15 August 2024, Volume 59 Issue 4

    

• Select all
  |

INTELLIGENT GEOPHYSICAL TECHNIQUE
• Select
  v_P/v_S prediction based on XGBoost algorithm and its application in reservoir detection

  TIAN Renfei, LI Shan, LIU Tao, JING Yang

  Oil Geophysical Prospecting. 2024, 59(4): 653-663. https://doi.org/10.13810/j.cnki.issn.1000-7210.2024.04.001

  Abstract ( )   Knowledge map   Save

  There are abundant oil and gas resources entrapped in the carbonate reservoirs of the Ordos Basin. However， exploration results showed that the Majiagou Formation in the Daniudi Gas Field had developed multiple kinds of faults with small fault throws due to complex origins， which brings many challenges to its exploration and development. To address these challenges， it is crucial to optimize the sensitive elastic parameters for reservoir prediction. Therefore， the relationship between seismic attributes and the velocity ratio of compressional to shear waves （v_P/v_S） in the reservoir has been established， based on the analysis of elastic-sensitive parameters in the Daniudi Gas Field. Then， a prediction method for the v_P/v_S based on the XGBoost algorithm and multiple seismic attributes is proposed. To further improve the performance and generalization ability of the model， the hyperparameters of the XGBoost algorithm are optimized by Bayesian algorithm. This approach aims to find the optimal combination of hyperparameters， ensuring improved performance of the model on both training and testing datasets. The XGBoost algorithm is applied to the Marmousi 2 model for predicting shear wave velocity， achieving a correlation coefficient between predicted and actual values exceeding 0.88. With root mean squared error and mean absolute percentage error below 6.55×10^-7 and 4% respectively， the accuracy and reliability of the proposed method are demonstrated. The method applied in the Daniudi Gas Field of the Ordos Basin has successfully identified gas-bearing reservoirs， and the results are consistent with actual dril-ling data. Both theoretical model and practical data indicate that XGBoost， as a powerful machine learning algorithm， exhibits high accuracy， which can provide an effective approach for directly predicting v_P/v_S from post-stack seismic attributes.
• Select
  Multiple attenuation algorithm based on improved pix2pix GAN network

  ZHANG Quan, LYU Xiaoyu, LEI Qin, HUANG Yixuan, PENG Bo, LI Yan

  Oil Geophysical Prospecting. 2024, 59(4): 664-674. https://doi.org/10.13810/j.cnki.issn.1000-7210.2024.04.002

  Abstract ( )   Knowledge map   Save

  The effective attenuation of seismic multiples plays a crucial role in the seismic data processing workflow. Despite the existence of numerous multiple attenuation methods， traditional approaches heavily rely on prior geological structure information and require extensive calculations， resulting in slow attenuation speed. This poses an even greater challenge for multiple attenuation under complex geological conditions. To overcome the limitations of traditional methods and improve efficiency， this paper applies the pix2pix GAN network to the problem of multiple attenuation and utilizes the feature learning capability of neural networks to improve the processing speed. It proposes an enhanced multiple attenuation method for the pix2pix GAN network， which integrates ResNet and U-Net as the network generator to avoid gradient vanishing or exploding phenomena used by deep netwoorks， while incorporating the SE attention mechanism. The improved generator can better perceive the characteristics of both first-order and multiples， thereby enhancing its performance. Additionally， a multi-scale discriminator is employed to discern detailed features and texture information on finer seismic images for accurate identification of authenticity. The input data for the network consists of full wave field data labeled as primary wave data， with training conducted using a dataset synthesized from two simple formation models and a public Sigbee2B model. Experimental results demonstrate that the improved GAN network exhibits superior accuracy in multiple attenuation compared to pix2pix GAN， effectively improving attenuation speed.
• Select
  High-resolution processing of seismic data using adaptive attention mechanism U-net

  ZHAO Ming, ZHAO Yan, SHEN Donghao, WANG Jianqiang, DAI Xiancai

  Oil Geophysical Prospecting. 2024, 59(4): 675-683. https://doi.org/10.13810/j.cnki.issn.1000-7210.2024.04.003

  Abstract ( )   Knowledge map   Save

  With the deepening of oil and gas exploration and development， thin and lithologic reservoirs have gradually become important exploration targets， which also leads to higher requirements for the resolution of seismic data. This paper presents a high-resolution seismic data processing method of U-net based on an adaptive attention mechanism. This method first uses the U-net structure to learn the feature representation of seismic data， extracts the abstract features of seismic data through the encoder of the down-sampling process， and then reconstructs and refines the features through the decoder of the up-sampling process. The attention mechanism is introduced in the process of up-sampling， which is utilized to adjust the attention of the network to diffe-rent seismic features. Therefore， the network can capture more details and features of the seismic data more effectively. The experimental results of synthetic seismic records of the Marmousi model and real data show that the new network has less error and is more stable than the original U-net， as it can effectively improve the prediction accuracy and realize the high-resolution processing of seismic data.
• Select
  Noise suppression of pumping unit based on multi-scale window generator network

  MA Yifan, WEN Wu, XUE Yajuan, WEN Xiaotao, XU Hong

  Oil Geophysical Prospecting. 2024, 59(4): 684-691. https://doi.org/10.13810/j.cnki.issn.1000-7210.2024.04.004

