Model Validation Criteria for System Identification in Time Domain

Hideo Muroi, Shuichi Adachi

Research output: Contribution to journalArticle

3 Citations (Scopus)

Abstract

The fit ratio is one of the most commonly used criteria to evaluate a result of system identification in the time domain. This criterion is given by the root mean squared error (RMSE) divided by the standard deviation of the measured signal. However, the fit ratio has some problems. For example, it can take negative values because it is not normalized, and it is easy to obtain a better value for a low-amplitude signal than for a high-amplitude signal. In this paper, we introduce some normalized criteria from the field of physical geography and consider criteria that resolve these problems. We evaluated these criteria through two case studies. We found that the correlation coefficient was effective to evaluate the phase, and a criterion obtained from the triangle inequality was effective for evaluating the gain and phase.

Original languageEnglish
Pages (from-to)86-91
Number of pages6
JournalIFAC-PapersOnLine
Volume48
Issue number28
DOIs
Publication statusPublished - 2015

Fingerprint

Identification (control systems)

Keywords

  • Correlation coefficients
  • Error criteria
  • System identification
  • Time domain
  • Validation

ASJC Scopus subject areas

  • Control and Systems Engineering

Cite this

Model Validation Criteria for System Identification in Time Domain. / Muroi, Hideo; Adachi, Shuichi.

In: IFAC-PapersOnLine, Vol. 48, No. 28, 2015, p. 86-91.

Research output: Contribution to journalArticle

@article{70e00383d25b44dea330f789b1d1f943,
title = "Model Validation Criteria for System Identification in Time Domain",
abstract = "The fit ratio is one of the most commonly used criteria to evaluate a result of system identification in the time domain. This criterion is given by the root mean squared error (RMSE) divided by the standard deviation of the measured signal. However, the fit ratio has some problems. For example, it can take negative values because it is not normalized, and it is easy to obtain a better value for a low-amplitude signal than for a high-amplitude signal. In this paper, we introduce some normalized criteria from the field of physical geography and consider criteria that resolve these problems. We evaluated these criteria through two case studies. We found that the correlation coefficient was effective to evaluate the phase, and a criterion obtained from the triangle inequality was effective for evaluating the gain and phase.",
keywords = "Correlation coefficients, Error criteria, System identification, Time domain, Validation",
author = "Hideo Muroi and Shuichi Adachi",
year = "2015",
doi = "10.1016/j.ifacol.2015.12.105",
language = "English",
volume = "48",
pages = "86--91",
journal = "IFAC-PapersOnLine",
issn = "2405-8963",
publisher = "IFAC Secretariat",
number = "28",

}

TY - JOUR

T1 - Model Validation Criteria for System Identification in Time Domain

AU - Muroi, Hideo

AU - Adachi, Shuichi

PY - 2015

Y1 - 2015

N2 - The fit ratio is one of the most commonly used criteria to evaluate a result of system identification in the time domain. This criterion is given by the root mean squared error (RMSE) divided by the standard deviation of the measured signal. However, the fit ratio has some problems. For example, it can take negative values because it is not normalized, and it is easy to obtain a better value for a low-amplitude signal than for a high-amplitude signal. In this paper, we introduce some normalized criteria from the field of physical geography and consider criteria that resolve these problems. We evaluated these criteria through two case studies. We found that the correlation coefficient was effective to evaluate the phase, and a criterion obtained from the triangle inequality was effective for evaluating the gain and phase.

AB - The fit ratio is one of the most commonly used criteria to evaluate a result of system identification in the time domain. This criterion is given by the root mean squared error (RMSE) divided by the standard deviation of the measured signal. However, the fit ratio has some problems. For example, it can take negative values because it is not normalized, and it is easy to obtain a better value for a low-amplitude signal than for a high-amplitude signal. In this paper, we introduce some normalized criteria from the field of physical geography and consider criteria that resolve these problems. We evaluated these criteria through two case studies. We found that the correlation coefficient was effective to evaluate the phase, and a criterion obtained from the triangle inequality was effective for evaluating the gain and phase.

KW - Correlation coefficients

KW - Error criteria

KW - System identification

KW - Time domain

KW - Validation

UR - http://www.scopus.com/inward/record.url?scp=84988476769&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84988476769&partnerID=8YFLogxK

U2 - 10.1016/j.ifacol.2015.12.105

DO - 10.1016/j.ifacol.2015.12.105

M3 - Article

VL - 48

SP - 86

EP - 91

JO - IFAC-PapersOnLine

JF - IFAC-PapersOnLine

SN - 2405-8963

IS - 28

ER -