Optimization of woven fabric parameters for ultraviolet radiation protection and comfort using artificial neural network and genetic algorithm. Issue 8 (November 2016)

• Record Type: Journal Article
• Authors: Majumdar, Abhijit; Das, Apurba; Hatua, Piyali; Ghosh, Anindya
• Is Part Of: Neural computing & applications. Volume 27:Issue 8(2016)
• Page Start: 2567
• Page End: 2576
• Publication Date: 2016-11
• Subjects: Artificial neural network -- Fitness function -- Genetic algorithm -- Optimization -- Ultraviolet radiation; Neural networks (Computer science) -- Periodicals; Neural circuitry -- Periodicals; Artificial intelligence -- Periodicals; Neural Networks (Computer) -- Periodicals; Réseaux neuronaux (Informatique) -- Périodiques; Réseaux nerveux -- Périodiques; Intelligence artificielle -- Périodiques; 006.32
• DOI: 10.1007/s00521-015-2025-6
• Languages: English
• ISSNs: 0941-0643

Abstract

Abstract Optimization of woven fabric parameters for ultraviolet protection factor (UPF) and comfort properties has been attempted using hybrid artificial neural network (ANN)–genetic algorithm (GA) system. ANN was used for developing the prediction models, and GA was employed as an optimization tool. Four feasible combinations of UPF, air permeability and moisture vapor transmission rate (MVTR) were chosen from the Pareto charts of UPF–air permeability and UPF–MVTR. Penalty function method was adopted to form a single objective function by combining the objectives and constraints related to UPF, air permeability and MVTR. The developed ANN–GA hybrid system was executed to obtain the solution set of input parameters for achieving the targeted fabric properties. To validate the developed ANN–GA-based fabric parameter optimization system, four fabric samples were woven using the solution sets of input parameters and functional properties of these engineered fabrics were evaluated. The targeted and achieved values of fabric properties of four validation samples were in reasonably good agreement.