This paper aims to encourage Front End Loading (FEL) to be used more effectively to increase the likelihood of delivering better project outcomes. It introduces a simple and pragmatic approach to assessing FEL which can be carried out in-house. Previous research has shown that, despite FEL being highly regarded, companies regularly sanction projects with insufficient levels of FEL. This has frequently resulted in projects not achieving the outcomes promised at the Final Investment Decision (FID) in terms of cost, time and production attained. This paper reviews reasons why FEL may not be used very effectively at present and proposes a solution to change this. An alternative method of assessing FEL has been developed which: is decision-based; can be carried out internally; and provides clarity on the factors that drive good FEL. In addition to assessing the status of activities carried out in the phase, the decision-based approach emphasises value-creation by considering key factors that could influence an increase or decrease in Decision Quality and thus the value created by the final outcome. The benchmarking and the decision-based approaches to FEL are very different. FEL benchmarking is external, objective, more bottom up; whereas the decision-based approach is internal, subjective and more top down. The benchmarking approach is more detailed, with a large number of individual activities assessed, and the progress on these aggregated to provide an overall benchmarking score. The decision-based approach is more of a big picture view. FEL benchmarking is well proven and its use is advocated. The decision-based FEL approach is unproven, but it has benefits that are not available from benchmarking and avoids some of the disadvantages. It encourages consideration to be given to activities that may result in value being created or destroyed; e.g. ensuring there are sufficient and appropriate alternatives during the Select phase, and that the benefits of flexibility are taken into account. A further advantage of the decision-based approach is that working through the FEL tools as a project team leads to a better joint understanding of the project and improves team integration. The two approaches consider FEL from different perspectives and have different benefits. They complement each other, and so the combination of the two approaches is more powerful than either on its own. It is suggested that the two approaches are worked in conjunction with each other to gain the benefits of both methods, provide a better understanding of FEL, and have a stronger basis for decision-making. A new way of assessing FEL has been developed which uses a decision-based approach aimed at increasing the value of project outcomes. Separate tools are provided for the Concept Select and FID phases.

History matching is a challenging reverse problem due to the inherited uncertainty from Earth modeling. History matching process mandates building one or more sets of representative reservoir models with minimum misfit between the reservoir models and observed field data. There are several commercial assisted history matching tools and workflows to facilitate and orchestrate this process. This paper provides performance assessment and process evaluation between two of the commonly used optimization algorithms: the Evolutionary Algorithm (EA) and the Ensemble Smoother with Multiple Data Assimilation (ES-MDA) in an automated and closed loop workflows. The workflow starts with a parametrization step to identify and quantify the static and dynamic uncertainties in the reservoir model. The objective function is then constructed to guide and measure the convergence of the optimization process. The workflow is then executed to start the model calibration and history matching process. The algorithms, EA and ES-MDA, take different routes during the optimization process. EA works by selecting the best cases (i.e. most fitting genes that resulted in a minimum misfit) in each cycle, and generates a number of simulation cases (new generation) based on the predefined gene selection criteria. The iterative ES-MDA algorithm maintains the same number of simulation cases in each iteration (as many as in the initial Ensemble) throughout the assimilation process. The EA and ES-MDA optimization algorithms have their advantages and disadvantages to the history matching process. For instance, both workflows preserve the geological consistency with the dynamic reservoir models. The automation that both workflows provide incorporates G&G (Geological and Geophysical) workflows to capture, and carryover the static uncertainty to ensure consistency throughout the history matching process. In addition, in both cases, multiple simulation models are generated to cover reservoir static and dynamic uncertainties, which are crucial to the predictive capabilities of the history matched models. Both algorithms, as known, are stochastics in nature (i.e. cannot guarantee one deterministic solution) and, therefore. the results are highly dependent on the initial ensemble. EA is highly effective in searching the solution space, it requires a large number of simulation runs to converge a solution. On the other hand, ES-MDA does not exceed the number of initial ensembles in each iteration regardless of the number of iterations. This paper presents an automated workflow that leads to multiple reliable history matched simulation models. All of the generated models are geologically consistent and capture all model uncertainties that are needed to be preserved and carried forward in prediction. Moreover, this helps in updating the models seamlessly whenever new data is available by simply re-running the workflow with the model updates.

