Performance enhancement for masonry creep predicting model using recurrent neural networks

Ahmed El-Shafie, Aboelmagd Noureldin, M. Reda Taha, Aini Hussain, Hassan Basri

Research output: Contribution to journalArticle

3 Citations (Scopus)

Abstract

It is well established that quasi-brittle materials experience visco-elastic creep strain under sustained loads. The creep strain represents the non-instantaneous strain that occurs with time when the stress is sustained. Most of the existing creep prediction models could achieve relatively low accuracy level because of the creep dependency on large number of parameters (e.g. relative humidity, stress level, age of loading). In addition, creep strain behavior is considered as a time-dependent visco-elastic property of masonry structures. This manuscript investigates the potential use of recurrent neural networks (RNN) for predicting creep of structural masonry. The main merit of using RNN is that RNN paradigm assembles the time-dependent process within its architecture during training. Thus, RNN becomes more capable of capturing time-dependent nonlinear relationships than the existing creep prediction model. Several network architectures are examined to enhance models' performance. The results showed that RNN architectures can reduce the creep prediction error by 30% when compared to feed-forward neural network models.

Original languageEnglish
Pages (from-to)29-38
Number of pages10
JournalEngineering Intelligent Systems
Volume17
Issue number1
Publication statusPublished - Mar 2009

Fingerprint

Recurrent neural networks
Creep
Network architecture
Feedforward neural networks
Brittleness
Loads (forces)
Atmospheric humidity

Keywords

  • Masonry creep
  • Prediction model
  • Recurrent neural network

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Control and Systems Engineering
  • Computer Science Applications

Cite this

Performance enhancement for masonry creep predicting model using recurrent neural networks. / El-Shafie, Ahmed; Noureldin, Aboelmagd; Taha, M. Reda; Hussain, Aini; Basri, Hassan.

In: Engineering Intelligent Systems, Vol. 17, No. 1, 03.2009, p. 29-38.

Research output: Contribution to journalArticle

@article{6c8685e8b5e54766b2a878e87cfaef33,
title = "Performance enhancement for masonry creep predicting model using recurrent neural networks",
abstract = "It is well established that quasi-brittle materials experience visco-elastic creep strain under sustained loads. The creep strain represents the non-instantaneous strain that occurs with time when the stress is sustained. Most of the existing creep prediction models could achieve relatively low accuracy level because of the creep dependency on large number of parameters (e.g. relative humidity, stress level, age of loading). In addition, creep strain behavior is considered as a time-dependent visco-elastic property of masonry structures. This manuscript investigates the potential use of recurrent neural networks (RNN) for predicting creep of structural masonry. The main merit of using RNN is that RNN paradigm assembles the time-dependent process within its architecture during training. Thus, RNN becomes more capable of capturing time-dependent nonlinear relationships than the existing creep prediction model. Several network architectures are examined to enhance models' performance. The results showed that RNN architectures can reduce the creep prediction error by 30{\%} when compared to feed-forward neural network models.",
keywords = "Masonry creep, Prediction model, Recurrent neural network",
author = "Ahmed El-Shafie and Aboelmagd Noureldin and Taha, {M. Reda} and Aini Hussain and Hassan Basri",
year = "2009",
month = "3",
language = "English",
volume = "17",
pages = "29--38",
journal = "Engineering Intelligent Systems",
issn = "1472-8915",
publisher = "CRL Publishing",
number = "1",

}

TY - JOUR

T1 - Performance enhancement for masonry creep predicting model using recurrent neural networks

AU - El-Shafie, Ahmed

AU - Noureldin, Aboelmagd

AU - Taha, M. Reda

AU - Hussain, Aini

AU - Basri, Hassan

PY - 2009/3

Y1 - 2009/3

N2 - It is well established that quasi-brittle materials experience visco-elastic creep strain under sustained loads. The creep strain represents the non-instantaneous strain that occurs with time when the stress is sustained. Most of the existing creep prediction models could achieve relatively low accuracy level because of the creep dependency on large number of parameters (e.g. relative humidity, stress level, age of loading). In addition, creep strain behavior is considered as a time-dependent visco-elastic property of masonry structures. This manuscript investigates the potential use of recurrent neural networks (RNN) for predicting creep of structural masonry. The main merit of using RNN is that RNN paradigm assembles the time-dependent process within its architecture during training. Thus, RNN becomes more capable of capturing time-dependent nonlinear relationships than the existing creep prediction model. Several network architectures are examined to enhance models' performance. The results showed that RNN architectures can reduce the creep prediction error by 30% when compared to feed-forward neural network models.

AB - It is well established that quasi-brittle materials experience visco-elastic creep strain under sustained loads. The creep strain represents the non-instantaneous strain that occurs with time when the stress is sustained. Most of the existing creep prediction models could achieve relatively low accuracy level because of the creep dependency on large number of parameters (e.g. relative humidity, stress level, age of loading). In addition, creep strain behavior is considered as a time-dependent visco-elastic property of masonry structures. This manuscript investigates the potential use of recurrent neural networks (RNN) for predicting creep of structural masonry. The main merit of using RNN is that RNN paradigm assembles the time-dependent process within its architecture during training. Thus, RNN becomes more capable of capturing time-dependent nonlinear relationships than the existing creep prediction model. Several network architectures are examined to enhance models' performance. The results showed that RNN architectures can reduce the creep prediction error by 30% when compared to feed-forward neural network models.

KW - Masonry creep

KW - Prediction model

KW - Recurrent neural network

UR - http://www.scopus.com/inward/record.url?scp=75149152870&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=75149152870&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:75149152870

VL - 17

SP - 29

EP - 38

JO - Engineering Intelligent Systems

JF - Engineering Intelligent Systems

SN - 1472-8915

IS - 1

ER -