基于GA-BP神经网络的郑州市土地集约利用时空差异分析  

The Evaluation of Land Use Intensity in Zhengzhou City Based on GA-BP Neural Network

作　　者：叶菁[1] 谢巧巧 胡君[1] 

机构地区：[1]中国地质大学(武汉)土地资源管理系,武汉430074

出　　处：《湖北农业科学》2015年第5期1270-1276,共7页Hubei Agricultural Sciences

基　　金：国家社科基金项目(14BJY057);中央高校基本科研业务费专项资金项目(CUGL120261)

摘　　要：对河南省郑州市2008-2012年城市发展状况进行分析,从土地投入强度、土地利用强度、土地利用效益及生态环境4个方面选取13个指标构建土地集约利用评价体系,采用GA-BP网络模型,评价5年内郑州市土地集约利用水平。结果表明:1)GA—BP网络能够提高网络收敛速度,优化网络结构,算法具有有效性;2)郑州市近五年来土地集约利用水平处于不断上升阶段,其中市区城市土地集约度分值分别为0.610 7,0.644 3,0.661 8,0.701 2,0.705 8;中心城区、县级市和开发区土地集约利用水平有一定差距,但都有不同程度的提高,需要根据各地区情况,合理规划,进行土地资源利用挖潜。The land use status of Zhengzhou city during 2008-2012 was analyzed. 13 indexes from four aspects including the land investment intensity, land use intensity, land use efficiency and the ecological environment were selected to construct the evaluation system of land intensive use. The GA-BP network model was used to evaluate the land intensive use level of Zhengzhou city in the last five years. The results showed that the GA-BP algorithm run with better efficiency. The level of intensive use of land was in the rising stages in Zhengzhou city. The land intensive use level of center city, county-level city and development zone had a certain gap, but all was improved to a certain degree. Some advices on intensive use of urban land were put forward.

关 键 词：土地集约利用 遗传算法 神经网络 郑州市 

分 类 号：F293.2[经济管理—国民经济]