Modification of PET fabrics by hyperbranched polymer: a comparative study of artificial neural networks (ANN) and statistical approach

Mahdi Hasanzadeh; Tahereh Moieni; Bentolhoda Hadavi Moghadam

doi:10.1515/polyeng-2012-0145

Article

Modification of PET fabrics by hyperbranched polymer: a comparative study of artificial neural networks (ANN) and statistical approach

Mahdi Hasanzadeh , Tahereh Moieni and Bentolhoda Hadavi Moghadam

Published/Copyright: June 18, 2013

Published by

Become an author with De Gruyter Brill

Submit Manuscript Author Information

From the journal Journal of Polymer Engineering Volume 33 Issue 5

Abstract

Hyperbranched polymers (HBPs) are highly branched, three-dimensional and polydisperse macromolecules and have been employed for modification of poly(ethylene terephthalate) (PET) fabrics. The PET fabrics treatment process parameters, like HBP concentration, temperature and time, play a major role in treatment yield and dyeability of treated PET fabrics by acid dyes. Two different quantitative models, comprising response surface methodology (RSM) and artificial neural networks (ANN), were developed for predicting color strength (K/S value) of treated fabrics. The experiments were conducted based on central composite design (CCD) and a mathematical model was developed. A comparison of the predicted color strength using RSM and ANN was studied. The results obtained indicated that both RSM and ANN models show a very good relationship between the experimental and predicted response values. However, the ANN model shows more accurate results than the RSM model.

Keywords: artificial neural network; color strength; hyperbranched polymer; PET fabric; response surface methodology

Corresponding author: Mahdi Hasanzadeh, Department of Textile Engineering, Amirkabir University of Technology, Tehran, Iran

References

[1] Jikei M, Kakimoto M. Prog. Polym. Sci. 2001, 26, 1233–1285.Search in Google Scholar

[2] Gao C, Yan D. Prog. Polym. Sci. 2004, 29, 183–275.Search in Google Scholar

[3] Schmaljohann D, Pötschke P, Hässler R, Voit BI, Froehling PE, Mostert B, Loontjens JA. Macromolecules 1999, 32, 6333–6339.10.1021/ma9902504Search in Google Scholar

[4] Seiler M. Fluid Phase Equilib. 2006, 241, 155–174.Search in Google Scholar

[5] Burkinshaw SM, Froehling PE, Mignanelli M. Dyes Pigm. 2002, 53, 229–235.Search in Google Scholar

[6] Khatibzadeh M, Mohseni M, Moradian S. Color. Technol. 2010, 126, 269–274.Search in Google Scholar

[7] Myers, RH, Montgomery, DC, Anderson-Cook, CM, Eds., Response Surface Methodology: Process and Product Optimization using Designed Experiments, 3rd ed., John Wiley and Sons: New York, 2009.Search in Google Scholar

[8] Gu SY, Ren J, Vancso GJ. Eur. Polym. J. 2005, 41, 2559–2568.Search in Google Scholar

[9] Shams Nateri A, Hasanzadeh M. J. Comput. Theor. Nanosci. 2009, 6, 1542–1545.Search in Google Scholar

[10] Dev VRG, Venugopal JR, Senthilkumar M, Gupta D, Ramakrishna S. J. Appl. Polym. Sci. 2009, 113, 3397–3404.Search in Google Scholar

[11] Galushkin, AL. Neural Networks Theory, Springer: Moscow, 2007.Search in Google Scholar

[12] Hasanzadeh M, Moieni T, Hadavi Moghadam B. Adv. Polym. Tech., 2013, 32.10.1002/adv.21345Search in Google Scholar

[13] McDonald, R. Color Physics for Industry, 2nd ed., Society of dyers and Colorists: England, 1997.Search in Google Scholar

[14] Nasouri K, Bahrambeygi H, Rabbi A, Shoushtari AM, Kaflou A. J. Appl. Polym. Sci. 2012, 126, 127–135.Search in Google Scholar

[15] Kasiri MB, Aleboyeh H, Aleboyeh A. Environ. Sci. Technol. 2008, 42, 7970–7975.Search in Google Scholar

Received: 2012-11-7

Accepted: 2013-5-20

Published Online: 2013-06-18

Published in Print: 2013-08-01

You are currently not able to access this content.

Articles in the same Issue

https://doi.org/10.1515/polyeng-2012-0145

Keywords for this article

artificial neural network; color strength; hyperbranched polymer; PET fabric; response surface methodology