Open AccessResearch on Track Irregularity Based on K Nearest Neighbor Algorithm
Author(s) -
Ting Bian,
Yulei Chen,
Kai Tang,
Xiaojie Huang,
Zhibo Cao,
Mi Yang
Publication year - 2019
Publication title -
iop conference series. materials science and engineering
Language(s) - English
Resource type - Journals
eISSN - 1757-899X
pISSN - 1757-8981
DOI - 10.1088/1757-899x/688/4/044002
Subject(s) - track (disk drive) , displacement (psychology) , acceleration , position (finance) , feature (linguistics) , computer science , inertial frame of reference , algorithm , inertial navigation system , k nearest neighbors algorithm , artificial intelligence , geodesy , computer vision , geology , physics , operating system , psychology , linguistics , philosophy , finance , classical mechanics , quantum mechanics , economics , psychotherapist
Aiming at the safety and comfort of urban rail transit operation, a machine learning KNN algorithm is proposed to predict the degree of track irregularity. There are many fundamental factors that cause the track to be uneven, mainly reflected in the lateral irregularity of the track and the vertical irregularity. In this paper, the inertial navigation system is used to realize the acquisition of the orbital dynamic acceleration data, and the secondary displacement of the acceleration can obtain the relevant displacement information. The measurement of the gauge distance can be achieved by using a laser displacement sensor. By extracting the different displacement feature data of the same position at the same time and calculating the distance between the feature values, the purpose of predicting the degree of track damage can be achieved.