|Theme||Visible||Selectable||Appearance||Zoom Range (now: 0)|
Accurate and frequent mud checking is essential for optimum drilling operations. Careful measurement and maintenance of drilling fluid properties (density and rheology) maintain the primary well control barrier and optimize fluid hydraulics and hole-cleaning ability. However, a full mud report while drilling is provided only once or twice a day. Moreover, the measurements are mostly performed using traditional equipment. Test interpretation is subjective and might be biased and difficult to verify independently.
In this paper, we present an automated mud skid unit (MSU), which performs continual drilling fluid sampling and measurements at variable temperatures. The unit provides the non-Newtonian rheological constants characterizing a yield-power law fluid as well as the real-time friction factor and critical Reynolds number using a pipe-viscometer measurement approach. Other important fluid properties such as pressurized density, oil/water ratio, and temperature are provided using high-quality in-line sensors. The unit is controlled by a programmable logic controller coupled with a Linux operating system for data analysis. The system sends real-time data to WITSML data servers and provides detailed mud reports to engineers working either on-site or remotely.
The MSU was deployed in the Permian Basin by an independent operator for automated mud monitoring during unconventional shale drilling operations. Rheology, density, and phase content measurements were compared with conventional mud reports provided by the on-site mud engineer. High accuracy (error<5%) was observed in mud rheology tests. The pressurized mud-density measurements provided by the MSU proved to be more accurate than nonpressurized mud balance measurements, which were affected by mud aeration. Moreover, the MSU provided mud check data 25 times more frequent than those generated by the mud engineer at temperatures of 50 and 65.5°C. Drilling-fluid-related issues, such as chemical overtreatment as well as sudden changes in mud density, rheology, and oil/water ratio, were reported immediately to the drilling crew. This paper provides details about the measurement technology as well as the results from the field deployment of the MSU.
D'Angelo, John (The University of Texas at Austin) | Pehlivanturk, Can (The University of Texas at Austin) | Ashok, Pradeepkumar (The University of Texas at Austin) | van Oort, Eric (The University of Texas at Austin) | Shahri, Mojtaba (Apache Corporation) | Vasicek, Alan (Apache Corporation) | Behounek, Michael (Apache Corporation)
Accurately estimating the directional tendency of a drill string is important to ensure tracking of the planned trajectory and to prevent over/undershooting targets. Predicting this tendency is often achieved through calibration of a physical model of wellbore propagation using previous directional drilling data. However, the amount of relevant prior information to use and the uncertainty in future predictions is not always clear.
The goal of the work reported here was to determine the amount of prior directional drilling data to use for calibration, and the frequency at which recalibration should occur to ensure accurate predictions of wellbore propagation. A procedure was designed to provide this information for a given wellbore propagation model in a given field. The particular model used in this study is a quaternion-based wellbore propagation model that describes the motion of the bit as a rigid body in space which moves based on fixed angular velocities and drilling inputs given in the form of a traditional slide sheet. However, the procedure can be used for any wellbore propagation model that requires calibration on prior directional drilling data.
The procedure consists of two tests. The first test (hold-out test) shows how the prediction error varies over an increasing amount of calibration data while holding the testing-set size constant. The second test (prediction test) shows how the prediction error varies over an increasing testing-set size while holding the amount of calibration data constant. These variations in prediction error can then be examined by looking at the spread of data over these tests.
The test methodology was applied on field data obtained from North American wells where bent sub motors were used for deviation control. The first test showed diminishing returns when incorporating more actions into the model calibration. In cases where bottom-hole assembly (BHA) changes were observed, calibrating on a larger number of prior actions slightly increased the test error. The second test allowed us to approximate the error propagation rate of our model predictions in the current field. The procedure described in this paper allows for such information to be quantified and visualized easily to improve the directional driller's understanding of the needs and reliability of the wellbore propagation model being used.
Optimum well construction operations require frequent and accurate measurements of drilling fluid properties. Optimized hydraulics, hole cleaning and well control requires proper management and characterization of fluid rheology and density. Currently, the methods by American Petroleum Institute (API) standards are used for drilling fluid property measurements. However, these methods are sporadic, labor-intensive and the data quality is highly dependent on the practicing engineer. These dated methods need to be replaced and improved by an automated, continuous and practical way of measuring and monitoring drilling fluid properties.
