ABSTRACT The modelling method of the air spring and its additional pneumatic system has been studied more frequently in recent years. As for the railway field, the study of the secondary suspension system of the bogie is the most popular, while little focus is on the air spring of the pantograph. In this paper, the modelling method of the air spring on the pantograph is investigated and developed. Four thermodynamic models are derived step by step, which can roughly be separated into two types. The first two models are established to express the relationship between the inner pressure and the vertical displacement of the air spring, and the effect of a regulator is illustrated and discussed. The latter two models that are developed from the first two models are set up to express the relationship between the vertical force and the vertical displacement of the air spring, where the effect of air spring material is illustrated and discussed. All four models are validated by comparing the simulation results with experimental results for frequency characteristics. Finally, the characteristics of the four proposed models are illustrated and discussed.

Steering torque constitutes the primary motorcycle control input for the lateral dynamics; consequently, estimating it is important. Conventionally, this is done with complete motorcycle models, requiring significant identification effort. The simplified models in the literature only describe the steering torque under specific cases. This work defined a steering assembly model with few parameters to estimate the steering torque analytically for stationary and transient manoeuvres. The model equations followed from existing motorcycle models through simplifying hypotheses; transfer functions describing the roll response and the Lane Change Roll Index (LCRI) were obtained from these equations. Measured steering torque signals from different datasets, including diverse motorcycle classes, were used as the reference for validation. A good agreement resulted between the estimated and measured torques, in the time and speed-acceleration domains and in terms of LCRI. When using the roll as the motorcycle response, manoeuvrability was highest at lower frequencies. The scooter was the most manoeuvrable; the sports and touring motorcycles were the least manoeuvrable at low and high frequencies, respectively. Concerning design parameters, the front-wheel spin inertia and front twist stiffness influenced manoeuvrability the most. The model allows recreating the steering torque signal for new and pre-existing datasets using commonly measured signals; the signal can describe the riding style and the effort required. Few parameters are required, facilitating its use and reducing the computational burden, allowing its use for steering assistance systems.

The rail pad damping is a critical parameter to evaluate its vibration reduction capacity, which has essential research significance. The mesh–type high damping rail pad (MTHDRP) is proposed for the first time in this paper, and the differences in damping performance between MTHDRP and traditional rail pads (PTRP and GRP) are compared and studied. It is concluded that the MTHDRP has the highest damping coefficient than any other rail pads. The dynamics model calculation has discovered that the increase of the rail pad damping coefficient can not only effectively decrease the vibration acceleration, wheel–rail force, and derailment coefficient, but also is beneficial for both vehicle safety and stability and track structure protection. The field test shows that the acceleration of the rail, clip, iron pad, and track bed in the MTHDRP laying section is significantly lower than that in the PTRP laying section, and the rail surface roughness improves significantly after the rail pad is replaced with the MTHDRP for half a year.

The paper shows the feasibility of building closed-form and fast-to-evaluate surrogate models via supervised kernel-based machine learning regressions behaving as digital twins for computationally expensive multibody simulations. The aforementioned surrogate models are adopted to predict the railway vehicle dynamics safety indexes defined in the international standards, depending on the wheel-rail forces, directly from the results of longitudinal train dynamics simulations. The digital twin models are trained with the outputs of Simpack multibody simulations of a reference freight wagon, to which the in-train forces calculated by an in-house MATLAB longitudinal train dynamics simulator are applied. Two machine learning techniques are considered: the support vector machine and the least-squares support vector machine regressions. Both techniques ensure a good accuracy even with a limited number of training samples. The derivation of the surrogate models can strongly enhance the modelling capabilities of common longitudinal train dynamics simulators, that cannot evaluate the wheel-rail contact forces. At the same time, the method shown in the paper allows to strongly reduce the total computational times, as the evaluation of the closed-form surrogate models allows to effectively estimate the safety indexes with no need for computationally demanding multibody simulations.

This work describes the development of a vehicle state estimator, designed to be employed onboard in real-time, based on the data acquisitions already available within the vehicle CAN bus infrastructure. The proposed estimation algorithm, taking into account the suspension elasto-kinematics, the longitudinal, lateral and combined dynamics of the tyres as well as the aligning interaction torques at the ground, represents at the current stage a significantly complete tool, since it also provides an estimation of road slope and banking angles, generalising the adoption of the presented vehicle model even on highly inclined roads. A further advantage of the presented estimator consists of its capability to identify in real-time and with a good level of accuracy the tyre/road interaction physical characteristics in terms of grip and cornering stiffness.

