期刊,Journal of Petroleum Science & Engineering 2022年215PA卷期_国家学术搜索

期刊信息/Journal information

Journal of Petroleum Science & Engineering

Elsevier Science B.V.

主办单位：Elsevier Science B.V.

国际刊号：0920-4105

Journal of Petroleum Science & Engineering/Journal Journal of Petroleum Science & Engineering

正式出版

收录年代

Stimulating relative permeability changes by low-frequency elastic waves: Theory and lab experiments

Alexander Y. RozhkoSerhii LozovyiMarcel Naumann

22页

查看更多>>摘要：A model is presented to describe how low-frequency vibrations can induce changes to relative permeabilities in dual-porosity dual-permeability rocks. We show that a combination of two physical processes may cause this effect: 1) the wave-induced two-phase fuid fow between soft pores (fractures or cracks) and stiff pores (matrix);; and 2) the contact angle hysteresis effect. These two effects lead to redistribution of fuid saturation between fractures and matrix, which may explain alteration of relative permeabilities by small transient strains (~10 6) in rocks where the fuid fow occurs primarily through the fracture network, and fuid storage occurs predominantly in the porous matrix. We assume that the frequency is low enough that viscous and inertial effects are negligible and that any stress-induced increments of pore pressure are uniform within wetting and nonwetting fuid phases. We demonstrate that pulse-like vibrations are much more effcient in changing relative permeabilities than sinusoidal-shaped vibrations. Furthermore, we show that two waveforms with the same amplitude and frequency but with different polarities could have opposite effects on the alteration of relative permeabilities. Specifcally, extensional pulses increase oil relative permeability and decrease water relative permeability in the water-wet rock. Contrary, compressional pulses decrease oil relative permeability and increase water relative permeability in the water-wet rock. To validate theoretical results, we developed a unique low-frequency laboratory setup and experimental methods. We performed drainage and imbibition cycles on a sandstone sample using oil and water as pore fuids. During fooding, the sample was excited by small-strain seismic pulses of a specifc shape, with varying polarization direction, frequency, and amplitude. The preliminary laboratory results validate the theoretical model qualitatively. The proposed model may contribute to environmentally friendly and low-cost methods for stimulating hydrocarbon production in conventional reservoirs without invasive chemicals or fracking. In unconventional reservoirs, however, this technology can be envisioned as supplementary to standard EOR/IOR techniques. Normally, the seismic wave energy is too low to increase absolute permeability or decrease oil viscosity;; only relative permeabilities can be affected.

原文链接:

NSTL

Multiphase Flow Through Chokes-An Evaluation of Frozen, Equilibrium, and Nonequilibrium Flow Models

Carl-Martin CarstensenStig Kare Kanstad

25页

查看更多>>摘要：This work has evaluated the most common published mechanistic multiphase choke models and compared them with a newly developed model. A typical assumption for most of the models is that the fow through the choke is adiabatic because it is argued that there is limited amount of time for heat and mass transfer. Such models are often referred to as “frozen fow” models and are widely used in the industry. The new model deviates from the frozen fow concept because it includes phase transfer. The depressurization through the choke is fast;; hence, thermodynamic equilibrium will not be established and a nonequilibrium coeffcient will account for this intermethate state. The goal was to investigate whether frozen fow and adiabatic expansion through the choke is a reasonable simplifcation or whether the concept of nonequilibrium fow could provide higher predictive ability. The predicted mass rates from all models have been calculated for two experimental multiphase datasets, and the mean relative error, mean absolute percentage error, root-mean-square deviation, and standard deviation have been calculated to compare the performances. The new model seems to be more robust and performs better than the more common frozen fow models. It has better overall predictive potential with a mean absolute percentage error of 2.6% and 5.2% for the multiphase datasets. The new model is generic and can be used with any given thermodynamic process, e.g., equilibrium fow, nonequilibrium fow, or frozen fow. For frozen fow, both adiabatic and polytropic gas expansion can be used, and the model performance will then be similar to previously published frozen fow models such as Al-Safran and Kelkar or Mwalyepelo. Further, a methodology for highly viscous fow has been proposed. All the present models originally assume frictionless fow, and high viscous fow is therefore not accounted for. A Reynolds number correction has been incorporated by adjusting the choke discharge coeffcient. The correction is based on a study of highly viscous fow through safety valves.

