Study on Deep Reinforcement Learning Model for Railway Vertical Alignment Design
Traditional intelligent algorithms require a fixed number of variables to remain unchanged during the calcula-tion process,while the number of slope-change points in the intelligent design of railway vertical alignment needs to be adaptively determined according to changes in terrain.Considering the characteristics of reinforcement learning being able to learn and interact with environmental data such as ground elevations and generated alignments to obtain the opti-mal strategies,in this paper,the method of deep reinforcement learning was applied to the intelligent design of the verti-cal alignments,and the method for the intelligent agent to decide the slope-change points in sequence from front to back was studied.A grade change point decision-making model was proposed for railway vertical alignment design to determine the expression forms of states,actions and rewards in the model.At the same time,combined with the char-acteristics of many design constraints in the vertical alignment design,an action masking mechanism was introduced to deal with the constraints,accelerate the convergence and improve the performance of the model.In addition,by intro-ducing the computation period into the state of the model,a single-network multi-strategy multi-objective processing method was proposed to generate multiple multi-objective strategies through a single network.The correctness and ef-fectiveness of the models for single-objective and multi-objective profile problems were verified through practical engi-neering cases.
NETL
NSTL
万方数据