  Abstract ( )   Knowledge map   Save

  The noise of the pumping unit strongly interferes with the exploration and development of old oil fields and seriously reduces the signal-to-noise ratio of seismic data. Therefore， a pumping unit noise suppression method based on a multi-scale window generator network is proposed. The constructed network is mainly composed of a double-layer encoder-decoder structure， and accurate denoising results can be obtained by combining characteristic information of different layers. The utilization of different-sized windows in different layers for feature extraction can effectively expand the sensing range of the neural network and extract more useful features from the pumping unit noise. To prevent the degradation of the network， residual connections are used respectively in each block of the encoder and decoder. The residual block of the encoder adopts the anti-bottleneck design with a large amount of convolution kernels in the middle and small at both ends to extract more features of seismic data. The decoder uses one-fifth of the convolutional layers of the encoder， speeding up model training and seismic data reconstruction. The network constructed in this way can effectively suppress pumping unit noise in seismic data by using multi-scale semantic information. Both simulated data and real data experimental results show that compared with DnCNN， GAN， and MLGNet， the proposed method can obtain high-quality denoising results and retain valid data to the greatest extent.
• Select
  Obtaining variation function parameters in modeling geological attribute based on U-Net and CNN deep learning

  FENG Gouqing, MO Haishuai, WU Baofeng

  Oil Geophysical Prospecting. 2024, 59(4): 692-701. https://doi.org/10.13810/j.cnki.issn.1000-7210.2024.04.005

  Abstract ( )   Knowledge map   Save

  In the modeling of geological attribute of oil and gas reservoirs， obtaining the variation function is especially critical， which is generally obtained by fitting the experimental variation function to acquire parameters such as varve， azimuth， and abutment value. However， when the number of sample points in the study area is insufficient， it will lead to a poor fitting effect， thereby affecting the quality of attribute modeling. To overcome the shortcomings of traditional experimental variation function modeling and make the most use of spatial data， this paper proposes a new method based on U-Net and CNN networks to predict the parameters of the variation function. The data points extracted from the porosity plane model obtained by sequential Gaussian simulation are taken as the benchmark. Using the U-Net network structure， the porosity distribution is reconstructed to maintain spatial correlation. Subsequently， a CNN network structure is applied to the sample set for deep lear- ning， thereby developing a model to predict the variation function. The practical application shows that the principal range direction obtained by the proposed method in this paper deviates by only 1.52°from that obtained using the experimental range function， closely matching the distribution direction of sedimentary microfacies. Meanwhile， the obtained principal and secondary ranges closely align with the experimental variation function， confirming the reliability of the model’s variation function results. At the same time， the method also simplifies the geological modeling workflow， reduces the subjectivity of finding the experimental variation function， and reduces the limitations posed by a small number of data points in the study area. It offers a novel approach for the predictive research of the variation function.
• Select
  High-resolution seismic data processing method based on supervised deep learning

  LI Fei, NIU Wenli, LIU Dawei, WANG Yonggang, HUANG Yan

  Oil Geophysical Prospecting. 2024, 59(4): 702-713. https://doi.org/10.13810/j.cnki.issn.1000-7210.2024.04.006

  Abstract ( )   Knowledge map   Save

  The resolution of seismic data directly influences the subsequent processing and interpretation precision， thus attracting considerable attention. Deep learning is widely used in solving reverse problems due to its capacity for automatic extraction of deep features and excellent nonlinear approximation. In the field of seismic exploration， the convolution operators in deep convolutional networks are consistent with the convolutional model of seismic data， which has the potential to significantly improve the resolution of seismic data through intelligent means. Currently， enhancing the resolution of seismic data through convolutional neural networks has become a research hotspot. The key to addressing this issue lies in designing suitable and effective network structures and loss functions for resolution enhancement. Therefore， a high-resolution seismic data processing method based on strong supervised deep learning is proposed. Drawing inspiration from image super-resolution reconstruction， this method makes full use of the spatial continuity of the underground structure， and a generative adversarial network structure is designed to enhance the longitudinal resolution of seismic data. Additionally， a loss function combining L₁ loss and multi-scale structural similarity loss is employed to improve the perceived quality of deep learning networks. The experimental results of seismic data and actual seismic data show that compared to the conventional loss function， the loss function presented in this study can significantly enhance the high-resolution processing performance of intelligent algorithms. It notably improves the continuity of the seismic events and enriches the high-frequency detail information of seismic data，and the feasibility and effectiveness of the proposed method are verified.
• Select
  An integrated method of seismic data reconstruction and denoising based on generative adversarial network

  ZHANG Yan, ZHANG Yiming, DONG Hongli, SONG Liwei

  Oil Geophysical Prospecting. 2024, 59(4): 714-723. https://doi.org/10.13810/j.cnki.issn.1000-7210.2024.04.008