原文链接:

NSTL

A water invasion unit numerical simulation model for the distribution of water and water invasion channel in complex edge water reservoir

Xiaobing HanXiaohua TanXiaoping Li

20页

查看更多>>摘要：In order to accurately and intuitively characterize the distribution of water and water invasion channel in complex edge water reservoir, a water invasion unit numerical simulation model (WINS) that can simulate complex edge water invasion dynamics is established, based on well geological data and production data. The model discretizes the edge water into water unit connected to the production well unit, and discretizes the locations with clear geological understanding into dense unit. The more dense units, the more accurate the water invasion dynamics. Based on the water unit, production well unit, dense unit, the material balance method is used to calculate the pressure of each unit, the water flooding front propulsion equation is used to calculate the saturation and water cut of each unit, and the genetic algorithm is combined to modify characteristic parameters of the water unit and the water invasion channel, fit automatically production data, realize the re-identification of water distribution and water invasion channel. Different edge water and high-permeability bands are constructed through numerical simulator, and the simulated production results are substituted into the model to automatically fit. The results show that the edge water distribution and water invasion dominant channel inversely performed by WINS are consistent with the results constructed by the numerical simulator. Compared with the process of constructing edge water at the reservoir by Eclipse numerical simulator, WINS saves the tedious steps of building a model, and its calculation efficiency is higher. Not only that, WINS can intelligently and automatically fit parameters, reducing errors caused by human interference.

原文链接:

NSTL

Incorporating injection stage into DFIT analysis for permeability estimation, and its significance

Y. CaiA. Dahi Taleghani

11页

查看更多>>摘要：Diagnostic fracture injection tests (DFIT) have been broadly used in unconventional reservoirs to derive properties such as initial formation pressure, formation permeability, and closure pressure. DFIT involve pumping a small volume of untreated fluid into the formation to form a small crack. Formation permeability is usually obtained by modeling fluid leakoff during well shut-in. Prevailing studies assume constant fluid pressure boundary conditions on the fracture walls, constant leakoff coefficient, or both. However, the results deduced from these assumptions can introduce tremendous errors because the fluid pressure inside the fracture dissipates fast as fluid leaks off. Therefore, this study proposes a material balance approach to analyze DFIT data and obtain formation permeability. The proposed analysis incorporates fluid leakoff during both injection and well shut-in periods. According to the material balance relationship, volume of water injected during hydraulic fracturing should be equal to the summation of leakoff volume during hydraulic fracturing, leakoff volume during shut-in period and residual fracture volume. Thus, we propose leakoff models for both fracture propagation and shut-in periods. These models are then validated through a few synthetic cases. The model accounting for leakoff during fracture propagation is modified to more accurately calculate the leakoff volume. The model for the shut-in period is solved using the Fourier transform and the eigenfunction expansion method. The results suggest that the proposed DFIT analysis approach can provide a good estimation of the formation permeability. The proposed solution can be used in the field even when pressure data contain significant levels of noise. We showed that more accuracy can be achieved using this method compared to available solutions in the literature.

原文链接:

NSTL

Experimental investigation on the creep behaviors of shale using nanoindentation technique and fractional constitutive models

Xian ShiXuefeng YangWenhan Yue

12页

查看更多>>摘要：Viscoelasticity is a key mechanical feature of shale that can influence a variety of oil and gas engineering issues. To describe the time-dependent creep behavior of shale at nano-scale, the nanoindentation technique was used. Moreover, Maxwell and Zener fractional derivative models were applied for the fitting data and the applicability and the accuracy of these fractional viscoelastic models are discussed. Moreover, the creep difference on samples with different mineral compositions may be compared using Energy Dispersive Spectra (EDS) mapping techniques. The experimental results indicate that there is a better match between fitting values and measured values versus time by fractional models compared to traditional viscoelastic models. Furthermore, fitting parameter changes with different holding times were investigated. The results demonstrate that the fractional Zener model yield the highest RMSE and R , indicating the optimal fitting results. The fractional Maxwell model likewise provides an excellent fit to the data and takes the least amount of time to compute, however, the creep response is infinite. In addition, both elastic parameters and fractional derivatives decrease while viscous parameters increase along with a reduction of Young's modulus and hardness, indicating microdamage accumulation induced by the viscoelastic behavior of shale. Comparing the results of clay-rich and quartz-rich samples showed reduced stiff shale creep, but creep-induced deformation led to more obvious deformation and mechanical strength reduction. This study not only investigates the use of nanoindentation in the creep analysis of composite shale but also shows that fractional models can be used to characterize shale viscoelasticity.

