Reservoir in the Central Structural Zone of the Xihu Sag is the Huagang Formation, dominated by natural gas reservoirs, and the reservoir in the Western Slope Zone is the Pinghu Formation, dominated by light oil and wet gas. In this paper, the Gas Drive Water Displacement- Magnetic Resonance Imaging (GWD-MRI) experiments are used to simulate the charging characteristics of the sandstone migration layer. The centrifugal- Magnetic Resonance (Cen-NMR) experiments simulate the short-term fast trap charging process, and the Semi-Permeable Baffle (SPB) charging experiment simulates the slow trap accumulation process. Studies have shown that there is a start-up pressure for the migration layer charging, and the start-up pressure of the core with a permeability of 0.3mD is about 0.6MPa. Experimental simulations have confirmed that there is a front zone with changed water saturation, with a width of about 1-1.5cm. The migration layer charging mainly has two actions, drive effect and carry effect. The drive effect reduces the water saturation to 70%-80%, and the carry effect can still reduce the water saturation by 5%-10%. The water saturation of rapid charging is mainly affected by the petro-physical. The porosity is high, the water saturation is low. The water saturation decreased significantly with the increase of centrifugal force when the centrifugal force is smaller. If the centrifugal force is greater than 0.8 MPa, the water saturation decreases slowly. The water saturation of the trap slowly charging is basically maintained at 40%-50%, which matches the water saturation of the airtight coring from 40%-55%, the petro-physical do not affect the final water saturation.

Few-shot object detection (FSOD) aims to address the challenge of requiring a substantial amount of annotations for training in conventional object detection, which is very labor-intensive. However, existing few-shot methods achieve high precision with the sacrifice of time-consuming for exhaustive fine-tuning, or take poor performance in novel-class adaptation. We presume the major reason is that the valuable correlation feature among different categories is insufficiently exploited, hindering the generalization of knowledge from base to novel categories for object detection. In this paper, we propose Few-Shot object detection via Correlation-RPN and Transformer Encoder-Decoder (CRTED), a novel training network to learn object-relevant features of inter-class correlation and intra-class compactness while suppressing object-agnostic features in the background with limited annotated samples. And we also introduce a 4-way tuple-contrast training strategy to positively activate the training progress of our object detector. Experiments over two few-shot benchmarks (Pascal VOC, MS-COCO) demonstrate that, our proposed CRTED without further fine-tuning can achieve comparable performance with current state-of-the-art fine-tuned works. The codes and pre-trained models will be released.