  Abstract ( )   Knowledge map   Save

  During the actual acquisition process， due to terrain conditions and human factors， seismic data can suffer from spatial under sampling or irregular sampling， as well as being contaminated by random noise， which hinders subsequent processing and interpretation. Current seismic data processing methods typically separate reconstruction and denoising into two stages， often introducing additional errors. The focus of the integrated reconstruction and denoising method is to accurately extract the effective features of seismic data under mixed interference from missing traces and noise. This paper proposes an integrated method for seismic data reconstruction and denoising based on conditional Wasserstein generative adversarial network （cWGAN）. Firstly， a ge-nerator model is constructed with the U-Net model as the basic network structure， and the event features of seismic data are extracted. Conditional constraints are then introduced into the discriminator model to guide the gradient optimization direction of the generator. Secondly， an error description model for reconstruction and denoising is established， and an integrated loss function is designed to address both tasks simultaneously. Finally， tests on synthetic and actual data demonstrate that the seismic data recovered by the proposed network model have a higher signal-to-noise ratio and good robustness.
• Select
  Seismic data de-noising method based on an attention mechanism U-Net

  CAO Jingjie, GAO Kangfu, XU Yinpo, WANG Naijian, ZHANG Chun, ZHU Yuefei

  Oil Geophysical Prospecting. 2024, 59(4): 724-735. https://doi.org/10.13810/j.cnki.issn.1000-7210.2024.04.011

  Abstract ( )   Knowledge map   Save

  Due to various factors such as instruments， equipment， and environment during field acquisition， there often exist various types of noise in seismic data， including surface waves，ghost waves， random noise， etc.， affecting the reliability and accuracy of seismic data processing and interpretation. Recently， methods based on artificial intelligence have become a research hotspot in seismic data denoising， as they have high computing efficiency and good numerical effects. U-Net is a classic convolutional neural network structure commonly used in image segmentation tasks. Attention mechanism （AM） is a technique that allows models to focus more on specific regions or features during the learning process. This paper constructs a U-Net with attention function by adding an AM module to the U-Net network and applies it to seismic data denoising. To address the boundary effects generated during the denoising process， the expansion filling method is used to segment data. This method has strong universality and can be used for other network models. By comparing the denoising effect of AU-Net and U-Net， it has been proved that AU-Net network has better denoising effect than that of the U-Net， which can better preserve weak signals.Meanwhile， AU-Net denoising method is more adap-table by transfer learning.
• Select
  Seismic random noise suppression method by adaptive dynamic filtering network

  XU Yankai, WANG Di, LI Yizhen, CAO Siyuan, HAO Yuexiang

  Oil Geophysical Prospecting. 2024, 59(4): 736-744. https://doi.org/10.13810/j.cnki.issn.1000-7210.2024.04.012

  Abstract ( )   Knowledge map   Save

  Due to the complex geological and environmental conditions， the signal-to-noise ratio of seismic data is relatively low， which has a negative impact on subsequent research. Therefore， the suppression of random noise in seismic data processing has been of great concern. The existing algorithms are unable to effectively suppress noise and preserve the effective signal. Therefore， this paper combines traditional methods with deep learning and puts forward a method based on an adaptive dynamic filtering network to suppress random noise in seismic data. The network is based on an encoder-decoder architecture. Firstly， the idea of channel attention mechanism （AM） is introduced to realize the feature integration of multi-scale data formed by dilated convolution through channel AM， providing accurate and rich feature representation for the network. Then， dynamic convolution is introduced to achieve the learning of high-frequency features of seismic data with low computational complexity， so as to preserve more detailed information. The experimental results of both synthetic data and actual data show that the adaptive dynamic filtering network can effectively suppress random noise in seismic data while retaining richer details of seismic data， and the signal-to-noise ratio of seismic data after proces- sing is significantly improved.
• Select
  The method of obstacle extraction about remote sensing image based on V-Transformer

  DENG Fei, LUO Wen, JIANG Xianyi, XU Yinpo, WANG Yan

  Oil Geophysical Prospecting. 2024, 59(4): 745-754. https://doi.org/10.13810/j.cnki.issn.1000-7210.2024.04.013

  Abstract ( )   Knowledge map   Save

  Obstacles in remote sensing images are the most important bases for the variable geometry of observation systems in seismic exploration. The traditional manual obstacle extraction methods are inefficient and susceptible to human factors， and difficult to ensure result consistency， making them unsuitable for complex surface environments and large numbers of obstacles. Current generalized methods for automatic obstacle extraction with convolutional neural networks are limited by the size of convolution kernels， unable to directly perform semantic interactions over long distances， and fail to accurately extract obstacles with large spans that are partially occluded （country roads， rivers， etc.）. Therefore， this study proposes a V-shaped fully self-attention network （MTNet） to extract obstacles from remote sensing images. Firstly， MTNet adopts an end-to-end V-shaped encoder-decoder structure to realize information interaction through skip connections；Secondly， the traditional convolutional layer is replaced by the Mix-Transformer block with long-range modeling capability to extract and reconstruct more accurate multi-scale features of the obstacle； Finally， the transposed convolution is replaced by the light-weight block extending layer for upsampling and image segmentation to reconstruct the obstacle information.Experimental results show that the network significantly outperforms existing methods in terms of accuracy and speed in segmenting obstacles， especially in road recognition.
PROCESSING TECHNIQUE
• Select
  Method of up-going and down-going wavefield separation in the frequency-wavenumber domain for dual-sensor seismic data

  GAO Shaowu, YU Wanhui, ZHANG Jianlei, SUN Pengyuan, MA Guangkai, WANG Kebin

  Oil Geophysical Prospecting. 2024, 59(4): 755-762. https://doi.org/10.13810/j.cnki.issn.1000-7210.2024.04.014