原文链接:

NSTL

Lithology identification by adaptive feature aggregation under scarce labels

Chenhui YuanYuping WuZerui Li

14页

查看更多>>摘要：Lithology identification plays an important role in petroleum exploration. However, due to the high cost of labeling by cores and cuttings, the application of semi-supervised method in lithology identification should be considered. There is a main problem in the past graph-based semi-supervised methods, that is, the connection strengths (i.e., edges) among samples are calculated before training and therefore fixed. However, these pre-handcrafted connections strengths determine the graph structure and therefore affect the performance of the trained model. Obviously, there inevitably exist inappropriate connection settings, which motivates us to consider the dynamic adjustment of these connection strengths via training. First, the graph attention network is used to complete the semi-supervised task in the process of lithology identification. This is the first time to our knowledge that the graph attention network is applied to lithology identification. This method can learn edge weights and mitigate interference due to the error of initial edge weights. Second, two methods are used in graph neural network to establish the connection between logging data and transforming them into graph structure data. Third, through a lot of experiments, the graph attention network using depth similarity (GAT-D) outperforms other algorithms on two metrics (macro-avg-fl score of 73.73% and macro-avg-re call of 80.54% in dataset A). Hence, converting logging data into graph data and then using graph neural network to realize semi-supervised classification could improve the performance of lithology identification.

原文链接:

NSTL

Optimization of CO2 huff-n-pufFEOR in the Bakken Formation using numerical simulation and response surface methodology

Samuel AfariKegang LingBillel Sennaoui

18页

查看更多>>摘要：CO2 huff-n-puff has been identified as a promising method for enhancing oil recovery in unconventional formations. However, the technique has not gained industry-wide adoption partly due to knowledge gaps in several aspects of this method. One such aspect is understanding the influence of critical operating parameters on oil recovery and gas utilization and their optimization. The purpose of this work is to study the impact of five critical operational parameters, namely, injection rate, injection period, soak period, production period and production bottom-hole pressure, on oil recovery and gas utilization. This work leverages compositional reservoir simulation and response surface methodology (RSM) for this purpose. The use of RSM enables the evaluation of the statistical significance of the influencing parameters and, most importantly, their interactions, which has often been neglected in the current literature. We also applied numerical and graphical techniques to find the combination of operating parameters that maximize oil recovery and gas utilization. The Bakken formation is used as a study area. Our findings suggest that the production bottom hole pressure (BHP) and production period are the most influencing parameters on oil recovery. Within our factor design space, the injection rate and period have little influence on oil recovery and may reduce gas utilization when increased. Contrary to several reports in the literature, increasing the soak period negatively influences the oil recovery factor. In addition, we unraveled several significant interactions among these parameters. Finally, numerical optimization indicates that CO2 huff-n-puff in the Bakken formation can yield more than 7% incremental recovery with a gas utilization factor of more than 2 bbl/Mscf. We believe the findings in this work can better guide operators to design efficient CO2 huff-n-puff strategies in the Bakken formation and other unconventional reservoirs.

原文链接:

NSTL

Influences of SO2 contamination in long term supercritical CO2 treatment on the physical and structural characteristics of the Zululand Basin caprock and reservoir core samples

Patience MavhengereNicola WagnerNandi Malumbazo

15页

查看更多>>摘要：Understanding the mineral and microstructure changes in sandstone samples during supercritical CO2 gas treatment is an important aspect of geological CO2 sequestration. To gain insight into the effects of SO2 contamination in CO2 in South Africa, two core samples extracted from the reservoir lateral seal (ZC) and Cenomanian sandstone aquifer (ZG) within the Zululand Basin were studied. The samples were treated with CO2 and CO2/SO2 gas streams at typical reservoir temperature and pressures (10 MPa and 316 K, and 17,5 MPa and 346 K) under supercritical conditions. High pressure Parr reactors were used for storing the rock-gas-water mixtures for up to two months using a watertrock ratio of 23:1. Tests were conducted using pure CO2 and a 99% by weight (wt.) CO2 and 1% (wt.) SO2 mixed gas stream. Due to the prohibitive costs associated with CO2 purification, the knowledge of the consequences of key impurities relevant to geological sequestration is critical. Pre-and post-CO2/CO2-SO2 treatment characterisation was conducted using X-ray Diffraction (XRD) Analyses, Fourier transform infrared spectroscopy (FTIR), and low-pressure gas adsorption (LPGA). Varying mineral alterations were observed in the CO2 treated samples, mainly comprising of calcite, plagioclase and smectite dissolution and the precipitation of quartz, plagioclase, calcite and smectite. Dissolution pores and pore clogging were observed and increased microstructure heterogeneity was reported. Increases in the adsorption capacity, surface area and pore volume were observed in all samples. After treatment with CO2-SO2 gas mixture, increases in mineral reactivity were observed in the ZC sample along with gypsum precipitation, indicating potential improvement in the lateral seal's self-sealing capacity. The introduction of SO2 lead to the increase in quartz and plagioclase precipitation and increased dissolution of smectite and stilbite in the ZG sample. The study presents a novel investigation of the changes expected to take place during CO2 injection in sandstone basins.

