Sigmoids superposition for signals approximation with a deadband for sweep in a sequence of quasi-rectangular pulses

In this article, we show how a specially constructed combination of smooth, monotonic non-linear functions and affine transformations can approximate dead-zone-looking one-dimensional signals, previously represented, as a rule, in a piecewise-linear form. Our results can help solve some open questio...

Full description

Saved in:
Bibliographic Details
Published in2020 International Conference on Electrotechnical Complexes and Systems (ICOECS) pp. 1 - 7
Main Authors Sologubov, A. Yu, Kirpichnikova, I. M.
Format Conference Proceeding
LanguageEnglish
Published IEEE 27.10.2020
Subjects
Aerospace electronics
approximation
Control systems
deadband
Fermi-Dirac statistics
intersection of sets
Mathematical model
nonlinear element
Signal processing
specific subset
Sun
superposition
Switches
Thermodynamics
Verhulst equation
Online AccessGet full text

Cover

More Information
Summary:In this article, we show how a specially constructed combination of smooth, monotonic non-linear functions and affine transformations can approximate dead-zone-looking one-dimensional signals, previously represented, as a rule, in a piecewise-linear form. Our results can help solve some open questions in analyzing systems with dead-zone-looking nonlinearities. In particular, in the study of such systems in which there are mechanical gaps, or the development of pulse-phase control systems. They can be well approximated by superposition of two symmetric or asymmetrical sigmoidal functions. Finally, numerical experiments show that the resulting formulas describe deadband-looking signals with a high degree of accuracy. This method is simple, because requires a minimum set of input parameters.
DOI:10.1109/ICOECS50468.2020.9278429