A nonlinear data analysis algorithm, namely empirical data decomposition （EDD） is proposed, which can perform adaptive analysis of observed data. Analysis filter, which is not a linear constant coefficient filter, i...A nonlinear data analysis algorithm, namely empirical data decomposition （EDD） is proposed, which can perform adaptive analysis of observed data. Analysis filter, which is not a linear constant coefficient filter, is automatically determined by observed data, and is able to implement multi-resolution analysis as wavelet transform. The algorithm is suitable for analyzing non-stationary data and can effectively wipe off the relevance of observed data. Then through discussing the applications of EDD in image compression, the paper presents a 2-dimension data decomposition framework and makes some modifications of contexts used by Embedded Block Coding with Optimized Truncation （EBCOT） . Simulation results show that EDD is more suitable for non-stationary image data compression.展开更多
Non-negative matrix factorization （NMF） is a technique for dimensionality reduction by placing non-negativity constraints on the matrix. Based on the PARAFAC model, NMF was extended for three-dimension data decompos...Non-negative matrix factorization （NMF） is a technique for dimensionality reduction by placing non-negativity constraints on the matrix. Based on the PARAFAC model, NMF was extended for three-dimension data decomposition. The three-dimension nonnegative matrix factorization （NMF3） algorithm, which was concise and easy to implement, was given in this paper. The NMF3 algorithm implementation was based on elements but not on vectors. It could decompose a data array directly without unfolding, which was not similar to that the traditional algorithms do, It has been applied to the simulated data array decomposition and obtained reasonable results. It showed that NMF3 could be introduced for curve resolution in chemometrics.展开更多
Since chemical processes are highly non-linear and multiscale,it is vital to deeply mine the multiscale coupling relationships embedded in the massive process data for the prediction and anomaly tracing of crucial pro...Since chemical processes are highly non-linear and multiscale,it is vital to deeply mine the multiscale coupling relationships embedded in the massive process data for the prediction and anomaly tracing of crucial process parameters and production indicators.While the integrated method of adaptive signal decomposition combined with time series models could effectively predict process variables,it does have limitations in capturing the high-frequency detail of the operation state when applied to complex chemical processes.In light of this,a novel Multiscale Multi-radius Multi-step Convolutional Neural Network(Msrt Net)is proposed for mining spatiotemporal multiscale information.First,the industrial data from the Fluid Catalytic Cracking(FCC)process decomposition using Complete Ensemble Empirical Mode Decomposition with Adaptive Noise(CEEMDAN)extract the multi-energy scale information of the feature subset.Then,convolution kernels with varying stride and padding structures are established to decouple the long-period operation process information encapsulated within the multi-energy scale data.Finally,a reconciliation network is trained to reconstruct the multiscale prediction results and obtain the final output.Msrt Net is initially assessed for its capability to untangle the spatiotemporal multiscale relationships among variables in the Tennessee Eastman Process(TEP).Subsequently,the performance of Msrt Net is evaluated in predicting product yield for a 2.80×10^(6) t/a FCC unit,taking diesel and gasoline yield as examples.In conclusion,Msrt Net can decouple and effectively extract spatiotemporal multiscale information from chemical process data and achieve a approximately reduction of 30%in prediction error compared to other time-series models.Furthermore,its robustness and transferability underscore its promising potential for broader applications.展开更多
The paper deals with the vehicle scheduling problem related to regional public transport. Linear programming methods are used to solve the problem. A mathematical model is created including the constraints and the obj...The paper deals with the vehicle scheduling problem related to regional public transport. Linear programming methods are used to solve the problem. A mathematical model is created including the constraints and the objective function minimizing costs and the number of vehicles. A minimum costs and a number of vehicles are forced at the same time by special economical input data analysis and an allocation of costs. Determining of the costs coefficients is done by three methods, which differs primarily by how much of the total costs they take into account. The decomposition of the set of lines into disjoint subsets can be used instead of the ＂direct＂ optimization. The decomposition has proven to be a suitable alternative in solving large optimization problems. The problem was applied to optimize vehicle scheduling in the region, which is situated in the north-east of the Czech Republic. There is used Xpress-IVE software, which solve the problem by simplex algorithm and branch and bound method. Research results show that there are large reserves in the organization of public transport. The implementation of the new vehicle scheduling would bring significant costs reductions in amount of at least 10% for the optimal solution and in amount of about 10% for the decomposition solution. The number of drivers could be decreased and the total time of the vehicles being outside the garage could be also reduced by at least 10%.展开更多
