首页期刊导航|Journal of Petroleum Science & Engineering
期刊信息/Journal information
Journal of Petroleum Science & Engineering
Elsevier Science B.V.
Journal of Petroleum Science & Engineering

Elsevier Science B.V.

0920-4105

Journal of Petroleum Science & Engineering/Journal Journal of Petroleum Science & Engineering
正式出版
收录年代

    End-to-end neural network approach to 3D reservoir simulation and adaptation

    E. IllarionovP. TemirchevD. Voloskov
    10页
    查看更多>>摘要：Reservoir simulation and adaptation (also known as history matching) are typically considered as separate problems. While a set of models are aimed at the solution of the forward simulation problem assuming all initial geological parameters are known, the other set of models adjust geological parameters under the fixed forward simulation model to fit production data. This results in many difficulties for both reservoir engineers and developers of new efficient computation schemes. We present a unified approach to reservoir simulation and adaptation problems. A single neural network model allows a forward pass from initial geological parameters of the 3D reservoir model through dynamic state variables to well's production rates and backward gradient propagation to any model inputs and variables. The model fitting and geological parameters adaptation both become the optimization problem over specific parts of the same neural network model. Standard gradient-based optimization schemes can be used to find the optimal solution. Using real-world oilfield model and historical production rates we demonstrate that the suggested approach allows reservoir simulation and history matching with a benefit of several orders of magnitude simulation speed-up. Finally, to propagate this research we open-source a Python-based framework DeepField that allows standard processing of reservoir models and reproducing the approach presented in this paper.
      原文链接:
    • NSTL

    Cement properties characterization from a section retrieved from an oil production well after 33 years of downhole exposure

    Katherine Beltran-JimenezDave GardnerSteinar Kragset
    15页
    查看更多>>摘要：A unique data set has been constructed based on the study of class G cement recovered from an oil well on the Norwegian continental shelf. The cement was placed between two casing sections and submitted to downhole conditions for 33 years. The aim of this research is to analyze the aging of the cement and its consequences for the well integrity. Additionally, the findings represent a baseline for future assessments of old age cement from oil and gas wells. The characterization includes;; petrophysical properties (porosity and permeability): mechanical properties (uniaxial compressive strength (UCS) and Young’s Modulus);; and compositional analysis using Computed Tomography (CT), Scanning Electron Microscopy (SEM), X-ray Fluorescence (XRF) and Quantitative X-ray Diffraction (XRD). Findings suggest that the overall decrease of cement integrity appears to be low, since the magnitudes of permeability and porosity fall within the ranges of early age cement. However, the chemical characterization shows that mud contamination may have an effect on the cement performance and is found to be relevant for a reduction in compressive strength.
      原文链接:
    • NSTL

    Rate of penetration prediction while drilling vertical complex lithology using an ensemble learning model

    Salaheldin ElkatatnyHany GamalAhmed Alsaihati
    15页
    查看更多>>摘要：The rate of penetration (ROP) accounts for a substantial portion of the overall drilling cost. Tlie drilling optimization process, which mostly involves the adjustment of the mechanical drilling parameters, is therefore of prime importance in ensuring efficient drilling. However, drilling formations with assorted types of lithology necessitate the involvement of more parameters to reduce uncertainty and enhance confidence when predicting the ROP. Tlie objective of this paper is to introduce an ensemble model based on random forest (RF), in which artificial neural network (ANN), and adaptive neuro-fuzzy inference system (ANFIS) are the base learner models, to predict the ROP across different lithological formations In this study, two types of actual field data of Well~(-1) were employed to build the model;; (i) mechanical drilling parameters collected from real-time sensors allocated at the rig site, and (ii) petrophysical properties obtained from conventional well logs. Well~(-2) with more than 2300 unseen data points was used to compare the capability of the base learners (i.e., ANN and ANFIS), standalone RF, and RF-meta model in predicting the ROP with two of the earliest published ROP empirical models (Maurer's and Bingham's models). Tlie results showed that the RF-meta model outperformed the base learners and Maurer's and Bingham's empirical models in predicting the ROP in Well~(-2) with a low absolute average percentage error (AAPE) of 7.8 % and a high coefficient of determination (R2) of 0.94.
      原文链接:
    • NSTL

    Depositional environment and organic matter accumulation of the Lower Cambrian Shuijingtuo Formation in the middle Yangtze area, China

