节点文献

一种基于R型因子分析和概率神经网络的冲击地压危险性等级评价模型

Evaluation model for the risk grade of rock burst based on the R-type factor analysis and a probabilistic neural network

  • 推荐 CAJ下载
  • PDF下载
  • 不支持迅雷等下载工具,请取消加速工具后下载。

【作者】 王佳信周宗红李克钢王海泉付自国李晓飞

【Author】 WANG Jiaxin;ZHOU Zonghong;LI Kegang;WANG Haiquan;FU Ziguo;LI Xiaofei;Faculty of Land Resource Engineering,Kunming University of Science and Technology;School of Resources and Safety Engineering,Central South University;

【通讯作者】 周宗红;

【机构】 昆明理工大学国土资源工程学院中南大学资源与安全工程学院

【摘要】 冲击地压是煤矿生产活动中常见的一种动力灾害,其危险性等级评价是冲击地压防治工作中施行解危措施的前提和基础。综合考虑冲击地压发生的地质条件和开采技术条件,选取煤层厚度、倾角、埋深、地质构造情况、煤层倾角变化、厚度变化、瓦斯浓度、顶板管理、采前卸压情况和响煤炮声等10个影响因素作为冲击地压危险性等级评价指标,但指标间或多或少存在一定的相关性,导致指标信息重叠。为了提高冲击地压危险性等级预测精度,借鉴一种R型因子分析对冲击地压10个指标进行降维处理,提取5个主因子作为新的评价指标。在R型因子分析的基础上,建立重庆砚石台煤矿冲击地压危险性等级评价的PNN模型,仿真结果表明:5种不同的训练和测试下PNN模型仍具有良好的评价效果,其正判率分别为94.29%,88.57%,88.57%,85.71%和82.86%,同时验证R型因子分析对冲击地压危险性等级评价结果。说明R型因子分析与PNN模型结合可为煤矿开采中冲击地压危险性等级评价提供一种思路。

【Abstract】 Rock burst is one of the common dynamic disasters in coal mine production. Hence, the evaluation of the risk grade is the prerequisite for the implementation of countermeasures in the prevention and control of rock burst. Factors of geological and mining technique conditions were considered comprehensively,and 10 factors,including the coal seam thickness,dip angle,depth,geological tectonic situation,change of coal seam dip angle,change of coal seam thickness,gas concentration,roof control, state of pressure relief before excavation and shooting,were selected as evaluation indexes of the risk grade for rock burst. However, there exist somewhat correlations, large or small, between these factors. In order to improve the prediction precision of rock burst, the method of R-type factor analysis was utilized to synthesize 5 main factors from these 10 factors as new evaluation indexes to eliminate the information overlapping of these factors, and reduce the input dimension. Based on this, a PNN model for the evaluation of risk grade of rock burst in Yanshitai coal mine in Chongqing was established. The simulation results show that the PNN model presents a favorable performance for 5 different kinds of training and test samples, and the distinguishing-positive rates are 94.29%, 88.57%, 88.57%, 85.71% and 82.86% respectively. Meanwhile, the evaluation performance of the R-type factor analysis model was validated. It indicates that the combination of R-type factor analysis and PNN model can provide a way for the risk grade evaluation of rock burst in coal mines.

【基金】 国家自然科学基金(51264018;51064012;41672303);中南大学研究生自主探索创新项目(2017ZZTS185)
  • 【文献出处】 振动与冲击 ,Journal of Vibration and Shock , 编辑部邮箱 ,2019年02期
  • 【分类号】TD324;TP183
  • 【被引频次】4
  • 【下载频次】321
节点文献中: