本文選題：城市建設(shè)用地　切入點(diǎn)：規(guī)模預(yù)測(cè)　出處：《江西師范大學(xué)》2014年碩士論文　論文類型：學(xué)位論文

【摘要】：土地是人類賴以生存和發(fā)展的物質(zhì)基礎(chǔ)，是城市化和城市建設(shè)的最基本的載體。我國(guó)是一個(gè)人口眾多，土地資源稀少的國(guó)家。隨著我國(guó)經(jīng)濟(jì)的飛速發(fā)展，城市化進(jìn)程的的加快，城市建設(shè)用地規(guī)模得到了迅速擴(kuò)張，而城市建設(shè)用地的稀缺性與城市化進(jìn)程之間的矛盾日益尖銳，建設(shè)用地甚至已成為制約城市發(fā)展的瓶頸科學(xué)合理地安排城市建設(shè)用地規(guī)模顯得尤為重要。為了落實(shí)“十分珍惜和合理用每一寸土地”和“保護(hù)耕地”的基本國(guó)策，保障糧食安全，維護(hù)社會(huì)安定，防止城市建設(shè)用地規(guī)模盲目擴(kuò)張，緩解人地矛盾勢(shì)在必行。 目前，我國(guó)大部分城市都編制了城市土地利用總體規(guī)劃，新一輪土地利用總體規(guī)劃修編即將開始，合理分配城市建設(shè)用地指標(biāo)是規(guī)劃修編的重中之重。怎樣做到合理地安排城市建設(shè)用地，這就要求有一套科學(xué)合理的對(duì)城市建設(shè)用地規(guī)模的預(yù)測(cè)體系和方法。本文從南昌市社會(huì)經(jīng)濟(jì)發(fā)展的實(shí)際情況出發(fā)，分析了南昌市城市建設(shè)用地規(guī)模和城市用地效益之間的關(guān)系，發(fā)現(xiàn)南昌市土地利用效益仍未達(dá)到最高，繼續(xù)擴(kuò)大城市規(guī)模是可行的�？紤]到引起城市建設(shè)用地規(guī)模變化的因素繁多而且復(fù)雜，先對(duì)影響南昌市城市建設(shè)用地規(guī)模的相關(guān)因子進(jìn)行了篩選，采用主成分分析和相關(guān)分析相結(jié)合的方法，確定了主要的影響因子為人口、人建設(shè)用地面積和城市化率；為了提高預(yù)測(cè)的合理性和準(zhǔn)確性，本文結(jié)合城市建設(shè)用地規(guī)模預(yù)測(cè)的理論和相關(guān)方法，采用了灰色GM(1，1)預(yù)測(cè)模型、多元線性回歸模型和BP神經(jīng)網(wǎng)絡(luò)預(yù)測(cè)模型分別對(duì)南昌市城市建設(shè)用地規(guī)模進(jìn)行了預(yù)測(cè)并對(duì)預(yù)測(cè)結(jié)果做了對(duì)比分析。主要研究結(jié)論有： （1）南昌市市區(qū)建設(shè)用地規(guī)模逐漸擴(kuò)大，土地利用效益也有所提高，但與國(guó)內(nèi)其他特大城市相比，南昌市城市規(guī)模仍然比較小，用地效益還有提高的空間。 （2）由于選取的影響因子比較多，各因子之間或多或少存在一定的共線性，在篩選時(shí)容易導(dǎo)致主次不分，重復(fù)選取。因此，在影響因子時(shí)，本文采用了主成分分析和相關(guān)分析相結(jié)合的方法，從15個(gè)影響因子中篩選出了幾個(gè)主要的影響因子，從而避免了因所選因子過多導(dǎo)致計(jì)算繁雜，預(yù)測(cè)不準(zhǔn)確的問題。 （3）本文通過對(duì)南昌市城市建設(shè)用地規(guī)模的灰色GM（1,1）預(yù)測(cè)，結(jié)果表明擬合情況較好，能夠比較準(zhǔn)確的預(yù)測(cè)出城市建設(shè)用地規(guī)模。但是由于灰色預(yù)測(cè)模型只是從城市建設(shè)用地規(guī)模本身的時(shí)間序列變化進(jìn)行預(yù)測(cè)，忽略了其他的一些影響因素，導(dǎo)致灰色GM（1,1）預(yù)測(cè)不適合對(duì)城市建設(shè)用地規(guī)模進(jìn)行長(zhǎng)期預(yù)測(cè)。 （4）本文建立了城市建設(shè)用地規(guī)模預(yù)測(cè)的BP神經(jīng)網(wǎng)絡(luò)模型，研究結(jié)果表明BP神經(jīng)網(wǎng)絡(luò)模型擬合函數(shù)良好，而且預(yù)測(cè)精度較高，預(yù)測(cè)結(jié)果準(zhǔn)確，適合適合對(duì)城市建設(shè)用地規(guī)模進(jìn)行長(zhǎng)期預(yù)測(cè)。 （5）本文用多元線性回歸模型對(duì)南昌市城市建設(shè)用地規(guī)模進(jìn)行了預(yù)測(cè)，得出了引起城市建設(shè)用地規(guī)模變化的主要影響因素是人口、城市化率、人均建設(shè)用地面積，并以此構(gòu)建了多元線性回歸方程。實(shí)證研究表明函數(shù)擬合情況良好、預(yù)測(cè)精度較高，預(yù)測(cè)結(jié)果準(zhǔn)確。 （6）文章通過對(duì)灰色GM（1,1）模型、多元線性回歸模型和BP神經(jīng)網(wǎng)絡(luò)模型的對(duì)比分析，從研究結(jié)果可以看出BP神經(jīng)網(wǎng)絡(luò)模型的預(yù)測(cè)精度最高，，其次為多元線性回歸模型，灰色GM（1,1）模型最差。在此基礎(chǔ)上把BP神經(jīng)網(wǎng)絡(luò)模型和多元線性回歸模型預(yù)測(cè)的平均值作為2015年和2020年南昌市城市建設(shè)用地規(guī)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枃鐢ㄥ鍏冪嚎鎬у洖褰掓ā鍨嬪鍗楁槍甯?jìng)鍩庡競(jìng)寰忚畡ㄥ湴瑙勬ā杩涜浜?jiǎn)棰勬祴,寰楀嚭浜?jiǎn)寮曡捣鍩庡競(jìng)寰忚畡ㄥ湴瑙勬ā鍙樺寲鐨勪富瑕佸奖鍝嶅洜绱犳槸錆h鍙

本文編號(hào)：1630514