Most cited articles
2011(1):1-22.
Abstract:
System identification is the theory and methods of establishing mathematical models of systems.The mathematical modeling has a long research history,but the system identification discipline has only several tens of years.In this short decades,system identification has achieved great developments,new identification methods are born one after another,and the research results cover the theory and applications of natural science and social sciences,including physics,biology,earth science,meteorology,computer science,economics,psychology,political science and so on.In this context,we come back to ponder some basic problems of system identification,which is not without benefits for the development of system identification.This is a paper of an introduction to system identification which briefly introduces the definition of identification,system models and identification models,the basic steps and purposes of identification,including the experimental design of identification and data preprocessing,and the types of identification methods,including the least squares identification methods,gradient identification methods,auxiliary model based identification methods,and multiinnovation identification methods,and hierarchical identification methods,etc
System identification is the theory and methods of establishing mathematical models of systems.The mathematical modeling has a long research history,but the system identification discipline has only several tens of years.In this short decades,system identification has achieved great developments,new identification methods are born one after another,and the research results cover the theory and applications of natural science and social sciences,including physics,biology,earth science,meteorology,computer science,economics,psychology,political science and so on.In this context,we come back to ponder some basic problems of system identification,which is not without benefits for the development of system identification.This is a paper of an introduction to system identification which briefly introduces the definition of identification,system models and identification models,the basic steps and purposes of identification,including the experimental design of identification and data preprocessing,and the types of identification methods,including the least squares identification methods,gradient identification methods,auxiliary model based identification methods,and multiinnovation identification methods,and hierarchical identification methods,etc
2011(2):1-97.
Abstract:
Control is the core of all scientific issues.Mathematical models are the basis of all control problems.The movement law of things described by equations is the mathematical model.The development of different disciplines is to establish the process of their mathematical models.This paper divides the mathematical models of the linear dynamic systems into three categories:the timeseries models,the equation error type models and the output error type models,and introduces basic mathematical models of linear systems in detail,including the discretization of continuoustime system models and model equivalence transform,singleinput singleoutput stochastic system,multivariable systems,multivariablelike systems,multipleinput and multipleoutput systems such as the main model,submodel and subsubmodel of the transfer function matrix,multipleinput singleoutput systems,and singleinput multipleoutput systems.
Control is the core of all scientific issues.Mathematical models are the basis of all control problems.The movement law of things described by equations is the mathematical model.The development of different disciplines is to establish the process of their mathematical models.This paper divides the mathematical models of the linear dynamic systems into three categories:the timeseries models,the equation error type models and the output error type models,and introduces basic mathematical models of linear systems in detail,including the discretization of continuoustime system models and model equivalence transform,singleinput singleoutput stochastic system,multivariable systems,multivariablelike systems,multipleinput and multipleoutput systems such as the main model,submodel and subsubmodel of the transfer function matrix,multipleinput singleoutput systems,and singleinput multipleoutput systems.
2014, 6(2):97-112.
Abstract:
For multivariate pseudo-linear regressive moving average systems,a multivariate extended stochastic gradient(ESG) algorithm is discussed.In order to reduce the computational cost of the identification algorithm,we decompose a multivariate system into several subsystems,and derive a partially coupled(subsystem) ESG algorithm and a partially coupled(subsystem) multi-innovation ESG algorithm according to the coupling identification concept and the multi-innovation identification theory.Furthermore,we extend these methods to multivariate pseudo-linear autoregressive moving average systems and present a partially coupled(subsystem) generalized extended stochastic gradient(GESG) algorithm and a partially coupled(subsystem) multi-innovation GESG algorithm.The computational efficiencies of the multivariate ESG algorithm,the partially coupled ESG algorithm and the partially coupled multi-innovation ESG algorithm are analyzed.
For multivariate pseudo-linear regressive moving average systems,a multivariate extended stochastic gradient(ESG) algorithm is discussed.In order to reduce the computational cost of the identification algorithm,we decompose a multivariate system into several subsystems,and derive a partially coupled(subsystem) ESG algorithm and a partially coupled(subsystem) multi-innovation ESG algorithm according to the coupling identification concept and the multi-innovation identification theory.Furthermore,we extend these methods to multivariate pseudo-linear autoregressive moving average systems and present a partially coupled(subsystem) generalized extended stochastic gradient(GESG) algorithm and a partially coupled(subsystem) multi-innovation GESG algorithm.The computational efficiencies of the multivariate ESG algorithm,the partially coupled ESG algorithm and the partially coupled multi-innovation ESG algorithm are analyzed.
