Back to Search
Start Over
A new methodology for optimal rail track condition measurement using acceleration signals
- Source :
- Measurement Science and Technology. 29:075901
- Publication Year :
- 2018
- Publisher :
- IOP Publishing, 2018.
-
Abstract
- This paper describes the possibility of estimating the health of a rail track in order to improve the comfort and ride quality of rail transportation by using the measured acceleration signal. The irregularities present in the track are the primary causes of vibration and these should be identified and the faults rectified to increase the safety of train travel. In this paper, a differential evolution technique based track measurement system (DE-TMS) is proposed to measure the health of the track. The abnormality of the track has been measured using micro electro mechanical system (MEMS) accelerometers which are mounted in the axle box. A new sensing method is used to estimate the exact location of the irregularity even when the signal is absent. The DE-TMS is simulated and compared with two other optimization techniques such as the particle swarm optimization technique and the covariance matrix adaptation with evolutionary strategy to evaluate the performance. The comparison results from the investigation show that the proposed DE-TMS provides cost-effective maintenance, and increase the safety of rail transportation. Finally, an experimental study was performed in four different irregularity cases and the results are estimated.
- Subjects :
- 050210 logistics & transportation
Computer science
Applied Mathematics
05 social sciences
Particle swarm optimization
02 engineering and technology
Ride quality
Accelerometer
Track (rail transport)
Acceleration
Axle
Control theory
0502 economics and business
0202 electrical engineering, electronic engineering, information engineering
020201 artificial intelligence & image processing
CMA-ES
Evolution strategy
Instrumentation
Engineering (miscellaneous)
Subjects
Details
- ISSN :
- 13616501 and 09570233
- Volume :
- 29
- Database :
- OpenAIRE
- Journal :
- Measurement Science and Technology
- Accession number :
- edsair.doi...........ac3dd49c983640a6a6c21a743af9d901