The Nanpu sag is an important secondary tectonic division of the Bohai Bay Basin and has an area of 1532 km2 (591 mi2), of which 1000 km2 (386 mi2) has been explored near Bohai. Recently, exploration has gradually shifted from continental to beach to shallow-marine settings, with significant exploration breakthroughs having occurred in the Nanpu 1, 2, and 3 tectonic belts, indicating good resource potential. However, a systematic resource assessment has yet to be performed, which strongly restricts the exploration progress. In this study, we improved upon the Pareto model and reevaluated the area’s resource potential using two new methods. In one method, the traps are divided into several types, including structural, stratigraphic, and buried hill traps, to predict their individual resource potential and total resources of this area; in the other method, the normalized fluid potential is used to divide the area into several accumulation systems to calculate the resource potential. The Nanpu sag resources are estimated at 1.45 × 109 t (10.6 × 109 bbl), which represents an increase of approximately 2.70 × 108 t (19.8 × 108 bbl) compared with previous evaluations because of the increase in exploration maturity and information validation. The main residual resource is distributed in beach to shallow-marine settings. The stratigraphic and buried hill traps present better resource potential using the trap-type method. Such information can guide the direction of exploration and selection of potential traps.

Tight oil is an important unconventional hydrocarbon resource. The differences in occurrence characteristics between light components (LCs) and heavy components (HCs) of tight oil profoundly affect its mobility and recovery. Current research has focused mainly on the rapid evaluation of the relative contents of LCs, whereas few studies have systematically analyzed the occurrence characteristics of LCs and HCs and their controlling factors. In this study, the differential occurrence characteristics between LCs and HCs are clarified on the basis of data from thin-section petrography, X-ray diffraction, nuclear magnetic resonance, confocal laser scanning microscopy, and reservoir pyrolysis analysis. An innovative quantitative characterization methodology for the relative occurrence volumes of LCs and HCs is proposed. On the basis of this method, the controlling factors that cause the different occurrence characteristics of LCs and HCs are elucidated. Furthermore, the occurrence characteristics of LCs and HCs in various source–reservoir combinations, physical properties, and development intensities of argillaceous laminae are summarized. Finally, an occurrence model of the crude oil components in the Chang 7 tight reservoir is established. The results show that LCs and HCs in the Chang 7 tight reservoir exhibit differences in occurrence volume, state, morphology, and pore size. These differences are primarily controlled by the hydrocarbon generation intensity of the source rock, the source-to-reservoir distance (SRD), and the content of oil-wet minerals in the reservoir. The source sandwich combination exhibits high physical properties, low hydrocarbon generation intensity, high SRD, and low oil-wet mineral content, resulting in relatively high LCs. The source–reservoir interbed and reservoir sandwich combinations feature a high content of argillaceous laminae, high hydrocarbon generation intensity, low SRD, and high oil-wet mineral content, resulting in relatively low LCs. There are three occurrence models of crude oil components in the Chang 7 tight reservoir: the charging force controlling model, the adsorption effect controlling model, and the argillaceous laminae controlling model. The results of this study provide significant guidance for predicting the fluidity of tight oil, accurately assessing the amount of recoverable tight oil resources, and achieving efficient extraction of tight oil.

