本文選題：鐵礦石　切入點(diǎn)：礦石品位　出處：《武漢理工大學(xué)》2015年碩士論文　論文類型：學(xué)位論文

【摘要】：隨著我國經(jīng)濟(jì)的迅速發(fā)展,我國對(duì)礦產(chǎn)資源的消費(fèi)需求也越來越大。近些年,采礦業(yè)發(fā)展迅速,這就對(duì)采礦技術(shù)提出了越來越高的要求。在開采礦產(chǎn)資源的過程當(dāng)中,品位可以用來衡量礦石的品性和質(zhì)量,品位是指礦石中有用組成部分的單位含量。礦石品位的高低直接反映了礦石品質(zhì)的優(yōu)劣程度。因此,找到一種合適的方法測(cè)量出礦石的品位,用來指導(dǎo)實(shí)際中的礦石的開采,將在很大程度上提高礦石的開采效率,增強(qiáng)了工廠的生產(chǎn)力和生產(chǎn)效益。本次課題研究的原始數(shù)據(jù)的采集來自于激光掃描儀的測(cè)量,利用MATLAB對(duì)數(shù)據(jù)進(jìn)行處理。首先對(duì)掃描的數(shù)據(jù)進(jìn)行三維數(shù)值模擬,得到多節(jié)車廂的三維網(wǎng)線圖,再以一節(jié)車廂的掃描數(shù)據(jù)作為樣本還原礦車表面的實(shí)際情況。其次,通過計(jì)算礦車表面小方塊的面積占整個(gè)表面積的比例,還原小車表面的塊礦比。對(duì)數(shù)據(jù)的處理只是課題的一部分,怎樣通過數(shù)據(jù)去求得車廂內(nèi)鐵礦石的體積和松散系數(shù)也是亟待解決的一個(gè)問題。因?yàn)檐噹麅?nèi)礦石是不規(guī)則的物體,所以,礦石的體積要用特殊的方法才能夠計(jì)算。裝載礦石的車廂是一個(gè)有規(guī)則長方體,這樣可以很容易的計(jì)算出礦車的底面積,車廂內(nèi)礦石的高度可以通過采集到的數(shù)據(jù)處理得到,再利用數(shù)值積分原理求得礦石的體積。求出了礦石的體積,進(jìn)而就可以算出礦石的松散系數(shù)(松散系數(shù)是指土石料松動(dòng)的體積與土石料未松動(dòng)時(shí)的自然體積的比值。)。通過構(gòu)造BP神經(jīng)網(wǎng)絡(luò)模型找出礦石的重量、體積及松散系數(shù)和礦石品位的某種對(duì)應(yīng)關(guān)系。通過礦石的重量、體積和松散系數(shù)求出礦石的品位,進(jìn)而與實(shí)驗(yàn)得出的品位相比較,礦石的塊礦比用于松散系數(shù)的修正,從而縮小實(shí)驗(yàn)品位與計(jì)算品位的誤差。構(gòu)造BP神經(jīng)網(wǎng)絡(luò)模型,不斷的去訓(xùn)練輸入和輸出的樣本,得到最佳的網(wǎng)絡(luò)權(quán)值矩陣,最終找到分析品位的計(jì)算方法,這種計(jì)算品位的方法將比實(shí)驗(yàn)方法更有效率,對(duì)于今后礦石品位的分析計(jì)量有著極其重要的作用。
[Abstract]:With the rapid development of China's economy, the consumption demand for mineral resources in China is also increasing. In recent years, the mining industry has developed rapidly, which puts forward higher and higher requirements for mining technology. In the process of mining mineral resources, Grade can be used to measure the quality and quality of ore. Grade refers to the unit content of useful components in ore. The grade of ore directly reflects the quality of ore. Finding a suitable way to measure the grade of the ore to guide the actual mining of the ore will greatly improve the mining efficiency of the ore. The raw data collected in this paper come from the measurement of laser scanner, and the data are processed by MATLAB. Firstly, the 3D numerical simulation of the scanned data is carried out. The three-dimensional net diagram of several cars is obtained, and then the scanning data of one car is taken as a sample to restore the actual situation of the car surface. Secondly, by calculating the proportion of the area of the small square on the surface of the car to the whole surface area, How to calculate the volume and loose coefficient of iron ore in the carriage through the data is also a problem to be solved urgently, because the ore in the car is an irregular object. Therefore, the volume of ore can only be calculated by a special method. The carriage containing the ore is a regular cuboid, so that the bottom area of the ore truck can be easily calculated. The height of the ore in the car can be obtained by processing the collected data, and the volume of the ore can be obtained by using the principle of numerical integration. Then the loose coefficient of ore can be calculated (loose coefficient is the ratio of the loose volume of soil and stone to the natural volume of soil and stone when it is not loose.) the weight of ore can be found by constructing a BP neural network model. According to the weight, volume and loose coefficient of the ore, the grade of the ore is calculated, and compared with the grade obtained from the experiment, the lump ore ratio of the ore is used to revise the loose coefficient. In order to reduce the error between experimental grade and calculation grade, the BP neural network model is constructed, the input and output samples are trained continuously, the optimal network weight matrix is obtained, and the calculating method of analysis grade is finally found. This method will be more efficient than the experimental method, and will play an extremely important role in the analysis and measurement of ore grade in the future.
【學(xué)位授予單位】：武漢理工大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2015
【分類號(hào)】：TD861.1