  Abstract ( )   Knowledge map   Save

  With the advancement of offshore oil and gas exploration and development， the application of dual-sensor ocean bottom receivers （OBC/OBN） is becoming increasingly widespread. As a key technology in dual-sensor data processing， the separation of the up-going and down-going wavefields determines the data processing quality and application effectiveness. To address the limitations of conventional methods that fail to separate the up-going and down-going wavefields， hydrophone and geophone seismic data are utilized in the frequency-wavenumber domain. By calculating the calibration filter factor for both types of data， with and without direct waves， precise calibration and effective separation of the up-going and down-going wavefields are achieved. The separated up-going wavefield data not only eliminates interference from multiple ghost waves but also improves the signal-to-noise ratio and resolution of seismic data. It thus provides high-fidelity up-going and down-going wavefield data for subsequent joint deconvolution and migration imaging processes. The data examples demonstrate the effectiveness and practicality of the proposed method.
• Select
  Estimation of mixed-phase wavelet and Q value of nonstationary seismic data using genetic algorithm

  ZHU Yaoxu, BAO Qianzong

  Oil Geophysical Prospecting. 2024, 59(4): 763-770. https://doi.org/10.13810/j.cnki.issn.1000-7210.2024.04.009

  Abstract ( )   Knowledge map   Save

  Seismic high-resolution processing or inversion is aimed to obtain an accurate reflection coefficient or elastic parameter model. However，the formation filtering effect blurs the formation reflection information in seismic records， so it is necessary to eliminate this filtering effect. Most of the existing resolution enhancement methods can not completely discard some assumptions about seismic wavelet and Q model. In order to obtain more realistic seismic wavelets，while simultaneously adaptively acquiring the Q model，this paper combines the seismic wavelet phase estimation with the Q model estimation，proposing an estimation method of mixed-phase wavelet and Q value for nonstationary seismic data based on genetic algorithm. First，the amplitude information of the initial seismic wavelet is obtained by fitting the well-side seismic record. Then the coding chain for genetic algorithm is constructed based on whether the root of the wavelet Z transform moves with the unit circle or not. On the other hand，the binary representation corresponding to the decimal Q model can also be characteri-zed by coded chain，so the global optimization algorithm can be used to simultaneously estimate the seismic mixed-phase wavelet and the Q model. Combining root transformation with genetic algorithms can continuously adjust the phase of the wavelet while adaptively generating the Q model. The time-varying wavelet matrix，along with well logging reflection coefficients，are used to obtain the synthetic seismic record，which is matched with the well-side seismic record. Finally，a reasonable mixed phase wavelet and formation Q model are obtained，and then the time-varying wavelet matrix is constructed for time-varying deconvolution. The phase of mixed-phase wavelet obtained by fitting the well-side seismic records and logging data is closer to the actual seismic wavelet. The theoretical data and practical data processing results confirm the effectiveness of this method.
• Select
  Damped multichannel singular spectrum analysis for diffraction separation based on the Cook-distance

  HUO Weiguang, CAO Jingjie, CHEN Xue, ZHAO Jingtao, ZHAO Shifeng, CAI Zhicheng

  Oil Geophysical Prospecting. 2024, 59(4): 771-781. https://doi.org/10.13810/j.cnki.issn.1000-7210.2024.04.010

  Abstract ( )   Knowledge map   Save

  Diffraction seismic exploration is a crucial technique for enhancing the lateral resolution of small-scale geological structure imaging. Diffracted waves in conventional seismic records can be obscured by strongerenergy reflected waves， making it necessary to separate these two wavefields for imaging purposes. Damped multi-channel singular spectrum analysis （DMSSA） is a rank-constrained denoising method that separates wavefields by decomposing the seismic data into a Hankel matrix and performing singular value decomposition. In this process， reflected and diffracted waves correspond to larger and smaller singular values respectively. However， this method relies on manually determining the rank of the reflected wavefield， which is impractical for processing large volumes of seismic data. To address the issue of manual selection， this paper proposes using Cook-distance to automatically calculate the rank of the reflected wavefield. By combining Cook-distance with the DMSSA algorithm， this paper achieved effective separation of reflected and diffracted waves. Experiments on common-shot gathers and post-stack data demonstrate that this method can successfully obtain high-quality diffraction wavefields， highlighting the effectiveness of the proposed approach.
• Select
  High-resolution seismic signal processing method based on matching pursuit and kernel principal component analysis

  DANG Tengyun, XU Tianji, QIAN Zhongping, ZOU Zhen, ZHANG Hongying

  Oil Geophysical Prospecting. 2024, 59(4): 782-789. https://doi.org/10.13810/j.cnki.issn.1000-7210.2024.04.015

  Abstract ( )   Knowledge map   Save

  Resolution is an important factor affecting the interpretation of seismic data. Low seismic signal resolution can lead to difficulties in identifying small faults and thin interbeds. To address this issue， this study proposes a high-resolution seismic signal processing method based on a matching pursuit algorithm and kernel principal component analysis （KPCA）. Firstly， the matching pursuit algorithm is utilized to iteratively obtain the most effective information on seismic signals through sparse decomposition. Next， the wavelet is replaced by a wideband Ricker wavelet for shaping processing， effectively suppressing the side lobes of the wavelet and improving the resolution of seismic data. Finally， the original seismic signals are mapped to a high-dimensional space through nonlinear mapping using KPCA， and the seismic signals are reconstructed in the high-dimensional space to eliminate redundant information. Practical applications demonstrate that the seismic signals processed by this method exhibit clearer waveforms and richer details， which are beneficial for fault identification and characterization of thin bed， thereby providing a data foundation for subsequent geological data interpretation and reservoir prediction.
MIGRATION AND IMAGING
• Select
  Reflection waveform inversion imaging driven by redatuming in ultra-deepwater area

