Identification of shallow cracks in rotating systems by utilizing convolutional neural networks and persistence spectrum under constant speed condition

The positive benefits of early faults detection in rotating systems have led scientists to develop automated methods. Although unbalancing is the most prevalent defect rotor systems, this fault normally accompanied by other defects such as crack. In article, an effective self-acting procedure addressed identifying shallow cracks throughout steady-state operation. To classify suffering with three various depths, firstly, healthy and cracked are modeled employing finite element method (FEM). following, systems' vibration signals calculated different situations numerically; for pre-processing stage, persistence spectrum implemented. Finally, using a supervised convolutional neural network (CNN), classified regarding crack depths. result testing step revealed that hybrid has rational capacity distinguishing operation where many methods somehow powerless.