Deepwater reservoirs often consist of highly laminated sand-shale sequences, where the formation layers are too thin to be resolved by conventional logging tools. To better estimate net sand and hydrocarbon volume in place, one may need to leverage the high resolutions offered by borehole image logs. Traditionally, explicit sand counting in thin beds has been done by applying a user-specified cutoff on a 1D resistivity curve extracted from electrical borehole images. These workflows require multiple preprocessing steps and log calibration, and the results are often highly sensitive to the cutoff selection, especially in high-salinity environments.
This paper presents a new method that estimates sand fractions directly from electrical borehole images without extracting an image resistivity curve or applying any preselected cutoffs. The processing is based on an artificial neural network, which takes the 2D borehole image array as input, and predicts sand fractions with the measurements from all button electrodes. A cumulative sand count can be computed after processing the borehole image logs along an entire well by summing up the estimated net sands. The neural network is trained and tested on a large dataset from wells in a deepwater reservoir with various degrees of laminations, and validated with sand fractions identified from core photos. Upon testing, a good match has been observed between the prediction and the target output. The results were also compared against another sand-counting method based on texture analysis, and showed advantages of yielding unbiased estimations and a lower margin of error.
Se, Yegor (Chevron Energy Technology Company) | Villegas, Mauricio (Chevron) | Iskakov, Elrad (Chevron) | Playton, Ted (Tengizchevroil) | Lindsell, Karl (Tengizchevroil) | Cordova, Ernesto (Chevron Energy Technology Company) | Turmanbekova, Aizhan | Wang, Haijing
Secondary oil recovery projects in naturally fractured carbonate reservoirs (NFR) often introduce uncertainties and challenges that are not common to conventional waterfloods. The recovery mechanism in NFRs relies on ability of the fracture network to deliver enough injected fluid to the matrix, as well as rate and magnitude of capillary interactions within the matrix rock, during which hydrocarbon displacement occurs. The imbibition measurements can be performed in the laboratory using core samples, but due to reservoir heterogeneity, certain limitations of the lab equipment and the quality of the core material, scalability of the core results to a reservoir model can be challenging.
This paper describes the design, execution and evaluation of the’ log-soak-log’ (LSL) pilot test conducted in a giant naturally fractured carbonate reservoir with a low-permeability matrix in Western Kazakhstan, where repeatable and reliable measurements of changes in water saturation were achieved across large intervals (tens of meters) using a time-lapse pulsed-neutron logging technique. Periodic measurements provided valuable observations of dynamic change in saturation and fluid level over time and allowed estimation of the rate and magnitude of imbibition in the slope margins, depositional settings and rock types of interest. Incorporation of the LSL results into reservoir models validated the ranges of water-oil relative permeability curves, residual oil saturation to water, irreducible water saturation, and capillary pressure assumptions. This validation constrained key subsurface uncertainty and updated the oil recovery forecast in several improved oil recovery (IOR) waterflood projects.
Numerous integrative approaches can be taken to link subsurface rock-type characterization to related openhole wireline log attributes. In this study, focus and emphasis was geared towards developing rock-typing models that link depositional environments to petrophysical property space trends and variations to then guide subsurface modeling. Multiple technical paths were taken, and tools used to link observed rock types in full-diameter conventional cores and related measured geological attributes to electrofacies and the refined petrofacies characterization. The data integration used a significant volume of core analytical and openhole wireline log suites including a base suite of triple-combo data (gamma ray, neutron, density, and resistivity) and expanding to include resistivity borehole image data. We present how the addition of various subsurface datasets impacts rock-typing efforts and accuracy. A cluster-based, least-mean-squares analytical result is observed and discussed in an unsupervised model application and is compared to a supervised model application. The relative importance of various attributes is discussed and used to recommend a workflow for Permian-focused rock typing that allows the subsurface characterization to be extrapolated to regional (basinwide) and local (single-well) scales. In short, we focus on sharing a workflow to effectively link core description (sedimentologic observations) and raw log analytics to refine and upscale rock property distributions for use in sequence stratigraphic frameworks, regional basin depositional models and multiscale modeling efforts.