  WANG Xuan, WANG Tengfei, CHENG Jiubing, WU Sihai, ZHU Feng, LIU Yuzhu

  Oil Geophysical Prospecting. 2024, 59(4): 790-799. https://doi.org/10.13810/j.cnki.issn.1000-7210.2024.04.007

  Abstract ( )   Knowledge map   Save

  Offshore petroleum exploration beyond the continental shelf is characterized by high risk and substantial investment， necessitating high precision in structural imaging and reservoir prediction. Traditional velocity modeling method based on ray theory often fails to meet the demands for fine imaging. In ultra-deepwater exploration， seismic observations are challenged by the thick seawater layer， resulting in extended recording times and immense data and model spaces， making large-scale application of wave theory-based velocity modeling difficult. To address this， this paper proposes a reflection waveform inversion imaging technology driven by redatuming. This technology mitigates the seismic wave propagation effects in the seawater layer， reducing data volume， compressing model space， and alleviating velocity-depth ambiguity， thereby enabling the practical application of full waveform inversion in deep-sea environments. Due to the limited detection distance in ultra-deepwater area， recording the first arrival wave with sufficient depth penetration is challenging， preventing its use in waveform inversion for velocity updates. Consequently， this paper employ reflection waveform inversion for the redatumed data to achieve a high-precision velocity model and imaging results. Synthetic data from theoretical models and actual data processing results from Block A in the South China Sea demonstrate that this technology significantly reduces data volume and enhances the efficiency of high-precision velocity modeling. It holds promising potential for improving ultra-deepwater oil and gas exploration.
• Select
  Seismic imaging technology in volcanic rock areas

  ZHANG Weiwei, YUAN Quanshe, XU Leyi, ZHANG Zhizhong, GUO Jia, GUO Jianqing

  Oil Geophysical Prospecting. 2024, 59(4): 800-810. https://doi.org/10.13810/j.cnki.issn.1000-7210.2024.04.017

  Abstract ( )   Knowledge map   Save

  With the continuous improvement of oil and gas exploration， some oil and gas fields with volcanic rocks and volcaniclastic rocks as reservoirs have been discovered. However， due to the significant shielding effect of volcanic rocks on seismic signals， the significant velocity variations of seismic waves in different layers and types of volcanic rocks， velocity modeling and imaging in volcanic rock development areas are extremely difficult.The EP region is located in the Yangjiang-Yitong Fault Zone， with great resource potential and a large number of structural traps. Meanwhile， the Paleogene and its overlying strata have developed multiple periods and layers of volcanic rocks. Due to unclear imaging of the volcanic rocks and their underlying strata， it is difficult to give a good definition of volcanic facies and to accurately evaluate the impact of volcanic rocks on the conditions of hydrocarbon generation， storage， and capping. This paper explores some new methods and new technology attempted for key issues such as multiple wave suppression， weak signal recovery under volcanic rocks， and precise velocity modeling of volcanic rocks. Firstly， MWD（Model based Water-layer related Demultiple）technology is used to predict volcanic multiples. Then， AVO residual amplitude compensation and low-frequency operator extrapolation random noise attenuation techniques are used to improve the quality of weak signals under volcanic rocks. Finally，first break tomography and grid-based tomography are used for accurate PSDM（Prestack Deep Migration） velocity modeling. The reprocessed PSDM results demonstrate substantial improvement in imaging of volcanic rocks and their underlying strata， indicating that the reprocessing techno- logy scheme can be a good reference for seismic imaging in volcanic rock areas.
COMPREHENSIVE RESEARCH
• Select
  ARD-AVA inversion based gas-bearing prediction technologies and application

  JI Yongzhen

  Oil Geophysical Prospecting. 2024, 59(4): 811-818. https://doi.org/10.13810/j.cnki.issn.1000-7210.2024.04.018

  Abstract ( )   Knowledge map   Save

  To further optimize AVA inversion technology， thus improving the accuracy of reservoir gas-bearing prediction， two key issues need to be addressed： the prior choice of inversion method and the choice of AVA inversion approximation and partial stack strategy. To this end， the variable specific values need to be estimated from different seismic channels by utilizing the hyper-parameters in automatic relevance determination （ARD） prior， rather than using fixed values （such as Gaussian prior and Cauchy prior） to constrain the inversion process. In this way， the lateral changes of reservoir geological features during the gas prediction can be better reflected， and high-precision prediction results can be obtained. Two criteria （parameter sensitivity and condition number） are referred to during the selection of the AVA inversion approximation and partial stack strategy， and recommendations for AVA approximation and the choice of maximum， minimum， and angle intervals are given. The proposed ARD-AVA inversion method， AVA approximation， and partial stack optimal strategy have demonstrated great effects in the gas-bearing prediction of channel-type gas reservoir. By comparing the prediction result with traditional methods， the feasibility and advantages of the proposed method have been proved by combining well logging and production data.
• Select
  Identification and correction methods of fault shadow zone: A case study of Shengping tectonics in Songliao Basin