With the current technology, only 3-10% of gas from tight shale is recovered economically through natural depletion, demonstrating a significant potential for enhanced shale gas recovery (ESGR). Experimental studies have demonstrated that shale kerogen/organic matter has a higher adsorption affinity for CO 2 than methane, CH 4 . CO 2 is preferentially adsorbed over CH 4 with a ratio of up to 5:1. This paper examines CO 2 -ESGR in compressible shale during huff-and-puff injection to better understand the parameters controlling its feasibility and effectiveness. A mathematical model will be presented where the CO 2 -CH 4 substitution mechanism is implemented in an injection-production setting representative of field implementation. A single hydraulic fracture surrounded symmetrically by a shale matrix will be considered. During primary depletion, gas is produced from the shale to the fracture by pressure driven flow and gas desorption. The porosity and permeability of matrix reduces due to compressibility. Stopping production and then injecting CO 2 through the fracture and into the shale leads to increase in pressure, causing both the preferential adsorption of CO 2 over CH 4 and the restoration of porosity and permeability. The shale matrix then releases more CH 4 in favor of the more chemically sorptive CO 2 . Restarting production from the well then allows more CH 4 gas to be produced. Experimental adsorption data were used to quantify the substitution parameters while typical field data were used for the operating and geometrical conditions. Additional investigated production mechanisms include gas diffusion and changes in apparent permeability and porosity due to compressibility (addressed in Section 2.3.2). Through a series of simulations, it was found that for the example studied, diffusion contributes the most to the performance of CO 2 -ESGR. Compressibility significantly reduces the matrix porosity and permeability during primary depletion, giving a positive effect to gas recovery. During CO 2 injection, matrix porosity and permeability mostly recovers depending on pressure increase in the reservoir. The gas production seemed little sensitive to fracture permeability. The pore radius, the diffusion coefficient and the rock compressibility, parameters related to flow in the matrix, all affected the performance of CO 2 -ESGR.

In Production operations, asset performance depends greatly on maximizing the run life of equipment while reducing the cost of maintenance. Often, E&P operators have a reactive approach to field maintenance resulting in uneccessary downtime in logistics, inventory management, diagnosing the issues, and in taking the recommended actions. This can lead to higher operating costs and non-productive time. E&P operators are aggressively looking to increase production with operational efficiency gains. In the unconventional fields, a large number of wells have been drilled and put in production with various artificial lift techniques. Proactive well and field production management requires digital enablement of operations, with no data silos and data flowing seamlessly from the subsurface to the hands of the operator. With huge amounts of data being collected, it is imperative to apply data-driven techniques to gain more insights that can be utilized to better manage production. A data-driven approach can provide huge benefits for organisations holding vast amount of reservoir, production, and facilities data. It could provide insights into non-linear multidimensional relationships between parameters so that the field development is better understood and optimized. It could allow companies using a proactive approach towards field operations and equipment maintenance resulting in additional cost savings. This paper presents case studies in which operators optimized production utilizing edge-driven Industrial Internet of Things (IIoT) solutions. These edge IIoT solutions enable fast-loop control through a combination of physics and data-driven workflows, which empowers the operator to proactively manage their assets and focus attention on potentially problematic wells. The solution’s architectural setup and ability to deliver fast-loop control workflows at the edge enables operators to successfully detect and manage potential issues and ultimately improve well performance. Additionally, this approach reduces the dependency upon domain experts to frequently analyze data. The high-frequency data capturing resulted in predicting equipment performance with confidence and allowing remote well management to reduce health, safety, and environment (HSE) risks while decreasing logistics and maintenance costs.

Objectives / Scope Time savings and bit longevity is one of the major challenges in Coal Seam Gas unconventional fields Onshore Queensland. Maximize rate of penetration based on best drilling parameter was the key target to tackle these issues. The dataset was accumulated from more than 80 bits run. A supervised machine-learning algorithm was used to classify drilling parameters that increase ROP and bit endurance. To achieve this goal, our focus was on a) optimizing PDC bit design, drilling hydraulics and b) developing a Drillers Roadmap to ensure optimal parameters were run in the field. Methods, Procedures, Process Operator and Service Company worked closely to select two PDC frames and benchmark performance between them. A novel approach was then taken to capture the rig sensors data. These steps were as follows: Capture big data using sensors integrated on the rig to produce drilling parameters data sets. An algorithm to clean large data sets with the use of Python Pandas libraries was created. Using supervised machine learning and clustering the data to identify which drilling parameters gave the best results. Process the data set using visualization tools to develop heat maps vs depth for each of the drilling parameters (WOB, RPM, flow rates and Torque). Create a Drillers Roadmap identify drilling parameter trends by depth and provide a "sweet spot" based on machine learning for fine tuning operational window. Results, Observations, Conclusions A formal process for optimizing performance was developed. After 6 wells, optimal operating ranges were identified. This process leads the driller to respond to changes in formation and deliver more consistent drilling performance with excellent results. ROP increased from 50m/hr to 150m/hr, driven by improved ROP in the lower section of the well. Time saving of more than 150hrs for the drilling campiagn. Refine PDC bit designs based on optimized parameters, therefore improving bit life through cutter and hydraulic enhancements. Novel / Additive information The innovation of this methodology is to gather drilling information, cleaning the data with Python using Pandas library thus making this process more efficient. After the data files were clean and QAQC, a supervised machine Learning algorithm was used to optimize ROP's by taking drilling datasets and fine-tuning the parameters (WOB, RPM, Flow rate) using drilling dynamics (vibration) and MSE models. Finally, innovative visualizations (heat maps) were produced in order to give a clear insight of the drilling information gathered by the rig sensors to the company men and drillers.

