Towards adaptive bogie design

Licentiate thesis, 2014

Suspension components play a crucial role in bogie dynamics behavior. In this regard, passive and active systems are developed to meet various design requirements. Adaptive suspension systems can adjust the bogie dynamics with respect to different operational scenarios and as a result improve the vehicle performance. Therefore, there is a great demand on design of adaptive suspension systems for high speed train bogies. Sensitivity analysis, optimization, and active vibration control techniques are the most important steps towards adaptive bogie design. In this thesis, the target is to cover these steps partly by formulating and solving the prescribed problems for some example railway vehicle models. Therefore, hierarchical levels of vehicle modelling are considered. The overall performance of a vehicle can be evaluated from different perspectives. In this thesis, the dynamics behaviour of railway vehicles is reflected by ride comfort, wear, safety, and in particular running stability, track shift force, and risk of derailment objective functions. The mathematical representation of the prescribed objectives as well as the evaluation procedure are described thoroughly. As an example on optimization problem with application in railway vehicles, the comfort/safety multiobjective optimization of a one car vehicle lateral dampers is considered. The genetic algorithm routine is used to solve this problem. In order to have a better wear estimation, a theoretical contact search approach is applied to calculate the creepages and wear. The optimization problem of primary dampers towards wear showed that one might achieve better wear performance by using active technology in bogie primary suspension components. Therefore, different on/off semi-active control strategies are integrated together with the magnetorheological dampers in bogie primary suspension and the corresponding effects on wear is explored on different operational scenarios. Finally, to have a better insight into adaptive bogie design, the global sensitivity analysis of bogie dynamics behavior with respect to suspension components is considered. The multiplicative version of the dimension reduction method is employed to provide the sensitivity indices. The result of such analysis can narrow down the number of input variables for the optimization and adaptive bogie design problems and improve the computational efficiency. All in all, this thesis deals with formulating and solving some example problems on the sensitivity analysis and optimization of railway vehicles. Furthermore, as an introduction to active bogie systems, the application of semi-active vibration control strategies in bogie primary suspension is also considered. The results of the current thesis can provide useful hints in design of adaptive suspension system for high speed train bogies.