A METHOD FOR ASSESSING THE IMPACT OF SIGNAL MEASUREMENT ERRORS AT THE INPUT AND OUTPUT OF A DYNAMIC SYSTEM ON THE QUALITY OF ITS IDENTIFICATION
Measuring Equipment and Metrology · 2026 · 87 (2) · 123-133
DOI: 10.23939/istcmtm2026.02.123 ↗Kurz gefasst
Fachartikel in der Zeitschrift Measuring Equipment and Metrology, 2026. Verfasst von Oleksandr Poliarus und 2 weiteren Autor:innen. Die Arbeit nennt 29 Literaturangaben.
Abstract (Originalsprache)
The paper proposes a method for assessing the impact of measurement errors of input and output signals on the quality of identification of dynamic systems. The problem of distinguishing between additive noise and random measurement errors, which in real observation conditions have similar statistical characteristics and complicate the construction of an adequate model of the system, is considered. The proposed approach is based on the use of an artificial neural network to classify noise and measurement errors based on spectral characteristics and the dependence of the absolute error on the signal amplitude. The neural network detects cases when noise or measurement errors dominate. The impact of measurement errors of input and output signals on the accuracy of determining the time constant of a dynamic system is studied. The modeling is performed for linear inertial systems of the first and second orders using white Gaussian noise as a model of random errors. It is shown that measurement errors of the output signal have a more significant impact on the quality of identification compared to measurement errors of the input signal. The quality of system identification and robustness of results were obtained using the well-known indicators Accuracy, Precision, Recall and F1-score. The obtained results can be used to form requirements for the accuracy of measuring instruments during the identification of technical systems.
Zitieren
APA
Poliarus, O., Medvedovska, Y., & Kuropiatnyk, M. (2026). A METHOD FOR ASSESSING THE IMPACT OF SIGNAL MEASUREMENT ERRORS AT THE INPUT AND OUTPUT OF A DYNAMIC SYSTEM ON THE QUALITY OF ITS IDENTIFICATION. Measuring Equipment and Metrology, 87(2), 123-133. https://doi.org/10.23939/istcmtm2026.02.123
Vancouver
Poliarus O, Medvedovska Y, Kuropiatnyk M. A METHOD FOR ASSESSING THE IMPACT OF SIGNAL MEASUREMENT ERRORS AT THE INPUT AND OUTPUT OF A DYNAMIC SYSTEM ON THE QUALITY OF ITS IDENTIFICATION. Measuring Equipment and Metrology. 2026;87(2):123-133. doi:10.23939/istcmtm2026.02.123
BibTeX
@article{Poliarus2026,
author = {Poliarus, Oleksandr and Medvedovska, Yana and Kuropiatnyk, Mykyta},
title = {A METHOD FOR ASSESSING THE IMPACT OF SIGNAL MEASUREMENT ERRORS AT THE INPUT AND OUTPUT OF A DYNAMIC SYSTEM ON THE QUALITY OF ITS IDENTIFICATION},
journal = {Measuring Equipment and Metrology},
year = {2026},
volume = {87},
number = {2},
pages = {123--133},
doi = {10.23939/istcmtm2026.02.123}
}
EUVIMED