This thesis aims at systematically studying the possibilities of minimising devastatingconsequences of high-speed derailments by appropriate measures and features in thetrain design, including the running gear. The course of events immediately afterderailments is studied with respect to whether the train stays upright and close to thetrack centre line or deviates laterally with probably serious consequences. There is abelief in the railway community that some trains can better cope with derailment thenothers, although this superiority is apparently hard to quantify.Firstly, an empirical database has been established containing as much relevantinformation as possible of past incidents and accidents occurred at higher speeds due tomechanical failure close to the interface between the running gear and the track, as wellas other causes that ultimately brought the train into a derailed condition. Although nevertwo derailments are the same, certain patterns appeared to crystallise after analysing thecourse of events immediately after the failure based on the descriptions available in eachincident or accident report. Ultimately, this led to that several critical vehicle parameterscould be distinguished as capable to influence the outcome of a derailment.Secondly, two of the critical vehicle features found in the first stage have been subject todetailed analysis by means of multi-body system (MBS) simulations. The first phase ofthe computer simulation program focused on studying the tendency of a wheelset toderail as a result of an axle journal failure on the outside of the wheel. The prederailmentcomputer simulation model has been validated with good results for twoauthentic Swedish events of axle journal failure.Thereafter, one of the newly found critical vehicle feature, the wheelset mechanicalrestrictions relative to the bogie frame, have been extensively studied on an X 2000power unit and trailer car model. The results show that a vertical mechanical restrictionof the wheelset relative to the bogie frame of approximately 50 to 60 mm is capable ofkeeping the wheelsets on the rails after an axle journal failure, for the studied conditions.An axle mounted brake disc constitutes the second critical vehicle feature that has thepotential to favourably influence the sequence of events in cases of wheel flangeclimbing. A minimal range of geometrical parameters for which the rail would safely fillthe gap between the brake disc and the wheel has been calculated.The third and last part of the thesis establishes the prerequisites necessary in order tostudy the remaining of the critical vehicle parameters found in the first part, whichrequires complete MBS simulations of derailed vehicles rolling on track structures, i.e.concrete sleepers. To accomplish this task, hysteresis data for the force as function ofconcrete material indentation, are aimed to be acquired by means of finite element (FE)simulations. Therefore, the intended FE model of wheel-concrete sleeper impact issubjected to a tentative validation procedure. A good agreement is observed whencomparing the FE model results with an authentic accident in terms of concrete sleeperindentation. Furthermore, preliminary results in terms of a wheelset tendency to reboundafter concrete sleeper impact are presented. / QC 20101125
| Identifer | oai:union.ndltd.org:UPSALLA1/oai:DiVA.org:kth-242 |
| Date | January 2005 |
| Creators | Brabie, Dan |
| Publisher | KTH, Farkost och flyg, Stockholm : KTH |
| Source Sets | DiVA Archive at Upsalla University |
| Language | English |
| Detected Language | English |
| Type | Licentiate thesis, monograph, info:eu-repo/semantics/masterThesis, text |
| Format | application/pdf |
| Rights | info:eu-repo/semantics/openAccess |
| Relation | Trita-AVE, 1651-7660 ; 2005:17 |
Page generated in 0.0025 seconds