南昌市城市建设用地规模预测研究

发布时间：2018-03-18 17:09

  本文选题：城市建设用地　切入点：规模预测　出处：《江西师范大学》2014年硕士论文　论文类型：学位论文

【摘要】：土地是人类赖以生存和发展的物质基础，是城市化和城市建设的最基本的载体。我国是一个人口众多，土地资源稀少的国家。随着我国经济的飞速发展，城市化进程的的加快，城市建设用地规模得到了迅速扩张，而城市建设用地的稀缺性与城市化进程之间的矛盾日益尖锐，建设用地甚至已成为制约城市发展的瓶颈科学合理地安排城市建设用地规模显得尤为重要。为了落实“十分珍惜和合理用每一寸土地”和“保护耕地”的基本国策，保障粮食安全，维护社会安定，防止城市建设用地规模盲目扩张，缓解人地矛盾势在必行。 目前，我国大部分城市都编制了城市土地利用总体规划，新一轮土地利用总体规划修编即将开始，合理分配城市建设用地指标是规划修编的重中之重。怎样做到合理地安排城市建设用地，这就要求有一套科学合理的对城市建设用地规模的预测体系和方法。本文从南昌市社会经济发展的实际情况出发，分析了南昌市城市建设用地规模和城市用地效益之间的关系，发现南昌市土地利用效益仍未达到最高，继续扩大城市规模是可行的。考虑到引起城市建设用地规模变化的因素繁多而且复杂，先对影响南昌市城市建设用地规模的相关因子进行了筛选，采用主成分分析和相关分析相结合的方法，确定了主要的影响因子为人口、人建设用地面积和城市化率；为了提高预测的合理性和准确性，本文结合城市建设用地规模预测的理论和相关方法，采用了灰色GM(1，1)预测模型、多元线性回归模型和BP神经网络预测模型分别对南昌市城市建设用地规模进行了预测并对预测结果做了对比分析。主要研究结论有： （1）南昌市市区建设用地规模逐渐扩大，土地利用效益也有所提高，但与国内其他特大城市相比，南昌市城市规模仍然比较小，用地效益还有提高的空间。 （2）由于选取的影响因子比较多，各因子之间或多或少存在一定的共线性，在筛选时容易导致主次不分，重复选取。因此，在影响因子时，本文采用了主成分分析和相关分析相结合的方法，从15个影响因子中筛选出了几个主要的影响因子，从而避免了因所选因子过多导致计算繁杂，预测不准确的问题。 （3）本文通过对南昌市城市建设用地规模的灰色GM（1,1）预测，结果表明拟合情况较好，能够比较准确的预测出城市建设用地规模。但是由于灰色预测模型只是从城市建设用地规模本身的时间序列变化进行预测，忽略了其他的一些影响因素，导致灰色GM（1,1）预测不适合对城市建设用地规模进行长期预测。 （4）本文建立了城市建设用地规模预测的BP神经网络模型，研究结果表明BP神经网络模型拟合函数良好，而且预测精度较高，预测结果准确，适合适合对城市建设用地规模进行长期预测。 （5）本文用多元线性回归模型对南昌市城市建设用地规模进行了预测，得出了引起城市建设用地规模变化的主要影响因素是人口、城市化率、人均建设用地面积，并以此构建了多元线性回归方程。实证研究表明函数拟合情况良好、预测精度较高，预测结果准确。 （6）文章通过对灰色GM（1,1）模型、多元线性回归模型和BP神经网络模型的对比分析，从研究结果可以看出BP神经网络模型的预测精度最高，，其次为多元线性回归模型，灰色GM（1,1）模型最差。在此基础上把BP神经网络模型和多元线性回归模型预测的平均值作为2015年和2020年南昌市城市建设用地规模。
[Abstract]:Land is the material basis for human survival and development, is the most basic carrier of urbanization and city construction. China is a country with large population, scarce land resources of the country. With the rapid development of China's economy, city urbanization accelerated, city construction land scale has been expanded rapidly, and the increasingly sharp contradiction between city construction with scarcity and the process of the city construction land, even science has become a bottleneck restricting the development of the city to arrange city scale of construction land is very important. In order to implement the "treasure and combined with the basic national policy of every inch of land" and "land protection". To ensure food security, maintaining social stability, to prevent the blind expansion of the scale of city construction land, alleviate the contradiction between human and land is imperative.
At present, most of the city in our country have developed city land use planning, a new round of land use planning is about to begin, the rational allocation of city construction land index is the priority among priorities planning. How to reasonably arrange the city construction land, which requires a set of scientific and reasonable for the city for the construction of prediction system and method scale. This paper from the actual situation of Nanchang's economic and social development of Nanchang city construction, analyzes the relationship between land use efficiency and scale of city, found the highest benefit has yet to land use in Nanchang City, to continue to expand the scale of the city is feasible. Considering the factors of city construction the scale of land use change are many and complex, the influence of the construction of Nanchang city were selected by correlation factor scale, combined with principal component analysis and correlation analysis Methods to determine the main influence factor for the population, land area and the rate of city construction; in order to improve the prediction accuracy and rationality, this paper forecast the land scale of city construction theory and related methods, using the grey GM (1,1) prediction model, multiple regression model and BP neural network prediction model of Nanchang city construction land scale was predicted and the prediction results were compared and analyzed. The main conclusions are as follows:
(1) the scale of urban construction land in Nanchang has gradually expanded, and the land use efficiency has also improved. However, compared with other large cities in China, the scale of Nanchang city is still relatively small, and the efficiency of land use still has room for improvement.
(2) because of the selection of the impact factor is more, some are more or less collinearity among the various factors in the selection, easily lead to confuse, repeated selection. Therefore, the influence factor, this paper uses the method of principal component analysis and correlation analysis are combined, selected from 15 factors that affect several main factors, so as to avoid the selected factor too much lead to complex problems of inaccurate prediction.
(3) based on the grey GM scale of urban construction in Nanchang city (1,1) forecast, the result showed that the fitting is better able to use the scale of city construction in a more accurate prediction. But because of the grey prediction model is from the city construction land itself time series to predict the change, ignoring some factors other, grey GM (1,1) prediction result is not suitable for long-term prediction of city construction land scale.
(4) the BP neural network model of urban construction land scale prediction is established in this paper. The results show that the BP neural network model has good fitting function, and the prediction accuracy is high, and the prediction accuracy is accurate, which is suitable for long-term prediction of urban construction land scale.
(5)鏈�鏂囩敤澶氬厓绾挎�у洖褰掓ā鍨嬪�瑰崡鏄屽競鍩庡競寤鸿�剧敤鍦拌�勬ā杩涜�屼簡棰勬祴,寰楀嚭浜嗗紩璧峰煄甯傚缓璁剧敤鍦拌�勬ā鍙樺寲鐨勪富瑕佸奖鍝嶅洜绱犳槸浜哄�

本文编号：1630514