A Multi-stage Linear Prediction Model For The Irregularity Of The Longitudinal Level Over Unit Railway Sections
Price
Free (open access)
Transaction
Volume
114
Pages
10
Page Range
641 - 650
Published
2010
Size
406 kb
Paper DOI
10.2495/CR100591
Copyright
WIT Press
Author(s)
H. Chang, R. Liu & Q. Li
Abstract
Various researchers, both at home and abroad, have developed models to predict the irregularity of longitudinal level changes, which take into account different influential factors. In order to help the railway maintenance department to more accurately grasp the pattern of irregularity of longitudinal level changes, the present authors have developed a new model that uses historical track geometry inspection data from track inspection cars to predict the irregularity of longitudinal level changes for unit railway sections (which are taken as 200 metres long in this paper). Various factors affect the irregularity of the track. These mainly include train operation factors, track structure and environmental factors. However, for a certain unit railway section, key consideration may be given to the impact of passing tonnage on the irregularity of the longitudinal level. This paper establishes linear regression equations that are only applicable to each particular unit section to fit the functional relationship between the irregularity of the longitudinal level and passing tonnage. The linear fit equations are obtained from standard deviations of inspected track level data at 200 intervals. The change curve of the irregularity of the longitudinal level is divided into several stages and these different stages are fitted with different linear regression equations. A multi-stage broken line is thus formed to approximate the pattern of changes in the irregularity of the longitudinal level. On this basis, the authors put forward a multi-stage linear prediction model for the irregularity of the longitudinal level. Finally, we use inspected data collected from the Beijing-Jiulong Railway Line in 2008 and 2009 to make predictions and validate the model.
Keywords
track, irregularity of longitudinal level, linear prediction model