2014, 6(3):193-210.
Abstract:
The auxiliary model identification idea,the multi-innovation identification theory and the coupling identification concept are the new ideas and principles for studying identification problems of complex systems.Combining them,this paper studies the identification methods of multivariable output-error-like systems and presents multivariate auxiliary model identification methods,multivariate auxiliary model based multi-innovation identification methods,interval-varying multivariate auxiliary model based multi-innovation identification methods.In order to reduce the computational complexity of the algorithms,we decompose the system into several sub-identification models and derive the partially coupled auxiliary model based identification methods and the partially coupled auxiliary model based multi-innovation identification methods,using the auxiliary model identification idea,the multi-innovation identification theory and the coupling identification concept.Finally,the computational efficiency,the computational steps and the flowcharts of some typical identification algorithms are discussed.
The auxiliary model identification idea,the multi-innovation identification theory and the coupling identification concept are the new ideas and principles for studying identification problems of complex systems.Combining them,this paper studies the identification methods of multivariable output-error-like systems and presents multivariate auxiliary model identification methods,multivariate auxiliary model based multi-innovation identification methods,interval-varying multivariate auxiliary model based multi-innovation identification methods.In order to reduce the computational complexity of the algorithms,we decompose the system into several sub-identification models and derive the partially coupled auxiliary model based identification methods and the partially coupled auxiliary model based multi-innovation identification methods,using the auxiliary model identification idea,the multi-innovation identification theory and the coupling identification concept.Finally,the computational efficiency,the computational steps and the flowcharts of some typical identification algorithms are discussed.
2009(1):1-15.
Abstract:
根据个人学习研究稳定性的心得体会，首先介绍了前苏联伟大的数学力学家Ｌｙａｐｕｎｏｖ院士的博士论文《运动稳定性的一般问题》在全世界产生的超过１个世纪的巨大影响．叙述了由该博士论文首创的几个巨大成就何以能奠定１门学科的基础，从而开创了１个新的重要的研究方向，以及留给后人很多很多研究的课题的理由．特别地，用事实和科学断语回答了“Ｌｙａｐｕｎｏｖ稳定性已领风骚１００多年，余晖还几何”的问题．明确表明１个观点：稳定性将是１个“永恒的主题”，不老的科学，定将永恒地给人启迪，洞察力，智慧和思想．
根据个人学习研究稳定性的心得体会，首先介绍了前苏联伟大的数学力学家Ｌｙａｐｕｎｏｖ院士的博士论文《运动稳定性的一般问题》在全世界产生的超过１个世纪的巨大影响．叙述了由该博士论文首创的几个巨大成就何以能奠定１门学科的基础，从而开创了１个新的重要的研究方向，以及留给后人很多很多研究的课题的理由．特别地，用事实和科学断语回答了“Ｌｙａｐｕｎｏｖ稳定性已领风骚１００多年，余晖还几何”的问题．明确表明１个观点：稳定性将是１个“永恒的主题”，不老的科学，定将永恒地给人启迪，洞察力，智慧和思想．
2010(6):557-561.
Abstract:
Lightning density and lightning current amplitude are two important parameters in lightning protection.In this article,time-weighted average method is introduced to combine data recorded by conventional method and lightning location system (LLS),Matlab is employed to realize the fitting of lightning current amplitude according to Anderson empirical formula.Results show that in Jiangsu province lightning activities occurred more in the southern regions than northern regions,more in the western regions than eastern regions.Lightning protection should be an important task particularly in regions surrounding Hongze Lake,Taihu Lake and Ningzhen hills; probability distribution of lightning current amplitude fits with Anderson experimental formula,and amplitude mainly distributed in the range from 20 kA to 40 kA.
