基于CUDA的地震信号频率补偿并行算法

doi:10.13810/j.cnki.issn.1000-7210.2022.05.025

摘要
图/表
参考文献
相关文章 (15)

全文: PDF (2532 KB) HTML []
输出: BibTeX | EndNote (RIS)

摘要基于压缩感知的地震信号频率补偿算法可有效拓宽地震信号的频谱，提高地震资料的分辨率。虽然该算法具有良好的拓频效果，但对于高维度、大规模的地震数据时效较低。经过分析发现该算法的计算瓶颈在于计算反射系数部分的大量代数运算和重构信号部分的卷积运算，为此，提出一种基于CUDA的并行方案对该算法进行并行优化。首先，改变地震数据的组织形式，使其存取效率更高，且更适合并行处理；然后，重新设计计算反射系数串行代码，利用CUDA(Compute Unified Device Architecture)平台调用GPU大量轻量级线程对其中的代数运算进行并行化；最后，利用卷积定理改变了时域信号卷积计算方式，采用cufft库函数将两个时域信号的卷积转换到频域进行计算。结果表明，在保证计算精度的前提下，与串行算法相比，并行算法在PC端获得了4倍以上的总体加速比。

	服务

	把本文推荐给朋友
	加入我的书架
	加入引用管理器
	E-mail Alert
	RSS
	作者相关文章
	张全
	张杰明
	雷芩
	彭博
	刘书妍

关键词 ：频率补偿, CUDA, 并行计算, 压缩感知

Abstract：The frequency compensation algorithm of seismic signals based on compressed sensing can effectively broaden the spectrum of seismic signals and improve the resolution of seismic data. Although the algorithm has a good frequency-broadening effect, it has low time efficiency in the face of high-dimensional and large-scale seismic data. Analysis shows that the bottleneck of the algorithm lies in massive algebraic operations for the reflection coefficient and convolution operations in signal reconstruction. Therefore, a parallel scheme based on CUDA is proposed for parallel optimization of the algorithm. Firstly, the organization form of seismic data is changed to make it more efficient and more suitable for parallel processing. Then, the serial code for computing the reflection coefficient is reconstructed, and a large number of lightweight threads of GPU are called by CUDA to parallelize the algebraic operations. Finally, the convolution calculation method of time-domain signals is changed by the convolution theorem, and the convolution operation of two time-domain signals is converted to the frequency domain by the cufft library function. The results reveal that the parallel algorithm achieves four times the overall speedup of the serial algorithm on the PC side on the pre-mise of ensuring computational accuracy.

Key words： frequency compensation CUDA parallel computing compressed sensing

收稿日期: 2021-10-12

PACS:

P631

基金资助:本项研究受油气藏地质及开发工程国家重点实验室开放基金项目“基于高性能计算与卷积神经网络的地震多次波压制方法研究”(PLN2021-21)和“基于联合深度学习的地震多次波压制方法研究”(PLN2021-25)联合资助

通讯作者: 彭博,四川省成都市新都区新都大道8号西南石油大学(成都校区)计算机科学学院，610500。Email:bopeng@swpu.edu.cn E-mail: bopeng@swpu.edu.cn

作者简介: 张全,讲师，硕士生导师，1985年生；2007年获西南石油大学计算机科学与技术专业学士学位，2011年获湘潭大学计算机软件与理论专业硕士学位，2015年获电子科技大学信息与通信工程专业工学博士学位；现为中国计算机学会会员，就职于西南石油大学计算机科学学院，主要从事地震信号处理和高性能计算等领域的教研

引用本文:

张全, 张杰明, 雷芩, 彭博, 刘书妍. 基于CUDA的地震信号频率补偿并行算法[J]. 石油地球物理勘探, 2022, 57(5): 1241-1249. ZHANG Quan, ZHANG Jieming, LEI Qin, PENG Bo, LIU Shuyan. Parallel algorithm of seismic signal frequency compensation based on CUDA. Oil Geophysical Prospecting, 2022, 57(5): 1241-1249.