The paper surveys interactions between complex and functional-analytic methods in the CauchyKovalevskaya theory. For instance, the behaviour of the derivative of a bounded holomorphic function led to abstract versions...The paper surveys interactions between complex and functional-analytic methods in the CauchyKovalevskaya theory. For instance, the behaviour of the derivative of a bounded holomorphic function led to abstract versions of the Cauchy-Kovalevskaya Theorem.Recent trends in the Cauchy-Kovalevskaya theory are based on the concept of associated differential operators. Since an evolution operator may posses several associated operators, initial data may be decomposed into components belonging to different associated spaces.This technique makes it also possible to solve ill-posed initial value problems.展开更多
For describing target motion in hypersonic vehicle defense,a parametric analyzing and modeling method on ballistic data is proposed based on time varying auto-regressive method.Ballistic data are regarded as non-stati...For describing target motion in hypersonic vehicle defense,a parametric analyzing and modeling method on ballistic data is proposed based on time varying auto-regressive method.Ballistic data are regarded as non-stationary random signal,where the hidden internal law is studied.Firstly,ballistic data are decomposed into smooth linear trend signal and non-stationary periodic skip signal with ensemble empirical mode decomposition method to avoid mutual interference between different modal data.Secondly,the linear trend signal and the periodic skip signal are modeled separately.The linear trend signal is approximated by power function regressive estimator and the periodic skip signal is modeled based on time varying auto-regressive method.In order to determine optimal model orders,a novel method is presented based on information theoretic criteria and the criteria of minimizing the mean absolute error.Finally,the consistency test is conducted by investigating the time-frequency spectrum characteristics and statistical properties of outputs of the parametric model established above and dynamics model under the same initial condition.Simulation results demonstrate that the parametric model established by the proposed method shares a high consistency with the original dynamics model.展开更多
Data envelopment analysis （DEA） is an effective non-parametric method for measuring the relative efficiencies of decision making units （DMUs） with multiple inputs and outputs. In many real situations, the internal...Data envelopment analysis （DEA） is an effective non-parametric method for measuring the relative efficiencies of decision making units （DMUs） with multiple inputs and outputs. In many real situations, the internal structure of DMUs is a two-stage network process with shared inputs used in both stages and common outputs produced by the both stages. For example, hospitals have a two-stage network structure. Stage 1 consumes resources such as information technology system, plant, equipment and admin personnel to generate outputs such as medical records, laundry and housekeeping. Stage 2 consumes the same set of resources used by stage 1 （named shared inputs） and the outputs generated by stage 1 （named intermediate measures） to provide patient services. Besides, some of outputs, for instance, patient satisfaction degrees, are generated by the two individual stages together （named shared outputs）. Since some of shared inputs and outputs are hard split up and allocated to each individual stage, it needs to develop two-stage DEA methods for evaluating the performance of two-stage network processes in such problems. This paper extends the centralized model to measure the DEA efficiency of the two-stage process with non split-table shared inputs and outputs. A weighted additive approach is used to combine the two individual stages. Moreover, additive efficiency decomposition models are developed to simultaneously evaluate the maximal and the minimal achievable efficiencies for the individual stages. Finally, an example of 17 city branches of China Construction Bank in Anhui Province is employed to illustrate the proposed approach.展开更多
The significant wave height prediction holds critical value for marine energy development,coastal infrastructure planning,and ensuring safety in maritime operations.The precision of such predictions carries substantia...The significant wave height prediction holds critical value for marine energy development,coastal infrastructure planning,and ensuring safety in maritime operations.The precision of such predictions carries substantial the oretical and practical weight.This survey delivers an exhaustive evaluation and integration of the latest studies and advances in the domain of significant wave height prediction,serving as a methodical guidepost for academicians.The study introduces an all-encompassing predictive framework for significan wave height,which not only integrates diverse established forecasting techniques but also paves the way for novel research trajectories and creative breakthroughs.The framework is structured into four principal layers i...feature selection,basic prediction,data decomposition,and parameter optimization.The ensuing sections meticulously dissect the methodologies within these strata,elucidating their core concepts,distinctive features merits,and constraints,and their applicability to significant wave height prediction.To wrap up,the study delves into fresh research inguiries and avenues pertinent to the discipline,thereby broadening the comprehension of significant wave height prediction.In essence,this scholarly article imparts critical knowledge beneficial to the realm of marine technology.展开更多
基金This project was supported by the National Natural Science Foundation of China (60532060)Hainan Education Bureau Research Project (Hjkj200602)Hainan Natural Science Foundation (80551).