Wheel polygonization intensifies the dynamic interaction between wheel and rail and reduces the service life of the locomotive and the rail system. The dynamic performance of the drive and transmission system (DATS) directly affects the wheel-rail interaction and further affects the wheel polygonization. In this study, a rigid-flexible locomotive coupled dynamics model of an HX locomotive is established. The DATS and the flexible characteristics of the wheelset are both considered. The calculation model of the wheel polygonal wear is established based on the local tangential contact distribution algorithm FaStrip and the USFD wear function. Effects of the DATS on the wheel-rail dynamic response and wheel polygonal wear are investigated. Evolution of the wear depth is studied under the different running speeds and the wheel profile initial states. The results indicate that the torsional vibration response of the DATS plays a significant role in promoting the development of the wheel polygonization. The more serious wear damage of the wheelset on the non-gear side (NGS) is both reflected in the actual service process and the numerically calculated results, which validates the established dynamics model and calculation method. This study could provide helpful reference for further understanding on the wheel polygonization phenomenon.

Pantograph-catenary (P-C) systems operating under cracks would induce potential risks to railways. Previous relevant research efforts lack consideration of the dynamic characteristics of P-C systems under crack damages. A rigid-flexible P-C coupling model with pantograph cracks for a metro line is developed in this study. The evolution law of P-C contact forces under the condition of cracked propagation is described, and the influence of speed, uplift force, span length, and pullout value arrangement on P-C contact force and the sensitivity of crack influence are explored. Results show that the dispersion of contact force values decreases continuously upon the existence and expansion of cracks on the pantograph. Proper selection of operating conditions is advised because of the sensitivity differences of the five influencing factors indicates. Metro trains should not be operated at 100 km/h for the rail section with an 8 m-span catenary, and further inspections on the pantograph should be scheduled for trains operating over 100 km/h. The uplift force should be adjusted to the 90-100 N range. The 8 m span length is suggested in a metro catenary, and the splayed arrangement of pullout value is recommended as a priority when the situation allows.

This paper examines the dynamic performance of the pantograph passing through an overlap section of the catenary system, using a China’s high speed railway with the 4-span overlap as an example. The pantograph and catenary system are simulated as a lumped mass model and finite element model, respectively. The model is verified according to EN50318:2018. The model is also assessed against an experimental study of the contact force and uplift of contact wire at 300 km/h. Then the contact force and uplift of contact wire at the speed of 150–350 km/h are calculated. The research shows that when the pantograph passes through the overlap, the maximum contact force increases beyond 14%, the standard deviation of the contact force increases by more than 26%, and the increase of the vertical displacement exceeds 38%, which becomes more apparent with the increase of speed. Under the influence of aerodynamic force, when the running speed increases from 150 to 350 km/h, the mean contact force in the knuckle-downstream and knuckle-upstream orientation increase by 59.49% and 97.85%, respectively. As a result, the contact force and vertical displacement increase greatly when passing through the overlap section, which could affect the current collection quality.

Couples of heavy-haul freight cars are simulated, using a 62-degree-of-freedom multi-body model developed on Simpack®, running on both tangent track and circular curve under different conditions. A benchmark simulation running at a speed of 70 km/h is set up, and then, different breaking scenarios with 2.5 kN, 5 kN, 7.5 kN and 10 kN of abnormal braking force application are simulated. In this paper, the effect of forces acting on the wheel–rail interface is compared using the derailment coefficient (L/V) and the material wear, defined by energy dissipation (Tγ). For the double-wagon setup placed in various positions along the train, the simulations run under no braking and abnormal braking conditions, where the braking force acts in only one of the wheels of the wheelset. This condition leads to an increase of up to 18% in L/V in one curve and up to 17% of reduction on the curve in the other direction. The wear increases 68% and decreases 71%, depending on the direction of the curve. These effects haven't been associated with longitudinal dynamic effects in previous studies, including in-car shocks and traction forces.