【參考文獻(xiàn)】

相關(guān)期刊論文 前8條

1 王解先;侯東亞;段兵兵;;三維激光掃描儀在堆積物體積計(jì)算中的應(yīng)用[J];測(cè)繪通報(bào);2013年07期

2 鄧滿蘭;易淼;;基于MATLAB編程的教學(xué)討論[J];考試周刊;2010年05期

3 趙小軍;林晨;黃柳仙;楊以月;;基于MATLAB圖像處理的車輛檢測(cè)與識(shí)別[J];數(shù)據(jù)采集與處理;2009年S1期

4 康永偉;鐘若飛;吳俁;;車載激光掃描儀外參數(shù)標(biāo)定方法研究[J];紅外與激光工程;2008年S3期

5 雷曉云;張麗霞;梁新平;;基于MATLAB工具箱的BP神經(jīng)網(wǎng)絡(luò)年徑流量預(yù)測(cè)模型研究——以塔城地區(qū)烏拉斯臺(tái)河為例[J];水文;2008年01期

6 毛建東;;動(dòng)態(tài)稱重系統(tǒng)的動(dòng)態(tài)補(bǔ)償和校正[J];食品與機(jī)械;2006年02期

7 ;Application of an expert system using neural network to control the coagulant dosing in water treatment plant[J];Journal of Control Theory and Applications;2004年01期

8 楊耀權(quán),施仁,于希寧,高鏜年;大型煤堆體積自動(dòng)檢測(cè)系統(tǒng)[J];計(jì)量技術(shù);1999年05期

相關(guān)會(huì)議論文 前1條

1 周愛民;鮑愛華;徐必根;徐樹嵐;;我國有色礦山的主要技術(shù)成就與差距[A];第六屆全國采礦學(xué)術(shù)會(huì)議論文集[C];1999年

相關(guān)博士學(xué)位論文 前1條

1 鄧偉;BP神經(jīng)網(wǎng)絡(luò)構(gòu)建與優(yōu)化的研究及其在醫(yī)學(xué)統(tǒng)計(jì)中的應(yīng)用[D];復(fù)旦大學(xué);2002年

相關(guān)碩士學(xué)位論文 前7條

1 王龍;井下礦石運(yùn)輸?shù)V車信息顯示牌動(dòng)態(tài)識(shí)別系統(tǒng)研究[D];武漢理工大學(xué);2011年

2 孟操;井下鐵礦石品位自動(dòng)計(jì)量系統(tǒng)的設(shè)計(jì)與開發(fā)[D];武漢理工大學(xué);2011年

3 李俊清;基于智能技術(shù)的鐵礦石品位仿真分析[D];武漢理工大學(xué);2011年

4 陳國利;中加礦業(yè)公司礦石品位動(dòng)態(tài)檢測(cè)及質(zhì)量控制系統(tǒng)試驗(yàn)研究[D];西安建筑科技大學(xué);2009年

5 陳敏;基于BP神經(jīng)網(wǎng)絡(luò)的混沌時(shí)間序列預(yù)測(cè)模型研究[D];中南大學(xué);2007年

6 吳昌友;神經(jīng)網(wǎng)絡(luò)的研究及應(yīng)用[D];東北農(nóng)業(yè)大學(xué);2007年

7 吳仕勇;基于數(shù)值計(jì)算方法的BP神經(jīng)網(wǎng)絡(luò)及遺傳算法的優(yōu)化研究[D];云南師范大學(xué);2006年

，

本文編號(hào)：1634409