Shear wave velocity (Vs) serves as a crucial petrophysical parameter for subsurface characterization, yet its acquisition remains challenging. While long short-term memory (LSTM) networks have emerged as the predominant solution for Vs prediction by synthesizing contextual relationships among conventional logging curves, existing implementations often overlook characteristic discrepancies between training and prediction datasets, leading to suboptimal performance. This study proposes an enhanced LSTM architecture integrated with a generative adversarial mechanism (LSTM-GAM) to address this limitation. The framework employs a dual-component structure: 1) A primary LSTM backbone that captures contextual dependencies across multi-logging sequences, and 2) An adversarial module where the generator minimizes reconstruction errors while the discriminator identifies essential feature representations common to both training and predictive data. This synergistic architecture not only preserves sequential correlations but also enhances cross-domain adaptability through adversarial feature alignment. We validate the model's efficacy using logging data from two vertical wells in the South China Sea. Comparative experiments demonstrate the proposed LSTM-GAM achieves superior prediction accuracy with a mean absolute error (MAE) of 59.4 m/s and determination coefficient (R²) of 0.9064, outperforming conventional LSTM network. Further ablation studies reveal consistent performance improvements across varied input configurations, confirming the method's enhanced generalization capability for Vs prediction tasks. The technical advancement provides an effective data-driven solution for shear wave velocity estimation in complex geological environments.

In view of the limitation of generalization ability faced by deep learning in fault identification, especially in the case of complex underground geological conditions and variable seismic data characteristics, it is often ineffective to directly use the network based on synthetic data training for fault prediction of real data. To overcome this challenge, this study proposes an innovative solution, which uses generative adversarial network-UNet (GAN-UNet) to extract features from data in depth. The network employs a U-net architecture as the backbone to simultaneously extract all features from forward-modeled synthetic data and real seismic data. These features are utilized as inputs for both the fault classifier and discriminator. The fault classifier distinguishes between fault and non-fault segments, while the discriminator employs adversarial mechanisms to differentiate whether input features originate from real seismic data or synthetic data. Once the discriminator, after training, cannot accurately discern the precise source of features, the network model has effectively uncovered the fundamental shared features between the two datasets. This approach demonstrates effective fault recognition in practical seismic data. To verify the effectiveness of the method, we applied it to the actual seismic data sets of the North Sea F3 block and the western deep basin. The experimental results show that compared with the traditional deep learning method, this method shows significant advantages in fault recognition. It not only improves the accuracy of fault identification, but also enhances the adaptability of the model to complex geological conditions.

Shear wave velocity is of great significance for accurate seismic data description and fluid tracking. However, the measurement of shear wave velocity is difficult, requiring high-precision equipment and professional operators. Conventional logging velocity often does not fully include the effective value of shear wave velocity, which brings serious challenges to the accurate exploration of oil and gas in deep strata. To meet demands of the industry, it is necessary to synthesize accurate shear wave logging from relevant conventional logging. Since the LSTM network is unable to compute sample points in parallel and has insufficient nonlinearity caused by sequential calculation, an auto-weighted sequence module is designed in this paper to extract context and better predict shear wave velocity. It takes each logging curve with a fixed step length to use a trainable matrix for auto-weighted rearrangement, and then uses the ELU function to activate. After the calculation of N such modules, the shear wave velocity is predicted by the fully connected layer. Comparing the prediction results of different methods, the auto-weighted neural network can indeed significantly improve the prediction accuracy, obtaining the highest accuracy with 39.21 of MAE and 0.9860 of R2. In addition, the proposal way is better than the LSTM network at different input lengths, with a significant improvement of about 5%. In conclusion, the proposed way can effectively extract the sequence information of the input data, thus it is an effective sequence modeling tool, which has good performance in the prediction of the shear wave.

Shear wave velocity (VS) is a vital prerequisite for rock geophysics. However, due to historical, cost, and technical reasons, the shear wave velocity of some wells is missing. To reduce the deviation of the description of underground oil and gas distribution, it is urgent to develop a high-precision neural network prediction method. In this paper, an attention module is designed to automatically calculate the weight of each part of the input value. Then, the weighted data are fed into the long short-term memory network to predict shear wave velocities. Numerical simulations demonstrate the efficacy of the proposed method, which achieves a significantly lower MAE of 38.89 compared to the LSTM network’s 45.35 in Well B. In addition, the relationship between network input length and prediction accuracy is further analyzed.

叠前反演在储层定量刻画中发挥着关键作用,反演精度受道集品质的影响较大.从提高叠前反演精度的角度出发,提出了 一种针对叠前道集"去噪-提频-谱平衡-剩余时差校正"组合的优化处理技术.通过RNA-3D随机噪声压制技术的应用,能提高CRP道集的信噪比;通过融合时频分析与谱白化技术,能提高CRP道集的垂向分辨率;通过谱平衡技术,能解决动校正拉伸导致的远近偏移距频率不匹配的问题;通过移动积分剩余时差校正技术,能解决同相轴不平的问题,且能较好保持AVO特征,不会造成中、远偏移距振幅畸变.在文昌A油田实际应用表明,该组合优化处理技术能较大程度改善CRP道集品质,提升储层预测的精度.

