| Presentation | 2021-12-10 Study on Detection Method of the Level Crossing Rod Breakage using the Machine Learning Hiroshi Shida, Noriyuki Shiraishi, Hiroshi Takahashi, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | The level crossing is the only part, which public road intersects a railroad. The level crossing rod is an important equipment, which blocks the railroad and the public road at the time of approaching a train. And it keeps the security of passenger, automobile and train, which pass the crossing. However, the breakdown of the level crossing rod, which is an important equipment of level crossing gate, often occurs by the automobile that carelessness crosses the crossing at the time of approaching a train. The railway company repairs it each time. Furthermore, the abnormal detection depends on the report of passenger. So, it is a problem to tell the report to maintenance center immediately. On the other hand, we study on approach for the abnormal detection of railway signaling equipment by the machine learning. In this paper, we discuss a result of analysis about long-term monitoring data of crossing gate by the Hotelling's method (Mahalanobis distance). In addition to, we report a possibility of detection of the breakdown of the level crossing rod. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | Railway signaling / Level crossing / Level crossing gate / Level crossing rod breakage / Time series data / Anomaly detection / Mahalanobis distance |
| Paper # | DC2021-62 |
| Date of Issue | 2021-12-03 (DC) |
| Conference Information | |
| Committee | DC |
|---|---|
| Conference Date | 2021/12/10(1days) |
| Place (in Japanese) | (See Japanese page) |
| Place (in English) | |
| Topics (in Japanese) | (See Japanese page) |
| Topics (in English) | |
| Chair | Hiroshi Takahashi(Ehime Univ.) |
| Vice Chair | Tatsuhiro Tsuchiya(Osaka Univ.) |
| Secretary | Tatsuhiro Tsuchiya(Nihon Univ.) |
| Assistant | |
| Paper Information | |
| Registration To | Technical Committee on Dependable Computing |
|---|---|
| Language | JPN |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | Study on Detection Method of the Level Crossing Rod Breakage using the Machine Learning |
| Sub Title (in English) | |
| Keyword(1) | Railway signaling |
| Keyword(2) | Level crossing |
| Keyword(3) | Level crossing gate |
| Keyword(4) | Level crossing rod breakage |
| Keyword(5) | Time series data |
| Keyword(6) | Anomaly detection |
| Keyword(7) | Mahalanobis distance |
| 1st Author's Name | Hiroshi Shida |
| 1st Author's Affiliation | West Japan Electric System(NESCO) |
| 2nd Author's Name | Noriyuki Shiraishi |
| 2nd Author's Affiliation | Shikoku Japan Railway Company(JR Shikoku) |
| 3rd Author's Name | Hiroshi Takahashi |
| 3rd Author's Affiliation | Ehime Univ Graduate School(Ehime Univ) |
| Date | 2021-12-10 |
| Paper # | DC2021-62 |
| Volume (vol) | vol.121 |
| Number (no) | DC-293 |
| Page | pp.pp.38-43(DC), |
| #Pages | 6 |
| Date of Issue | 2021-12-03 (DC) |