K均值奇异值分解（K-SVD）算法是一种行之有效的地震资料去噪方法,但由于其稀疏分解存在不确定性,需要引入正则项对其改进。为此,在常规粒子群算法的基础上,提出了一种自适应动态粒子群算法优化正则化参数的正则化近似K-SVD（RAK-SVD）去噪方法。首先通过修改字典原子和相关参数,解决了由于常规粒子群算法的惯性参数固定不变,导致后期搜索效率下降的问题；其次将正则化系数引入近似K-SVD（AK-SVD）方法,明显提升了去噪效果；最后利用自适应动态粒子群算法自动优选AK-SVD方法中的正则化参数,提高了稀疏分解的确定性,在对强反射信号进行去噪的同时加强了对弱信号的保护。模型测试和实际应用均表明,该方法有利于弱信号的提取和识别,不仅能够显著改善弱地震信号的去噪效果,还提升了计算效率。该方法具有一定的实际应用价值。
Abstract
The K-means singular value decomposition （K-SVD） algorithm is an effective seismic data denoising method. However, due to the uncertainty problem of its sparse decomposition, it is necessary to be improved by introducing regularization terms. Therefore, a regularization approximation K-SVD （RAK-SVD） denoising method for optimizing regularization parameters by using an adaptive dynamic particle swarm optimization algorithm based on a conventional particle swarm optimization algorithm was proposed. Firstly, by modifying the dictionary atoms and related parameters, the problem of decreased search efficiency in the later stage due to the fixed inertia parameters of the conventional particle swarm optimization algorithm was solved. Secondly, regularization coefficients were introduced into the approximate K-SVD method, which significantly improved the denoising effect. Finally, the adaptive dynamic particle swarm optimization algorithm was used to automatically optimize the regularization parameters in the AK-SVD method, which improved the determinacy of sparse decomposition and enhanced the protection of weak signals while denoising strong reflection signals. Model tests and practical applications have shown that this method is beneficial for extracting and identifying weak signals. It can not only significantly improve the denoising effect of weak seismic signals but also enhance computational efficiency. This method has certain practical application value.
关键词
自适应动态粒子群算法 /
K-SVD字典 /
正则化 /
去噪
{{custom_keyword}} /
Key words
adaptive dynamic particle swarm algorithm /
K-SVD dictionary /
regularization /
denoising
{{custom_keyword}} /
中图分类号：
P631
{{custom_clc.code}}
({{custom_clc.text}})
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] MU Xing. Seismic weak signal separation based on blind signal processing[J]. Petroleum Geology and Recovery Efficiency, 2012, 19(5):47-49.
穆星. 基于盲信号处理技术的地震弱信号分离方法[J]. 油气地质与采收率, 2012, 19(5):47-49.
[2] XU Yankai, SHUANG Kai. Detection of transient weak signal based on wavelet transform and singular value decomposition[J]. Journal of China University of Petroleum(Edition of Natural Science), 2014, 38(3):181-185.
徐彦凯, 双凯. 小波奇异值分解的瞬变弱信号检测[J]. 中国石油大学学报(自然科学版), 2014, 38(3):181-185.
[3] DUAN Xiangyang, WANG Yongsheng, SU Yong-sheng. Feature extraction methods based on singular value decomposition[J]. Journal of Vibration and Shock, 2009, 28(11):30-33.
段向阳, 王永生, 苏永生. 基于奇异值分解的信号特征提取方法研究[J]. 振动与冲击, 2009, 28(11):30-33.
[4] LI Yue, YANG Baojun, ZHAO Xueping, et al. An algorithm of chaotic vibrator to detect weak events in seismic prospecting records[J]. Chinese Journal of Geophysics, 2005, 48(6):1428-1433.
李月, 杨宝俊, 赵雪平, 等. 检测地震勘探微弱同相轴的混沌振子算法[J]. 地球物理学报, 2005, 48(6):1428-1433.
[5] BAI Dawei, DU Bingrui, ZHANG Penghui. Weak signal processing technology of low frequency GPR based on Hilbert-Huang transform and its application to permafrost area gas hydrate exploration[J]. Geophysical and Geochemical Exploration, 2017, 41(6):1060-1067.
白大为, 杜炳锐, 张鹏辉. 基于希尔伯特-黄变换的低频探地雷达弱信号处理技术及其在天然气水合物勘探中的应用[J]. 物探与化探, 2017, 41(6):1060-1067.
[6] LI Jianfeng. Research and Application on the Method of Seismic Signal Processing Based on Blind Source Separation[D]. University of Petroleum(East China), Qingdao, Shandong, 2012.
李建锋. 基于盲源分离的地震信号处理方法研究及应用[D]. 山东青岛:中国石油大学(华东), 2012.
[7] AHARON M, ELAD M, BRUCKSTEIN A. K-SVD:an algorithm for designing overcomplete dictionaries for sparse representation[J]. IEEE Transactions on Signal Processing, 2006, 54(11):4311-4322.
[8] RUBINSTEIN R, ZIBULEVSKY M, ELAD M. Efficient implementation of the K-SVD algorithm using batch orthogonal matching pursuit:CS-2008-08[R]. Israel Institute of Technology, Haifa, Israel, 2008.
[9] IROFTI P, DUMITRESCU B. Regularized algorithms for dictionary learning[C].2016 International Conference on Communications (COMM), Bucharest, Romania, June2016, 439-442.
[10] DUMITRESCU B, IROFTI P. Regularized K-SVD[J]. IEEE Signal Processing Letters, 2017, 24(3):309-313.
[11] HANSEN P C. Analysis of discrete ill-posed problems by means of the L-curve[J]. SIAM Review, 1992, 34(4):561-580.
[12] XU Y B, PEI Y, DONG F. An extended L-curve method for choosing a regularization parameter in electrical resistance tomography[J]. Measurement Science and Technology, 2016, 27(11):114002.
[13] NIE Duxian, YUAN Liguo, WEN Youwei. App-lying particle swarm optimization algorithm to select regularization parameter[J]. Computer Engineering and Applications, 2009, 45(12):195-197.
聂笃宪, 袁利国, 文有为. 应用粒子群优化算法选择正则化参数[J]. 计算机工程与应用, 2009, 45(12):195-197.
[14] YU Ruiyan. Optimal choosing of Tikhonov regularization parameter based on chaos particle swarm optimization algorithm[J]. Journal of Mathematical Study, 2011, 44(1):101-106.
余瑞艳. 基于混沌粒子群算法的Tikhonov正则化参数选取[J]. 数学研究, 2011, 44(1):101-106.
[15] XU Longxiu, XIN Chaoshan, NIU Dongxiao, et al. Power consumption prediction model of least squares support vector machine based on adaptive particle swarm optimization[J]. Science Technology and Engineering, 2019, 19(6):136-141.
徐龙秀, 辛超山, 牛东晓, 等. 基于自适应粒子群参数优化的最小二乘支持向量机用电量预测模型[J]. 科学技术与工程, 2019, 19(6):136-141.
[16] LIAO Qingling, DOU Zhenhai, SUN Kai, et al. Load forecasting based on adaptive particle swarm optimization algorithm optimizing support vector machine[J]. Modern Electronics Technique, 2022, 45(3):125-129.
廖庆陵, 窦震海, 孙锴, 等. 基于自适应粒子群算法优化支持向量机的负荷预测[J]. 现代电子技术, 2022, 45(3):125-129.
{{custom_fnGroup.title_cn}}
脚注
{{custom_fn.content}}
PDF(17668 KB)
