Technical Brief

A Data-Driven, Statistical Feature-Based, Neural Network Method for Rotary Seal Prognostics

[+] Author and Article Information
Madhumitha Ramachandran

School of Industrial and Systems Engineering,
University of Oklahoma,
Felgar Hall, Rm 147, 865 Asp Ave.
Norman, OK 73019
e-mail: madhumitha23@ou.edu

Zahed Siddique

School of Aerospace and Mechanical Engineering,
University of Oklahoma,
Felgar Hall, Rm 212, 865 Asp Ave.
Norman, OK 73019
e-mail: zsiddique@ou.edu

1Corresponding author.

Manuscript received September 19, 2018; final manuscript received January 18, 2019; published online April 4, 2019. Assoc. Editor: Francesco Lanza di Scalea.

ASME J Nondestructive Evaluation 2(2), 024501 (Apr 04, 2019) (6 pages) Paper No: NDE-18-1034; doi: 10.1115/1.4043191 History: Received September 19, 2018; Accepted April 04, 2019

Failure of the rotary seal is one of the foremost causes of breakdown in rotary machinery, and such a failure can be catastrophic, resulting in costly downtime and large expenses. Assessing the performance degradation of the rotary seal is very important for maintenance decision-making. Although significant progress has been made over the last 5 years to understand the degradation of seals using experimental verification and numerical simulation, there is a research gap on the data-driven-based tools and methods to assess the health condition of rotary seals. In this paper, we propose a data-driven-based performance degradation assessment approach to classify the running/health condition of rotary seals, which was not considered in the previous studies. Statistical time domain features are extracted from friction torque-based degradation signals collected from a rotary setup. Wrapper-based feature selection approach is used to select relevant features, with multilayer perceptron neural network utilized as a classification technique. To validate the proposed methodology, an accelerated aging and testing procedure is developed to capture the performance of rotary seals. The study findings indicate that multilayer perceptron (MLP) classifier built using features related to the amplitude of torque signal has a better classification accuracy for unseen data when compared with logistic regression and random forest classifiers.

Copyright © 2019 by ASME
Your Session has timed out. Please sign back in to continue.


