Production from unconventional reservoirs is influenced by well spacing and induced fracture placement as well as the interaction between hydraulic fractures and in-place natural fracture systems. The purpose of modeling of these complex systems is to evaluate how production can be maximized while maintaining operational efficiencies, which promote reduced well pad footprints and effective fracture stage placement.
Comingled flow conduits in unconventional reservoirs exist as amalgamated fracture systems, and multidisciplinary characterization with analysis from geologists, geophysicists, and engineers is necessary to maintain a consistent subsurface representation. To extend model fidelity in the description of complex fracture systems, a workflow was developed to evaluate the spatial constraint of natural fractures based on use of a fault indicator in cases where correlation exists between faults and fractures as a result of exerted structural controls. Use of the fault likelihood attribute in the development of an unconventional reservoir confirms that some of the better producing wells have been completed near heavily faulted zones; however, such increased productivity can also be hindered when pressure communication is established between wells associated with the same fault block.
An examination of an Eagle Ford formation was conducted, highlighting how a consistent subsurface description not only enabled increased efficiency in future wells and hydraulic fracture placement but also promoted reduced drilling and completion costs as well as increased field productivity. This was achieved by combining fault likelihood constrained natural fracture network (NFN) as well as dynamic simulation of the stimulated and external reservoir volume, incorporating a petro-elastic model (PEM) to preserve geologic continuity between seismic attributes and the simulation.
We present the
Levchenko, Pavel (Tengizchevroil) | Iskakov, Elrad (Tengizchevroil) | Chalak, Amanbay (Tengizchevroil) | Zhumagulov, Kanat (Tengizchevroil) | Turmanbekova, Aizhan (Tengizchevroil) | Galimzhanov, Saken (Tengizchevroil)
The PDF file of this paper is in Russian.
Korolev field is a large Devonian-Carboniferous carbonate buildup with a flow system dominated by natural fractures. Currently TCO is looking into potential IOR opportunities at Korolev field, which might help to unlock additional resources beyond the scope of current development plans. Therefore, characterization and modeling of the fracture system is of fundamental importance for a new flow- simulation model to assess and predict IOR performance.
The fracture modeling workflow closely integrates matrix and fracture modeling, which facilitates identification of important parameters for fracture distribution early in the modeling process. Fracture prediction is based on correlations with various geological parameters, such as stratigraphy, depositional facies, mechanical properties and geomorphological features, which provides a soft probability trend for distribution of fracture parameters.
Fracture network characterization based on analysis of well log and core data only is very limited in scale. Pressure Transient Tests (PTT) and Pulse Tests provide important insights into characteristics of fracture network at the larger scale than the conventional wireline data allows. Therefore, it is important to incorporate dynamic dataset as a fracture characterization constraint during modelling of fracture distribution. Most of the wells at Korolev field have good quality pressure buildup and pulse test data. TCO developed a workflow to incorporate dynamic data into the fracture modeling process for the full- field dual porosity, dual permeability (DPDK) model. The first step in the workflow is to calibrate fracture density distribution to match well productivity indices (PI) observed in the field. The next step involves dynamic simulation of pressure buildup tests and their comparison to the actual measured data. The last step is to validate the geologic model with available pulse test data.
Dynamic data integration required multiple iterations and loopbacks to fracture characterization and property distribution. Close collaboration between fracture experts, earth scientists and reservoir engineers along the whole process was essential for successful implementation of dynamic data into fracture characterization and modeling. Calibration with the available dynamic data led to better understanding of spatial distribution of fracture properties and provided important additional constraint for the fracture model construction. Improved fracture model at Korolev is the key factor for more reliable production forecasts and evaluation of future development opportunities.
