Lateral scanning logging while drilling data processing using convolutional neural networks

This article discusses a new approach to processing lateral scanning logging while drilling data based on a combination of three-dimensional numerical modeling and convolutional neural networks. We prepared dataset for training neural networks. Dataset contains realistic synthetic resistivity images and geoelectric layer boundary layouts, obtained based on true values of their spatial orientation parameters. Using convolutional neural networks two algorithms have been developed and programmatically implemented: suppression of random noise and detection of layer boundaries on the resistivity images. The developed algorithms allow fast and accurate processing of large amounts of data, while, due to the absence of full-connection layers in the neural networks’ architectures, it is possible to process resistivity images of arbitrary length.

NanoGram Detection of Drilling Fluids Additives for Uncertainty Reduction in Surface Logging

Manual sampling rock cuttings off the shale shaker for lithology and petrophysical characterization is frequently performed during mud logging. Knowing the depth origin where the cuttings were generated is very important for correlating the cuttings to the petrophysical characterization of the formation. It is a challenge to accurately determine the depth origin of the cuttings, especially in horizontal sections and in coiled tubing drilling, where conventional logging while drilling is not accessible. Additionally, even in less challenging drilling conditions, many factors can contribute to an inaccurate assessment of the depth origin of the cuttings. Inaccuracies can be caused by variation of the annulus dimension used to determine the lag time (and thus the depth of the cuttings), by the shifting or scrambling of cuttings during their return trip back to the surface, and by the mislabelling of the cuttings during sampling. In this work, we report the synthesis and application of polystyrenic nanoparticles (NanoTags) in labeling cuttings for depth origin assessment. We have successfully tagged cuttings using two NanoTags during a drilling field test in a carbonate gas well and demonstrated nanogram detection capability of the tags via pyrolysis-GCMS using an internally developed workflow. The cuttings depth determined using our tags correlates well with the depth calculated by conventional mud logging techniques.

Dynamic Formation Evaluation to Reduce Uncertainty and Confirm Completed Intervals in Brown Fields

Uncertainties in fluid typing and contacts within Sarawak Offshore brown field required a real time decision. To enhance reservoir fluid characterisation and confirm reservoir connectivity prior to well final total depth (TD). Fluid typing while drilling was selected to assure the completion strategy and ascertain the fluvial reservoir petrophysical interpretation. Benefiting from low invasion, Logging While Drilling (LWD) sampling fitted with state of ART advanced spectroscopy sensors were deployed. Pressures and samples were collected. The well was drilled using synthetic base mud. Conventional logging while drilling tool string in addition to sampling tool that is equipped with advanced sensor technology were deployed. While drilling real time formation evaluation allowed selecting the zones of interest, while fluid typing was confirmed using continually monitored fluids pump out via multiple advanced sensors, contamination, and reservoir fluid properties were assessed while pumping. Pressure and sampling were performed in drilling mode to minimise reservoir damage, and optimise rig time, additionally sampling while drilling was performed under circulation conditions. Pressures were collected first followed by sampling. High success in collecting pressure points with a reliable fluid gradient that indicated a virgin reservoir allowed the selection of best completion strategy without jeopardising reserves, and reduced rig time. Total of seven samples from 3 different reservoirs, four oil, and three formation water. High quality samples were collected. The dynamic formation evaluation supported by while drilling sampling confirmed the reservoir fluid type and successfully discovered 39ft of oil net pay. Reservoir was completed as an oil producer. The Optical spectroscopy measurements allowed in situ fluid typing for the quick decision making. The use of advanced optical sensors allowed the sample collection and gave initial assessment on reservoir fluids properties, as a result cost saving due to eliminating the need for additional Drill Stem Test (DST) run to confirm the fluid type. Sample and formation pressures has confirmed reservoir lateral continuity in the vicinity of the field. The reservoir developed as thick and blocky sandstone. Collected sample confirmed the low contamination levels. Continuous circulation mitigated sticking and potential well-control risks. This is the first time in surrounding area, advanced optical sensors are used to aid LWD sampling and to finalize the fluid identification. The innovative technology allowed the collection of low contamination. The real-time in-situ fluid analysis measurement allowed critical decisions to be made real time, consequently reducing rig downtime. Reliable analysis of fluid type identification removed the need for additional run/service like DST etc.

