QuickSearch:   Number of matching entries: 0.

Search Settings

    AuthorTitleYearJournal/ProceedingsDOI/URL
    Matthias Himmelsbach, Andreas Kroll On optimal test signal design and parameter identification schemes for dynamic Takagi-Sugeno fuzzy models using the Fisher information matrix 2022 International Journal of Fuzzy Systems, vol. 24, no. 2, pp. 1012-1024  DOI  
    Abstract: This paper is concerned with the analysis of optimization procedures for optimal experiment design for locally affine Takagi-Sugeno (TS) fuzzy models based on the Fisher Information Matrix (FIM). The FIM is used to estimate the covariance matrix of a parameter estimate. It depends on the model parameters as well as the regression variables. Due to the dependency on the model parameters good initial models are required. Since the FIM is a matrix, a scalar measure of the FIM is optimized. Different measures and optimization goals are investigated in three case studies.
    BibTeX:
    @article{2022-Himmelsbach-IJFS,
     abstract = {This paper is concerned with the analysis of optimization procedures for optimal experiment design for locally affine Takagi-Sugeno (TS) fuzzy models based on the Fisher Information Matrix (FIM). The FIM is used to estimate the covariance matrix of a parameter estimate. It depends on the model parameters as well as the regression variables. Due to the dependency on the model parameters good initial models are required. Since the FIM is a matrix, a scalar measure of the FIM is optimized. Different measures and optimization goals are investigated in three case studies.},
     author = {Matthias Himmelsbach and Andreas Kroll},
     doi = {10.1007/s40815-021-01185-9},
     journal = {International Journal of Fuzzy Systems},
     keywords = {Optimal experiment design; Fisher information matrix;
    Takagi-Sugeno models; Nonlinear system
    identification},
     language = {english},
     mrtnote = {peer,OED},
     number = {2},
     owner = {gringard},
     pages = {1012--1024},
     timestamp = {2020.12.22},
     title = {On optimal test signal design and parameter identification
    schemes for dynamic {T}akagi-{S}ugeno fuzzy models using
    the {F}isher information
    matrix},
     volume = {24},
     year = {2022}
    }
    
    
    Matthias Himmelsbach, Andreas Kroll Zum strukturvorwissensbasierten Testsignalentwurf mittels Vorsteuerung für die Identifikation von Takagi-Sugeno-Modellen 2022 at - Automatisierungstechnik, vol. 70, no. 2, pp. 119-133  DOI  
    Abstract: In diesem Beitrag wird eine Methode zum Entwurf von Testsignalen für die Identifikation lokal-affiner dynamischer Takagi-Sugeno-SISO-Modelle vorgestellt. Diese Methode verwendet ein zuvor identifizierten Prozessmodell und nutzt so erlangtes Wissen über die Struktur mittels modellprädiktiver Regelung aus. Im Rahmen einer akademischen Fallstudie wird die Methode demonstriert.
    BibTeX:
    @article{2022-MH-at-MPC,
     abstract = {In diesem Beitrag wird eine Methode zum Entwurf von Testsignalen für die Identifikation lokal-affiner dynamischer Takagi-Sugeno-SISO-Modelle vorgestellt. Diese Methode verwendet ein zuvor identifizierten Prozessmodell und nutzt so erlangtes Wissen über die Struktur mittels modellprädiktiver Regelung aus. Im Rahmen einer akademischen Fallstudie wird die Methode demonstriert.},
     author = {Matthias Himmelsbach and Andreas Kroll},
     doi = {10.1515/auto-2021-0043},
     journal = {at -- Automatisierungstechnik},
     journaltitle = {Computational Intelligence},
     keywords = {Takagi-Sugeno-Modelle; optimaler Experimententwurf; Testsignalentwurf; modellprädiktive
    Regelung},
     language = {german},
     mrtnote = {peer,OED},
     number = {2},
     owner = {himmelsbach},
     pages = {119--133},
     pubstate = {submitted},
     timestamp = {2020.11.06},
     title = {Zum strukturvorwissensbasierten Testsignalentwurf mittels Vorsteuerung für die Identifikation von Takagi-Sugeno-Modellen},
     volume = {70},
     year = {2022}
    }
    