    Lin ChenBaomin ZhangXiaohong Chen
    14页
    查看更多>>摘要：The organic-rich shale of the Lower Cambrian Shuijingtuo Formation in the Yangtze area is considered a high-quality source rock in South China. Geochemical proxies are used to reveal the formation and preservation conditions of the Shuijingtuo shale in Yichang in the middle Yangtze area. Geochemical proxies for paleoredox (U/Th, V/Cr, Ni/Co, and Uef versus Moef), paleoproductivity (Ba/Al, Cu/Al, and Ni/Al), and terrigenous influx (Al and Ti) have been employed to interpret the mechanism of organic matter (OM) accumulation. The total organic carbon (TOC) contents of the Shuijingtuo shale is between 0.19 wt% and 4.77 wt% (with an average of 1.38 wt%). The shale can be divided into two intervals based on TOC content, the lower interval with high TOC (TOC >2 wt%) and the upper interval with low TOC (TOC <2 wt%). The results of the geochemical proxies show that the lower interval shale was formed in an anoxic setting with fair paleoproductivity and relatively low terrigenous influx. The upper interval shale was deposited mainly under oxic conditions, with gradually enhanced oxidation, decreased productivity, and relatively high terrigenous influx compared to the lower interval shale. The water mass inferred by the Mo/TOC ratios was a restricted environment with episodic upwelling in the lower interval. The TOC contents display a positive correlation with both the paleoredox and paleoproductivity proxies, indicating that anoxic/dysoxic and high productivity favored OM accumulation. The Co x Mn versus Cd/Mo relationship also suggests that the OM accumulation is chiefly determined by preservation in a restricted environment. These data indicate that the redox condition acts as the dominant factor for OM accumulation. A depositional model is proposed that emphasizes the importance of preservation for OM accumulation of this shale.
      原文链接:
    • NSTL

    Heterogeneous distribution and potential significance of solid bitumen in paleo-oil reservoirs;; Evidence from oil cracking experiments and geological observations

    Pingping LiTing LiHuayao Zou
    9页
    查看更多>>摘要：The solid bitumen (SB) residing in reservoirs after oil cracking can reduce reservoir quality. To build a quantitative standard for the SB content for paleo-oil reservoir identification and to evaluate the heterogeneous distribution of SB and its effect on reservoir quality, we conducted a cracking experiment on eight dolostone samples injected with oil and analyzed the SB content (vol%) and porosity of 706 reservoir samples from the Sichuan Basin that had undergone oil cracking. The SB contents of the experimental samples (0.95~(-1).48 %) injected with normal oil were higher than those of the experimental samples (0.22-0.49 %) injected with light oil. Furthermore, for the experimental samples injected with the same type of oil, the SB contents of the samples with higher paleoporosities (PPs) were also higher than those of the samples with lower PPs. For the reservoir samples from the Sichuan Basin, the SB content exhibits a generally positive correlation with PP. When the PP was <2.5 %, the SB content was relatively low (<1.0 %);; and when the PP was >2.5 %, the SB content increased significantly and was generally > 1.0 %. The data obtained from the oil cracking experiments and reservoir samples from the Sichuan Basin suggests that SB contents of 1.0 % and 0.4 % can be used as the lower limits for normal and light oil reservoir identification, respectively. The oil cracking experiments verified that SB preferentially resided in the smaller pores and throats, and the SB saturations of the samples with lower PPs were larger than those of the samples with higher PPs. For the reservoir samples from the Sichuan Basin, a paleo-oil reservoir with a PP > 8.0 % can be an effective gas reservoir, but some of the paleo-oil reservoirs with PPs of 2.5 %-8.0 % are ineffective gas reservoirs due to SB filling. As a result, the heterogeneous distribution of SB should be carefully considered in gas reservoir evaluation.
      原文链接:
    • NSTL

    Diagenetic evolution and its impact on reservoir porosity in Lower Triassic Jialingjiang Formation in eastern Sichuan Basin, SW China;; Evidence from petrography, trace elements, and Isotopic Studies

