Accurate capacity and State of Charge(SOC)estimation are crucial for ensuring the safety and longevity of lithium-ion batteries in electric vehicles.This study examines ten machine learning architectures,Including Dee...Accurate capacity and State of Charge(SOC)estimation are crucial for ensuring the safety and longevity of lithium-ion batteries in electric vehicles.This study examines ten machine learning architectures,Including Deep Belief Network(DBN),Bidirectional Recurrent Neural Network(BiDirRNN),Gated Recurrent Unit(GRU),and others using the NASA B0005 dataset of 591,458 instances.Results indicate that DBN excels in capacity estimation,achieving orders-of-magnitude lower error values and explaining over 99.97%of the predicted variable’s variance.When computational efficiency is paramount,the Deep Neural Network(DNN)offers a strong alternative,delivering near-competitive accuracy with significantly reduced prediction times.The GRU achieves the best overall performance for SOC estimation,attaining an R^(2) of 0.9999,while the BiDirRNN provides a marginally lower error at a slightly higher computational speed.In contrast,Convolutional Neural Networks(CNN)and Radial Basis Function Networks(RBFN)exhibit relatively high error rates,making them less viable for real-world battery management.Analyses of error distributions reveal that the top-performing models cluster most predictions within tight bounds,limiting the risk of overcharging or deep discharging.These findings highlight the trade-off between accuracy and computational overhead,offering valuable guidance for battery management system(BMS)designers seeking optimal performance under constrained resources.Future work may further explore advanced data augmentation and domain adaptation techniques to enhance these models’robustness in diverse operating conditions.展开更多
Power quality is a crucial area of research in contemporary power systems,particularly given the rapid proliferation of intermittent renewable energy sources such as wind power.This study investigated the relationship...Power quality is a crucial area of research in contemporary power systems,particularly given the rapid proliferation of intermittent renewable energy sources such as wind power.This study investigated the relationships between power quality indices of system output and PSD by utilizing theories related to spectra,PSD,and random signal power spectra.The relationship was derived,validated through experiments and simulations,and subsequently applied to multi-objective optimization.Various optimization algorithms were compared to achieve optimal system power quality.The findings revealed that the relationships between power quality indices and PSD were influenced by variations in the order of the power spectral estimation model.An increase in the order of the AR model resulted in a 36%improvement in the number of optimal solutions.Regarding optimal solution distribution,NSGA-II demonstrated superior diversity,while MOEA/D exhibited better convergence.However,practical applications showed that while MOEA/D had higher convergence,NSGA-II produced superior optimal solutions,achieving the best power quality indices(THDi at 4.62%,d%at 3.51%,and cosφat 96%).These results suggest that the proposed method holds significant potential for optimizing power quality in practical applications.展开更多
This study tracked the characteristics of atmospheric wet deposition of the toxic element arsenic(As)at both urban(Guangzhou(GZ))and forested(Dinghushan Natural Reserve(DHS))sites within the Pearl River Delta(PRD)regi...This study tracked the characteristics of atmospheric wet deposition of the toxic element arsenic(As)at both urban(Guangzhou(GZ))and forested(Dinghushan Natural Reserve(DHS))sites within the Pearl River Delta(PRD)region between 2016 and 2019,examining its correlation with rainfall patterns.Additionally,by employing backward trajectory analysis and the potential source contribution function(PSCF)in conjunction with pertinent emission inventories,we pinpointed the main pathways of atmospheric arsenic transport and evaluated the emission contributions from priority source areas.The study revealed that the atmospheric arsenic wet deposition fluxes at the GZ and DHS sites exhibited a trend of increase followed by a decrease over the four-year period.Wet season deposition fluxes were more than triple those of the dry season,with urban site showing a difference of over four times.Notably,wet season As deposition at both sites was predominantly affected by heavy rainfall from marine air masses,constituting 31%of the total deposition.The predominant trajectory directions contributing to arsenic deposition at GZ and DHS were northeast(55%)and south(53%),respectively.The primary source areas for both sites were largely outside the PRD region,with the GZ site having 80%to 95%of its source area in the non-PRD region,compared to 69%to 88%at the DHS site.Furthermore,non-PRD areas contributed approximately 65%to arsenic emissions for both sites,with the industrial sector being the dominant emission source,exceeding 97%of the total emissions.展开更多
Nested simulation encompasses the estimation of functionals linked to conditional expectations through simulation techniques.In this paper,we treat conditional expectation as a function of the multidimensional conditi...Nested simulation encompasses the estimation of functionals linked to conditional expectations through simulation techniques.In this paper,we treat conditional expectation as a function of the multidimensional conditioning variable and provide asymptotic analyses of general nonparametric least squared estimators on sieve,without imposing specific assumptions on the function’s form.Our study explores scenarios in which the convergence rate surpasses that of the standard Monte Carlo method and the one recently proposed based on kernel ridge regression.We use kernel ridge regression with inducing points and neural networks as examples to illustrate our theorems.Numerical experiments are conducted to support our statements.展开更多