    
    Matthias Himmelsbach, Andreas Kroll Toolbox zum Testsignalentwurf für Standardtestsignale für die Identifikation von Eingrößensystemen: Prozessmodellfreie und -basierte Methoden 2020 30. Workshop Computational Intelligence, KIT Scientific Publishing, Berlin, GMA-FA 5.14, 26.-27. November  URL  
    BibTeX:
    @inproceedings{HimmelsbachGMACI2020,
     address = {Berlin},
     author = {Matthias Himmelsbach and Andreas Kroll},
     booktitle = {30. Workshop Computational Intelligence},
     month = {26.-27. November},
     mrtnote = {nopeer,FuzzyT2,pke,OED},
     organization = {GMA-FA 5.14},
     owner = {gringard},
     publisher = {KIT Scientific Publishing},
     timestamp = {2020.08.18},
     title = {Toolbox zum Testsignalentwurf für Standardtestsignale für die Identifikation von Eingrößensystemen: Prozessmodellfreie und -basierte Methoden},
     url = {https://www.rst.e-technik.tu-dortmund.de/cms/de/Veranstaltungen/GMA-Fachausschuss/index.html},
     year = {2020}
    }
    
    
    Matthias Gringard, Andreas Kroll On considering the output in space-filling test signal designs for the identification of dynamic Takagi-Sugeno models 2020 21st IFAC World Congress, vol. 53, no. 2, pp. 1200-1205, Elsevier, Berlin, Germany, IFAC, 12-17. July  URL  
    Abstract: In this contribution the optimal experiment design for the identification of linear affine Takagi-Sugeno (TS) models is discussed. The parameters of these models can be separated into local model and partition parameters which also allows for a separation of the design. The presented optimal design is based on the Fisher Information Matrix (FIM) and its objective is to minimize the parameter estimation uncertainty. In this contribution parameters of a standard test signal type (multi-step) are optimized to achieve this objective. The effectivity of FIM-based designs for nonlinear models depends on initial identification including structural decisions. The general functionality of the presented method is demonstrated on a case study.
    BibTeX:
    @inproceedings{GringardIFAC2020,
     abstract = {In this contribution the optimal experiment design for the identification of linear affine Takagi-Sugeno (TS) models is discussed. The parameters of these models can be separated into local model and partition parameters which also allows for a separation of the design. The presented optimal design is based on the Fisher Information Matrix (FIM) and its objective is to minimize the parameter estimation uncertainty. In this contribution parameters of a standard test signal type (multi-step) are optimized to achieve this objective. The effectivity of FIM-based designs for nonlinear models depends on initial identification including structural decisions. The general functionality of the presented method is demonstrated on a case study.},
     address = {Berlin, Germany},
     author = {Matthias Gringard and Andreas Kroll},
     booktitle = {21st IFAC World Congress},
     language = {english},
     month = {12-17. July},
     mrtnote = {peer,OED},
     number = {2},
     organization = {IFAC},
     owner = {gringard},
     pages = {1200-1205},
     publisher = {Elsevier},
     timestamp = {2017.12.13},
     title = {On considering the output in space-filling test signal designs for the identification of dynamic {T}akagi-{S}ugeno
    models},
     url = {https://www.ifac2020.org/},
     volume = {53},
     year = {2020}
    }
    
    
    Matthias Gringard, Andreas Kroll Zur Homogenisierung von Testsignalen für die nichtlineare Systemidentifikation 2019 at - Automatisierungstechnik, vol. 67, no. 10, pp. 820-832   
    BibTeX:
    @article{GringardAT2019,
     author = {Matthias Gringard and Andreas Kroll},
     journal = {at -- Automatisierungstechnik},
     mrtnote = {peer,OED},
     number = {10},
     owner = {gringard},
     pages = {820--832},
     timestamp = {2017.12.13},
     title = {Zur Homogenisierung von Testsignalen für die
    nichtlineare Systemidentifikation},
     volume = {67},
     year = {2019}
    }
    
