Research on Fault Diagnosis Method for High-speed Railway Signal Equipment Based on Deep Learning Integration

被引：0

作者：

Li X. ^{[1
]}

Zhang P. ^{[2
]}

Shi T. ^{[3
]}

Li P. ^{[1
]}

机构：

[1] Institute of Computing Technology, China Academy of Railway Sciences Corporation Limited, Beijing

[2] Standards & Metrology Research Institute, China Academy of Railway Sciences Corporation Limited, Beijing

[3] China Academy of Railway Sciences Corporation Limtted, Beijing

来源：

Tiedao Xuebao/Journal of the China Railway Society | 2020年 / 42卷 / 12期

关键词：

ADASYN data synthesis; Deep learning; Fault diagnosis; High-speed railway signal equipment; Integrated learning;

D O I：

10.3969/j.issn.1001-8360.2020.12.013

中图分类号：

学科分类号：

摘要：

A fault diagnosis method for high-speed railway signal equipment was proposed based on the text data of high-speed railway equipment faults. According to the fault text data of signal turnout equipment and the expert experience, the two-level turnout fault diagnosis categories were constructed. To address the imbalance of fault samples of signal turnout equipment, the ADASYN sample synthesis method was used to synthesize fault samples with few categories. In the fault diagnosis model, TF-IDF was used to extract text features, while deep learning BiGRU and BiLSTM were applied to classify texts, and combined weight calculation method was designed to integrate the learning results of deep learning. Experimental verification conducted using the signal turnout fault data generated by high-speed railway from 2009 to 2018 proves that the deep learning integration method can further improve the performance of fault diagnosis and classification of signal equipment. © 2020, Department of Journal of the China Railway Society. All right reserved.

引用

页码：97 / 105

页数：8

共 13 条

[1] (2018)
[2] CHAWLA N V, BOWYER K W, HALL L O, Et al., SMOTE: Synthetic Minority Over-sampling Technique[J], Journal of Artificial Intelligence Research, 16, 1, pp. 321-357, (2002)
[3] HE H B, BAI Y, GARCIA E A, Et al., ADASYN: Adaptive Synthetic Sampling Approach for Imbalanced Learning[C], IEEE International Joint Conference on Neural Networks, pp. 1322-1328, (2008)
[4] YANG Lianbao, LI Ping, XUE Rui, Et al., Intelligent Fault Classification of Railway Signal Equipment Based on Imbalanced Text Data Mining, Journal of the China Railway Society, 40, 2, pp. 59-66, (2018)
[5] WU L, STEVEN H, YU N H, Et al., Semantics-preserving Bag-of-words Models and Applications, IEEE Transactions on Image Processing, 19, 7, pp. 1908-1920, (2010)
[6] ZHAI C X, LAFFERTY J., A Study of Smoothing Methods for Language Models Applied to Information Retrieval, Acm Transactions on Information Systems, 22, 2, pp. 179-214, (2004)
[7] PETER D T, PANTEL P., From Frequency to Meaning: Vector Space Models of Semantics, Journal of Artificial Intelligence Research, 37, 1, pp. 141-188, (2010)
[8] ZHAO Yang, XU Tianhua, Text Mining Based Fault Diagnosis for Vehicle On-board Equipment of High Speed Railway Signal System, Journal of the China Railway Society, 37, pp. 53-59, (2015)
[9] WANG C L, JIANG F J, YANG H X., A Hybrid Framework for Text Modeling with Convolutional RNN, 23th ACM SIGKDD International Conference, (2017)
[10] LEI T, BARZILAY R, JAAKKOLA T., Molding CNNs for Text: Non-linear, Non-consecutive Convolutions[J], Indiana University Mathematics Journal, 58, 3, pp. 1151-1186, (2015)

← 1 2 →