The SPE has split the former "Management & Information" technical discipline into two new technical discplines:
- Management
- Data Science & Engineering Analytics
SPE Disciplines
Conference
Theme
Author
Concept Tag
Country
Genre
Geophysics
Industry
Oilfield Places
Technology
Source
File Type
The SPE has split the former "Management & Information" technical discipline into two new technical discplines:
Layer | Fill | Outline |
---|
Theme | Visible | Selectable | Appearance | Zoom Range (now: 0) |
---|
Fill | Stroke |
---|---|
Abstract This work presents experimental studies on a new tool concept to address casing-casing-annulus (CCA) pressure leak challenges in the drilling industry. The new method uses an intervention-type tool that allows for exiting the casing, cleaning cement behind, and injecting any required sealant to block fluid migration on the annular side. Addressing such CCA challenges is essential for increasing the production time and maintaining wellbore pressure integrity. A combination of 3D modeling and experimental studies is used to evaluate the feasibility of the new concept for addressing CCA fluid migration challenges. This study focuses on the development and evaluation of a tool that allows accessing and sufficiently cleaning cement in multiple CCAs. We have successfully tested a scaled tool. This tool can punch a small hole in a casing at a unique angle and clean cement behind it by drilling spirals on the annular side. The new method for accessing the annular side of the casing and cleaning cement behind it has been developed and successfully tested using scaled model rigs. Studies have involved an early proof of the concept in plastic and steel. We have also simulated cement with fluid communication channels behind the casing with a successful attempt of removing it. The experimental test results are being used to further develop a robust, downhole field-deployable tool and method that captures the essential features required to access and operate in CCA areas. The current study suggests that a significant section of cement can be removed by the proposed method: One small-diameter hole is drilled in the casing, and then a cement removing assembly is run in a spiral motion on the annular side of this casing. A suitable sealant can be injected in the created void in cement to stop potential fluid migration. This experimental study suggests that the CCA can be accessed and resealed with a minimum time and equipment if required. This CCA milling-injection system (patent pending) utilizes a novel, easily-deployable tool. This tool enables milling access into the annular side of designated casings, and enables cleaning the cement behind it. The new system only mills one hole in the casing limiting its damage and providing the ability to clean a significant section of the cement at the desired depth. This helps address potential CCA leaks, saves time and cost.
Hitchcock, Graham (Aramco Overseas Company) | Watson, Andrew (OilToolTeq Ltd) | Alshalan, Meshari (Aramco)
Abstract The drilling of a secondary wellbore departing from a primary wellbore, known as a sidetrack, is used for many reasons during wellbore construction. Sidetrack techniques and procedures are well established with all the major service companies. A typical sidetrack operation would consist of setting a deflector (whipstock), cutting a hole in the existing casing (window), and then drilling into the formation at the desired exit angle and direction (kick off) to produce a starting hole (rathole) for subsequent drilling. The sidetrack milling assembly would then be Pulled Out of Hole and the drilling bottom hole assembly (BHA) run in hole to continue drilling the secondary wellbore trajectory. If for some reason, the casing exit and desired departure of the sidetrack is not obtained, this will remain unknown until the subsequent drilling operation with Logging While Drilling (LWD) tools are run in hole and have extended the rathole. The failure to obtain and recognize the desired kickoff departure prior subsequent drilling operations can lead to significant Non Productive Time (NPT) and subsequent rectification costs. This paper describes the development and successful testing of a compact sensor package located near the front of the sidetrack milling assembly. The sensor package is activated at surface and records inclination data during the sidetrack operation. When the sidetrack milling assembly is pulled out of hole the sensor is read wirelessly using a handheld reader once the milling assembly is at surface. The reader gives the operator a visual confirmation, or otherwise, that the desired sidetrack kick off departure has been achieved allowing informed decisions to be made on the next operational stage.
Abstract This paper describes the development and field trial test of a new well intervention service tool for lifting observation wells for oil saturation logging. The Well Intervention Service Tool (WIST) is a high efficiency, slim hole pumping system that is conveyed and powered through a wireline power cable to appropriate depth in order to lift wellbore fluids. A multi-resettable inflatable packer system was developed to work in conjunction with the permanent magnet motor pumping system, which allows the pump to be positioned and repositioned in the well as required. This improves operational efficiency, enabling jobs to be conducted in a single run, rather than multiple runs as would be the case with conventional packer design. A field trial test was conducted to qualify the prototype system. This paper will describe the test result and future improvement plans. This lightweight cable deployed pump system is an alternative to coiled tubing and nitrogen lift jobs, which can often be expensive and logistically challenging, requiring large amount of equipment and personnel at the wellsite to perform the activity. Solutions to the issues encountered during this work have since been successfully implemented in further testing.
