Abstract

In prior work (Biswas & Chatterjee 2014 Proc. R. Soc. A 470, 20130817 (doi:10.1098/rspa.2013.0817)), we developed a six-state hysteresis model from a high-dimensional frictional system. Here, we use a more intuitively appealing frictional system that resembles one studied earlier by Iwan. The basis functions now have simple analytical description. The number of states required decreases further, from six to the theoretical minimum of two. The number of fitted parameters is reduced by an order of magnitude, to just six. An explicit and faster numerical solution method is developed. Parameter fitting to match different specified hysteresis loops is demonstrated. In summary, a new two-state model of hysteresis is presented that is ready for practical implementation. Essential Matlab code is provided.

Highlights

  • In this paper, we follow up on our recent work on lowdimensional modelling of frictional hysteresis [1]

  • A 470, 20130817), we developed a six-state hysteresis model from a high-dimensional frictional system

  • We show results of fitting the hysteresis model to some arbitrary input data

Read more

Summary

Introduction

We follow up on our recent work on lowdimensional modelling of frictional hysteresis [1]. The net result is a two-state hysteresis model that captures minor loops under small reversals within larger load paths and is ready for practical numerical implementation (simple Matlab code is provided). The famous Bouc–Wen model ([9,10]; see [11]) is one-dimensional but fails to form minor loops under small reversals within larger load paths. With this background, we recently studied [1] a frictional hysteretic system given by μ sgn(x) + Kx = bf (t),. As far as we know, the two-state model developed here has no parallel in the literature

New frictional system
Reduced-order model
Choice of basis functions
Slip criterion
Slip direction
Reduced-order model using incremental map
Final reduced-order model using a differential equation
Fitting parameters to given data
Results and discussion
Closing note
Finding the slip direction η
A A ccT A
Matlab code for computing η
Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call