Gowida, Ahmed (King Fahd University of Petroleum & Minerals) | Elkatatny, Salaheldin (King Fahd University of Petroleum & Minerals) | Abdulraheem, Abdulazeez (King Fahd University of Petroleum & Minerals)
Formation density plays a central role to identify the types of downhole formations. It is measured in the field using density logging tool either via logging while drilling (LWD) or more commonly by wireline logging, after the formations have been drilled, because of operational limitations during the drilling process that prevent the immediate acquisition of formation density.
The objective of this study is to develop a predictive tool for estimating the formation bulk density (RHOB) while drilling using artificial neural networks (ANN). The ANN model uses the drilling mechanical parameters as inputs and petrophysical well-log data for RHOB as outputs. These drilling mechanical parameters including the rate of penetration (ROP), weight on bit (WOB), torque (T), standpipe pressure (SPP) and rotating speed (RPM), are measured in real time during drilling operation and significantly affected by the formation types. A dataset of 2,400 data points obtained from horizontal wells was used for training the ANN model. The obtained dataset has been divided into a 70:30 ratio for training and testing the model, respectively.
The results showed a high match with a correlation coefficient (R) between the predicted and the measured RHOB of 0.95 and an average absolute percentage error (AAPE) of 0.71%. These results demonstrated the ability of the developed ANN model to predict RHOB while drilling based on the drilling mechanical parameters using an accurate and low-cost tool. The black-box mode of the developed ANN model was converted into white-box mode by extracting a new ANN-based correlation to calculate RHOB directly without the need to run the ANN model. The new model can help geologists to identify the formations while drilling. Also, by tracking the RHOB trends obtained from the model it helps drilling engineers avoid many interrupting problems by detecting hazardous formations, such as overpressured zones, and identifying the well path, especially while drilling horizontal sections. In addition, the continuous profile of RHOB obtained from the developed ANN model can be used as a reference to solve the problem of missing and false logging data.
Petrophysical analysis of downhole logs requires accurate knowledge of matrix properties, commonly referred to as matrix adjustments. In organic-rich shale, the presence of abundant kerogen (solid and insoluble sedimentary organic matter) has a disproportionate impact on matrix properties because kerogen is compositionally distinct from all inorganic minerals that comprise the remainder of the solid matrix. As a consequence, matrix properties can be highly sensitive to kerogen properties. Moreover, the response of many downhole logs to kerogen is similar to their response to fluids. Relevant kerogen properties must be accurately known to separate tool responses to kerogen (in the matrix volume) and fluids (in the pore volume), to arrive at accurate volumetric interpretations. Unfortunately, relevant petrophysical properties of kerogen are poorly known in general and nearly always unknown in the formation of interest.
A robust method of “thermal maturity-adjusted log interpretation” replaces these unknown or assumed kerogen properties with a consistent set of relevant properties specifically optimized for the organic shale of interest, derived from only a single estimate of thermal maturity of the kerogen. The method is founded on the study of more than 50 kerogens spanning eight major oil- and gas-producing sedimentary basins, 300 Ma of depositional age, and thermal maturity from immature to dry gas (vitrinite reflectance, Ro, ranges from 0.5 to 4%). The determined kerogen properties include measured chemical (C, H, N, S, O) composition and skeletal (grain) density, as well as computed nuclear properties of apparent log density, hydrogen index, thermal- and epithermal-neutron porosities, macroscopic thermal-neutron capture cross section, macroscopic fast-neutron elastic scattering cross section, and photoelectric factor. For kerogens relevant to the petroleum industry (i.e., type II kerogen with thermal maturity ranging from early oil to dry gas), it is demonstrated that petrophysical properties are controlled mainly by thermal maturity, with no observable differences between sedimentary basins. As a result, universal curves are established relating kerogen properties to thermal maturity of the kerogen, and the curves apply equally well in all studied shale plays. Sensitivity calculations and field examples demonstrate the importance of using a consistent set of accurate kerogen properties in downhole log analysis. Thermal maturity-adjusted log interpretation provides a robust estimate of these properties, enabling more accurate and confident interpretation of porosity, saturation, and hydrocarbon in place in organic-rich shales.
