img

QQ群聊

img

官方微信

高级检索

黄金科学技术 ›› 2023, Vol. 31 ›› Issue (5): 794-802.doi: 10.11872/j.issn.1005-2518.2023.05.046

• 采选技术与矿山管理 • 上一篇    下一篇

基于改进DCNN法的微震信号自动识别模型及应用

杨轶男1(),胡建华2(),周坦1,赵风文1,王牧帆1   

  1. 1.中南大学资源与安全工程学院,湖南 长沙 410083
    2.福州大学紫金地质与矿业学院,福建 福州 350108
  • 收稿日期:2023-03-21 修回日期:2023-09-11 出版日期:2023-10-31 发布日期:2023-11-21
  • 通讯作者: 胡建华 E-mail:857674908@qq.com;hujh21@126.com
  • 作者简介:杨轶男(1998-),女,黑龙江哈尔滨人,硕士研究生,从事矿山安全及充填研究工作。857674908@qq.com
  • 基金资助:
    国家自然科学基金项目“深地环境下胶结充填体多场多尺度力学行为试验与损伤机理研究”(52274182)

Automatic Recognition Model of Microseismic Signal Based on Improved DCNN Method and Its Application

Yinan YANG1(),Jianhua HU2(),Tan ZHOU1,Fengwen ZHAO1,Mufan WANG1   

  1. 1.School of Resources and Safety Engineering, Central South University, Changsha 410083, Hunan, China
    2.Zijin School of Geology and Mining, Fuzhou University, Fuzhou 350108, Fujian, China
  • Received:2023-03-21 Revised:2023-09-11 Online:2023-10-31 Published:2023-11-21
  • Contact: Jianhua HU E-mail:857674908@qq.com;hujh21@126.com

摘要:

为精准识别矿山微震信号,采用改进深度卷积神经网络(DCNN)法,通过傅里叶变换得到的频谱图叠加原始图像的空间域图作为微震信号的识别对象,提出了一种基于改进DCNN法的微震信号自动识别与分类模型,建立了某铅锌矿的IMS微震监测信号数据库的训练集、验证集和测试集,并通过实际工程数据验证了方法的可靠性。结果表明:利用频谱图和空间域图在BGR通道上堆叠的特征值作为DCNN输入的方法,构建的微震信号自动识别模型精度更高且泛化能力更强,该模型能够高效地提取微震信号特征;采用F1值、ROC曲线和AUC值3种性能度量进行评价,验证了改进方法的可行性、有效性和可靠性。

关键词: DCNN法, 微震监测, 信号识别, 傅里叶变换, 自动识别模型, 深部开采

Abstract:

In order to accurately identify mine microseismic signals,an improved deep convolutional neural network (DCNN) method was adopted.The spatial domain image of the original image superimposed by the spectrum map obtained by Fourier transform was used as the identification object of microseismic signals.An automatic identification and classification model of microseismic signals based on an improved DCNN method was proposed. The training set,verification set and test set of the IMS microseismic monitoring signal of a lead-zinc mine were established,and the reliability of the method was verified by actual engineering data. The results show that the automatic recognition model of microseismic signals with higher precision and better generalization ability is constructed by using the eigenvalues stacked on the BGR channel of the spectrum map and the spatial domain image as the input of DCNN,which can extract the features efficiently. By evaluating the F1 value,ROC curve and AUC value,the feasibility,effectiveness,and reliability of the improved method is verified.

