Exploratory Data Analysis on the Poincaré Features of the Electrical Torque Oriented to the Severity Diagnosis of a Gearbox Tooth Breakage

This work develops the exploratory data analysis of the Poincaré features extracted from the electrical torque of an induction motor which is coupled to a gearbox under different severity levels of the failure mode related to the tooth breakage. This research was developed because of the sensibility of the electrical torque faced to the mechanical faults in the motor load. Several features were extracted from the Poincaré plots of the electrical torque, and feature selection was applied by using different techniques such as ReliefF, Chi-2 and a cluster validity index called Composing Density Between and With clusters CD_bw. Boxplots and t-SNE plots were used to visualize the data distribution and perform the data analysis regarding nine different fault severity levels of the tooth breakage. These graphics show Poincaré features are discriminative to separate some set of fault severities, and they can be used as inputs for machine learning based classification models.