    
    Matthias Gringard, Andreas Kroll Optimal Experiment Design for Identifying Dynamical Takagi-Sugeno Models with Minimal Parameter Uncertainty 2018 18th IFAC Symposium on System Identification, Stockholm, Sweden, IFAC, 09-11. July  URL  
    Abstract: In this contribution the optimal experiment design for the identification of linear affine Takagi-Sugeno (TS) models is discussed. The parameters of these models can be separated into local model and partition parameters which also allows for a separation of the design. The presented optimal design is based on the Fisher Information Matrix (FIM) and its objective is to minimize the parameter estimation uncertainty. In this contribution parameters of a standard test signal type (multi-step) are optimized to achieve this objective. The effectivity of FIM-based designs for nonlinear models depends on initial identification including structural decisions. The general functionality of the presented method is demonstrated on a case study.
    BibTeX:
    @inproceedings{GringardSYSID2018,
     abstract = {In this contribution the optimal experiment design for the identification of linear affine Takagi-Sugeno (TS) models is discussed. The parameters of these models can be separated into local model and partition parameters which also allows for a separation of the design. The presented optimal design is based on the Fisher Information Matrix (FIM) and its objective is to minimize the parameter estimation uncertainty. In this contribution parameters of a standard test signal type (multi-step) are optimized to achieve this objective. The effectivity of FIM-based designs for nonlinear models depends on initial identification including structural decisions. The general functionality of the presented method is demonstrated on a case study.},
     address = {Stockholm, Sweden},
     author = {Matthias Gringard and Andreas Kroll},
     booktitle = {18th IFAC Symposium on System Identification},
     month = {09-11. July},
     mrtnote = {peer,OED},
     organization = {IFAC},
     owner = {gringard},
     timestamp = {2017.12.13},
     title = {Optimal Experiment Design for Identifying Dynamical Takagi-Sugeno Models with Minimal Parameter Uncertainty},
     url = {https://www.ifac-control.org/events/system-identification-18th-sysid-2018},
     year = {2018}
    }
    
    
    Matthias Gringard, Andreas Kroll Zum optimalen Offline-Testsignalentwurf für die Identifikation dynamischer TS-Modelle: Steuerfunktionen zur optimalen Schätzung der Partitionsparameter 2018 28. Workshop Computational Intelligence, pp. 39 - 60, KIT Scientific Publishing, Dortmund, GMA-FA 5.14, 29.-30. November  DOI , URL  
    BibTeX:
    @inproceedings{GringardGMA2018,
     address = {Dortmund},
     author = {Matthias Gringard and Andreas Kroll},
     booktitle = {28. Workshop Computational Intelligence},
     doi = {10.5445/KSP/1000085935},
     isbn = {9783731508458},
     month = {29.-30. November},
     mrtnote = {nopeer,FuzzyT2,pke,OED},
     organization = {GMA-FA 5.14},
     owner = {gringard},
     pages = {39 -- 60},
     publisher = {KIT Scientific Publishing},
     timestamp = {2017.08.31},
     title = {Zum optimalen Offline-Testsignalentwurf für die Identifikation dynamischer TS-Modelle: Steuerfunktionen zur optimalen Schätzung der Partitionsparameter},
     url = {https://www.rst.e-technik.tu-dortmund.de/cms/de/Veranstaltungen/GMA-Fachausschuss/index.html},
     year = {2018}
    }
    
