Monitoring of Railpad Long-term Condition in Turnouts Using Extreme Value Distributions

by   Pegah Barkhordari, et al.

The railpad is a key element in railway infrastructures that plays an essential role in the train-track dynamics. Presence of worn or defective railpads along railway track may lead to large wheel/rail interaction forces, and a high rate of deterioration for track components. Despite the importance of railpad, the track infrastructure managers use no inspection tool for monitoring in-service railpads over time. In this paper, a novel data-driven monitoring tool for long-term performance analysis of in-service railpads is developed based on train-induced vibration data collected by a track-side measurement system. The monitoring tool consists of a method for track resonance frequencies estimation, a temperature-frequency model for describing railpad behavior with respect to ambient temperature, and a generalized likelihood ratio test based on the generalized extreme value distribution for detecting changes in the railpad status over time. To evaluate the performance of the proposed monitoring system, the status of railpads at four different locations along a railway turnout is monitored over a period of 18 months. It is shown that the monitoring system can successfully detect changes in railpad properties over the considered period.



There are no comments yet.


page 1

page 2


A Wearable Medical Sensor for Provisional Healthcare

Thispaper presents the design and realization of a context-aware wireles...

Sequential Score Adaptation with Extreme Value Theory for Robust Railway Track Inspection

Periodic inspections are necessary to keep railroad tracks in state of g...

Accelerometer-based Bed Occupancy Detection for Automatic, Non-invasive Long-term Cough Monitoring

We present a machine learning based long-term cough monitoring system by...

A Transfer Learning-based State of Charge Estimation for Lithium-Ion Battery at Varying Ambient Temperatures

Accurate and reliable state of charge (SoC) estimation becomes increasin...

Impact of Thermal Throttling on Long-Term Visual Inference in a CPU-based Edge Device

Many application scenarios of edge visual inference, e.g., robotics or e...

Who supervises the supervisor? Model monitoring in production using deep feature embeddings with applications to workpiece inspection

The automation of condition monitoring and workpiece inspection plays an...

Long-term continuous assessment of SRAM PUF and source of random numbers

The qualities of Physical Unclonable Functions (PUFs) suffer from severa...
This week in AI

Get the week's most popular data science and artificial intelligence research sent straight to your inbox every Saturday.