Unconventional Engineering Toward Efficient Geosteering and Well Placement - Logging-While-Drilling in an Oil-Based Mud Environment

This paper presents a success story of deploying new technology to improve geosteering operations in an unconventional horizontal well. A new-generation logging-while-drilling (LWD) imaging tool, that provides high resolution resistivity and ultrasonic images in an oil-based mud environment, was tested while drilling a long lateral section of an unconventional horizontal well. In addition to improving the geosteering operations, this tool has proven the ability to eliminate the wireline image log requirements (resistivity and ultrasonic), hence reducing rig time significantly. The LWD bottomhole-assembly (BHA) included the following components: gamma ray (GR), density, neutron, resistivity, sonic, density imager, and the newly deployed dual imager (resistivity and ultrasonic). The dual imager component adds an additional 15-ft sub to the drilling BHA, which includes four ultrasonic sensors orthogonal to each other, and two electromagnetic sensors diametrically opposite to each other (reference figure 1). This new technology was deployed in an unconventional horizontal well to help geosteer the well in the intended zone, which led to an improvement in well placement, enhanced the evaluation of the lateral facies distribution, and allowed better identification of natural fractures. The dual images provided the necessary information for interpreting geological features, drilling induced features, and other sedimentological features, thus enhancing the multistage hydraulic fracturing stimulation design. In addition, an ultrasonic caliper was acquired while drilling the curve and lateral section, providing a full-coverage image of the borehole walls and cross-sectional borehole size. The unique BHA was designed to fulfill all the directional drilling, formation evaluation and geosteering requirements. A dynamic simulation was done to confirm the required number of stabilizers, and their respective locations within the BHA, to reduce shock and vibration, borehole tortuosity and drilling related issues, thereby improving over-all performance. Real-time drilling monitoring included torque and drag trending, back-reaming practices and buckling avoidance calculations, which were implemented to support geosteering, and for providing a smooth wellbore for subsequent wireline and completion operations run in this well. A new generation dual-image oil-based mud environment LWD tool was successfully deployed to show the multifaceted benefits of enhanced geo-steering/well placement, formation evaluation, and hydraulic fracturing design in an unconventional horizontal well. Complexities in the multifunctioning nature of the BHA were strategically optimized to support all requirements without introducing any significant risk in operation.

Carbonate Reservoir Characterization with High-Resolution Logging-While-Drilling Dual-Images in Oil-Based Mud

Fracture evaluation and vuggy feature understanding are of prime importance in carbonate reservoirs. Commonly the related features are extracted from high resolution borehole images in water-based mud environments. To reduce the formation damage from drilling fluids, many wells are drilled with oil-based muds (OBM) in carbonate reservoirs. There are no appropriate measurements to resolve the reservoir characterization in OBM with the existing technologies in horizontal wells—especially in real-time—to make decisions at an early stage. In this paper, we would like to introduce a workflow for geological characterization using a new dual-images logging while drilling tool in oil-based mud. This new tool provides high resolution resistivity and ultrasonic images at the same time. Structural features, such as bedding boundaries, faults, fractures can be identified efficiently from resistivity images; while detailed sedimentary features, for example, cross beddings, vugs, stylolite are easily characterized using ultrasonic images. Benefiting from the dual images, an innovative workflow was proposed to estimate the vug feature more accurately; and the fractures can be identified from images and classified based on tool measurement principles. One case study from the Middle East demonstrated the benefits of this new measurement. A near well structure model was constructed from bed boundaries picked from borehole images. The fractures were picked and classified confidently using the dual images. Additionally, fracture density statistics are available along the well trajectory. The vug features were extracted efficiently, which indicates the secondary porosity development information. Rock typing is achieved by combining fracture and vug analysis to provide zonation for completion and production stimulation. The dual-images provide the capability for geological characterization in carbonate reservoir in an oil-based mud environment. The image-based rock typing helps segment the drain-hole for completion and production stimulation. The reservoir mapping with rock typing provides detailed information for in-filling well design.

Hydraulic Workover Unit Utilization for New Well Openhole Drilling with Directional Motor and Logging-While-Drilling Bottom-Hole-Assembly in M Area

Achieving a number of well targets in M Area is an important objective for MK, one of the oil and gas operators in Indonesia. An economic challenge is present due to marginal gas reservoirs in shallow zone. The conventional swamp rig unit requires significant costs for site preparation work and in some cases no longer fulfils the economic criteria. The objective was to drill the same one-phase well (OPW) architecture as the swamp rig normally drills, but at lower costs using a hydraulic workover unit (HWU). Drilling the 8½-in hole section OPW architecture using HWU was challenging, not only on the equipment rating and capability, but also on the deck space limitation part. The fit-for-purpose directional and logging-while-drilling (LWD) system was utilized in this project consisting of customized low-torque excellent hydraulics drill bit design, a positive displacement motor (PDM) with aggressive bend setting to achieve directional objective (with max 3.8°/30-m dogleg severity), annular-pressure-while-drilling (APWD) measurement to ensure equivalent circulating density (ECD) is maintained, and combined electromagnetic propagation resistivity and sonic slowness measurement coupled with high-speed telemetry measurement-while-drilling (MWD) tool to get an accurate and timely formation evaluation. The HWU deck space limitation was solved by implementing a single combined directional drilling (DD), MWD, mudlogging cabin, in addition to the remote operation control implementation to further reduce carbon footprint. Five wells were drilled safely and successfully in this campaign. Drilling efficiency improved with up to 109% ROP increase as compared to the first well, showing the progressive learning curve and excellent teamwork from all involved parties. The directional bottom hole assembly (BHA) was capable of delivering up to 4–5°/30-m dogleg, not only achieving the directional objective, but also penetrating the reservoir targets with tight tolerances. The drill bit delivered very good ROP, reaching 60.4 m/h (about 66% of average OPW ROP achieved by swamp rig). This campaign also successfully reduced the overall site preparation cost by up to 30%, enabling MK to drill wells that were initially not feasible to be drilled using swamp rig within the time frame and budget. Thanks to the success, this new method is currently under study for industrialization. The HWU drilling campaign provided a valuable learning experience, is considered as a proven drilling method, and served as a benchmark for other operators in Indonesia. HWU drilling has proven to be an efficient drilling method and capable of delivering the one-phase-well. This paper presents a unique case study of new well open hole drilling with the HWU and its applicability in M Area. Most studies in the past were HWU drilling in re-entry or sidetrack cases.