    
    Andreas Kroll, Axel Dürrbaum On optimal experiment design for identifying premise and conclusion parameters of Takagi-Sugeno models: nonlinear regression case 2017 Applied Soft Computing, vol. 60, pp. 407 - 422  DOI , URL  
    Abstract: Optimal Experiment Design (OED) is a well-developed concept for regression problems that are linear-in-the-parameters. In case of experiment design to identify nonlinear Takagi-Sugeno (TS) models, non-model-based approaches or OED restricted to the local model parameters (assuming the partitioning to be given) have been proposed. In this article, a Fisher Information Matrix (FIM) based OED method is proposed that considers local model and partition parameters. Due to the nonlinear model, the FIM depends on the model parameters that are subject of the subsequent identification. To resolve this paradoxical situation, at first a model-free space filling design (such as Latin Hypercube Sampling) is carried out. The collected data permits making design decisions such as determining the number of local models and identifying the parameters of an initial TS model. This initial TS model permits a FIM-based OED, such that data is collected which is optimal for a TS model. The estimates of this first stage will in general not be ideal. To become robust against parameter mismatch, a sequential optimal design is applied. In this work the focus is on D-optimal designs. The proposed method is demonstrated for three nonlinear regression problems: an industrial axial compressor and two test functions.
    BibTeX:
    @article{2017-Kroll_Duerrbaum-ASOC-OED,
     abstract = {Optimal Experiment Design (OED) is a well-developed concept for regression problems that are linear-in-the-parameters. In case of experiment design to identify nonlinear Takagi-Sugeno (TS) models, non-model-based approaches or OED restricted to the local model parameters (assuming the partitioning to be given) have been proposed. In this article, a Fisher Information Matrix (FIM) based OED method is proposed that considers local model and partition parameters. Due to the nonlinear model, the FIM depends on the model parameters that are subject of the subsequent identification. To resolve this paradoxical situation, at first a model-free space filling design (such as Latin Hypercube Sampling) is carried out. The collected data permits making design decisions such as determining the number of local models and identifying the parameters of an initial TS model. This initial TS model permits a FIM-based OED, such that data is collected which is optimal for a TS model. The estimates of this first stage will in general not be ideal. To become robust against parameter mismatch, a sequential optimal design is applied. In this work the focus is on D-optimal designs. The proposed method is demonstrated for three nonlinear regression problems: an industrial axial compressor and two test functions.},
     author = {Andreas Kroll AND Axel Dürrbaum},
     doi = {10.1016/j.asoc.2017.07.015},
     journal = {Applied Soft Computing},
     keywords = {Takagi-Sugeno fuzzy models; optimal experiment design; design of experiments, nonlinear regression; nonlinear system
    identification},
     language = {english},
     mrtnote = {peer,OED},
     mrturla = {https://mrt-pc1.mrt.maschinenbau.uni-kassel.de/MRT/Bibliothek/Publikationen/2017-Kroll_Duerrbaum-ASoC-OED-submitted-PUB.pdf},
     owner = {duerrbaum},
     pages = {407 -- 422},
     timestamp = {2017.07.10},
     title = {On optimal experiment design for identifying premise and conclusion parameters of Takagi-Sugeno models: nonlinear regression
    case},
     url = {https://reader.elsevier.com/reader/sd/pii/S1568494617304246},
     volume = {60},
     year = {2017}
    }
    
    
    Matthias Gringard, Andreas Kroll On the parametrization of APRBS and multisine test signals for the identification of nonlinear dynamic TS-models 2016 IEEE Symposium Series of Computational Intelligence 2016, pp. 39 - 60, Athens, Greece, IEEE, 06.-09. December  URL  
    BibTeX:
    @inproceedings{GringardSSCI2016,
     address = {Athens, Greece},
     author = {Matthias Gringard and Andreas Kroll},
     booktitle = {IEEE Symposium Series of Computational Intelligence
    2016},
     month = {06.-09. December},
     mrtnote = {peer,OED},
     organization = {IEEE},
     owner = {gringard},
     pages = {39 -- 60},
     timestamp = {2016.06.08},
     title = {On the parametrization of APRBS and multisine test signals for the identification of nonlinear dynamic TS-models},
     url = {https://ssci2016.cs.surrey.ac.uk/IEEE%202016.htm},
     year = {2016}
    }
    