In this paper, we introduce a helical-pipe viscometer system for automated mud measurements. An automated flow-loop was constructed to perform experiments with Newtonian and non-Newtonian fluids. Flow tests were conducted in two straight- and two helical-pipe test-sections. Frictional pressure loss data was obtained from each section simultaneously. Flow rate, density and temperature data were obtained using a Coriolis flow meter.
It was observed that the toroidal geometry results in secondary flow due to the centrifugal forces. Therefore, elevated frictional pressure losses and delayed transition to turbulent flow were observed in the experiments from the helical-pipe test-sections. A machine-learning regression model was trained using the results from the experimental data for friction factor predictions by the helical pipe pressure loss measurements. The proposed approach was compared with both the experimental data and other theoretical equations reported in the literature. It was observed that the existing correlations in the literature for non-Newtonian fluids are not accurate for the characterization of Herschel Bulkley fluids. Using the trained machine learning model, an algorithm was developed and tested to prove the applicability of the system for automated rheological property determination. Very good accuracy was observed when comparing helical pipe viscometer measurements with standard API measurements.
Helical pipe viscometers offer distinct advantages over standard pipe viscometers, such as compact size and more versatile frictional pressure loss profiles. They are also of interest for the hydraulic design of coiled tubing and slim-hole operations, where the maximum allowable flow rate is often limited by frictional pressure loss.
One of the primary functions of the drilling fluid is to transport cuttings from the bit to the surface. This transport is mainly a function of fluid properties (rheology and density), pump rate, wellbore trajectory and geometry, drillstring rotation and eccentricity. Because these interrelated parameters all have a significant effect on cuttings transport, it is important that hole cleaning modeling is comprehensive and inclusive to yield useful, reliable results. Inaccurate models that either do not consider significant effects or oversimplify them can lead to inadequate hole cleaning advice, resulting in non-productive and invisible lost time.
A novel cuttings transport model that includes relevant effects such as pipe rotation, eccentricity and the annular blockage is presented. Local velocity profiles are calculated for a given fluid and compared against the local critical velocity for cuttings transport. Then, the location and the magnitude of annulus blockage are numerically assessed. After an annulus becomes partially blocked with cuttings deposited in a cuttings bed, continuity and momentum equations are solved for the blocked annulus to estimate the new local velocities. The annulus is divided into small sections and the pressure profile is calculated with the progression of time. By doing so, a realistic representation of a wellbore's pressure and velocity profiles is obtained.
The results from this study are in very good agreement with large-scale flow loop experiments and field observations, e.g. showing that axial flow alone is not enough for effective hole cleaning in high deviation and horizontal wells. The simulations show that in such wells the cuttings tend to deposit even at elevated axial flow rates without pipe rotation in a fully eccentric annulus, due to a significantly reduced flow rate where the drillstring is close to the borehole or previous casing string. The results also provided a way to predict "pack-off" events, when the string can become stuck, the equivalent circulating density can spike, and lost circulation can occur. The non-uniform settling of the cuttings along the wellbore will cause non-uniform distribution of pressure losses, with local pressure spikes when cuttings blockage becomes substantial. A pack-off sensitivity analysis is presented in this work. The overall aim of this work is to provide more comprehensive cuttings transport modeling for the industry to conduct safe and improved hole cleaning operations.
Poor hole cleaning leads to problems such as pack-offs, stuck pipe incidents, and lost circulation events due to increased equivalent circulating density (ECD). Hole cleaning issues can be mitigated by using a digital twinning system that constantly monitors borehole condition with real-time data and process models, and suggests optimal actions. The solution space of possible actions, however, is very large causing computational hurdles for real time implementation. This paper describes a time-efficient digital twinning approach that uses reinforcement learning (RL) to simulate scenarios corresponding to multiple hole cleaning actions.
