Assessment of impact detection tchniques for aeronautical application: ANN vs. LSSVM
File(s)
Accepted version
Author(s)
Yue, N
Sharif Khodaei, Z
Type
Journal Article
Abstract
The Impact localisation in composite panels is assessed using two machine
learning techniques: least square support vector machines (LSSVM) and artificial
neural networks (ANN) with local strain signals from piezoelectric sensors. Sensor
signals from impact experiments on a composite plate as well as signals simulated by a
finite element model are used to train and test models. A comparative study shows that
LSSVM achieves better accuracy than ANN on identifying location of impacts for a
combination of large mass impact and small mass impact, in particular when less data
is available for training which is more appropriate for real aeronautical application.
Additionally, LSSVM is more capable of identifying new impact events which have not
been considered in the training process.
learning techniques: least square support vector machines (LSSVM) and artificial
neural networks (ANN) with local strain signals from piezoelectric sensors. Sensor
signals from impact experiments on a composite plate as well as signals simulated by a
finite element model are used to train and test models. A comparative study shows that
LSSVM achieves better accuracy than ANN on identifying location of impacts for a
combination of large mass impact and small mass impact, in particular when less data
is available for training which is more appropriate for real aeronautical application.
Additionally, LSSVM is more capable of identifying new impact events which have not
been considered in the training process.
Date Issued
2016-10-11
Date Acceptance
2016-08-17
Citation
Journal of Multiscale Modeling, 2016, 07
ISSN
1756-9745
Publisher
World Scientific Publishing
Journal / Book Title
Journal of Multiscale Modeling
Volume
07
Copyright Statement
© 2016 World Scientific Publishing Europe Ltd.
Subjects
Science & Technology
Physical Sciences
Mathematics, Interdisciplinary Applications
Mathematics
Passive sensing
impact detection and characterization in composites
meta-model
ANN
SVM
SUPPORT VECTOR MACHINES
PIEZOELECTRIC STRAIN SENSORS
STIFFENED COMPOSITE PANELS
IDENTIFYING IMPACTS
NEURAL-NETWORK
SYSTEM
IDENTIFICATION
PLATES
PREDICTION
LOCATION
Publication Status
Published
Article Number
1640005