基于SOM神经网络模型的耕地非农化压力区域差异研究——以湖北省为例

发布时间：2018-03-27 20:17

本文选题：SOM神经网络　切入点：耕地非农化　出处：《华中农业大学学报(社会科学版)》2017年04期

【摘要】：基于2006-2014年湖北省16市的面板数据,从经济因素、社会因素和资源禀赋三方面构建湖北省耕地非农化压力评价指标体系,运用SOM神经网络模型对湖北省的耕地非农化压力区域差异进行实证分析。结果表明:(1)湖北省16个主要城市的耕地非农化压力存在一定的区域差异。2006年以西南、东北地区往省内中部地区递减;2008年其区域差异变化不大;2010年则从西南-东北走向往边缘化发展;2012年呈现以武汉市、宜昌市为耕地非农化高压中心并辐射周边城市,形成了两条共同发展的平行线;直至2014年,武汉市耕地非农化压力趋于平缓,武汉市周边小城市的耕地非农化压力有小幅度上升。(2)SOM神经网络模型具有较强的聚类和容错能力,适用于开展耕地非农化压力的区域差异研究。研究结果反映了湖北省耕地非农化压力的空间异质性,为土地资源区际优化配置提供了基础。
[Abstract]:Based on the panel data of 16 cities in Hubei Province from 2006 to 2014, the evaluation index system of non-agricultural pressure of cultivated land in Hubei Province is constructed from three aspects: economic factors, social factors and resource endowment. The SOM neural network model is used to analyze the regional difference of cultivated land non-agricultural pressure in Hubei Province. The results show that there are some regional differences in non-agricultural pressure of cultivated land in 16 major cities of Hubei Province. The northeast region decreased to the central part of the province; in 2008, the regional difference did not change much; in 2010, it moved from southwest to northeast to marginalize; in 2012, Wuhan City, Yichang City as the non-agricultural high pressure center of cultivated land and radiation surrounding cities, Until 2014, the non-agricultural pressure of cultivated land in Wuhan tended to be gentle, and the non-agricultural pressure of cultivated land in small cities around Wuhan increased by a small margin. The SOM neural network model had strong clustering and fault-tolerant ability. The results reflect the spatial heterogeneity of the non-agricultural pressure of cultivated land in Hubei Province and provide the basis for the optimal allocation of land resources.
【作者单位】：华中农业大学公共管理学院;
【基金】：国家自然科学基金项目“城市群城镇用地扩张多尺度耦合机理研究”(41101098);国家自然科学基金项目“城镇用地扩张与耕地功能演变的时空耦合机理研究——以武汉市为例”(41371113) 国家社会科学基金青年项目“基于生产力总量平衡的耕地区域布局优化及补偿机制研究”(13CGL092)
【分类号】：F323.211

【相似文献】