In oil and gas industry, machineries and mechanical components are designed with high reliability to meet the demand of the oil field. Rotating machinery is a widely used equipment and any failure of critical components within the machinery could lead to delays and large expenses. Failure of rotary seal is one of the foremost causes of breakdown in rotary machinery and such a failure can affect the other process operations in oil and gas plants. Assessing seal degradation and severity estimation are very important for maintenance decision-making. Extracting meaningful and sensitive features that can show seal degradation from raw signals is a challenging task of degradation assessment. However, no extensive works are dedicated in this area of seals. In this paper, we perform accelerated aging and testing to capture the behavior of seals through their cycle of operation and demonstrated a statistical time domain feature based approach for extracting the sensitive features that can show seal degradation. Out of eleven statistical features extracted, seven extracted features such as mean, RMS, maximum, squared mean rooted absolute amplitude, impulse factor, crest factor, margin factor are found to be significant factors which have a potential to differentiate severity levels in seals. The findings from our work show that our approach has a potential to assess the severity in seals. As a possible extension, extracted features can be used to build a classification model to classify severity in seals which could be of great interest to the users and manufacturers of rotary seals.
Statistical Time Domain Feature Based Approach to Assess the Performance Degradation of Rotary Seals
- Views Icon Views
- Share Icon Share
- Search Site
Ramachandran, M, & Siddique, Z. "Statistical Time Domain Feature Based Approach to Assess the Performance Degradation of Rotary Seals." Proceedings of the ASME 2018 International Mechanical Engineering Congress and Exposition. Volume 13: Design, Reliability, Safety, and Risk. Pittsburgh, Pennsylvania, USA. November 9–15, 2018. V013T05A071. ASME. https://doi.org/10.1115/IMECE2018-87857
Download citation file: