Public transport passenger flow spatial and temporal distribution simulation method and simulation system based on individual activity chain

Abstract

The invention discloses a public transport passenger flow spatial and temporal distribution simulation method and simulation system based on an individual activity chain, and belongs to the field of transportation characteristic simulation. The method comprises the following steps that firstly, a database related to resident travelling conditions, traffic zones and urban road networks is built; temporal-spatial information of an activity and traveling chain of a resident in the workday is simulated based on the reinforcement learning theory; sampling expansion is carried out according to the sampling rate of different zone samples, the activity and traveling chains of all urban residents are obtained, and whole urban public transport passenger flow spatial and temporal distribution is extracted from the activity and traveling chains. Based on individual activity needs, the activity and traveling arrangement behaviors of the residents are simulated through a reinforcement learning algorithm, spatial and temporal distribution information of urban traffic is obtained through simulation, and the final spatial and temporal distribution information of public transport passenger flow is extracted on this basis.

Description

technical field [0001] The invention relates to a simulation method and system capable of simultaneously obtaining the time-space characteristics information of individual travelers' bus travel, and belongs to the field of traffic travel characteristic simulation. Background technique [0002] With the continuous increase of car ownership, more and more urban residents choose cars to travel, resulting in increasingly serious urban road traffic congestion. Compared with new high-quality roads, reasonable control of traffic demand and priority development of public transportation can solve the congestion problem more scientifically and effectively. The spatio-temporal distribution of bus passenger flow is the basic information for the research of bus priority policy, and it is of great significance to establish an accurate and widely applicable simulation system for the spatio-temporal distribution of bus passenger flow. Activity participation is the root of traffic demand, s...

AI Technical Summary

Problems solved by technology

[0004] Existing traffic prediction models based on individual activity chains mainly simulate individual travel behaviors from the perspectives of individual activity start time, duration, and travel mode, but the problem is that they can only determine the time dimension or space dimension of travel. Finally, the micro-simulation of traffic travel in another dimension cannot be simulated at the same time for the travel distribution of time and space

However, the spatiotemporal distribution of traffic travel often affects each other, and different travel locations take different travel times, so there are often large differences between the simulation results of previous models and the actual traffic conditions

Images