本文選題：砂巖型鈾礦 + 灰色系統(tǒng)GM（1��； 參考：《吉林大學(xué)》2015年碩士論文

【摘要】：砂巖型鈾礦是目前世界上重要的鈾礦產(chǎn)出類(lèi)型之一，，也是鈾資源總量中占據(jù)優(yōu)勢(shì)的鈾礦礦種。世界著名的中亞鈾成礦帶為砂巖型鈾礦成礦帶；鈾礦資源作為重要的核能源，已經(jīng)成為國(guó)民經(jīng)濟(jì)能源結(jié)構(gòu)中的重要組成部分。本文首先對(duì)世界鈾礦及砂巖型鈾礦進(jìn)行概述，根據(jù)地質(zhì)作用對(duì)典型鈾礦床進(jìn)行分類(lèi)。其次，對(duì)國(guó)外砂巖型鈾礦從四個(gè)不同時(shí)期進(jìn)行現(xiàn)狀闡述；將國(guó)外砂巖型鈾礦與我國(guó)砂巖型鈾礦現(xiàn)狀進(jìn)行對(duì)比分析，分析表明目前我國(guó)鈾礦資源仍以砂巖型鈾礦為主，其中以北方中新生代盆地為主要找礦目標(biāo)。北方巨型盆地包括松遼盆地、塔里木盆地、二連盆地、鄂爾多斯盆地等等。作者在綜合前人關(guān)于砂巖型鈾礦的分類(lèi)模式的基礎(chǔ)上，依據(jù)6種分類(lèi)標(biāo)準(zhǔn)重新對(duì)砂巖型鈾礦進(jìn)行劃分。在分類(lèi)研究的同時(shí)，重點(diǎn)探討了我國(guó)北方砂巖型鈾礦成礦模式，并將其與中亞砂巖型鈾礦及北美砂巖型鈾礦成礦模式進(jìn)行類(lèi)比，類(lèi)比區(qū)域包括烏茲別克斯坦的中卡茲爾庫(kù)姆礦區(qū)、哈薩克斯坦的門(mén)庫(kù)杜克礦區(qū)及美國(guó)的加斯希爾斯礦區(qū)。結(jié)果表明，砂巖型鈾礦勘查研究以及儲(chǔ)量分布構(gòu)成我國(guó)鈾礦資源的重要組成部分。 我國(guó)經(jīng)濟(jì)的快速發(fā)展必然導(dǎo)致能源供應(yīng)緊張，鈾礦資源已經(jīng)成為我國(guó)能源消費(fèi)中的主體，呈現(xiàn)供不應(yīng)求的狀態(tài)，鈾礦資源產(chǎn)出與消費(fèi)矛盾日益突出。本文收集了我國(guó)1998-2013年鈾資源消費(fèi)量數(shù)據(jù)，以Matlab計(jì)算軟件為平臺(tái)，分別建立三種定量模型對(duì)2014年與2015年鈾資源消費(fèi)量進(jìn)行預(yù)測(cè)。預(yù)測(cè)模型包括灰色系統(tǒng)GM（1,1）模型、一元線性回歸模型及BP神經(jīng)網(wǎng)絡(luò)分析。同時(shí)，分別給出了上述三種方法的后驗(yàn)差檢驗(yàn)、顯著性檢驗(yàn)及誤差百分比值，并根據(jù)2013年鈾消費(fèi)量的實(shí)際值與預(yù)測(cè)值分別給出三種方法所得結(jié)果的誤差率。預(yù)測(cè)表明，三種定量模型方法預(yù)測(cè)的結(jié)果相當(dāng)，預(yù)測(cè)的誤差都符合檢驗(yàn)標(biāo)準(zhǔn)。最終選定誤差值最小的方法，即BP神經(jīng)網(wǎng)絡(luò)方法來(lái)對(duì)2014年和2015年鈾資源消費(fèi)量進(jìn)行預(yù)測(cè)，預(yù)測(cè)值分別為7764噸和8210噸。得出的鈾資源消費(fèi)量為預(yù)測(cè)估算值，對(duì)今后的我國(guó)鈾礦資源供需平衡具有一定的指導(dǎo)作用。
[Abstract]:Sandstone-type uranium deposits are one of the most important types of uranium deposits in the world at present, and they are also the dominant uranium deposits in the total uranium resources. The world famous uranium metallogenic belt in Central Asia is a sandstone-type uranium metallogenic belt, and uranium resources, as an important nuclear energy, have become an important part of the energy structure of the national economy. This paper first summarizes the world uranium deposits and sandstone-type uranium deposits and classifies typical uranium deposits according to geological processes. Secondly, the present situation of sandstone-type uranium deposits in foreign countries is expounded in four different periods, and the present situation of sandstone-type uranium deposits abroad is compared with that of sandstone-type uranium deposits in China. The analysis shows that sandstone-type uranium deposits are still the main resources of uranium deposits in China at present. The Mesozoic and Cenozoic basins in the north are the main prospecting targets. Northern giant basins include Songliao Basin, Tarim Basin, Erlian Basin, Ordos Basin and so on. Based on the previous classification models of sandstone-type uranium deposits, the author reclassifies the sandstone-type uranium deposits according to six classification criteria. At the same time, the metallogenic model of sandstone-type uranium deposits in northern China is discussed, and the metallogenic model of sandstone-type uranium deposits in Central Asia and North America is compared with that of sandstone-type uranium deposits in North America. Analogy areas include Uzbekistan's Middle Kazkum, Kazakhstan's Menkouduk, and the United States' Gashires. The results show that the study of sandstone-type uranium deposits and the distribution of reserves constitute an important part of uranium resources in China. The rapid development of China's economy will inevitably lead to a tight supply of energy. Uranium resources have become the main body of energy consumption in China, showing a state that supply exceeds supply, and the contradiction between the output and consumption of uranium resources is increasingly prominent. In this paper, we collect the data of uranium resource consumption from 1998 to 2013 in China, and establish three quantitative models to predict uranium resource consumption in 2014 and 2015 based on Matlab software. The prediction model includes grey system GM-1) model, univariate linear regression model and BP neural network analysis. At the same time, the posterior error test, significance test and error percentage value of the three methods are given, and the error rates of the three methods are given according to the actual value of uranium consumption in 2013 and the predicted value. The prediction results show that the prediction results of the three quantitative model methods are similar, and the errors of the prediction are all in line with the test standard. Finally, the method of minimum error, BP neural network method, was chosen to forecast uranium resource consumption in 2014 and 2015. The predicted values were 7764 tons and 8210 tons respectively. The uranium resource consumption is predicted and estimated, which has a certain guiding effect on the balance of supply and demand of uranium resources in China in the future.
【學(xué)位授予單位】：吉林大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2015
【分類(lèi)號(hào)】：P619.14