Logging while drilling – efficient time management and reliable base for estimating volumetric parameters of a reservoir

The relevance and economic efficiency of the technology of recording geophysical parameters while drilling are shown. A comparative analysis of the input geophysical data recorded while and after drilling in directional wells is carried out. The main conclusions from the analysis of reservoir properties and net pays identified based on wireline logging (WL) and logging while drilling (LWD) data are presented.

Using Supervised Machine Learning Algorithms for Automated Lithology Prediction from Wireline Log Data

This paper presents advancements in machine learning and cloud deployment that enable rapid and accurate automated lithology interpretation. A supervised machine learning technique is described that enables rapid, consistent, and accurate lithology prediction alongside quantitative uncertainty from large wireline or logging-while-drilling (LWD) datasets. To leverage supervised machine learning, a team of geoscientists and petrophysicists made detailed lithology interpretations of wells to generate a comprehensive training dataset. Lithology interpretations were based on applying determinist cross-plotting by utilizing and combining various raw logs. This training dataset was used to develop a model and test a machine learning pipeline. The pipeline was applied to a dataset previously unseen by the algorithm, to predict lithology. A quality checking process was performed by a petrophysicist to validate new predictions delivered by the pipeline against human interpretations. Confidence in the interpretations was assessed in two ways. The prior probability was calculated, a measure of confidence in the input data being recognized by the model. Posterior probability was calculated, which quantifies the likelihood that a specified depth interval comprises a given lithology. The supervised machine learning algorithm ensured that the wells were interpreted consistently by removing interpreter biases and inconsistencies. The scalability of cloud computing enabled a large log dataset to be interpreted rapidly; >100 wells were interpreted consistently in five minutes, yielding >70% lithological match to the human petrophysical interpretation. Supervised machine learning methods have strong potential for classifying lithology from log data because: 1) they can automatically define complex, non-parametric, multi-variate relationships across several input logs; and 2) they allow classifications to be quantified confidently. Furthermore, this approach captured the knowledge and nuances of an interpreter's decisions by training the algorithm using human-interpreted labels. In the hydrocarbon industry, the quantity of generated data is predicted to increase by >300% between 2018 and 2023 (IDC, Worldwide Global DataSphere Forecast, 2019–2023). Additionally, the industry holds vast legacy data. This supervised machine learning approach can unlock the potential of some of these datasets by providing consistent lithology interpretations rapidly, allowing resources to be used more effectively.

A Solution to Tough Logging Conditions (Lwf): Intro to Tough Logging Conditions, Logging While Fishing, Operation Procedure and Conditions

This review work summarised new generation logging techniques such Tough Logging Conditions (TLC) & Logging While Fishing (LWF) and their advancement in drilling operations. The production of Oil & gas from the stage of exploration to production should need a lot of data for economic and safe operations. The conditions of the sub-surface cannot be simply predicted unless with some measured parameters under the LOGGING term. Logging is defined as a continuous record of Petro’s physical parameters of rock against time and depth. Instead of conventional logging techniques of wireline such as SP, Gamma-ray, Neutron, Calliper log, etc, logging while drilling, logging while fishing set them aside of their extended applications. Logging while Fishing is a new generation technology that allows unfailing operations of logging tool by a special installation even in cut and thread operation also aids economic and time enhancement. Tough logging conditions are a technique applied either when the hole has highly deviated or when you need to control the position of a tool. This project includes the study and interpretation of above discussed new generation logs. These tools offer all types of logging carried out on wireline except the SP logging. Logging while drilling provides real-time measurements of physical parameters while drilling operation itself which avoids an additional running of tools causing trips and sticking of drill pipe. The data is stored in the bottom assembled logging tool.