岩性油气藏 ›› 2026, Vol. 38 ›› Issue (3): 132–140.doi: 10.12108/yxyqc.20260311

• 地质勘探 • 上一篇    下一篇

基于GPU加速的SSA三维微地震实时定位方法及应用

曹立斌1(), 郑马嘉2,3, 陈倩1, 伍亚2, 程利敏1   

  1. 1 成都博赫科技有限公司， 成都 610213
    2 中国石油西南油气田公司 开发事业部， 成都 610051
    3 中国石油勘探开发研究院， 北京 100083
  • 收稿日期:2025-08-12 修回日期:2025-09-12 出版日期:2026-05-01 发布日期:2026-02-11
  • 第一作者:曹立斌(1976—)，男，硕士，高级工程师，主要从事微地震监测技术研究及其推广应用方面的工作。地址：（610213）四川省成都市天府新区麓山大道一段。Email:12001134@163.com。
  • 基金资助:
    中国石油天然气股份有限公司西南油气田公司科技项目“川南地区筇竹寺组深层页岩气‘双高’采集方法试验及处理解释技术研究”(2024D101-03-04)

GPU-accelerated source scanning algorithm for real-time 3D microseismic monitoring and its application ​

CAO Libin1(), ZHENG Majia2,3, CHEN Qian1, WU Ya2, CHENG Limin1   

  1. 1 Chengdu Bohe Technology Co., Ltd.， Chengdu 610213, China
    2 Development Department, PetroChina Southwest Oil & Gasfield Company, Chengdu 610051, China
    3 Research Institute of Petroleum Exploration and Development, CNPC, Beijing 100083, China
  • Received:2025-08-12 Revised:2025-09-12 Online:2026-05-01 Published:2026-02-11

RichHTML

0

PDF (PC)

4

摘要：

针对微地震监测技术在页岩气压裂改造过程中存在弱信号识别困难、实时处理效率低以及微地震对裂缝响应弱等问题，基于震源扫描算法（SSA），融合多级塌陷网格搜索方法，提出了一种高精度微地震事件识别方法，并利用GPU并行计算技术实现了实时定位。研究结果表明：①通过波场能量扫描与叠加机制，对虚拟震源实施动校正与相干叠加强化P波弱信号特征，结合“粗网格全局扫描+细网格局部精搜”多级塌陷网格搜索策略提升定位效率与精度，引入GPU并行加速的窄带快速推进法（FMM）后，旅行时计算效率提升约16倍。②数值试验显示，低噪比条件下，多级塌陷网格搜索定位结果逼近真实震源，精度优于单一粗网格搜索。某钻井平台应用中，其与传统方法所得事件点均呈北东向条带状展布，事件深度延伸可达300~400 m，人工裂缝事件点更集中于井筒周围，能识别更多弱能量事件。③四川页岩气压裂施工监测中，可从多维度区分水力压裂裂缝与诱发天然裂缝特征，空间上，前者聚集于当前压裂段井筒附近、形态规则，后者远离压裂段、分布不规则且非对称；时间上，前者事件集中于施工期，后者具显著后效性，停泵后30分钟内仍持续产生；波形上，前者主频高、直达波清晰、结构简单，后者主频低、能量持续久、伴随多个续至波；相对震级上，前者普遍偏低，后者偏高且对应更大破裂尺度。结合相对震级与事件空间展布特征，可构建“天然裂缝响应带”与“压裂主控带”分布模型，为压裂效果评价及地质解释提供可靠依据。

关键词: 页岩气, GPU加速, 震源扫描算法, 多级网格搜索, 事件识别, 微地震定位, 裂缝响应, 压裂监测

Abstract:

Microseismic monitoring technology faces challenges, such as difficulty in weak signal identification, low real-time processing efficiency, and weak microseismic response to fractures in shale gas fracturing ope-rations. To address these issues, a high-precision microseismic event identification method was proposed based on source scanning algorithm (SSA), incorporating a multi-level collapse grid search approach. Real-time loca-lization was achieved using GPU parallel computing technology. The results show that:（1）Through the wavefield energy scanning and superposition mechanism, dynamic correction and coherent superposition are performed on virtual sources to enhance characteristics of weak P-wave signals. Combined with the multi-level collapsing grid search strategy of “coarse grid global scanning + fine grid local precise searching”, the positioning efficiency and accuracy are improved simultaneously. After introducing the GPU-parallelized narrow band fast marching method (FMM), the traveltime calculation efficiency is increased by about 16 times. （2） Numerical experiments show that under low signal-to-noise ratio conditions, the positioning result of the multi-level collapse grid search is close to the real source, and its accuracy is superior to that of the single coarse grid search. In the application of a drilling platform, event points obtained by this method and the conventional method show a northeastward zonal distribution, the event depth can extend up to 300-400 m, the hydraulic fracture event points are more concentrated around the wellbore, and more weak energy events can be identified. （3） In the monitoring of shale gas fracturing construction in Sichuan, characteristics of hydraulic fractures and induced natural fractures can be distinguished from multiple dimensions. Spatially, the former gathers near the wellbore of the current fracturing section with regular morphology, while the latter is far from the fracturing section with irregular and asymmetric distribution. Temporally, events of the former are concentrated during the fracturing construction period, while the latter has significant aftereffect and continues to occur within 30 minutes after pump shutdown. In terms of waveform, the former has high dominant frequency, clear direct wave and simple structure, while the latter has low dominant frequency, long energy duration and multiple subsequent waves. In terms of relative magnitude, the former is generally low, while the latter is high, corresponding to a larger fracture scale. Combined with the relative magnitude and spatial distribution characteristics of events, distribution models of “natural fracture response zone” and “fracturing main control zone” can be constructed, providing a reliable basis for fracturing effect evaluation and geological interpretation.

Key words: shale gas, GPU acceleration, source scanning algorithm (SSA), multi-level grid search, event identification, microseismic localization, fracture response, fracturing monitoring

中图分类号: 

  • TE02

图1

地面微地震监测系统和震源空间网格剖分示意图"

图2

塌陷网格搜索示意图 注：黄色圆点为一次剖分网格节点；黑色圆点为一次搜索出的能量最大的网格节点；红色圆点为二次剖分网格节点。"

图3

基于GPU并行加速的窄带快速推进法进行旅行时三维射线追踪效果"

图4

正演速度模型"

图5

基于正演速度模型合成地震记录"

图6

基于正演速度模型的微地震成像结果"

图7

四川盆地某钻井平台微地震监测观测到的时-空域道集"

图8

四川盆地某钻井平台微地震定位结果平面投影图 注：图中不同颜色的圆点代表着不同井、不同井段微地震监测事件点，圆点大小指示事件大小，圆点越大微地震事件震级越大；分段射孔簇中不同颜色的点表示不同的压裂段的射孔，相同颜色表示同一压裂段。"

图9

四川盆地某钻井平台强能量事件定位质控图"

图10

四川盆地某钻井平台中等能量事件定位质控图"

图11

四川盆地某钻井平台弱能量事件定位质控图"

图12

四川盆地某钻井平台典型井组诱发天然缝与水力压裂缝空间上的差异（平面投影图） 注：图中不同颜色的圆点代表着不同采集时间微地震监测事件点，而圆点大小指示微地震事件的震级大小；蓝色线为井轨迹，井轨迹中不同颜色的点指示不同压裂段的射孔，相同颜色为同一压裂段。"

图13

四川盆地某钻井平台典型井诱发天然缝与水力压裂缝空间上的差异（平面投影图） 注：图中不同颜色的圆点代表着不同压裂段微地震监测事件点，圆点大小指示微地震事件的震级大小；蓝色曲线为井轨迹，井轨迹中不同颜色的点指示不同压裂段的射孔，相同颜色为同一压裂段。"

图14

四川盆地某钻井平台诱发天然缝与水力压裂缝时间上的差异（平面投影图） 注：图中圆点大小指示微地震事件的震级大小；不同颜色的射孔簇指示不同的压裂段及射孔位置。"

图15

四川盆地某钻井平台压裂30分钟后诱发天然缝与水力压裂缝分布差异（平面投影图） 注：图中圆点大小指示微地震事件的震级大小；不同颜色的射孔簇指示不同的压裂段及射孔位置。"

图16

四川盆地某钻井平台诱发天然缝响应与水力压裂缝响应波形的差异"

图17

四川盆地某钻井平台诱发天然裂缝与三维裂缝预测叠合图（红色实线为微地震诱发天然缝延伸走向线） 注：图中不同颜色的圆点代表着不同压裂段的微地震监测事件点，圆点大小指示微地震事件的震级大小；NF1~NF11为天然裂缝；黑色线条为天然裂缝带；分段射孔簇图例中不同颜色表示不同的射孔段。"

