Control of resistance spot welding

This paper discusses recent progress in the development of a feedback controller for the resistance spot welding process. Two key elements of our control architecture are: operating condition detection and nugget growth estimation. In this paper the operating condition problem is addressed with respect to three well-defined operating conditions. We employ a principal component-based design for the operating condition detector. The nugget growth estimation problem is addressed by employing an affine input-output model and neural networks for local feedback