文摘A nonlinear data analysis algorithm, namely empirical data decomposition （EDD） is proposed, which can perform adaptive analysis of observed data. Analysis filter, which is not a linear constant coefficient filter, is automatically determined by observed data, and is able to implement multi-resolution analysis as wavelet transform. The algorithm is suitable for analyzing non-stationary data and can effectively wipe off the relevance of observed data. Then through discussing the applications of EDD in image compression, the paper presents a 2-dimension data decomposition framework and makes some modifications of contexts used by Embedded Block Coding with Optimized Truncation （EBCOT） . Simulation results show that EDD is more suitable for non-stationary image data compression.
文摘Non-negative matrix factorization （NMF） is a technique for dimensionality reduction by placing non-negativity constraints on the matrix. Based on the PARAFAC model, NMF was extended for three-dimension data decomposition. The three-dimension nonnegative matrix factorization （NMF3） algorithm, which was concise and easy to implement, was given in this paper. The NMF3 algorithm implementation was based on elements but not on vectors. It could decompose a data array directly without unfolding, which was not similar to that the traditional algorithms do, It has been applied to the simulated data array decomposition and obtained reasonable results. It showed that NMF3 could be introduced for curve resolution in chemometrics.
文摘Since chemical processes are highly non-linear and multiscale,it is vital to deeply mine the multiscale coupling relationships embedded in the massive process data for the prediction and anomaly tracing of crucial process parameters and production indicators.While the integrated method of adaptive signal decomposition combined with time series models could effectively predict process variables,it does have limitations in capturing the high-frequency detail of the operation state when applied to complex chemical processes.In light of this,a novel Multiscale Multi-radius Multi-step Convolutional Neural Network(Msrt Net)is proposed for mining spatiotemporal multiscale information.First,the industrial data from the Fluid Catalytic Cracking(FCC)process decomposition using Complete Ensemble Empirical Mode Decomposition with Adaptive Noise(CEEMDAN)extract the multi-energy scale information of the feature subset.Then,convolution kernels with varying stride and padding structures are established to decouple the long-period operation process information encapsulated within the multi-energy scale data.Finally,a reconciliation network is trained to reconstruct the multiscale prediction results and obtain the final output.Msrt Net is initially assessed for its capability to untangle the spatiotemporal multiscale relationships among variables in the Tennessee Eastman Process(TEP).Subsequently,the performance of Msrt Net is evaluated in predicting product yield for a 2.80×10^(6) t/a FCC unit,taking diesel and gasoline yield as examples.In conclusion,Msrt Net can decouple and effectively extract spatiotemporal multiscale information from chemical process data and achieve a approximately reduction of 30%in prediction error compared to other time-series models.Furthermore,its robustness and transferability underscore its promising potential for broader applications.
文摘The paper deals with the vehicle scheduling problem related to regional public transport. Linear programming methods are used to solve the problem. A mathematical model is created including the constraints and the objective function minimizing costs and the number of vehicles. A minimum costs and a number of vehicles are forced at the same time by special economical input data analysis and an allocation of costs. Determining of the costs coefficients is done by three methods, which differs primarily by how much of the total costs they take into account. The decomposition of the set of lines into disjoint subsets can be used instead of the ＂direct＂ optimization. The decomposition has proven to be a suitable alternative in solving large optimization problems. The problem was applied to optimize vehicle scheduling in the region, which is situated in the north-east of the Czech Republic. There is used Xpress-IVE software, which solve the problem by simplex algorithm and branch and bound method. Research results show that there are large reserves in the organization of public transport. The implementation of the new vehicle scheduling would bring significant costs reductions in amount of at least 10% for the optimal solution and in amount of about 10% for the decomposition solution. The number of drivers could be decreased and the total time of the vehicles being outside the garage could be also reduced by at least 10%.
