Characterizing spring durability for automotive ride using artificial neural network analysis

Y. S. Kong, Shahrum Abdullah, D. Schramm, Mohd. Zaidi Omar, Sallehuddin Mohamed Haris, T. Bruckmann, F. Kracht

Research output: Contribution to journalArticle

Abstract

This paper presents the establishment of a relationship between coil spring fatigue life and automotive vertical vibration using neural network. During an automotive suspension design process, the suspension components are designed with the consideration of structure strength and fatigue life as well as the effects toward automotive ride. Hence, it is important to have a functional mathematical model to predict the fatigue life and automotive life simultaneously. To build the mathematical model, a multibody kinematic quarter model of suspension system was constructed to simulate force and acceleration time histories from the suspension system and the sprung mass of the vehicle model. The force time histories were used to predict the fatigue life of the coil spring while the acceleration time histories were converted into ISO vertical vibration index. A neural network model was created and used to fit the spring fatigue life and vehicle vertical vibration into a mathematical function. The neural network with 1 hidden layer and 2 neurons has shown a good fitting of the data with coefficient of determination as high as 0.88, 0.98, 0.96 for training, validation and testing, respectively. This constructed neural network serves to predict the vehicle vertical vibration using the spring fatigue life and suspension natural frequencies as input, and hence reduce the automotive suspension design process.

Original languageEnglish
Pages (from-to)47-53
Number of pages7
JournalInternational Journal of Engineering and Technology(UAE)
Volume7
Issue number3
DOIs
Publication statusPublished - 1 Jan 2018

Fingerprint

Electric network analysis
Durability
Fatigue
Fatigue of materials
Suspensions
Neural networks
Vibration
Suspensions (components)
Mathematical models
Theoretical Models
Neural Networks (Computer)
Vibrations (mechanical)
Neurons
Natural frequencies
Biomechanical Phenomena
Kinematics
Testing

Keywords

  • Curve fitting
  • Neural network
  • Spring fatigue
  • Vertical vibration

ASJC Scopus subject areas

  • Biotechnology
  • Computer Science (miscellaneous)
  • Environmental Engineering
  • Chemical Engineering(all)
  • Engineering(all)
  • Hardware and Architecture

Cite this

Characterizing spring durability for automotive ride using artificial neural network analysis. / Kong, Y. S.; Abdullah, Shahrum; Schramm, D.; Omar, Mohd. Zaidi; Mohamed Haris, Sallehuddin; Bruckmann, T.; Kracht, F.

In: International Journal of Engineering and Technology(UAE), Vol. 7, No. 3, 01.01.2018, p. 47-53.

Research output: Contribution to journalArticle

@article{392a6278bcfe48d4ba1a7482f51e1785,
title = "Characterizing spring durability for automotive ride using artificial neural network analysis",
abstract = "This paper presents the establishment of a relationship between coil spring fatigue life and automotive vertical vibration using neural network. During an automotive suspension design process, the suspension components are designed with the consideration of structure strength and fatigue life as well as the effects toward automotive ride. Hence, it is important to have a functional mathematical model to predict the fatigue life and automotive life simultaneously. To build the mathematical model, a multibody kinematic quarter model of suspension system was constructed to simulate force and acceleration time histories from the suspension system and the sprung mass of the vehicle model. The force time histories were used to predict the fatigue life of the coil spring while the acceleration time histories were converted into ISO vertical vibration index. A neural network model was created and used to fit the spring fatigue life and vehicle vertical vibration into a mathematical function. The neural network with 1 hidden layer and 2 neurons has shown a good fitting of the data with coefficient of determination as high as 0.88, 0.98, 0.96 for training, validation and testing, respectively. This constructed neural network serves to predict the vehicle vertical vibration using the spring fatigue life and suspension natural frequencies as input, and hence reduce the automotive suspension design process.",
keywords = "Curve fitting, Neural network, Spring fatigue, Vertical vibration",
author = "Kong, {Y. S.} and Shahrum Abdullah and D. Schramm and Omar, {Mohd. Zaidi} and {Mohamed Haris}, Sallehuddin and T. Bruckmann and F. Kracht",
year = "2018",
month = "1",
day = "1",
doi = "10.14419/ijet.v7i3.17.16622",
language = "English",
volume = "7",
pages = "47--53",
journal = "International Journal of Engineering and Technology(UAE)",
issn = "2227-524X",
publisher = "Science Publishing Corporation Inc",
number = "3",

}

TY - JOUR

T1 - Characterizing spring durability for automotive ride using artificial neural network analysis

AU - Kong, Y. S.

AU - Abdullah, Shahrum

AU - Schramm, D.

AU - Omar, Mohd. Zaidi

AU - Mohamed Haris, Sallehuddin

AU - Bruckmann, T.

AU - Kracht, F.

PY - 2018/1/1

Y1 - 2018/1/1

N2 - This paper presents the establishment of a relationship between coil spring fatigue life and automotive vertical vibration using neural network. During an automotive suspension design process, the suspension components are designed with the consideration of structure strength and fatigue life as well as the effects toward automotive ride. Hence, it is important to have a functional mathematical model to predict the fatigue life and automotive life simultaneously. To build the mathematical model, a multibody kinematic quarter model of suspension system was constructed to simulate force and acceleration time histories from the suspension system and the sprung mass of the vehicle model. The force time histories were used to predict the fatigue life of the coil spring while the acceleration time histories were converted into ISO vertical vibration index. A neural network model was created and used to fit the spring fatigue life and vehicle vertical vibration into a mathematical function. The neural network with 1 hidden layer and 2 neurons has shown a good fitting of the data with coefficient of determination as high as 0.88, 0.98, 0.96 for training, validation and testing, respectively. This constructed neural network serves to predict the vehicle vertical vibration using the spring fatigue life and suspension natural frequencies as input, and hence reduce the automotive suspension design process.

AB - This paper presents the establishment of a relationship between coil spring fatigue life and automotive vertical vibration using neural network. During an automotive suspension design process, the suspension components are designed with the consideration of structure strength and fatigue life as well as the effects toward automotive ride. Hence, it is important to have a functional mathematical model to predict the fatigue life and automotive life simultaneously. To build the mathematical model, a multibody kinematic quarter model of suspension system was constructed to simulate force and acceleration time histories from the suspension system and the sprung mass of the vehicle model. The force time histories were used to predict the fatigue life of the coil spring while the acceleration time histories were converted into ISO vertical vibration index. A neural network model was created and used to fit the spring fatigue life and vehicle vertical vibration into a mathematical function. The neural network with 1 hidden layer and 2 neurons has shown a good fitting of the data with coefficient of determination as high as 0.88, 0.98, 0.96 for training, validation and testing, respectively. This constructed neural network serves to predict the vehicle vertical vibration using the spring fatigue life and suspension natural frequencies as input, and hence reduce the automotive suspension design process.

KW - Curve fitting

KW - Neural network

KW - Spring fatigue

KW - Vertical vibration

UR - http://www.scopus.com/inward/record.url?scp=85051001512&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85051001512&partnerID=8YFLogxK

U2 - 10.14419/ijet.v7i3.17.16622

DO - 10.14419/ijet.v7i3.17.16622

M3 - Article

VL - 7

SP - 47

EP - 53

JO - International Journal of Engineering and Technology(UAE)

JF - International Journal of Engineering and Technology(UAE)

SN - 2227-524X

IS - 3

ER -