Surrogate-assisted management of water-flooding, as a reservoir production optimization strategy, dynamically adjusts development schemes at each production step, leading to improved economic benefits and enhanced recovery. However, many computationally expensive numerical simulation runs are needed to build the surrogate model in most existing surrogate-assisted reservoir production optimization methods. With that in mind, this study proposes an efficient intelligent optimization method based on active learning strategy and surrogate ensemble for multi-objective reservoir production optimization named ALSA-MOPO. In this proposed ALSA-MOPO method, three frequently-used surrogate models, the radial basis function network, Gaussian process regression, and support vector regression are adopted to construct the surrogate ensemble. In addition, an active learning strategy is adopted to reduce the sample of establishing surrogate model and query for the worse and best samples based on the surrogate ensemble using particle swarm optimization which are infilled to the dataset for improving the accuracy and quality of the surrogate ensemble. The ALSA-MOPO method stands out due to its unique use of an active learning strategy to enhance the accuracy of the surrogate model, combined with a surrogate ensemble to improve robustness. Furthermore, two synthetic reservoirs with different scales and one complex fault block reservoir were utilized to test the effectiveness and practicability of the ALSA-MOPO method. The optimization results indicated that the ALSA-MOPO framework outperformed numerical simulation-based methods by approximately 50, 20, and 35 times in the three respective cases.

This study addresses the challenge of understanding hydrocarbon reservoir distribution in lacustrine rift-basins, particularly in the Fushan Depression of the South China Sea, where intricate tectonic and sedimentary dynamics complicate analysis. Effective source kitchens play a critical role in controlling hydrocarbon accumulation patterns, yet existing studies often rely on statistical methods limited by exploration biases. To overcome these shortcomings, 3D basin modeling was employed to assess the influence of effective source rocks on the distribution of hydrocarbons. The findings reveal that reservoirs in lacustrine rift-basins are closely linked to effective source rocks, exhibiting lateral migration patterns that form rings around the source kitchen, as well as vertical migration patterns concentrated above the effective source rock centers. Hydrocarbon migration is generally short-range, extending laterally up to 12 km and vertically up to 2 km. Notably, oil and gas pools are predominantly located in sedimentary facies transition zones, with key resources concentrated within a 3–6 km radius of the source rock in the lower petroleum system and at depths of 3–4 km in the upper system. This work quantitatively assesses the role of effective source rocks, offering crucial insights for future exploration in similar basins. The novelty lies in its comprehensive modeling approach, which enhances understanding of hydrocarbon distribution beyond prior statistical frameworks, providing a robust foundation for exploration and development strategies.

A significant portion of tight sandstone reservoirs commonly displays intricate fluvial channels or fault systems. Despite various attempts at analytical/semi-analytical modeling of multistage-fractured horizontal wells (MFHWs) in unconventional reservoirs, the majority of studies have focused on scenarios with homogeneous original physical properties, neglecting cases where MFHWs traverse multiple regions in channelized heterogeneous reservoirs. Comprehending the influence of heterogeneous and leaky faults on the performance of MFHWs is essential for efficient development. This study presents an innovative semi-analytical model to analyze the pressure transient behavior of MFHWs with secondary fractures as they traverse multiple regions in banded channel heterogeneous reservoirs, particularly considering the presence of partially-communicating faults. The approach combines the source method and Green’s function method to obtain solutions, introducing a novel technique for discretizing fractures without discretizing interfaces. The effects of the reservoir heterogeneity, partially-communicating faults and fractures system on pressure behavior are analyzed. The results indicate that the pressure behavior of MFHWs passing through regions with different physical properties exhibits distinctive characteristics, differing from both the homogeneous case and the heterogeneous cases where the well does not traverse distinct regions. Permeability heterogeneity influences the curves of all other flow regimes, except the early and late flow regimes. Faults affect transient pressure behavior only when not positioned in the middle of each two primary fractures. Region area heterogeneity primarily influences the medium flow regimes. This work provides valuable insights into the performance of MFHWs in channelized heterogeneous reservoirs, offering technical support for well testing in these reservoirs.