    Ziye LuWenli XuHuaguo Wen
    18页
    查看更多>>摘要：The Lower Triassic Jialingjiang Formation, characterised by a carbonate-evaporite succession, is one of the important pay layers in the eastern Sichuan Basin. Several diagenetic events were identified in this study, including meteoric dissolution, dolomitisation, anhydrite/gypsum replacement/cementation, TSR, BSR, and multiple calcite cementation (C1-C3). Dolomitisation and anhydrite/gypsum replacement/cementation are the major diagenetic events affecting reservoir porosity. The dolostone is composed of Dl, D2, and D3 dolomite. Dl dolomite is very finely crystalline, dominating in the low-energy, restricted depositional facies. D2 dolomite is finely crystalline, constituting a major portion of the dolostones. D3 dolomite is depositional fabric retentive dolomite, including dolo-wacke-/packstone, dolo-grainstone, and dolo-boundstone. Four types of anhydrites (Al-A4) were identified in the core and thin sections. Al is stratabound anhydrite. A2 is the anhydrite in dolostones, filling in the interparticle spaces in dolo-grainstone and the intraparticle spaces in dolo-wackestone. A3 is the anhydrite in limestones, which occurs non-fabric-selectively. A4 is the anhydrite filling in the fractures. The diagenetic anhydrites (A2-A4) are interpreted to postdate to major dolomitisation. Dl and D2 dolomites are characterised by 8~(13)Cvpdb values of-3.4-+7.3 %o and 818Ovpdb values of-4.7~(-2).4 %o. The ￡EE patt erns of D1-D3 dolomites are characterised by light rare earth element (LREE) depletion relative to heavy rare earth elements (HREE) ((Pr/Tm)_N of 0.54~(-1).16, 0.81 + 0.20), middle rare earth elements (MREE) enrichment (BSI values of 1.06~(-1).38,1.26 ± 0.10), and negative Ce anomalies (Ce/Ce ratios of 0.74-0.94, 0.83 ± 0.07). The diagenetic anhydrites (A2-A4) show δ~(34)S_V.cdt values of+25.9-+45.7 %o, similar to those of the stratabound anhydrite (+29.6~(-1)-43.0 %o). The large variations in the 8~(13)Cvpdb values of Dl and D2 dolomites are attributed to methanogenesis and BSR. The δ_(18)Ovpdb values and REE characteristics of D1-D3 dolomites suggest that the carbonates are dolomitised by the mesohaline fluid. The δ_(34)Sv_cdt values of A1-A4 indicate that the diagenetic anhydrites are formed by redistribution in a closed system. The petrographical and geochemical data suggest that the Jialingjiang carbonate evolved in a closed system during burial diagenesis. Primary pores dominate in the Jialingjiang carbonate reservoirs, and the boundstone and grainstone show the best porosities based on the porosityversus lithofacies data. Therefore, we propose that the lithofacies, instead of the diagenetic events, are the key factors of the Jialingjiang carbonate. The resisting compaction of dolomitisation helps preserve the reservoir porosity, and anhydrite is the major diagenetic event destroying the reservoir porosity.
      原文链接:
    • NSTL

    The effect of paleo-environment on hydrocarbon generation potential;; Example from Vaca Muerta Formation, Neuquen Basin, Argentina

    A. MatachowskaG.P. LisM. Mastalerz
    14页
    查看更多>>摘要：A 137-m continuous core from the Jurassic-Early Cretaceous marine derived oil shale with the maturity Ro about 0.7 %, representing the oil window in the Vaca Muerta Formation, Neuquen Basin, Argentina, was examined using geological, mineralogical, petrographic, and geochemistry techniques. Three distinct intervals were identified within the core;; the upper carbonate-rich section with intense bioturbation, indicating dysoxic to oxic conditions that resulted in poor organic matter preservation;; a middle section with decreasing carbonate content, suboxic bottom conditions, and higher organic matter content;; and a lower section with higher detrital input and low carbonate content, suboxic to anoxic water conditions, and enhanced preservation of organic matter. Based on the presence of indicator fossils and geochemical parameters, a general trend of increasing organic matter content with lower paleo-oxygen levels was observed. The proportion of solid bitumen within organic matter increases with decreasing oxygen content, suggesting that organic matter deposited in anoxic conditions is more prone to transformation and hydrocarbon generation. There is a negative impact of increasing paleo-oxygen levels on the quantity and quality of organic matter, represented by its potential to generate hydrocarbons.
      原文链接:
    • NSTL

    Genetic types of mudstone in a closed-lacustrine to open-marine transition and their organic matter accumulation patterns;; A case study of the paleocene source rocks in the east China sea basin

    Chao FuShengli LiShunli Li
    19页
    查看更多>>摘要：Organic mudstone can be a high-quality source rock in petroleum basins. The accumulation patterns of various types of mudstone can also serve as a useful record of paleogeography. Due to the impact of tectonic movements, the East China Sea Basin (ECSB) saw its deposition setting transform from closed-lacustrine to open marine settings. This provides an opportunity to explore organic matter accumulation patterns within such a transformation setting. This study classified the constituent mudstones according to their differences in composition and sedimentary structure as follows;; coaly mudstone or carbonaceous mudstone, calcareous mudstone, clayey mudstone, and siliceous mudstone. Considering the different sedimentary structures, siliceous mudstone can be subdivided into laminated, wavy, and slumped mudstones. Using integrated multiple indices of the paleo-environment, which includes clay mineral content, trace mineral content, and micro-paleontology assemblages, we reconstructed the paleography setting during the Paleocene and inferred that during this time, the ECSB was characterized by sedimentary differentiation. We discerned two different organic mudstone accumulation patterns by combining the results from rock-eval pyrolysis and paleo-environmental indices;; 1) A closed-lacustrine mudstone accumulation pattern characterized by abundant sediment supply and dominated by allogenous mudstone formation by the flocculation process. The fresh to slightly salty water environment provided adequate nutrition to microorganisms and an appropriate temperature led to the increased accumulation of type Hi and II2 kerogen. 2) An open-marine mudstone accumulation pattern which revealed a sediment-starved setting. The flocculation process and fine-graded turbidities were the main processes leading to mudstone accumulation. More plants were concentrated in shallow water with declining salinity, leading to peat and mass carbon mudstone accumulation.
      原文链接:
    • NSTL

