Recognition of plating state and quantitative evaluation of the processed surface using hyperspectral data

Abstract

Plating is a quite important technology used in a variety of sectors such as industrial products and decorative accessories. It is a special processing technology that handles various chemical substances and consists of long serial processes, plating is likely to encounter quality problems. Therefore, it is necessary to realize a method to maintain and manage plating process and evaluate the quality of plated products. The purpose of this study is the state recognition in the plating process and the quantitative evaluation of the quality of the processed surface using hyperspectral data for electroless nickel plating. The hyperspectral data is a combination of two-dimensional spatial information and spectral information with high wavelength resolution. Actual process recognition system with a hyperspectral camera was constructed to obtain and analyze the hyperspectral data of both the plating solution and the processed surface based on multivariate analysis. The non-contact and high-speed monitoring method could be realized to estimate phosphorous acid concentration which causes deterioration of plating solution. Furthermore, the quantitative evaluation of plating surface with newly defined parameters was proposed, and the monitoring results confirmed the validity of the proposed method.

Keywords

• hyperspectral data
• plating process
• multivariate analysis
• condition monitoring
• concentration estimation
• surface evaluation