Frequency response matrix estimation from partially missing data

Diana Ugryumova; Rik Pintelon; Gerd Vandersteen

Abstract

The Frequency Response Matrix (FRM) gives a nonparametric description of a Multiple-Input-Multiple-Output (MIMO) dynamic system in the frequency domain. FRM estimation quickly provides information about the behavior of a system from input-output measurements without making too many assumptions. In model estimation, we often assume to have access to the full measurement set of a system. Nevertheless, measurement devices, sensors and data communication links are prone to failures, which can result in partially missing measurement data. Using the conventional identification methods in this case could result in biased FRM estimates. Thus, our mayor goal is to find an accurate FRM estimate from partially missing output data. To achieve this we extend a recently developed method called Local Polynomial Method [Pintelon et al ,MSSP, 2011] by adding the missing values as extra unknowns. We generalize the extended LPM in the presence of missing output data for single-input-single-output systems (presented at ERNSI 2013) to the multivariate case. The new method could also be used in case there are samples missing at the input and output of the system, provided that the reference signal is completely known.

Original language	English
Title of host publication	Presentation of poster at ERNSI 2014, European Research Network on System Identification, Oostende, Belgium, September 21-24, 2014
Publication status	Published - 21 Sept 2014
Event	ERNSI 2014 - Thermae Palace Hotel, Ostend, Belgium Duration: 21 Sept 2014 → 24 Sept 2014

Workshop

Workshop	ERNSI 2014
Period	21/09/14 → 24/09/14
Other	Modelling of dynamical systems is fundamental in almost all disciplines of science and engineering, ranging from life science to plant-wide process control. Engineering uses models for the design and analysis of complex technical systems. System identification concerns the construction, estimation and validation of mathematical models of dynamical physical or engineering phenomena from experimental data. This is the 23rd version of the European Workshop on System Identification, the first one being held in Saint-Malo in 1992. All through these years the workshop has maintained the scope of bringing together European researchers in the area of System Identification, in an informal setting that gives ample opportunities for participants to meet. The workshop program is composed of lectures from invited speakers, lectures from members of the ERNSI community, and poster presentations by -particularly- the PhD students and postdocs that are active in the network.

Keywords

Frequency Response Matrix (FRM)

Handle.net

Persistent link

SRP19: Strategic Research Programme: Center for model-based system improvement - From Computer-Aided Engineering to Model-Aided Engineering
Vandersteen, G., Rolain, Y., Wambacq, P., Kuijk, M., Vandersteen, G., Rolain, Y., Wambacq, P. & Kuijk, M.
1/11/12 → 31/10/24
Project: Fundamental
DWTC282: Dynamical systems, control and optimization
Pintelon, R., Vandewalle, J., Aeyels, D., Sepulchre, R., Kinnaert, M., Vande Wouwer, A., Blondel, V., Winkin, J., Boyd, S. & Leonard, N.
1/04/12 → 30/09/17
Project: Fundamental

Cite this

@inbook{e1377be71f524177bd571ccf86a3b5c4,

title = "Frequency response matrix estimation from partially missing data",

abstract = "The Frequency Response Matrix (FRM) gives a nonparametric description of a Multiple-Input-Multiple-Output (MIMO) dynamic system in the frequency domain. FRM estimation quickly provides information about the behavior of a system from input-output measurements without making too many assumptions. In model estimation, we often assume to have access to the full measurement set of a system. Nevertheless, measurement devices, sensors and data communication links are prone to failures, which can result in partially missing measurement data. Using the conventional identification methods in this case could result in biased FRM estimates. Thus, our mayor goal is to find an accurate FRM estimate from partially missing output data. To achieve this we extend a recently developed method called Local Polynomial Method [Pintelon et al ,MSSP, 2011] by adding the missing values as extra unknowns. We generalize the extended LPM in the presence of missing output data for single-input-single-output systems (presented at ERNSI 2013) to the multivariate case. The new method could also be used in case there are samples missing at the input and output of the system, provided that the reference signal is completely known.",

keywords = "Frequency Response Matrix (FRM)",

author = "Diana Ugryumova and Rik Pintelon and Gerd Vandersteen",

year = "2014",

month = sep,

day = "21",

language = "English",

booktitle = "Presentation of poster at ERNSI 2014, European Research Network on System Identification, Oostende, Belgium, September 21-24, 2014",

note = "ERNSI 2014 ; Conference date: 21-09-2014 Through 24-09-2014",

}

TY - CHAP

T1 - Frequency response matrix estimation from partially missing data

AU - Ugryumova, Diana

AU - Pintelon, Rik

AU - Vandersteen, Gerd

PY - 2014/9/21

Y1 - 2014/9/21

N2 - The Frequency Response Matrix (FRM) gives a nonparametric description of a Multiple-Input-Multiple-Output (MIMO) dynamic system in the frequency domain. FRM estimation quickly provides information about the behavior of a system from input-output measurements without making too many assumptions. In model estimation, we often assume to have access to the full measurement set of a system. Nevertheless, measurement devices, sensors and data communication links are prone to failures, which can result in partially missing measurement data. Using the conventional identification methods in this case could result in biased FRM estimates. Thus, our mayor goal is to find an accurate FRM estimate from partially missing output data. To achieve this we extend a recently developed method called Local Polynomial Method [Pintelon et al ,MSSP, 2011] by adding the missing values as extra unknowns. We generalize the extended LPM in the presence of missing output data for single-input-single-output systems (presented at ERNSI 2013) to the multivariate case. The new method could also be used in case there are samples missing at the input and output of the system, provided that the reference signal is completely known.

AB - The Frequency Response Matrix (FRM) gives a nonparametric description of a Multiple-Input-Multiple-Output (MIMO) dynamic system in the frequency domain. FRM estimation quickly provides information about the behavior of a system from input-output measurements without making too many assumptions. In model estimation, we often assume to have access to the full measurement set of a system. Nevertheless, measurement devices, sensors and data communication links are prone to failures, which can result in partially missing measurement data. Using the conventional identification methods in this case could result in biased FRM estimates. Thus, our mayor goal is to find an accurate FRM estimate from partially missing output data. To achieve this we extend a recently developed method called Local Polynomial Method [Pintelon et al ,MSSP, 2011] by adding the missing values as extra unknowns. We generalize the extended LPM in the presence of missing output data for single-input-single-output systems (presented at ERNSI 2013) to the multivariate case. The new method could also be used in case there are samples missing at the input and output of the system, provided that the reference signal is completely known.

KW - Frequency Response Matrix (FRM)

M3 - Meeting abstract (Book)

BT - Presentation of poster at ERNSI 2014, European Research Network on System Identification, Oostende, Belgium, September 21-24, 2014

T2 - ERNSI 2014

Y2 - 21 September 2014 through 24 September 2014

ER -

Frequency response matrix estimation from partially missing data

Abstract

Workshop

Keywords

Handle.net

Fingerprint

Projects

SRP19: Strategic Research Programme: Center for model-based system improvement - From Computer-Aided Engineering to Model-Aided Engineering

DWTC282: Dynamical systems, control and optimization

Cite this