Lightning density and lightning current amplitude are two important parameters in lightning protection.In this article,time-weighted average method is introduced to combine data recorded by conventional method and lightning location system (LLS),Matlab is employed to realize the fitting of lightning current amplitude according to Anderson empirical formula.Results show that in Jiangsu province lightning activities occurred more in the southern regions than northern regions,more in the western regions than eastern regions.Lightning protection should be an important task particularly in regions surrounding Hongze Lake,Taihu Lake and Ningzhen hills; probability distribution of lightning current amplitude fits with Anderson experimental formula,and amplitude mainly distributed in the range from 20 kA to 40 kA.
2011(3):193-226.
Abstract:
System identification is the theory and methods of establishing the mathematical models of systems.This paper discusses some basic issues involved in system identification,including the identification accuracy,the identification methods,the input signal design,the parameter identifiability and system identifiability,the open-loop identifiability and closed-loop identifiability,the identifiability and the controllability and observability,the identifiability and the input signal,and the excitation signal and the excitation conditions relate to the convergence of identification methods,and the convergene theorems of some typical identification algorithms.
System identification is the theory and methods of establishing the mathematical models of systems.This paper discusses some basic issues involved in system identification,including the identification accuracy,the identification methods,the input signal design,the parameter identifiability and system identifiability,the open-loop identifiability and closed-loop identifiability,the identifiability and the controllability and observability,the identifiability and the input signal,and the excitation signal and the excitation conditions relate to the convergence of identification methods,and the convergene theorems of some typical identification algorithms.
2014, 6(6):481-504.
Abstract:
Because the state space model contains both the unknown states and the unknown parameters,and they involve the nonlinear product relations,which makes the identification problem more complicated.In order to solve this problem,this paper studies the combined state and parameter estimation methods for canonical state space systems.The interactive estimation theory is used to derive the combined state and parameter estimation algorithms by means of the recursive or iterative scheme.When computing the parameter estimates,the unknown states in the information vector of the identification algorithms are replaced with their estimates,the obtained parameter estimates are used to design the parameter estimates based observer or the parameter estimates based Kalman filtering algorithm to estimate the states of the systems.They form an interactive estimation process (a hierarchical estimation process).Along this line,from the recursive scheme or the iterative scheme,this paper presents the observer based or the Kalman filtering based stochastic gradient (SG) identification algorithm,recursive least squares (LS) identification algorithm,multi-innovation SG algorithm,multi-innovation LS identification algorithm,and the model decomposition based identification methods.Finally,the computational efficiency,the computational steps and the flowcharts of some typical algorithms are discussed.
Because the state space model contains both the unknown states and the unknown parameters,and they involve the nonlinear product relations,which makes the identification problem more complicated.In order to solve this problem,this paper studies the combined state and parameter estimation methods for canonical state space systems.The interactive estimation theory is used to derive the combined state and parameter estimation algorithms by means of the recursive or iterative scheme.When computing the parameter estimates,the unknown states in the information vector of the identification algorithms are replaced with their estimates,the obtained parameter estimates are used to design the parameter estimates based observer or the parameter estimates based Kalman filtering algorithm to estimate the states of the systems.They form an interactive estimation process (a hierarchical estimation process).Along this line,from the recursive scheme or the iterative scheme,this paper presents the observer based or the Kalman filtering based stochastic gradient (SG) identification algorithm,recursive least squares (LS) identification algorithm,multi-innovation SG algorithm,multi-innovation LS identification algorithm,and the model decomposition based identification methods.Finally,the computational efficiency,the computational steps and the flowcharts of some typical algorithms are discussed.
2015, 7(1):1-23.
Abstract:
Typical block-oriented structure nonlinear systems include the basic input nonlinear systems,the output nonlinear systems,the input-output nonlinear systems and the feedback nonlinear systems.The input nonlinear systems include the input nonlinear equation-error type systems and the input nonlinear output-error type systems.Taking the input nonlinear equation-error autoregressive systems (namely the input nonlinear controlled autoregressive autoregressive (IN-CARAR) systems as an example,this paper studies and presents stochastic gradient (SG) identification methods,multi-innovation SG methods,recursive least squares (LS) identification methods and multi-innovation LS identification methods for IN-CARAR systems based on the over-parameterization model,the key term separation principle and the data filtering technique,the model decomposition technique.These methods can be extended to other input nonlinear equation-error systems,input nonlinear output-error type systems,output nonlinear equation-error type systems and output nonlinear output-error systems,and feedback nonlinear systems.Finally,the computational efficiency,the computational steps and the flowcharts of several typical identification algorithms are discussed.
