基于自适应动态粒子群优化的RAK-SVD方法  

RAK⁃SVD method based on adaptive dynamic particle swarm optimization

作　　者：乐友喜[1] 姚晓辰 付俊楠 葛传友 YUE Youxi;YAO Xiaochen;FU Junnan;GE Chuanyou(School of Geosciences,China University of Petroleum(East China),Qingdao,Shandong 266580,China)

机构地区：[1]中国石油大学(华东)地球科学与技术学院,山东青岛266580

出　　处：《石油地球物理勘探》2024年第3期494-503,共10页Oil Geophysical Prospecting

摘　　要：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。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 algo-rithm 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,regu-larization 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 automati-cally 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 sig-nals.It can not only significantly improve the denoising effect of weak seismic signals but also enhance computa-tional 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 

分 类 号：P631[天文地球—地质矿产勘探]