原文链接:

NSTL

Top-of-line corrosion via physics-guided machine learning: A methodology integrating field data with theoretical models

Passaworn SilakornNawatjantrakulchaiNapat Wararatkul

14页

查看更多>>摘要：An artificial learning method for predicting top-of-line corrosion (TOLC) in oil/gas production from an existing oil field production data was presented, along with verifications and a few real-world applications. The developed algorithm was aimed to boost fundamental understanding of the day-to-day unit operations and periodic maintenance and inspection of oil/gas pipelines via a particular inline inspection technique (magnetic-flux-leakage measurements) and corrosion inhibitor applications. A recurrent neural network was utilized to establish a model for oil field production while a series of small neural networks representing the fundamental physics of TOLCs were included. The relevant physical models are those relating to (i) electrochemistry of corrosion mechanisms, (ii) gas-liquid phase flows and atomizations, and (iii) the condensation rates at the top of the lines. By incorporating these networks into the operation data, the developed learning algorithm gives accurate TOLC prediction for the complex oil field structures. The results show that the estimation models based on the physics-guided artificial neural network present the root-mean-square error within 10%. The developed software also offers insights into the essential parameters for the better prevention of the TOLC in pipelines.

原文链接:

NSTL

A novel approach to estimate rock strength parameters from multistage triaxial tests (the A-HAS method)

Abo Taleb T. Al-HameediHusam H. AlkinaniShari Dunn-Norman

11页

查看更多>>摘要：Rock strength parameters are vital to creating robust geomechanical models for wellbore stability, sand production, and hydraulic fracturing. Static rock strength parameters can be estimated from single stage triaxial tests (SST) or multistage triaxial tests (MST). Although SST tests are the most accurate way to estimate rock strength parameters, SST tests require at least four to five core plugs to estimate rock strength parameters. Thus, MST tests are used as an alternative method to estimate rock strength parameters from only one core plug to overcome the limitation of materials and save time and money. However, the currently used method to estimate rock strength parameters is the yield-based only method which can significantly underestimate rock strength (especially unconfined compressive strength (UCS)) and can hinder creating a robust geomechanical model. In this work, the authors proposed a novel method called the A-HAS method which is a correction to the yield-based method that is currently used. Using core plugs from three different zones (Nahr Umr shale, Zubair shale, and Zubair sandstone in the Rumaila field, Iraq), SST and MST tests were executed. Rock strength parameters were estimated based on SST tests for each zone (Base Case). Then, rock strength parameters were estimated based on four different methods;; the currently used yield-based only method (Case 1) from MST, corrected maximum compressive strength (MCS) (Case 2 or the A-HAS method) from MST, and Cases 3 and 4 which are based on corrections from both SST and MST tests. The results have shown that regardless of the method used, coefficient of friction (μ_i) and internal friction angle (φ) were not significantly affected while UCS and cohesive strength (S0) were highly affected. Comparing UCS and S0 results of the yield-based method (Case 1) and the A-HAS method (Case 2) to the Base Case, the A-HAS method has significantly outperformed the yield-based method in all three zones tested. The A-HAS method has resulted in errors of 4%, 3%, and 2% for Nahr Umr shale, Zubair shale, and Zubair sandstone, respectively while the yield-based method resulted in errors of 41%, 31%, 83% for Nahr Umr shale, Zubair shale, and Zubair sandstone, respectively. The error of the yield-based method is very significant and may hinder creating robust geomechanical models for wellbore stability, sand production, hydraulic fracturing, and many other applications. The authors recommend utilizing the novel A-HAS method to obtain rock strength parameters (especially UCS and S0) from MST tests and not utilize the currently used yield-based method since the A-HAS has proven to be effective and produced better results in the three tested zones than the yield-based method. By using the A-HAS method, the limitations of the MST tests can be overcome and reliable results from MST tests can be achieved which will contribute to saving time, money, and material.

原文链接:

NSTL