Monitoring,understanding and predicting Origin-destination(OD)flows in a city is an important problem for city planning and human activity.Taxi-GPS traces,acted as one kind of typical crowd sensed data,it can be used ...Monitoring,understanding and predicting Origin-destination(OD)flows in a city is an important problem for city planning and human activity.Taxi-GPS traces,acted as one kind of typical crowd sensed data,it can be used to mine the semantics of OD flows.In this paper,we firstly construct and analyze a complex network of OD flows based on large-scale GPS taxi traces of a city in China.The spatiotemporal analysis for the OD flows complex network showed that there were distinctive patterns in OD flows.Then based on a novel complex network model,a semantics mining method of OD flows is proposed through compounding Points of Interests(POI)network and public transport network to the OD flows network.The propose method would offer a novel way to predict the location characteristic and future traffic conditions accurately.展开更多
Dear Editor, This letter deals with the problem of algorithm recommendation for online fault detection of spacecraft. By transforming the time series data into distributions and introducing a distribution-aware measur...Dear Editor, This letter deals with the problem of algorithm recommendation for online fault detection of spacecraft. By transforming the time series data into distributions and introducing a distribution-aware measure, a principal method is designed for quantifying the detectabilities of fault detection algorithms over special datasets.展开更多
Machine fault diagnostics are essential for industrial operations,and advancements in machine learning have significantly advanced these systems by providing accurate predictions and expedited solutions.Machine learni...Machine fault diagnostics are essential for industrial operations,and advancements in machine learning have significantly advanced these systems by providing accurate predictions and expedited solutions.Machine learning models,especially those utilizing complex algorithms like deep learning,have demonstrated major potential in extracting important information fromlarge operational datasets.Despite their efficiency,machine learningmodels face challenges,making Explainable AI(XAI)crucial for improving their understandability and fine-tuning.The importance of feature contribution and selection using XAI in the diagnosis of machine faults is examined in this study.The technique is applied to evaluate different machine-learning algorithms.Extreme Gradient Boosting,Support Vector Machine,Gaussian Naive Bayes,and Random Forest classifiers are used alongside Logistic Regression(LR)as a baseline model because their efficacy and simplicity are evaluated thoroughly with empirical analysis.The XAI is used as a targeted feature selection technique to select among 29 features of the time and frequency domain.The XAI approach is lightweight,trained with only targeted features,and achieved similar results as the traditional approach.The accuracy without XAI on baseline LR is 79.57%,whereas the approach with XAI on LR is 80.28%.展开更多
In this paper, a two-dimensional(2D) DOA estimation algorithm of coherent signals with a separated linear acoustic vector-sensor(AVS) array consisting of two sparse AVS arrays is proposed. Firstly,the partitioned spat...In this paper, a two-dimensional(2D) DOA estimation algorithm of coherent signals with a separated linear acoustic vector-sensor(AVS) array consisting of two sparse AVS arrays is proposed. Firstly,the partitioned spatial smoothing(PSS) technique is used to construct a block covariance matrix, so as to decorrelate the coherency of signals. Then a signal subspace can be obtained by singular value decomposition(SVD) of the covariance matrix. Using the signal subspace, two extended signal subspaces are constructed to compensate aperture loss caused by PSS.The elevation angles can be estimated by estimation of signal parameter via rotational invariance techniques(ESPRIT) algorithm. At last, the estimated elevation angles can be used to estimate automatically paired azimuth angles. Compared with some other ESPRIT algorithms, the proposed algorithm shows higher estimation accuracy, which can be proved through the simulation results.展开更多
This work presents an advanced and detailed analysis of the mechanisms of hepatitis B virus(HBV)propagation in an environment characterized by variability and stochas-ticity.Based on some biological features of the vi...This work presents an advanced and detailed analysis of the mechanisms of hepatitis B virus(HBV)propagation in an environment characterized by variability and stochas-ticity.Based on some biological features of the virus and the assumptions,the corresponding deterministic model is formulated,which takes into consideration the effect of vaccination.This deterministic model is extended to a stochastic framework by considering a new form of disturbance which makes it possible to simulate strong and significant fluctuations.The long-term behaviors of the virus are predicted by using stochastic differential equations with second-order multiplicative α-stable jumps.By developing the assumptions and employing the novel theoretical tools,the threshold parameter responsible for ergodicity(persistence)and extinction is provided.The theoretical results of the current study are validated by numerical simulations and parameters estimation is also performed.Moreover,we obtain the following new interesting findings:(a)in each class,the average time depends on the value ofα;(b)the second-order noise has an inverse effect on the spread of the virus;(c)the shapes of population densities at stationary level quickly changes at certain values of α.The last three conclusions can provide a solid research base for further investigation in the field of biological and ecological modeling.展开更多
The SiS molecule,which plays a significant role in space,has attracted a great deal of attention for many years.Due to complex interactions among its low-lying electronic states,precise information regarding the molec...The SiS molecule,which plays a significant role in space,has attracted a great deal of attention for many years.Due to complex interactions among its low-lying electronic states,precise information regarding the molecular structure of SiS is limited.To obtain accurate information about the structure of its excited states,the high-precision multireference configuration interaction(MRCI)method has been utilized.This method is used to calculate the potential energy curves(PECs)of the 18Λ–S states corresponding to the lowest dissociation limit of SiS.The core–valence correlation effect,Davidson’s correction and the scalar relativistic effect are also included to guarantee the precision of the MRCI calculation.Based on the calculated PECs,the spectroscopic constants of quasi-bound and bound electronic states are calculated and they are in accordance with previous experimental results.The transition dipole moments(TDMs)and dipole moments(DMs)are determined by the MRCI method.In addition,the abrupt variations of the DMs for the 1^(5)Σ^(+)and 2^(5)Σ^(+)states at the avoided crossing point are attributed to the variation of the electronic configuration.The opacity of SiS at a pressure of 100 atms is presented across a series of temperatures.With increasing temperature,the expanding population of excited states blurs the band boundaries.展开更多
