Skip to Main Content
With the increasing energy consumption in Chinese mainline railways amid the worldwide carbon emission concerns, the need for energy-efficient locomotive operation becomes urgent. Locomotive operation is directly linked to speed limits imposed by the train ahead through signaling. In China's mainline railways, speed limits for locomotive operation change frequently because of relatively short headways in a highly congested network. Whenever the speed limit changes, the locomotive operation must be determined again quickly to adapt to the new speed limit. As a result, the energy-efficient locomotive operation is a real-time optimization problem with time-varying constraints, in which the tradeoff between solution optimality and computational time is essential, but it has not been considered adequately in previous studies. This study develops a fuzzy predictive control approach, continuously providing locomotive operation instructions, with respect to the prevailing speed limits, to reduce energy consumption of train movement. The proposed approach is implemented in an onboard decision support system to assist drivers. The system is tested on the Ning'xi line in China. The results indicate that energy consumption on train operations is reduced by 4%, without increasing the runtime between stations, while the computational requirement satisfies the demand of real-time solutions. Extensive simulations show that the proposed approach is able to provide sufficient solution optimality in reasonable computational time under different operation settings.