A method is proposed to achieve optimal train operation while coping with real-time running performance variations of rolling stock. The chief objective of the method is to attain accurate punctuality and maximum energy efficiency in train operation. The performance perturbations in question are those confronted unpredictively in on-site operations, concerning train weight, tractive force, braking force, and running resistance. Two steps comprise the proposed method; (1) real-time system-identification of train dynamics from actual movements, and (2) real-time optimisation of target running profiles using revised dynamics data. The two steps cooperate to adapt optimal operation to changing environments by reflecting altered performance into the characteristics of target running profiles in real-time. Simulation results showed that the proposed method, compared with a method with no real-time adaptability, was capable of restoring punctuality even in case of deteriorated running performance, and simultaneously reducing energy consumption up to approximately 2
展开▼
