At times, petrophysicists are expected to evaluate potential of the well in time-constraint situations while maintaining consistency of the parameters and interpretation. Other than that, some challenges may also occur when working with older wells where the dataset are not as complete as current wells and processing parameters are not transferable. In this case study, class-based machine learning (CBML) approach is used to perform petrophysical evaluation to identify potential hydrocarbon zones in the target wells. The objective is to find solution to improve efficiency and consistency in those challenging situations. A class-based machine learning (CBML) workflow uses cross-entropy clustering (CEC)-Gaussian mixture model (GMM)- hidden Markov model (HMM) workflow that identifies locally stationary zones sharing similar statistical properties in logs, and then propagates zonation information from training wells to other wells (Jain, et al., 2019). The workflow is divided into two (2) main steps: training and prediction. Key wells which best represent the formation in the field are used to train the model. This approach automatically generates the number of cluster (class) using unsupervised or supervised depending on the input data. The model from key wells data is then used to reconstruct inputs and outputs along with uncertainty and outlier flags. This allows expert to QC and validate the generated class which is the most crucial part of the workflow. Once the model from the key wells has been built, it is applied to predict the same set of zones in the new wells that require interpretation and predict output curves. The result matched well over the good data interval with the petrophysical interpretation result from conventional approach. While in the bad interval, some discrepancies can be observed. The discrepancy was identified easily from the uncertainty and outlier flags which helps petrophysicists to identify which interval to fix or re-evaluate. Some requirements to condition the input were observed (no missing value over the input and outlier) to get the best result. A number of inputs used in the model need to be consistent over the set of wells used in the training and prediction target. This machine learning workflow speeds-up the petrophysical analysis process, reduce analyst bias and improve consistency result between one well to another within the same field. This machine learning application can also generate auto log QC, zonation class for rock typing also reconstructed logs which enrich the petrophysical interpretation even for wells with limited logs availability. This paper offers practical examples and lessons learned of CBML approach application to perform petrophysical evaluation and identify potential zones while being in time-constrained and limited resource situations.
D-01 was an exploration well requiring a Plug-and-Abandonment (P&A) procedure with sustained casing pressure up to 2,000 psi on the B annulus. The presence of Sustained Casing Pressure (SCP) is one of the major technical challenges to decommission and abandon the well safely. Several attempts to secure the well using the perforation-and-squeeze method were performed – but failed. It was decided to perform section milling operations to create a viable rock-to-rock barrier. In this operation, the key factor in determining success, is selecting the correct depth to mill safely and secure the annular pressure source. A comprehensive approach was taken to determine the optimum depth for the section milling by evaluating existing open-hole and cased-hole data. Additionally, triple-detector Pulsed Neutron Log (PNL) was also performed prior to the section milling operation. The triple-detector PNL tool offered not only behind casing porosity (TPHI) and sigma (SIGM) measurement, but also a relatively new measurement in the oil and gas industry called Fast Neutron Cross Section (FNXS), which were expected to provide more accurate gas detection and gauge the condition near the borehole. By combining all the logs and reservoir data, the milling interval was selected and the section milling and subsequent cement plug operations were performed. Evaluation of existing open-hole and cased-hole logs provided geological and petrophysical insights which were useful in determining the hydrocarbon source charging the B-annulus. Further analysis on PNL data provided indication of possible gas pockets in the B-annulus. This information was used to distinguish between shallower formation sources or gas pockets that were not yet bled off. The operation on D-01 successfully resolved the B-annulus issue and the well was properly abandoned per regulatory requirements. Considering the complexity and high cost of section milling operations, a thorough review of data and pre-job logging increases the probability of selecting the optimum intervals needed to successfully complete P&A operations on SCP wellbores.
Nowadays, the stakes of operating hydrocarbon producing wells in mature field are getting higher. Advanced technologies are needed, and the industry must improve the cost efficiency of maturing assets to compensate for declining production and high fixed costs. One of the methods developed for efficient operation in producing hydrocarbon in mature field is one-phase-well (OPW), which is a well architecture without 9-5/8" surface casing. For safety reason, the conventional pressure temperature fluid analysis (PT-FA) logging can no longer be performed in OPW; alternative approaches are required to determine the potential of reservoirs. Pulsed neutron log (PNL) is proposed as a one-stop-shop solution to determine main reservoir characteristics and fluid status in anticipation of future OPW implementation. Latest generation PNL technology utilizes high-counting neutron generator coupled to high-resolution nuclear detectors to measure accurate oil and gas saturation by means of carbon/oxygen ratio (COR), in-situ total organic carbon, sigma, neutron porosity and novel fast neutron cross section (FNXS) measurements, while simultaneously providing accurate lithology volumes and porosity by means of advanced elemental spectroscopy combined with cased-hole porosity (TPHI). To evaluate the robustness and applicability of the method, the latest generation PNL was run in three recently drilled wells which have complete open-hole (OH) logs dataset along with fluid analysis (FA) and mudlog information. The PNL data were processed and interpreted independently, without utilizing any input from the OH log data and without the support of fluid analysis and mudlog information. A criterion based on comparison and correlation between fluid volumes and saturation defined by the PNL and the fluid analysis from FA was defined. For all the wells, the PNL interpretation results matched the hydrocarbon information from FA with success ratio higher than 90 percent over the logged interval, confirming its ability to perform accurate standalone evaluation and its value as alternative technology for effective fluid analysis. Because of the complexity of the system, conditions and potential risks must be properly analyzed case-by-case to encourage more massive application in the future.