  YU Jie, WANG Jingyi, ZHANG Zhiming, LIU Xiaowen, MU Lun, LIU Zongbao

  Oil Geophysical Prospecting. 2024, 59(4): 819-827. https://doi.org/10.13810/j.cnki.issn.1000-7210.2024.04.019

  Abstract ( )   Knowledge map   Save

  Stratigraphic dislocations often cause lateral variations of velocity， resulting in the phenomenon of fault shadow zone in time-domain seismic data， which affects geological interpretation and trap evaluation. To this end， taking the Shengping tectonics in the Songliao Basin as an example， the phenomenon of fault shadow zone is analyzed， and three identification methods are proposed： forward modeling， well-seismic combined stratigraphic correlation， and average velocity mapping. Based on these analyses， two correction methods， namely， high-precision velocity field correction and time-domain horizon correction， are proposed for the fault shadow zone. The results indicate that the high-precision velocity field correction method effectively corrects the data volume and has broader adaptability， while the time-domain horizon correction method allows for quick and straightforward horizon corrections and provides valuable insights for similar areas.
• Select
  Post-stack seismic impedance inversion method with multi-channel correlation under adaptive window

  PENG Zhen, XU Huiqun, ZHANG Wei

  Oil Geophysical Prospecting. 2024, 59(4): 828-836. https://doi.org/10.13810/j.cnki.issn.1000-7210.2024.04.020

  Abstract ( )   Knowledge map   Save

  At present， the main way to improve the lateral continuity of inversion is still the regularization method based on the assumption of prior information of model parameters. When the prior information contradicts the actual situation， the regularization method will have problems. Therefore， based on the theory and method of multi-channel inversion， a post-stack seismic impedance inversion method with multi-channel correlation under an adaptive window is proposed.Firstly， using the idea of active learning， the position of key features of seismic data is screened by querying the original seismic data.Secondly， the seismic data are divided into time windows of different scales according to the above locations， which is helpful for fault information protection and abnormal body boundary identification. Thirdly， the correlation between seismic traces in the adaptive window is calculated according to the dot multiplication.Finally， the “weighted sum” seismic trace is used to obtain the reconstructed seismic trace， which is the constraint condition of the inversion objective function.With the help of the idea of active learning and local correlation of seismic data， the proposed method can better reflect the correlation between multiple channels in the adaptive window and can be processed in pa-rallel.The model test and the field data test show that the proposed method has good lateral continuity of inversion results， and can take into account the special structural information. However， the computational efficiency is low， and there are some limitations in the shallow and deep areas with large speed differences， making the proposed method suitable for narrow areas with special structures.
• Select
  High-precision fault prediction technologies and applications

  ZHANG Xichen, HAN Ruidong, DU Changjiang, LI Lei, CHEN Maoshan, FENG Jiameng

  Oil Geophysical Prospecting. 2024, 59(4): 837-847. https://doi.org/10.13810/j.cnki.issn.1000-7210.2024.04.021

  Abstract ( )   Knowledge map   Save

  Using fault-sensitive attributes to identify faults is limited in accuracy and applicability， mainly able to identify major large-scale faults. As medium and small-scale faults lack sufficient resolution and continuity， the predicted number of faults is inaccurate. The credibility of predicting faults solely relying on a single attribute is relatively lower. In addition， for a long time， fault interpretation has been mainly done manually， seriously affecting working efficiency. To overcome the limitations of a single method， thus identifying more fault layers， improving the resolution and continuity of fault prediction， and eliminating the “double boundary” effect， this paper creates a high-precision artificial intelligence fault prediction process to gradually improve the accuracy and efficiency of fault interpretation. Firstly， post-stack seismic data preprocessing is carried out to improve the quality of original seismic data and establish a foundation for generating high-precision fault attributes. Secondly， artificial intelligence fault prediction is carried out to further improve the prediction accuracy and process the artificial intelligence prediction results in the “fault enhancement-skeletonization-ant body” way. Again， multi-attribute fusion is utilized to enrich multi-scale fault information. Finally， with high-precision fault prediction results and intelligent technology， the efficiency of fault interpretation is improved. The high-precision artificial intelligence fault prediction process was applied to the actual data of Zone X in the Bohai Bay Basin， and the results showed that the profile normal faults are mainly composed of stepped and Y-shaped combinations. Most of them are SE orientation and form simultaneously with the fault depression. The dominant orientation of the fault on the plane is NE， and forms a comb-like combination with secondary faults in the near SN direction， indicating that the area may have undergone sinistral strike slip while deforming extensionally.
• Select
  Seismic S-wave modeling technology and its application in deep-water gas field reservoir prediction

  AO Wei, ZHANG Weiwei, WANG Xudong, YANG Xueqi, LI Zhiye

  Oil Geophysical Prospecting. 2024, 59(4): 848-855. https://doi.org/10.13810/j.cnki.issn.1000-7210.2024.04.022

