MEG Source Localization Via Deep Learning

We present a deep learning solution to the problem of localization of magnetoencephalography (MEG) brain signals. The proposed deep model architectures are tuned to single and multiple time point MEG data, and can estimate varying numbers of dipole sources. Results from simulated MEG data on the cortical surface of a real human subject demonstrated improvements against the popular RAP-MUSIC localization algorithm in specific scenarios with varying SNR levels, inter-source correlation values, and number of sources. Importantly, the deep learning models had robust performance to forward model errors resulting from head translation and rotation and a significant reduction in computation time, to a fraction of 1 ms, paving the way to real-time MEG source localization.

Geochemical Characteristics and Oil Source Correlation of Minfeng Area, Dongying Depression, China

The identification of the oil-source correlation plays a significant role in petroleum exploration and development. In this study, we identify the oil-source correlation by a hierarchical cluster analysis method combined with traditional methods. The results shed light on the oil-source correlation in Minfeng area and revealed the oil migration and accumulation process. The crude oil in different structural belts and different horizons has different geochemical characteristics. According to the four types of crude oil and their planner distribution, it was considered that the crude oil mainly migrates along with favorable sand bodies and unconformity surfaces in the lateral direction and then charged and accumulated in the glutenite of Sha3 and Sha4 members since the area from sag to Yan Jia Oil and the gas field was lacking of oil source faults. Further analysis shows that the traps of fault blocks in Yong’anzhen are formed in the same phase, while the crude oil generated in the early stage is charged and accumulated in the fault block of the near source. Along with increasing of the buried depth of source rocks, the overlying source rocks gradually entered into the hydrocarbon generation phase, when crude oil started to charge in the fault blocks farther away.

Fault Mesh Petroleum Plays in the Donghetang Area, Tabei Uplift, Tarim Basin, Northwestern China, and Its Significance for Hydrocarbon Exploration

— The hydrocarbon formation mechanism and potential targets in clastic strata from the Tabei Uplift, Tarim Basin, are documented using the fault mesh petroleum plays theory, based on integrating seismic, well log, well core, and geochemical data. The reservoirs in the Donghetang area are typical allochthonous and far-source fault mesh petroleum plays. There are two sets of fault meshes in the study area: (1) the combination of the Donghe sandstone and Permian–Triassic strata and (2) the combination of the fourth and third formations in the Jurassic strata. The fault mesh petroleum play in the Jurassic is a secondary reservoir that originates from the Carboniferous Donghe sandstone reservoir adjustment based on source correlation. The fault mesh carrier systems show the fully connected, fault–unconformity–transient storage relay, fault–transient storage–unconformity relay, and transient storage–fault relay styles, according to the architecture of the fault mesh. Based on the characteristics of the fault mesh petroleum plays, the reservoirs are divided into three categories (upper-, inner-, and margin-transient storage styles) and 15 styles. Integrated analysis of the hydrocarbon generation and faulting time periods reveals that there were four periods of hydrocarbon charging, with the first three stages charging the reservoirs with oil and the last stage charging the reservoirs with gas. There are multiple stages of reservoir accumulation and adjustment in the fault mesh in the study area. These stages of fault mesh accumulation and adjustment are the main reason why the reservoir distribution multiple vertical units have different hydrocarbon properties. Fault-block and lithologic reservoirs related to the inner- and upper-transient storage styles are the main exploration targets in the clastic strata in the study area.