Comparison of RSM and ANN Modeling Approaches in Predicting the Laser Phase Transformation Hardening Parameters on the Heat Input and Hardened-Bead Profile Quality of Unalloyed Titanium

Abstract

In the present work, laser transformation hardening (LTH) of unalloyed titanium, nearer to ASTM Grade 3 of chemical composition was investigated using CW 2kW, Nd: YAG laser. The laser process variables such as laser power, scanning speed, and focused position play a major role in deciding the laser hardened bead quality. Two methods, Response Surface Methodology (RSM) and Artificial Neural Network (ANN) were used to predict the heat input and hardened bead geometry parameters such as hardened bead width, hardened depth, and angle of entry of hardened bead profile of laser transformation hardened unalloyed titanium sheet of 1.6 mm thickness, keeping argon gas flow rate of 10 lpm as fixed input parameter. This paper also describes the comparison of the heat input and hardened bead geometry parameters such as hardened bead width, hardened depth, angle of entry of hardened bead profile (AEHB) predictive models based on Artificial Neural Networks (ANN) and Response Surface Methodology (RSM). The experimental plan was based on a 3-factor, 3-level Box–Behnken surface statistical design matrix method. Linear and quadratic polynomial equations were developed for proper process parametric study for its optimal performance characteristics using Response Surface Methodology. The experimental results under optimum conditions were compared with the simulated values obtained from the RSM and ANN models. Adequacy of the developed models was tested by analysis of variance (ANOVA) technique. A multilayer feed forward neural network with a Levenberg-Marquardt back-propagation (LMBP) algorithm was adopted to develop the relationships between the laser hardening process parameters and heat input, hardened bead width, hardened depth and angle of entry of hardened bead profile. The performance of the developed ANN models were compared with the second-order RSM mathematical models of heat input, hardened bead width, hardened depth and angle of entry of hardened bead profile (AEHB).There was good agreement between the experimental and simulated values of RSM and ANN. The comparison clearly indicates that the ANN models provide more accurate prediction compared to the RSM models. The details of experimentation, model development, testing, and performance comparison of RSM and ANN models are presented in the paper. The results of Box–Behnken design of RSM models and ANN models also indicate that the proposed models predict the responses adequately within the limits of input parameters being used. It is suggested that regression equations can be used to find optimum conditions for hardened bead geometry parameters.

Keywords: LTH, Nd:YAG laser, RSM,ANN, Box-Behnken

Cite this Article:

Duradundi Sawant Badkar, Krishnashankar Pandey, BuvanashekarA.N., Comparison of RSM and ANN Modeling Approaches in Predicting the Laser Phase Transformation Hardening Parameters on the Heat Input and Hardened-Bead Profile Quality of Unalloyed Titanium. Journal of Mater-ials & Metallurgical Engineering.2015; 5(1): 36–59p.