东濮凹陷上古生界地层油气显示活跃,烃源条件较好,分布范围较广,特别是近年在文留、胡庆地区相继发现自生自储油气藏,展示了上古生界良好的勘探前景.目前上古生界油气勘探还未形成规模储量,尤其是有效储层的地震预测是上古生界勘探的卡脖子技术之一.文中以胡庆地区上古生界储层为例,首先,通过针对性资料处理提升古生界地震资料品质;然后,利用钻井试油结果、精细岩石物理分析与正演模拟,建立上古生界甜点储层识别标准及地震响应特征;最后,采用"分频振幅描河道、联合反演找砂体、频率属性检油气"等递进式技术对甜点砂岩进行预测.甜点砂岩预测结果与钻井资料吻合率达80％以上.

银额盆地拐子湖凹陷沙漠探区特殊的地形、地貌条件对地震资料的激发和接受产生不利的影响,其表层被巨厚且松散的沙土覆盖,沙丘厚度在50～100 m间.在该区开展可控震源采集时,受可控震源机械特征和近地表结构双重影响,在近道范围内,形成"黑三角"干扰,导致地震记录信噪比低."黑三角"干扰具有能量强、频带宽、分布广、形态差异大等特征,在共炮点域、共检波点域等单一排列内,无法准确统计其振幅强度,导致压制"黑三角"噪声后残留噪声过多,为此提出先按照共炮—偏移距域进行数据重排,然后进行去噪处理的思路,基于银额盆地拐子湖地区的实际可控震源资料应用,验证了该方法在有效压制"黑三角"的同时,能有效保护近道反射信号,特别是深层弱反射信息,也验证了该方法在提高资料信噪比上的有效性.

东濮凹陷是一个典型的陆相盐湖断陷盆地,砂泥岩薄互层是常见的沉积叠加样式,同时西南洼地区沙二下储层以浅水三角洲沉积体系为主,单层砂体薄,岩性横向变化快,且受到复杂断裂影响,储层边界刻画不清.为此,提出了一种基于地震相控储层反演技术流程:首先,通过精细岩性标定及井震统计分析,优选反映砂岩厚度的敏感地震属性;然后,运用多属性融合技术刻画地震相,并以地震相进行约束,建立反演低频模型,开展叠后反演;最后,通过坐标旋转与叠前反演技术相结合,实现储层的精细预测.应用结果表明,该技术流程能够有效解决薄储层砂体边界刻画不清以及垂向识别能力不足的问题,可为东部其他区域砂泥岩薄互层地震识别提供借鉴.

地质统计学反演融合随机建模与地震确定性反演,有效地综合了地质、测井和原始地震数据,该方法可以有效避开常规确定性反演遇到的数据正则化问题,同时通过地质统计学反演得到的属性体,其垂向分辨率不仅可以接近井资料,同时横向分辨率也得到较好地保障,并且相对于确定性反演而言,具有较高的抗噪性能.通过该方法在中原油田东濮老区的应用,证实利用地质统计学反演能有效地提高砂泥岩薄互层的刻画能力.

波阻抗是地震波传播速度与介质密度的乘积，波阻抗反演在地震反演中具有十分重要的地位。通过介绍波阻抗反演技术的基本原理、具有的优势以及波阻抗反演的流程，并结合砂体理论模型说明了波阻抗反演的重要性。

Pre-stack elastic impedance inversion involves the calculation of elastic impedance curve,while the conventional calculations are based on the assumption that K is a constant,but the assumption does not match with the actual situation,which results in the deviation in calculating elastic impedance curve and extracting the lithology parameters.Therefore,we introduced K scan method to optimize the selection of K.The method includes three steps.Firstly, different K was selected to compute elastic impedance.Then,the elastic impedance was transformed into reflectivity.Finally,the difference of the reflectivity obtained from the second step and accurate Zeoppritz equation was calculated,and the minimum difference was the optimal K.application on theoretical model and actual logging curve show that the method can be used to select best K and has good applicability.

The elastic wave impedance inversion is a "bright point" in the seismic inversion technique in recent years, and also a "hot point" in the inversion research. Elastic wave impedance inversion is a seismic inversion technique which combines conventional acoustic impedance inversion with AVO inversion. It uses different offset gathers data and S-wave、P-wave and density data to carry out joint inversion of elastic parameters related to lithologies and oil/gas-bearing ability, and determine the reservoir properties and oil-bearing ability. The basic principle and method of the pre-stack elastic wave impedance inversion are briefly expounded, and the application case is given.

With the increasing needs of exploration precision, seismic anisotropy has to be taken as a critical issue. Thus the research of seismic wave propagation characteristic of the anisotropic media has a certain value. Based on the theory of the weak anisotropy proposed by Thomsen, this paper analysis the effects of Thomsen anisotropic parameters on three body waves of weak anisotropic VTI media by numerical calculating the theoretical model. The results show that the anisotropic characters of qP-wave at larger、intermediate and small angle depend on e、δ respectively, the anisotropic characters of qSV-wave at intermediate angle depend on (e-δ), the anisotropic characters of qSH-wave at large angle depend on γ,and the anisotropic characters of group velocity which may represent cusps is more obvious than phase velocity. Through these research and analysis, we further understand the law of phase velocity and group velocity according to the anisotropy parameters in the weak anisotropic media.