[1] MAXWELL S. Microseismic:Growth born from success[J]. The Leading Edge, 2010, 29(3):338-343.
[2] BELYADI H, FATHI E, BELYADI F. Hydraulic fracturing in unconventional reservoirs:Theories,operations,and economic analysis[M]. 2nd ed. Cambridge: Gulf Professional Publishing, 2019.
[3] 罗志锋, 黄静云, 何天舒, 等. 碳酸盐岩储层酸压裂缝高度延伸规律:以川西栖霞组为例[J]. 岩性油气藏, 2020, 32(2):169-176.
LUO Zhifeng, HUANG Jingyun, HE Tianshu, et al. Extending regularity of fracture height by acid fracturing in carbonate re-servoir:A case study of Qixia Formation in western Sichuan[J]. Lithologic Reservoirs, 2020, 32(2):169-176.
[4] 杜金玲, 林鹤, 纪拥军, 等. 地震与微地震融合技术在页岩油压后评估中的应用[J]. 岩性油气藏, 2021, 33(2):127-134.
DU Jinling, LIN He, JI Yongjun, et al. Application of seismic and microseismic data fusion technology in postfracturing evaluation of shale oil reservoir[J]. Lithologic Reservoirs, 2021, 33(2):127-134.
[5] 胡俊杰, 宋毅, 朱炬辉, 等. 基于微地震监测的长宁区块页岩气水平井压裂效果评估[J]. 油气井测试, 2025, 34(1):35-42.
HU Junjie, SONG Yi, ZHU Juhui, et al. Evaluating hydraulic-fracture effectiveness in Changning shale gas block from microseismic monitoring[J]. Well Testing, 2025, 34(1):35-42.
[6] 赵超峰, 贾振甲, 田建涛, 等. 基于井中微地震监测方法的压裂效果评价:以吉林探区Y22井为例[J]. 岩性油气藏, 2020, 32(2):161-168.
ZHAO Chaofeng, JIA Zhenjia, TIAN Jiantao, et al. Fracturing effect evaluation based on borehole microseismic monitoring method:A case study from well Y22 in Jilin exploration area[J]. Lithologic Reservoirs, 2020, 32(2):161-168.
[7] 刘博, 徐刚, 纪拥军, 等. 页岩油水平井体积压裂及微地震监测技术实践[J]. 岩性油气藏, 2020, 32(6):172-180.
LIU Bo, XU Gang, JI Yongjun, et al. Practice of volume fracturing and microseismic monitoring technology in horizontal wells of shale oil[J]. Lithologic Reservoirs, 2020, 32(6):172-180.
[8] 刘晗, 张建中. 微震信号自动检测的STA/LTA算法及其改进分析[J]. 地球物理学进展, 2014, 29(4):1708-1714.
LIU Han, ZHANG Jianzhong. STA/LTA algorithm analysis and improvement of microseismic signal automatic detection[J]. Progress in Geophysics, 2014, 29(4):1708-1714.
[9] MA Ju, WU Shuang, ZHAO Yuan, et al. Cooperative P-wave velocity measurement with full waveform moment tensor inversion in transversely anisotropic media[J]. Sensors, 2022, 22(5):1-12.
[10] ROSS Z E, MEIER M A, HAUKSSON E, et al. Generalized seismic phase detection with deep learning[J]. Bulletin of the Seismological Society of America, 2018, 108(5A):2894-2901.
[11] LIU Lei, SONG Weiqi, ZENG Chao, et al. Microseismic event detection and classification based on convolutional neural network[J]. Journal of Applied Geophysics, 2021, 192:1-10.
[12] HAN Jiajun, VAN DER BAAN M. Microseismic and seismic denoising via ensemble empirical mode decomposition and adaptive thresholding[J]. Geophysics, 2015, 80(6):KS69-KS80.