Several extended reach-restricted completion-smart wells required matrix acid stimulation using Coiled Tubing (CT). These wells are equipped with Mechanical Formation Isolation Valve (MFIV) in the tubing end which needs to be reopened by using CT to access mother-bore for performing acid job. CT simulation results for these wells showed coil can reach only 60-70% of open hole interval. Calibrated friction coefficients were used in CT simulation based on the data base developed from other wells in the same area. Only CT-Tractor simulation showed 100% reach with maximum tractor over pull of 3200 lbs from the lock-up depth. World’s slimmest CT tractor was developed in collaboration specifically to address the challenges of extending CT reach past lock-up depth during acidizing and other intervention jobs in minimum restriction extended reach well. The world’s slimmest tractor (2 1/8" outside diameter; OD) was able to provide a pulling force of 3200 lbs and was powered by intervention fluid pumped through coiled tubing. The tractor has many failsafe features and self-cleaning mechanism of gripper arms to overcome the challenges encountered while tractoring in open hole. Four smart wells (with minimum restriction 2.81") were successfully matrix acid stimulated while the CT-Tractor combination were able to extend the reach of CT beyond lock-up depth. CT-Tractor was run-in-hole (RIH) to lock-up depth and then tractor was activated from that depth to extend the reach of CT. At total depth (TD), the tractor was isolated and de-activated by dropping steel ball. The wells were acid stimulated while pulling-out-of-hole (POH). Pre-flush was pumped while RIH at low pumping rate to avoid activating tractor before lock-up depth and after lock-up, pre-flush was pumped through tractor to activate the tractor. This paper outlines the tractor characteristics, matrix acid stimulation jobs design parameters, detailed operational data captured and the learning curve achieved during job execution.

The strong domestic need for oil in China requires further exploration in unconventional reservoirs, such as volcanic and shale oil reservoirs. Sweet zone identification is one of the most critical missions in formation evaluation. The complex mineralogy and the low porosity in unconventional reservoirs result in little contrast of resistivities between oil-producing zones and water zones. Reducing uncertainties affecting the type of fluid in the reservoir and its potential movability is key to defining the optimal landing zones for future horizontal drains. In conventional reservoirs, nuclear magnetic resonance (NMR) logging based on cutoff analysis is the optimum choice for evaluating the porosity and pore geometry of hydrocarbon bearing reservoirs. However, the routine single-dimension T2 measurement is not sufficient for fluid typing in unconventional reservoirs because of the overlap of signals of various fluids in the T2 domain. This contribution presents case studies from a new generation NMR tool providing continuous 2D T1-T2 measurements. The T1-T2 measurements enable separation and quantification of different fluids in the pores. Fluid typing can be done by integration of other wireline logs such as spectroscopy and dielectric. Case studies are presented from volcanic and shale oil reservoirs in Xinjiang Oilfield of PetroChina. In the volcanic reservoir case, dedicated 2D data analytics technic is used to extract relevant signals from 2D NMR T1-T2 measurements. An integrated workflow combining wireline borehole electrical image, nuclear spectroscopy, dielectric dispersion and NMR T1-T2 gives an insight into the fluid composition in the pore system. Besides accurate measurements of lithology independent porosity and pore geometry from NMR, T1-T2 specific measurement of clay bound water volume and total water volume match well with the same measurements estimated from nuclear spectroscopy and dielectric dispersion logs. It lowers the uncertainties on fluid types in the volcanic reservoir where resistivities fail to differentiate wet zones and oil zones. In the shale oil reservoir case, a reservoir producibility index (RPI) derived from the integration of NMR T1-T2, wireline borehole electrical image, and nuclear spectroscopy is shown to be quite efficient in sweet zone identification and ranking. This paper discusses a novel application of a new generation NMR T1-T2 logging method in unconventional reservoirs, which helps the operator ascertain the potential of these reservoirs. The oil zones identified by the new method have promising oil productions. The workflow illustrated in the two case studies can be applied and extended to other unconventional plays in China.

