Subway connection line generation method based on Q-learning algorithm

• Main Authors: SUN HAO, QU HUA, PU XIUXIA, LI JUNHUI, DENG JIERONG, WANG HONGGANG, GUO JIANGUO, ZHANG YANGCUN, ZHENG DONGDONG
• Format: Patent
• Language: Chinese; English
• Published: 09.08.2022
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; DATA PROCESSING SYSTEMS OR METHODS, SPECIALLY ADAPTED FORADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORYOR FORECASTING PURPOSES; ELECTRIC DIGITAL DATA PROCESSING; PHYSICS; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE,COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTINGPURPOSES, NOT OTHERWISE PROVIDED FOR

The invention discloses a subway connection line generation method based on a Q-learning algorithm, and the method comprises the steps: carrying out the digital expression of actual passing roads, forming an abstract road network, abstracting a road network into nodes and edges according to a graph structure in a data structure, enabling the nodes to be the intersection points of the roads, enabling the edges to be the roads with passability, and enabling the nodes to be the intersection points of the roads and the edges to be the roads with passability; then aggregated OD data is formed after node aggregation of the abstract road network, then on the basis of completion of real road network abstraction, planning of a connection line is performed, travel OD data is taken as a reference, the maximum number of accumulated passengers of the connection line is taken as an optimization target, a Q-learning algorithm is used to optimize a road passing weight, and finally generation of the connection line is complet