基于BP神经网络的急倾斜煤层耦合致裂方案优化  

Optimization of coupled fracturing scheme for steeply inclined coal seam based on BP neural network

作　　者：张磊[1,2] 刘小明[1,2] 来兴平[1,2] 高语蔚[3] 崔峰 杨毅然[1,2] ZHANG Lei;LIU Xiao-ming;LAI Xing-ping;GAO Yu-wei;CUI Feng;YANG Yi-ran(College of Energy Scienee and Engbwering,Xi＇ an University of Science and Technology,Xi＇ an 710054,China;Key Laboratory of Western Mine Exploitation and Hazard Prevention,Ministry of Education,Xi＇ an University of Scienee and Technology,Xi＇ an 710054,China;College of Computer Science and Engineering,Xi＇ an University of Scienee and Technology,Xi＇ an 710054,Chin)

机构地区：[1]西安科技大学能源学院,陕西西安710054 [2]西安科技大学西部矿井开采及灾害防治重点实验室,陕西西安710054 [3]西安科技大学计算机科学与技术学院,陕西西安710054

出　　处：《西安科技大学学报》2018年第3期367-374,共8页Journal of Xi’an University of Science and Technology

基　　金：国家自然科学基金(51504184;U1361206);国家重点基础研究发展计划(973)项目(2015CB251602);陕西省博士后科研项目(2016BSHEDZZ29)

摘　　要：为准确优化急倾斜煤层顶煤耦合致裂方法,根据碱沟煤矿+495水平B1+2工作面开采技术条件,梳理制约顶煤冒放性的7个主要因素;构建了预测顶煤冒放性的BP神经网络模型,并对不同的耦合致裂方案效果进行了预测与分析。结果表明,随顶煤裂化率增加,顶煤冒放性逐步提高,当劣化率大于43%时,弱化后顶煤达到Ⅰ类冒放性标准。考虑经济和实施环境等,选定注水压力为5 MPa,炸药单耗为0.3 kg/m^3为最优方案参数。优化方案实施后,碱沟煤矿顶煤冒放性由第Ⅳ类向第Ⅰ类转化,顶煤冒放性与回采率大幅提升,为安全高效开采提供了理论依据。In order to accurately optimize the coupling cracking method for the roof of steeply inclined coal seams,seven factors that restrict top coal caving are made clear according to the mining conditions of B1 ＋ 2 working face at ＋ 495 level in Jiangou Coal Mine. A BP neural network model for predicting top coal caving property is constructed,and Matlab was adopted to train the network model to obtain the optimized model. The optimized model was then used to predict and analyze the effect of coupling cracking scheme when various weakening parameters was applied. Results show that the caving ability of top coal increases gradually with the increase of cracking rate. When the deterioration rate is greater than 43%,the top coal would reach the class I caving level after weakening. Considered the economic and implementation environment,we selected 5 MPa as water injection pressure and 0. 3 kg/m3 as the unit explosive consumption for the optimal parameters. After the implementation of the optimized scheme,the caving property of the top coal in the coal mine is transformed from class Ⅳ level to classⅠ level,and the caving and recovery rate of top coal is greatly improved,which provides a theoretical basis for the safe and efficient mining of steeply inclined coal seams.

关 键 词：急倾斜煤层 BP神经网络 耦合致裂 裂化率 方案优化 

分 类 号：TD32[矿业工程—矿井建设]