Detection of Tram Wheel Faults Using MEMS-Based Sensors
Micro-electromechanical-systems (MEMS) based sensors are used for monitoring the state of machines in condition-based maintenance tasks. This approach is applied at tram depots for the purpose of identifying faulty wheels on trams in order to eliminate defective trams at the entry or dispatch gates....
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-08-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/22/17/6373 |
| _version_ | 1797493236875395072 |
|---|---|
| author | Yohanis Dabesa Jelila Wiesław Pamuła |
| author_facet | Yohanis Dabesa Jelila Wiesław Pamuła |
| author_sort | Yohanis Dabesa Jelila |
| collection | DOAJ |
| description | Micro-electromechanical-systems (MEMS) based sensors are used for monitoring the state of machines in condition-based maintenance tasks. This approach is applied at tram depots for the purpose of identifying faulty wheels on trams in order to eliminate defective trams at the entry or dispatch gates. The application of MEMS-based sensors for the detection of wheel faults is the focus of this study. A method for processing of the collected sensor data is developed. It is based on assessing the energy of vibrations at different frequency bands. Maximal Overlap Discrete Wavelet Packet Transform (MODWPT) is used for obtaining a description of the sensor data. The task of finding the energy threshold for detecting faulty wheels, frequency band and parameters of MODWPT which most distinctly distinguish the wheels is the goal of the method. The weighted difference (DW) between the extreme values of energy in a frequency band for normal and faulty wheels is proposed as the measure of the ability to distinguish the wheels. The search for the solution is formulated as a discrete optimisation problem of maximising this measure. Both the simulation and experimental results indicate that faulty wheels have greater vibration energy than normal wheels. The properties of this approach are discussed and evaluated. |
| first_indexed | 2024-03-10T01:17:08Z |
| format | Article |
| id | doaj.art-7318b1d3af7a42d0af3a01d07bec6c28 |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-10T01:17:08Z |
| publishDate | 2022-08-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-7318b1d3af7a42d0af3a01d07bec6c282023-11-23T14:07:23ZengMDPI AGSensors1424-82202022-08-012217637310.3390/s22176373Detection of Tram Wheel Faults Using MEMS-Based SensorsYohanis Dabesa Jelila0Wiesław Pamuła1Department of Transport Systems, Traffic Engineering and Logistics, Faculty of Transport and Aviation Engineering, Silesian University of Technology, 40-019 Katowice, PolandDepartment of Transport Systems, Traffic Engineering and Logistics, Faculty of Transport and Aviation Engineering, Silesian University of Technology, 40-019 Katowice, PolandMicro-electromechanical-systems (MEMS) based sensors are used for monitoring the state of machines in condition-based maintenance tasks. This approach is applied at tram depots for the purpose of identifying faulty wheels on trams in order to eliminate defective trams at the entry or dispatch gates. The application of MEMS-based sensors for the detection of wheel faults is the focus of this study. A method for processing of the collected sensor data is developed. It is based on assessing the energy of vibrations at different frequency bands. Maximal Overlap Discrete Wavelet Packet Transform (MODWPT) is used for obtaining a description of the sensor data. The task of finding the energy threshold for detecting faulty wheels, frequency band and parameters of MODWPT which most distinctly distinguish the wheels is the goal of the method. The weighted difference (DW) between the extreme values of energy in a frequency band for normal and faulty wheels is proposed as the measure of the ability to distinguish the wheels. The search for the solution is formulated as a discrete optimisation problem of maximising this measure. Both the simulation and experimental results indicate that faulty wheels have greater vibration energy than normal wheels. The properties of this approach are discussed and evaluated.https://www.mdpi.com/1424-8220/22/17/6373MEMSfaulty wheel detectionMODWPT transformstramdiscrete optimization problemDW function |
| spellingShingle | Yohanis Dabesa Jelila Wiesław Pamuła Detection of Tram Wheel Faults Using MEMS-Based Sensors Sensors MEMS faulty wheel detection MODWPT transforms tram discrete optimization problem DW function |
| title | Detection of Tram Wheel Faults Using MEMS-Based Sensors |
| title_full | Detection of Tram Wheel Faults Using MEMS-Based Sensors |
| title_fullStr | Detection of Tram Wheel Faults Using MEMS-Based Sensors |
| title_full_unstemmed | Detection of Tram Wheel Faults Using MEMS-Based Sensors |
| title_short | Detection of Tram Wheel Faults Using MEMS-Based Sensors |
| title_sort | detection of tram wheel faults using mems based sensors |
| topic | MEMS faulty wheel detection MODWPT transforms tram discrete optimization problem DW function |
| url | https://www.mdpi.com/1424-8220/22/17/6373 |
| work_keys_str_mv | AT yohanisdabesajelila detectionoftramwheelfaultsusingmemsbasedsensors AT wiesławpamuła detectionoftramwheelfaultsusingmemsbasedsensors |