Application of Neural Network for the Prediction of Loss in Mechanical Properties of Aramid Fabrics After Thermal Aging

Banu Özgen Keleş

Article Details

Application of Neural Network for the Prediction of Loss in Mechanical Properties of Aramid Fabrics After Thermal Aging

Authors:

Banu Özgen Keleş

Publisher:

Ege University

Main Topics:

Wearable Materials

Journal:

Textile and Apparel Vol.34 Issue 1, pp.77-86

Document Type:

Article

Publication Date:

3/2024

Document Language:

English

Document ID (Org No.):

2001314862

pISSN:

1300-3356

Keywords:

Abstract:

 Aramid fabrics are used to produce most of the flame resistant protection clothes to fulfil the protection requirements. Even though aramid fibers have good thermal stability and flame resistance properties, fabrics used in protective clothing age and loss some of their essential functions under various environmental and operational conditions during their lifetime. These conditions cause serious limitations in the use of clothing. In this study, various woven fabrics produced from aramid Nomex, Kevlar fabrics were exposed to accelerated aging tests under varying temperature and time period in order to construct Neural Network models to predict weight loss and tensile strength loss percentages of the fabrics. The results of Artificial Neural Network models demonstrate that regression values are 0.98405 for weight loss percentages and 0.99935 for tensile strength loss percentages of the fabrics. Accordingly, the proposed Artificial Neural Network models are correctly constituted and the losses in determined fabric properties is successfully predicted.

 Aramid fabrics are used to produce most of the flame resistant protection clothes to fulfil the protection requirements. Even though aramid fibers have good thermal stability and flame resistance properties, fabrics used in protective clothing age and loss some of their essential functions under various environmental and operational conditions during their lifetime. These conditions cause serious limitations in the use of clothing. In this study, various woven fabrics produced from aramid Nomex, Kevlar fabrics were exposed to accelerated aging tests under varying temperature and time period in order to construct Neural Network models to predict weight loss and tensile strength loss percentages of the fabrics. The results of Artificial Neural Network models demonstrate that regression values are 0.98405 for weight loss percentages and 0.99935 for tensile strength loss percentages of the fabrics. Accordingly, the proposed Artificial Neural Network models are correctly constituted and the losses in determined fabric properties is successfully predicted.

 Aramid fabrics are used to produce most of the flame resistant protection clothes to fulfil the protection requirements. Even though aramid fibers have good thermal stability and flame resistance properties, fabrics used in protective clothing age and loss some of their essential functions under various environmental and operational conditions during their lifetime. These conditions cause serious limitations in the use of clothing. In this study, various woven fabrics produced from aramid Nomex, Kevlar fabrics were exposed to accelerated aging tests under varying temperature and time period in order to construct Neural Network models to predict weight loss and tensile strength loss percentages of the fabrics. The results of Artificial Neural Network models demonstrate that regression values are 0.98405 for weight loss percentages and 0.99935 for tensile strength loss percentages of the fabrics. Accordingly, the proposed Artificial Neural Network models are correctly constituted and the losses in determined fabric properties is successfully predicted.

 Aramid fabrics are used to produce most of the flame resistant protection clothes to fulfil the protection requirements. Even though aramid fibers have good thermal stability and flame resistance properties, fabrics used in protective clothing age and loss some of their essential functions under various environmental and operational conditions during their lifetime. These conditions cause serious limitations in the use of clothing. In this study, various woven fabrics produced from aramid Nomex, Kevlar fabrics were exposed to accelerated aging tests under varying temperature and time period in order to construct Neural Network models to predict weight loss and tensile strength loss percentages of the fabrics. The results of Artificial Neural Network models demonstrate that regression values are 0.98405 for weight loss percentages and 0.99935 for tensile strength loss percentages of the fabrics. Accordingly, the proposed Artificial Neural Network models are correctly constituted and the losses in determined fabric properties is successfully predicted.

Article Details

Application of Neural Network for the Prediction of Loss in Mechanical Properties of Aramid Fabrics After Thermal Aging

Authors:

Banu Özgen Keleş

Publisher:

Ege University

Main Topics:

Wearable Materials

Journal:

Textile and Apparel Vol.34 Issue 1, pp.77-86

Document Type:

Article

Publication Date:

3/2024

Document Language:

English

Document ID (Org No.):

2001314862

pISSN:

1300-3356

Aramid yarns

accelerated thermal aging

weight loss

tensile strength loss

artificial neural network

Issues

Article Details

Application of Neural Network for the Prediction of Loss in Mechanical Properties of Aramid Fabrics After Thermal Aging

Authors: Banu Özgen Keleş

Publisher: Ege University

Main Topics: Wearable Materials

Journal: Textile and Apparel Vol.34 Issue 1, pp.77-86

Document Type: Article

Publication Date: 3/2024

Document Language: English

Document ID (Org No.): 2001314862

pISSN: 1300-3356

Aramid yarns

accelerated thermal aging

weight loss

tensile strength loss

artificial neural network

Issues

Authors:

Banu Özgen Keleş

Publisher:

Ege University

Main Topics:

Wearable Materials

Journal:

Textile and Apparel Vol.34 Issue 1, pp.77-86

Document Type:

Article

Publication Date:

3/2024

Document Language:

English

Document ID (Org No.):

2001314862

pISSN:

1300-3356