Abstract
In the current context of climate change and economic crisis, it is imperative to increase the energy efficiency of all human activities, and particularly those related to transport. Railways present a clear advantage in terms of efficiency when compared to other transport means, but it is still necessary to reduce their energy consumption so as to minimise their environmental impact and increase their competitiveness. Within this framework, an extensive monitoring campaign has been carried out in the metro and tram system in Valencia (Spain) aiming to measure the energy consumption under several conditions and scenarios. Measuring devices were placed in one metro unit (4300 series from Vossloh) and the energy consumed in the pantograph, the resistances and the auxiliary equipment was measured while the train operated normally. The data thus gathered was then processed and analysed in order to fully characterise the main trends of energy consumption and to identify possible improvements. The second part of the paper presents the development of a neural network capable of estimating the metro energy consumption. The network was trained using part of the data measured during the monitoring campaign, and the results obtained are shown and discussed. Once fully trained, the network will become a useful tool to further study the energy consumption of the metro system as it will allow testing different hypothetical operation scenarios and improving the overall energy efficiency of the system.
