Abstract
Estimation of the maximum size of inclusions in a large volume of steel is an important issue in relation to clean steels. Statistics of extremes can be used for the estimation, and one such method, developed by Murakami and coworkers, is now used by the steel industry for characterising steel specimens. An alternative method, based on the generalised Pareto distribution, has recently been developed and has advantages. A key factor in such methods is the extent to which they can discriminate between steels of different cleanness levels. In the present paper, this discrimination is demonstrated using data from polished metallographic specimens. The confidence limit intervals are estimated using the maximum likelihood method.
Get full access to this article
View all access options for this article.