QQ群聊

• CN 62-1112/TF
• ISSN 1005-2518
• 创刊于1988年

## Optimization of Micro-seismic Monitoring Network Layout in Linglong Gold Mine

CUI Yu,, LI Xibing,, DONG Longjun, BAI Lü

School of Resources and Safety Engineering，Central South University，Changsha 410083，Hunan，China

 基金资助: 国家重点研发计划项目“深部高应力诱导与能量调控理论”.  2016YFC0600706

Received: 2018-05-27   Revised: 2018-09-13   Online: 2019-07-08

Abstract

Due to the using of large-scale mining equipment and the improvement of production management level in recent decades，long-term mining has led to the depletion of shallow mineral resources. Because of the complex geological conditions and high ground stress in deep mining，high-energy rock burst，earthquake，large-area goaf instability and other dynamic disasters are more likely to occur in the process of deep mining.Moreover，these geological hazards are difficult to accurately predict and prevent by traditional monitoring techniques.Micro-seismic monitoring technology can monitor micro-seismic events in the form of elastic waveforms released by rock mass during deformation and fracture in real time.It can also determine the location and energy parameters of micro-seismic events，so as to evaluate the safety of rock mass activity and stability.This is the main monitoring means of dynamic disasters in existing mines，and has been widely used in engineering fields with high risk of rock burst.The mining depth of Linglong gold mine in Shandong has exceeded 1 000 meters，and a micro-seismic monitoring system has been built in the deep part of the mine.Layout of network is the first and most important part of the construction of micro-seismic monitoring system，and it is the key factor affecting the effect of micro-seismic monitoring.Generally，that is need to be focused on are technology and economy factors.Technology is feasible to ensure the accuracy of monitoring data within the scope of system monitoring.Economy is reasonable to ensure that equipment and construction costs are reasonable.Due to the complexity of underground engineering，the layout of micro-seismic monitoring network is greatly limited，so it is usually necessary to compare several schemes to select the most suitable one.However，because the program optimization of micro-seismic monitoring system is a comprehensive evaluation problem involving multiple indicators，the traditional empirical analogy method is more subjective and difficult to achieve quantitative judgement.Based on the principal component analysis （PCA），a comprehensive optimization analysis model for micro-seismic monitoring network was established.First，a scientific and reasonable evaluation index system needs to be established.When determining the evaluation index，the principal component analysis can eliminate the influence of the correlation among the indicators，and does not need to consider the independence of the indicators.At the same time，it can simplify the data structure of the evaluation index and transform it into a few comprehensive indicators.Therefore，it is necessary to fully and comprehensively consider the various influencing factors in the system construction and operation stage，and try to select more evaluation indicators to make the evaluation results more comprehensive and accurate.Based on the actual situation of the construction and operation of the micro-seismic monitoring system in the Dakaitou mining area of Linglong gold mine，and combined with the evaluation parameters of previous related projects，eight indicators were selected from two aspects of economic and technical conditions to build a comprehensive evaluation index system.After the calculation of the model，eight original indicators were replaced by two new composite indicators，and the latter included about 91.9% information of the original data，which greatly simplifies the data structure of the scheme evaluation.Finally，based on the difference of the original data itself，the information contribution rate is used as the weight coefficient of the new comprehensive index，which avoids the error of subjective weight.It is a more scientific and simple weighting method.The comprehensive evaluation value of each scheme is calculated，and the scheme Ⅲ with the comprehensive evaluation value of 0.71 was the optimal scheme.The principal component analysis model provides a concise and effective comprehensive evaluation method for the optimization of micro-seismic monitoring network schemes.

Keywords： deep mining ; micro-seismic monitoring ; sensor layout ; multivariate statistical analysis ; program evaluation ; PCA ; weight ; sensitivity

CUI Yu, LI Xibing, DONG Longjun, BAI Lü. Optimization of Micro-seismic Monitoring Network Layout in Linglong Gold Mine[J]. Gold Science and Technology, 2019, 27(3): 417-424 doi:10.11872/j.issn.1005-2518.2019.03.417

