Condition monitoring of ductile cast iron with acoustic emission entropy during tensile loading: Energy parameter reporting

In this study, ductile cast iron (EN-GJS700-2) was evaluated to identify the initiation of cracking and failure. Uniform uniaxial tensile tests were performed on the specimens, extracted from the web of an automobile crankshaft under displacement-controlled conditions according to Iso-6892 standard. The crack initiation detection and failure were performed by the acoustic emission information entropy method. The results of this study showed that the detection of crack initiation in the method based on acoustic emission parameters is associated with uncertainties, while Shannon logarithmic cumulative entropy and Kullback-Liebler logarithmic cumulative relative entropy can be used with good accuracy to detect crack initiation and failure. Another achievement of this study is the presentation of two experimental failure criteria based on acoustic emission entropy, which makes the evaluation approach based on acoustic emission entropy for real-time monitoring purposes to be very practical. Among the two logarithmic cumulative entropies of Shannon and Kullback-Liebler relative, the results showed that Kullback-Liebler relative entropy is more suitable for detecting crack initiation.