Time-varying wavelet extraction based on time-frequency analysis and adaptive segmentation
Wang Rongrong1, Dai Yongshou2, Li Chuang3, Zhang Peng2, Tan Yongcheng2
1. Hisense(Shandong) refrigerator Co. Ltd., Qingdao, Shandong 266736, China;
2. College of Information and Control Engineering, China University of Petroleum(East China), Qingdao, Shandong 266580, China;
3. School of Geosciences, China University of Petroleum(East China), Qingdao, Shandong 266580, China
Abstract:There are two effective methods of time-varying seismic wavelet extraction. First, after dividing the non-stationary seismogram into several time windows, the seismogram in each window can be treated as stationary one and then the seismic wavelets are extracted piecewise. Second, the non-stationary seismogram is transformed into time-frequency domain and the time-varying wavelets are extracted at each time point. To solve the problems that both two methods are difficult to extract the accurate amplitude and phase spectra of time-varying wavelets simultaneously, this paper presents a time-varying wavelet extraction method based on time-frequency analysis and adaptive segmentation, in which amplitude spectra are extracted by quadratic spectrum modeling in time-frequency domain and segmented phase spectra are extracted by adaptive molecular decomposition method. In order to solve the problem of matching time-varying amplitude spectra and segmented phase spectra, we extent the segmented phase spectra to every time point, and then the time-varying wavelets can be estimated completely and accurately by combining the amplitude and phase spectra at each time point. The reflection coefficient sequence can be inversed from non-stationary seismogram by using extracted time-varying wavelets and evaluated based on evaluation criteria. In this way, the accuracy of time-varying wavelet extraction can be evaluated quantitatively. Simulation and real data processing results show that, taking the distortion of amplitude and phase into consideration when the seismic signal spreads in underground, the proposed method can adapt well to the non-stationary nature of actual seismic data and obtain time-varying seismic wavelets accurately.
王蓉蓉, 戴永寿, 李闯, 张鹏, 谭永成. 时频分析与自适应分段相结合的时变子波提取方法[J]. 石油地球物理勘探, 2016, 51(5): 850-862.
Wang Rongrong, Dai Yongshou, Li Chuang, Zhang Peng, Tan Yongcheng. Time-varying wavelet extraction based on time-frequency analysis and adaptive segmentation. OGP, 2016, 51(5): 850-862.
Dai Yongshou,Wang Junling,Wang Weiwei et al.Seismic wavelet extraction via cumulant based ARMA model approach with linear and nonlinear combination.Chinese Journal of Geophysics,2008,51(6):1851-1859.
Liu Xiwu,Liu Hong,Li Youming et al.Local waves decomposition and its applications to seismic signal time2frequency analysis.Progress in Geophysics,2007,22(2):365-375.
Gao Jinghuai,Wang Lingling,Zhao Wei.Enhancing resolution of seismic traces based on the changing wavelet model of the seismogram.Chinese Journal of Geophysics, 2009,52(5):1289-1300.
Lu Ziwei,Zhong Xiuhao,Zhang Yan et al.Feature point extracting and fitting method based on scanning input engineering curves.Journal of Liaoning Shihua University, 2014,34(1):65-68.
[13]
袁三一,王尚旭.子波相位不准对反演结果的影响.应用地球物理,2011,8(1):48-59.
Yuan Sanyi,Wang Shangxu.Influence of inaccurate wavelet phase estimation on seismic inversion.Applied Geophysics, 2011,8(1):48-59.
[14]
王有新,周兴元.最小熵反褶积和集约化准则.石油地球物理勘探,1993,28(6):678-684.
Wang Youxin,Zhou Xingyuan.Minimum entropy deconvelution and parsimonious criterion.OGP,1993,28(6):678-684.
He Tonglin,Zheng Peng,Liu Yulin et al.Nonlinear blind deconvolution based on Kurtosis criterion and decision directed algorithm.Journal of University of Electronic Science and Technology of China,2007,36(2):186-189.
[17]
程乾生.信号数字处理的数学原理.北京:石油工业出版社, 1993.
[18]
Levy S,Oldenburg D W.Automatic phase correction of common-midpoint stacked data.Geophysics,1987,52(1):51-59.
[19]
Wiggins R A.Minimum entropy deconvolution. Geoexploration,1978,16(1):21-35.
[20]
Longbottom J,Walden A T,White R E.Principles and application of maximum kurtosis phase estimation. Geophysical Prospecting,1988,36(2):115-138.
[21]
White R E.Maximum kurtosis phase correction.Geophysical Journal,1988,95(2):371-389.