Behaviour Analysis of Modeling and Model Evaluating Methods in System Identification for a Multiprocess Station

被引:0
作者
Beula, A. Annie Steffy [1 ]
Peter, Geno [2 ]
Alexander Stonier, Albert [3 ]
Vignesh, K. Ezhil [1 ]
Ganji, Vivekananda [4 ]
机构
[1] Stella Marys Coll Engn, Elect & Elect Engn, Ganapathipuram, India
[2] Univ Technol Sarawak, Sch Engn & Technol, CRISD, Sibu, Sarawak, Malaysia
[3] Vellore Inst Technol, Sch Elect Engn, Vellore, India
[4] Debre Tabor Univ, Dept Elect & Comp Engn, Debre Tabor, Ethiopia
关键词
INFORMATION CRITERIA; MAXIMUM-LIKELIHOOD; NONLINEAR-SYSTEMS; ALGORITHM; SELECTION;
D O I
10.1155/2024/7741473
中图分类号
O1 [数学];
学科分类号
0701 ; 070101 ;
摘要
Systems are designed to perform specific task by giving certain input which produces the required output in an orderly manner known as process. The input, output, and the state variables should be known that will help in interacting with the system. The relation between these variables can be brought out by building a model that resembles or expresses the original performance of the system. The parameters of the model are estimated using the least squares approximation, maximum likelihood, maximum log-likelihood, and Bayesian parameter estimation methods by utilizing the experimental data from the multiprocess station. The selected parameters are converted to nine different transfer function models that represent the given dynamic system. The models framed are analyzed by the criterion curve technique using seven criterion functions evaluating the fitness of the model. Order of the model is found from Hankel matrix representation methods such as singular value decomposition and determinant method. Response of the models is compared with the original response to choose the best fit model by calculating ISE standard. All the above methods are used to model the system without physical and theoretical laws which is known as system identification.
引用
收藏
页数:13
相关论文
共 26 条
  • [1] Alves V. A. O., 2012, 2012 IEEE INT C CONT, DOI [10.1109/cca.2012.6402422, DOI 10.1109/CCA.2012.6402422]
  • [2] Modified information criteria and selection of long memory time series models
    Baillie, Richard T.
    Kapetanios, George
    Papailias, Fotis
    [J]. COMPUTATIONAL STATISTICS & DATA ANALYSIS, 2014, 76 : 116 - 131
  • [3] Maximum Likelihood Estimation and Non-Linear Least Squares Fitting Implementation in FPGA Devices for High Resolution Hodoscopy
    Blasco, Jose M.
    Sanchis, Enrique
    Gonzalez, Vicente
    Martin, Jose D.
    Egea, Francisco J.
    Barrientos, Diego
    Granero, Domingo
    Sanchis-Sanchez, Enrique
    [J]. IEEE TRANSACTIONS ON NUCLEAR SCIENCE, 2013, 60 (05) : 3578 - 3584
  • [4] A two-level hybrid evolutionary algorithm for modeling one-dimensional dynamic systems by higher-order ODE models
    Cao, HQ
    Kang, LS
    Guo, T
    Chen, YP
    de Garis, H
    [J]. IEEE TRANSACTIONS ON SYSTEMS MAN AND CYBERNETICS PART B-CYBERNETICS, 2000, 30 (02): : 351 - 357
  • [5] Identification of multivariable nonlinear systems in the presence of colored noises using iterative hierarchical least squares algorithm
    Jafari, Masoumeh
    Salimifard, Maryam
    Dehghani, Maryam
    [J]. ISA TRANSACTIONS, 2014, 53 (04) : 1243 - 1252
  • [6] Javed Anum, 2019, 2019 International Conference on Applied and Engineering Mathematics (ICAEM), P133, DOI 10.1109/ICAEM.2019.8853743
  • [7] Order selection criteria for vector autoregressive models
    Karimi, Mahmood
    [J]. SIGNAL PROCESSING, 2011, 91 (04) : 955 - 969
  • [8] Ljung L., 1991, MATLAB USERS GUIDE
  • [9] Naung Y, 2018, PROCEEDINGS OF THE 2018 IEEE CONFERENCE OF RUSSIAN YOUNG RESEARCHERS IN ELECTRICAL AND ELECTRONIC ENGINEERING (EICONRUS), P1801, DOI 10.1109/EIConRus.2018.8317455
  • [10] Parameters-Transfer Identification for Dynamic Systems and Recursive Form
    Ping, Xiaojing
    Luan, Xiaoli
    Zhao, Shunyi
    Ding, Feng
    Liu, Fei
    [J]. IEEE SIGNAL PROCESSING LETTERS, 2023, 30 : 1302 - 1306