    
    Matthias Gringard, Andreas Kroll On the systematic parametrization of APRBS and multisine test signals for nonlinear system identification 2016 26. Workshop Computational Intelligence, pp. 119-138, KIT Scientific Publishing, Dortmund, GMA-FA 5.14, 24.-25. November  DOI , URL  
    BibTeX:
    @inproceedings{GringardGMA2016,
     address = {Dortmund},
     author = {Matthias Gringard and Andreas Kroll},
     booktitle = {26. Workshop Computational Intelligence},
     doi = {10.5445/KSP/1000060007},
     isbn = {9783731505884},
     month = {24.-25. November},
     mrtnote = {nopeer,FuzzyT2,pke,OED},
     organization = {GMA-FA 5.14},
     owner = {gringard},
     pages = {119-138},
     publisher = {KIT Scientific Publishing},
     timestamp = {2016.06.08},
     title = {On the systematic parametrization of APRBS and multisine test signals for nonlinear system identification},
     url = {https://www.rst.e-technik.tu-dortmund.de/cms/de/Veranstaltungen/GMA-Fachausschuss},
     year = {2016}
    }
    
    
    Axel Dürrbaum, Andreas Kroll On robust experiment design for identifying locally affine Takagi-Sugeno models 2016 IEEE International Conference on Systems, Man, and Cybernetics (SMC 2016), Budapest, Hungary, IEEE Systems, Man, and Cybernetics, October, 9 - 12  DOI , URL  
    Abstract: Optimal experiment design (OED) is a well-developed concept for linear regression and linear dynamical modeling problems. In case of nonlinear models, the dilemma is that in order to evaluate the Fisher Information Matrix (FIM) for experiment design, the parameters to be estimated are required to evaluate the FIM. In case of locally affine Takagi-Sugeno (TS) models and D-optimal designs, even a "robust" sequential OED may not be sufficiently robust against wrong assumptions on partition parameters. As remedy, a two stage experiment design is proposed: It uses a space-filling design to estimate good initial TS model parameters. These are used to initialize a robust sequential FIM-based OED. The method is demonstrated for a nonlinear regression problem.
    BibTeX:
    @inproceedings{2016-ad_ak-SMC2016-OED,
     abstract = {Optimal experiment design (OED) is a well-developed concept for linear regression and linear dynamical modeling problems. In case of nonlinear models, the dilemma is that in order to evaluate the Fisher Information Matrix (FIM) for experiment design, the parameters to be estimated are required to evaluate the FIM. In case of locally affine Takagi-Sugeno (TS) models and D-optimal designs, even a "robust" sequential OED may not be sufficiently robust against wrong assumptions on partition parameters. As remedy, a two stage experiment design is proposed: It uses a space-filling design to estimate good initial TS model parameters. These are used to initialize a robust sequential FIM-based OED. The method is demonstrated for a nonlinear regression problem.},
     address = {Budapest, Hungary},
     author = {Axel Dürrbaum and Andreas Kroll},
     booktitle = {IEEE International Conference on Systems, Man, and
    Cybernetics (SMC 2016)},
     doi = {10.1109/SMC.2016.7844506},
     language = {english},
     month = {October, 9 -- 12},
     mrtnote = {peer,OED,poster:kroll},
     organization = {IEEE Systems, Man, and Cybernetics},
     owner = {duerrbaum},
     timestamp = {2016.06.27},
     title = {On robust experiment design for identifying locally
    affine Takagi-Sugeno models},
     url = {https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=7844506},
     year = {2016}
    }
    
    
    Andreas Kroll Zum optimalen Testsignalentwurf für die Partitionierung und Teilmodellparametrierung dynamischer Takagi-Sugeno-Modelle: Problemstellung und Lösungsansätze 2016 26. Workshop Computational Intelligence, pp. 97-118, KIT Scientific Publishing, Dortmund, GMA-FA 5.14 "Computational Intelligence", 24.-25. November  DOI  
    BibTeX:
    @inproceedings{GMA_Kroll_2016,
     address = {Dortmund},
     author = {Andreas Kroll},
     booktitle = {26. Workshop Computational Intelligence},
     doi = {10.5445/KSP/1000060007},
     language = {german},
     month = {24.-25. November},
     mrtnote = {nopeer,OED},
     organization = {GMA-FA 5.14 "Computational Intelligence"},
     owner = {duerrbaum},
     pages = {97-118},
     publisher = {KIT Scientific Publishing},
     timestamp = {2010.07.06},
     title = {Zum optimalen Testsignalentwurf für die Partitionierung und Teilmodellparametrierung dynamischer Takagi-Sugeno-Modelle: Problemstellung und
    Lösungsansätze},
     year = {2016}
    }
    