Typical block-oriented structure nonlinear systems include the basic input nonlinear systems,the output nonlinear systems,the input-output nonlinear systems and the feedback nonlinear systems.The input nonlinear systems include the input nonlinear equation-error type systems and the input nonlinear output-error type systems.Taking the input nonlinear equation-error autoregressive systems (namely the input nonlinear controlled autoregressive autoregressive (IN-CARAR) systems as an example,this paper studies and presents stochastic gradient (SG) identification methods,multi-innovation SG methods,recursive least squares (LS) identification methods and multi-innovation LS identification methods for IN-CARAR systems based on the over-parameterization model,the key term separation principle and the data filtering technique,the model decomposition technique.These methods can be extended to other input nonlinear equation-error systems,input nonlinear output-error type systems,output nonlinear equation-error type systems and output nonlinear output-error systems,and feedback nonlinear systems.Finally,the computational efficiency,the computational steps and the flowcharts of several typical identification algorithms are discussed.
2010(4):353-356.
Abstract:
Advanced impact testing device was employed to study the performance of the MOV under the lightning impulsion.Residual voltage change of the MOV active energy coordination,along with the coordination forms,were comprehensively analyzed in impact testing.Testing results showed that avalanche breakdown was the main damage form when the MOV absorbs more energy in the transient process.The test of MOV s AEC in serial and parallel connection indicated that compared with a single MOV,several MOV s parallel connec...
Advanced impact testing device was employed to study the performance of the MOV under the lightning impulsion.Residual voltage change of the MOV active energy coordination,along with the coordination forms,were comprehensively analyzed in impact testing.Testing results showed that avalanche breakdown was the main damage form when the MOV absorbs more energy in the transient process.The test of MOV s AEC in serial and parallel connection indicated that compared with a single MOV,several MOV s parallel connec...
2011(2):119-128.
Abstract:
Observation of concentration of pollutants in atmosphere by OPSIS AB DOAS system and Thermo SCIENTIFIC EMS system was performed in rural areas of Nanjing during the winter of 2009 and the spring of 2010.Based on technical characteristics of the two systems,quality control of the observation data and comparison of the two systems were carried.The characteristics of seasonal variations of mass concentration of NO2,O3 and SO2 were analyzed.The comparison of pollution characteristics between urban and rural areas measured by the two different systems during the autumn of 2009 was also carried.Quite good correlation can be established between the observation data of the two systems.The observation data of DOAS system are influenced by water vapor and aerosols in atmosphere,so they are generally higher than that of EMS system,with a rate from 14％ to 25％.The concentration of SO2 and NO2 are higher in winter,while that of O3 is lower,however,quite the reverse is true in spring.There is negative correlation between the curve of NO2 and O3.The concentration of NO2 in urban areas is higher than that in rural areas,and the diurnal variation shows two peaks,while that shows one peak in rural areas.The diurnal variation of O3 shows one peak in both urban and rural areas,but the former has a bigger variation range.The diurnal variation of SO2 shows one peak in urban areas and two peaks in rural areas.
Observation of concentration of pollutants in atmosphere by OPSIS AB DOAS system and Thermo SCIENTIFIC EMS system was performed in rural areas of Nanjing during the winter of 2009 and the spring of 2010.Based on technical characteristics of the two systems,quality control of the observation data and comparison of the two systems were carried.The characteristics of seasonal variations of mass concentration of NO2,O3 and SO2 were analyzed.The comparison of pollution characteristics between urban and rural areas measured by the two different systems during the autumn of 2009 was also carried.Quite good correlation can be established between the observation data of the two systems.The observation data of DOAS system are influenced by water vapor and aerosols in atmosphere,so they are generally higher than that of EMS system,with a rate from 14％ to 25％.The concentration of SO2 and NO2 are higher in winter,while that of O3 is lower,however,quite the reverse is true in spring.There is negative correlation between the curve of NO2 and O3.The concentration of NO2 in urban areas is higher than that in rural areas,and the diurnal variation shows two peaks,while that shows one peak in rural areas.The diurnal variation of O3 shows one peak in both urban and rural areas,but the former has a bigger variation range.The diurnal variation of SO2 shows one peak in urban areas and two peaks in rural areas.