  Abstract ( )   Knowledge map   Save

  Due to the lack of shear wave （S-wave） data in deep-water area， the P-wave model is generally converted to the S-wave model through linear relationship or phase control in the modeling process of pre-stack inversion. However， those methods may bring uncertainty due to human influence， and pre-stack inversion results cannot fully and reliably represent the formation characteristics and reservoir distribution.Therefore， combining with the production demand， this paper proposes to extract Poisson's ratio information in three-dimensional space based on wide angle seismic data as the conversion relationship between P-wave and S-wave data， and then obtain the low-frequency model of S-wave constrained by Poisson’s ratio.The specific steps are as follows： Firstly， the study evaluates the incident angle which is sensitive to the lithology in the formation. Secondly， the study extracts Poisson’s ratio elastic impedance （PEI） with post-stack inversion based on far-angle seismic. Thirdly， the study converts the P-wave model into the S-wave model with the PEI. The exploration practice in the Baiyun deep-water area shows that pre-stack inversion results can clearly reflect the geological and lithologic information， and the plane distribution of No. 2 sand body is consistent with the gas/water boundary and the real drilling results， which confirms the reservoir distribution， expands the proved natural gas reserves， and achieves good economic benefits.
• Select
  A multi-scale attribute fusion based on multi-wave seismic

  PAN Hui, GAO Jianhu, GUI Jinyong, LI Shengjun, CHEN Qiyan

  Oil Geophysical Prospecting. 2024, 59(4): 856-864. https://doi.org/10.13810/j.cnki.issn.1000-7210.2024.04.023

  Abstract ( )   Knowledge map   Save

  Compared with the multi-wave seismic data acquisition and processing methods， the progress of multi-wave seismic data interpretation methods is relatively slow， which makes it difficult to demonstratethe superio-rity of multi-wave seismic technology. Conventional seismic attribute extraction and analysis are mostly based on P-wave seismic data， without fully utilizing reservoir P-wave and S-wave information. Therefore， a multi-scale attribute fusion technology is proposed to enhance the oil and gas sensitivity of multi-wave composite attributes. For an actual 2D seismic profile， firstly， a Gaussian pyramid is constructed to generate various compo-site attributes of different resolutions. Secondly， all attributes of different scales are fused to form an effective multi-scale enhanced attribute. Thirdly， by leveraging image superposition theory， different multi-scale enhanced attributes are mixed and superimposed， which can effectively highlight the reservoir development area and retain the differences between various attributes， thus better describing the oil and gas bearing properties of the sand body. The model test shows that for complex two-dimensional models with small differences between oil-bearing sandstone and surrounding rock， the multi-wave seismic fusion attribute based on RGB can detect reservoirs and identify the location of reflection interfaces. The actual data test shows that the proposed method can accurately distinguish the distribution of channel sand body in the target area.
• Select
  Application of guided fuzzy C-means clustering algorithm in joint inversion comprehensive interpretation

  CHEN Yizhou, LIU Jiang, TU Qicui, LI Bingying, LOU Min

  Oil Geophysical Prospecting. 2024, 59(4): 865-874. https://doi.org/10.13810/j.cnki.issn.1000-7210.2024.04.024

  Abstract ( )   Knowledge map   Save

  There are differences in the inversion results of different geophysical methods， and the key to obtaining accurate underground knowledge is a final reasonable interpretation based on the joint inversion results of different methods. A guided fuzzy C-means（FCM） clustering algorithm is proposed for this purpose， and based on the fuzzy C-means （FCM） clustering algorithm， this paper includes the existing geologic understanding， introduces prior constraint information to guide the determination of the clustering centers， and provides a comprehensive quantitative interpretation of the results of the geophysical joint inversion， aiming at reducing the subjectivity and limitations of traditional manual interpretation. The model test shows that the guided FCM clustering technology is more effective than the traditional FCM clustering technologies， especially its ability to effectively distinguish different geological bodies when processing inversion data of complex geological structures. The results of practical data applications demonstrate the great potential of the guided FCM clustering technology in the comprehensive interpretation of multi-attribute geophysical joint inversion results. This technology not only makes geophysical data interpretation more scientific but also provides a more reliable and accurate tool for underground resource exploration.
• Select
  Direct inversion method for fluid factors and anisotropy parameters in VTI media

  PAN Fei, LI Shengjun, Qin Dewen, CHEN Kang, GAO Jianhu, ZHANG Guangzhi

  Oil Geophysical Prospecting. 2024, 59(4): 875-886. https://doi.org/10.13810/j.cnki.issn.1000-7210.2024.04.025

  Abstract ( )   Knowledge map   Save

  Reservoir fluid identification is an important research direction in the field of oil and gas exploration. Shale reservoirs have strong transverse isotropy with a vertical axis of symmetry （VTI）， undeniably impacting fluid factor estimation. Based on the Rüger longitudinal wave reflection coefficient equation， this paper proposes an inversion method for fluid factor， Young’s modulus， and anisotropy parameters based on the elastic impedance of VTI media. Firstly， a reflection coefficient approximate formula and an elastic impedance equation represented by Russell fluid factor， Young’s modulus， and equivalent anisotropic parameters are derived. Then， the fluid factor， Young’s modulus， and anisotropic parameters are estimated through elastic impedance inversion， and the influence of anisotropic parameters on the inversion results is compared. Model testing and practical application of work area data show that this method can reasonably and accurately predict fluid factors， Young’s modulus， and anisotropic parameters， providing a new method for identifying shale reservoir fluids， predicting brittle parameters， and predicting the anisotropic development degree.
NON-SEISMIC
• Select
  Simulation study of secondary oil and gas reservoir detection using marine controlled source electromagnetic method