Key words: DCNN method, microseismic monitoring, signal identification, Fourier transform, automatic recognition model, deep mining

中图分类号: 

  • TD326

表1

4种模型在TensorFlow平台上的实现"

网络模型数据库标准识别率/%
Le Net-5Mnist99.1
Alex Net(top-5)Image Net201280.2
Google Net(top-5)Image Net201288.9
ResNet152(top-5)Image Net201296.4

图1

傅里叶变换处理过程"

图2

微震事件样本图"

图3

微震事件样本的预处理过程"

图4

傅里叶变换前后图像对比"

图5

ResNet18网络结构模型"

图6

训练集与验证集的精度和损失曲线"

表2

模型预测结果与真实结果"

测试集类型真实情况/个预测结果/个

准确率

/%

微震事件非微震事件微震事件非微震事件
正例50004574391.4
反例05004845290.4

图7

模型的ROC曲线与AUC值"

图8

微震信号和爆破信号示例图"

表3

微震信号识别结果"

信号编号输入类别输出类别识别结果
微震信号1微震信号微震信号正确
微震信号2微震信号微震信号正确
微震信号3微震信号微震信号正确
爆破信号1爆破信号非微震信号正确
爆破信号2爆破信号非微震信号正确
爆破信号3爆破信号非微震信号正确
Cai Meifeng, Xue Dinglong, Ren Fenhua,2019.Current status and development strategy of metal mines[J]. Chinese Journal of Engineering,41(4):417-426.
Dong L J, Tang Z, Li X B,et al,2020.Discrimination of mining microseismic events and blasts using convolutional neural networks and original waveform[J].Journal of Central South University,27(10):3078-3089.
Fu Xuan, Huang Linqi, Chen Jiangzhan,et al,2022.Meeting the challenge of high geothermal ground temperature environment in deep mining—Research on geothermal ground temperature simulation platform of rock true triaxial testing machine[J].Gold Science and Technology,30(1):72-84.
Gao Youwen, Zhou Benjun, Hu Xiaofei,2018.Research on image recognition of convolution neural network based on data enhancement[J].Computer Technology and Development,28(8):62-65.
Gao Zhenyu,2018.Research and Application of Image Classification Method Based on Deep Convolutional Neural Network[D].Hefei:University of Science and Technology of China.
Gentili S, Michelini A,2006.Automatic picking of P and S phases using a neural tree[J].Journal of Seismology,10(1):39-63.
Hou Wei, Huo Haiying, Guo Xiaopeng,et al,2014.Research on microseismic monitoring and control technology of dynamic pressure in deep fully-mechanized top coal caving face[J].Coal Science and Technology,42(3):36-38.
Hu J H, Zhou T, Ma S W,et al,2022.Rock mass classification prediction model using heuristic algorithms and support vector machines:A case study of Chambishi copper mine[J].Scientific Reports,12:928.doi:https://doi.org/10.1038/s41598-022-05027-y .
doi: https://doi.org/10.1038/s41598-022-05027-y
Hu Naixun, Chen Tao, Zhen Na,et al,2021.Object-oriented open pit extraction based on convolutional neural network[J].Remote Sensing Technology and Application,36(2):265-274.
Jiang Xinmeng,2017.Application and Research of Convolution Neural Network Based on Tensor Flow[D].Wuhan:Central China Normal University.