12 Coupled multi-innovation stochastic gradient type identification methods for multivariate systems
2014, 6(1):1-16.
Abstract:
For multivariate linear regression systems,using the coupling identification concept and the multi-innovation identification theory,this paper discusses a multivariate stochastic gradient algorithm,a multivariate multi-innovation stochastic gradient algorithm,and an interval-varying multivariate multi-innovation gradient algorithm,decomposes a multivariate system into several subsystems,and presents a coupled subsystem stochastic gradient algorithm,a coupled stochastic gradient algorithm,a coupled subsystems multi-innovation stochastic gradient algorithm and a coupled multi-innovation stochastic gradient algorithm.These methods are extended to multivariate pseudo-linear moving average systems and multivariate pseudo-linear autoregressive moving average systems.Finally,this paper gives the steps and diagrams for computing the parameter estimates using several typical coupled stochastic gradient algorithms and coupled multi-innovation stochastic gradient algorithms.
For multivariate linear regression systems,using the coupling identification concept and the multi-innovation identification theory,this paper discusses a multivariate stochastic gradient algorithm,a multivariate multi-innovation stochastic gradient algorithm,and an interval-varying multivariate multi-innovation gradient algorithm,decomposes a multivariate system into several subsystems,and presents a coupled subsystem stochastic gradient algorithm,a coupled stochastic gradient algorithm,a coupled subsystems multi-innovation stochastic gradient algorithm and a coupled multi-innovation stochastic gradient algorithm.These methods are extended to multivariate pseudo-linear moving average systems and multivariate pseudo-linear autoregressive moving average systems.Finally,this paper gives the steps and diagrams for computing the parameter estimates using several typical coupled stochastic gradient algorithms and coupled multi-innovation stochastic gradient algorithms.
2012, 4(2):97-124.
Abstract:
Hierarchical identification is an important branch of system identification.The hierarchical identification principle is developed on the bisis of the "decomposition-coordination principle" in the hierarchical control for a large-scale system.It is able to not only solve problems that the identification algorithms require heavy computational burden for a lareg-scale systems with many parameters and high dimensions problem,but also solve identification problems for bilinear-parameter systems,multi-linear-parameter systems and nonlinear systems with complex structures.In this paper,firstly we describe the hierarchical identification principle,the Jacobi iteration and Gauss-Seidel iteration for linear systems with a set of equations Ax=b,and give the family of iterative methods for linear equations;secondly,we study hierarchical least squares based and hierarchical gradient based iterative algorithms for general matrix equations and coupled matrix equations in the light of the Jacobi iteration and the hierarchical identification principle;thirdly,we present a two-stage recursive least squares algorithm(i.e.,a simple hierarchical least squares algorithm) for equation error models and a hierarchical least squares identification algorithm for linear regression models.Finally,the hierarchical identification methods are introdiced for multivariable CARMA-like systems using the hierarchical identification principle.
Hierarchical identification is an important branch of system identification.The hierarchical identification principle is developed on the bisis of the "decomposition-coordination principle" in the hierarchical control for a large-scale system.It is able to not only solve problems that the identification algorithms require heavy computational burden for a lareg-scale systems with many parameters and high dimensions problem,but also solve identification problems for bilinear-parameter systems,multi-linear-parameter systems and nonlinear systems with complex structures.In this paper,firstly we describe the hierarchical identification principle,the Jacobi iteration and Gauss-Seidel iteration for linear systems with a set of equations Ax=b,and give the family of iterative methods for linear equations;secondly,we study hierarchical least squares based and hierarchical gradient based iterative algorithms for general matrix equations and coupled matrix equations in the light of the Jacobi iteration and the hierarchical identification principle;thirdly,we present a two-stage recursive least squares algorithm(i.e.,a simple hierarchical least squares algorithm) for equation error models and a hierarchical least squares identification algorithm for linear regression models.Finally,the hierarchical identification methods are introdiced for multivariable CARMA-like systems using the hierarchical identification principle.