The presence of shale in thin beds reservoirs affects formation evaluation where the standard conventional log analyses are not designed to properly correct this effect. The conventional logging tools, with low vertical resolution, are not able to characterize these thin beds. This implies that log values do not represent the true bed or layer properties, but rather an average of multiple beds. Muda Formation are characterized by thin bed layers, made up of clastic rock sequences with dominant lithology of sandstone inter-bedded with shale, siltstone, and organic material as confirmed by drilling cuttings, logs response, and also supported by observation from sidewall cores. There are many uncertainties related to the presence of thin beds, primarily sand, silt, shale or their combination in term of their petrophysical properties and lateral extent. Inadequate reservoir characterization can cause significant amounts of oil and gas to remain unidentified. Accurate petrophysical parameters determination play an important role in the development plan of a field. The lateral and vertical variations in the petrophysical properties of the reservoir lead to different scenarios of the field development. The study of Muda Formation in this structure has integrated the sidewall core and log data. The contribution of the thin sand laminae to the average log response resulted in underestimating the porosity (Ф) and hydrocarbon saturation (Sh). The advanced measurement, like the resistivity anisotropy, proved quite useful as the vertical and horizontal resistivity across these beds leading to measurable electrical anisotropy. The resistivity measured perpendicular to the bedding is significantly higher than resistivity measured parallel to the bedding. The situation occurs due to high resistivity sand layers interbedded with low resistivity shale layers. The true sand porosity and hydrocarbon saturation were calculated using the laminated sand shale sequence and calibrated with core data. The study led to the more realistic petrophysical estimation of the sand shale laminae. A combination and integration of high-resolution image log for sand count, nuclear magnetic resonance (NMR) for porosity evaluation and triaxial resistivity for volumetric model through Laminated Sand Analysis approach are found useful to solve thin bed reservoir issue.
Pertamina EP has recently drilled an exploration well in Akasia Maju Field located in the West Java Basin Area, Indonesia. The well penetrated the main reservoir over clastics play of Lower – Upper Cibulakan Formation. Recognizing these reservoirs from standard logging data (Gamma Ray, Neutron-Density and Resistivity) were very challenging, as they had the characteristics of tight, low contrast, low resistivity, limited to no neutron density crossover, and in some cases, high GR. Consequently, these reservoirs were easily missed on the previous exploration activity. Considering this situation, Pertamina EP has decided to change their wireline logging strategy, from the standard wireline logging suite to the combination of advanced wireline logging suite. This advanced combination was comprised of borehole resistivity image logs, nuclear magnetic resonance, and dielectric logs. These were mainly used to help identify potential reservoirs, optimize DST intervals, and distinguish between high water saturation and residual pay zones. In addition, the formation tester tool was run to measure formation pressures, obtain fluid samples, and determine fluid types and contacts. This combination has successfully revealed the presence of hydrocarbon zones on the intervals that had no clear indication of hydrocarbon occurrence in both sandstone and limestone zones of the Cibulakan Formation. In these hydrocarbon zones, the textural analysis of resistivity image logs indicates that they were correlated either with the well sorted rock fraction (in sandstone) or with the secondary porosity development (in limestone). Meanwhile, the nuclear magnetic resonance reveals potential reservoir candidates. The potential hydrocarbon zones were indicated by the prominence of specific porosity bins. Particularly, porosity bin number 6 (100-300 ms) and porosity bin number 7 (300-1000 ms). This is not usually the case, as hydrocarbon association cannot be determined using porosity bin distribution alone. The formation tester on Upper Cibulakan formation revealed the medium mobility zone in which pressure and fluid analysis data were acquired using the conventional formation tester. However, in the Lower Cibulakan formation, the mobility data showed very low to low mobility zone. Hence, the advanced formation tester was used to obtain pressure data and perform conclusive fluid analysis. Using this data combination, Pertamina EP managed to optimally select four DST intervals in Cibulakan Formation. The DST result was satisfactory, with the maximum oil rate reaching up to 1700 BOPD and the gas rate reaching up to 6.1 MMSCFD, positioning this well as a significant recent hydrocarbon discovery in Indonesia. From this case study, it can be concluded that the use of advanced wireline logging suite yielded definitive results. * Pertamina EP **Schlumberger
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
customersupport@researchsolutions.com
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.