• Journal of the Korean Society for Railway
• /
• v.14 no.3
• /
• pp.203-210
• /
• 2011

This study formulates the theoretical wheel-set model to evaluate wheel-climbing derailments of rolling stock due to collision, and verifies this theory with dynamic simulations. The impact forces occurring during collision are transmitted from a car body to axles through suspensions. As a result of combinations of horizontal and vertical forces applied to axles, rolling stock may lead to derailment. The derailment type will depend on the combinations of the horizontal and vertical forces, flange angle and friction coefficient. According to collision conditions, the wheel-lift, wheel-climbing or roll-over derailments can occur between wheel and rail. In this theoretical derailment model of wheelset, the wheel-climbing derailment types are classified into Climb-over, Climb/roll-over, and pure Roll-over according to derailment mechanism between wheel and rail, and we proposed the theoretical conditions to generate each derailment mechanism. The theoretical wheel-set model was verified by dynamic simulations.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.37 no.10
• /
• pp.1207-1218
• /
• 2013

It is difficult to predict derailment with the existing derailment coefficient like Nadal's formula which is based on the contact forces between one wheel and rail. A new derailment coefficient model developed on a wheelset is able to make a better estimate about the climb derailment, slip derailment, roll over derailment, and mixed derailment types of these. Moreover, not only the mechanical factors considered in the existing derailment coefficients but also other various factors affecting derailment such as wheel unloading and loading, diameter of wheel, and locations of axle-box bearings can be covered with this new derailment coefficient model. That is, the derailment patterns which couldn't be solved with the existing formulas such as Nadal's and Weinstock's models can be analyzed with this wheelset derailment coefficient model because of considering various factors causing derailment. Finally, the validity of the new derailment coefficient model is verified using dynamic model simulations.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.38 no.7
• /
• pp.699-709
• /
• 2014

This paper proposes a new method for predicting the derailment of a running train under cross-wind conditions, using the single wheelset derailment theory. The conventional theories used for predicting the derailment due to cross-winds were developed under the assumption that derailment will always be of the roll-over type, thus neglecting other possible types such as wheel-climbing, which may occur under special driving conditions. In addition, these theories do not consider running conditions such as dynamic wheel-rail interactions and friction effects. The new method considers the effects of dynamic wheel-rail interaction as well as those of lateral acceleration, rail cant, and cross-winds. The results of this method were compared and verified with those of the conventional methods and numerical simulations.

• Proceedings of the KSR Conference
• /
• 2011.10a
• /
• pp.1868-1873
• /
• 2011

In this paper, a theoretical formulation of a simplified wheel-set model for collision-induced derailments was evaluated by numerical simulations for the wheel-climb derailment and wheel-lift derailment types. The derailment types were classified into the wheel-climb derailment and the wheel-lift derailment according to the friction force direction of the wheel-flange. The wheel-climb derailment type was classified into Climb-up, Climb/Roll-over, and Roll-over-C, and wheel-lift derailment type was classified into Slip-up, Slip/Roll-over and Roll-over-L. To verify the theoretical equations derived for the wheel-climb derailment and the wheel-lift derailment, dynamic simulations using RecurDyn of Functionbay and Ls-Dyna of LSTC were performed and compared for some examples. The derailment predictions of the suggested theoretical formulation were in good agreement with those of the numerical simulations. The direction of the frictional force between the wheel-flange and the rail can be well predicted using the suggested derailment formulation at a initial derailment.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.37 no.2
• /
• pp.177-185
• /
• 2013

A new theoretical derailment coefficient model of wheel-climb derailment is proposed to consider the influence of wheel unloading. The derailment coefficient model is based on the theoretical derailment model of a wheelset that was developed to predict the derailment induced by train collisions. Presently, in domestic derailment regulations, a derailment coefficient of 0.8 is allowable using Nadal's formula, which is for a flange angle of $60^{\circ}$ and a friction coefficient of 0.3. However, theoretical studies focusing on different flange angles to justify the derailment coefficient of 0.8 have not been conducted. Therefore, this study theoretically explains a derailment coefficient of 0.8 using the proposed derailment coefficient model. Furthermore, wheel unloading of up to 50% is accepted without a clear basis. Accordingly, the correlation between a wheel unloading of 50% and a derailment coefficient of 0.8 is confirmed by using the proposed derailment coefficient model. Finally, the validity of the proposed derailment coefficient model is demonstrated through dynamic simulations.

• Journal of the Korean Society for Railway
• /
• v.17 no.1
• /
• pp.28-34
• /
• 2014

In this paper, a concept design of a rail type ultra-high-speed train is proposed and its dynamic characteristics are analyzed. Instead of the existing solid axle, a new type bogie system and independently rotating wheels are applied in the proposed train. In order to analyze the dynamic characteristics, a multibody dynamic model of a vehicle is developed and the basic validity is verified by eigenvalue analysis. Also, it is shown that the critical speed is improved in comparison to that of existing high-speed train model HEMU-430X. Finally, through 7000R curved track driving analysis at a speed of 550 km/h, the lateral force of the wheels and the derailment quotient are estimated and the applicability of the new concept railway vehicle is confirmed.

• 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.