closed form with fast Fourier transforms(FFTs),which contributes to a highly efficient implementation.Extensive evaluations on the OTB100 and VOT2016 datasets demonstrate that the proposed tracker performs favorably against several state-of-the-art algorithms in terms of accuracy and robustness.Especially,our tracker is able to run in real-time with 28 fps on a single CPU.
基金This work was supported by the Guangdong Provincial Key Laboratory(No.2020B121201001)the National Natural Science Foundation of China(NSFC)(Nos.61502239 and 62002148)+3 种基金Natural Science Foundation of Jiangsu Province of China(No.BK20150924)the Program for Guangdong Introducing Innovative and Enterpreneurial Teams(No.2017ZT07X386)Shenzhen Science and Technology Program(No.KQTD2016112514355531)Research Institute of Trustworthy Autonomous Systems(RITAS).
文摘A constrained multi-objective optimization model for the low-carbon vehicle routing problem(VRP)is established.A carbon emission measurement method considering various practical factors is introduced.It minimizes both the total carbon emissions and the longest time consumed by the sub-tours,subject to the limited number of available vehicles.According to the characteristics of the model,a region enhanced discrete multi-objective fireworks algorithm is proposed.A partial mapping explosion operator,a hybrid mutation for adjusting the sub-tours,and an objective-driven extending search are designed,which aim to improve the convergence,diversity,and spread of the non-dominated solutions produced by the algorithm,respectively.Nine low-carbon VRP instances with different scales are used to verify the effectiveness of the new strategies.Furthermore,comparison results with four state-of-the-art algorithms indicate that the proposed algorithm has better performance of convergence and distribution on the low-carbon VRP.It provides a promising scalability to the problem size.