本文選題：頂板突水 + 三支決策粗糙集��； 參考：《山東科技大學(xué)》2017年碩士論文

【摘要】：在我國(guó),煤炭?jī)?chǔ)存量比較豐富,不過(guò)其使用量也極大,并且對(duì)促進(jìn)我國(guó)國(guó)民經(jīng)濟(jì)發(fā)展具有深遠(yuǎn)影響而成為我國(guó)重要的基礎(chǔ)能源。在煤炭產(chǎn)量節(jié)節(jié)升高之時(shí),對(duì)于煤層的開(kāi)采也越來(lái)越深入,因此開(kāi)采地層環(huán)境也變得更為復(fù)雜,造成煤層頂板突水發(fā)生事故的幾率增加,因此作者在前人的基礎(chǔ)上引入三支決策粗糙集理論,用于研究煤層頂板突水危險(xiǎn)性評(píng)價(jià),為煤礦安全生產(chǎn)盡一份力量。文章選取礦井頂板突水的主要因素:頂板隔水層隔水強(qiáng)度,導(dǎo)水裂隙帶高度,含水層水壓,上覆含水層的富水程度,地質(zhì)構(gòu)造發(fā)育情況五個(gè)因素。將五個(gè)因素代入SPSS22中,根據(jù)顯著性選出頂板隔水層隔水強(qiáng)度,導(dǎo)水裂隙帶高度,地質(zhì)構(gòu)造發(fā)育情況三個(gè)因素,排除含水層水壓,上覆含水層的富水性這兩個(gè)因素。三支決策粗糙集將貝葉斯風(fēng)險(xiǎn)分析和概率性風(fēng)險(xiǎn)關(guān)系引入粗糙集中,設(shè)置了三種決策規(guī)則。之后將頂板隔水層隔水強(qiáng)度,導(dǎo)水裂隙帶高度,地質(zhì)構(gòu)造發(fā)育情況三個(gè)因素代入二元logistic方程,求出每個(gè)樣本數(shù)據(jù)的突水的條件概率。然后采取一種自適算法求閥值α, β,根據(jù)樣本突水概率與閥值α,β的關(guān)系將樣本分成突水,不突水,延遲決策三類(lèi)。通過(guò)上述三支決策粗糙集理論建立的煤礦頂板突水預(yù)測(cè)系統(tǒng)將操作予以簡(jiǎn)化,不失為一種用于煤礦頂板突水預(yù)測(cè)的有效方法。將三支決策粗糙集理論用于對(duì)老公營(yíng)子煤礦與鮑店煤礦5304-2工作面驗(yàn)證,老公營(yíng)子煤礦頂板發(fā)生頂板突水的概率大于α,鮑店煤礦5304-2工作面頂板發(fā)生的概率小于β,符合三支決策規(guī)律,從而證實(shí)了三支決策粗糙集方法在頂板突水預(yù)測(cè)中的可行性。礦井頂板突水預(yù)測(cè)系統(tǒng)由Visual Basic語(yǔ)言編程而來(lái),分為煤礦數(shù)據(jù)信息文件中的礦井板突水預(yù)測(cè)系統(tǒng)以及管理系統(tǒng)2個(gè),同時(shí)礦井?dāng)?shù)據(jù)文件中的管理子系統(tǒng)為突水預(yù)測(cè)提供數(shù)據(jù)支持,并且針對(duì)頂板突水因素信息進(jìn)行管理。根據(jù)計(jì)算相關(guān)的條件概率與閾值而設(shè)計(jì)的礦井頂板突水預(yù)測(cè)子系統(tǒng),能夠?qū)ΦV井頂板突水情況作出合理判斷。
[Abstract]:In China, coal storage is relatively rich, but its use is also great, and has a far-reaching impact on promoting the development of China's national economy, and has become an important basic energy in China. When the coal production is increasing, the mining of coal seam is getting deeper and deeper, so the mining environment becomes more complicated, resulting in an increase in the probability of water inrush from the roof of coal seam. Therefore, the author introduces the three-branch decision rough set theory on the basis of predecessors, which is used to study the risk assessment of coal seam roof water inrush, so as to make a contribution to the safety of coal mine production. In this paper, the main factors of water inrush from mine roof are selected as follows: water insulation intensity of roof partition layer, height of water-conducting fissure zone, water pressure of aquifer, water enrichment degree of overlying aquifer, and development of geological structure. In this paper, five factors are added to SPSS22. According to the significance, three factors are selected, such as the water insulation intensity of roof partition layer, the height of water-conducting fissure zone and the development of geological structure, and the water pressure of aquifer and the water enrichment of overlying aquifer are excluded. Three sets of decision rough sets introduce Bayesian risk analysis and probabilistic risk relation into rough sets and set up three kinds of decision rules. After that, three factors, such as the water insulation intensity, the height of the water-conducting fissure zone and the development of the geological structure, are substituted into the binary logistic equation, and the conditional probability of water inrush for each sample data is obtained. Then an adaptive algorithm is adopted to calculate the threshold 偽, 尾. According to the relationship between the probability of water inrush and the threshold 偽, 尾, the samples are divided into three categories: water inrush, no water inrush and delayed decision. The coal mine roof water inrush prediction system established by the above three decision rough set theory simplifies the operation and is an effective method for the prediction of coal mine roof water outburst. The theory of three-branch decision rough set is applied to verify the 5304-2 coal face of Cengyingzi Coal Mine and Baodian Coal Mine. The probability of roof water inrush in Laoyingzi coal mine is greater than 偽, and the probability of roof water inrush in Baodian coal mine 5304-2 face is less than 尾, which accords with the rule of three branches of decision making, which proves the feasibility of three branches decision rough set method in predicting roof water inrush. The prediction system of mine roof water inrush is programmed by Visual Basic language. It is divided into two parts: mine board water inrush prediction system and management system. Meanwhile, the management subsystem in mine data file provides data support for water inrush prediction. And for roof water inrush factor information management. Based on the calculation of the relative conditional probability and threshold value, the prediction subsystem of mine roof water inrush can make a reasonable judgment on the situation of mine roof water inrush.
【學(xué)位授予單位】：山東科技大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2017
【分類(lèi)號(hào)】：TD745.2

