Design of feedforward compensator for the travelling wave ultrasonic motor based on non-uniform rational B-splines curve neural network

Hung Wen Wu, Ming-Yang Cheng

Research output: Contribution to journalArticle

Abstract

In this study, a novel feedforward compensator which is based on the non-uniform rational B-splines (NURBS) curve neural network (NURBSCNN), is developed for the travelling wave ultrasonic motor (TWUSM). The proposed NURBSCNN consists of three hidden layers, in which the blending functions are adopted as the activation functions for the neurons in the first hidden layer rather than the commonly used sigmoid functions. Simulations are performed to demonstrate the feasibility of the proposed schemes. Simulation results reveal that the proposed design approach exhibits satisfactory performance.

Original languageEnglish
Pages (from-to)1671-1676
Number of pages6
JournalInformation Technology Journal
Volume9
Issue number8
DOIs
Publication statusPublished - 2010 Jan 1

Fingerprint

Ultrasonic waves
Splines
Neural networks
Neurons
Chemical activation

All Science Journal Classification (ASJC) codes

  • Computer Science (miscellaneous)

Cite this

@article{bc53e2381c9b45b4b3bc8883c0a111a0,
title = "Design of feedforward compensator for the travelling wave ultrasonic motor based on non-uniform rational B-splines curve neural network",
abstract = "In this study, a novel feedforward compensator which is based on the non-uniform rational B-splines (NURBS) curve neural network (NURBSCNN), is developed for the travelling wave ultrasonic motor (TWUSM). The proposed NURBSCNN consists of three hidden layers, in which the blending functions are adopted as the activation functions for the neurons in the first hidden layer rather than the commonly used sigmoid functions. Simulations are performed to demonstrate the feasibility of the proposed schemes. Simulation results reveal that the proposed design approach exhibits satisfactory performance.",
author = "Wu, {Hung Wen} and Ming-Yang Cheng",
year = "2010",
month = "1",
day = "1",
doi = "10.3923/itj.2010.1671.1676",
language = "English",
volume = "9",
pages = "1671--1676",
journal = "Information Technology Journal",
issn = "1812-5638",
publisher = "Asian Network for Scientific Information",
number = "8",

}

TY - JOUR

T1 - Design of feedforward compensator for the travelling wave ultrasonic motor based on non-uniform rational B-splines curve neural network

AU - Wu, Hung Wen

AU - Cheng, Ming-Yang

PY - 2010/1/1

Y1 - 2010/1/1

N2 - In this study, a novel feedforward compensator which is based on the non-uniform rational B-splines (NURBS) curve neural network (NURBSCNN), is developed for the travelling wave ultrasonic motor (TWUSM). The proposed NURBSCNN consists of three hidden layers, in which the blending functions are adopted as the activation functions for the neurons in the first hidden layer rather than the commonly used sigmoid functions. Simulations are performed to demonstrate the feasibility of the proposed schemes. Simulation results reveal that the proposed design approach exhibits satisfactory performance.

AB - In this study, a novel feedforward compensator which is based on the non-uniform rational B-splines (NURBS) curve neural network (NURBSCNN), is developed for the travelling wave ultrasonic motor (TWUSM). The proposed NURBSCNN consists of three hidden layers, in which the blending functions are adopted as the activation functions for the neurons in the first hidden layer rather than the commonly used sigmoid functions. Simulations are performed to demonstrate the feasibility of the proposed schemes. Simulation results reveal that the proposed design approach exhibits satisfactory performance.

UR - http://www.scopus.com/inward/record.url?scp=78651354285&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=78651354285&partnerID=8YFLogxK

U2 - 10.3923/itj.2010.1671.1676

DO - 10.3923/itj.2010.1671.1676

M3 - Article

VL - 9

SP - 1671

EP - 1676

JO - Information Technology Journal

JF - Information Technology Journal

SN - 1812-5638

IS - 8

ER -