Fusion of multiple parameters of signals obtained by vector magnetic flux observation for evaluation of stress loaded steel samples

Abstract

In this paper, a multiple signals features data fusion approach was performed in order to increase the reliability of assessment of the damage state of the non-oriented electrical steel elements. The signals obtained by a transducer with two sensitivity axes operating under rotating AC field for samples cut out from the steel sheet at three different angles were utilized to create database. Several signals features were considered in the process of definition of feature vector corresponding to the damage stage. Next, the database was transformed in order to abstract the information content. Finally, the function combining transformed multiple parameters was defined allowing the damage stage assessment.