Virtual sensor design for coating thickness estimation in a hot dip galvanising line based on interpolated SOM local models

The galvanising process is usually complex and difficult to model. However, as a result of production requirements this process usually works on a reduced set of working points leading to process data with a cluster structure. An accurate description of process data can be given at a low computational cost by specifically assigning a local model to each cluster in process data space. This paper describes a virtual sensor design for coating thickness estimation in a hot dip galvanising line based on local models using SOM and GRNN neural networks.