Search

The strong heterogeneity characteristics of deep-buried clastic low-permeability reservoirs may lead to great risks in hydrocarbon exploration and development, which makes the accurate identification of reservoir lithofacies crucial for improving the obtained exploration results. Due to the very limited core data acquired from deep drilling, lithofacies logging identification has become the most important method for comprehensively obtaining the rock information of deep-buried reservoirs and is a fundamental task for carrying out reservoir characterization and geological modeling. In this study, a machine learning method is introduced to lithofacies logging identification, to explore an accurate lithofacies identification method for deep fluvial-delta sandstone reservoirs with frequent lithofacies changes. Here Sangonghe Formation in the Central Junggar Basin of China is taken as an example. The K-means-based synthetic minority oversampling technique (K-means SMOTE) is employed to solve the problem regarding the imbalanced lithofacies data categories used to calibrate logging data, and a probabilistic calibration method is introduced to correct the likelihood function. To address the situation in which traditional machine learning methods ignore the geological deposition process, we introduce a depositional prior for controlling the vertical spreading process based on a Markov chain and propose an improved Bayesian inversion process for training on the log data to identify lithofacies. The results of a series of experiments show that, compared with the traditional machine learning method, the new method improves the recognition accuracy by 20%, and the predicted petrographic vertical distribution results are consistent with geological constraints. In addition, SMOTE and probabilistic calibration can effectively handle data imbalance problems so that different categories can be adequately learned. Also the introduction of geological prior has a positive impact on the overall distribution, which significantly improves the accuracy and recall rate of the method. According to this comprehensive analysis, the proposed method greatly enhanced the identification of the lithofacies distributions in the Sangonghe Formation. Therefore, this method can provide a tool for logging lithofacies interpretation of deep and strongly heterogeneous clastic reservoirs in fluvial-delta and other depositional environments.

It is of great significance to study the spatial-temporal variations of the thickness of the gas hydrate stability zone (GHSZ) to understand the decomposition, migration, accumulation and dissipation of gas hydrate, the corresponding relationship between bottom-simulating reflectors (BSRs) and gas hydrate, and the distribution of heterogeneous gas hydrate. We selected the Dongsha region in the South China Sea (SCS) as the research object to calculate the spatial-temporal variation of the GHSZ since 10 Ma, analyzed the main factors affecting the thickness of the GHSZ, discussed the dynamic accumulation processes of gas hydrate, and proposed an accumulation model of gas hydrate in the Dongsha region. The results show that the thicknesses of the GHSZ in the study area were between 0 m and 100 m from 10 to 5.11 Ma, and the relatively higher bottom water temperature (BWT) was the key factor leading to the thinner thickness of the GHSZ during this period. From 5.11-0 Ma, the thickness of the GHSZ gradually increased but showed several fluctuations in thickness due to changes in the geothermal gradient, seawater depth, BWT, and other factors. The decrease in the BWT was the main factor leading to GHSZ thickening from 5.11 to 0 Ma. The thicknesses of the GHSZ are between 110 m and 415 m at present. The present spatial distribution features show the following characteristics. The GHSZ in the deep canyon area is relatively thick, with thicknesses generally between 225 m and 415 m, while the GHSZ in other areas is relatively thin, with thicknesses between 110 m and 225 m. Based on the characteristics of the GHSZ, two hydrate accumulation models are proposed: a double-BSRs model due to thinning of the GHSZ and a multilayer hydrate model due to thickness changes of the GHSZ, with single or multiple BSRs.

