Artificial Neural Network Through Energy Value of Empirical Mode Decomposition Feature Extraction based: Application on Bearing Fault Diagnosis

Abstract

This research aims to create a system by utilizing artificial intelligence to detect damage to bearings in a single-phase induction motor. Vibration signals are used as input data into artificial intelligence. The complexity of vibration signals requires a method for extracting features in signal data. This research uses the Empirical Mode Decomposition (EMD) feature extraction method, this method produces a signal in the form of Intrinsic Mode Function (IMF). Energy calculations are obtained from the IMF. The method of artificial intelligence chosen in this research is Backpropagation Neural Network. As a result of the overall experiments conducted, it can be concluded that the Backpropagation Neural Network with input features from the EMD method produces an accuracy of 91%.