Pure 3D elastic wave numerical simulations require huge computation resources,which is not conducive to scientific research when computer configuration is not high.2D elastic wave numerical simulation cannot accurately represent 3D situation,and the simulation results are relatively unsatisfactory by using the traditional 3D pseudo spectral method when the model and spatial distribution of wave field are complex.Thus,based on high accurate one-order elastic wave equation expressed with velocity and stress,this paper uses 2.5D numerical simulation method to calculate the partial derivatives of y spatial orientation by Fourier transforming and the partial derivative of x,y spatial orientation and time domain by finite difference.This method can hence compute the 3D elastic wave field in 2D media.This paper also simulates the multiwave wave field in anisotropy media using 2.5D numerical simulation method,which shows that the 2.5D numerical simulation method is an efficient and accurate forward method.This method can also be adapted to complex models.Through wave field analysis,the authors further understand the propagation law of the seismic wave in anisotropy media.

In view of the fact that layers not only show anisotropic behavior,but also show intrinsic viscoelastics,thus an accurate description of wave propagation in earth medium requires a model that accounts for the anisotropic and viscoelastic behavior of rocks. This paper established a wave equation and the difference scheme of 2D viscoelastic VTI media based on the Boltzmann superposition principle,starting from the constitutive equations. The simulation wave field propagation in elastic and viscoelastic VTI medium used grid high order finite difference,and compared and analyzed the effects of energy attenuation and frequency absorption related to different Q factor pair. The results of numerical simulation show:the Q factor related to dilation inelastic deformation affects energy attenuation of qP wave,and the smaller the Q factor is,the greater the qP-wave attenuation. The Q factor related to shear inelastic deformation affects energy attenuation of qSV wave,and the smaller the Q factor,the greater the qSV-wave attenuation. While the frequency of qP wave and qSV wave in the anisotropic viscoelastic tend to be of low frequency,absorption of high frequency is obvious and the effective frequency band narrows. The research work is useful for further understanding of seismic wave propagation in anisotropic viscoelastic media.

4D seismic exploration technology has been developed rapidly in recent years.However,two measurement data observed repeatedly are of non-uniformity because of various factors.These differences exist mainly in four areas: Signal time-delay、energy difference of signal、bandwidth difference of signal、phase difference of signal.In order to effectively protect the differences resulted from fluid and oil-gas reservoir and remove the differences resulted from others factors,we need calibrate the monitor data.This article describes the theory of cross-equilibration technique,and applies it in the 4D seismic data.The result shows that the non-anticipate differences had been removed well,which further confirms the correctness and effectiveness of this technique.

The application of the conventional seismic inversion in reservoir prediction is constrained by many factors, while the technology of random inversion combines effectively the random simulation technology and conventional seismic inversion, and integrates geology、logging and 3-D seismic data, which can make up for the lack of enough information about logging and improve prediction accuracy, what is more, this technology can play a positive role in the rolling stage of exploitation. The technology was applied in a three-dimensional work area to predict permeable sandstone in layers Eq41, Eq42and Eq43, which achieves good results and plays a positive role in the rolling stage of exploitation.

In the oil-gas fields with high exploitation degree,time-lapse seismic technology can be used to predict the exploitation of oil reservoir and the distribution of remaining oil,and to enhance the oil recovery.Time-lapse seismic data(monitoring data) and the early seismic data(basic data) have differences in the geophysical parameters. The differences include reservoir exploitation factors(expected difference) and non-reservoir factors(non-expected difference). Therefore,the non-expected difference in time-lapse seismic data should be removed without damaging the expected difference. Cross-equilibration technique can effectively remove non-expected difference and reflect the reservoir exploitation information and remaining oil distribution information on difference profile.In one oilfield, the seismic data with a 7-year interval in acquisition time was utilized.After the non-uniform analysis for time-lapse seismic data, and cross-equilibration technique was used to correct time,amplitude, phase and frequency,etc.the non-expected difference was effectively removed,and the difference profile reflected the present reservoir exploitation situation as well as the distribution of remaining oil.

新疆塔里木盆地沙雅隆起艾协克-桑台塔木油气区是地震工作最早、勘探程度较高地区.1994年以来,地矿部西北石油局加大了对该区的勘探力度,先后在桑塔木、艾协克、塔里木乡、牧场北、艾协克北、桑塔木东北部署了三维地震勘探工作,并对三维地震资料进行了处理与解释,落实了有利圈闭.在此基础上对有利区块开展了储层反演工作,确定了工区储层展布特征,为该区下一步开发部署提供了依据.

Because of the limitation of the conventional stock analysis method of stock trend in culmination and nadir,a new wavelet network model for learning is obtained according to the multiresolution analysis in the wavelet space.The experiments show that the wavelet network not only has the rapid constringency speed of approaching share price trend,but also has high precision and the stock trend in culmination and nadir is evident too.