• Proceedings of the KSR Conference
• /
• 2009.05a
• /
• pp.66-72
• /
• 2009

In order to forecast the progress of the track irregularity, we should observe the long-term track quality and divide a track into some separated divisions which have homogeneous property. For this, we define the division of track which has homogeneous property as a 'Segment' and manage the 'TQI(Track Quality Index)' using track induction data based on each segment. In this study, we introduce some methods of estimating track quality and figure out the TQIs of sample section using new FRA TQI method. In addition, we conducted a basic study of the forecasting model for the progress of track irregularity by analyzing track maintenance data.

• Proceedings of the KSR Conference
• /
• 2003.05a
• /
• pp.768-773
• /
• 2003

The inspection of track irregularity, which is the most important index for the evaluation of the dynamic safety of the rolling stock, is performed by setting up the testing train set. The self-diagnosis for the various rolling stocks and railways can be obtained if it is possible to take the simultaneous inspection of track irregularity for the commercial vehicle while it is running and to build up a dynamic safety evaluation system. It is expected to have some good effects, such as preventing accident with the low dynamic safety, cutting cost for the testing train set and evaluating the exact influence on the rolling stock and railway. In this study, innertial measuring method, which allows us to directly measure the track irregularity from the commercial vehicle, will be considered and some overseas cases will be explored as well.

• Proceedings of the KSR Conference
• /
• 2000.11a
• /
• pp.450-457
• /
• 2000

This paper establishes a scheme to estimate track quality using standard deviation and deterioration rate of track irregularities. To provide index of quality, standard deviation and deterioration rate of track irregularities are analyed. As an index, standard deviation and deterioration rate were applied to decision of corrective maintenance and replacement maintenance, respectively. Further, this paper proposes a basic scheme for a modernization of mechanize track maintenance that using MTT(Multiple Tie Tamper) and Ballast Cleaner.

• Journal of the Korean Society for Railway
• /
• v.19 no.4
• /
• pp.547-554
• /
• 2016

This study considered how to estimate the deterioration rate of the track quality index, which represents track geometric irregularity. Most existing studies have used a simple linear regression and regarded the slope of the regression equation as the progress rate. In this paper, we present a Bayesian approach to estimate the track irregularity progress. This Bayesian approach has many advantages, among which the biggest is that it can formally include the prior distribution of parameters which can be derived from historic data or from expert experiences; then, the rate can be expressed as a probability distribution. We investigated the possibility of applying the Bayesian method to the estimation of the deterioration rate by comparing our bayesian approach to the conventional linear regression approach.

• Journal of the Korean Society for Railway
• /
• v.13 no.3
• /
• pp.304-308
• /
• 2010

This paper presents a methodology for identifying track irregularity using a wavelet transfer function. An equivalent wavelet SISO (single-input single-output) transfer function is defined by the measured track geometry and the acceleration data measured at a bogie of a train. All the measured data with various sampling frequencies were rearranged according to the constant 25cm reference recording distance of the track recording vehicle used in the field. Before applying the wavelet transform, measured data were regressed by eliminating those out of the range. The inverse wavelet transfer function is also formulated to estimate track geometry. The closeness of the estimated track geometry to the actual one is evaluated by the coherence function and also by FRF (frequency response function). A track irregularity index is defined by comparing the variance of the estimation error from the intact condition and that from the current condition. A simulation study has been carried out to examine the proposed algorithm.

• International Journal of Railway
• /
• v.7 no.1
• /
• pp.8-15
• /
• 2014

Recently, in the total railroad construction field, concrete slab tracks have been adapted in place of ballast tracks. Because ballast tracks need frequent maintenance and are difficult to handle due to increasing maintenance costs, eventually concrete slab tracks were selected as an alternative. However, owing to the hydration heat reactions and temperature-affected shrinkage of the concrete, a variety of studies to solve maintenance problems related to concrete slab tracks are underway. This study analyzed characteristics of TQI values evaluating track irregularity, investigated the relationship between crack progress and TQI, and then evaluated the correlation between the two values. Through our analysis, we found that there is a need to supplement the problems of the current method and develop a track evaluation index which takes cracking into account. We also concluded that TQI is the main decision-making tool in track maintenance.

• Journal of the Korean Society for Railway
• /
• v.12 no.4
• /
• pp.506-511
• /
• 2009

An algorithm for identifying track irregularities along the railway is presented. A baseline frequency-domain transfer function based on the equivalent SlSO(Single Input Single Output) model is defined at the intact condition between the measured track geometry of the ground displacement and the acceleration measured at a location in a train. The pre-defined transfer function at the intact condition is used inversely to predict track geometry in time with the currently measured acceleration at the same location in a train. The predicted track geometry is compared in time with that of the baseline values at the intact condition. The difference between them is calculated as an error in time and used to identify the track irregularities. An irregularity index is proposed as the ratio between the moving variance of the error at the current inspection and that at the intact condition. A 3D numerical simulation study has been carried out with a train model to verify the validity of the presented algorithm. In the analysis for the simulation, the track geometry has been considered as the displacement boundary condition varying in time.

• Journal of Korean Society of Disaster and Security
• /
• v.5 no.2
• /
• pp.71-79
• /
• 2012

Recently, in the total railroad construction field, the concrete slab track is adapted, not ballast track. Because the ballast track have the characteristics doing the ongoing maintenance and difficult to handle increasing maintenance costs, eventually the concrete slab track is selected as an alternative. However, owing to the hydration heat reactions and temperature affected shrinkage cracks related to concrete itself, a variety of studies to solve maintenance problems related to concrete slab track are underway. This study analysed characteristics of TQI values evaluating the track irregularity, searched the relationship between crack progress and TQI, and then evaluated of the correlation between the two values. Through this method, there is a need to complete the problems of the current method, only TQI is the main decision making tool in track maintenance, and also the need for the development of evaluation index considering the crack.

• Journal of the Korean Geosynthetics Society
• /
• v.19 no.1
• /
• pp.45-53
• /
• 2020

There is an increasing tendency to try to make predictive analysis using measurement data based on machine learning techniques in the railway industries. In this paper, it was predicted that Track quality index (TQI) using vehicle acceleration data based on the machine learning method. The XGB (XGBoost) was the most accurate with 85% in the all data sets. Unlike the SVM model with a single algorithm, the RF and XGB model with a ensemble system were considered to be good at the prediction performance. In the case of the Surface TQI, it is shown that the acceleration of the z axis is highly related to the vertical direction and is in good agreement with the previous studies. Therefore, it is appropriate to apply the model with the ensemble algorithm to predict the track quality index using the vehicle vibration acceleration data because the accuracy may vary depending on the applied model in the machine learning methods.