[13] LI Huijian, WANG Runqiu, CAO Siyuan, et al. Weak signal detection using multiscale morphology in microseismic monitoring[J]. Journal of Applied Geophysics, 2016, 133:39-49.
[14] LIU K, LAW S S, XIA Y, et al. Singular spectrum analysis for enhancing the sensitivity in structural damage detection[J]. Journal of Sound and Vibration, 2014, 333(2):392-417.
[15] ZHANG Xingli, ZHAO Zhenhua, JIA Ruisheng, et al. Identification of microseismic signals based on multiscale singular spectrum entropy[J]. Shock and Vibration, 2020, 2020(4):1-12.
[16] 李文军, 李丽, 陈棋福. 用震源扫描算法(SSA)研究列车源的运动[J]. 地球物理学报, 2008, 51(4):1146-1151.
LI Wenjun, LI Li, CHEN Qifu. Research on the movement of vibration source of train by means of SSA[J]. Chinese Journal of Geophysics, 2008, 51(4):1146-1151.
[17] LI Lei, TAN Jingqiang, SCHWARZ B, et al. Recent advances and challenges of waveform‐based seismic location methods at multiple scales[J]. Reviews of Geophysics, 2020, 58(1):1-47.
[18] HOON K, SHAN Shaoju. The source-scanning algorithm:Mapping the distribution of seismic sources in time and space[J]. Geophysical Journal International, 2004, 157(2):589-594.
[19] ZOU An, LI Jing, GILL C D, et al. RTGPU:Real-time GPU scheduling of hard deadline parallel tasks with fine-grain utilization[J]. IEEE Transactions on Parallel and Distributed Systems, 2023, 34(5):1450-1465.
[20] 刘帅, 季晓慧, 芦俊, 等. CPU与GPU协同并行的多分量地震数据各向异性叠前时间偏移[J]. 石油地球物理勘探, 2019, 54(1):65-72.
LIU Shuai, JI Xiaohui, LU Jun, et al. Multi-component aniso-tropy prestack time migration based on collaborative parallel computing with CPU and GPU[J]. Oil Geophysical Prospec-ting, 2019, 54(1):65-72.
[21] XIE Jiangang, GUO Zichao, LIU Hai, et al. GPU acceleration of time gating based reverse time migration using the pseudospectral time-domain algorithm[J]. Computers and Geosciences, 2018, 117:57-62.
[22] 巴振宁, 赵靖轩, 吴孟桃, 等. 基于CPU-GPU异构并行的复杂场地近断层地震动谱元法模拟[J]. 地震学报, 2022, 44(1):182-193.
BA Zhenning, ZHAO Jingxuan, WU Mengtao, et al. Simulation of near-fault ground motions in complex sites based on CPU-GPU heterogeneous parallelism by spectral element method[J]. Acta Seismologica Sinica, 2022, 44(1):182-193.
[23] XUE Qingfeng, WANG Yibo, ZHAN Yi, et al. An efficient GPU implementation for locating micro-seismic sources using 3D elastic wave time-reversal imaging[J]. Computers and Geosciences, 2015, 82:89-97.
[24] 张猛. 基于GPU并行加速的黏声最小二乘逆时偏移及应用[J]. 岩性油气藏, 2022, 34(1):148-153.
ZHANG Meng. Least-squares reverse time migration in visco-acoustic medium based on GPU parallel acceleration and its app-lication[J]. Lithologic Reservoirs, 2022, 34(1):148-153.