One of the keys to successful and environmentally responsible well stimulation programs in coal seam gas development is to establish consistent procedures for the safeguarding, planning and executing activities across multiple wells. The aim of this paper is to show how a novel application of petrophysical program scripting can be used to make the stimulation process more efficient, consistent and compliant across assets with varying requirements. A macro embedded in petrophysical evaluation software applies a series of rules to rank coals by thickness, allocate a series of perforations and stimulation type based upon coal rank and spacing and then produces actionable treatment schedules which are seamlessly implemented in well stimulation operations at well sites. To do this, the macro grades all coals within the well by thickness based upon a cut-off on the density log, with the thickest coal being graded highest. The macro then identifies the top ranked coal and places perforations based on user defined logic, geological information from offset wells, permeability attributes of the target coal layer(s), depth and vertical separation between adjoining coal targets. Based on the stimulation type assigned, a stimulation schedule is generated that includes estimates of fluid volumes, proppant volumes, injection rates, proppant ramp type and stipulates flush conditions (over-flush or under-flush). Coals thicker than a maximum perforation size are perforated in an upper, middle and lower configuration. Most coals are thinner than the maximum allowable perforation interval and so the macro looks up and down the borehole to include thinner coals within a potential perforation window. The system then generates the stimulation schedule as described above. The macro continues to allocate perforations and stimulation schedules for each validated coal interval and sequentially tries to maximise the total target coal interval along the wellbore. Certain environmental constraints are included in the macro logic to maintain local and regional commitments. For example, coal zones in proximity of permeable non-coal layers i.e. interburden are automatically excluded from stimulation. Multiple advantages of this system have been realised including, a) effective QA/QC as outputs can be directly plotted against the well logs giving the user a quick and easy visual check b) actionable instructions that site based teams can execute including exact perforation depths and stimulation schedules c) provide realistic materials and costs estimates that ensure efficient planning and logistics, d) monitor and document any variations between allocated schedule versus actual execution, e) provide estimate of expected net coal connectivity at a well, development package and asset level which feeds into production and recovery forecasts, f) plan future optimisation studies or pilots and g) most importantly offers a consistent, efficient and compliant framework that can be applied across multiple assets, engineering teams and service providers. This paper focuses on capabilities and advantages of using a macro to automate stimulation design allocation for CSG multi-well (>100 wells) assets. Details of individual stimulation designs for Walloons Coal measures are mentioned in other publications ( Kirk-Burnnand et al., 2015 and Flottmann et al. 2018 ) and hence not covered here.

Bulk permeability of coal is a critical parameter in coalbed methane (CBM) or coal seam gas (CSG) well completion designs and field development planning. The estimation of permeability can be made by well testing either during drilling or production; however, well tests are costly, time sensitive and resource-intensive. Therefore, field-wide estimates are often dependent on production data history-matching, which has a high degree of uncertainty. In this paper, we present a new attempt to apply machine learning approach to estimate coal permeability using drilling data. We first extract important parameters from well test analyses, which are obtained using a packer element testing (PET) tool from four wells in the Surat Basin, Australia. Then drilling data from the wells are processed and fed into different artificial neural networks (ANNs), which include multi-layer perceptrons (MLP), convolutional neural network (CNN), and recurrent neural network (RNN). Two types of models are constructed: regression model with permeability values, and classification model with permeability class intervals (i.e., low, medium, and high permeability values). The evaluation metrics include R 2 for regression models and confusion matrix for classification models. Results show that the drilling mud losses are generally higher in coal layers and lower in interburden formations. The predicted medium and high coal permeabilities from MLP and CNN are in good agreement with measured permeability values from PET data. For the classification task, the CNN achieved an overall accuracy of 99%. Thus, an improved coal permeability map with a higher resolution and less calibration against PET data can be developed quickly to aid production data history matching. The developed machine learning model demonstrated a potential to be applied to new wells to predict the coal permeability for the Surat Basin as well as other CSG appraisal projects. This model can allow more rapid optimisation of well spacing, improved downhole pump design, more targeted well stimulation, and overall project economics.

The Jurassic Plover Formation is one of two reservoirs in the Ichthys Field, North West Shelf of Australia. It consists of fluvial to shallow-marine sandstones, shales and igneous rocks. The objective of this study is to build multiple scenario-based models to optimise development planning in preparation for the upcoming production phase. We have integrated data and interpretations of thin sections, cores, well logs and seismic data to create multiple geological concepts for the field and to identify key geological uncertainties. As the reservoir is geologically complex and many uncertainties were initially identified, it is essential to single out those uncertainties which have a significant impact on the development planning. We have established the key uncertainties and optimal model design for practical use through multi-disciplinary discussions and by running sensitivity models to check the production performance. A rock type (RT) scheme has been devised based on detailed petrographic observations and justified in terms of sedimentology and diagenesis. Using the scheme, a wide range of permeability variations in the sandstones has been captured and modelled. Environments of deposition (EOD) are firstly interpreted at core and well-log scales, then upscaled to the model zone scale. The EOD interpretations are laterally extended using lithology (sandstone, shale and igneous rock) probability maps derived from quantitative seismic interpretation (QSI). Multiple EOD scenarios are generated to capture the possible range of reservoir distributions. Each EOD is characterised by a unique net-sand porosity and RT proportion based on the well data. These values are used to define multiple possible porosity trends and RT proportions, guided by the EOD maps. The distribution and quality of the reservoir sandstones have been identified as key uncertainties. Another key uncertainty is reservoir compartmentalisation, thought to be mainly caused by sheet-like igneous intrusions. Subtle seismic lineaments are regarded as possible indications of such igneous intrusions, and multiple compartmentalisation scenarios have been prepared based on our understanding of igneous activity across the field. Reservoir structure and water saturation are also recognised as key uncertainties. Integrating the key uncertainties, we have established a practical modelling workflow and built multiple scenario-based models to cover a sufficient range of geological uncertainty. The workflow is also adaptive for future history matching, enabling us to flexibly edit the model properties under geological constraints. A decision tree for development planning, which defines a series of decisions for the well sequence depending on the well results, will be prepared based on the multiple scenario-based models delivered in this study. This will enable us to prepare for any potential decision-making in advance. The development planning will be continuously optimised throughout the production phase by simply selecting the scenario-based models most in line with the well results.