2014, 6(5):385-404.
Abstract:
According to the hierarchical identification principle,this paper presents the hierarchical stochastic gradient algorithms and the hierarchical gradient based iterative algorithms,the hierarchical least squares algorithms and the hierarchical least squares based iterative algorithms for multivariable equation-error-like systems and multivariable equation-error ARMA-like systems,and further derives the hierarchical multi-innovation gradient algorithms and the hierarchical multi-innovation least squares algorithms.In order to reduce computational burdens,this paper derives the filtering based hierarchical identification algorithms and the filtering based hierarchical multi-innovation identification algorithms for multivariable equation-error ARMA-like systems using the filtering technique. Finally,the computational efficiency and the computational steps of some typical identification algorithms are discussed.
According to the hierarchical identification principle,this paper presents the hierarchical stochastic gradient algorithms and the hierarchical gradient based iterative algorithms,the hierarchical least squares algorithms and the hierarchical least squares based iterative algorithms for multivariable equation-error-like systems and multivariable equation-error ARMA-like systems,and further derives the hierarchical multi-innovation gradient algorithms and the hierarchical multi-innovation least squares algorithms.In order to reduce computational burdens,this paper derives the filtering based hierarchical identification algorithms and the filtering based hierarchical multi-innovation identification algorithms for multivariable equation-error ARMA-like systems using the filtering technique. Finally,the computational efficiency and the computational steps of some typical identification algorithms are discussed.
2015, 7(4):289-312.
Abstract:
Systems have two categories, one is linear and the other is nonlinear.The linear systems have uniform descriptions and the nonlinear systems have countless categories and have no uniform descriptions.The linear-parameter systems are a special class of nonlinear systems and are linear on the parameter space.For pseudo-linear-parameter systems, this paper studies and presents the auxiliary model based multi-innovation(MI) identification methods, the data filtering based auxiliary model MI identification methods, the model decomposition based auxiliary model MI identification methods, and the filtering based decomposition MI identification methods.Finally, the computational efficiency, the computational steps and the flowcharts of several typical identification algorithms are discussed.
Systems have two categories, one is linear and the other is nonlinear.The linear systems have uniform descriptions and the nonlinear systems have countless categories and have no uniform descriptions.The linear-parameter systems are a special class of nonlinear systems and are linear on the parameter space.For pseudo-linear-parameter systems, this paper studies and presents the auxiliary model based multi-innovation(MI) identification methods, the data filtering based auxiliary model MI identification methods, the model decomposition based auxiliary model MI identification methods, and the filtering based decomposition MI identification methods.Finally, the computational efficiency, the computational steps and the flowcharts of several typical identification algorithms are discussed.
2020, 12(2):150-158.
DOI: 10.13878/j.cnki.jnuist.2020.02.002
Abstract:
Synthetic Aperture Radar (SAR) technology has unique advantages in forest resource monitoring due to its all-day,all-weather imaging capability and sensitivity to vertical forest structure information.Therefore,SAR has become a research focus of current forest resources remote sensing survey technology.Firstly,the development background,development trajectory and related knowledge of SAR forest resources monitoring technology are introduced.Then,the technological developments of polarimetric SAR,interferometric SAR,polarimetric SAR interferometry and tomographic SAR in research of forest land cover type classification,change detection and forest parameter quantification estimation are emphasized.Finally,the existing problems and development trends for forest resource monitoring research and application of SAR are summarized and analyzed.
Synthetic Aperture Radar (SAR) technology has unique advantages in forest resource monitoring due to its all-day,all-weather imaging capability and sensitivity to vertical forest structure information.Therefore,SAR has become a research focus of current forest resources remote sensing survey technology.Firstly,the development background,development trajectory and related knowledge of SAR forest resources monitoring technology are introduced.Then,the technological developments of polarimetric SAR,interferometric SAR,polarimetric SAR interferometry and tomographic SAR in research of forest land cover type classification,change detection and forest parameter quantification estimation are emphasized.Finally,the existing problems and development trends for forest resource monitoring research and application of SAR are summarized and analyzed.
Address：No. 219, Ningliu Road, Nanjing, Jiangsu Province
Postcode：210044
Phone：025-58731025
