Train operation control in urban railways is challenging due to its high dynamics, complex environment, and level of comfort and safety. To address these challenges, authors propose a new Deep Reinforcement-based Train Operation (DRTO) method which includes: i) a deterministic deep reinforcement learning algorithm, ii) a dynamic incentive system, which is used to ensure safe operation in a multi-train environment, and iii) an event-driven method, which is used to improve the DRTO performance based on an event-driven strategy. To evaluate the performance, authors thoroughly compare the proposed method with other operation control solutions on both synthetic and real datasets. Our results demonstrate that DRTO is effective in: i) decreasing the energy consumption of train operation, ii) increasing passenger comfort and iii) achieving a good trade-off between efficiency and safety. In addition, the effectiveness of the event-driven strategy and the dynamic incentive system is also demonstrated in the experiments.