Formation evaluation including porosity and directional permeability assessment in complex carbonate formations is challenging due to the anisotropic and heterogeneous nature of the formation. The pore structure can be captured by digital rock technology, which makes the constriction of numerical whole core, and the numerical simulations of chemical enhanced oil recovery on digital rock possible. We introduce a new workflow to construct the digital rock based on micron-CT scanned sequential images of the carbonate sample. Firstly, we apply the modified Otsu's global thresholding method to convert the grey-scale micron-CT scanned images to binary images, and we systematically divide the whole core 3D segmented image into small blocks or representative elementary volume (totally 10143 blocks including inactive blocks). Then we estimate the porosity of each block, and use Lattice Boltzmann Method (LBM) to simulate fluid flow in randomly selected blocks and accurately compute the directional permeability of these blocks by simulating Darcy flow through them. The permeability tensor of other blocks is estimated by applying Kriging method based on the estimated porosity of all the blocks and the accurately calculated permeability tensor of the selected blocks. Then we simulate the polymer flooding on the constructed digital rock and show that the simulation process is computational efficient. We calculate the porosity and permeability of the constructed digital whole core and compare them with laboratory measurements. The image analyzed porosity of whole digital rock is comparable with laboratory measurements by only 1% difference. The simulated permeability value with calibration of the whole digital rock is also in good agreement with experimental result by 3.5% difference. Then we simulate the polymer flooding on the constructed digital rock and show that the simulation process is computational efficient. The proposed digital whole core construction workflow integrates the accuracy of LBM simulation and the efficiency of kriging method. The accurately constructed digital whole core only have 10143 blocks including inactive blocks, which greatly reduces the heavy computation time of chemical enhanced oil recovery simulation on digital rock and makes the proposed method more practical in routine analysis and modeling.

Drilling is considered one of the most challenging and costly operations in the oil and gas industry. Several initiatives were applied to reduce the cost and increase the effectiveness of drilling operations. One of the frequent difficulties that faces these operations is unexpected drilling troubles that take place and stops the operation, resulting in losing a lot of time and money, and could lead to safety issues culminating in a fatality situation. For that, the industry is in continues efforts to prevent drilling troubles. Part of these efforts is utilizing the artificial intelligence (AI) technologies to identify troubles in advance and prevent them before maturing to a serious situation. Multiple approaches were tried; however, errors and significant deviation were observed when comparing the prediction results to the actual drilling data. This could be due to the improper design of the artificial intelligent technology or inappropriate data processing. Therefore, searching for dynamic and adequate artificial intelligent technology and encapsulated data processing model is very essential. This paper presents an effective data-mining methodology to determine the most efficient artificial intelligent technology and the applicable data processing techniques, to identify the early symptoms of drilling troubles in real-time. This methodology is CRISP-DM that stands for Cross Industry Standard Process for Data Mining. This methodology consists of the following phases: Business Understanding, Data Understanding, Data Preparation, Modeling, Evaluation and Deployment. During these phases, multiple data-quality techniques were applied to improve the reliability of the real-time data. The developed model presented a significant improvement in identifying the drilling troubles in advance, compared to the current practice. Parameters such as hook-load and bit-depth, were studied. Actual data from several oil fields were used to develop and validate this smart model. This model provided the drilling engineers and operation crew with bigger window to mitigate the situation and resolve it, prevent the occurrence of several drilling troubles, result in big time and cost savings. In addition to the time and cost savings, CRISP-DM provided the artificial intelligent experts and the drilling domain experts with a framework to exchange knowledge and sharply increase the synergy between the two domains, which lead to a common and clear understanding, and long-term successful drilling and AI teams collaboration. The novelty of this paper is the introduction of data-mining CRIPS methodology for the first time in the prediction of drilling troubles. It enabled the development of a successful artificial intelligence model that outperformed other drilling troubles prediction practices.