    
    Matthias Gringard, Andreas Kroll Zur Homogenisierung von Breitbandtestsignalen für die nichtlineare Systemidentifikation am Beispiel eines nichtlinearen Stellantriebs 2015 25. Workshop Computational Intelligence, pp. 145 - 162, KIT Scientific Publishing, Dortmund, GMA-FA 5.14, 26.-27. November  DOI , URL  
    BibTeX:
    @inproceedings{GringardGMA2015,
     address = {Dortmund},
     author = {Matthias Gringard and Andreas Kroll},
     booktitle = {25. Workshop Computational Intelligence},
     doi = {0.5445/KSP/100004962},
     isbn = {9783731504320},
     month = {26.-27. November},
     mrtnote = {nopeer,FuzzyT2,pke,OED},
     organization = {GMA-FA 5.14},
     owner = {gringard},
     pages = {145 -- 162},
     publisher = {KIT Scientific Publishing},
     timestamp = {2015.11.11},
     title = {Zur Homogenisierung von Breitbandtestsignalen für die nichtlineare Systemidentifikation am Beispiel eines nichtlinearen
    Stellantriebs},
     url = {https://www.rst.e-technik.tu-dortmund.de/cms/de/Veranstaltungen/GMA-Fachausschuss/index.html},
     year = {2015}
    }
    
    
    Andreas Kroll, Axel Dürrbaum On joint optimal experiment design for identifying partition and local model parameters of Takagi-Sugeno models 2015 Proceedings of the 17th IFAC Symposium on System Identification (SysID), pp. 1427 - 1432, Beijing, China, October 19-21  DOI  
    Abstract: Optimal Experiment Design (OED) is a well-developed concept for regression problems that are linear-in-their-parameters or for linear dynamical models. In case of nonlinear Takagi-Sugeno models either non-model-based experiment design or OED restricted to the local model parameters has been examined. This article proposes a joint design of local model and partition parameters that bases on the Fisher Information Matrix (FIM). For this purpose, a symbolic description of the joint FIM is derived. Its heterogeneous structure can make it badly conditioned, complicating computation of the determinant for D-optimal design. This problem is relaxed using determinant decomposition. A theoretical analysis and a case study show that experiment design for local model and partition parameters may significantly differ from each other.
    BibTeX:
    @inproceedings{Duerrbaum-2015-SysID,
     abstract = {Optimal Experiment Design (OED) is a well-developed concept for regression problems that are linear-in-their-parameters or for linear dynamical models. In case of nonlinear Takagi-Sugeno models either non-model-based experiment design or OED restricted to the local model parameters has been examined. This article proposes a joint design of local model and partition parameters that bases on the Fisher Information Matrix (FIM). For this purpose, a symbolic description of the joint FIM is derived. Its heterogeneous structure can make it badly conditioned, complicating computation of the determinant for D-optimal design. This problem is relaxed using determinant decomposition. A theoretical analysis and a case study show that experiment design for local model and partition parameters may significantly differ from each other.},
     address = {Beijing, China},
     author = {Andreas Kroll and Axel Dürrbaum},
     booktitle = {Proceedings of the 17th IFAC Symposium on System
    Identification ({SysID})},
     doi = {doi:10.1016/j.ifacol.2015.12.333},
     keywords = {Nonlinear system identification, optimal experiment design
    Takagi-Sugeno fuzzy
    systems},
     language = {english},
     month = {October 19-21},
     mrtnote = {peer,presenter:Dürrbaum,oed},
     pages = {1427 -- 1432},
     timestamp = {2015.02.13},
     title = {On joint optimal experiment design for identifying partition and local model parameters of Takagi-Sugeno
    models},
     year = {2015}
    }
    
    

    Created by JabRef on 20.11.24.