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

Abstract

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) prediction. The Genetic Algorithm (GA) and Ordinary Least Squares (OLS) optimization methods will be used to develop suitable prognostic parameters from the selected features. Both time and frequency domain analysis will be investigated. Steady-state data obtained from electric motor accelerated degradation testing is used for method validation. Ten three-phase 5HP induction were run through temperature and humidity accelerated degradation cycles on a weekly basis. Of those, five presented similar degradation pathways due to bearing failure modes. The results show that the OLS method, on average, generated the best prognostic parameter performance using a combination of time domain features. However, the best single model performance was obtained using the GA methodology. In this case, the estimated RUL nearly coincided with the true RUL with an absolute percent error averaging under 5% near the end of life.

Demonstration of a Response Time Based Remaining Useful Life (RUL) Prediction for Software Systems

“…Categorization improves the estimation of story points, impact factors and CPV by providing the necessary scaling and grouping. During the formulation of prognostic model(s) and estimation of the RUL, the goal should be to combine several features into an optimal prognostic parameter that can easily be modeled with a chosen function [28]. The rationale behind the fusion of the features is to combine several information sources into one predictive parameter that has improved robustness and an underlying trend related to the overall condition of the software.…”

Auxiliary Particle Filter for Prognostics and Health Management

“…The accelerated degradation testing was adapted from the studies of Upadhyaya et al (1997) , 1974). The detailed experiment plan can be found in (Barbieri et al, 2015). The list of collected dataset variables is shown in Table 2.…”

“…(Gnanaprakasam & Chitra, 2014). A series of accelerated run-to-failure aging experiments was conducted to simulate degradation and extremely noisy measurement data were collected by low-precision sensors (Sharp, 2012;Barbieri et al, 2015).…”

Development of a Technology Roadmap for Online Monitoring of Nuclear Power Plants

“…The use of data from the sources (Section 2) and data analytics methods (Section 3), in a PRA in a manner similar to what is currently being performed requires a systematic approach to convert the outcome from data analytics into a probabilistic risk of system, structure, or component (SSC) failure. Approaches that quantify component degradation from measurement data to estimate the RUL, or performing prognosis for condition-based maintenance, are being researched (Yadav et al 2018, Kim and Heo 2018, Barbieri et al 2015, Hu et al 2016, Coble and Hines 2011. These methods would capture SSC performance and reliability parameters in real time and provide data-based knowledge, which can be used to enable better RUL estimation.…”