The Optimization And Control Research Of Spatial Weigthed Bus Network

With the development of city,the traffic congestion has become a big challenge.Bus public transport system is the main part of the city public transport.In recent years,with the breakthrough progress in the study of complex networks,more and more researchers apply complex network theory to public bus network,but few researchers consider of the time and space characteristics ofpublic bus network,and also very few network study is combined with the control theory of complex network.Taking consider of these problems,this paper applies complex network theory,and gives full consideration of its unique time and space properties,combines with statistical knowledge,to take optimization and control research of public bus network.This paper mainly completes the following several aspects:(1)Based on spatial point pattern analysis and spatial stratified sampling theory,this paper designs a spatial stratified sampling scheme based on grid,to seek a method for simplifying large-scale complex networks.The paper collects a total number of 3350 bus stops' latitude and longitude coordinates in Wuhan city.The spatial point pattern analysis shows that the spatial distribution of bus stops is significant aggregation pattern.Applying sampling method,106 sample points that have the same spatial distribution with the intact bus stops are extracted,and they can cover the whole Wuhan city.At the same time,comparing sampling points and all points,the distributions of the number of bus line via each bus stop are basically identical.It is show that the sampling plan is rational and valid because it selects representative sample points.(2)Based onmaximum simulation traffic,a bus network optimization model is established.In this paper,firstly Poisson process is introduced to simulate traffic between bus stops with the consideration of time and space characteristics,and the traffic is divided into three categories: the journey,the destination and waiting for in situ.The net traffic is defines as the sum of journey and destination minus waiting for in situ traffic.In order to guarantee the connectivity of the network,combining with the maximum spanning tree model,the bus network optimization model based on Poisson process is established.Based on sampling points,solving the optimization model,the result shows that compared with the actual bus network,optimization simulation networks have greatly increaseson traffic,which suggests that the optimization model in this article established has good reference value for actual bus network planning.(3)The strictly control theory and method of complex network is applied to actual bus network in Wuhan city and the optimization simulation networks.It is found that the number of minimum driving nodes set of actual bus network is one.While the optimization simulation networks are all about 68 with the number of connect edges changing.It shows that the control ability of optimization simulation networks is weaker than the actual bus network.The driving nodes of optimization simulation networks are mixed distribution in all nodes,which implies the driving nodes can control near nodes but having little impact of far nodes.Finally through the analysis of degree and node weight,it is found that the linear correlation of degree and node weight may lead to the same drive node set.