Motor Fault Diagnosis Using CNN Based Deep Learning Algorithm Considering Motor Rotating Speed
Motor fault is a major problem in unmanned oriented systems such as smart factories. Recently, some studies have been actively carried out to make such a fault diagnosis unattended by a deep learning algorithm. However, these studies do not take into account the speed at which motors are driven, and therefore they are not appropriate for the actual system. In this paper, the experiment was performed to develop a deep learning algorithm that considered the motor speed. The method is that regard the vibration signal as an image and uses the algorithm, CNN, which is suitable for this processing. Adopting the suitable model reduced over-fitting and increased accuracy by reducing the model complexity. As a result, it is shown that fault diagnosis can be performed considering the motor driving speed by using the deep learning algorithm.