  CHEN Hangyu, LIU Ying, LIU Mengyuan

  Oil Geophysical Prospecting. 2024, 59(4): 887-898. https://doi.org/10.13810/j.cnki.issn.1000-7210.2024.04.026

  Abstract ( )   Knowledge map   Save

  Marine controlled source electromagnetic method plays a unique role in offshore oil and gas development， which is often employed to identify the oil and gas reservoir and assist in its location. In offshore oil and gas exploration， there are often secondary oil and gas reservoirs near the main oil and gas reservoir， which is hard to distinguish during exploration. In oil and gas exploitation monitoring， the adoption of a time-lapse controlled source electromagnetic method to detect the existence of secondary oil and gas reservoirs can effectively reduce the cost and improve exploration efficiency. Firstly， the two-dimensional forward simulation of the marine controlled source electromagnetic method is conducted to simulate various models with different thicknesses of the secondary oil and gas reservoir， location and distance compared to the main oil and gas reservoir， and different buried depths of the main oil and gas reservoir. The forward electromagnetic response shows obvious electromagnetic differences. Then， according to the difference in normalized amplitudes， the influence of the secondary oil and gas reservoir on the electromagnetic response is analyzed. Finally，OCCAM inversion is employed to invert simple oil and gas models under different production conditions and complex geological conditions. The results show that with the development of the production process， the response of the seconda-ry oil and gas reservoir is increasingly clear， which provides reliable reference data for offshore oil and gas exploitation monitoring.
REVIEW
• Select
  A review of artificial intelligence-based seismic first break picking methods

  YI Simeng, TANG Donglin, ZHAO Yunliang, LI Henghui, DING Chao

  Oil Geophysical Prospecting. 2024, 59(4): 899-914. https://doi.org/10.13810/j.cnki.issn.1000-7210.2024.04.027

  Abstract ( )   Knowledge map   Save

  Seismic first break picking plays a crucial role in providing vital information concerning subsurface structures and seismic activities， thereby holding significance for seismic exploration and geological research. The automatic and accurate picking of first-arrival waves from low signal-to-noise ratio data has garnered consi-derable attention from scholars. This paper provides a comprehensive review of artificial intelligence-based methods employed for seismic picking. It presents an in-depth analysis of the principles， characteristics， and developmental trajectory of five distinct types of methods： clustering， support vector machines （SVM）， backpropagation neural network （BPNN）， convolutional neural networks （CNN）， and recurrent neural networks （RNN）. Clustering， SVM and BPNN methods demonstrate a relatively intuitive and interpretable nature， albeit requiring manual feature extraction. Conversely， CNN and RNN methods possess the ability to autonomously learn seismic data features， yet they rely on substantial volumes of labeled data to facilitate their learning process. Furthermore， this paper discusses the challenges and future research directions of seismic first break picking. Specifically， it emphasizes the imperative need to further advance the real-time capabilities for picking first break under extremely low signal-to-noise ratios and to further develop the lightweight of the network.
• Select
  Ultra-deep oil and gas reservoir seismic prospecting technologies progress

  LI Yalin

  Oil Geophysical Prospecting. 2024, 59(4): 915-924. https://doi.org/10.13810/j.cnki.issn.1000-7210.2024.04.028

  Abstract ( )   Knowledge map   Save

  The ultra-deep oil and gas reservoir gradually becomes an important growth point of explored reserves and production nationwide and worldwide， and the research and application of ultra-deep oil and gas reservoir seismic prospecting technology are of great significance. To this end， with the ultra-deep oil and gas reservoir in Tarim as the research target， the difficulties faced by the seismic and geological conditions and seismic prospecting of ultra-deep oil and gas reservoirs are analyzed. The formation， development， and application results and effects of the third-generation seismic prospecting technology for ultra-deep oil and gas reservoir are summarized， and the development strategy of this technology is prospected. The research suggests that the first generation of ultra-deep oil and gas reservoir seismic prospecting technology mainly based on wide-line and 3D geometry， high-velocity layer excitation， and pre-stack time migration has settled the low signal-to-noise ratio in ultra-deep seismic data. The second generation of ultra-deep oil and gas reservoir seismic technology mainly based on 3D， wide azimuth， higher-density geometry， and prestack depth migration has preliminarily resolved the inaccurate location of ultra-deep seismic data imaging. The third generation of ultra-deep reservoir seismic technology is mainly based on high-density and wide-azimuth 3D geometry， enhanced energy and noise reduction for seismic excitation， deep layer spectrum spread and fidelity， true surface TTI depth migration imaging，segmented fault interpretation and combination， OVT domain phase-controlled inversion and interpretation， etc. This techno-logy has basically overcome the high-precision imaging problem of ultra-deep seismic data and effectively supports the discovery and exploration of the Fuman ultra-deep ten-billion-ton oilfield and the Bozi-Dabei ultra-deep trillion-cubic-meter gas field. The energy enhancement and noise reduction， the spectrum spread and fidelity， the focus location， and the decoupling mapping will be prioritized in the future development of ultra-deep oil and gas reservoir seismic prospecting technology.
PERSONEGE
• Select
  Geophysicist XIONG Jinliang

  Oil Geophysical Prospecting. 2024, 59(4): 925-925.

  Abstract ( )   Knowledge map   Save