Digital twinning of cuttings transport in a wellbore requires the development of a system with multiple integrated physics and data-based models to quantitatively detect symptoms of impending issues early, and to take actions to mitigate these problems. First, the state of the borehole is quantified in terms of ECD, concentration of cuttings in flow, and cuttings bed height. Next, a solution space of all legal and feasible actions based on the current system state is obtained. Finally, the alternate action sequences are evaluated to suggest the most suitable path forward.
To classify the state of the borehole in terms of ECD, cuttings bed height and cuttings concentration in the flow stream, hydraulics and cuttings transport models were implemented, tested and validated against real field and experimental data. Based on the state of the system, allowable actions were evaluated in terms of some combination of hole cleaning parameters that can be controlled in near real-time, such as flow rate, rotary speed, mud density, mud rheology and the weight on bit for rate of penetration (ROP) control. To limit the size of the solution space, those actions that were unrealistic for implementation in real-time were discarded. Also, any actions that violated the limits of the available drilling margin were discarded. Finally, by defining immediate rewards associated with different states-action pairs based on their effect on the wellbore stability and hole condition, the concepts of Markov reward process (MRP) and scenario realizations were used to suggest the optimal action for a given state.
The predicted output of the algorithm for multiple operational scenarios was validated by comparing it with actions that a hole cleaning / extended reach drilling (ERD) expert would have taken when given similar scenarios. The automated process of identifying hole cleaning system states, generating multiple viable hole cleaning action sequences, and finally evaluating the probability of successful hole cleaning for multiple actions in real-time and deciding the best course of action, is a novel contribution of this work. It will benefit practitioners who struggle with hole cleaning / stuck pipe-related non-productive time and pave the way for hole cleaning automation.
van Oort, Eric (The University of Texas at Austin) | Juenger, Maria (The University of Texas at Austin) | Aldin, Munir (Metarock Laboratories) | Thombare, Akshay (Metarock Laboratories) | McDonald, Mike (National Silicates, an affiliate of PQ Corporation)
Well abandonment is one of the biggest challenges in the oil and gas industry, both in terms of cost and effort as well as the technical hurdles associated with wellbore isolation for an indefinite term. A mechanism that may be exploited to simplify well abandonments is using natural shale formations for the creation of annular barriers. Currently, uncemented annuli often require casing milling and pulling before abandonment plugs can be set, which necessitates the use of a drilling rig. This is an expensive, time- and labor-intensive process, particularly offshore. However, shale creep may naturally form a barrier behind uncemented casing sections. With a qualified annular shale barrier in place, the well may only require the setting of abandonment plugs within the existing casing string(s), a task that can often be done rigless and with significantly less effort.
The work described in this paper presents the results of a rock mechanical investigation into the creep behavior of North Sea shales and their ability to form effective annular barriers. Field core from the Lark-Horda shale was used to conduct dedicated, customized experiments that simulated the behavior of shale confined under downhole effective stress, pressure and temperature conditions to fill in an annular space behind a simulated casing string. Full scale tri-axial rock mechanics equipment was used for testing cylindrical shale samples obtained from well-preserved field core in a set-up that mimicked an uncemented casing section of a well. The deformation behavior of the shale was monitored for days to weeks, and the formation of the annular barrier was characterized using dedicated strain measurements and pressure pulse decay probing of the annular space.
The large-scale lab results clearly show that the Lark-Horda shales will form competent low permeability annular barriers when left uncemented, as confirmed using pressure-pulse decay measurements. They also show that experimental conditions influence the rate of barrier formation: higher effective stress, higher temperature and beneficial manipulation of the annular fluid chemistry all have a significant effect. This then opens up the possibility of activating shale formations that do not naturally create barriers by themselves into forming them, e.g. by exposing them to low annular pressure, elevated temperature, different annular fluid chemistry, or a combination. The results are in very good agreement with field observations reported earlier by several North Sea operators.
Ucherek, Jared (The University of Texas at Austin) | Lawal, Tesleem (The University of Texas at Austin) | Prinz, Matthew (The University of Texas at Austin) | Li, Lisa (The University of Texas at Austin) | Ashok, Pradeepkumar (The University of Texas at Austin) | van Oort, Eric (The University of Texas at Austin) | Gobert, Tatiana (Occidental Petroleum Corporation) | Mejia, Juan (Occidental Petroleum Corporation)
Activity codes recorded by drillers are very useful for quantifying invisible lost time (ILT). However, classifying more than 100 activity codes accurately and consistently across various rig operations becomes infeasible for human operators. We propose an auto-suggestive system that guides the drillers to the correct codes based on memos they enter into the system. This aims to both eliminate manual classification errors and improve memo entry.
