The “Universal” Set of Quantitative Parameters for Reading of the Trendless Sequences

Abstract

In this paper, we want to demonstrate a set of “universal” parameters that help to read quantitatively any trendless sequence (TLS). This set will be very useful in order to select the “pattern” noise from the tested one and thereby to solve the problem of calibration of random fluctuations and express some qualitative inputs in terms of these “universal” parameters. This set of quantitative parameters allows to compare the TLS(s) of different nature (acoustic, mechanical, electrochemical, vibrational, etc.) with each other. The model example based on different random sequences generated by different distributions and the TLS(s) based on real data justify this approach and confirm its universal character. Besides, using the same algorithm we analyzed the acoustic noise recorded from the frictionless bearings (FB) in a normal state and the distorted noise from the FBs with artificially created defects. The proposed algorithm allows detecting the desired defect that initially had a qualitative description only. We do suppose that the proposed “universal” scheme free from uncontrollable errors can find a wide application in the solution of many practical problems.