[25] 徐田录, 吴承美, 张金凤, 等. 吉木萨尔凹陷二叠系芦草沟组页岩油储层天然裂缝特征与压裂模拟[J]. 岩性油气藏, 2024, 36(4):35-43.
XU Tianlu, WU Chengmei, ZHANG Jinfeng, et al. Natural fracture characteristics and fracture network simulation in shale reservoirs of Permian Lucaogou Formation in Jimsar Sag[J]. Lithologic Reservoirs, 2024, 36(4):35-43.
[26] GRIGOLI F, CESCA S, AMOROSO O, et al. Automated seismic event location by waveform coherence analysis[J]. Geophysical Journal International, 2014, 196(3):1742-1753.
[27] 赵忠华, 王成, 初海红, 等. 基于等效慢度的快速推进法求解三维地震波走时[J]. 地球物理学报, 2024, 67(10):3904-3914.
ZHAO Zhonghua, WANG Cheng, CHU Haihong, et al. 3D tra-veltime computation using the effective-slowness-based fast marching method[J]. Chinese Journal of Geophysics, 2024, 67(10):3904-3914.
[28] GHARTI H N, OYE V, ROTH M, et al. Automated microearthquake location using envelope stacking and robust global optimization[J]. Geophysics, 2010, 75(4):MA27-MA46.
[1] 郑马嘉, 刘勇, 伍亚, 陈俊宇, 陈颖, 钟梓月, 詹淋, 范存辉. 四川盆地西南部寒武系筇竹寺组页岩储层特征及成藏条件[J]. 岩性油气藏, 2026, 38(3): 79-93.
[2] 刘若彤, 张大智, 隋立伟, 肖利梅, 田亚, 孙山, 彭丹丹, 李建智. 川中合川地区志留系龙马溪组页岩岩相测井智能识别方法[J]. 岩性油气藏, 2026, 38(3): 94-106.
[3] 肖富强, 肖卫东, 姜智东, 高磊, 赵正威, 潘晓飞, 陈富贵, 邹勇军. 下扬子地区萍乐坳陷二叠系乐平组煤系页岩储层特征及勘探潜力[J]. 岩性油气藏, 2026, 38(1): 100-114.
[4] 陆江, 王健, 吴楠, 李程善, 冯子飞. 鄂尔多斯盆地西缘奥陶系乌拉力克组页岩气勘探潜力[J]. 岩性油气藏, 2025, 37(5): 34-48.
[5] 王振, 王兴志, 朱逸青, 杨雨然, 杨一茗, 康家豪, 黄柏文, 吕豪. 四川盆地德阳—安岳裂陷槽寒武系筇竹寺组地层划分及勘探意义[J]. 岩性油气藏, 2025, 37(5): 97-110.
[6] 詹淋, 范存辉, 唐雯, 杨西燕, 刘冬玺, 李博, 杨昕睿. 四川盆地东部南雅地区二叠系吴家坪组构造特征及其控藏机制[J]. 岩性油气藏, 2025, 37(5): 133-144.
[7] 杨雪, 杨雨然, 张景缘, 田鹤, 王青, 宋芳, 张赛柯, 陈瑶. 川北地区开江—梁平海槽二叠系海相页岩特征及优质储层形成机制[J]. 岩性油气藏, 2025, 37(3): 108-119.
[8] 罗冰, 文华国, 廖义沙, 张兵, 姚永君, 温思宇, 杨凯. 川东北地区二叠系吴家坪组二段页岩储层特征及有利区分布[J]. 岩性油气藏, 2025, 37(1): 1-12.
[9] 吴松, 冯冰, 于继良, 蓝宝锋, 李龙, 王胜, 沈家宁, 李刚权. 黔北正安地区安场向斜奥陶系五峰组—志留系龙马溪组页岩气富集规律[J]. 岩性油气藏, 2025, 37(1): 182-193.
[10] 闫建平, 来思俣, 郭伟, 石学文, 廖茂杰, 唐洪明, 胡钦红, 黄毅. 页岩气井地质工程套管变形类型及影响因素研究进展[J]. 岩性油气藏, 2024, 36(5): 1-14.
[11] 杨学锋, 赵圣贤, 刘勇, 刘绍军, 夏自强, 徐飞, 范存辉, 李雨桐. 四川盆地宁西地区奥陶系五峰组—志留系龙马溪组页岩气富集主控因素[J]. 岩性油气藏, 2024, 36(5): 99-110.
[12] 包汉勇, 赵帅, 张莉, 刘皓天. 川东红星地区中上二叠统页岩气勘探成果及方向展望[J]. 岩性油气藏, 2024, 36(4): 12-24.
[13] 申有义, 王凯峰, 唐书恒, 张松航, 郗兆栋, 杨晓东. 沁水盆地榆社—武乡区块二叠系煤系页岩储层地质建模及“甜点”预测[J]. 岩性油气藏, 2024, 36(4): 98-108.
[14] 程静, 闫建平, 宋东江, 廖茂杰, 郭伟, 丁明海, 罗光东, 刘延梅. 川南长宁地区奥陶系五峰组—志留系龙马溪组页岩气储层低电阻率响应特征及主控因素[J]. 岩性油气藏, 2024, 36(3): 31-39.
[15] 段逸飞, 赵卫卫, 杨天祥, 李富康, 李慧, 王嘉楠, 刘钰晨. 鄂尔多斯盆地延安地区二叠系山西组页岩气源储特征及聚集规律[J]. 岩性油气藏, 2024, 36(3): 72-83.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!

陇ICP备17006252号
版权所有© 2018 中国石油集团西北地质研究所有限公司《岩性油气藏》编辑部

地址：甘肃省兰州市城关区雁儿湾路535号(陇ICP备17006252号)    电话：0931-8686158;8686058;8686288    邮箱：yxyqc@petrochina.com.cn

甘公网安备 62010202002827号

本系统由北京玛格泰克科技发展有限公司设计开发