Remaining Useful Life Prediction Using Enhanced Convolutional Neural Network on Multivariate Time Series Sensor Data

Manassakan SANAYHA; Peerapon VATEEKUL

doi:10.48048/wjst.2019.4144

Authors

Manassakan SANAYHA Department of Computer Engineering, Faculty of Engineering, Chulalongkorn University, Bangkok 10330
Peerapon VATEEKUL Department of Computer Engineering, Faculty of Engineering, Chulalongkorn University, Bangkok 10330

DOI:

https://doi.org/10.48048/wjst.2019.4144

Keywords:

Multivariate time series, deep learning, convolutional neural network, remaining useful life

Abstract

All machines in power plants need high reliability and to be continuous run at all times in the production process. The Remaining Useful Life (RUL) prediction of machines is an estimation for planning maintenance activities in advance to save the cost of corrective and preventive maintenance. Most existing models analyze sensor data separately. This univariate analysis never considers the relationship between sensors and time simultaneously. In this paper, we applied a Convolutional Neural Network (CNN), which considered both dimensions of and sensors; a multivariate time series analysis. Furthermore, we applied many techniques to enhance the framework of deep learning, including dropout, L² Regularization, and the Adaptive Gradient Descent (AdaGrad). For the experiment, we conducted our method and showed the performance in term of Root Mean Square Error (RMSE) on a standard benchmark and for real-case datasets.

Downloads

Download data is not yet available.

Metrics

Metrics Loading ...

References

P Baraldi, M Compare, S Sauco and E Zio. Ensemble neural network-based particle filtering for prognostics. Mech. Syst. Signal Process. 2013; 41, 288-300.

S Porotsky. Remaining useful life estimation for systems with non-trendability behavior. In: Proceedings of the 2012 IEEE Conference on Prognostics and Health Management. 2012, p. 1-6.

K Javed, R Gouriveau and N Zerhouni. A new multivariate approach for prognostics based on extreme learning machine and fuzzy clustering. IEEE Trans. Cybernet. 2015; 45, 2626-39.

Z Yang, P Baraldi and E Zio. A comparison between extreme learning machine and artificial neural network for remaining useful life prediction. In: Proceedings of the 2016 Prognostics and System Health Management Conference. 2016, p. 1-7.

L Ren, J Cui, Y Sun and X Cheng. Multi-bearing remaining useful life collaborative prediction: A deep learning approach. J. Manuf. Syst. 2017; 43, 248-56.

Z Zhao, L Bin, X Wang and W Lu. Remaining useful life prediction of aircraft engine based on degradation pattern learning. Reliab. Eng. Syst. Saf. 2017; 164, 74-83.

GS Babu, P Zhao and XL Li. Deep convolutional neural network based regression approach for estimation of remaining useful life. In: Proceedings of the 21st International Conference Database Systems for Advanced Applications. 2016, p. 214-28.

NY Lecu, ULB To, Y Bengio and RP Haffne. Gradient-based learning applied to document recognition. Proc. IEEE 1998; 86, 2278-324.

A Saxena, K Goebel, D Simon and N Eklund. Damage propagation modeling for aircraft engine run-to-failure simulation. In: Proceedings of the International Conference on Prognostics and Health Management. 2008, p. 1-9.

Predictive Modeling Of Turbofan Degradation (Part 1). Data Scientist In A Box. Available at: https://datascientistinabox.com/2015/11/30/predictive-modeling-of-turbofan-degradation-part-1, , accessed June 2017.

O Ditlevsen and R Olesen. Statistical analysis of the virkler data on fatigue crack growth. Eng. Fract. Mech. 1986; 25, 177-95.

OF Eker, F Camci and IK Jennions. Major challenges in prognostics: Study on benchmarking prognostics datasets. In: Proceedings of the 1st European Conference of the Prognostics and Health Management Society. 2012.

Pumps & Systems. Steps to Successful Installation of Vertical Circulating Water Pumps, Available at: https://www.pumpsandsystems.com/pumps/november-2015-steps-successful-installation-vertical-circulating-water-pumps, accessed January 2019.