Page 248 - Adaptive Identification and Control of Uncertain Systems with Nonsmooth Dynamics

CHAPTER 16

Hysteresis Dynamics and

Modeling

16.1 INTRODUCTION
Hysteresis is commonly encountered in many practical plants, such as servo
motors, smart materials, shape memory alloys, piezoelectric ceramics, tele-
scopic actuators, etc. Hysteresis can be represented by both dynamic input-
output and static constitutive relationships, which could limit both static
and dynamic performance of feedback control systems, and thus have been
taken as a typical non-smooth dynamics. Hysteresis is a phenomenon which
is either useful or harmful depending on the application. It is useful if one
is trying to build a memory or to record a phenomenon; however, it is
harmful when trying to build a linear transducer or a low loss device. In
either case, intelligent materials and magnetostrictive materials with hys-
teresis dynamics have been recently used in medical, aerospace, ship, and
other fields [1].
Therefore, hysteresis modeling, identification, and control are of partic-
ular interests in both academic and engineering fields for decades. However,
the precise control of the systems with hysteresis is not a trivial task, since
the existence of hysteresis could seriously affect the control accuracy, and
sometimes lead to significant oscillations and even cause instability. If the
hysteresis dynamics can be modeled accurately, one can introduce appro-
priate compensation schemes.
Hysteresis models can be roughly divided into two categories [2]: phys-
ical model and phenomenological model. The physical model is related
to the physical properties, where the model parameters vary with the ob-
ject, leading to difficulties in the modeling. The commonly used physical
models include Jiles-Altherton model [3,4], Bouc-Wen model [5,6], and
so on. On the other hand, the phenomenological model is able to de-
scribe the hysteresis phenomenon, but cannot involve the physical param-
eters. The phenomenological models mainly include Preisach model [7],
Prandtl-Ishlinskii (PI) model [8], Krasnoselskii-Pokrovskii (KP) model [9]
and backlash model [10], which have been widely studied in the literature.
In the following sections of this chapter, we will introduce several widely
Adaptive Identification and Control of Uncertain Systems with Non-smooth Dynamics.
DOI: https://doi.org/10.1016/B978-0-12-813683-6.00021-0 249
Copyright © 2018 Elsevier Inc. All rights reserved.