Applied Geophysics最新文献

Buiding data-driven models using machine learning methods has gradually become a common approach for studying reservoir parameters. Among these methods, deep learning methods are highly effective. From the perspective of multi-task learning, this paper uses six types of logging data—acoustic logging (AC), gamma ray (GR), compensated neutron porosity (CNL), density (DEN), deep and shallow lateral resistivity (LLD) and shallow lateral resistivity (LLS) —that are inputs and three reservoir parameters that are outputs to build a porosity saturation permeability network (PSP-Net) that can predict porosity, saturation, and permeability values simultaneously. These logging data are obtained from 108 training wells in a medium-low permeability oilfield block in the western district of China. PSP-Net method adopts a serial structure to realize transfer learning of reservoir-parameter characteristics. Compared with other existing methods at the stage of academic exploration to simulating industrial applications, the proposed method overcomes the disadvantages inherent in single-task learning reservoir-parameter prediction models, including easily overfitting and heavy model-training workload. Additionally, the proposed method demonstrates good anti-overfitting and generalization capabilities, integrating professional knowledge and experience. In 37 test wells, compared with the existing method, the proposed method exhibited an average error reduction of 10.44%, 27.79%, and 28.83% from porosity, saturation, permeability calculation. The prediction and actual permeabilities are within one order of magnitude. The training on PSP-Net are simpler and more convenient than other single-task learning methods discussed in this paper. Furthermore, the findings of this paper can help in the re-examination of old oilfield wells and the completion of logging data.

In this paper, using natural earthquake P-wave arrival time data recorded by the seismic network in the surrounding area of Madoi, the three-dimensional fine P-wave crustal velocity structure at depths above 60 km in the epicenter of the Madoi Ms7.4 earthquake was inverted using the double-difference seismic tomography method. On the basis of the relocation of the source of the aftershock sequence, we summarized the strip-shaped distribution characteristics along the strike of the Jiangcuo fault, revealing the significant heterogeneity of the crustal velocity structure in the source area. Research has found that most of the Madoi Ms7.4 aftershocks were located in the weak area of the high-speed anomaly in the upper crust. The focal depth changed with the velocity structure, showing obvious fluctuation and segmentation characteristics. There was a good correspondence between the spatial distribution and the velocity structure. The high-velocity bodies of the upper crust in the hypocenter area provided a medium environment for earthquake rupture, the low-velocity bodies of the middle crust formed the deep material, and the migration channel and the undulating shape of the high-speed body in the lower crust corroborated the strong pushing action in the region. The results confirmed that under the continuous promotion of tectonic stress in the Madoi area, the high-speed body of the Jiangcuo fault blocked the migration of weak materials in the middle crust. When the stress accumulation exceeded the limit, the Madoi Ms7.4 earthquake occurred. Meanwhile, the nonuniform velocity structure near the fault plane determined the location of the main shock and the spatiotemporal distribution of the aftershock sequence.

Solar quiet daily variation (Sq) are dependent on local time. Herein, we applied the moving superposition method to separate the Sq component of correction observatory data and performed a time difference correction on the Sq component according to the longitudinal difference between the correction observatory and the field station while maintaining the time of other data components. The data were then reconstructed and used for diurnal-variation correction to improve the accuracy of the daily variations correction resu; lts The moving superposition method employs data of “nonmagnetic disturbance days” obtained 15 d before and after to perform the superposing average calculation on a daily basis, aiming to obtain the Sq of continuous morphological changes. The effect of longitude correction was tested using the observatory record and field survey data. The average correction distance of the test observatories was 2114 km, and the correction accuracies of the H (horizontal component of geomagnetic field), D (geomagnetic declination), and Z (vertical component of geomagnetic field) were improved by 28.4%, 45.0%, and 21.7%, respectively; the average correction distance of the field stations was 2130 km, and the correction accuracies of the F (geomagnetic total intensity), D, I (geomagnetic inclination) components were improved by 35.2%, 26.7%, and 13.9%, respectively. The test results also demonstrated that the longitude correction effect was greater with an increased correction distance.