The first hydraulic-jet-fracturing multizone completion application was performed in an offshore high-permeability oil well located in the Bozhong oil field in Bohai Bay, China. A detailed case study is presented, highlighting operational challenges and lessons learned from the offshore multizone completion in the Bozhong field. Most offshore high-permeability oilwell completions in the Bozhong field employ tubing conveyed perforation (TCP) followed by frac-packing as multi-zone completions, which is practical in this field. Unlike conventional multizone frac-pack completions, to run perforation guns, the perforating must be completed first, followed by running isolation packers to complete the multizone frac-packing treatments. For hydraulic-jet-fracturing multizone completions, the hydraulic jetting tools are delivered to the target depth using drill pipe; following that, slurry fluid is pumped through the drill pipe to cut through the casing to create fluid flowing channels, which communicate with the pay zone formation. Lastly, the fracturing operation is completed by pumping both proppant slurry through the drill pipe and cleaning fluid through the annulus simultaneously, which creates dynamic hydraulic diversion as functions of isolation between zones. After completing one treatment zone, the tools are pulled to the next target depth, repeating the procedures, until the entire well is complete. After the fracturing treatment, the operator runs production tubing and other accessories, putting the well into production as usual. A detailed completion design and step-by-step operational procedures were prepared based on well data and work conditions to help ensure the effectiveness and safety of the offshore operation. Although some challenges were encountered during the operation, it was overall successfully performed, and significant performance improvement was achieved. After being in production for months, the candidate well has maintained a longer high-production-rate period than the offset wells completed using conventional multizone frac-pack completions. This hydraulic-jet-fracturing multizone completion has proven successful in the Bozhong oil field. The experiences and lessons learned can help benefit future multizone completion optimization for other similar offshore high-permeability oil wells.

The Greater Enfield Project (GEP) is a challenging offshore oil development, designed to produce from the Laverda Canyon, Cimatti and Norton over Laverda oil fields. Six water injection wells are required to provide pressure support and sweep oil to three production wells in the Laverda Canyon and Cimatti oil accumulations to improve oil recovery. The GEP injection wells are a critical aspect of water flood design in a complex field, new to Woodside and with limited global benchmarks. A specific drilling and completion fluid system (Reservoir drilling fluid, completion fluid and chemical filter cake breaker) combined with a unique clean up and displacement technique have been adopted to provide high and sustainable matrix injection performance. While filter cake breakers have been previously used in the industry, they are typically combined with a flowback for filter cake removals. Filter cake clean up by means of flowback was discounted for GEP due to cost and the inability of some wells to naturally flow during early life. All GEP injection wells were completed in 2018-2019, one of which is globally the longest horizontal water injection well completed to date based on the Rushmore data base. Fieldwide injection commenced in July 2019 with favorable results. This paper summarises the key design aspects adopted to deliver successful matrix injection performance, presents the improvements implemented during offshore execution and provides an insight into the early life injection performance.

As the oil and gas industry enters the digital era, openness is a key enabler to realizing the vision of transforming the industry for the better. The practice of reservoir engineering and reservoir simulation is no exception. In this paper, an openness mechanism in a reservoir simulator using Python scripting language is introduced. It empowers engineers to utilize simulation in new ways. It extends simulator capabilities and enables people to implement flexible-control logic to solve field management challenges. The new openness mechanism in the simulator allows engineers to program and include Python scripts in a simulation model. These scripts interrogate and interact with the simulator. The scripts are executed by the simulator while running the model. Flexibility is available to execute the scripts at every Newton iteration, before and after every simulation timestep, at specified times, or when a criterion is met. Simulation model properties can be queried through the scripts, such as well connections, well properties, group properties, grid region properties, network entities, current simulation time, etc. The scripts can set properties such as well constraints, well and connection productivity index (PI), group production target, pausing or stopping the run, etc. Customized control logic, if not directly available in the simulator, can be implemented in the scripts that interrogate and drive the simulator. Such customization can be packaged as Python libraries and shared with team members, enabling continuous value creation. Public Python libraries, such as NumPy, pandas and pywin32 or win32com, can also be loaded in the scripts to extend even further what the simulator can do. The openness mechanism is demonstrated on case examples. They include customized action to acidize wells when production drops, approximating the geomechanical effect of decreasing well pressure, modeling the effect of fines in injected water on well injectivity, and connecting to a network simulator. Examples are also given on customized reporting for model diagnostics and result interpretation, setting production constraints based on economics calculated in an Excel sheet with complex fiscal regime, advanced gas accounting, management of sulphur content, dual-optimization to meet gas demand while honoring oil treatment capacity, and integrated asset modeling from reservoir to surface networks to processing facilities. the ability to extend built-in functionalities of a reservoir simulator and customize field management controls using user scripting language. It embraces innovation and enables continuous value creation in reservoir simulation.