链接本文:

http://www.ogp-cn.com/CN/10.13810/j.cnki.issn.1000-7210.2022.05.025 或 http://www.ogp-cn.com/CN/Y2022/V57/I5/1241

[1]	管路平,唐权钧. 地震信号的高低频成分补偿[J]. 石油物探,1990,29(3):35-45.GUAN Luping,TANG Quanjun. High/low frequency compensation of seismic signal[J]. Geophysical Prospecting for Petroleum,1990,29(3):35-45.
[2]	谷玉田,魏继东,张旭,等. 陆上勘探低频信号的接收与恢复[J]. 石油物探,2021,60(4):539-545.GU Yutian,WEI Jidong,ZHANG Xu,et al. Sensing and recovery of low-frequency signals in onshore surveys[J]. Geophysical Prospecting for Petroleum,2021,60(4):539-545.
[3]	ZHI L X,CHEN S Q,LI X Y. Amplitude variation with angle inversion using the exact Zoeppritz equa-tions:Theory and methodology[J]. Geophysics,2016,81(2):N1-N15.
[4]	ZHANG Junhua,ZHANG Binbin,ZHANG Zaijin,et al. Low-frequency data analysis and expansion[J]. Applied Geophysics,2015,12(2):212-220,275.
[5]	KROODE F T,BERGLER S,CORSTEN C,et al. Broadband seismic data:The importance of low fre-quencies[J]. Geophysics,2013,78(2):WA3-WA14.
[6]	张军华,张在金,张彬彬,等. 地震低频信号对关键处理环节的影响分析[J]. 石油地球物理勘探,2016,51(1):54-62.ZHANG Junhua,ZHANG Zaijin,ZHANG Binbin,et al. Low frequency signal influences on key seismic data processing procedures[J]. Oil Geophysical Prospecting,2016,51(1):54-62.
[7]	魏继东. 检波器反褶积对低频信息的补偿作用[J]. 石油地球物理勘探,2016,51(2):224-231.WEI Jidong. Geophone deconvolution low-frequency compensation for seismic data[J]. Oil Geophysical Prospecting,2016,51(2):224-231.
[8]	TONN R. The determination of the seismic quality factor Q from VSP data:a comparison of different computational methods[J]. Geophysical Prospec-ting,1991,39(1):1-27.
[9]	段中钰,李婷婷,肖勇,等. 基于压缩感知的SR-ADMM地震数据重建[J]. 石油地球物理勘探,2021,56(6):1220-1228.DUAN Zhongyu,LI Tingting,XIAO Yong,et al. Seismic data Reconstruction by SR-ADMM algorithm based on compressed sensing[J]. Oil Geophysical Prospecting,2021,56(6):1220-1228.
[10]	张军华,王静,王延光,等. 基于压缩感知的反射系数域沿层L2范数约束去强屏蔽方法[J]. 石油地球物理勘探,2022,57(2):405-413.ZHANG Junhua,WANG Jing,WANG Yanguang,et al. A strong shielding removal of reflection coefficient domain based on compressed sensing with L2 norm constraint along layer[J]. Oil Geophysical Prospecting,2022,57(2):405-413.
[11]	闫海洋,周辉,刘海波,等. FK和Shearlet域联合压缩感知数据重构技术[J]. 石油地球物理勘探,2022,57(3):557-569,491.YAN Haiyang,ZHOU Hui,LIU Haibo,et al. Com-pressed sensing data reconstruction technology in joint FK and Shearlet domain[J]. Oil Geophysical Prospecting,2022,57(3):557-569,491.
[12]	CANDōS E J. Compressive sampling[C]. Procee-dings on the International Congress of Mathemati-cians,2006,1433-1452.
[13]	韩红平. 压缩感知中信号重构算法的研究[D]. 江苏南京:南京邮电大学,2012.HAN Hongping. Research on Signal Reconstruction Algorithm of Compressive Sense[D]. Nanjing Uni-versity of Posts and Telecommunications,Nanjing,Jiangsu,2012.
[14]	韩立国,张莹,韩利,等. 基于压缩感知和稀疏反演的地震数据低频补偿[J]. 吉林大学学报(地球科学版),2012,42(增刊3):259-264.HAN Liguo,ZHANG Ying,HAN Li,et al. Com-pressed sensing and sparse inversion based low-frequency information compensation of seismic data[J]. Journal of Jilin University (Earth Science Edition),2012,42(S3):259-264.
