【作者】 杨睿；

【导师】 朱彤；

【作者基本信息】 长安大学 ， 交通运输工程（专业学位）， 2022， 硕士

【摘要】 路侧停车作为城市公共停车系统的重要组成部分,与交通出行流动特性、城市用地功能分布及交通设施建设水平息息相关。研究建成环境、交通设施的空间分布对路侧停车泊位利用率的影响机理,能够为城市路侧停车规划、运营管理部门制定和调整发展策略提供理论依据,也是城市综合交通规划和静态交通规划中需着重考虑的问题。通过综述,发现以往研究对于路侧停车泊位利用率与城市路网形态、建成环境之间的关联机理研究尚有不足。为了解决以上问题,论文主要从两个方面研究时空因素对路侧停车泊位利用率的影响。首先,基于城市路侧停车泊位利用率的空间异质性分布特点,在对使用地理加权回归模型(GWR)应用与方法总结的基础上,引入考虑空间影响因素的影响尺度差异性的多尺度地理加权回归模型(MGWR),对路侧停车泊位利用率与建成环境变量的影响作用的空间异质性进行研究。最后在双层检验模式下,对最小二乘回归模型(OLS)和地理加权回归及其衍生模型(GWR、MGWR)结果进行参数对比和效果检验。其次,基于空间句法理论及s DNA工具对研究区域内路网结构进行量化分析,在结果中选取评价路网形态的路网接近度与路网穿行度指标,重新建立MGWR模型,分析路网形态变量对路侧停车泊位利用率的影响作用及空间分布特征。研究结果表明:(1)地理加权回归模型(GWR、MGWR)的回归优度明显强于普通多元线性回归模型(OLS),而地理加权回归及衍生模型中的MGWR回归效果检验参数更优,该模型针对考虑路网形态和不考虑路网形态这两种情况,回归结果中的最大Adjusted R2分别为81.4%和86.8%,且MGWR模型的AIC_c值、RSS值是三种模型中最小的,分别为335.95、47.561,说明MGWR模型无论在解释能力还是研究尺度差异性上的表现均优于其他两个模型。(2)回归结果表明,医院密度、企业密度、休闲绿地及景区密度对路侧停车泊位利用率具有显著的正向影响作用,值得注意的是,休闲绿地及景区密度作为本研究中路侧停车需求的显著影响变量在其他研究中尚未被揭示。站点车位数、地铁邻近度、土地利用混合度对路侧停车泊位利用率具有显著的负向影响作用;在工作日和节假日,建成环境变量的回归系数分布表现了明显差异。(3)路网接近度与路侧停车泊位利用率呈显著的正相关关系,而路网穿行度与路侧停车泊位利用率呈显著的负相关关系,且路网接近度对泊位利用率的影响程度大于路网穿行度的影响。更多还原

【Abstract】 As an important part of the urban public parking system,on-street parking is closely related to the characteristics of traffic flow,the functional distribution of urban land and the construction level of transportation facilities.Studying the influence mechanism of the built environment and the spatial distribution of traffic facilities on the utilization rate of on-street parking spaces can provide a theoretical basis for urban on-street parking planning and operation management departments to formulate and adjust development strategies issues considered.Through the review,it is found that the previous studies on the correlation mechanism between the utilization rate of on-street parking spaces and the urban road network form and built environment are still insufficient.In order to solve the above problems,the paper mainly studies the influence of space-time factors on the utilization of on-street parking spaces from two aspects.Firstly,based on the spatial heterogeneity distribution characteristics of urban roadside parking space utilization,on the basis of summarizing the application and method of using the geographically weighted regression model(GWR),a multi-scale geography that considers the influence scale differences of spatial influencing factors is introduced.A weighted regression model(MGWR)was used to study the spatial heterogeneity of the effects of on-street parking space utilization and built environment variables.Finally,the results of the least squares regression model(OLS),geographically weighted regression and its derivative models(GWR,MGWR)were compared and tested under the double-layer test mode.Secondly,based on the space syntax theory and s DNA tools,the road network structure in the study area is quantitatively analyzed,and the road network proximity and road network passability indicators for evaluating road network morphology are selected from the results,and the MGWR model is re-established to analyze the road network morphological variables influence and spatial distribution characteristics of on-street parking space utilization.The research results show that:(1)The regression goodness of geographically weighted regression models(GWR,MGWR)is significantly stronger than that of the ordinary multiple linear regression model(OLS),while the MGWR regression effect test parameters in the geographically weighted regression and derivative models are better,and the For the model considering the road network shape and not considering the road network shape,the maximum Adjusted R2in the regression results are 81.4%and 86.8%respectively,and the AIC_c value and RSS value of the MGWR model are the smallest among the three models,respectively.are335.95 and 47.561,indicating that the MGWR model outperforms the other two models in both explanatory power and research scale differences.(2)The regression results show that hospital density,enterprise density,leisure green space and scenic spot density have a significant positive effect on the utilization rate of on-street parking spaces.It is worth noting that the density of recreational green space and scenic spots is the demand for on-street parking in this study.The significant influencing variables have not been revealed in other studies.The number of parking spaces,subway proximity,and land use mix have a significant negative impact on the utilization of on-street parking spaces;on weekdays and holidays,the distribution of regression coefficients of built environment variables shows significant differences.(3)There is a significant positive correlation between the road network proximity and the utilization rate of roadside parking spaces,while the road network passing degree has a significant negative correlation with the utilization rate of roadside parking spaces.The degree of influence is greater than that of the traversal degree of the road network.更多还原