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 邢一飛;張誠(chéng);王建輝;;2007～2014年我國(guó)煤礦突水事故分析及規(guī)律研究[J];煤炭技術(shù);2016年07期

2 張雁;;老公營(yíng)子煤礦煤層頂板突水機(jī)理[J];煤田地質(zhì)與勘探;2015年04期

3 李博;;灰色關(guān)聯(lián)-層次分析法的煤層頂板突水危險(xiǎn)性評(píng)價(jià)模型[J];河南理工大學(xué)學(xué)報(bào)(自然科學(xué)版);2015年03期

4 楊培功;張雁;郭英杰;;老公營(yíng)子煤礦頂板突水機(jī)理與防水安全煤巖柱留設(shè)研究[J];煤礦安全;2015年05期

5 李坤;曾一凡;尚彥軍;武強(qiáng);何萬(wàn)通;;基于GIS的“三圖 雙預(yù)測(cè)法”的應(yīng)用[J];煤田地質(zhì)與勘探;2015年02期

6 楊志超;張成龍;吳奕;安薇薇;朱海兵;龔燈才;;基于粗糙集和RBF神經(jīng)網(wǎng)絡(luò)的變壓器故障診斷方法研究[J];電測(cè)與儀表;2014年21期

7 石黎;史玉珍;;基于粗集和BP神經(jīng)網(wǎng)絡(luò)的城市水資源可持續(xù)利用評(píng)價(jià)模型[J];水電能源科學(xué);2014年06期

8 嚴(yán)群;孫杰;王秋生;王慶峰;田力;;基于模糊聚類(lèi)神經(jīng)網(wǎng)絡(luò)的煤炭資源等級(jí)劃分方法——以?xún)?nèi)蒙古煤炭資源預(yù)測(cè)區(qū)為例[J];中國(guó)煤炭地質(zhì);2014年05期

9 馮雨;謝守祥;;中國(guó)煤炭產(chǎn)業(yè)周期性影響因素研究[J];中國(guó)煤炭;2014年01期

10 謝騁;商琳;;基于三支決策粗糙集的視頻異常行為檢測(cè)[J];南京大學(xué)學(xué)報(bào)(自然科學(xué)版);2013年04期

相關(guān)碩士學(xué)位論文 前6條

1 魏真;基于SVR的煤層頂板水害分析模型研究[D];中國(guó)礦業(yè)大學(xué);2015年

2 羅斌;基于GIS的頂板水害條件分析及多源信息綜合評(píng)價(jià)[D];中國(guó)礦業(yè)大學(xué);2014年

3 喬育鋒;遺傳算法和BP神經(jīng)網(wǎng)絡(luò)在煤礦突水預(yù)測(cè)中的應(yīng)用研究[D];西安建筑科技大學(xué);2011年

4 秦軍;煤炭市場(chǎng)及其穩(wěn)定供給的策略研究[D];浙江大學(xué);2009年

5 王苗;基于灰色理論的礦井頂板涌水量預(yù)測(cè)模型的建立與應(yīng)用[D];山東科技大學(xué);2007年

6 俞海玲;基于模糊物元模型的礦井頂板涌水預(yù)測(cè)方法研究[D];山東科技大學(xué);2006年

，

本文編號(hào)：1790842