Citation

Xing Y, Wang J, Shangguan W, Peng C, Zhu L. China Saf. Sci. J. 2022; 32(5): 112-118.

Copyright

(Copyright © 2022, China Occupational Safety and Health Association, Publisher Gai Xue bao)

DOI

10.16265/j.cnki.issn1003-3033.2022.05.2389

PMID

unavailable

Abstract

In order to address deviation of decision-making boundary of track circuit diagnosis model due to imbalanced monitoring data and slow training speed caused by massive data, a fault diagnosis method based on data resampling and ensemble learning algorithm was proposed. Firstly, imbalanced data were processed by feature synthesis and resampling including random down-sampling and Synthetic Minority Oversampling Technique (SMOTE). Secondly, a fault diagnosis module for massive monitoring data was constructed based on LightGBM algorithm which could be trained efficiently, training and diagnosis flow was designed, and key parameters were selected by grid search and cross-validation. Finally, Macro-F1, which was not affected by imbalanced data, was introduced as an evaluation indicator of the model. The results show that the comprehensive performance of each diagnosis model for imbalanced data can be improved by feature synthesis and data resampling. Compared with other algorithms, LightGBM is the best in terms of comprehensive performance and training time, ensuring superiority and rapidity when faced with massive data. © PHYSOR 2022 China Safety Science Journal. All rights reserved.

Language: zh