Neural process optimization of pulse plating in the electronic industry

The purpose of the project is to optimize the pulse plating process
using mathematical models.
The traditional electronic industry uses a chemical bath with
additives to make the through-hole plating on printed circuit
boards.
The pulse plating process can make the through plating without the
additives and even improve the quality of the copper. So there is both
solution to a environmental problem and a quality problem
Based on the measurements from plated printed circuit boards the
mathematical model is used find the optimal pulse parameters. In this
project there is used artificial neural nets to solve the problem.
With the final model it should be possible to "feed" the neural net
with a CAD-design. Based on the design and the "knowledge" of the
neural net it will be possible to compute the optimal pulse
parameters.