### 1.1 主成分分析的原理

$Xn×p=x11x12⋯x1p⋮⋮⋮⋮xn1xn2⋯xnp$

### 图1

Fig.1   General steps of principal component analysis

### 1.2 PCA优化分析模型的算法

（1）数据预处理。

（2）计算协方差矩阵$∑$及其特征值和特征向量。

，数据极值正规化处理后的方差$Dzj=Dxj∕∆j2$，一致化决策矩阵$Z=Z1,Z2,⋯,ZP$p个指标两两之间的协方差为

$CovZi,Zj=EZi-EZiZj-EZj$

$∑ij=1n-1∑k=1nZki-Z¯iZkj-Z¯j$

$i$$j$为负向变量，则：

$Fi=eiZ=ei1Z1+ei2Z2+⋯+eiPZP$

（3）确定主成分贡献率和综合评价值。

$p×p$阶协方差矩阵$∑$p个特征值的和等于总方差，也等于各主成分的方差之和[21]。即

$wk=λk/∑i=1pλi$
$αm=∑i=1mλi/∑i=1pλi$

### 2 工程应用实例

Table 1  Evaluation indexes of the layout scheme of

$X1$/万元$X2$/万元$X3$/万元$X4$/m$X5$/m$X6$$X7$/%$X8$/年
I9016.58.630310.5555
12425.512.52820-1.2755
10217.69.61414-1.6858
15832.016.22419-1.4856
12323.612.41616-2.0808
24248.231.01615-1.6898

$Z=1.001.001.000.000.000.000.000.000.780.720.830.130.650.680.640.000.920.970.961.001.000.840.901.000.550.510.660.380.710.760.900.330.780.780.830.880.881.000.771.000.000.000.000.880.880.841.001.00$

### 2.1 原始数据预处理

$X$进行指标类型统一化和无量纲化处理，对负向指标X1X6、正向指标X7X8分别使用式（2）、式（3）进行极差正规化处理，将原始指标矩阵$X$转换成一致化指标决策矩阵$Z$

### 2.2 选定主成分

Table 2  Eigenvalue and variance analysis table

$F1$0.84068.24668.246
$F2$0.29223.73091.976
$F3$0.0897.24099.216
$F4$0.0080.62099.836
$F5$0.0020.164100.00
$F6$00100.00
$F7$00100.00
$F8$00100.00

### 图2

Fig.2   Scree plot of principal component analysis

### 2.3 计算综合评价值

Table 3  F1、F2 scores and comprehensive evaluation values of each scheme

-1.618-0.201-1.25
-0.608-0.324-0.53
0.4631.4180.71
0.167-0.502-0.12
0.5040.9250.61
1.241.3170.58

### 图3

Fig.3   Parameter setting of the sensor

### 图4

Fig.4   Simulation cloud of micro-seismic monitoring system performance of scheme Ⅲ

## 3 结论

（1）根据玲珑金矿大开头矿段地压微震在线监测系统构建时矿山的实际情况，选取了经济和技术等8个指标构建了基于主成分分析的微震台网布设综合评价指标体系，全面考虑了微震监测系统建设过程中的各方面要求。

（2）将评价模型应用于玲珑金矿微震台网布设的方案优化中，在涵盖了原始数据91.9%信息量的情况下，用2个新的综合评价指标（主成分）替代了原始的8个指标，新的综合指标之间互不相关，极大地简化了方案评价的数据结构。

（3）以选定主成分的方差占总方差的比例作为各自的信息贡献率，经加权处理后作为综合指标的权重。这种赋权方法是基于原始数据本身的差异，不需要专家评判或评分，避免了主观权重的误差，综合评价结果只与原始数据有关。利用模型对大开头矿段的微震监测方案进行优化分析，得出方案Ⅲ综合性能最好。

## 参考文献 原文顺序 文献年度倒序 文中引用次数倒序 被引期刊影响因子

［M］.北京科学出版社2014.

Li Xibing .

Rock Dynamics Fundamentals and Applications

［M］.BeijingScience Press2014.

［J］.现代矿业，20111）：1-14.

Wang Yunmin .

Opportunities and challenges to metal mine mining industry and the technical countermeasures

［J］.Modern Mining20111）：1-14.

［J］.中国有色金属学报，2017276）：1236-1262.

Li Xibing Zhou Jian Wang Shaofeng et al .

Review and practice of deep mining for solid mineral resources

［J］.The Chinese Journal of Nonferrous Metals2017276）：1236-1262.

［C］//香山第175次科学会议.北京：中国环境科学出版社，2002101-108.

Li Xibing Gu Desheng .

The hazard control and cataclastic mutagenesis induced by high stress in hard rock mining at depth

［C］//The 175th Xiangshan Science Congress.Beijing：China Environmental Science Press2002101-108.

［J］.岩石力学与工程学报，20113010）：2057-2067.

