A comparative analysis of surface roughness in robot spray painting using nano paint by Taguchi – fuzzy logic-neural network methods. Issue 1 (1st January 2023)

• Record Type: Journal Article
• Authors: sethuramalingam, Prabhu; Kiran, J. R. V. Sai; Uma, M; Thushar, T
• Is Part Of: Australian journal of mechanical engineering. Volume 21:Issue 1(2023)
• Page Start: 203
• Page End: 220
• Publication Date: 2023-01-01
• Subjects: Nano paint -- Taguchi method -- fuzzy logic -- SEM analysis -- neural network -- sensitivity analysis; Mechanical engineering -- Periodicals; Mechanical engineering -- Australia -- Periodicals; Periodicals; 621.05
• DOI: 10.1080/14484846.2020.1842157
• Languages: English
• ISSNs: 1448-4846

Abstract

ABSTRACT: Taguchi L9 orthogonal array is used to conduct the IRB1410 robot spray painting experiments to investigate the surface characteristics of Cold Rolled close Annealed (CRCA) steel workpiece. The main parameter of spray painting is considered as Distance (D), Pressure (P) and Speed (S) of the Robot. The nanopaint is prepared through the ultra-sonication process for robot spray painting applications. To optimise the robot process parameters and to achieve the accuracy level, the fuzzy expert system is integrated with Taguchi analysis. The developed truth valued fuzzy logic model is assessed the surface roughness by robot nanopaint with Carbon Nanotube (CNT). ANOVA is used to identify the noteworthy factors influencing surface roughness in robot painting process and validated using the empirical regression model. Based on the model, the influence of Robot painting process parameters on surface roughness is analysed by using the sensitivity analysis method. Further, the Neural Network model is used to predict the thickness variation of Nano painted workpiece and compared the results with experimental values.