Research Article
Detection of Rail Bottom Damage Defects Based on Recurrent Neural Network
@INPROCEEDINGS{10.1007/978-3-030-94182-6_36, author={Fengguang Zhou and Qinjun Zhao and Yuan Xu and Qinhua Xu and Tao Shen}, title={Detection of Rail Bottom Damage Defects Based on Recurrent Neural Network}, proceedings={IoT and Big Data Technologies for Health Care. Second EAI International Conference, IoTCare 2021, Virtual Event, October 18-19, 2021, Proceedings, Part II}, proceedings_a={IOTCARE PART 2}, year={2022}, month={6}, keywords={Rail damage detection Ultrasonic detect Recurrent Neural Network Adaptive moment estimation optimizer}, doi={10.1007/978-3-030-94182-6_36} }- Fengguang Zhou
Qinjun Zhao
Yuan Xu
Qinhua Xu
Tao Shen
Year: 2022
Detection of Rail Bottom Damage Defects Based on Recurrent Neural Network
IOTCARE PART 2
Springer
DOI: 10.1007/978-3-030-94182-6_36
Abstract
Today, the development of railway network has covered most of the country. This not only brings convenience to people, but also undoubtedly brings a lot of dangerous rail damages which will threaten the running safety of trains. One of the best ways to ensure the safety of railway transportation is to detect the possible damaged position in the rail before accidents. Under the circumstance, this paper studies the detection and identification of rail damages, establishes a set of rail bottom damage detection system based on Recurrent Neural Network, the detection accuracy is about 90%. Then, an adaptive moment estimation optimizer is added to the network model to realize the dynamic adjustment of the learning rate, and the training accuracy is also improved to over 95%. This system can quickly and accurately find out the damage at the bottom of the rail, which not only reduces the labor intensity of workers, but also improves the safety factor of railway transportation.