Dong Longjun Li Xibing Tang Lizhong et al .

Mathematical functions and parameters for micro-seismic source location without pre-measuring speed

［J］.Chinese Journal of Rock Mechanics and Engineering20113010）：2057-2067.

［J］.地下空间与工程学报，200951）：122-128.

Li Shulin .

Discussion on micro-seismic monitoring technology and its applications to underground projects

［J］.Chinese Journal of Underground Space and Engineering200951）：122-128.

Yang C X Luo Z Q Hu G B et al .

Application of a microseismic monitoring system in deep mining

［J］.Journal of University of Science and Technology Beijing，2007141）：6-8.

Bertoncini C A Hinders M K .

Fuzzy classification of roof fall predictors in microseismic monitoring

［J］.Measurement，20104310）：1690-1701.

［J］.黄金科学技术，2016241）：76-79.

Zhang Hongshan Song Wenzhi Li Qiutao et al .

Analysis of micro-seismicity activity induced by deep ore body mining at Jinqingding gold mine，Shandong Province

［J］.Gold Science and Technology2016241）：76-79.

［J］.岩石力学与工程学报，20052412）：2048-2053.

Li Shulin Yin Xiangang Zheng Wenda et al .

Research of multi-channel micro-seismic monitoring system and its application to Fankou lead-zinc mine

［J］.Chinese Journal of Rock Mechanical and Engineering20052412）：2048-2053.

Kijko A .

An algorithm for the optimum distribution of a regional seismic network —Ⅱ.An analysis of the accuracy of location of local earthquakes depending on the number of seismic stations

［J］.Pure and Applied Geophysics，19771154）：1011-1021.

Mendecki A J Aswegen G V Mountfort P I .

A Guide to Routine Seismic Monitoring in Mines

［M］// Jager A J，Ryder J A.A Handbook on Rock Engineering Practice for Tabular Hard Rock Mines.JohannesburgThe Safety in Mines Research Advisory Committee1999287-309.

［J］.地球物理学报，2010532）：457-465.

Gong Siyuan Dou Linming , Cao Anye et al .

Study on optimal configuration of seismological observation network for coal mine

［J］.Chinese Journal of Geophysics2010532）：457-465.

［J］.北京科技大学学报，20133512）：1538-1545.

Gao Yongtao Wu Qingliang Wu Shunchuan et al .

Optimization of micro-seismic monitoring networks based on the theory of D-optimal design

［J］.Journal of University of Science and Technology Beijing20133512）：1538-1545.

［J］.中国安全生产科学技术，2016127）：82-86.

Zhou Yongyong Li Xibing Liu Zhixiang et al .

Fuzzy optimum approach for three-dimensional underground micro-seismic network with high precision

［J］.Journal of Safety Science and Technology2016127）：82-86.

［J］.中国安全生产科学技术，20151112）：180-185.

Lu Dandan Song Wenhua Zhang Guichuan et al .

Evaluation on vulnerability of hazard bearing body in petrochemical enterprises based on AHP-entropy weight method

［J］.Journal of Safety Science and Technology20151112）：180-185.

［J］.黄金科学技术，2015236）：58-63.

Li Xibing Zhu Wei Liu Weijun et al .

Research on rock mass blastability based on principal component analysis and RBF neural network

［J］.Gold Science and Technology2015236）：58-63.

［J］.黄金科学技术，2018261）：49-55.

Liu Qiang Li Xibing Liang Weizhang .

PCA-RF model for the classification of rock mass quality and its application

［J］.Gold Science and Technology2018261）：49-55.

［J］.河北工业大学学报，1999281）：94-97.

Li Yanshuang Zeng Zhenxiang Zhang Min et al .

Application of primary component analysis in the methods of comprehensive evaluation for many indexes

［J］.Journal of Hebei University of Technology1999281）：94-97.

［J］.统计研究，1995126）：47-51.

Bai Xuemei Zhao Songshan .

Discussion on some problems of comprehensive evaluation method of principal component analysis

［J］.Statistics Research1995126）：47-51.

Johnson R A. Wichern D W .

［M］.4版.北京清华大学出版社2001.

Johnson R A Wichern D W .

Applied Multivariate Statistical Analysis

［M］.4th Edition.BeijingTsinghua University Press2001.

［R］.长沙中南大学2014.

Li Xibing .

Research report of seismic monitoring and safety early warning system in Yongshaba mine

［R］.ChangshaCentral South University2014.

/

 〈 〉