Silicon monoxide(SiO)(silicon[Si]mixed with silicon dioxide[SiO_(2)])/graphite(Gr)composite material is one of the most commercially promising anode materials for the next generation of high-energy-density lithium-ion...Silicon monoxide(SiO)(silicon[Si]mixed with silicon dioxide[SiO_(2)])/graphite(Gr)composite material is one of the most commercially promising anode materials for the next generation of high-energy-density lithium-ion batteries.The major bottleneck for SiO/Gr composite anode is the poor cyclability arising from the stress/strain behaviors due to the mismatch between two heterogenous materials during the lithiation/delithiation process.To date,a meticulous and quantitative understanding of the highly nonlinear coupling behaviors of such materials is still lacking.Herein,an electro–chemo–mechanics-coupled detailed model containing particle geometries is established.The underlying mechanism of the regulation between SiO and Gr components during electrochemical cycling is quantitatively revealed.We discover that increasing the SiO weight percentage(wt%)reduces the utilization efficiency of the active materials at the same 1C rate charging and enhances the hindering effects of stress-driven flux on diffusion.In addition,the mechanical constraint demonstrates a balanced effect on the overall performance of cells and the local behaviors of particles.This study provides new insights into the fundamental interactions between SiO and Gr materials and advances the investigation methodology for the design and evaluation of next-generation high-energydensity batteries.展开更多
Dear Editor,In this letter,a novel data-driven adaptive predictive control method is proposed using the triangular dynamic linearization technique.The proposed method only contains one time-varying parameter with expl...Dear Editor,In this letter,a novel data-driven adaptive predictive control method is proposed using the triangular dynamic linearization technique.The proposed method only contains one time-varying parameter with explicit physical meaning,which can prevent severe deviation in parameter estimation.Specifically,a triangular dynamic linearization(TDL)data model is employed to predict future system outputs,and then to correct inaccurate predictive outputs,a feedback regulator is designed.An autotuned weighing factor is introduced to alleviate the computational burden in practical applications and further improve output tracking performance.Closed-loop stability conditions are derived by rigorous analysis.Simulation results are provided to demonstrate the efficacy of the proposed method.展开更多
News media profiling is helpful in preventing the spread of fake news at the source and maintaining a good media and news ecosystem.Most previous works only extract features and evaluate media from one dimension indep...News media profiling is helpful in preventing the spread of fake news at the source and maintaining a good media and news ecosystem.Most previous works only extract features and evaluate media from one dimension independently,ignoring the interconnections between different aspects.This paper proposes a novel news media bias and factuality profiling framework assisted by correlated features.This framework models the relationship and interaction between media bias and factuality,utilizing this relationship to assist in the prediction of profiling results.Our approach extracts features independently while aligning and fusing them through recursive convolu-tion and attention mechanisms,thus harnessing multi-scale interactive information across different dimensions and levels.This method improves the effectiveness of news media evaluation.Experimental results indicate that our proposed framework significantly outperforms existing methods,achieving the best performance in Accuracy and F1 score,improving by at least 1%compared to other methods.This paper further analyzes and discusses based on the experimental results.展开更多
Regularized system identification has become the research frontier of system identification in the past decade.One related core subject is to study the convergence properties of various hyper-parameter estimators as t...Regularized system identification has become the research frontier of system identification in the past decade.One related core subject is to study the convergence properties of various hyper-parameter estimators as the sample size goes to infinity.In this paper,we consider one commonly used hyper-parameter estimator,the empirical Bayes(EB).Its convergence in distribution has been studied,and the explicit expression of the covariance matrix of its limiting distribution has been given.However,what we are truly interested in are factors contained in the covariance matrix of the EB hyper-parameter estimator,and then,the convergence of its covariance matrix to that of its limiting distribution is required.In general,the convergence in distribution of a sequence of random variables does not necessarily guarantee the convergence of its covariance matrix.Thus,the derivation of such convergence is a necessary complement to our theoretical analysis about factors that influence the convergence properties of the EB hyper-parameter estimator.In this paper,we consider the regularized finite impulse response(FIR)model estimation with deterministic inputs,and show that the covariance matrix of the EB hyper-parameter estimator converges to that of its limiting distribution.Moreover,we run numerical simulations to demonstrate the efficacy of ourtheoretical results.展开更多
