Model predictive control of battery-powered trains for catenary-free operation

Rail transportation is recently making a comeback with stunning results from technological viewpoint. While the efficiency of trains is well known, many aspects related to energy management still need to be tackled, including sustainability and optimization issues. These issues are central to the control community, and in this context model predictive control (MPC) is a powerful control approach for its capability of guaranteeing constraints satisfaction, while minimizing a predefined cost function. In this article, we exploit these advantages to provide more efficient control of the electric equipment inside railway vehicles. More precisely, the proposed MPC approach is capable of addressing the challenging scenario of partially catenary-free tracks for trains. These are equipped with batteries, which have to supply traction motors and parallel-connected auxiliary loads in an efficient manner, while reducing the amount of losses over the electric lines. Simulation results, based on real data provided by the industrial partner Alstom rail transport, show the effectiveness of the proposal.