Krizhevsky A, Sutskever I, Hinton G E,2012.ImageNet classification with deep convolutional neural networks[J].Communications of the ACM,60(6):84-90.
Li Xibing, Zhou Jian, Wang Shaofeng,et al,2017.Review and practice of deep mining for solid mineral resources[J]. The Chinses Journal of Nonferrous Metals,27(6):1236-1262.
Pang J M, Chen K, Shi J P,et al,2020.Libra R-CNN:Towards ba-lanced learning for object detection[C]//2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition(CVPR).California:IEEE.doi:10.1109/CVPR.2019.00091 .
doi: 10.1109/CVPR.2019.00091
Perol T, Gharbi M, Denolle M,2018.Convolutional neural network for earthquake detection and location[J].Science Advances,4(2):e1700578.
Ross Z E, Meier M A, Hauksson E,et al,2018.Generalized seismic phase detection with deep learning[J].Bulletin of the Seismological Society of America,108(5A):2894-2901.
Tian Xiao, Wang Mingjun, Zhang Xiong,et al,2022.Discrimination of earthquake and quarry blast based on multi-input convolutional neural network[J].Chinese Journal of Geophysics,65(5):1802-1812.
Wang J, Teng T L,1995.Artificial neural network-based seismic detector[J].Bulletin of the Seismological Society of America,85(1):308-319.
Wang Runpei,2020.Study on the Prediction of Deformation and Rock Pressure of Overburden Monitored by Distributed Optical Fiber Based on Machine Learning[D].Xi’an:Xi’an University of Science and Technology.
Wei Mengyi, Tan Yuyang, Mao Zhonghua,et al,2018.Automatic microseismic event detection and arrival picking based on waveform cross-correlation[J].Acta Scientiarum Naturalium Universitatis Pekinensis,54(2):361-372.
Wu Aixiang, Wang Yong, Zhang Minzhe,et al,2021.New development and prospect of key technology in underground mining of metal mines[J].Metal Mine,50(1):1-13.
Wu Xiaoyang,2008.Detection and Tracking of Moving Object in Image Sequence Based on OpenCV[D].Hangzhou:Zhejiang University.
Yang Chen,2018.Research on the Analysis of Underground Mine Microseismic Monitoring Signal [D].Wuhan:Wuhan University of Science and Technology.
Yao Kaiyi, Li Yingyu,2018.Neural-network-based seismic phase automatic pickup method [J].Electronic Design Engineering,26(22):1-5.
Yu Shibo, Yang Xiaocong, Yuan Ye,et al,2020.Research on destress effect of ground pressure control for the time-space mining sequence at depths[J].Gold Science and Technology,28(3):345-352.
Zhang Minmin, Xu Heping, Wang Xiaojie,et al,2017.Application of Google TensorFlow machine learning framework[J].Cyber Security and Data Governance,36(10):58-60.
Zhao Hongbao, Liu Rui, Gu Tao,et al,2021.Research on automatic picking method of microseismic signal P wave based on deep learning mode[J].Chinese Journal of Rock Mechanics and Engineering,40(Supp.2):3084-3097.
