An application of neural network solutions to laser assisted paint stripping process of hybrid epoxy-polyester coatings on aluminum substrates M. Barletta ⁎ , A. Gisario Department of Mechanical Engineering, University of Rome “Tor Vergata”, via del Politecnico, 1-00133 Rome, Italy Received 4 April 2005; accepted in revised form 29 September 2005 Available online 14 November 2005 Abstract A neural network approach is used to model a paint stripping process performed using a 1.5 kW continuous diode laser source on an aluminum substrate coated with approximately 80 μm of a hybrid epoxy-polyester resin. Two different coating colors, namely RAL 8087 (dark) and RAL 1013 (bright), are examined in order to analyze the influence of the difference in absorption of the laser energy by the surface on process. Experimental analysis was performed first in order to find the trend of paint stripping factor (PSF) according to leading process parameters such as laser power, scan speed, defocus length, and number of passes. A statistical approach is used to discuss the experimental data found. Two neural network models are investigated, namely Multi-Layer Perceptron (MLP) and Radial Basic Function (RBF), with MLP being more reliable and effective in modeling experimental results. A sensitivity analysis on the MLP model is used to show the significance of all the input data employed. As a result of sensitivity analysis, a check between experimental and calculated trends for each investigated variable was performed, which revealed an appreciable fit between data displayed. Following this, a regression model to predict the trend of PSF according to laser fluence was developed. Finally, the regression model and the MLP model were compared and showed the high degree of accuracy of neural network solution in predicting the experimental results. © 2005 Elsevier B.V. All rights reserved. Keywords: Laser cleaning; Diode laser; Epoxy-polyester coating; Neural network model 1. Introduction Laser cleaning (LC) is an emerging laser process with great potential for use in stripping paint and coatings, in surface cleaning (i.e. contaminant and soil removal), and in metal polishing (i.e. oxides and stain removal) [1]. The process selectively vaporizes the undesired layers clinging onto the workpiece surface by using controlled irradiation emitted by a focused laser beam. Unlike mechanical and chemical stripping, laser cleaning is a ‘free contact’ and ‘free tool’ process, with no significant liquid and solid residuals being generated during operations and no ‘tool’ wear or need for ‘tool’ regeneration. Therefore, the main reason for the increasing diffusion of LC is that the undesired layers can be removed from workpiece surface avoiding techniques [2–5] which, making use of solvents [2,3], abrasives or tools [4,5], severely impact on the environment, thereby failing to respect the ever more stringent legislation and regulations. A few mechanisms have been identified which explain the leading phenomena in LC, accounting for various part geo- metries, laser process conditions, and the complex interaction of many thermal and mechanical factors [6,7]. Although these interactions are not fully understood yet, improving knowledge of the process in terms of theoretical models and control strategies [8], as well as continued development of adaptive and/ or predictive LC systems [9], means the process will offer the industry an ever increasing and a significant potential value in overcoming traditional operational cleaning problems. 2. Background and motivation Several applications of different laser systems to paint and coating stripping process can be found in the literature [10–14]. Most applied techniques are based on a pulsed ArF system using photolytic and photoacoustic effects [10,11]. Very few relevant applications of continuous laser sources have been Surface & Coatings Technology 200 (2006) 6678 – 6689 www.elsevier.com/locate/surfcoat ⁎ Corresponding author. E-mail address: barletta@mail.mec.uniroma2.it (M. Barletta). 0257-8972/$ - see front matter © 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.surfcoat.2005.09.030