This paper underscores two important computational aspects associated with algorithms for data-driven inverse analysis and characterization of heat deposition processes. The first involves the fact that the mathematical foundation of these algorithms suggests that a very large class of temperature fields, associated with different types of heat deposition processes and different domain geometries can be represented parametrically by a relatively small set of functions. They include linear combinations of solutions to the heat conduction equation. Issues concerning the use of algorithms based on different types of parametric representations related to different types of heat deposition processes are discussed. A prototype analysis is presented to demonstrate many characteristics of these algorithms that are significant for their practical application. The second aspect relates to the filtering property of inversion itself due to the nature of the kernel function involved with the description of the heat source. Both of these aspects are crucially beneficial to developing methodologies that effect the real-time dynamic data-driven production parameter adjustments of manufacturing processes involving heat deposition.

This content is only available via PDF.
You do not currently have access to this content.