Zhao Y, Takano K,1999.An artificial neural network approach for broadband seismic phase picking[J].Bulletin of the Seismological Society of America,89:670-680.
Zhu Wei,2022.Research on Intelligent Identification Model of Microseismic and Blasting in Deep Mines[D].Changsha:Central South University.
蔡美峰,薛鼎龙,任奋华,2019.金属矿深部开采现状与发展战略[J].工程科学学报,41(4):417-426.
傅璇,黄麟淇,陈江湛,等,2022.迎接深部开采高地温环境的挑战——岩石真三轴试验机地温模拟平台研究[J].黄金科学技术,30(1):72-84.
高友文,周本君,胡晓飞,2018.基于数据增强的卷积神经网络图像识别研究[J].计算机技术与发展,28(8):62-65.
高震宇,2018.基于深度卷积神经网络的图像分类方法研究及应用[D].合肥:中国科学技术大学.
侯玮,霍海鹰,郭晓朋,等,2014.深部综放工作面动压微震监测及控制技术研究[J].煤炭科学技术,42(3):36-38.
胡乃勋,陈涛,甄娜,等,2021.基于卷积神经网络的面向对象露天采场提取[J].遥感技术与应用,36(2):265-274.
姜新猛,2017.基于TensorFlow的卷积神经网络的应用研究[D].武汉:华中师范大学.
李夕兵,周健,王少锋,等,2017. 深部固体资源开采评述与探索[J].中国有色金属学报,27(6):1236-1262.
田宵,汪明军,张雄,等,2022.基于多输入卷积神经网络的天然地震和爆破事件识别[J].地球物理学报,65(5):1802-1812.
王润沛,2020.基于机器学习的分布式光纤监测覆岩变形矿压预测研究[D].西安:西安科技大学.
魏梦祎,谭玉阳,毛中华,等,2018.基于波形互相关的微地震事件自动识别及初至拾取[J].北京大学学报(自然科学版),54(2):361-372.
吴爱祥,王勇,张敏哲,等,2021.金属矿山地下开采关键技术新进展与展望[J].金属矿山,50(1):1-13.
吴晓阳,2008.基于OpenCV的运动目标检测与跟踪[D].杭州:浙江大学.
杨晨,2018.地下矿山微震监测信号分析研究[D].武汉:武汉科技大学.
姚开一,李英玉,2018.基于神经网络的地震震相自动拾取方法[J].电子设计工程,26(22):1-5.
于世波,杨小聪,原野,等,2020.深部区域采矿时序的地压调控卸荷效应研究[J].黄金科学技术,28(3):345-352.
章敏敏,徐和平,王晓洁,等,2017.谷歌TensorFlow机器学习框架及应用[J].网络安全与数据治理,36(10):58-60.
赵洪宝,刘瑞,顾涛,等,2021.基于深度学习模式的微震信号P波自动拾取方法研究[J].岩石力学与工程学报,40(增2):3084-3097.
朱玮,2022.深部矿山微震与爆破智能辨识模型研究[D].长沙:中南大学.
[1] 张君, 杨清平, 刘芳芳, 张金钟, 徐刚强, 李晓松. 深井规模化开采矿山与分布式微震监测系统设计研究[J]. 黄金科学技术, 2023, 31(4): 659-668.
[2] 郭对明,李国清,侯杰,胡乃联. 基于FLUENT的深井掘进巷道局部通风参数优化[J]. 黄金科学技术, 2022, 30(5): 753-763.
[3] 陈立强,赵国彦,李洋,毛文杰,党成凯,方博扬. 开挖卸荷扰动下的深部巷道支护及其效果评价[J]. 黄金科学技术, 2022, 30(3): 438-448.
[4] 傅璇,黄麟淇,陈江湛,吴阳春,李夕兵. 迎接深部开采高地温环境的挑战——岩石真三轴试验机地温模拟平台研究[J]. 黄金科学技术, 2022, 30(1): 72-84.
[5] 唐宇,王少锋. 单向受限应力下镐型截齿破岩特性及其影响因素分析[J]. 黄金科学技术, 2021, 29(5): 669-679.
[6] 景岳,王少锋,鲁金涛. 矿岩开挖松动区厚度预测及非爆机械化开采判据[J]. 黄金科学技术, 2021, 29(4): 525-534.
[7] 党明智,张君,贾明涛. 黄土坡铜锌矿微震监测技术应用与灾害预警方法研究[J]. 黄金科学技术, 2020, 28(2): 246-254.
[8] 崔宇,李夕兵,董陇军,白吕. 玲珑金矿微震监测台网布设优化[J]. 黄金科学技术, 2019, 27(3): 417-424.
[9] 王春,王成,熊祖强,程露萍,王怀彬. 动力扰动下深部出矿巷道围岩的变形特征[J]. 黄金科学技术, 2019, 27(2): 232-240.
[10] 杨世兴,付玉华,占飞. 围压与料浆浓度对尾砂充填体峰后损伤演化研究[J]. 黄金科学技术, 2019, 27(1): 97-104.
[11] 陈冲,李夕兵,冯帆. 诱导巷道的围岩松动破坏区数值研究[J]. 黄金科学技术, 2018, 26(6): 771-779.
[12] 卢富然, 陈建宏. 基于AHP和熵权TOPSIS模型的岩爆预测方法[J]. 黄金科学技术, 2018, 26(3): 365-371.
[13] 刘晓明,赵君杰,彭平安,毕林,代碧波. 有效微震信号自动识别技术研究[J]. 黄金科学技术, 2017, 25(3): 84-91.
[14] 贾敏涛,汪群芳,吴冷峻. 深部开采热环境控制技术研究现状及展望[J]. 黄金科学技术, 2017, 25(2): 83-88.
[15] 张洪山,宋文志,李秋涛,赵兴东,李洋洋,刘强. 山东金青顶矿区深部矿体开采诱发微震活动分析[J]. 黄金科学技术, 2016, 24(1): 76-79.
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!