Quality prediction for Laser Seam Welding

This nugget is developed by the University of
Porto. It is about quality prediction for laser seam welding. In this nugget
you can learn how to develop a model for quality prediction in a well-defined
industrial process. It is split into problem definition and motivation, goal,
scenario description, solution implementation and conclusions. The first phase
of a production process is the commissioning and ramp-up phase where you will
have problems with system integration and calibration. The second phase is the
operation phase where problems can occur with installation and configuration of
new elements or with maintenance.

The laser seam
welding process is explained and how to build a model using an artificial
neural network architecture. Exploratory data analysis is used to understand
the process including performance visualization.