Detecting wheel slip from railway operational data through a combined wavelet, long short-term memory and neural network classification method (2024)

Namoano, B., Emmanouilidis, C., & Starr, A. (2024). Detecting wheel slip from railway operational data through a combined wavelet, long short-term memory and neural network classification method. Engineering Applications of Artificial Intelligence, 137(Part B), Article 109173. https://doi.org/10.1016/j.engappai.2024.109173

Namoano, Bernadin ; Emmanouilidis, Christos ; Starr, Andrew. / Detecting wheel slip from railway operational data through a combined wavelet, long short-term memory and neural network classification method. In: Engineering Applications of Artificial Intelligence. 2024 ; Vol. 137, No. Part B.

@article{c084c0f2d6b1421a991ef7cddf0e4249,

title = "Detecting wheel slip from railway operational data through a combined wavelet, long short-term memory and neural network classification method",

abstract = "Detecting wheel slip is important in railway operations, to prevent damage to wheels and tracks, reduce maintenance costs, improve safety and enhance passenger comfort. Slip activity is characterised by reduced adhesion between the wheel and the rail and limits effective braking or acceleration, causing also operational risks. It is influenced by environmental conditions, vehicle load, track and axle quality, contaminants, inclines, rail oxidation, and braking forces. This paper introduces an innovative method for wheel slip detection in operational trains, utilizing wavelet analysis combined with Long-Short Term Memory (LSTM) modelling. This method analyzes operational data to effectively identify wheel slip, showing promising results when compared to traditional classification-based machine learning methods such as decision trees, forests, logistic regression, na{\"i}ve Bayes, and support vector machines. This novel approach addresses the complexities of wheel slip detection and is capable of identifying the conditions leading to slip several seconds prior to the commencement of the slip event, offering a practical solution for real-world railway systems.",

keywords = "Anomaly detection, Classification, Railway operations, Train wheel slip, Wavelet analysis, Long short term memory",

author = "Bernadin Namoano and Christos Emmanouilidis and Andrew Starr",

year = "2024",

month = nov,

doi = "10.1016/j.engappai.2024.109173",

language = "English",

volume = "137",

journal = "Engineering Applications of Artificial Intelligence",

issn = "0952-1976",

number = "Part B",

}

Namoano, B, Emmanouilidis, C & Starr, A 2024, 'Detecting wheel slip from railway operational data through a combined wavelet, long short-term memory and neural network classification method', Engineering Applications of Artificial Intelligence, vol. 137, no. Part B, 109173. https://doi.org/10.1016/j.engappai.2024.109173

Detecting wheel slip from railway operational data through a combined wavelet, long short-term memory and neural network classification method. / Namoano, Bernadin; Emmanouilidis, Christos; Starr, Andrew.
In: Engineering Applications of Artificial Intelligence, Vol. 137, No. Part B, 109173, 11.2024.

Research output: Contribution to journal › Article › Academic › peer-review

TY - JOUR

T1 - Detecting wheel slip from railway operational data through a combined wavelet, long short-term memory and neural network classification method

AU - Namoano, Bernadin

AU - Emmanouilidis, Christos

AU - Starr, Andrew

PY - 2024/11

Y1 - 2024/11

N2 - Detecting wheel slip is important in railway operations, to prevent damage to wheels and tracks, reduce maintenance costs, improve safety and enhance passenger comfort. Slip activity is characterised by reduced adhesion between the wheel and the rail and limits effective braking or acceleration, causing also operational risks. It is influenced by environmental conditions, vehicle load, track and axle quality, contaminants, inclines, rail oxidation, and braking forces. This paper introduces an innovative method for wheel slip detection in operational trains, utilizing wavelet analysis combined with Long-Short Term Memory (LSTM) modelling. This method analyzes operational data to effectively identify wheel slip, showing promising results when compared to traditional classification-based machine learning methods such as decision trees, forests, logistic regression, naïve Bayes, and support vector machines. This novel approach addresses the complexities of wheel slip detection and is capable of identifying the conditions leading to slip several seconds prior to the commencement of the slip event, offering a practical solution for real-world railway systems.

AB - Detecting wheel slip is important in railway operations, to prevent damage to wheels and tracks, reduce maintenance costs, improve safety and enhance passenger comfort. Slip activity is characterised by reduced adhesion between the wheel and the rail and limits effective braking or acceleration, causing also operational risks. It is influenced by environmental conditions, vehicle load, track and axle quality, contaminants, inclines, rail oxidation, and braking forces. This paper introduces an innovative method for wheel slip detection in operational trains, utilizing wavelet analysis combined with Long-Short Term Memory (LSTM) modelling. This method analyzes operational data to effectively identify wheel slip, showing promising results when compared to traditional classification-based machine learning methods such as decision trees, forests, logistic regression, naïve Bayes, and support vector machines. This novel approach addresses the complexities of wheel slip detection and is capable of identifying the conditions leading to slip several seconds prior to the commencement of the slip event, offering a practical solution for real-world railway systems.

KW - Anomaly detection

KW - Classification

KW - Railway operations

KW - Train wheel slip

KW - Wavelet analysis

KW - Long short term memory

U2 - 10.1016/j.engappai.2024.109173

DO - 10.1016/j.engappai.2024.109173

M3 - Article

SN - 0952-1976

VL - 137

JO - Engineering Applications of Artificial Intelligence

JF - Engineering Applications of Artificial Intelligence

IS - Part B

M1 - 109173

ER -

Namoano B, Emmanouilidis C, Starr A. Detecting wheel slip from railway operational data through a combined wavelet, long short-term memory and neural network classification method. Engineering Applications of Artificial Intelligence. 2024 Nov;137(Part B):109173. doi: 10.1016/j.engappai.2024.109173