摘要
为提高矿震监测系统定位精度,减少监测盲区,降低监测成本,基于分布式的思想,提出一种基于手机移动传感网络的矿震定位方法.首先以矿区附近工人及家属等使用的智能手机建立手机移动传感网络,其次对模拟震源点网格化,构建基于标准差的目标函数,提出改进的萤火虫寻优策略,并使用拐点回溯法以及手机移动传感网络排除离散点策略(EDPS)降低定位误差,最后通过矿震模拟实验进行验证.实验结果表明:在手机移动传感网络无到时误差理想情况下,所有模拟震源点都能够准确收敛至震源位置,定位误差小于1 m.但手机相较于检波器到时误差较高,且定位误差与到时误差具有相关性,当手机到时误差为-1.0~1.0 s时,传统算法定位误差为216 m,无法实现高精度定位.通过研究目标函数值与定位误差间的关系,提出并使用拐点回溯法以及EDPS两种优化方法,算法绝对定位误差降低至73 m,当到时误差为-0.2~0.2 s时,绝对定位误差降低至17 m,定位精度提高76.1%.基于手机移动传感网络的矿震群智定位方法,为矿震监测提供了一种新方法,未来可考虑与井下微震系统联合,在节省监测成本、提高定位精度方面具有重要意义.
To improve the positioning accuracy of a mining-induced seismicity monitoring system,reduce the monitoring blind area,and reduce the monitoring cost,based on the distributed idea,this paper proposes a positioning method of mining-induced seismicity based on the smartphone sensor network.First,smartphones used by workers and their families near the mining area were utilized to establish a mobile sensor network.Second,the simulated source points were meshed,and the objective function based on the standard deviation was constructed.An improved firefly optimization strategy was proposed.The inflection point backtracking method and smartphone sensor network exclude the discrete points strategy,namely,EDPS,to reduce the positioning error.Verification is done by the simulation experiment of the mining-induced seismicity location.Experimental results show that under the ideal condition of no arrival time error in the smartphone sensor network,all simulated source points can converge to the source position accurately with a positioning error of less than 1 m.However,compared to the detector,the arrival error of the smartphone is higher,and the positioning error is correlated with the arrival error.When the mobile phone arrival error is−1.0–1.0 s,the traditional algorithm positioning error is 216 m,which cannot achieve high-accuracy positioning.Researching the relationship between objective function value and positioning error,this work proposes and uses two optimization methods:(1)inflection point backtracking method and(2)EDPS.The absolute positioning error of the algorithm is reduced to 73 m.When the time error is−0.2-0.2 s,the absolute positioning error is reduced to 17 m,and the positioning accuracy is improved by 76.1%.The location method of the mining-induced seismicity based on the crowdsensing of a phone mobile sensor network provides a new method for mining-induced seismicity monitoring.It can be considered to combine with an underground microseismic system in the future,which is of great significance in sa
作者
罗浩
冯天真
于靖康
潘一山
张利
LUO Hao;FENG Tian-zhen;YU Jing-kang;PAN Yi-shan;ZHANG Li(College of Information,Liaoning University,Shenyang 110036,China;School of Environment,Liaoning University,Shenyang 110036,China)
出处
《工程科学学报》
EI
CSCD
北大核心
2022年第12期2017-2028,共12页
Chinese Journal of Engineering
基金
国家自然科学基金资助项目(51704138)
国家重点研发计划资助项目(2017YFC0804208)
辽宁省教育厅科学技术研究资助项目(LQN201910)。
关键词
矿震
智能手机
群智定位
到时误差
排除离散点策略
mining-induced seismicity
smart phone
group intelligence positioning
arrival error
exclude discrete points strategy(EDPS)