The assembly of a protein complex is very important for its biological function,which can be investigated by determining the order of assembly/disassembly of its protein subunits.Although static structures of many pro...The assembly of a protein complex is very important for its biological function,which can be investigated by determining the order of assembly/disassembly of its protein subunits.Although static structures of many protein com-plexes are available in the protein data bank,their assembly/disassembly orders of subunits are largely unknown.In addition to experimental techniques for studying subcomplexes in the assembly/disassembly of a protein complex,computational methods can be used to predict the assembly/disassembly order.Since sampling is a nontrivial issue in simulating the assembly/disassembly process,coarse-grained simulations are more efficient than atomic simulations are.In this work,we developed computational protocols for predicting the assembly/disassembly orders of protein complexes via coarse-grained simulations.The protocols were illustrated via two protein complexes,and the predicted assembly/disassembly orders were consistent with the available experimental data.展开更多
This paper addresses the problem of containment control for heterogeneous multi-agent systems subject to Markovian randomly switching topologies and unbounded communication delays.The objective is to design a distribu...This paper addresses the problem of containment control for heterogeneous multi-agent systems subject to Markovian randomly switching topologies and unbounded communication delays.The objective is to design a distributed control strategy that ensures the output of each follower converges to the convex hull formed by the outputs of a group of leaders in mean square sense.A novel distributed observer is proposed by tackling both Markovian randomly switching topologies and unbounded delays.Then,a distributed state feedback controller and a distributed output feedback controller are developed based on the distributed observer,respectively.Finally,simulation results are provided to demonstrate the effectiveness of the proposed controllers.展开更多
For given simple graphs H1,H2,...,Hc,the multicolor Ramsey number R(H1,H2,...,Hc) is defined as the smallest positive integer n such that for an arbitrary edge-decomposition{Gi}ci=1of the complete graph K_n,at least o...For given simple graphs H1,H2,...,Hc,the multicolor Ramsey number R(H1,H2,...,Hc) is defined as the smallest positive integer n such that for an arbitrary edge-decomposition{Gi}ci=1of the complete graph K_n,at least one Gihas a subgraph isomorphic to Hi.Let m,n1,n2,...,nc be positive integers andΣ=Σci=1(ni-1).Some bounds and exact values of R(K1,n1,...,K1,nc,Pm) have been obtained in literature.Wang (Graphs Combin.,2020) conjectured that ifΣ?≡0 (mod m-1) andΣ+1≥(m-3)2,then R(K1,n1,...,K1,nc,Pm)=Σ+m-1.In this note,we give a new lower bound and some exact values of R(K1,n1,...,K1,nc,Pm) provided m≤Σ,Σ≡k (mod m-1),and 2≤k≤m-2.These results partially confirm Wang’s conjecture.展开更多
Chemical oxygen demand (COD) is an important index to measure the degree of water pollution. In this paper, near-infrared technology is used to obtain 148 wastewater spectra to predict the COD value in wastewater. Fir...Chemical oxygen demand (COD) is an important index to measure the degree of water pollution. In this paper, near-infrared technology is used to obtain 148 wastewater spectra to predict the COD value in wastewater. First, the partial least squares regression (PLS) model was used as the basic model. Monte Carlo cross-validation (MCCV) was used to select 25 samples out of 148 samples that did not conform to conventional statistics. Then, the interval partial least squares (iPLS) regression modeling was carried out on 123 samples, and the spectral bands were divided into 40 subintervals. The optimal subintervals are 20 and 26, and the optimal correlation coefficient of the test set (RT) is 0.58. Further, the waveband is divided into five intervals: 17, 19, 20, 22 and 26. When the number of joint intervals under each interval is three, the optimal RT is 0.71. When the number of joint subintervals is four, the optimal RT is 0.79. Finally, convolutional neural network (CNN) was used for quantitative prediction, and RT was 0.9. The results show that CNN can automatically screen the features inside the data, and the quantitative prediction effect is better than that of iPLS and synergy interval partial least squares model (SiPLS) with joint subinterval three and four, indicating that CNN can be used for quantitative analysis of water pollution degree.展开更多
The characteristics of interval systems under the framework of extremum algebra is a meaningful research direction.In this paper we study the characterizations of various kinds of solvabilities of interval linear ineq...The characteristics of interval systems under the framework of extremum algebra is a meaningful research direction.In this paper we study the characterizations of various kinds of solvabilities of interval linear inequalities in the max-min algebra,including the weak solvability,strong solvability,tolerance solvability,strongly tolerance solvability,control solvability and strongly control solvability.Furthermore,we analyze the existence of solutions and the solvabilities,and show these two concepts are equivalent in certain situations.In addition,we find the maximum solutions corresponding to different kinds of solvabilities.展开更多
The proliferation of robot accounts on social media platforms has posed a significant negative impact,necessitating robust measures to counter network anomalies and safeguard content integrity.Social robot detection h...The proliferation of robot accounts on social media platforms has posed a significant negative impact,necessitating robust measures to counter network anomalies and safeguard content integrity.Social robot detection has emerged as a pivotal yet intricate task,aimed at mitigating the dissemination of misleading information.While graphbased approaches have attained remarkable performance in this realm,they grapple with a fundamental limitation:the homogeneity assumption in graph convolution allows social robots to stealthily evade detection by mingling with genuine human profiles.To unravel this challenge and thwart the camouflage tactics,this work proposed an innovative social robot detection framework based on enhanced HOmogeneity and Random Forest(HORFBot).At the core of HORFBot lies a homogeneous graph enhancement strategy,intricately woven with edge-removal techniques,tometiculously dissect the graph intomultiple revealing subgraphs.Subsequently,leveraging the power of contrastive learning,the proposed methodology meticulously trains multiple graph convolutional networks,each honed to discern nuances within these tailored subgraphs.The culminating stage involves the fusion of these feature-rich base classifiers,harmoniously aggregating their insights to produce a comprehensive detection outcome.Extensive experiments on three social robot detection datasets have shown that this method effectively improves the accuracy of social robot detection and outperforms comparative methods.展开更多