Abstract A new down hole system and method to use for releasing stuck pipes is presented. New system design, features and limits are compared to commonly used techniques for releasing stuck pipe showing benefits of the new system when dealing with differential stuck pipe incidents. The new down hole system is capable to deliver much greater forces when compared to jars and other down hole accelerators near the stuck point. This system can generate over 40G`s lateral forces continuously down the hole acting on the stuck pipe area. The system can be integrated into a Bottom Hole Assembly (BHA) and activated once drill string become stuck or run as a part of the remediate assembly. Different aspects of two types of assemblies are described outlining the benefits and drawbacks. The author will discuss in details the background and rationale to the new technology, including a review of differential sticking challenges and functionality of this new system. The new system was compared to the most commonly used techniques for releasing differentially stuck pipe. Previously not releasable stuck pipe forces of over 1,000,000 lb. can now be overcome with the presented new approach to generate down hole forces near the stuck place. Flexibility in system integration and deployment allows for further optimization in BHA design and cost affective fishing operations in dedicated hole sections. This new approach can be implemented to release the most challenging stuck pipe mechanisms in drilling to minimize NPT and cost associated with stuck pipe, remedial operations and sidetracks. Similar approach can be utilized to release differentially stuck pipes, tubing and completions. The novelty of this stuck pipe release system is the entire down hole system and operations of the overall system using new approach to generate large shocks down the hole. Additional novelty is related to flexibility during integration and deployment of this system. Similar to current shock tools, this system can be placed in BHA, fishing type assemblies and also pumped down inside of the stuck drill string to save time and cost.
Mahalingam, Sakethraman (Aramco Overseas Company) | Munro, Gavin (GM Flow Measurement) | Arsalan, Muhammad (Saudi Aramco) | Gawski, Victor (GM Flow Measurement)
Abstract A traditional fixed size Venturi meter has a turndown of about 8:1 under dry gas conditions that may drop to as low as 3:1 under wet-gas flow. When the well conditions change, a replacement of the original Venturi meter with one of a different size is needed. In this paper, we present the design, development and testing of an Adjustable cone meter that has the ability to adapt itself to the flow conditions automatically and provide a turndown of as much as a 54:1 under dry gas conditions and as much as 20:1 under wet-gas conditions. The patented feature of the Adjustable cone meter is the adjustable sleeve that moves over the cone when the flow rate decreases below a preset value causing an increase in the differential pressure across the meter. In addition, traditional Venturi meters have only one differential pressure measurement and the sensor tends to overestimate the flow when there is liquid present in the flow (wet-gas). The Adjustable cone meter has two differential pressure sensors and the second measurement is used to estimate the liquid content in wet-gas. Two meters were manufactured and tested at the National Engineering Laboratory in East Kilbride, Scotland under gas flow rates of up to 18 MMscfd. Based on the differential pressure measurements under varying flow conditions, algorithms were developed to measure the dry gas and liquid fraction. An over-reading model of the meter and a liquid fraction estimation model based on the pressure loss ratio was derived from an additional differential pressure measurement. The model was used to not only to quantify the gas and liquid flow rates but also the estimated error in each measurement. The measurements show that the Adjustable Cone meter is able to provide low uncertainty in both dry and wet gas conditions and met the conditions outlined in ISO 5167-5. The Adjustable cone meter is a much needed innovation in the area of differential pressure measurement.
A new approach to the inversion and joint inversion of geophysical data is described. We take advantage of the domains of local optimization and of the machine learning (ML) or deep learning (DL) technique to generate efficient optimization schemes to reduce uncertainties in the model parameter estimations, exploit the image segmentation capability of DL techniques, and guarantee compliance with the requirement of physics for the wave propagation. The domains of physics driven (Phy) optimization, based on data misfit functionals, and of DL optimization, based on model misfit (loss), are coupled by multiple penalty functions imposed on the common model term of the physical domain such as performed in a joint inversion approach. The procedure is complemented by network retraining with partial inversion results to augment the network knowledge base and enable more physics-oriented DL predictions. After several iterations, the procedure tends to converge to models satisfying both physics and DL optimization schemes by providing at the same time better resolution and accuracy in parameter estimation. The developed method is demonstrated on synthetic and field transient EM data.
Mahalingam, Sakethraman (Aramco Overseas Company) | Munro, Gavin (GM Flow Measurement) | Arsalan, Muhammad (Saudi Aramco) | Gawski, Victor (GM Flow Measurement)
Abstract When the gas flow rate of a well significantly changes, the flow rate can fall below that of the operating range of a traditional fixed size Venturi meter, necessitating the replacement of the original meter with one of a smaller size. However, with an adjustable cone meter, the internal reconfiguration feature allows it to automatically switch from high operating flow range to low operating flow range and there is no requirement to disassemble the meter from the flow line assembly. Adjustable cone meters were designed, developed and tested at the wet-gas flow loop at National Engineering Laboratory in East Kilbride, Scotland. After calibrating the meter with dry nitrogen gas, the meter was tested with increasing amounts of liquid being injected into the flowline, upstream of the meter. The liquid caused the differential pressure measurement on the meter to over-read. Based on the differential pressure measurements under varying flow conditions, algorithms were developed to measure the dry gas and liquid fraction. The data obtained from the tests such as differential pressure, pressure, temperature, liquid density were used to build an over-reading model of the meter and a liquid fraction estimation model based on pressure loss ratio derived from an additional differential pressure measurement. The model was used to not only to quantify the gas and liquid flow rates but also the estimated error in each measurement. The measurements show that the Adjustable Cone meter is able to provide low uncertainty in both dry and wet gas conditions and offers a turndown ratio of up to 54:1 in dry gas conditions. In addition, the automatic adjustment of the meter from high flow to low flow positions avoids the need for manual intervention that involves additional risk and cost.
