首页|基于随机森林模型的合肥市主城区轨道交通站域空间活力与建成环境关联性测度

基于随机森林模型的合肥市主城区轨道交通站域空间活力与建成环境关联性测度

扫码查看
以合肥市主城区轨道交通站域为例,基于百度热力指数等多源数据分析其空间活力特征,采用随机森林模型测度地理区位、土地利用、功能设施、开发强度、交通便利度以及环境品质6个维度的建成环境要素与轨道交通站域空间活力的关联性.同时,结合K-means聚类算法,划分轨道交通站域类型并识别典型站域存在问题,从而针对不同类型轨道交通站域提出空间活力优化策略.结果表明:①合肥市主城区轨道交通站域空间活力在时序上呈现"晚高峰高于早高峰,休息日晚高峰高于工作日晚高峰"特征,在空间上呈现"内高外低"由二环线向外逐渐递减的空间分异格局;②综合工作日与休息日早/晚/非高峰时段建成环境变量对轨道交通站域空间活力的相对重要性排序,由大到小依次为:功能设施>开发强度>地理区位>土地利用>交通便利度>环境品质;③合肥市主城区轨道交通站域可划分为成熟型、成长型、初熟型、孕育型4类,聚类特征明显,在空间上呈现"圈层结构".其中,成熟型站域集聚分布在二环线内,成长型站域沿二环线呈环状分布,初熟型站域主要呈零散状分布在二环线外滨湖区与经开区,孕育型站域主要分布在二环线外城市各区,分布范围较广.研究旨在从建成环境视角为轨道交通站域空间活力提升与低碳交通策略提供政策启示.
Measuring the correlation between spatial vitality of metro station domain and built up environment in the main urban area of Hefei based on random forest model
Taking the metro station domains in the main urban area of Hefei as an example,this study ana-lyzed its spatial vitality characteristics based on multi-source data such as the Baidu heat index.A random forest model was used to measure the correlation of six built environment elements,including geographic loca-tion,land use,functional facilities,development intensity,accessibility,and environmental quality,with the spatial vitality of metro station domains.Furthermore,combined with the K-means clustering algorithm,this study classified the types of metro station domains and identified the deficiencies of typical station domains,and then proposed spatial vitality optimization strategies for different types of metro station domains in a tar-geted manner.The results show that:1)The spatial vitality of the metro station domains in the main urban area of Hefei was higher in the evening peak than in the morning peak,and higher in the evening peak on weekdays than in the evening peak on weekends.Spatial vitality also showed a differentiation pattern of"high inside and low outside",with vitality values decreasing from the second ring road to the outside;2)The relative import-ance of each built environment variable on the spatial vitality of metro station domains during the morning,evening,and off-peak hours on weekdays and weekends was ranked in descending order:functional facilities>development intensity>geographic location>land use>accessibility>environmental quality;3)The metro station domains in the main urban area of Hefei were divided into 4 types:mature,growing,low matur-ity and breeding,showing the significant clustering characteristic of"circle structure".The mature-type station domains were clustered within the second ring road,the growing-type station domains were distributed in a cir-cular pattern along the second ring road,the low maturity-type station domains were mainly scattered outside the second ring road in the Binhu District and the Economic Development District,and the breeding-type sta-tion domains were mainly distributed outside the second ring road in various districts of the city with a wide range.The study aims to provide policy insights for the enhancement of spatial vitality of metro station do-mains and low-carbon transportation strategies from a built environment perspective.

metro station domainspatial vitalitybuilt up environmentBaidu heat indexrandom forest model

方云皓、赵丽元、顾康康、袁建峰

展开 >

华中科技大学建筑与城市规划学院,湖北武汉 430074

湖北省城镇化工程技术研究中心,湖北武汉 430074

安徽建筑大学建筑与规划学院,安徽合肥 230601

武汉市规划研究院,湖北武汉 430014

展开 >

轨道交通站域 空间活力 建成环境 百度热力指数 随机森林模型

国家自然科学基金中央高校基本科研业务费专项

523780562023WKZDJC009

2024

地理科学
中国科学院 东北地理与农业生态研究所

地理科学

CSTPCDCSSCICHSSCD北大核心
影响因子:3.117
ISSN:1000-0690
年,卷(期):2024.44(5)
  • 13