Remaining Useful Life Prediction of Kernel Density Estimation Based on Adaptive Window Width

Hui Shi, Hui Kang, Zhizhuang Zhang, Xiuquan Sun, Fengshou Gu

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

The kernel density estimation method is completely based on the data instead of the assumption distribution models considering the data-driven remaining useful life prediction. Among them, window width is an important parameter, and the selection of window width directly affects the accuracy of the final prediction results. To solve these problems, a kernel density remaining useful life prediction method based on the real-time updating of the adaptive window width is proposed. Firstly, when the new degraded data is obtained, the k-nearest neighbor method is used to update the data density in real-time, and the window width of the adaptive kernel density estimation is used. Further, the kernel density estimation model is constructed based on the degraded data of the system, and the probability density function of the remaining useful life is obtained, and the remaining useful life of the system is predicted. Finally, the proposed method is applied to the degradation modeling and remaining useful life prediction of gears. Results obtained demonstrate the accuracy and effectiveness of the method.

Original languageEnglish
Title of host publicationProceedings of IncoME-V & CEPE Net-2020
Subtitle of host publicationCondition Monitoring, Plant Maintenance and Reliability
EditorsD. Zhen, D. Wang, T. Wang, H. Wang, B. Huang, J. K. Sinha, A. D. Ball
PublisherSpringer, Cham
Pages665-682
Number of pages18
Volume105
ISBN (Electronic)9783030757939
ISBN (Print)9783030757922
DOIs
Publication statusPublished - 16 May 2021
Event5th International Conference on Maintenance Engineering: IncoME-V and the 2020 Annual Conference of the Centre for Efficiency and Performance Engineering Network: IncoME-V & CEPE Net-2020 - Zhuhai, China
Duration: 23 Oct 202025 Oct 2020
Conference number: 5

Publication series

NameMechanisms and Machine Science
PublisherSpringer
Volume105
ISSN (Print)2211-0984
ISSN (Electronic)2211-0992

Conference

Conference5th International Conference on Maintenance Engineering: IncoME-V and the 2020 Annual Conference of the Centre for Efficiency and Performance Engineering Network
CountryChina
CityZhuhai
Period23/10/2025/10/20

Fingerprint

Dive into the research topics of 'Remaining Useful Life Prediction of Kernel Density Estimation Based on Adaptive Window Width'. Together they form a unique fingerprint.

Cite this