Detection of the pre-failure state of the switch by the active power diagram
https://doi.org/10.52170/1815-9265_2023_67_40
Abstract
The existing technical diagnostics and monitoring systems cannot fully predetermine the state of the switch motor. First and foremost, this is associated with a large number of failure modes (in control scheme, internal components of the drive, external factors), as well as a lack of monitored parameters capable of fully describing a faulty unit. However, there are indirect assessment methods based on artificial intelligence technologies. This study examined the power characteristics obtained from diagnostic devices of power parameters. The preprocessing and modeling results justify the selection of suitable unsupervised machine learning algorithms.
About the Author
V. A. Kanarskiy
Far Eastern State Transport University
Russian Federation
Russian Federation
Vadim A. Kanarskiy – Postgraduate of the Computer Engineering and Computer Graphics Department, Lecturer of the Automation, Telemechanics, and Communication Department
Khabarovsk
References
1. Analysis of the operational activities of the automation and telemechanics infrastructure at the end of 2019. Automation and Telemechanics Department of the Central Infrastructure Directorate. Moscow: Russian Railways; 2019. 101 с. (In Russ.).1. Analysis of the operational activities of the automation and telemechanics infrastructure at the end of 2019. Automation and Telemechanics Department of the Central Infrastructure Directorate. Moscow: Russian Railways; 2019. 101 с. (In Russ.).
2. Kotov V. K., Pavlovskiy A. A., Pavlovskiy E. A. Diagnostics of the electric point drives by the current parameters. Automation, communication, informatics. 2015;(7):12–17. (In Russ.).
3. Ivanov A. A., Legonkov A. K., Molodtsov V. P. Automate diagnostics of the power parameters of the arrow electric drive. Problems of safety and reliability of microprocessor complexes. Proceedings of scientific and practical conference, St. Petersburg, May 27–28, 2015. St. Petersburg: St. Petersburg State University of Railway Transport of Emperor Alexander I; 2015. P. 110–117. (In Russ.).
4. Kanarskiy V. A. Prediction of pumping station failures using machine learning without a teacher // Bulletin of RosNow: Complex Systems: Models, Analysis and Control. 2022;(4):63–74. (In Russ.). DOI 10.18137/RNU.V9187.22.04.P.1.
5. Kanarskiy V. A. Improvement of the arrow drives diagnostics using artificial intelligence. Intelligent transport systems. Proceedings of the II International Scientific and Practical Conference (Moscow, May 25, 2023). Moscow: Russian University of Transport; 2023. С. 578–583. (In Russ.).
6. Harris C. R., Millman, K. J., van der Walt S. J. [et al.]. Array programming with NumPy. Nature. 2020;585(7825):357–362. DOI 10.1038/s41586-020-2649-2.
7. Christ M., Braun N., Neuffer J., Kempa-Liehr A. W. Time Series FeatuRe Extraction on basis of Scalable Hypothesis tests (tsfresh – A Python package). Neurocomputing. 2018;307(7):72–77. DOI 10.1016/j.neucom.2018.03.067.
8. Géron A. Hands-On Machine Learning with ScikitLearn, Keras, and TensorFlow: Concepts, Tools, and Techniques to Build Intelligent Systems. Sebastopol, CA: O’Reilly Media; 2019. 600 p.
9. Zhao Y., Nasrullah Z., Li Z. PyOD: A Python Toolbox for Scalable Outlier Detection. Journal of Machine Learning Research (JMLR). 2019;20(96):1–7.
10. Kanarskiy V. A. Investigation of machine learning efficiency in signal point monitoring. Reliability. 2023;(1):38–44. (In Russ.). https://doi.org/10.21683/1729-2646-2023-23-1-38-44.
Review
For citations:
Kanarskiy V.A. Detection of the pre-failure state of the switch by the active power diagram. Bulletin of Siberian State University of Transport. 2023;(4):40-46. (In Russ.) https://doi.org/10.52170/1815-9265_2023_67_40
Views:
7
Email this article 