Local Linear Neural Networks Based on Principal Component Analysis

L. Ramrath; M. Münchhof; R. Isermann

doi:10.1109/ACC.2006.1657185

Local Linear Neural Networks Based on Principal Component Analysis

L. Ramrath, M. Münchhof, R. Isermann

Institute of Robotics and Cognitive Systems

Technical University Darmstadt

Abstract

A new method for the estimation of process parameters based on the Principal Component Analysis is developed. The estimator yields optimal estimation results in the case of Errors In Variables (EIV) problems which are characterized by corrupted measurements of input and output signals. As the residual generation in fault detection methods often feature EIV characteristics, the estimator can be used to identify linear models for residual calculation. To overcome the limitations on linear models, the developed estimator is integrated into the LOLIMOT approach which is able to identify nonlinear processes. The estimator is used as an alternative to the standard Least Squares estimator to identify the parameters of the local linear models. Comparative results show the better suitability of the developed estimator for the residual generation in EIV-setups.

Original language	English
Title of host publication	2006 American Control Conference
Number of pages	6
Publisher	IEEE
Publication date	01.12.2006
Pages	3050-3055
Article number	1657185
ISBN (Print)	1-4244-0209-3
ISBN (Electronic)	1-4244-0210-7
DOIs	https://doi.org/10.1109/ACC.2006.1657185
Publication status	Published - 01.12.2006
Event	2006 American Control Conference - Minneapolis, United States Duration: 14.06.2006 → 16.06.2006 Conference number: 69440

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 3 Good Health and Well-being
SDG 9 Industry, Innovation, and Infrastructure

Access to Document

10.1109/ACC.2006.1657185

Cite this

@inproceedings{0bbd88afb84d455bb499de78bd6ff39a,

title = "Local Linear Neural Networks Based on Principal Component Analysis",

abstract = "A new method for the estimation of process parameters based on the Principal Component Analysis is developed. The estimator yields optimal estimation results in the case of Errors In Variables (EIV) problems which are characterized by corrupted measurements of input and output signals. As the residual generation in fault detection methods often feature EIV characteristics, the estimator can be used to identify linear models for residual calculation. To overcome the limitations on linear models, the developed estimator is integrated into the LOLIMOT approach which is able to identify nonlinear processes. The estimator is used as an alternative to the standard Least Squares estimator to identify the parameters of the local linear models. Comparative results show the better suitability of the developed estimator for the residual generation in EIV-setups.",

author = "L. Ramrath and M. M{\"u}nchhof and R. Isermann",

year = "2006",

month = dec,

day = "1",

doi = "10.1109/ACC.2006.1657185",

language = "English",

isbn = "1-4244-0209-3",

series = "Proceedings of the American Control Conference",

publisher = "IEEE",

pages = "3050--3055",

booktitle = "2006 American Control Conference",

note = "2006 American Control Conference ; Conference date: 14-06-2006 Through 16-06-2006",

}

TY - GEN

T1 - Local Linear Neural Networks Based on Principal Component Analysis

AU - Ramrath, L.

AU - Münchhof, M.

AU - Isermann, R.

N1 - Conference code: 69440

PY - 2006/12/1

Y1 - 2006/12/1

N2 - A new method for the estimation of process parameters based on the Principal Component Analysis is developed. The estimator yields optimal estimation results in the case of Errors In Variables (EIV) problems which are characterized by corrupted measurements of input and output signals. As the residual generation in fault detection methods often feature EIV characteristics, the estimator can be used to identify linear models for residual calculation. To overcome the limitations on linear models, the developed estimator is integrated into the LOLIMOT approach which is able to identify nonlinear processes. The estimator is used as an alternative to the standard Least Squares estimator to identify the parameters of the local linear models. Comparative results show the better suitability of the developed estimator for the residual generation in EIV-setups.

AB - A new method for the estimation of process parameters based on the Principal Component Analysis is developed. The estimator yields optimal estimation results in the case of Errors In Variables (EIV) problems which are characterized by corrupted measurements of input and output signals. As the residual generation in fault detection methods often feature EIV characteristics, the estimator can be used to identify linear models for residual calculation. To overcome the limitations on linear models, the developed estimator is integrated into the LOLIMOT approach which is able to identify nonlinear processes. The estimator is used as an alternative to the standard Least Squares estimator to identify the parameters of the local linear models. Comparative results show the better suitability of the developed estimator for the residual generation in EIV-setups.

UR - https://www.scopus.com/pages/publications/34047203451

U2 - 10.1109/ACC.2006.1657185

DO - 10.1109/ACC.2006.1657185

M3 - Conference contribution

SN - 1-4244-0209-3

T3 - Proceedings of the American Control Conference

SP - 3050

EP - 3055

BT - 2006 American Control Conference

PB - IEEE

T2 - 2006 American Control Conference

Y2 - 14 June 2006 through 16 June 2006

ER -

Local Linear Neural Networks Based on Principal Component Analysis

Abstract

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this