基于ANN模型的城市土地集约利用中观评价研究

发布时间：2018-03-05 02:12

本文选题：集约利用　切入点：中观评价　出处：《河北联合大学》2014年硕士论文　论文类型：学位论文

【摘要】：随着城市化和工业化的快速发展，我国城市土地资源面临的空间压力和环境压力越来越大，出现了城市外延扩展严重、土地利用效益低下、用地结构不合理等现象，造成人地矛盾、经济与环境矛盾的日益突出。所以，集约利用土地对经济循环发展、建设节约型社会、提高土地使用效率具有重大的意义。在土地集约利用评价方法方面，以土地集约利用的一些相关理论为基础，对模糊综合评价模型、PRS模型、TOPSIS模型、多因素综合评价模型、BP神经网络模型等的特点进行比较，综合分析，特别针对BP神经网络模型进行研究，建立BP神经网络模型，对其进行仿真训练，得出该模型应用于土地集约利用评价中，能很好的解决各种评价指标之间的非线性关系，消除人为确定指标权重值带来的主观因素影响。在城市功能区划分方面，通过对城市土地利用特点进行分析，，以GIS技术作为支撑，研究了中观尺度下城市土地功能区和样本片区的划分方法。从土地利用强度、土地投入水平、土地利用效率多个方面定性分析城市整体、各功能区之间的区别，针对不同类型的功能用地选取不同的评价指标，构建了一套完整的评价指标体系。以唐山市中心城区为例进行实证研究。选取2009年的Quirk Bird高分辨率遥感影像作为数据源，在中观尺度下将唐山中心城区划分为居住功能区、工业功能区和商业功能区，并对各功能区从土地利用强度、土地投入和土地产出三个方面建立各自适用的评价指标体系，借助BP人工神经网络模型测算不同功能区土地集约利用水平，并对其集约利用潜力值进行计算。得出唐山市居住用地集约度处于中下等水平，具有一定的挖掘空间，主要的挖潜方向为新城扩展和旧城改造；工业用地集约度水平在中高等以上，集约度较高的区域为高新技术开发区以及东部工业区，挖掘空间较小；商业用地集约利用处于中等水平，集约度值较高的为唐山市的各大商场，还有一定的挖掘空间，但相对较小。
[Abstract]:With the rapid development of urbanization and industrialization, urban land resources in China are facing more and more space pressure and environmental pressure, such as serious urban extension and expansion, low efficiency of land use, unreasonable structure of land use and so on. Therefore, intensive use of land is of great significance to the development of economy cycle, the construction of economical society and the improvement of land use efficiency. In the aspect of land intensive utilization evaluation method, based on some related theories of land intensive utilization, the characteristics of fuzzy comprehensive evaluation model, PRS model and TOPSIS model, multifactor comprehensive evaluation model and BP neural network model are compared. Comprehensive analysis, especially for the study of BP neural network model, the establishment of BP neural network model, simulation training to its conclusion that the model is used in land intensive use evaluation, It can solve the nonlinear relationship between various evaluation indexes and eliminate the subjective factors which caused by the artificial determination of index weight. In the urban functional area division, the characteristics of urban land use are analyzed. Based on GIS technology, this paper studies the method of dividing urban land function area and sample area in mesoscale scale. It qualitatively analyzes the whole city from the aspects of land use intensity, land input level and land use efficiency. According to the differences between different functional areas, a complete evaluation index system is constructed according to the different evaluation indexes of different types of functional land. Taking Tangshan central city as an example, the Quirk Bird high-resolution remote sensing image of 2009 is selected as data source, and Tangshan central urban area is divided into residential function area, industrial function area and commercial function area in mesoscale scale. The evaluation index system of land use intensity, land input and land output are established for each functional area, and the land intensive utilization level of different functional areas is calculated by using BP artificial neural network model. It is concluded that the intensity of residential land in Tangshan is at the middle and lower level and has a certain excavation space. The main direction of tapping potential is the expansion of the new city and the transformation of the old city. The industrial land intensive degree level is above the middle and higher level, the high intensity degree area is the high-tech development zone and the eastern industrial zone, the excavating space is small, the commercial land intensive utilization is in the middle level, The high degree of intensity is the major shopping malls in Tangshan City, there is a certain excavation space, but relatively small.
【学位授予单位】：河北联合大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：F301.2;F224

【参考文献】