The emergence of Real-Time Well Construction Performance Monitoring Centers has significantly improved the service delivery for Operators across numerous offshore oil fields in Australia. New technologies and workflows were implemented for 3 Australian offshore wells, with the primary objective being to improve drilling efficiency while managing the associated risks. Additional objectives included optimizing daily operational performance, thus delivering time savings for the Operator and highlighting areas of possible improvements. The workflows combined proven technologies, software and wellbore surveillance services, which delivered risk-free wells ahead of Authority for Expenditure (AFE), by targeting the reduction of nonproductive time (NPT) and invisible lost time (ILT), together with risk management. The workflow began with a preliminary field analysis to identify the main challenges and areas that needed improvement. The key learnings were then consolidated on a comprehensive drilling road map by well section that outlined prevention and mitigation measures, expected rate of penetration (ROP) and recommended drilling parameters. The study also identified performance benchmarks such as average and on bottom ROP and connection time. Key performance indicators (KPI) were developed to monitor progress and track the well with respect to well improvements versus the benchmarks identified. The selection of the monitoring tools and modules required during the Execution Stage were developed based on the risks and KPIs identified during the field study and were further tailored to target specific challenges such as stuck pipe prevention, hole cleaning, shock and vibration and connection time. The workflows enabled the flawless execution of 3 offshore wells 29.0 days ahead of AFE which saved the Operator USD 15.5M. The closed loop monitoring enabled real time interventions which aided in preventing risks such as stuck pipe, ensured efficient shoe-to-shoe drilling and avoided potential lost-in-hole (LIH) costs. This case study demonstrates the use of a novel approach to increase well construction efficiency by eliminating lost time while enhancing risk control. The main basis for success of this workflow was using existing and cost-effective technologies while capitalizing on the renewed synergy between different departments such as Drilling, Mudlogging and Well Operations analysis. The workflow was and can be customized based on different needs, by combining specific modules for monitoring, analysis and wellbore surveillance services, to increase the efficiency of any well construction.

The focus on digitalization and how to reap the benefits from it, is increasing in the oil and gas industry. To make this happen subsea by using machine learning and neural networks on mechanical products, we need to understand how we can digitize them. In this paper we will show how we can generate a true digital twin that could enable the step change in how we monitor and understand the mechanical products placed out of reach for normal preventive maintenance. We look at the industry’s current information philosophy regarding mechanical products; from idea to development and testing to installation and operation. Where the products are currently designed to end up with a nominal 3D model, test components and the verification that they are fit for purpose is made through a test scope simulating certain scenarios defined by the industry. The products are then installed, and their condition is assumed to be fine until evidence of the opposite is apparent, and the time window to perform mitigative intervention is gone. Creating true digital twins of mechanical products will require more data. In recent tests we have focused on generating data points to understand the response/behaviour of the products in multiple scenarios and utilized this data to describe the behaviour accurately and numerically; from this we can generate a digital representation of what state the products are in. The results are based on recent work in developing a digital twin for mechanical products. In this work we have proven through documented testing and quantitative analysis that we can generate a validated numerical model for mechanical products. This will form the basis for understanding the state of the products and predict when intervention/maintenance is imminent for the operator. Using this model and method, condition monitoring of the mechanical products can now be enabled with relative few datapoints extracted in time intervals, and through the use of its numerical representation, establish an actual condition history of the product itself. Further this model would enable a deeper understanding of the actual operational effects, to which the product is exposed during its lifetime, leading to more precise and cost-efficient industry requirements and system knowledge.

The Sangomar Field Development – Phase1, located offshore Senegal, will consist of 23 wells targeting stacked reservoirs, with horizontal lengths up to 1,500m. Due to the long drilling campaign, wells are likely to be drilled after first production, targeted for 2023. At the time of drilling, reservoir pressure in these wells will be affected by nearby producers and injectors, which will reduce the mud window in the boreholes, affecting the drillability. Vulnerable wells are identified by a deterministic borehole stability assessment based on mud weight windows and a probabilistic well-by-well analysis of reservoir pressure at the time of drilling. The estimated mud weight windows are unique to each well and a function of completion type, overburden thickness and well azimuth. Capturing reservoir uncertainties within the uncertainty framework results in 216 models which are reduced to a more manageable 20 models for simulation using a distance-based K-medoid algorithm. Maps of maximum depletion and inflation are generated for each of the medoids. With this work, the importance of probabilistically assessing reservoir pressures is demonstrated in the context of developing a robust drilling sequence, highlighting that results can vary significantly depending on the reservoir models used. Potential magnitudes of depletion and inflation in some scenarios are significantly greater than initially anticipated. The combination of borehole stability assessment and probabilistic well-by-well analysis of reservoir pressure allows clear identification of wells at increased risk of borehole stability problems caused by injection or depletion. Subsequently, the drilling sequence is updated to mitigate borehole stability risks by executing potential problem wells prior to first production and prioritising key water injectors, enabling management of post-production depletion while continuing to meet the other objectives such as maximising early production potential. This paper outlines an innovative workflow which captures the range of subsurface uncertainties to define probabilistic impact of depletion and injection on borehole stability. Within the literature we have not seen any examples of depletion and inflation being assessed probabilistically on multiple models or using a map-based format. Depletion and inflation are also often discussed in the context of producing fields, not during the first phase of drilling, as is the case for the Sangomar Field.