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 李朝陽(yáng);魏毅;;基于MATLAB灰色GM(1,1)模型的大氣污染物濃度預(yù)測(cè)[J];環(huán)境科學(xué)與管理;2012年01期

2 陳戴生;李勝祥;蔡煜琦;;我國(guó)中、新生代盆地砂巖型鈾礦沉積環(huán)境研究概述[J];沉積學(xué)報(bào);2006年02期

3 張金帶;徐高中;林錦榮;彭云彪;王果;;中國(guó)北方6種新的砂巖型鈾礦對(duì)鈾資源潛力的提示[J];中國(guó)地質(zhì);2010年05期

4 權(quán)志高,李占雙;新疆十紅灘砂巖型鈾礦床基本特征及成因分析[J];地質(zhì)論評(píng);2002年04期

5 張金帶;李友良;簡(jiǎn)曉飛;;我國(guó)鈾資源勘查狀況及發(fā)展前景[J];中國(guó)工程科學(xué);2008年01期

6 闕為民;王海峰;牛玉清;張飛鳳;谷萬(wàn)成;;中國(guó)鈾礦采冶技術(shù)發(fā)展與展望[J];中國(guó)工程科學(xué);2008年03期

7 趙忠華;中國(guó)北東部中新生代盆地地浸砂巖型鈾礦成礦條件及找礦方向[J];地質(zhì)與資源;2003年01期

8 張字龍;韓效忠;姚春玲;;砂巖型鈾礦床鈾源條件評(píng)價(jià)方法探討[J];世界核地質(zhì)科學(xué);2006年02期

9 范磊,張運(yùn)陶,程正軍;基于Matlab的改進(jìn)BP神經(jīng)網(wǎng)絡(luò)及其應(yīng)用[J];西華師范大學(xué)學(xué)報(bào)(自然科學(xué)版);2005年01期

10 楊殿忠,夏斌,吳國(guó)干;吐哈盆地西南部砂巖鈾礦層間氧化帶發(fā)育特征[J];中國(guó)科學(xué)(D輯:地球科學(xué));2003年07期

本文編號(hào)：1874875