Machocki, Krzysztof Karol (Aramco Overseas Company) | Marhoon, Zahrah (Saudi Aramco) | Shaarawi, Amjad (Saudi Aramco) | Sehsah, Ossama (Saudi Aramco) | Dixon, Tom (EFC) | Muhammad, Jamal uddin (Weatherford)
Abstract Managed pressure drilling (MPD) is a technology that allows for precise wellbore pressure control, especially in formations of uncertain geomechanical properties (in specific: Fracture pressure and pore pressure gradients). The rotating control device (RCD) is the backbone to the MPD equipment. A new condition monitoring system was developed to improve the reliability of the RCD elements and to eliminate its catastrophic failures during MPD jobs. The new method to increase the reliability of an RCD is based on understanding and quantifying the factors affecting the lifetime of the RCD components. The condition monitoring system was designed to be attached onto the RCD and collect data from an array of sensors during the MPD jobs. Sensors are measuring: vibrations, acoustic emissions, rotation, pipe movement, temperatures and contamination level in the coolant fluid. System is capable to display the measurements in the real time to the operator, giving early warnings to take actions in order to prevent catastrophic failures of the RCD during the job. Data is also recorded to allow further processing and analysis using ML and AI techniques. The authors will discuss in detail the background and rationale to the new technology, including a review of the condition monitoring system, its elements, and functionality. The system design and intended operation will be explained including, sensors and data collection points in the condition monitoring process. No catastrophic failures of the RCD were encountered when the RCD condition monitoring system was installed and running in the field up to date. The measurements collected from the array of sensors and presented in the real time to the MPD operators, allows to monitor changes in condition of the critical RCD elements. From the system design, sensor type, and frequency of data inputs, it was concluded that the quantification of some parameters affecting the lifetime of RCD could be successfully performed in post analysis, using advanced AI techniques. This condition monitoring system can optimize the MPD operations, making MPD jobs safer and reducing the Non Productive Time. The novelty of this condition monitoring system is in the approach of measuring and displaying critical values to the operator during the job and possibility to quantification of the factors affecting the RCD elements lifetime.
Kazei, Vladimir (KAUST) | Ovcharenko, Oleg (KAUST) | Plotnitskii, Pavel (KAUST) | Peter, Daniel (KAUST) | Silvestrov, Ilya (Saudi Aramco) | Bakulin, Andrey (Saudi Aramco) | Zwartjes, Paul (Aramco Overseas Company) | Alkhalifah, Tariq (KAUST)
Abstract Strong near-surface heterogeneity poses a major challenge for seismic imaging of deep targets in arid environments. Inspired by the initial success of deep learning applications to inverse problems, we investigate the possibility of building near-surface models directly from raw elastic data including surface and body waves in arid conditions. Namely, we train a convolutional neural network to map the data into the model directly in a supervised way on a part SEAM Arid synthetic dataset and evaluate its performance on a different part of the same dataset. The main feature of our approach is that we estimate the model as a set of 1D vertical velocity profiles, utilizing relevant subsets of input data from neighboring locations. This effectively reduces the data and label spaces for a more practical neural network application.
As a relatively new technology, although the joint migration inversion technology has great potential in exploiting multiples in the seismic data for both velocity model building and seismic migration, the state-of-the-art joint migration inversion adopts an over-simplified full wavefield modeling engine, which assumes that reflection and transmission effects are independent on the incident angles of the wavefield. As pointed out by previous work, this simplified full wavefield modeling method cannot accurately model the amplitude versus offset effect in the data, and so the inversion results from the joint migration inversion technology are consequently compromised. In this paper, we propose a new full wavefield modeling engine using physically correct one-way propagation, reflection and transmission operators. We first introduce the framework of our theory, and next apply our new full wavefield modeling engine to a synthetic model. Using the wavefields simulated by a finite-difference modeler based upon the two-way wave equation as a control, the success of our new full wavefield modeling method can be clearly appreciated, although some unwanted simulation artifacts unavoidably exist for the time being. Presentation Date: Monday, October 12, 2020 Session Start Time: 1:50 PM Presentation Time: 3:30 PM Location: Poster Station 12 Presentation Type: Poster