Drilling in deep high-pressure high-temperature (HPHT) abrasive sandstone pose significant challenges: low rate of penetration (ROP), bit wear, differential sticking, and wellbore instability issues. These issues are magnified when attempting to drill long laterals in the direction of minimum stress. This paper focuses on the use of Managed Pressure Drilling (MPD) and Artificial Intelligence (AI) analytics to improve ROP. MPD is normally used to help drilling in formations with narrow mud weight window, it achieves this by controlling the surface backpressure to keep the annular pressure in the wellbore above the pore pressure and below the fracture gradient. One key benefit of using MPD is that high mud weight is no longer required, since the Equivalent Circulating Density (ECD) is going to be managed to maintain the overbalance. An example of a well that was drilled using MPD solely for ROP improvement is presented in this paper. This well achieved almost double the ROP of a control well, which was drilled in the same formation with no MPD. Essentially most of the drilling parameters used, which include, pump rate, revolution per minute (RPM), weight on bit (WOB), and other drilling practices, are controlled by the people on the rig. Incorporating AI analytics in the equation, help minimizes human intervention and could achieve further improvement in ROP. After the ROP improvement observed while using MPD, both technologies were combined in a well drilling the same formation. An example is presented for the well drilled using both technologies.
Drilling in high pressure high temperature (HPHT) deep gas reservoirs, with multiple shallow different pressure horizons, requires special techniques which include application of Managed Pressure Drilling (MPD), revising casing setting depths, improving casing strength, and refining mud design. This paper focuses on application of MPD in HPHT gas wells and also describes briefly other techniques which can improve drilling performance and reduce non-productive time.
Objectives/Scope This paper summarizes mitigation practices to safely conduct slick line memory production logging (PLT) in high pressure sour gas wells. The paper addresses the enhancements that were made to overcome the lifting force on the tool due to high flow rates and prolonged exposure time. Methods, Procedures, Process Well intervention in a live well poses several safety and operational considerations. These considerations are more critical when operating at or near high-pressure, high-temperature (HPHT) conditions in sour gas reservoirs. Well intervention in these types of reservoirs is associated with specific risks and, consequently, require a modified tool string designed specifically to withstand such a demanding environment. The paper describes all the problems that can occur in performing such jobs, and their mitigation practices that will lead to a successful job. Results, Observations, Conclusions There are several conveyance methods to run PLT in gas wells, such as: slickline, wireline, and coil tubing. In this job, the wireline could not be utilized due to its material which is not compatible with sour gas. Coil tubing is generally run in horizontal wells, but for a vertical well, CT is not a cost-effective option. Hence, slickline in memory mode was chosen. The slickline material is of utmost importance, and must be verified to withstand sour environment. Moreover, the pressure control equipment (PCE), H2S partial pressure, job safety analysis (JSA), job schedule, and job time, are all important factors that were carefully analyzed to successfully perform this job. Novel/Additive Information This paper might be the first in the literature describing how to safely perform slick line memory mode PLT in high-pressure high temperature sour gas environment.
Background Since the introduction of the first electrical resistivity well log by Marcel and Conrad Schlumberger in 1927, the field of petrophysical well logging experienced significant technological advancements [3]. One of the new technologies was Logging While Drilling (LWD), which allows for real time data streaming and acquisition from the initial drilling depth to the target depth. The target depth sometimes reaches more than 25,000 feet, resulting in wealth of captured data [7]. As special logging probes scan given subsurface intervals, a long list of diverse readings is collected as functions of either depth or time [4]. Unfortunately, most of the obtained data cannot be used as is; several processing, calibration and interpretation activities must be performed on the stored raw data to extract useful insights about the penetrated formations [5]. While these data processing activities are plausible for one particular hydrocarbon reservoir using conventional processing techniques, performing field-wide petrophysical studies can be a real challenge. However, big data technologies can be seen as a potential solution as petrophysical data satisfies the main characteristics of big data. Such characteristics include the high volume, velocity, extreme variety of measurement types and formats, and the uncertain veracity of data attained from several vendors and sensors. In this paper, we first review the major challenges limiting geoscientists, geophysicists and petroleum engineers from fully exploiting petrophysical data. Then, we propose a big data-based framework which can help overcome some of these challenges by capitalizing on advanced processing techniques. Finally, we discuss the results of applying the framework on a defined business case.
Measurement of Special Core Analysis (SCAL) parameters is a costly and time-intensive process. Some of the disadvantages of the current techniques are that they are not performed in-situ, and can destroy the core plugs, e.g., mercury injection capillary pressure (MICP). The objective of this paper is to introduce and investigate the emerging techniques in measuring SCAL parameters using Nuclear Magnetic Resonance (NMR) and Artificial Intelligence (Al). The conventional methods for measuring SCAL parameters are well understood and are an industry standard. Yet, NMR and Al - which are revolutionizing the way petroleum engineers and scientists describe rock/fluid properties - have yet to be utilized to their full potential in reservoir description. In addition, integration of the two tools will open a greater opportunity in the field of reservoir description, where measurement of in-situ SCAL parameters could be achieved. This paper shows the results of NMR lab experiments and Al analytics for measuring capillary pressures and permeability. The data set was divided into 70% for training and 30% for validation. Artificial Neural Network (ANN) was used and the developed model compared well with the permeability and capillary pressure data measured from the conventional methods. Specifically, the model predicted permeability 10% error. Similarly, for the capillary pressures, the model was able to achieve an excellent match. This active research area of prediction of capillary pressure, permeability and other rock properties is a promising emerging technique that capitalizes on NMR/AI analytics. There is significant potential is being able to evaluate wettability in-situ. Core-plugs undergoing Amott-Harvey experiment with NMR measurements in the process can be used as a building block for an NMR/AI wettability determination technique. This potential aspect of NMR/AI analytics can have significant implications on field development and EOR projects The developed NMR-Al model is an excellent start to measure permeability and capillary pressure in-situ. This novel approach coupled with ongoing research for better handling of in-situ wettability measurement will provide the industry with enormous insight into the in-situ SCAL measurements which are currently considered as an elusive measurement with no robust logging technique to evaluate them in-situ.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.