Trimonova, M. (Institute of Geosphere Dynamics RAS) | Baryshnikov, N. (Institute of Geosphere Dynamics RAS) | Zenchenko, E. (Institute of Geosphere Dynamics RAS) | Zenchenko, P. (Institute of Geosphere Dynamics RAS) | Turuntaev, S. (Institute of Geosphere Dynamics RAS)
In the article the results of laboratory studies on unstable fracture propagation in injection well are presented alongside with the verification of previously proposed model verification using the results of experimental studies. The experimental studies were carried out in accordance with the similarity criterion between the experimentally studied material and the rock properties under the in-situ state. The similarity criterion was chaosen according to the possibilitiesof the experimental complex. The joint study of the possibilities of the experimental complex, mathematical model and similarity theory makes a basis for the application of mathematical model of water-induced hydraulic fracture propagation while modeling the reservoir exploitation process.
In this paper, the effect of local changes in the stress-strain state of rocks is analyzed, and an appropriate approach is proposed for modeling the geometry of hydraulic fractures on horizontal wells under conditions of a dense drilling network, including Zipper Frac technology. The main source of factual information on the development of cracks are the pressure curves obtained during injection tests, mini-fracs and main frac jobs, including fracs on offset wells and microseismic monitoring of fracturing.
At Vinogradov field as the optimal selected row production system. The analysis of the accumulated experience made it possible to identify promising areas for increasing the efficiency of the production system, including reduction in the distance between rows, a change in well azimuths relative to the principal stresses, and the extension of horizontal sections of the wells.
In this paper, we analyze the results of multi stage HF in conjunction with modeling the geometry of cracks in various software products, taking into account and without taking into account the local change in the stress-strain state of the rocks near hydraulic fractures. The simulation results are cross-validated with actual information on the various wells of the development object under consideration.
Modeling and analysis of the geometry of cracks in neighboring wells, oriented both longitudinally and transversely with respect to the direction of maximum horizontal stress, is carried out. Including the mutual influence of stages of Zipper Frac technology is analyzed.
Taking into account the effect of local stress redistribution made it possible to achieve a high degree of correspondence between the results of modeling available actual information from the target wells and the environment wells (pressure responses, microseismic results and production figures). Without taking this effect into account, it was not possible to achieve such a degree of compliance.
The conclusion is made that the effect of local stress redistribution both during fracturing at adjacent ports of a single horizontal well and fracturing of adjacent horizontal wells has significant influence on geometry and, in particular, to the fracture height, which is critical in the geological conditions of the considered object.
An approach to modeling and planning of the multi stage frac on horizontal wells in conditions of complex reservoirs similar to the object under consideration was developed. In particular, new approaches to the optimization of the sequence of Zipper Frac stages are proposed and recommendations are made for carrying out fracturing in the conditions of both longitudinal and transverse drilling mesh in relation to the direction of the maximum horizontal stress.
The urgency of conducting microseismic studies in the process of hydraulic fracturing is considered in the article.
The increased interest in this area in domestic geophysics is a reflection of the current challenges of the oil industry: an increase in the share of non-traditional and hard-to-recover reserves in the structure of the assets being developed. Increased interest of researchers in the issue of microseismic monitoring makes it necessary to develop methodological recommendations for the most effective positioning of field observation systems, as well as for choosing the most correct method for solving the inverse problem. Achieving this goal is impossible without digital modeling, which will allow, in conditions of a priori known solution, to develop recommendations for optimizing each of the MSM.
Within the framework of the article, the first stage of this large-scale work is considered, which consists in setting up a digital experiment - obtaining synthetic seismograms of MSM.
The unsteady recovery of oil and gas prices in early 2017 led to an increase in drilling and hydraulic fracturing operations in North America liquid-rich shale plays. The increasing number of producing wells, in addition to re-fractured wells, impose the need for optimizing field development strategies and shale gas supply chain networks that maximize profitability. Moreover, operators must account for undulating natural gas demands both locally and externally in a persistently low oil price environment. In this paper, we adopt supervised machine learning approaches to forecast local natural gas demand as well as to guide well re-frac candidature. Both issues serve as critical inputs for maximizing the net present value (NPV) for selected field development operations. An optimized water management structure is also incorporated in the proposed framework to account for associated produced-water recycling.