[15]	张莹. 基于压缩感知和稀疏反演的低频补偿方法研究[D]. 吉林长春:吉林大学,2013.ZHANG Ying. A Study on Low-frequency Compen-sation Based on Compressed Sensing and Sparse In-version[D]. Jilin University,Changchun,Jilin,2013.
[16]	丁燕,杜启振,刘力辉,等. 基于压缩感知和宽带俞式低通整形滤波器的地震低频信息特征分析与补偿[J]. 地球物理学报,2019,62(6):2267-2275.DING Yan,DU Qizhen,LIU Lihui,et al. Feature analysis and compensation of seismic low-frequency based on compressed sensing and broad-band Yu-type low-passing shaping filter[J]. Chinese Journal of Geophysics,2019,62(6):2267-2275.
[17]	张军华,臧胜涛,单联瑜,等. 高性能计算的发展现状及趋势[J]. 石油地球物理勘探,2010,45(6):918-925.ZHANG Junhua,ZANG Shengtao,SHAN Lianyu,et al. Development status and trends for high per-formance computing[J]. Oil Geophysical Prospecting,2010,45(6):918-925.
[18]	赵虎,武泗海,尹成,等. 基于OpenACC编程模型的逆时偏移多级并行的设计与优化[J]. 石油地球物理勘探,2018,53(6):1307-1313,1325.ZHAO Hu,WU Sihai,YIN Cheng,et al. Multi-level parallel design and optimization for reverse time mi-gration based on OpenACC programming model[J]. Oil Geophysical Prospecting,2018,53(6):1307-1313,1325.
[19]	张全,林柏栎,杨勃,等. CPU-GPU异构平台的抛物线Radon变换并行算法[J]. 石油地球物理勘探,2020,55(6):1263-1270.ZHANG Quan,LIN Baiyue,YANG Bo,et al. Parabolic Radon transform parallel algorithm for CPU-GPU heterogeneous platform[J]. Oil Geophysical Prospecting,2020,55(6):1263-1270.
[20]	张全,林柏栎,彭博,等. 基于CUDA的地震相干体并行算法[J]. 地质与勘探,2020,56(1):147-153.ZHANG Quan,LIN Baiyue,PENG Bo,et al. Seismic coherence parallel algorithm based on CUDA[J]. Geology and Exploration,2020,56(1):147-153.
[21]	吴吉忠,左虎. 叠前衰减补偿时间偏移及GPU实现[J]. 石油地球物理勘探,2019,54(1):84-92.WU Jizhong,ZUO Hu. Attenuation compensation in prestack time migration and its GPU implementation[J]. Oil Geophysical Prospecting,2019,54(1):84-92.
[22]	陆基孟.地震勘探原理[M]. 北京:石油工业出版社,1982.LU Jimeng. Principles of Seismic Exploration[M]. Petroleum Industry Press,Beijing,1982.
[23]	隋宇涵. 地震时变子波估计与盲稀疏脉冲反褶积[D]. 黑龙江哈尔滨:哈尔滨工业大学,2020.SUI Yuhan. Seismic Time-Varying Wavelet Estimation and Blind Sparse-Spike Deconvolution[D]. Harbin Institute of Technology,Harbin,Heilongjiang,2020.
[24]	戴永寿,张彧豪,张鹏,等. 时变地震子波提取研究方法综述[J]. 石油物探,2020,59(2):169-176.DAI Yongshou,ZHANG Yuhao,ZHANG Peng,et al. A review on time-varying seismic wavelet extraction[J]. Geophysical Prospecting for Petroleum,2020,59(2):169-176.
[25]	BECK A,TEBOULLE M. A fast iterative Shrinkage-Thresholding algorithm for linear inverse problems[J]. SIAM Journal on Imaging Sciences,2009,2(1):183-202.
[26]	李庆扬,王能超,易大义. 数值分析[M]. 第5版. 北京:清华大学出版社,2008.LI Qingyang,WANG Nengchao,YI Dayi. Numerical Analysis[M]. 5th Edition. Tsinghua University Press,Beijing,2008.
[27]	周竹生,何继善,赵荷晴. 利用广义共轭梯度算法求解地震道反演问题[J]. 石油地球物理勘探,1998,33(4):439-447,476-572.ZHOU Zhusheng,HE Jishan,ZHAO Heqing. Solving a seismic trace inversion problem by using gene-ralized conjugate gradient algorithm[J]. Oil Geophy-sical Prospecting,1998,33(4):439-447,476-572.
[28]	刘宇航,黄建平,杨继东,等. 弹性波全波形反演中的四种优化方法对比[J]. 石油地球物理勘探,2022,57(1):118-128.LIU Yuhang,HUANG Jianping,YANG Jidong,et al. Comparison of four optimization methods in elastic full-waveform inversion[J]. Oil Geophysical Pro-specting,2022,57(1):118-128.