Predicting mechanical properties of cold- rolled low carbon steel based on magnetic parameter measurement using ANFIS model

Abstract

In this paper, a novel method for predicting mechanical properties of cold- rolled low carbon steel based on magnetic parameter measurement using Adaptive Neuro Fuzzy Inference System (ANFIS) is presented. The Yield Stress (YS) and Ultimate Tensile Strength (UTS) are predicted using two ANFIS models on the basis of B-H curve parameter measurement. B-H curve parameter measurement is carried out using a measurement system specially developed for this project. Using this system, remanence (Br), coercive force (Hc), harmonic components of the field intensity, and flux density are extracted and used as input parameters of the ANFIS models. The individual influence of different input parameters is evaluated and compared with metallurgical test results. The ANFIS models show good performance and the results are in agreement with the experimental data. The developed models can be used as an on-line, non-destructive evaluation technique in steel mill factories.