正在加载图片...
第4期 郭德勇等:煤与瓦斯突出预测灰色理论一神经网络方法 .357 余五次预测结果与生产实践相一致,说明所建立的 参考文献 网络模型可靠,经过灰色关联分析,突出特征值的 [1]焦作矿业学院瓦斯地质研究室,瓦斯地质概论·北京:煤炭工 提取比较合理,在神经网络预测中,根据具体情况不 业出版,1990,44 断吸收新样本,使待识别样本神经元的最大值、最小 [2]王宏图,鲜学福,王昌贤.四川盆地典型高瓦斯突出矿井瓦斯赋 值包含在训练样本中,这样可以提高识别的准确性, 存的地质特征.煤炭学报,1999,24(1):11 [3]郭德勇,韩德馨,王新义·煤与瓦斯突出的构造物理环境及其 4 结论 应用.北京科技大学学报,2002,24(6):581 [4]孙东玲,董钢峰,梁运培,煤与瓦斯突出预测指标临界值的选 (1)对平顶山八矿煤与瓦斯突出控制因素关联 取对预测准确率的影响,煤炭学报,2001,26(1):75 度计算结果表明,影响该矿煤与瓦斯突出的关键因 [5]Guo D Y.Han D X.Jiang GJ.Research on geological structure 素是地质构造和反映煤体结构指标的坚固性系数, mark of coal and gas outburst in Pingdingshan mining area.JChi- na Univ Min Technol.2002.11(1):72 其次为软分层厚度的变化, [6]Shepherd J.Rixon L K.Griffiths L.Outbursts and geological (2)煤与瓦斯突出预测参数具有模糊性、随机 structure in coal mines:A review.Int J Rock Mech Min Sci Ge- 性,即呈现灰色特性的因素,应用灰色关联分析法 omech Abstr.1981.18(4):267 能定量地处理煤与瓦斯突出因素设计中的灰色因 [7]Cao Y X.He DD.Cglick D.Coal and gas outburst in footwalls 素.通过灰色关联度分析,能够撇开复杂因素而对 of reverse faults.Int J Coal Geol.2001,46(1):47 [8]伍爱友,肖红飞,王从陆,等.煤与瓦斯突出控制因素加权灰 主要因素进行分析,对提高突出预测的准确性具有 色关联模型的建立与应用.煤炭学报,2005,30(1):58 一定指导意义, [9]傅立·灰色系统理论及其应用·北京:科学技术文献出版社, (③)选择控制平顶山八矿煤与瓦斯突出的10 1992:35 个主要控制因素,利用神经网络的方法建立了煤与 [10]施鸿宝。神经网络及其应用.西安:西安交通大学出版社, 瓦斯突出预测神经网络系统,对煤与瓦斯突出进行 1987:79 [11]郝吉生,袁崇孚.模糊神经网络技术在煤与瓦斯预测中的应 了预测,预测结果证明运用人工神经网络模型预测 用.煤炭学报,1999,24(6):77 突出危险性是可行的 Prediction method of coal and gas outburst using the grey theory and neural net- work GUO Deyong),LI Nianyou),PEI Dawen3),ZHENG Dengfeng) 1)Resource and Safety Engineering School,China University of Mining and Technology (Beijing).Beijing 100083,China 2)Sichuan Weitian Mine Safety Science and Technology Evaluate and Consultation Co Chengdu 610083,China 3)Pingdingshan Coal (group)Co.Ltd..Pingdingshan 467000.China ABSTRACI The grey theory and neural network method were applied to coal and gas outburst forecast.Main controlling factors of coal and gas outburst were filtered by the grey correlation method of the grey system theo- ry.The mathematical model and systematic structure of artificial neural network were founded to forecast the risk of coal and gas outburst.The effectiveness of the risk forecast in the outburst zone of Pingdingshan No.8 Coal Mine was demonstrated the grey theory and neural artificial network as a new means is available. KEY WORDS coal and gas outburst;outburst forecast;gray relevancy;artificial neural network余五次预测结果与生产实践相一致‚说明所建立的 网络模型可靠.