Use of design of experiments to improve a neural network model in order to
predict the thickness of the chromium layer in a hard chromium plating process
F. Sánchez Lasheras1, J.A. Vilán Vilán2, P.J. García Nieto3, J.J. del Coz Díaz4
Research Department, Tecniproject SL, c/ Marqués de Pidal Nº7 – 33004 – Oviedo, (Spain)
2
Departament of Mechanical Engineering, University of Vigo, 36310 Vigo (Spain)
3
Department of Mathematics, University of Oviedo, 33007 Oviedo (Spain)
4
Construction Department, University of Oviedo, 33204 Gijón (Spain)
Emails: fsanchez@tecniproject.com; jvilan@uvigo.es; lato@orion.ciencias.uniovi.es;
juanjo@constru.uniovi.es
1
Abstract
The hard chromium plating process aims at creating a coating of hard and wear-resistant
chromium with a thickness of some micron directly on the metal part without the
insertion of copper or nickel layers. Chromium plating features high levels of hardness
and resistance to wear and it is thanks to these properties that they can be applied in a
huge range of sectors. Resistance to corrosion of a hard chromium plate depends on the
thickness of the coating, adherence and micro-fissures of the latter. This micro-fissured
structure is what provides the optimal hardness of the layers. The electro-deposited
chromium layer is not uniformly distributed. In particular, there are zones such as sharp
edges or points where deposits are highly accentuated, while deposits are virtually non
existent in holes or in the undercuts. Hard chromium plating process is one of the most
effective ways for protecting the base material against hostile environment or improving
surface properties of base material. However, in electroplating industry, electro-platers
are faced with many problems and undesirable results on chromium plated materials.
Problems such as matt deposition, milky white chromium deposition, rough or sandy
chromium deposition, insufficient thickness and hardness are the most common
problems faced in the electroplating industry. This article presents an artificial neural
network model to predict the thickness of the layer in a hard chromium plating process.
Neural networks are non-linear statistical data modelling tools. They can be used to
model complex relationships between inputs and outputs or to find patterns in data. In
order to improve the neural network performance, the design of experiments theory
(DOE) is used. The DOE is a structured, organized method for determining the
relationship between factors affecting a process and the output of that process, in this
case, the thickness of the chromium layer. In the present work the purpose of using
DOE is double: define the optimal experiments which maximize the ratio of the model
accuracy and minimize the number of experiments necessaries. Finally, conclusions of
this study are exposed.