ANN and Non-Integer Order Modeling of ABS Solenoid Valves

This brief presents an electric model of an on/off solenoid valve used to regulate brake pressure in anti-lock braking control systems (ABS). The model has been derived with the purpose to test and validate some coil driver integrated circuits in a Hardware-in-the-loop approach. The valve model is c...

Full description

Saved in:
Bibliographic Details
Published in:IEEE transactions on control systems technology 2011-05, Vol.19 (3), p.628-635
Main Authors: Branciforte, M, Meli, A, Muscato, G, Porto, D
Format: Article
Language:English
Subjects:
ABS
ABS resins
Anti-lock braking system (ABS)
Applied sciences
artificial neural network
Artificial neural networks
Circuit testing
Control system synthesis
Control systems
Drives
Duhem model
Dynamical systems
Exact sciences and technology
Fluid dynamics
fractional order system
Fundamental areas of phenomenology (including applications)
Hysteresis
Integrated circuit modeling
Integrated circuit testing
Learning theory
Mechanical engineering. Machine design
Neural networks
Nonlinear dynamics
nonlinear model
Physics
Pressure control
Shafts, couplings, clutches, brakes
Solenoid valves
Solenoids
Solid mechanics
Structural and continuum mechanics
Valves
Vibration, mechanical wave, dynamic stability (aeroelasticity, vibration control...)
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:This brief presents an electric model of an on/off solenoid valve used to regulate brake pressure in anti-lock braking control systems (ABS). The model has been derived with the purpose to test and validate some coil driver integrated circuits in a Hardware-in-the-loop approach. The valve model is composed by magneto-dynamic, mechanical and fluid dynamic subsystems. The global system is highly nonlinear and presents hysteresis phenomena due to the ferromagnetic effect. A non-integer order system has been adopted to simply represent the high order dynamics present into the system, while Duhem equations coupled with two different neural networks are used to model the hysteresis. Several experimental trials have been performed in order to validate the suitability of the proposed approach.
ISSN:1063-6536
1558-0865
DOI:10.1109/TCST.2010.2049999