文摘Accurate capacity and State of Charge(SOC)estimation are crucial for ensuring the safety and longevity of lithium-ion batteries in electric vehicles.This study examines ten machine learning architectures,Including Deep Belief Network(DBN),Bidirectional Recurrent Neural Network(BiDirRNN),Gated Recurrent Unit(GRU),and others using the NASA B0005 dataset of 591,458 instances.Results indicate that DBN excels in capacity estimation,achieving orders-of-magnitude lower error values and explaining over 99.97%of the predicted variable’s variance.When computational efficiency is paramount,the Deep Neural Network(DNN)offers a strong alternative,delivering near-competitive accuracy with significantly reduced prediction times.The GRU achieves the best overall performance for SOC estimation,attaining an R^(2) of 0.9999,while the BiDirRNN provides a marginally lower error at a slightly higher computational speed.In contrast,Convolutional Neural Networks(CNN)and Radial Basis Function Networks(RBFN)exhibit relatively high error rates,making them less viable for real-world battery management.Analyses of error distributions reveal that the top-performing models cluster most predictions within tight bounds,limiting the risk of overcharging or deep discharging.These findings highlight the trade-off between accuracy and computational overhead,offering valuable guidance for battery management system(BMS)designers seeking optimal performance under constrained resources.Future work may further explore advanced data augmentation and domain adaptation techniques to enhance these models’robustness in diverse operating conditions.
基金funded by the Inner Mongolia Nature Foundation Project,Project number:2023JQ04.
文摘Power quality is a crucial area of research in contemporary power systems,particularly given the rapid proliferation of intermittent renewable energy sources such as wind power.This study investigated the relationships between power quality indices of system output and PSD by utilizing theories related to spectra,PSD,and random signal power spectra.The relationship was derived,validated through experiments and simulations,and subsequently applied to multi-objective optimization.Various optimization algorithms were compared to achieve optimal system power quality.The findings revealed that the relationships between power quality indices and PSD were influenced by variations in the order of the power spectral estimation model.An increase in the order of the AR model resulted in a 36%improvement in the number of optimal solutions.Regarding optimal solution distribution,NSGA-II demonstrated superior diversity,while MOEA/D exhibited better convergence.However,practical applications showed that while MOEA/D had higher convergence,NSGA-II produced superior optimal solutions,achieving the best power quality indices(THDi at 4.62%,d%at 3.51%,and cosφat 96%).These results suggest that the proposed method holds significant potential for optimizing power quality in practical applications.
基金supported by the National Natural Science Foundation of China(Nos.42121004,42275107,and 42077205)the National Key Research and Development Plan(No.2023YFC3706202)+1 种基金the Foundational and Applied Basic Research in Guangzhou in 2023(No.2023A04J0251)the Special Fund Project for Science and Technology Innovation Strategy of Guangdong Province(No.2019B121205004)。
文摘This study tracked the characteristics of atmospheric wet deposition of the toxic element arsenic(As)at both urban(Guangzhou(GZ))and forested(Dinghushan Natural Reserve(DHS))sites within the Pearl River Delta(PRD)region between 2016 and 2019,examining its correlation with rainfall patterns.Additionally,by employing backward trajectory analysis and the potential source contribution function(PSCF)in conjunction with pertinent emission inventories,we pinpointed the main pathways of atmospheric arsenic transport and evaluated the emission contributions from priority source areas.The study revealed that the atmospheric arsenic wet deposition fluxes at the GZ and DHS sites exhibited a trend of increase followed by a decrease over the four-year period.Wet season deposition fluxes were more than triple those of the dry season,with urban site showing a difference of over four times.Notably,wet season As deposition at both sites was predominantly affected by heavy rainfall from marine air masses,constituting 31%of the total deposition.The predominant trajectory directions contributing to arsenic deposition at GZ and DHS were northeast(55%)and south(53%),respectively.The primary source areas for both sites were largely outside the PRD region,with the GZ site having 80%to 95%of its source area in the non-PRD region,compared to 69%to 88%at the DHS site.Furthermore,non-PRD areas contributed approximately 65%to arsenic emissions for both sites,with the industrial sector being the dominant emission source,exceeding 97%of the total emissions.
文摘Nested simulation encompasses the estimation of functionals linked to conditional expectations through simulation techniques.In this paper,we treat conditional expectation as a function of the multidimensional conditioning variable and provide asymptotic analyses of general nonparametric least squared estimators on sieve,without imposing specific assumptions on the function’s form.Our study explores scenarios in which the convergence rate surpasses that of the standard Monte Carlo method and the one recently proposed based on kernel ridge regression.We use kernel ridge regression with inducing points and neural networks as examples to illustrate our theorems.Numerical experiments are conducted to support our statements.
基金This work is supported by Shandong Provincial Natural Science Foundation,China under Grant No.ZR2017MG011This work is also supported by Key Research and Development Program in Shandong Provincial(2017GGX90103).
