Sensor-Based Degradation Prediction and Prognostics for Remaining Useful Life Estimation – Validation on Experimental Data of Electric Motors

Federico Barbieri, J. Wesley Hines, Michael Sharp, and Mauro Venturini
Publication Target: 
Publication Issue: 
Special Issue Nuclear Energy PHM
Submission Type: 
Full Paper
ijphm_15_019.pdf719.35 KBMay 27, 2015 - 12:47pm

Prognostics is an emerging science of predicting the health condition of a system and/or its components, based upon current and previous system status, with the ultimate goal of accurate prediction of the Remaining Useful Life (RUL). Based on this assumption, components/systems can be monitored to track the health state during operation. Acquired data are generally processed to extract relevant features related to the degradation condition of the component/system. Often, it is beneficial to combine several of these degradation parameters through an optimization process to develop a single parameter, called prognostic parameter, which can be trended to estimate the RUL. The approach adopted in this paper consists of a prognostic procedure which involves prognostic parameter generation and General Path Model (GPM). The Genetic Algorithm (GA) and Ordinary Least Squares (OLS) estimation methods will be used to develop suitable prognostic parameters from the selected features. Both time and frequency domain analysis will be investigated. The degradation data used in this paper for methodology validation is steady-state data obtained from electric motor degradation processes. Ten three-phase motors with an electric power of 5 HP were run through temperature and humidity accelerated degradation cycles on a weekly basis and five of them presented similar degradation pathways, due to bearing failure modes. The results show that the OLS method is, on average, preferable with respect to the GA method. However, the best model performance was obtained by using GA methodology for prognostic parameter generation by means of a specific combination of time domain features. In this case, the estimated RUL provided by the model nearly coincided with the true RUL and the absolute percent error values resulted on average under 5% near the end of life. Moreover, for the considered data set, the use of more data points during feature extraction did not offer better predictive performance.

Publication Year: 
Publication Volume: 
Publication Control Number: 
Page Count: 
Submission Keywords: 
Data-driven prognostics
motor prognostics.
Submission Topic Areas: 
Data-driven methods for fault detection, diagnosis, and prognosis
Submitted by: 

follow us

PHM Society on Facebook Follow PHM Society on Twitter PHM Society on LinkedIn PHM Society RSS News Feed