摘要
城市空间联系强度是测定区域经济活跃程度和一体化进程的重要指标。综合运用城市流模型和SOM神经网络分类器,将我国地级以上城市作为切入点,将中国大陆31个省级行政区作为基本区域单位,通过计算2008年的城市流强度,考察不同区域,特别是省际城市空间联系的差异性,并对城市空间联系水平进行分类。结果显示,我国大陆形成了沿海三大城市空间联系辐射中心,并呈连续的面状分布。不同省级区域间的城市空间联系水平和对外服务功能存在巨大差异,地带性和区域性空间分异明显。经济发展水平较低的省级区域,其城市空间联系的集中化指数存在偏高或偏低现象。根据城市流计算结果,构建了城市空间联系的指标体系,将我国城市空间联系水平分为非常密切、较密切、较差和差4类。
The intensity of urban spatial relation is a main index for measuring activity and integration of regional economies. Comprehensive using urban-flow model and SOM neural network, by using cities at prefecture level as breakthrough point, taking provinces as fundamental unit, explore regional variation of urban spatial relationship and classify it by urban-flow indexes. The results show that, there are three major radiation center of urban spatial relation in mainland of China. The urban spatial relation is different in different provincial region. The urbanflow centralization index of under-developed provinces is too high or too low. According to calculation results of urban flow, constructing index system of urban spatial relation, classifing provinces into four groups, namely very close, close, far, further.
出处
《地域研究与开发》
CSSCI
北大核心
2012年第1期40-44,共5页
Areal Research and Development
基金
国家自然科学基金项目(41101114
41071094)