文摘Monitoring,understanding and predicting Origin-destination(OD)flows in a city is an important problem for city planning and human activity.Taxi-GPS traces,acted as one kind of typical crowd sensed data,it can be used to mine the semantics of OD flows.In this paper,we firstly construct and analyze a complex network of OD flows based on large-scale GPS taxi traces of a city in China.The spatiotemporal analysis for the OD flows complex network showed that there were distinctive patterns in OD flows.Then based on a novel complex network model,a semantics mining method of OD flows is proposed through compounding Points of Interests(POI)network and public transport network to the OD flows network.The propose method would offer a novel way to predict the location characteristic and future traffic conditions accurately.
基金supported by the National Key R&D Program of China (2021YFB1715000)the National Natural Science Foundation of China (U1811461, 62022013, 12150007, 62103450, 61832003, 62272137)。
文摘Dear Editor, This letter deals with the problem of algorithm recommendation for online fault detection of spacecraft. By transforming the time series data into distributions and introducing a distribution-aware measure, a principal method is designed for quantifying the detectabilities of fault detection algorithms over special datasets.
基金funded by Woosong University Academic Research 2024.
文摘Machine fault diagnostics are essential for industrial operations,and advancements in machine learning have significantly advanced these systems by providing accurate predictions and expedited solutions.Machine learning models,especially those utilizing complex algorithms like deep learning,have demonstrated major potential in extracting important information fromlarge operational datasets.Despite their efficiency,machine learningmodels face challenges,making Explainable AI(XAI)crucial for improving their understandability and fine-tuning.The importance of feature contribution and selection using XAI in the diagnosis of machine faults is examined in this study.The technique is applied to evaluate different machine-learning algorithms.Extreme Gradient Boosting,Support Vector Machine,Gaussian Naive Bayes,and Random Forest classifiers are used alongside Logistic Regression(LR)as a baseline model because their efficacy and simplicity are evaluated thoroughly with empirical analysis.The XAI is used as a targeted feature selection technique to select among 29 features of the time and frequency domain.The XAI approach is lightweight,trained with only targeted features,and achieved similar results as the traditional approach.The accuracy without XAI on baseline LR is 79.57%,whereas the approach with XAI on LR is 80.28%.
基金supported by the National Natural Science Foundation of China (62261047,62066040)the Foundation of Top-notch Talents by Education Department of Guizhou Province of China (KY[2018]075)+3 种基金the Science and Technology Foundation of Guizhou Province of China (ZK[2022]557,[2020]1Y004)the Science and Technology Research Program of the Chongqing Municipal Education Commission (KJQN202200637)PhD Research Start-up Foundation of Tongren University (trxyDH1710)Tongren Science and Technology Planning Project ((2018)22)。
文摘In this paper, a two-dimensional(2D) DOA estimation algorithm of coherent signals with a separated linear acoustic vector-sensor(AVS) array consisting of two sparse AVS arrays is proposed. Firstly,the partitioned spatial smoothing(PSS) technique is used to construct a block covariance matrix, so as to decorrelate the coherency of signals. Then a signal subspace can be obtained by singular value decomposition(SVD) of the covariance matrix. Using the signal subspace, two extended signal subspaces are constructed to compensate aperture loss caused by PSS.The elevation angles can be estimated by estimation of signal parameter via rotational invariance techniques(ESPRIT) algorithm. At last, the estimated elevation angles can be used to estimate automatically paired azimuth angles. Compared with some other ESPRIT algorithms, the proposed algorithm shows higher estimation accuracy, which can be proved through the simulation results.
基金supported by the NSFC(12201557)the Foundation of Zhejiang Provincial Education Department,China(Y202249921).
文摘This work presents an advanced and detailed analysis of the mechanisms of hepatitis B virus(HBV)propagation in an environment characterized by variability and stochas-ticity.Based on some biological features of the virus and the assumptions,the corresponding deterministic model is formulated,which takes into consideration the effect of vaccination.This deterministic model is extended to a stochastic framework by considering a new form of disturbance which makes it possible to simulate strong and significant fluctuations.The long-term behaviors of the virus are predicted by using stochastic differential equations with second-order multiplicative α-stable jumps.By developing the assumptions and employing the novel theoretical tools,the threshold parameter responsible for ergodicity(persistence)and extinction is provided.The theoretical results of the current study are validated by numerical simulations and parameters estimation is also performed.Moreover,we obtain the following new interesting findings:(a)in each class,the average time depends on the value ofα;(b)the second-order noise has an inverse effect on the spread of the virus;(c)the shapes of population densities at stationary level quickly changes at certain values of α.The last three conclusions can provide a solid research base for further investigation in the field of biological and ecological modeling.
