本文選題：圍巖穩(wěn)定性分類 + 支護決策��； 參考：《安徽理工大學》2015年碩士論文

【摘要】：隨著我國經濟的發(fā)展,能源的需求量不斷上升,煤炭的消耗也越來越大,煤礦開采向深部延伸。隨著開采深度的增加,巷道圍巖的支護難度加大,這對支護技術提出了新的要求。錨桿支護是目前煤礦巖石巷道的主要支護方式之一,現有支護決策系統(tǒng)存在不足,要做改進。 本文主要以兩淮礦區(qū)巖石巷道為工程背景,收集了三十多條巷道工程地質、水文地質情況,以及支護方案的數據,對樣本數據初步分類處理,選擇了其中三十條作為決策系統(tǒng)訓練樣本,其余三條作為驗證樣本。同時,借助MATLAB軟件開發(fā)了適合兩淮礦區(qū)的“深井巷道圍巖穩(wěn)定性分類及支護決策系統(tǒng)”。 通過決策系統(tǒng)的開發(fā)研究過程,得到以下幾點結論： 1、得出了巷道變形失穩(wěn)以及支護的影響因素,和相應支護設計指標的選取原則。 2、基于工程巖體分級理論和專家評分法原則,對樣本巷道圍巖進行了分類,為決策系統(tǒng)提供了訓練樣本。 3、基于人工神經網絡預測模型,構建了決策系統(tǒng)圍巖分類與支護網絡模型；此網絡輸入層神經節(jié)點數為8,隱含層為1層,隱含層節(jié)點數為13,輸出層節(jié)點數為16。 4、基于MATLAB軟件開發(fā)出了適應兩淮礦區(qū)不同埋深巖石巷道的圍巖分類與支護決策系統(tǒng),利用GUI界面設計了三大系統(tǒng)功能模塊：(1)神經網絡模型的建立與選擇模塊；(2)巷道圍巖穩(wěn)定性分類模塊；(3)巷道支護設計模塊。 5、選擇了II7226N底抽巷作為試驗巷道,分別用FLAC3D軟件數值模擬和現場巷道變形觀測的手段對決策系統(tǒng)進行驗證,試驗結果表明本支護決策系統(tǒng)具有較強的實用性和可靠性,可以為巷道支護施工提供依據。
[Abstract]:With the development of economy in China, the energy demand is rising and the consumption of coal is increasing. With the increase of mining depth, it is more difficult to support the surrounding rock of roadway, which puts forward new requirements for supporting technology. Bolt support is one of the main supporting methods of rock roadway in coal mine at present. The existing supporting decision system is deficient and should be improved. In this paper, taking the rock roadway of Lianghuai mining area as the engineering background, more than 30 tunnel engineering geology, hydrogeological conditions and supporting scheme data are collected, and the sample data are preliminarily classified and processed. Thirty of them were selected as training samples for decision-making system and the other three as validation samples. At the same time, with the help of MATLAB software, a "classification and support decision system of surrounding rock stability of deep roadway" is developed for Lianghuai mining area. Through the development and research process of the decision-making system, the following conclusions are obtained: 1, the deformation and instability of roadway and the influencing factors of support are obtained. Secondly, based on the classification theory of engineering rock mass and the principle of expert scoring method, the rock surrounding rock of the sample roadway is classified. A training sample is provided for the decision making system. 3. Based on the artificial neural network prediction model, the wall rock classification and support network model of the decision system is constructed, the number of ganglion points in the input layer of the network is 8, and the hidden layer is one layer. The number of hidden layer nodes is 13 and the output layer nodes are 16.4. Based on MATLAB software, a wall rock classification and support decision system for different buried rock roadways in Lianghuai mining area is developed. Using GUI interface, three system function modules are designed: (1) neural network model establishment and selection module; (2) roadway surrounding rock stability classification module; (3) roadway support design module. 5, select II7226N bottom roadway as test roadway, The numerical simulation of FLAC3D software and the field observation of roadway deformation were used to verify the decision system. The experimental results show that the supporting decision system has strong practicability and reliability, and can provide the basis for roadway support construction.
【學位授予單位】：安徽理工大學
【學位級別】：碩士
【學位授予年份】：2015
【分類號】：TD353

【參考文獻】

相關期刊論文 前10條

1 倪志偉;;BP網絡中激活函數的深入研究[J];安徽大學學報(自然科學版);1997年03期

2 周鳴爭;人工神經網絡在自動控制系統(tǒng)中的應用[J];安徽機電學院學報;2000年02期

3 王國欣,王旭東,肖樹芳;廣義RQD在巖體各向異性中的應用研究[J];吉林大學學報(地球科學版);2002年03期

4 W.S.德肖維茨;H.H.愛因斯坦;吳玉忠;;人工智能在巖石力學中的應用[J];國外金屬礦采礦;1988年06期

5 張磊,胡春,錢鋒;BP算法局部極小問題改進的研究進展[J];工業(yè)控制計算機;2004年09期

6 吳凌云;BP神經網絡學習算法的改進及其應用[J];信息技術;2003年07期

7 祁瑞芳;新奧法與我國地下工程[J];哈爾濱建筑工程學院學報;1987年02期

8 馬世志,張茂林,靖洪文,陳坤福;巷道圍巖穩(wěn)定性分類方法評述[J];建井技術;2004年05期

9 張煜東;吳樂南;王水花;;專家系統(tǒng)發(fā)展綜述[J];計算機工程與應用;2010年19期

10 孔德森,蔣金泉;深部巷道圍巖穩(wěn)定性預測與錨桿支護優(yōu)化[J];礦山壓力與頂板管理;2002年02期

相關博士學位論文 前1條

1 隆金玲;Sum-of-Product神經網絡和徑向基函數神經網絡的逼近能力研究[D];大連理工大學;2008年

，

本文編號：2052587