经过灰色关联分析‚突出特征值的 提取比较合理‚在神经网络预测中‚根据具体情况不 断吸收新样本‚使待识别样本神经元的最大值、最小 值包含在训练样本中‚这样可以提高识别的准确性. 4 结论 (1) 对平顶山八矿煤与瓦斯突出控制因素关联 度计算结果表明‚影响该矿煤与瓦斯突出的关键因 素是地质构造和反映煤体结构指标的坚固性系数‚ 其次为软分层厚度的变化. (2) 煤与瓦斯突出预测参数具有模糊性、随机 性‚即呈现灰色特性的因素‚应用灰色关联分析法 能定量地处理煤与瓦斯突出因素设计中的灰色因 素.通过灰色关联度分析‚能够撇开复杂因素而对 主要因素进行分析‚对提高突出预测的准确性具有 一定指导意义. (3) 选择控制平顶山八矿煤与瓦斯突出的10 个主要控制因素‚利用神经网络的方法建立了煤与 瓦斯突出预测神经网络系统‚对煤与瓦斯突出进行 了预测‚预测结果证明运用人工神经网络模型预测 突出危险性是可行的. 参 考 文 献 [1] 焦作矿业学院瓦斯地质研究室.瓦斯地质概论.北京:煤炭工 业出版‚1990:44 [2] 王宏图‚鲜学福‚王昌贤.四川盆地典型高瓦斯突出矿井瓦斯赋 存的地质特征.煤炭学报‚1999‚24(1):11 [3] 郭德勇‚韩德馨‚王新义.煤与瓦斯突出的构造物理环境及其 应用.北京科技大学学报‚2002‚24(6):581 [4] 孙东玲‚董钢峰‚梁运培.煤与瓦斯突出预测指标临界值的选 取对预测准确率的影响.煤炭学报‚2001‚26(1):75 [5] Guo D Y‚Han D X‚Jiang G J.Research on geological structure mark of coal and gas outburst in Pingdingshan mining area.J Chi￾na Univ Min Technol‚2002‚11(1):72 [6] Shepherd J‚Rixon L K‚Griffiths L.Outbursts and geological structure in coal mines:A review.Int J Rock Mech Min Sci Ge￾omech Abstr‚1981‚18(4):267 [7] Cao Y X‚He D D‚Cglick D.Coal and gas outburst in footwalls of reverse faults.Int J Coal Geol‚2001‚46(1):47 [8] 伍爱友‚肖红飞‚王从陆‚等.煤与瓦斯突出控制因素加权灰 色关联模型的建立与应用.煤炭学报‚2005‚30(1):58 [9] 傅立.灰色系统理论及其应用.北京:科学技术文献出版社‚ 1992:35 [10] 施鸿宝.神经网络及其应用.西安:西安交通大学出版社‚ 1987:79 [11] 郝吉生‚袁崇孚.模糊神经网络技术在煤与瓦斯预测中的应 用.煤炭学报‚1999‚24(6):77 Prediction method of coal and gas outburst using the grey theory and neural net￾work GUO Deyong 1)‚LI Nianyou 1‚2)‚PEI Dawen 1‚3)‚ZHENG Dengfeng 1) 1) Resource and Safety Engineering School‚China University of Mining and Technology (Beijing)‚Beijing100083‚China 2) Sichuan Weitian Mine Safety Science and Technology Evaluate and Consultation Co.‚Chengdu610083‚China 3) Pingdingshan Coal (group) Co.Ltd.‚Pingdingshan467000‚China ABSTRACT The grey theory and neural network method were applied to coal and gas outburst forecast.Main controlling factors of coal and gas outburst were filtered by the grey correlation method of the grey system theo￾ry.The mathematical model and systematic structure of artificial neural network were founded to forecast the risk of coal and gas outburst.The effectiveness of the risk forecast in the outburst zone of Pingdingshan No.8 Coal Mine was demonstrated the grey theory and neural artificial network as a new means is available. KEY WORDS coal and gas outburst;outburst forecast;gray relevancy;artificial neural network 第4期 郭德勇等: 煤与瓦斯突出预测灰色理论-神经网络方法 ·357·
<<向上翻页
©2008-现在 cucdc.com 高等教育资讯网 版权所有