The Global Sensitivity Analysis On Ride Quality Of Railway Vehicle With Stochastic Parameters

With its advantages, such as speediness, convenience, economic, environmentalprotection, high safety etc, railway transportation has become an important mean oftransport in modern society. At present, rail transit enterprises are developed actively overcountries. The construction of high-speed railway and urban rail transit is also speeding upin our country. However, when an overloaded train runs in high speed, wheel/railinteraction forces will be aggravated, the ride quality of vehicle declines. Therefore, it isnecessary to study the ride quality of high-speed vehicle in deepth, and the analysis ofvehicle system parameters has an important reference value to improve its ride quality.This article’s main research content and work are as follows:(1) A vertical multi-rigid-body dynamics model of railway vehicle based ongeneralized Ruzicka vibration isolation system is established. The complex modal analysismethod is adopted to obtain the modal parameters and inherent vibration modes of vehiclesystem. Under the action of random track irregularities, the random vibration responsespectrum of vehicle system and vertical Sperling index are solved quickly by thepseudo-excitation method. Contrast to the classic vehicle vertical model with10degrees offreedom, how the change of rubber stiffness at damper ends affects the vibration responsebehavior and ride quality indexs is analyzed.(2) A Part of vehicle system parameters’ uncertainties are considered as Gaussianmodels. Stochastic simulation method is used to obtain of the means, standard deviationsand variation coefficients of Sperling index under the situations of single parameter andmulti parameter variation. Then, the probability value scope of Sperling index is indicated.(3) Based on the stochastic parameter structure of railway vehicle, Extended FourierAmplitude Sensitivity Test method is used to obtain the global sensitivities of stochasticsuspension parameters on the Sperling index. Then, the effect of Extended FourierAmplitude Sensitivity Test is discussed with the correlation coefficients between the samples of Sperling index and suspension parameters.In this paper, the research work provides valuable references for reliability analysisand robust design of railway vehicle dynamics performance.