Machine Learning Approach for Process Optimization of Black Nickel Electroplating

Abstract Optimization of electroplating processes is crucial yet time-consuming. In an electroplating process like black nickel plating, achieving a consistent coating becomes challenging without boric acid, which serves as a pH buffer and widens the operational process window. Despite its benefits, the use of boric acid conflicts with the growing emphasis on sustainable and eco-friendly manufacturing. This complication intensifies the intricacy of process optimization. To tackle this challenge, this study employs machine learning techniques to gain insights into optimizing a boric acid-free black nickel plating process. Specifically, three machine learning models were developed to predict coating defects, coating colour, and coating mass. Drawing from the predictions of these models, an optimization algorithm is proposed to select process parameters ensuring a flawless coating with the desired colour and mass, which is validated by experimental result. The machine learning models were able to give accurate predictions for black nickel electroplating process which behaves in a non-linear manner. This research illustrates how machine learning can enhance the manufacturing industry, particularly in the surface finishing sector, by providing an effective approach to optimize processes in non-linear scenarios, thereby improving product quality and work productivity.