A nonlinear programming technique to estimate the parameters for a temperature normalized version of the Malvern equation is presented. The general conjugate gradient algorithm is used to minimize a least-squares function formulated from data for A533B steel. A constrained solution (which provides preferential treatment for a particular data point) is also given. The results satisfactorily represent experimental data over a temperature range from −73C to 260C. The approach is well suited to design applications where data at a specific temperature of interest are limited or nonexistent.

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