Predicting the remaining useful life of ball bearing under dynamic loading using supervised learning
A study of ball bearing remaining useful life predictor using supervised learning
Problem Statement
Bearing failure under dynamic loading conditions is studied in this research paper. Data from the IMS NASA bearing dataset which contains bearing vibrations for multiple axes till failure was utilized to train models and classify the bearing degradation stage. Further, the bearing life or remaining useful life of the bearing was calculated using regression techniques and various algorithms. The findings of this study are published in IEEM (International conference on industrial engineering and engineering management) and the paper presented in Macau.
Technology used
Python, Pandas, Sklearn, MATLAB