摘要
城市土地的集约利用直接关系到区域土地的可持续利用,如何衡量土地的集约利用程度是一个复杂而重要的过程。以广州市1990—2007年建设用地数据、各类社会经济数据为基础,构建适合广州市建设用地集约利用评价的指标体系,综合运用主成分分析和BP神经网络建立建设用地集约利用评价模型,计算其集约利用程度,并与前人通过模糊综合评价法得出的结论进行比较。结果表明:1990—2007年广州市整体集约利用水平呈上升趋势,1990—1994年、1996—1997年处于粗放利用,1995年、1998—2003年处于中度集约,2004—2007年处于高度集约。主成分分析和BP神经网络的评价方法与模糊综合评价法的结果平均误差为4.37%。
The degree of urban intensive land utilization is directly related to the problem of the sustainable use in regional land, how to measure the degree of intensive land utilization is a complex and important process. Based on 1990--2007 Guangzhou City construction land data and various socio-economic data, an evaluation index system for intensive use of construction land was established. Integration of principal component analysis and BP neural network evaluation model were used to calculate land intensity, and compare with the re suit from former through fuzzy comprehensive evaluation. The results showed that the overall intensity of land use in Guangzhou City had progressively increased, the level of intensive use was at a primary stage during period from 1990 to 1994 and 1996 to 1997, in 1995 and 1998--2003 at a moderate intensive stage, in 2004 2007 at a highly intensive stage. The average error between principal component analysis, BP neural network evaluation methods and fuzzy comprehensive evaluation method is 4.37%.
出处
《水土保持研究》
CSCD
北大核心
2014年第2期162-166,共5页
Research of Soil and Water Conservation
基金
广东省教育部产学研结合项目(2012B091100054)
关键词
主成分分析
BP神经网络
建设用地
土地集约利用
广州市
principal component analysis（PCA）
BP neural network
construct land
intensive land utilization
Guangzhou City