The method for extracting activity codes from memos can be broken into the following steps. The first step consists of filtering unnecessary text and vectorizing the memos. The vectors are then re-weighted using the term frequency-inverse document frequency (TFIDF) statistical measure. Next, data resampling helps to create a uniform set of labels for the training data, because there are quite a few important activity codes that appear infrequently with respect to others. Finally, a classifier is trained. It is shown that the finalized model can be used as a real-time auto-suggestive mechanism during the drillers’ data input process. Moreover, its use for cleaning up historical datasets is also explored.
This method was implemented on a large historical dataset consisting of 150 wells, and ILT analysis was performed with the original dataset and with the auto-classified dataset. Comparing these results clearly showed that performing analysis on a dataset that has not been properly classified can lead to incorrect and misleading conclusions. Also, this method did not require a manual re-labeling of the dataset for model training. This makes the algorithm readily applicable for any end-user, irrespective of the number of activity codes used. Various classifiers including logistic regression, support vector machine, random forests, naïve Bayes, and multi-layered perceptron were implemented and tested. Given comparable performances, we conclude that a simple and interpretable logistic regression model is best for real-time classification. Tests were also performed to see how many typed words in a memo would be needed before the correct activity code was identified. The results are detailed in this paper.
This is the first body of work that has taken drillers’ memos and converted them into activity codes, without the need for a human-classified training dataset. The real-time classifier is very powerful in ensuring clean data at the source and will be particularly useful when implemented on reporting systems for classifying rig activities by IADC activity codes. We further demonstrate the use of the classifier for cleansing historical datasets such that ILT analysis can be done more accurately.
Gu, Qifan (University of Texas at Austin) | Fallah, AmirHossein (University of Texas at Austin) | Ambrus, Adrian (NORCE Norwegian Research Centre AS) | Ma, Zheren (Quantum Reservoir Impact) | Chen, Dongemi (University of Texas at Austin) | Ashok, Pradeep (University of Texas at Austin) | van Oort, Eric (University of Texas at Austin)
Riser gas unloading events in subsea well construction are hazardous and difficult to control. When a gas influx enters the wellbore and dissolves in nonaqueous fluids (NAFs), it may go unnoticed because the pit gain on the surface may be minimal and remain below the detection threshold when using conventional well control indicators. Once the dissolved gas is circulated up and comes out of the solution at pressure and temperature conditions below the bubblepoint, it can quickly displace a large volume of mud in the riser or the chokeline. When this breaking out of gas occurs at a shallow depth, it leaves little time for the rig crew to react.
In this paper, we present a novel managed pressure drilling (MPD) approach for riser gas unloading control that makes use of a pressurized riser drilling (PRD) controller. The PRD method employs the constant bottomhole pressure (CBHP) controller for normal operations (e.g., drilling and circulation of kick), whereas it offers a more comprehensive way to manage and control a riser gas unloading behavior. The PRD choke controller dynamically applies a backpressure on the dissolved gas–NAF mixture in the riser to delay, minimize, or even prevent the gas breaking out at locations closer to the surface. The control algorithm considers the pressure limits of the riser and of the openhole formations and can adjust for kick uncertainties (e.g., whether the kick is a gas, liquid, or both, its volume and distribution).
The proposed PRD controller consists of three operation modes: pressure control mode, flow control mode, and solubility control mode, with each mode applicable to its corresponding operating condition. The controller can automatically switch among different modes on the basis of the observed kick behavior, thereby gaining the ability to compensate for the limitations of the individual control modes and, more importantly, to deal with kicks agnostically, i.e., independent of their nature. The proposed controller is evaluated by simulating different riser gas unloading scenarios. Here, two distinct cases are given special consideration: (1) the case in which the subsea blowout preventers (BOPs) remain open after the kick passes them and the controller regulates the pressure using the maximum allowable surface pressure (MASP) or downhole fracture gradient as an upper limit and (2) the case in which the subsea BOPs are closed after the kick passes them and the controller regulates the pressure within the riser pressure limits when the kick is circulated to the surface using the riser booster pump.
