Open AccessDetección de fallas en máquinas rotatorias utilizando parámetros no lineales
Author(s) -
Erick Eduardo Huesca-Lazcano,
Oscar Flores-Ramírez,
Gabriel Romero-Rodríguez,
Karla Cecilia Apan-Araujo
Publication year - 2019
Publication title -
revista de ingeniería mecánica
Language(s) - English
Resource type - Journals
ISSN - 2531-2189
DOI - 10.35429/jme.2019.12.3.25.33
Subject(s) - nonlinear system , computer science , lyapunov exponent , voltage , vibration , control engineering , representation (politics) , electronic engineering , control theory (sociology) , engineering , artificial intelligence , electrical engineering , acoustics , physics , control (management) , quantum mechanics , politics , chaotic , law , political science
With the development of modern electronics and the increase in processing power it is now possible to install many and diverse sensors in a single type of machinery. Temperature, vibration, pressure, voltage, etc. they are variables that are commonly monitored in rotating machinery. These variables together contain all the information related to the condition of the machine. In case of a malfunction, this will be reflected in one or more of the monitored variables. These changes can be so subtle that they can not be noticed directly in the time series. Therefore, it is necessary to transform this information into a new and more useful representation. In the present work an alternative methodology is exposed for the analysis and diagnosis of malfunctions present in components of rotating machinery. This methodology is based on the processing of time series obtained from the sensors installed in the machinery, without considering the model of this. The extraction of nonlinear parameters is presented as an alternative, among which the maximum exponent of Lyapunov stands out, as an indicator of the state of the machine. In conjunction with traditional parameters it makes it possible to detect faults masked due to non-linear behavior of dynamic systems.