Optimization of the velocity profile for the economic train operation

초록

As the concerns to the environment and the fuel efficiency are raised, the interest about the economical train operation has been increased. Since the train generally moves by following the pre-defined path, it is possible to derive the optimal operation strategy that is the most economical if it is assumed that the precise track and train information are provided. The operation strategy contains how to accelerate the train to the right velocity and then how much distance to run by coasting as well as from which point to start braking system in order to minimize the energy consumption. The operation strategy is represented as the velocity profile. The velocity limits which are defined by the each segment on the path and the track data information that contains the gradient and the curvature should be considered in order to derive the economic velocity profile. However, these informations represent as discontinuous function as the displacement, and therefore, it is difficult to derive the close form optimal solution. Moreover, the traction and the braking characteristics as well as the detailed train model should be considered in order to yield the energy to accelerate or to decelerate the train. However, it is difficult to take these characteristics into consideration because these are non-linear and sometimes represent discrete values as Fig. 1. In this study, the traction motor and the braking system models as well as train model are established according to the measurement data and the optimal velocity profile is derived when a train moves from a station to the next station. Genetic algorithms are employed and the energy consumption result is compared to that by the conventional velocity profile.