Speed optimal-setting of electric multiple units based on energy-efficient operation

The operation process of electric multiple units (EMUs) is characterized by complex running environment, and frequent change of operating conditions. It is difficult to accomplish energy-efficient operation using the given V-S curve (velocity versus position curve). And the alternation of coasting points will have different running time and energy consumption, then appropriate coasting scheme is an effective energy-saving strategy. Based on the traction energy consumption and running time model, we exploit a data-driven RBF (radial basis function) neural network to build a model that connects the coasting points and their corresponding energy consumption and running time. Finally, we adopt GA (genetic algorithms) to optimize the outputs of the RBF, and the speed optimal-setting profile satisfying punctual, energy-efficient operation is obtained. Simulation results show the effectiveness of this method.