    Hydrocarbon retention and leakage in traps bounded by active faults;; A case study from traps along the NDG fault in the Qinan area, Bohai Bay Basin, China

    Xianqiang SongHaixue WangXiaofei Fu
    16页
    查看更多>>摘要：Fault-bounded traps, especially along active faults, have complicated hydrocarbon retention and leakage mechanisms. For successful exploration studying hydrocarbon leakage is necessary. Three traps were drilled in the Eslx submember in the hanging wall of the NDG fault. A comparison of the actual hydrocarbon column heights and trap heights in these three traps shows that only the NDG-3 trap is underfilled;; the other traps are completely filled. The paleo-oil column indicates that the reason for the underfilling of the NDG-3 trap is poor retention rather than a small charge volume. The fault sealing capacity is assessed using well and seismic data. The active fault history and throws were measured to analyze the role of fault activation in oil leakage. The results strongly indicate that across-fault membrane leakage is highly unlikely to have caused the oil leakage. Instead, post-charge reactivation probably allowed hydrocarbon leakage from the NDG-3 trap. Furthermore, elimination of the influence of top-seal leakage also implies that post-charge reactivation is the most likely mechanism for oil leakage in the NDG-3 trap. The critical threshold of post-charge reactivation throw for hydrocarbon preservation and leakage is 202~(-2)08 m. When the post-charge reactivation throw is < 202 m, upfault hydrocarbon leakage occurs with difficulty, and hydrocarbons are preserved in fault-bounded traps. When the post-charge reactivation throw is > 208 m, upfault hydrocarbon leakage is more likely to occur. Although the hydrocarbons that accumulate in deep traps may be lost by fault reactivation, this process also supplies hydrocarbons to shallow traps, which offer an opportunity for exploration. These results can strengthen our understanding of the hydrocarbon leakage mechanisms and locations from traps bounded by faults across the Bohai Bay Basin. This is of particular significance in exploring for hydrocarbons and in hydrocarbon production since seal failure is the major cause of dry wells.
      原文链接:
    • NSTL

    Data-driven sequence labeling methods incorporating the long-range spatial variation of geological data for lithofacies sequence estimation

    Gyeong-Tae ParkJina JeongIrina Emelyanova
    14页
    查看更多>>摘要：The use of geophysical well-log data for interpreting the stratigraphic lithofacies sequence is cost effective. In this study, several data-driven lithofacies sequence estimation models are developed, where long-and short-term memory (LSTM) and bidirectional LSTM (BLSTM) are applied to efficiently complement the long-range spatial variation of successive well-log and lithofacies measurements. During the development, the models using the autoregressive (AR) input variables of lithofacies are designed to incorporate the lithofacies sequence pattern into the estimation. The performances of the proposed models are comparatively validated with an artificial deep neural network (DNN)-based model that does not consider long-range variation. Accordingly, a total of six estimation models are examined;; DNN, AR-DNN, LSTM, AR-LSTM, BLSTM, and AR-BLSTM. For model implementation, synthetic data and actual data acquired from the Satyr-5 well in Western Australia are used. For the synthetic data, the results indicate that the incorporation of nonstationary statistical information improves the performance of BLSTM-based models. In addition, AR-input information is effective with respect to the estimation of the vertical thickness of lithofacies. The advantage of using AR inputs can also be observed for actual data, where AR-based models perform significantly better than the other models. Quantitatively speaking, the fitness of DNN-, LSTM-, and BLSTM-based models is 81.79 %, 84.53 %, and 85.08 %, respectively, whereas that of AR-DNN-, AR-LSTM-, and AR-BLSTM-based models is 85.43 %, 85.72 %, and 86.56 %, respectively. The proposed models are expected to be useful with respect to interpreting the heterogeneity of lithofacies distribution in a cost-effective and computationally efficient way. Particularly, BLSTM-based models are widely applicable because they perform well regardless the spatial statistics of lithofacies sequences.
      原文链接:
    • NSTL