基金Project supported by the Natural Science Foundation of Heilongjiang Province,China(Grant No.LH2022A026)the National Key Research and Development Program of China(Grant No.2022YFA1602500)+2 种基金the National Natural Science Foundation of China(Grant No.11934004)Fundamental Research Funds in Heilongjiang Province Universities,China(Grant No.145109309)Foundation of National Key Laboratory of Computational Physics(Grant No.6142A05QN22006)。
文摘The SiS molecule,which plays a significant role in space,has attracted a great deal of attention for many years.Due to complex interactions among its low-lying electronic states,precise information regarding the molecular structure of SiS is limited.To obtain accurate information about the structure of its excited states,the high-precision multireference configuration interaction(MRCI)method has been utilized.This method is used to calculate the potential energy curves(PECs)of the 18Λ–S states corresponding to the lowest dissociation limit of SiS.The core–valence correlation effect,Davidson’s correction and the scalar relativistic effect are also included to guarantee the precision of the MRCI calculation.Based on the calculated PECs,the spectroscopic constants of quasi-bound and bound electronic states are calculated and they are in accordance with previous experimental results.The transition dipole moments(TDMs)and dipole moments(DMs)are determined by the MRCI method.In addition,the abrupt variations of the DMs for the 1^(5)Σ^(+)and 2^(5)Σ^(+)states at the avoided crossing point are attributed to the variation of the electronic configuration.The opacity of SiS at a pressure of 100 atms is presented across a series of temperatures.With increasing temperature,the expanding population of excited states blurs the band boundaries.
文摘Silicon monoxide(SiO)(silicon[Si]mixed with silicon dioxide[SiO_(2)])/graphite(Gr)composite material is one of the most commercially promising anode materials for the next generation of high-energy-density lithium-ion batteries.The major bottleneck for SiO/Gr composite anode is the poor cyclability arising from the stress/strain behaviors due to the mismatch between two heterogenous materials during the lithiation/delithiation process.To date,a meticulous and quantitative understanding of the highly nonlinear coupling behaviors of such materials is still lacking.Herein,an electro–chemo–mechanics-coupled detailed model containing particle geometries is established.The underlying mechanism of the regulation between SiO and Gr components during electrochemical cycling is quantitatively revealed.We discover that increasing the SiO weight percentage(wt%)reduces the utilization efficiency of the active materials at the same 1C rate charging and enhances the hindering effects of stress-driven flux on diffusion.In addition,the mechanical constraint demonstrates a balanced effect on the overall performance of cells and the local behaviors of particles.This study provides new insights into the fundamental interactions between SiO and Gr materials and advances the investigation methodology for the design and evaluation of next-generation high-energydensity batteries.
基金supported in part by the National Natural Science Foundation of China(62173002,52301408,62173255)the Beijing Natural Science Foundation(4222045).
文摘Dear Editor,In this letter,a novel data-driven adaptive predictive control method is proposed using the triangular dynamic linearization technique.The proposed method only contains one time-varying parameter with explicit physical meaning,which can prevent severe deviation in parameter estimation.Specifically,a triangular dynamic linearization(TDL)data model is employed to predict future system outputs,and then to correct inaccurate predictive outputs,a feedback regulator is designed.An autotuned weighing factor is introduced to alleviate the computational burden in practical applications and further improve output tracking performance.Closed-loop stability conditions are derived by rigorous analysis.Simulation results are provided to demonstrate the efficacy of the proposed method.
基金funded by“the Fundamental Research Funds for the Central Universities”,No.CUC23ZDTJ005.
文摘News media profiling is helpful in preventing the spread of fake news at the source and maintaining a good media and news ecosystem.Most previous works only extract features and evaluate media from one dimension independently,ignoring the interconnections between different aspects.This paper proposes a novel news media bias and factuality profiling framework assisted by correlated features.This framework models the relationship and interaction between media bias and factuality,utilizing this relationship to assist in the prediction of profiling results.Our approach extracts features independently while aligning and fusing them through recursive convolu-tion and attention mechanisms,thus harnessing multi-scale interactive information across different dimensions and levels.This method improves the effectiveness of news media evaluation.Experimental results indicate that our proposed framework significantly outperforms existing methods,achieving the best performance in Accuracy and F1 score,improving by at least 1%compared to other methods.This paper further analyzes and discusses based on the experimental results.
基金supported in part by the National Natural Science Foundation of China(No.62273287)by the Shenzhen Science and Technology Innovation Council(Nos.JCYJ20220530143418040,JCY20170411102101881)the Thousand Youth Talents Plan funded by the central government of China.
文摘Regularized system identification has become the research frontier of system identification in the past decade.One related core subject is to study the convergence properties of various hyper-parameter estimators as the sample size goes to infinity.In this paper,we consider one commonly used hyper-parameter estimator,the empirical Bayes(EB).Its convergence in distribution has been studied,and the explicit expression of the covariance matrix of its limiting distribution has been given.However,what we are truly interested in are factors contained in the covariance matrix of the EB hyper-parameter estimator,and then,the convergence of its covariance matrix to that of its limiting distribution is required.In general,the convergence in distribution of a sequence of random variables does not necessarily guarantee the convergence of its covariance matrix.Thus,the derivation of such convergence is a necessary complement to our theoretical analysis about factors that influence the convergence properties of the EB hyper-parameter estimator.In this paper,we consider the regularized finite impulse response(FIR)model estimation with deterministic inputs,and show that the covariance matrix of the EB hyper-parameter estimator converges to that of its limiting distribution.Moreover,we run numerical simulations to demonstrate the efficacy of ourtheoretical results.
基金This work was supported by the National Key Research and Development Program of China(2021YFA1301504)the Chinese Academy of Sciences Strategic Priority Research Program(XDB37040202)the National Natural Science Foundation of China(91953101).