Considering a shale gas network in the Marcellus, supply chain optimization was achieved using a mixed integer non-linear programming (MINLP) formulation. The MINLP formulation relies on at least 4 major efforts including reservoir simulation, which in turn relies on output from a feed forward Neural Network (NN) algorithm. The trained NN algorithm was deemed suitable for recommending re-frac candidates, necessary decision variables for multiphase reservoir simulation. Finally, NPV optimization relied on a four-layer Long Short-Term Memory (LSTM) recurrent neural network, developed for forecasting local shale gas demand. Both neural network algorithms were scripted using python.
The well pads in the case study superstructure are modelled using compositional reservoir simulation. Each reservoir model is tuned by history matching production and pressure data for designated producing wells. Alternative field development strategies (including re-fracturing) are then simulated for a 10-year planning horizon to generate gas and water rate decline profiles. Using the LSTM developed, local gas demand is forecasted using data sets created from multivariate time-dependent local and global variables affecting shale gas demand. The LSTM algorithm is derived by convolving some features from the raw data set, with fined-tuned weights, with the objective of minimizing a pre-defined demand error function. 17 non-redundant parameters were mined for 250 wells in the AOI. The t-SNE technique was then used to visualize related low dimensional manifolds within the high-dimensional data set. A NN algorithm was then used to obtain probabilities of misclassification. The results obtained show that application of this integrated approach can give operators strategic foresight, thereby preventing erroneous feedback of project profitability but allowing early time decision-making that maximizes shale asset NPV.
Haddad, E. (Schlumberger) | Wells, P. (Schlumberger) | Fredette, M. (Schlumberger) | Toniolo, J. (Schlumberger) | Mallick, A. (Schlumberger) | Nguyen, H. (Schlumberger) | Bammi, S. (Schlumberger) | Laronga, R. J. (Schlumberger) | Kherroubi, J. (Schlumberger) | He, A. (Schlumberger) | Gelman, A. (Schlumberger) | Jarrot, A. (Schlumberger) | Fratarcangeli, D. (Tapstone Energy) | Alcorn, T. (Tapstone Energy) | Tipton, T. (Tapstone Energy)
Microelectrical imaging is a well-known and highly versatile geological and reservoir characterization technique that produces representative and photorealistic images of the formations intersected by a wellbore to form the basis of a thorough and reliable geological interpretation. These images are used to characterize geological structures, natural fractures, faults and interpret sedimentary features and rock facies. This paper introduces the world's first through-the-bit microelectrical imaging tool, also the world's smallest tool in the genre, at 2-1/8 in diameter and 140 lbs. The new tool provides the lowest-cost, lowest-risk method to obtain high-quality images in lateral wells for applications such as fracture characterization of unconventional reservoirs.
We present the electrical and mechanical design innovations that enabled repackaging the performance of the industry- standard microelectrical imaging tool into a ‘nano’ format tough enough to withstand the rigors of through-the-bit conveyance, often in laterals that exceed two miles' length. The basic physics of the industry-standard are maintained with some obvious changes to the geometry. A simple and elegant twelve-arm bowspring design maximizes coverage of the borehole wall, while being robust enough to prevent pads from being ripped off downhole, a well-known fault of existing imaging tools in lateral wells. In-pad front-end signal processing of twelve buttons ensures strong signal-to-noise while demanding further innovation in miniaturization. Notwithstanding its diminutive size, the new tool delivers images of 5mm nominal resolution and 76% circumferential coverage in six inch boreholes drilled with water-base fluids.
We discuss implications of the new design for the image data processing chain, as development of significant new and tool- specific processing methods was necessary. For example, the irregularity of tool movement in long laterals, the lack of wireline cable depth measurement during logging, and the multiple pad levels necessitated the application of new depth correction techniques that smartly combine physics-based and image-based approaches. On another point, considering the lack of real-time QC during memory acquisition, the data acquisition strategy was designed to provide comprehensive auxiliary data to give the processor maximum flexibility to quality control and correct the signal processing.
