An optimal automatic train operation (ATO) control using genetic algorithms (GA)

This paper shows the form of the optimal solution and how to minimize energy of the train driving control that can be included into automatic train operation (ATO) systems. We consider the case where a train is to be driven by automatic operation mode along a nonconstant gradient curve and with speed limits. Using the genetic algorithms (GA), we constructed an optimal train driving strategy. The results are compared with P. Howlett's optimization method using the constrained optimal technique (Lagrange function and Kuhn-Tucker equations) in view of energy cost benefit. For the case studies, we used a railway track of Seoul City MRT system. As a result of the test, we verified that the proposed algorithm could be of effective energy cost benefit.