A new acoustic tool has been developed to measure formation acoustic properties through casing. This measurement is important for oil and gas production in mature fields, and for wells that are cased without logging due to borehole stability issues. Conventional acoustic logging through casing in poorly bonded boreholes has been a difficult task due to the presence of overwhelming casing waves that mask the formation acoustic signal. To overcome this difficulty, we developed an acoustic tool using dual-source transmitters and the processing technique for the data acquired by the tool. This paper elaborates the operating principle of the new dual-source technology and demonstrates its application to casedhole acoustic logging. By using the dual-source design, the overwhelming casing waves from the poorly bonded casing are largely suppressed. On the basis of the casing-wave suppression and the condition that the formation is acoustically slower than casing, the formation acoustic-wave amplitude is significantly enhanced in the dual-source data-acquisition process. Subsequent processing of the data reliably obtains the acoustic velocity of the formation. The new tool has been tested in many cased wells with proven performance for various cement-bond conditions. The success of this technology makes casedhole acoustic logging an effective operation that can be routinely used to obtain reliable formation information through casing for slow to moderately fast formations.
Depth matching well logs acquired from multiple logging passes in a single well has been a longstanding challenge for the industry. The existing approaches employed in commercial platforms are typically based on classical cross-correlation and covariance measures of two signals, followed by manual adjustments. These solutions do not satisfy the rising demand to minimize user intervention to proceed towards automated data interpretation. We aimed at developing a robust and fully automatic algorithm and workflow for depth matching gamma-ray logs, which are commonly used as a proxy to match the depth of other well logs measured in multiple logging passes within the same well. This was realized by a supervised machine-learning approach through a fully connected neural network. The training dataset was obtained by manually labeling a limited set of field data. As it is unrealistic to expect a perfect model from the initial training with limited manually labeled data, we developed a continuously self-evolving depth-matching framework. During the use of depth-matching service, this framework allows taking the user input and feedback to further train and improve the depth-matching engines. This is facilitated by an automatic quality-control module for that we developed a dedicated metric by combining a few different algorithms. We use this metric to assess the quality of the returned results from the depth-matching engine. The users review the results and do manual adjustments if some intervals are not ideally depth matched by the engine. Those manual adjustments can be used to further improve the machine-learning model. A well-designed framework enables automatic and continuous self-evolving of the depth-matching service.
A key aspect of the developed framework is its generalization potential because it is independent of the signal type. It could be easily extended for other log types, especially when the correlation thereof is not obvious, provided that a sufficiently large volume of labeled data is available. This framework has been prototyped and tested on field data.
Sheng, Xiaofei (Tianjin University) | Shen, Jianguo (Tianjin University) | Shen, Yongjin (Beijing Huahui Shengshi Energy Technology) | Zhu, Liufang (Logging Company of Shengli Petroleum Engineering Co.) | Zang, Defu (Logging Company of Shengli Petroleum Engineering Co.)
Transient electromagnetic (TEM) logging is a promising noncontact method for through-casing formation conductivity measurements. We studied the through-casing TEM logging method based on the processing of TEM logging data measured in a production well. Similar to Doll’s work in borehole induction logging, we presented the expressions of the ‘useful signal’ and the ‘useless signal’ in casedhole logging based on which, the methods of removing the ‘useless signal’ and obtaining the formation conductivity curve are introduced. We analyzed the influence of the casing on the TEM signals, described the characteristics of TEM response signals, and obtained the ‘useful signal’ carrying formation conductivity data. Casedhole formation conductivity curves, which are subsequently compared with the known openhole conductivity log, are obtained by dealing with the ‘useful signal’. We identified the characteristics of casedhole formation conductivity curves, and some problems that need to be considered in their practical application. Due to the influence of the casing, the radial detection depth of the TEM logging tool in a cased hole is small, so the detection result is mainly the equivalent conductivity of the cement ring and formation near the outer casing wall. Although the casedhole conductivity curves are in good agreement with openhole logging results in regular formations, due to the influence of the casing and the changes in the physical environment in the well, complete consistency is unrealistic for these two kinds of curves in all well intervals. Therefore, a thorough analysis is required before practical application. Moreover, the effects of well temperature and casing deformation must be corrected for accordingly.