文摘The assembly of a protein complex is very important for its biological function,which can be investigated by determining the order of assembly/disassembly of its protein subunits.Although static structures of many protein com-plexes are available in the protein data bank,their assembly/disassembly orders of subunits are largely unknown.In addition to experimental techniques for studying subcomplexes in the assembly/disassembly of a protein complex,computational methods can be used to predict the assembly/disassembly order.Since sampling is a nontrivial issue in simulating the assembly/disassembly process,coarse-grained simulations are more efficient than atomic simulations are.In this work,we developed computational protocols for predicting the assembly/disassembly orders of protein complexes via coarse-grained simulations.The protocols were illustrated via two protein complexes,and the predicted assembly/disassembly orders were consistent with the available experimental data.
文摘This paper addresses the problem of containment control for heterogeneous multi-agent systems subject to Markovian randomly switching topologies and unbounded communication delays.The objective is to design a distributed control strategy that ensures the output of each follower converges to the convex hull formed by the outputs of a group of leaders in mean square sense.A novel distributed observer is proposed by tackling both Markovian randomly switching topologies and unbounded delays.Then,a distributed state feedback controller and a distributed output feedback controller are developed based on the distributed observer,respectively.Finally,simulation results are provided to demonstrate the effectiveness of the proposed controllers.
基金National Natural Science Foundation of China (Grant No. 12071453)the National Key R and D Program of China (Grant No. 2020YFA0713100)the Innovation Program for Quantum Science and Technology (Grant No. 2021ZD0302902)。
文摘For given simple graphs H1,H2,...,Hc,the multicolor Ramsey number R(H1,H2,...,Hc) is defined as the smallest positive integer n such that for an arbitrary edge-decomposition{Gi}ci=1of the complete graph K_n,at least one Gihas a subgraph isomorphic to Hi.Let m,n1,n2,...,nc be positive integers andΣ=Σci=1(ni-1).Some bounds and exact values of R(K1,n1,...,K1,nc,Pm) have been obtained in literature.Wang (Graphs Combin.,2020) conjectured that ifΣ?≡0 (mod m-1) andΣ+1≥(m-3)2,then R(K1,n1,...,K1,nc,Pm)=Σ+m-1.In this note,we give a new lower bound and some exact values of R(K1,n1,...,K1,nc,Pm) provided m≤Σ,Σ≡k (mod m-1),and 2≤k≤m-2.These results partially confirm Wang’s conjecture.
文摘Chemical oxygen demand (COD) is an important index to measure the degree of water pollution. In this paper, near-infrared technology is used to obtain 148 wastewater spectra to predict the COD value in wastewater. First, the partial least squares regression (PLS) model was used as the basic model. Monte Carlo cross-validation (MCCV) was used to select 25 samples out of 148 samples that did not conform to conventional statistics. Then, the interval partial least squares (iPLS) regression modeling was carried out on 123 samples, and the spectral bands were divided into 40 subintervals. The optimal subintervals are 20 and 26, and the optimal correlation coefficient of the test set (RT) is 0.58. Further, the waveband is divided into five intervals: 17, 19, 20, 22 and 26. When the number of joint intervals under each interval is three, the optimal RT is 0.71. When the number of joint subintervals is four, the optimal RT is 0.79. Finally, convolutional neural network (CNN) was used for quantitative prediction, and RT was 0.9. The results show that CNN can automatically screen the features inside the data, and the quantitative prediction effect is better than that of iPLS and synergy interval partial least squares model (SiPLS) with joint subinterval three and four, indicating that CNN can be used for quantitative analysis of water pollution degree.
基金supported by the Natural Science Foundation of Zhejiang Province(No.LY21A010021)。
文摘The characteristics of interval systems under the framework of extremum algebra is a meaningful research direction.In this paper we study the characterizations of various kinds of solvabilities of interval linear inequalities in the max-min algebra,including the weak solvability,strong solvability,tolerance solvability,strongly tolerance solvability,control solvability and strongly control solvability.Furthermore,we analyze the existence of solutions and the solvabilities,and show these two concepts are equivalent in certain situations.In addition,we find the maximum solutions corresponding to different kinds of solvabilities.
基金Funds for the Central Universities(grant number CUC24SG018).
文摘The proliferation of robot accounts on social media platforms has posed a significant negative impact,necessitating robust measures to counter network anomalies and safeguard content integrity.Social robot detection has emerged as a pivotal yet intricate task,aimed at mitigating the dissemination of misleading information.While graphbased approaches have attained remarkable performance in this realm,they grapple with a fundamental limitation:the homogeneity assumption in graph convolution allows social robots to stealthily evade detection by mingling with genuine human profiles.To unravel this challenge and thwart the camouflage tactics,this work proposed an innovative social robot detection framework based on enhanced HOmogeneity and Random Forest(HORFBot).At the core of HORFBot lies a homogeneous graph enhancement strategy,intricately woven with edge-removal techniques,tometiculously dissect the graph intomultiple revealing subgraphs.Subsequently,leveraging the power of contrastive learning,the proposed methodology meticulously trains multiple graph convolutional networks,each honed to discern nuances within these tailored subgraphs.The culminating stage involves the fusion of these feature-rich base classifiers,harmoniously aggregating their insights to produce a comprehensive detection outcome.Extensive experiments on three social robot detection datasets have shown that this method effectively improves the accuracy of social robot detection and outperforms comparative methods.
