I am glad to announce that Dr. Sean Wu, Sr. Wells Engineer of Shell, has accepted our invitation to serve as associate editor of SPE Drilling & Completion (SPEDC). Dr. Wu will be mainly involved in reviewing manuscripts related to drilling dynamics and machine learning.
In this issue, I would like to highlight two papers related to real-time machine learning in drilling and in completion. The first paper is Classifying Cutting Volume at Shale Shakers in Real-Time Via Video Streaming Using Deep-Learning Techniques and the second paper is Near Real-Time Hydraulic Fracturing Event Recognition Using Deep Learning Methods. These two papers demonstrate how machine-learning technology can be applied in real-time drilling and completion to save cost and improve productivity.
Drilling. Human observation of cuttings at the shale shaker cannot provide a consistent evaluation of the hole-cleaning condition and becomes a bottleneck in drilling management. The paper entitled Classifying Cutting Volume at Shale Shakers in Real-Time Via Video Streaming Using Deep-Learning Techniques proposes a real-time deep-learning model to classify the volume of cuttings from a shale shaker on an offshore drilling rig by analyzing the real-time monitoring video stream. Compared with the traditional video-analytics method, which is time-consuming, the proposed model can implement a real-time classification and achieve remarkable accuracy. A deep-learning-based object-detection approach is implemented to help the classification model find the region containing the cuttings flow, and a convolutional-neural-network-based classification model is pre-trained with videos collected from previous drilling operations. Compared with results manually labeled by engineers, the model can achieve highly accurate results in real time without dropping frames.
Hole enlargement while drilling is a necessity to achieve the desired well geometry in certain Gulf of Mexico deepwater plays. The inclusion of a hole enlargement tool in the drillstring increases the chances of dysfunctions such as excessive lateral vibration, stick/slip, and backward whirl vibrations. Enhancing Reamer Drilling Performance in Deepwater Gulf of Mexico Wells presents a novel programmatic approach to model rate of penetration for reamers and improve drilling efficiency. Three field implementations demonstrate value added by the reamer drilling optimization (RDO) methodology.
Freestanding drilling riser (FSDR), a new type of riser in deepwater drilling, though not in commercial use, can significantly reduce the engineering sensitivity to severe weather compared with conventional risers. In their paper entitled Study on the Mechanical Characteristics and Operating Envelope of Freestanding Drilling Riser in Deepwater Drilling, the authors study the optimal installation depth of the near surface disconnection package (NSDP) and the optimal number of buoyancy cans in the FSDR system. Analysis results show that the NSDP should be installed 200 m below the sea surface to avoid strong wave-current profile and enhance the performance of the FSDR in freestanding mode. The optimal number of buoyancy cans is six and the vessel offset should be less than 2% of the water depth to ensure the safety of the FSDR in normal drilling mode.
Completion. It has often been reported that the peak production of a well drilled in tight formations is highly dependent on the fracture-contact area. However, at present, there is no efficient approach to estimate the fracture surface area for each fracture stage. In their paper Fracture Surface Area Estimation from Hydraulic-Fracture Treatment Pressure Falloff Data, the authors propose a method to calculate the fracture surface area on the basis of the falloff data after each stage of the main hydraulic-fracture treatment. The approach is applied to all stages in a horizontal well that exhibit the fracture-closure behavior. It shows some promise as a potential way to estimate fracture surface areas that could allow an early estimate of the expected well performance.
The paper entitled Surface-Modified Graphite Nanoplatelets To Enhance Cement Sheath Durability proposes a novel cement additive made of graphite nanoplatelets (GNPs) for improved hydraulic isolation and durability of oil and gas wells. The purpose of this research is containing or at least minimizing the intrinsic and developed flow paths through the cementitious matrix with the help of surface-modified GNPs that possess high-surface-area-to-volume ratios. In this study, the authors focus on the effect of surface-modified GNPs on the overall mechanical properties of both cement slurry and hardened cement slurry affecting the permeability of cement. They present two dispersion methods based on physical and chemical treatments of the surface properties of GNPs. They found that an optimum 0.2 vol% concentration of acid-functionalized GNPs improves the compressive and the shear bond strength of the prepared cement by approximately 42 and 175% as compared to the plain cement, respectively.
Proppant diagenesis occurs when minerals form on the proppant surface and/or around the embedment crater at high-temperature and/or high-stress conditions. The diagenesis process has not been investigated in the case of carbonate-rich shale formations. Therefore, the objectives of the paper Proppant Diagenesis in Carbonate-Rich Eagle Ford Shale Fractures are to experimentally investigate the proppant diagenesis process during hydraulic fracturing of the Eagle Ford Shale Formation and to determine the role of the proppant in the process. A thermodynamic modeling study was conducted and confirmed the possibility of formation of the observed precipitate and overgrowth minerals at the equilibrium state of the rock and proppant mixture in water. The study contributes to the understanding of the scale formation and the mechanisms that damage fracture conductivity in the Eagle Ford Shale. Results impact the choice of fluid and proppant for fracturing optimization and long-term production sustainability in the Eagle Ford Shale reservoirs.
The paper Near Real-Time Hydraulic Fracturing Event Recognition Using Deep Learning Methods provides the technical details of developing models to enable automated stage-wise analyses to be implemented within the real-time completion (RTC) analytics system. The models detect the hydraulic-fracture stage start and end, identify the ball seat operation, and categorize periods of pump rate. The presented solution provides real-time automated interpretations of hydraulic-fracture events, enabling auto-generation of key performance indicator reports, dispelling the need for manual labeling, and eliminating human bias and errors. It replaces the manual tasks in the RTC workflow/data pipeline and paves the way for a fully automated RTC system.One of the critical issues that occur in many oil and gas wells is the failure of the cement sheath because of debonding from the casing string or from the formation. This results in the formation of microannuli, which can become pathways for fluid migration. Cement shrinkage during setting is regarded as one of the main causes of the formation of microannuli. In the paper Smart Expandable Fiber Additive To Prevent Formation of Microannuli, a new class of polymer-based expandable additives in the form of fibers is incorporated into the cement to compensate for shrinkage and thereby help prevent the formation of microannuli in oil and gas wells. The cement expansion, fluid loss, gel strength, compressive strength, ductility, and tensile strength of the samples containing these fibers are examined using destructive and nondestructive methods. The proposed class of expandable additives can help operators reach sustainable well integrity by increasing the contact stress at the cement–casing and the cement–formation interfaces to prevent fluid migration and the propagation of cracks.
Shilin Chen, SPE Drill & Compl Executive Editor,
Chief Technical Advisor, Halliburton Drill Bits and Services