The sustained increase in global demand for cleaner fuels continues to drive the gas industry growth. Liquefied natural gas (LNG) has been a key enabler for this growth by making sizeable remote gas re-serves, which are unreachable by pipeline, accessible to the major and emerging gas markets. Every segment of the LNG supply chain has its own set of technical challenges. On the upstream side, many gas resources require significant pre-treatment before liquefaction, and the feed gas to the LNG facility is typically a mixture of various compositions from multiple sources; this composition mix evolves over the life of the project. The main challenge is development planning for the contributing reservoirs under the constraints imposed by the processing facility– managing reservoir deliverability, scheduling & sequencing of wells, and downtime management while maintaining the inlet stream specification. To aid with long-term planning for such assets, a virtual field management system is needed that can emulate a real-world hydrocarbon producing asset by capturing all operational constraints, resource lim-its, and complex operating logic. This paper describes a comprehensive field management framework that can create an integrated vir-tual asset by coupling reservoir, wells, network, and facilities models and provides an advisory system for efficient asset management. The field management component can replicate any operational logic, exercises holistic control over the sub-surface model, integrates with the surface network model, and provides optimization capabilities. This paper demonstrates this for a complex LNG asset that is fed by sour gas of different compositions from multiple reservoirs. We describe the different levels of constraints the asset needs to operate under, including treatment plant capacity, the LNG production capacity, the contractual LNG specifications, the disposal of gas impurities and imposes them on the model by utilizing a flexible and extensible logic framework. Con-straints applied at different levels can be mutually competing and their combination with recovery opti-mization goals increases complexity. The unified field management system uses a robust scheme to bal-ance the coupled system under these constraints while optimizing overall recovery. The optimization is enabled through the ability provided by the field management system to query and retroactively control flow entities during the simulation at the desired frequency. Customization through scripting was necessary to implement this advanced logic and was enabled by the extensible nature of the field management framework. This extensibility, along with native capabili-ties, ensures that any level of complexity can be captured, and the workflow described in this paper can be applied to any hydrocarbon producing asset for short-term and long-term development planning.

There was always the challenge to match Permeability estimated from logs to Permeability derived from well tests. The main reason for this is the difference in scale, both vertically in the borehole (net contributing sands) as well as radially out into the formation. The well test that takes into consideration many hours of fluid flow into the borehole was always deemed more representative than permeability derived from logs that measures a smaller area and volume. However, this perception should not relegate log-derived permeability to an insignificant parameter within dynamic models. When wells are stimulated prior to well test (e.g. under-balanced perforation or chemical stimulation for clean-up), it is expected that the permeability from the well test be enhanced as seen in Iago-2 and Gorgon-3 wells. This paper takes core data from wells in the Carnarvon Basin, create log relationships to predict permeability that match these core data, and compare these to wireline formation tester mobility and to well tests. The results are very promising, and the workflow proposed can be applied to any well in the basin. One of the objectives of this paper is to create a workflow that can be replicated easily and to use raw logs that are available across all wells, in order to reduce the uncertainty in the predicted permeability. The reservoir sands of the Mungaroo formation are easily recognised by the cross over in the neutron-density logs, in both gas zones and water zones. This is the criteria used by operators to obtain formation pressure tests (MDT/RFT) and this is the same criteria used in this paper to define reservoir sands. Only those core data acquired in reservoir sands are used as the "Learning" dataset to predict permeability. Several learning datasets were created, and these were blind tested on other wells that were not part of the learning dataset. The results of these predicted permeabilities were cross plotted against core permeability that have been over-burden corrected and depth shifted to wireline logs. Where the match is not satisfactory, new learning datasets are derived and this step of the workflow is repeated. At the end, there are four groups of learning datasets that are used as the final results. These four groups of datasets are associated with four sets of equations and these provides a very good match between the predicted log-derived permeabilities and core plug permeabilities. When compared to mobilities from formation pressure testers, the predicted permeabilities are a very good match. These were then compared to well test permeabilities showing an overall good match. This gives confidence that the predicted permeabilities from the four sets of equations are good and can be applied to any new well targeting the Mungaroo Formation in the Carnarvon Basin.