Simulation results show that the proposed controller can quickly and robustly control the riser gas unloading situations with complicated transient conditions, without fracturing downhole formations or jeopardizing the pressure integrity of the riser. The developed PRD controller aims to help mitigate some of the concerns about riser gas unloading when the dissolved gas is allowed to pass the subsea BOPs and enter the riser and to facilitate the implementation of more automated subsea well control using MPD technology in the foreseeable future.
van Oort, Eric (The University of Texas) | Juenger, Maria (The University of Texas) | Liu, Xiangyu (The University of Texas, Currently with Nalco Champion) | McDonald, Michael (National Silicates, an affiliate of PQ Corporation)
Ordinary Portland cement (OPC) has been the material of choice for oil & gas well cementing and abandonment for many decades now. However, there are drawbacks to the use of OPC for cementing and abandonment purposes, particularly in wells with higher temperatures. OPC is brittle and does not re-heal when cracked. It is easily contaminated by mud and spacer fluids. Furthermore, it has relatively low tensile strength and low strength when bonding to rock formations and casing. Moreover, the production of OPC is the 2nd largest source of CO2 emissions in the world. At the CODA industry-affiliate consortium at the University of Texas at Austin dedicated to well construction, decommissioning and abandonment, work is ongoing to find technically superior alternatives to OPC. Particularly promising materials are so-called geopolymers, formed by activating an alumino-silicate material such as fly ash (a waste material that is often discarded) with an alkali. It was found that these geopolymer materials offer more ductile strength and failure behavior, considerable resistance to contamination, higher tensile strength and bond strength, and an ability to re-heal when damaged. The results obtained for geopolymers formed by activating flyash with potassium and sodium silicates indicate that these may be well-suited for achieving long-term thermal well integrity.
Enayatpour, Saeid (Hildebrand Department of Petroleum and Geosystems Engineering, The University of Texas at Austin, TX, USA) | Thombare, Akshay (Metarock Laboratories, Houston, TX, USA) | Aldin, Munir (Metarock Laboratories, Houston, TX, USA) | van Oort, Eric (Hildebrand Department of Petroleum and Geosystems Engineering, The University of Texas at Austin, TX, USA)
Oil and gas wells produce hydrocarbons for a limited number of years, and at the end of their production life they need to be plugged and abandoned. This process has to be done in a safe and economic way. Creep deformation of shale rock in uncemented casing sections may simplify well abandonments considerably. Creep can close the annular gap between a shale formation and an uncemented section of a casing string, generating a barrier that prevents hydrocarbons from flowing to the surface on the annular side. Wells with such a "shale-as-a-barrier" generated by creep now only require abandonment plugs on the inside of the casing, without the need for installation of additional annular barriers. This may eliminate such operations as casing milling and casing pulling, thereby allowing e.g. offshore abandonments to be done rigless, at significantly reduced cost.
This paper presents the first results of an experimental investigation and numerical modeling study into the nature of the "shale-as-a-barrier" phenomenon. Specifically, we focus on laboratory and field scale numerical simulation of creep behavior of North Sea Lark shale rock for oil and gas well plug and abandonment purposes. In our Finite Element simulations of the shale creep phenomenon, we have used the time-hardening creep model, which assumes a non-linear relationship between creep strain and stress, temperature and time. The model parameters were obtained from a curve fit of laboratory experimental results conducted for a creeping shale. Then, using the experimentally-derived parameters, numerical simulation was performed for a laboratory scale model and result was validated against laboratory results. Once this validation had taken place, the model size was extended to the field scale for prediction of annular closure time and barrier formation. Simulations show a strong correlation between rock stiffness and annular gap closure time, as expected; hence, the success of any "shale-as-a-barrier" project is a distinct function of shale rock stiffness. Lowering near-wellbore stiffness artificially may accelerate annular barrier creation of slowly creeping shale formations.