We review the results of seventy-five jobs conducted in North American unconventional wells, and examine the details of specific case studies. In many specific plays, a growing number of operators recognize the geology—in particular the distribution of natural fractures and faults along the lateral, as the key factor in completion performance. We find that the new and efficiently acquired images are a powerful tool to identify and characterize these features, underlining a strategy to eliminate negative surprises and improve lateral completion performance.
Contribution of Natural Fracture Network (NFN) to the productivity of shale wells is a well-established fact. So much so that it is widely believed throughout the industry that all other things being equal, the presence and the extent of Natural Fracture Network in shale determines the productivity of a shale well. While deterministic mapping of the NFN in shale through direct measurements is not an option, stochastic techniques are used to generate NFN to be used in numerical simulation. Among tools used to provide some indication of NFN in shale are wellbore image logs and seismic surveys. However, limitations of these tools for this purpose is well documented. In this paper, we present an alternative technique that makes use of a large number of field measurements and incorporates artificial intelligence in order to generate NFN map in shale.
In this technique Artificial intelligence, infers the presence and the extent of the NFN from well productivity. Parameters that impact the productivity of shale wells are: well construction … trajectory, reservoir characteristics, completion, stimulation/hydraulic fracturing, and operational conditions (wellhead pressure, choke size). Natural Fracture Network falls under the category of reservoir characteristics. In this approach Artificial Intelligence uses 12 measured parameters representing the volume of fluid available for production and operational conditions as well as completion and hydraulic fracturing practices to infer the presence and the extent of NFN.
These field measurements are analyzed on a well by well basis to generate a dimensionless score for the presence and the intensity of the NFN that are then mapped throughout the reservoir using geo-statistics. Through two case studies, we demonstrate that the discrepancies in correlation between reservoir, completion and frac job parameters and the well productivity can be explained through the presence and the extent of the Natural Fracture Network.
The results of this process is (a) a distribution map of the presence and the extent of the Natural Fracture Network throughout the reservoir and (b) a large number of examples with detail comparisons between wells that justify the generated map.
Multi-well pad has been considered as the most efficient horizontal-well drilling technique in unconventional reservoir development since it not only greatly maximizes the oil production, but also significantly reduces environmental impact and operation costs by drilling group of wells on a single pad. To optimize both hydraulic fracture parameters of each well and well placement simultaneously is still largely unexplored and remains to be a challenging task. Conventional optimization techniques, such as genetic algorithm, particle swarm optimization, and differential evolution algorithm are inadequate to optimize production performance in the multi-well pad, because it may take hours to days to run a single reservoir simulation, leading to an unaffordable computational cost for the optimization processes.
To speed up the search process of global optimization in reservoir simulations, a novel optimization framework for computationally expensive simulations is developed based on Bayesian optimization algorithm. The newly developed optimization algorithm constructs a probabilistic model for the objective function and then exploits this model to make decisions about where in search space to next evaluate the function. In this study, Gaussian Process (GP) is utilized to construct the prior distribution over the objective function. Then, the posterior over functions is obtained based on the prior distribution and evaluations of objective functions. Finally, acquisition function is developed through maximizing the expected improvement over the current best from the posterior, allowing us to determine the next point to evaluate in search space.
It is shown that GP Bayesian optimization framework can successfully optimize the hydraulic fracture parameters and horizontal well placement simultaneously in tight oil reservoirs. 19 parameters involve well spacing, well length, fracture spacing, fracture half-length, and fracture conductivity in a four-well pad were optimized and a high net present value (NPV) was achieved. The oil recovery and NPV of the optimum scenario derived through the Bayesian optimization technique are increased by 36.0% and 55.7% respectively in comparison with a field reference case. The proposed Bayesian optimization framework is found to be a promising and efficient optimization strategy, which takes full advantage of the information available from previous evaluations of objective function, in handling the computationally expensive optimization problems.