基于滑动时步和预警阈值的双参数地音预警模型研究
|
摘要
基于地音预警原理,选取地音活动量偏差值DEV作为地音预警的指标,根据预警时效性和实用性制定了预警规则,提出使用不同滑动时步n和预警阈值DEV_0组合优化预警模型的思想。利用多因素无重复试验方差分析对影响地音报准率和误报率的n,DEV_0双参数进行分析,得出n,DEV_0对预警效果均具有显著影响,且DEV_0是主要影响因素。通过构建效用加权双目标函数,求得预警模型的最优解组合为滑动时步取8步,预警阈值取70%,此时对应地音的报准率67%,漏准率33%,误报率29%。
Based on underground sound early warning principle,and then deviation value of terrestrial sound activity DEV was selected as underground sound early warning index,warning principle was constituted according timeliness and practicability of early warning,the idea that combination optimization early warning model with different sliding step n and warning thresholds DEV_0 was put forward.Underground sound precision and false alarm rate,and DEV_0 two parameters were analyzed by multiple factors and unrepeated trials variance analysis,the parameters n and DEV_0 play an important role on warning results,the DEV_0 was the main influence factors.According effectiveness weighing double aims function,the optimal solution combination of early warning was derived as sliding steps was 8 and warning thresholds was 70%,the corresponding underground sound precision was 67%,missing rate was 33% and false rate was 29%.
Based on underground sound early warning principle,and then deviation value of terrestrial sound activity DEV was selected as underground sound early warning index,warning principle was constituted according timeliness and practicability of early warning,the idea that combination optimization early warning model with different sliding step n and warning thresholds DEV_0 was put forward.Underground sound precision and false alarm rate,and DEV_0 two parameters were analyzed by multiple factors and unrepeated trials variance analysis,the parameters n and DEV_0 play an important role on warning results,the DEV_0 was the main influence factors.According effectiveness weighing double aims function,the optimal solution combination of early warning was derived as sliding steps was 8 and warning thresholds was 70%,the corresponding underground sound precision was 67%,missing rate was 33% and false rate was 29%.
引文
[1]潘俊锋,王书文,刘少虹,等.浅部矿井静载荷主导型冲击地压监测方法与实践[J].煤炭科学技术,2016,44(6):64-70,98.
[2]夏永学,康立军,齐庆新,等.基于微震监测的5个指标及其在冲击地压预测中的应用[J].煤炭学报,2010,35(12):2011-2016.
[3]刘金海,翟明华,郭信山,等.震动场、应力场联合监测冲击地压的理论与应用[J].煤炭学报,2014,39(2):353-363.
[4]聂百胜,何学秋,王恩元,等.电磁辐射法预测煤矿冲击地压[J].太原理工大学学报,2000,31(6):609-611.
[5]刘金海.煤矿冲击地压监测预警技术新进展[J].煤炭科学技术,2016,44(6):71-77.
[6]夏永学.不同冲击启动类型的地音前兆信息识别[J].中国煤炭,2015,41(3):49-53.
[7]孙书亮,王元杰,乔中栋,等.地音监测技术在冲击矿压预测预报中的应用 [J].煤矿开采,2009,14(5): 80-82.
[8]窦林名,何学秋,Bernard Drzezls.冲击矿压危险性评价的地音法[J].中国矿业大学学报,2000,29(1):85-88.
[9]邓志刚,任勇,王传朋,等.微震地音数据综合分析法初探[J].煤矿开采,2010,5(1):8-10,14.
[10]郭照庄,霍东升,孙月芳.密度核估计中窗宽选择的一种新方法[J].佳木斯大学学报 (自然科学版),2008,26(3):401-403.
[11]傅强,张兴敏.时变参数模型的最优滚动窗宽选择标准及应用[J].中国管理科学,2018,26(8):21-30.
[12]夏永学,蓝航,魏向志.基于微震和地音监测的冲击危险性综合评价技术研究[J].煤炭学报,2011,36(S2): 358-364.
[13]李岩,马江军,左政,等.地音监测技术在常村煤矿冲击地压预警中的作用[J].中国煤炭,2016,42(1):22-26.
[14]许洪泽,周梅.基于假设性检验的地面大气电场预警阀值确定方法[J].气象与减灾研究,2017,40(1):56-63.
[15]刘恒.基于数据驱动的过程参数报警阈值优化[D].北京:北京化工大学,2012.
[16]肖丹卉.结合优先级的多变量过程报警阈值优化方法[D].北京:北京化工大学,2014.
[2]夏永学,康立军,齐庆新,等.基于微震监测的5个指标及其在冲击地压预测中的应用[J].煤炭学报,2010,35(12):2011-2016.
[3]刘金海,翟明华,郭信山,等.震动场、应力场联合监测冲击地压的理论与应用[J].煤炭学报,2014,39(2):353-363.
[4]聂百胜,何学秋,王恩元,等.电磁辐射法预测煤矿冲击地压[J].太原理工大学学报,2000,31(6):609-611.
[5]刘金海.煤矿冲击地压监测预警技术新进展[J].煤炭科学技术,2016,44(6):71-77.
[6]夏永学.不同冲击启动类型的地音前兆信息识别[J].中国煤炭,2015,41(3):49-53.
[7]孙书亮,王元杰,乔中栋,等.地音监测技术在冲击矿压预测预报中的应用 [J].煤矿开采,2009,14(5): 80-82.
[8]窦林名,何学秋,Bernard Drzezls.冲击矿压危险性评价的地音法[J].中国矿业大学学报,2000,29(1):85-88.
[9]邓志刚,任勇,王传朋,等.微震地音数据综合分析法初探[J].煤矿开采,2010,5(1):8-10,14.
[10]郭照庄,霍东升,孙月芳.密度核估计中窗宽选择的一种新方法[J].佳木斯大学学报 (自然科学版),2008,26(3):401-403.
[11]傅强,张兴敏.时变参数模型的最优滚动窗宽选择标准及应用[J].中国管理科学,2018,26(8):21-30.
[12]夏永学,蓝航,魏向志.基于微震和地音监测的冲击危险性综合评价技术研究[J].煤炭学报,2011,36(S2): 358-364.
[13]李岩,马江军,左政,等.地音监测技术在常村煤矿冲击地压预警中的作用[J].中国煤炭,2016,42(1):22-26.
[14]许洪泽,周梅.基于假设性检验的地面大气电场预警阀值确定方法[J].气象与减灾研究,2017,40(1):56-63.
[15]刘恒.基于数据驱动的过程参数报警阈值优化[D].北京:北京化工大学,2012.
[16]肖丹卉.结合优先级的多变量过程报警阈值优化方法[D].北京:北京化工大学,2014.