A novel machine learning algorithm for interval systems approximation based on artificial neural network

In recent years, order-reduction techniques based on artificial intelligence algorithms have become a topic of interest in the structural dynamics community. In this paper, we describe a new artificial intelligence technique based on the artificial neural network used to reduce a large interval syst...

Full description

Saved in:
Bibliographic Details
Published in:Journal of intelligent manufacturing 2023-06, Vol.34 (5), p.2171-2184
Main Authors: Zerrougui, Raouf, Adamou-Mitiche, Amel B. H., Mitiche, Lahcene
Format: Article
Language:English
Subjects:
Algorithms
Approximation
Artificial intelligence
Artificial neural networks
Business and Management
Control
Dynamic stability
Dynamical systems
Machine learning
Machines
Manufacturing
Mathematical analysis
Mechatronics
Methods
Neural networks
Polynomials
Processes
Production
Robotics
Simulation
Spectrum analysis
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:In recent years, order-reduction techniques based on artificial intelligence algorithms have become a topic of interest in the structural dynamics community. In this paper, we describe a new artificial intelligence technique based on the artificial neural network used to reduce a large interval system. Applied to reduce the degree of the polynomial numerator and denominator each separately, by allowing them to learn automatically from the original system, this machine learning phase allows the algorithm to improve over time and control performance of the approximation, maintaining as much as possible the stability of the dynamic system, and reducing errors between the original system and the reduced system that are presented as a very acceptable approximation, a comparison study is presented between existing works and the proposed technique, with the help of examples from literature.
ISSN:0956-5515
1572-8145
DOI:10.1007/s10845-021-01874-0