| Popis: |
The on-board collection of data related to train operation enables a better calibration of the train resistances, which are fundamental for the elaboration of optimized train control solutions. Here, the possibility to implement a calibration of train resistances models based on power measurements is investigated. For this purpose, a huge dataset of train runs data collected by a Swiss train operator has been analysed. The train runs of three train types, operating on three different lines, have been extracted and used for the calibration phase. The calibration model is formulated as an optimization problem for parameters fitting and a Global Optimization Multi Start approach is used for finding the suboptimal solution. The performances of the model are therefore discussed together with possible further investigation. |
