Machine learning aided identification of train weights from railway sleeper vibration

Sakdirat Kaewunruen, Jessada Sresakoolchai, Arol Thamba

Research output: Contribution to journalArticlepeer-review

1 Citation (Scopus)
251 Downloads (Pure)


Within the UK and many other countries, the weight of a train is a critical factor in determining how much the train operator must pay to use the railway network. This study explores the use of different machine learning algorithms to predict the weight of a train based on the vibration signals recorded in a railway sleeper. The vibration signals are split into two groups: time-domain and frequency-domain signals. Then, different algorithms are developed for each group, to see which domain offers the best representation of a vibration signal for use within models to determine target values. From the study, it can be seen that machine learning has the potential to predict the weight of a train effectively. This insight can lead to the use of mobile sensors in practice, such as the application of wireless accelerometers connected with a smartphone, to help engineers audit and provide assurance of the train access integrity.
Original languageEnglish
Pages (from-to)151-159
Number of pages9
JournalInsight - Non-Destructive Testing and Condition Monitoring
Issue number3
Publication statusPublished - 1 Mar 2021

Bibliographical note

Funding Information: The authors wish to thank the European Commission for the financial sponsorship of the H2020-RISE Project No 691135 ‘RISEN: Rail Infrastructure Systems Engineering Network’, which enables a global research network that addresses the grand challenge of railway infrastructure resilience and advanced sensing in extreme environments (


  • Deep learning
  • Machine learning
  • Weight detection

ASJC Scopus subject areas

  • Mechanics of Materials
  • Mechanical Engineering
  • Metals and Alloys
  • Materials Chemistry


Dive into the research topics of 'Machine learning aided identification of train weights from railway sleeper vibration'. Together they form a unique fingerprint.

Cite this