文摘The paper surveys interactions between complex and functional-analytic methods in the CauchyKovalevskaya theory. For instance, the behaviour of the derivative of a bounded holomorphic function led to abstract versions of the Cauchy-Kovalevskaya Theorem.Recent trends in the Cauchy-Kovalevskaya theory are based on the concept of associated differential operators. Since an evolution operator may posses several associated operators, initial data may be decomposed into components belonging to different associated spaces.This technique makes it also possible to solve ill-posed initial value problems.
文摘For describing target motion in hypersonic vehicle defense,a parametric analyzing and modeling method on ballistic data is proposed based on time varying auto-regressive method.Ballistic data are regarded as non-stationary random signal,where the hidden internal law is studied.Firstly,ballistic data are decomposed into smooth linear trend signal and non-stationary periodic skip signal with ensemble empirical mode decomposition method to avoid mutual interference between different modal data.Secondly,the linear trend signal and the periodic skip signal are modeled separately.The linear trend signal is approximated by power function regressive estimator and the periodic skip signal is modeled based on time varying auto-regressive method.In order to determine optimal model orders,a novel method is presented based on information theoretic criteria and the criteria of minimizing the mean absolute error.Finally,the consistency test is conducted by investigating the time-frequency spectrum characteristics and statistical properties of outputs of the parametric model established above and dynamics model under the same initial condition.Simulation results demonstrate that the parametric model established by the proposed method shares a high consistency with the original dynamics model.
基金Acknowledgments The authors thank the editors and two anonymous referees for their helpful comments and suggestions that substantially improved the quality of this work. This research has been supported by grants from National Natural Science Foundation of China （71224001） and China Postdoctoral Science Foundation funded project （2015M571135）.
文摘Data envelopment analysis （DEA） is an effective non-parametric method for measuring the relative efficiencies of decision making units （DMUs） with multiple inputs and outputs. In many real situations, the internal structure of DMUs is a two-stage network process with shared inputs used in both stages and common outputs produced by the both stages. For example, hospitals have a two-stage network structure. Stage 1 consumes resources such as information technology system, plant, equipment and admin personnel to generate outputs such as medical records, laundry and housekeeping. Stage 2 consumes the same set of resources used by stage 1 （named shared inputs） and the outputs generated by stage 1 （named intermediate measures） to provide patient services. Besides, some of outputs, for instance, patient satisfaction degrees, are generated by the two individual stages together （named shared outputs）. Since some of shared inputs and outputs are hard split up and allocated to each individual stage, it needs to develop two-stage DEA methods for evaluating the performance of two-stage network processes in such problems. This paper extends the centralized model to measure the DEA efficiency of the two-stage process with non split-table shared inputs and outputs. A weighted additive approach is used to combine the two individual stages. Moreover, additive efficiency decomposition models are developed to simultaneously evaluate the maximal and the minimal achievable efficiencies for the individual stages. Finally, an example of 17 city branches of China Construction Bank in Anhui Province is employed to illustrate the proposed approach.
基金supported by the Open Project of Xiangjiang Laboratory(No.22XJ02003)National Science Fund for Outstanding Young Scholars(No.62122093)+1 种基金Science&Technology Project for Young and Middle-aged Talents of Hunan(No.2023TJ-Z03)University Fundamental Research Fund(Nos.23-ZZCX-JDZ-28 and ZK21-07).
文摘The significant wave height prediction holds critical value for marine energy development,coastal infrastructure planning,and ensuring safety in maritime operations.The precision of such predictions carries substantial the oretical and practical weight.This survey delivers an exhaustive evaluation and integration of the latest studies and advances in the domain of significant wave height prediction,serving as a methodical guidepost for academicians.The study introduces an all-encompassing predictive framework for significan wave height,which not only integrates diverse established forecasting techniques but also paves the way for novel research trajectories and creative breakthroughs.The framework is structured into four principal layers i...feature selection,basic prediction,data decomposition,and parameter optimization.The ensuing sections meticulously dissect the methodologies within these strata,elucidating their core concepts,distinctive features merits,and constraints,and their applicability to significant wave height prediction.To wrap up,the study delves into fresh research inguiries and avenues pertinent to the discipline,thereby broadening the comprehension of significant wave height prediction.In essence,this scholarly article imparts critical knowledge beneficial to the realm of marine technology.