Borehole measurements are often subject to uncertainty resulting from the effects of mud-filtrate invasion. Accurate interpretation of these measurements relies on properly understanding and incorporating mud-filtrate invasion effects in the calculation of petrophysical properties. Although attempts to experimentally investigate mud-filtrate invasion and mudcake deposition have been numerous, the majority of published laboratory data are from experiments performed using linear rather than radial geometry, homogeneous rock properties, and water-based (WBM) rather than oil- or synthetic oil-based drilling mud (OBM or SOBM).
We introduce a new experimental method to accurately reproduce conditions in the borehole and near-wellbore region during, and shortly after the drilling process, when the majority of wellbore measurements are acquired. Rather than using a linear-flow apparatus, the experiments are performed using cylindrical rock cores with a hole drilled axially through the center. Radial mud-filtrate invasion takes place while injecting pressurized drilling mud into the hole at the center of the core while the outside of the core is maintained at a lower pressure. During the experiments, the core sample is rapidly and repeatedly scanned using high-resolution X-ray microcomputed tomography (micro-CT), allowing for visualization and quantification of the time-space distribution of mud filtrate and mudcake thickness. Because of the size of the core sample, the developed experimental method allows for accurate evaluation of the influence of various rock properties, such as the presence of spatial heterogeneity and fluid properties, including WBM versus OBM, on the processes of mud-filtrate invasion and mudcake deposition. Results indicate that our experimental procedure reliably captures the interplay between the spatial distributions of fluid properties and rock heterogeneities during the process of mud-filtrate invasion.
Determining the potential of shale-gas reservoirs involves an exhaustive process of calculating the volume of total gas, or original gas in place (OGIP). The calculation of total gas relies on calibrating wireline logs to core data, which are considered to be an empirical validation or ‘ground truth’. However, inconsistency in sample preparation and analytical techniques within, and between laboratories creates significant uncertainty in calculating the free- and adsorbed-gas components, which constitute total gas. Here, we present an analytical program performed on samples of core to elucidate the causes of uncertainty in calculation of total gas. The findings of this program are used to propose improved methods of calculating total gas from core.
Free gas calculated from properties, such as porosity and water saturation measured on core, was found to be highly dependent on laboratory analytical protocols. Differences in sample preparation and water extraction methods led to relative differences of 20% in water saturation and 10% in porosity observed between laboratories, leading to differences of 35% in calculations of free gas in place (FGIP).
Adsorbed gas was evaluated using methane adsorption testing to study the changes in Langmuir parameters in samples with a wide variety of water saturations, clay content, and total organic content over a range of temperatures. It was found that the storage capacity of adsorbed gas artificially increased by a factor of two to three when the experimental temperature exceeded the boiling point of water. This increase is related to the expulsion of clay-bound water and subsequent availability of clay surfaces for methane adsorption.
Total gas in place (TGIP) is the sum of free and adsorbed gas volume estimates. The interaction and overlap of pore space between these two volume components are also important to consider. It is proposed to use a simplistic monolayer-based correction of volume of adsorbed gas from the free-gas volume based on a composite pore-size distribution from scanning electron microscopy (SEM) point-counting and nitrogen-adsorption data.
Pressurized sidewall-core samples were acquired at reservoir conditions to measure free- and adsorbed- gas volumes during controlled depressurization under laboratory conditions. This provided a baseline measurement for comparison with calculations from traditional measurements to understand which laboratory protocol and sample preparation technique provided the most robust results.
This study has elucidated methods to reduce the uncertainty in gas-in-place calculations and better understand resource distribution in dry-gas source rocks.