The current study examines the methane gas hydrate stability zone (GHSZ) in the Shenhu area in the northern South China Sea (SCS) as an example to calculate the thickness of the GHSZ and reconstruct its evolution since 8.2 Ma. Two mechanisms for typical double BSRs in the Shenhu area are shown, and the relationship between the evolving thickness of the GHSZ and the dynamic accumulation of NGHs at typical stations in the Shenhu area is clarified. The results show that the thickness of the GHSZ varies over time with overall thickening in the Shenhu area. The current thickness of the GHSZ is between 160.98 and 267.94 m. Two mechanisms of double BSRs in the Shenhu area are summarized: the double BSRs pattern based on changes in formation temperature, pressure and other conditions and the double BSRs pattern based on differences in gas source and composition. The formation process and occurrence characteristics of double BSRs and hydrate at site SH-W07-2016 in the Shenhu area are also closely related to the changes in thickness of the GHSZ. In addition, the age when gas source first enters the GHSZ has a considerable influence on the dynamic accumulation process of hydrate. Since the formation of hydrate above the BSR at site SH-W07-2016, the GHSZ has experienced up to two periods of thickening and two periods of thinning at this site. With the changes in the thickness of the GHSZ, up to two stages of hydrate formation and at most two stages of hydrate decomposition have occurred. This paper is of great value for understanding the formation of multiple bottom-simulating reflectors (BSRs) as well as the migration, accumulation and dissipation of natural gas hydrate (NGH) during the dynamic accumulation process.

Micro-nano scale pores can accurately and fastly be measured by the nuclear magnetic resonance (NMR) technique, which provides a new method to quantitatively characterize pore structures in tight sandstone. Based on the method of calibration of mercury pressure data for NMR T2 spectrum, for the measurement inaccuracy due to the mercury saturation less than 100% in tight sandstone, the mercury pressure curve and T2 spectrum is used to cumulate from the maximum pore on the right boundary to the small pores in the left, the range of pore-throat radius measured by the mercury injection in the leftward cumulative curve is selected as a comparable interval of NMR pore-throat radius, and the longitudinal interpolation method and the least square method are utilized to construct the distribution curve of pore-throat radius transformed by T2 spectrum. The modified method is used to obtain NMR T2 spectrum, conversion coefficient of pore-throat radius and pore-throat radius distribution of the Carboniferous-Permian tight sandstone gas reservoirs in the eastern Linqing depression, and characteristics of reservoir pore structures are quantitatively investigated; in addition, in combination with analysis of thin section and scanning electron microscopy, the reservoir effectiveness and cause of the pore structure variability in the tight sandstone are also well studied. The results show that the NMR pore-throat radius curve obtained by the modified method has a high consistency with the mercury injection curve, and the NMR test accuracy of tight sandstone is significantly improved. In the study area, the pore-throat radius of the Carboniferous-Permian tight sandstone mainly ranges from 0.002 to 2 μm, the pore is generally submicro-nano scale, but the pore-throat radius distribution of different types of sandstone varies significantly. The lithic quartz sandstone is rich in siliceous matter and poor in plastic detritus and matrix, generally dominated by submicro-scale pore-throats including micro-scale pore-throats; lithic feldspar sandstone and quartz-rich feldspar lithic sandstone are rich in quartz and poor in plastic detritus and matrix, dominated by submicro-nano scale pore-throats (nano-scale pore-throats predominantly); the lithic fragment-rich feldspar lithic sandstone and lithic sandstone are poor in quartz and rich in plastic detritus and matrix, mainly dominated by nano-scale pore-throats smaller than 0.05 μm. Micropetrographic components are key factors to control pore structure difference and reservoir effectiveness, and the reservoir quality may be macroscopically controlled by sedimentary microfacies; the lithic quartz sandstones of coarse- and fine-grained point bar/riverbed microfacies are the most favorable reservoirs; the lithic feldspar sandstone of fine-grained point bar microfacies, the quartz-rich feldspar lithic sandstone of fine-grained distributary channel and barrier bar microfacies are relatively favorable reservoirs, while both lithic fragment-rich feldspar lithic sandstone and lithic sandstone of tidal-flat facies are ineffective reservoirs with very poor porosity and permeability. Keywords: NMR T2 spectrum, Mercury injection, Conversion coefficient, Pore-throat, Rock fabric, Microfacies, Tight sandstone, Linqing depression