Guo, F., Jia, X., Lv, M., Wang, L., Salant, R. F., and Wang, Y., 2014, “The Effect of Aging in Oil on the Performance of a Radial Lip Seal,” Tribol. Int., 78, pp. 187–194. [CrossRef]
Jennewein, B., and Frölich, D., 2012, “Simulation of the Radial Force of Radial Shaft Seal Rings at Different Temperatures and Aging Conditions,” Proceedings of the 17th International Sealing Conference, Stuttgart, Germany, Sept. 13–14, pp. 526–545.
Lou, W., Zhang, W., Liu, X., Dai, W., and Xu, D., 2017, “Degradation of Hydrogenated Nitrile Rubber (HNBR) O-Rings Exposed to Simulated Servo System Conditions,” Polym. Degrad. Stab., 144, pp. 464–472. [CrossRef]
Liu, X., Zhang, W., Lou, W., Huang, Y., and Dai, W., 2017, “Investigation on Thermal Oxidative Aging of Nitrile Rubber (NBR) O-Rings Under Compression Stress,” IOP Conference Series: Materials Science and Engineering, IOP Publishing, Chengdu, China, Oct. 9–11, p. 012003.
Pazur, R., Cormier, J., and Korhan-Taymaz, K., 2014, “Service Life Determination of Nitrile O-Rings in Hydraulic Fluid,” Rubber Chem. Technol., 87(2), pp. 239–249. [CrossRef]
Mosallam, A., Medjaher, K., and Zerhouni, N., 2016, “Data-Driven Prognostic Method Based on Bayesian Approaches for Direct Remaining Useful Life Prediction,” J. Intell. Manuf., 27(5), pp. 1037–1048. [CrossRef]
Guo, F., Jia, X., Huang, L., Salant, R. F., and Wang, Y., 2013, “The Effect of Aging During Storage on the Performance of a Radial Lip Seal,” Polym. Degrad. Stab., 98(11), pp. 2193–2200. [CrossRef]
Guo, F., Jia, X., Longke, W., Salant, R. F., and Wang, Y., 2014, “The Effect of Wear on the Performance of a Rotary Lip Seal,” ASME J. Tribol., 136(4), p. 041703. [CrossRef]
Salant, R., 2012, “Problems and Prospects for the Simulation of Dynamic Elastomeric Seal Wear and Life,” Proceedings of the 17th International Sealing Conference, Stuttgart, Germany, Sept. 13–14, pp. 39–52.
Ramachandran, M., and Siddique, Z., 2018, “Statistical Time Domain Feature Based Approach to Assess the Performance Degradation of Rotary Seals,” Proceedings of the ASME 2018 International Mechanical Engineering Conference & Exposition IMECE 2018, Pittsburgh, PA, Nov. 9–15, p. V013T05A071.
Yang, F., Habibullah, M. S., Zhang, T., Xu, Z., Lim, P., and Nadarajan, S., 2016, “Health Index-Based Prognostics for Remaining Useful Life Predictions in Electrical Machines,” IEEE Trans. Ind. Electron., 63(4), pp. 2633–2644. [CrossRef]
Wu, Q., Yang, X., and Zhou, Q., 2012, “Pattern Recognition and its Application in Fault Diagnosis of Electromechanical System,” J. Inf. Comput. Sci., 9(8), pp. 2221–2228.
Yu, J.-B., 2011, “Bearing Performance Degradation Assessment Using Locality Preserving Projections,” Expert Syst. Appl., 38(6), pp. 7440–7450. [CrossRef]
Kuhn, M., 2012, “Variable Selection Using the Caret Package.” https://www.idg.pl/mirrors/CRAN/web/packages/caret/vignettes/caretSelection.pdf. Accessed December 17, 2018.
Sharma, A., Amarnath, M., and Kankar, P., 2016, “Feature Extraction and Fault Severity Classification in Ball Bearings,” J. Vib. Control, 22(1), pp. 176–192. [CrossRef]
Brzezinski, A., Wang, Y., Choi, D., Qiao, X., and Ni, J., 2008, “Feature-Based Tool Condition Monitoring in a Gear Shaving Application,” ASME 2008 International Manufacturing Science and Engineering Conference Collocated with the 3rd JSME/ASME International Conference on Materials and Processing, Evantson, IL, Oct. 7–10, pp. 65–74.
Lilo, M. A., Latiff, L., Abu, A. B. H., Al Mashhadany, Y. I., and Ilijan, A. K., 2015, “Gas Turbine Bearing and Vibration Classification of Using Multi-Layer Neural Network,” 2015 International Conference on Smart Sensors and Application (ICSSA), Kuala Lumpur, Malaysia, May 26–28, pp. 20–23.
Kowalski, C. T., and Orlowska-Kowalska, T., 2003, “Neural Networks Application for Induction Motor Faults Diagnosis,” Math. Comput. Simul., 63(3–5), pp. 435–448. [CrossRef]


Grahic Jump Location
Fig. 1

Cross section of a rotary seal on a rotary shaft

Grahic Jump Location
Fig. 2

Flowchart of the proposed methodology

Grahic Jump Location
Fig. 3

Cross section of seal housing

Grahic Jump Location
Fig. 4

Schematic of a test setup to test rotary seals

Grahic Jump Location
Fig. 5

Cross section of seal aging cylinder with aggressive aging fluid and aging fixture

Grahic Jump Location
Fig. 6

Filtered torque signal illustrating breakout and steady-state torque

Grahic Jump Location
Fig. 7

Architecture of multilayer perceptron neural network

Grahic Jump Location
Fig. 8

Development of aging signs

Grahic Jump Location
Fig. 9

Effect of seal wear on extracted breakout torque feature (class 1: healthy; class 2: slightly worn; class 3: significantly worn; and class 4: failed)

Grahic Jump Location
Fig. 10

Effect of seal wear on certain steady-state torque statistical features (class 1: healthy; class 2: slightly worn; class 3: significantly worn; and class 4: failed)

Grahic Jump Location
Fig. 11

Feature subset selection using RFE



Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles
Related eBook Content
Topic Collections

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In