The Lufeng Sag and Huizhou Sag, both located in the Zhu-1 Depression, have similar geographical locations, but their reservoir characteristics in the Paleogene Wenchang Formation show obvious differences. Primary intergranular pores are mainly developed in the Lufeng Sag. However, secondary pores are the main reservoir space in the Huizhou depression. Overall, the reservoir properties of the Lufeng Sag are better than those of the Huizhou Sag. To analyse the differences between the Paleogene reservoirs in these two areas, this study mainly uses assay data, such as rock thin sections, scanning electron microscope images, drilling, and logging, to analyse the differential development mechanisms of high-quality reservoirs, and two types of reservoir development models were concluded. The results show that the anti-compaction primary porosity preservation mode is mainly developed in the Lufeng Sag. High compositional maturity quartz sandstone is the congenital condition of primary porosity development. The top and bottom calcareous cementation formed of the large set of thick sand bodies increases the rock’s anti-compaction ability. The early shallow burial slows down the compaction action of overlying strata. Under the low geothermal temperature, it can delay the time for deep reservoirs to enter the middle diagenetic stage. The reservoirs in the Huizhou Sag are developed with the secondary dissolution pore development model. The Wenchang Formation reservoir in the Huizhou Sag has a large area of contact with source rocks, and organic acids can migrate to sandstone reservoirs for dissolution. Additionally, the secondary dissolution pores are more developed because the Wenchang Formation reservoirs in the Huizhou Sag contain more easily dissolved substances.

Assessing and forecasting hydrocarbon resource potential (HRP) is of great significance. However, due to the complexity and uncertainty of geological conditions during hydrocarbon accumulation, it is challenging to accurately establish HRP models. This study employs machine learning methods to construct a HRP assessment model. First, nine primary controlling factors were selected from the five key conditions for HRP: source rock, reservoir, trap, migration, and accumulation. Subsequently, three prediction models were developed based on the backpropagation (BP) neural network, BP-Bagging algorithm, and BP-AdaBoost algorithm, with hydrocarbon resources abundance as the output metric. These models were applied to the Dongpu Depression in the Bohai Bay Basin for performance evaluation and optimization. Finally, this study examined the importance of various variables in predicting HRP and analyzed model uncertainty. The results indicate that the BP-AdaBoost model outperforms the others. On the test dataset, the BP-AdaBoost model achieved an R² value of 0.77, compared to 0.73 for the BP-Bagging model and only 0.64 for the standard BP model. Variable importance analysis revealed that trap area, sandstone thickness, sedimentary facies type, and distance to faults significantly contribute to HRP. Furthermore, model accuracy is influenced by multiple factors, including the selection and quantification of geological parameters, dataset size and distribution characteristics, and the choice of machine learning algorithm models. In summary, machine learning provides a reliable method for assessing HRP, offering new insights for identifying high-quality exploration blocks and optimizing development strategies.

In tight reservoirs deposited in diverse sedimentary settings, the pore structure governs tight oil enrichment features and sweet-spot distribution. Taking the tight sandstone reservoirs of the lower third member of the Shahejie Formation in the Linnan Subsag of Bohai Bay Basin in China as an example, this study employs XRD to delineate petrological characteristics, while porosity and permeability measurements are used to quantify physical properties. In addition, thin section, SEM, HPMI, NMR, fractal theory, and cathodoluminescence experiments are applied to investigate pore structure characteristics and influencing factors. The results reveal two sedimentary systems: turbidity current and delta front deposits. Turbidite reservoirs exhibit the coarse pore-coarse throats (Type A), medium pore-medium throats (Type B), and fine pore-medium throats (Type C) pore structures. Delta front reservoirs are characterized by medium-pore-coarse-throat (Type D), medium-pore-fine-throat (Type E), and fine-pore-fine-throat (Type F) pore structures. Turbidite reservoirs show more favorable pore structures for oil exploration compared to delta fronts, in which lithofacies and diagenetic facies are the key influences. A genetic model identifies the highest-quality Type A forms in fine sandstone lithofacies under medium compaction–medium cementation–strong dissolution, with pore diameters averaging 10.84 μm in turbidite reservoirs. Conversely, the poorest Type F forms in argillaceous layered siltstone lithofacies under strong compaction, cementation, and weak dissolution diagenetic facies in delta fronts, with pore diameters averaging 0.88 μm. Consequently, the control effect of the pore quality means that Type A has the highest and Type F has the lowest oil-bearing capacity. These findings provide valuable guidance for the classification, evaluation, and exploration of tight oil sweet spots.