Automatic train regulation with energy saving using dual heuristic programming

Abstract

Issues regarding environment sustainability and energy saving have been receiving much concern in worldwide railway society. Energy saving via train operation and regulation is a cost-effective way and has become an issue to be considered while performing train operation and regulation. Automatic train regulation (ATR) plays an important role in maintaining metro service quality; however, designing ATR is an optimisation problem with large scale, high non-linearity, heavy constraints and stochastic characteristics. Considering issues regarding energy saving in the ATR design would further complicate the optimisation problem. A model describing traffic environment associated with energy consumption was investigated and developed; and then the dual heuristic programming was employed for designing ATR with energy saving via coasting and station dwell control. A simulation test of the ATR design was conducted with field data. The simulation shows better traffic regulation with less energy consumption is attainable with the ATR design.