Autonomous scheduling method and system for train

Abstract

The invention provides an autonomous scheduling method and system for a train. A simulation module receives rail transit data and simulates an actual rail transit system; the simulation module and a deep reinforcement learning module perform interactive training; the deep reinforcement learning module obtains a trained scheduling decision model; the deep reinforcement learning module transmits thetrained scheduling decision model to a scheduling scheme module; the simulation module simulates a current train running state and outputs the current train running state to the scheduling scheme module; the scheduling scheme module generates a scheduling scheme based on the current train running state; and the scheduling scheme module transmits the scheduling scheme to the actual rail transit system. The actual rail transit system is simulated through the simulation module; the scheduling model is trained through the deep reinforcement learning module; the deep reinforcement learning moduletrains the scheduling model, so that each train adjusts the operation control strategy of the train according to an operation environment, and therefore, train operation energy consumption and passenger waiting time are reduced on the premise that train operation safety and punctuality are ensured; and the real-time performance and flexibility of scheduling are relatively high.

Description

technical field [0001] The invention belongs to the field of rail transit, and in particular relates to a method and system for autonomous train dispatching. Background technique [0002] Under the existing transportation organization model, the transportation plan is usually compiled based on the passenger flow demand forecast of the stage. In a short period of time, due to the fluctuation of real-time passenger flow demand, there is a certain mismatch between transportation supply and transportation demand, which leads to a decrease in the service level of the transportation system. At the same time, due to the influence of various external factors during the operation of the train, the train operation gradually deviates from the operation schedule and the established energy-saving control curve, and it is difficult to guarantee the punctuality and energy-saving performance of the train operation. [0003] The existing research on train operation scheduling is mainly cent...

AI Technical Summary

Problems solved by technology

[0004] However, in the process of train operation, the surrounding environment and transportation requirements that need to be considered are extremely complex and present strong dynamic time-varying characteristics, and centralized scheduling methods are often difficult to adapt to application scenarios that require high real-time and flexibility

Under the nonlinear and real-time requirements of the dispatching tasks of the rail transit system, the dispatching organization method based on optimization not only has the problem of too long solution time, but also its flexibility is restricted in the dynamic dispatching scenario

At the same time, most simulation-based methods only propose policy support or simple policy generation logic for train operation scheduling scheme generation, which is difficult to apply to train operation scheduling in complex scenarios

Images