Patent Application
20190024485
The present invention generally relates to systems and methods for intelligent and autonomous/semi-autonomous control of cementing and liner hanging operations.
Boreholes, or wells, are drilled deep into the earth for many applications such as carbon dioxide sequestration, geothermal production, and hydrocarbon exploration and production. In all of the applications, the boreholes are drilled such that they pass through or allow access to a material (e.g., a gas or fluid) contained in a formation located below the earth's surface. Different types of tools and instruments may be disposed in the boreholes to perform various tasks and measurements. A well, e.g., for production, is generally completed by placing a casing (also referred to herein as a “liner” or “tubular”) in the wellbore. The spacing between the liner and the wellbore inside, referred to as the “annulus,” is then filled with cement. The liner and the cement may be perforated to allow the hydrocarbons to flow from the reservoirs to the surface via a production string installed inside the liner.
The disclosure herein provides intelligent control over one or more aspects of such completion operations.
In aspects, the present disclosure provides a method for controlling a well completion operation. The method may include conducting the well completion operation, estimating at least one operation parameter while conducting the well completion operation using at least one sensor, using a controller to determine at least one parameter adjustment relating to the well completion operation, and generating a command relating to the well completion operation. The parameter adjustment determination may be done using the at least one model and the at least one estimated operating parameter. The command generation may be based on the determined at least one parameter adjustment.
In other aspects, the present disclosure provide an apparatus for controlling a well completion operation. The apparatus may include a completion system configured to conduct the well completion operation, at least one sensor configured to estimate at least one parameter while conducting the well completion operation, and a controller. The controller may have access to at least one model configured to determine a parameter relating to the well completion operation. The controller may be configured to determine at least one parameter adjustment relating to the well completion operation using the at least one model and the at least one estimated parameter and generate a command relating to the well completion operation based on the determined at least one parameter adjustment relating to the well completion operation.
Illustrative examples of some features of the disclosure thus have been summarized rather broadly in order that the detailed description thereof that follows may be better understood, and in order that the contributions to the art may be appreciated. There are, of course, additional features of the disclosure that will be described hereinafter and which will form the subject of the claims appended hereto.
For detailed understanding of the present disclosure, references should be made to the following detailed description of the preferred embodiment, taken in conjunction with the accompanying drawings, in which like elements have been given like numerals and wherein:
In aspects, the present disclosure provides methods and related systems that intelligently guide, in an automated or semi-automated fashion, the various steps of completing a well by hanging and cementing a liner. Cementing can be a complicated procedure where the amount of cement, placement of the cement, and quality of the cement bond are important considerations. Embodiments of the present disclosure use a pre-defined process for such operations. A feedback loop may compare modeling versus real-time measurements to optimize the overall process. Additionally, operating parameters may be adjusted minute-by-minute and potential issues may be identified concurrently. Embodiments of the present disclosure use detailed pre-planning and real-time information to advise on what step(s) to take and what, if any, adjustments must be made to operating parameters. These actions may be fully automated, with human interaction only in case of unforeseen challenges. Alternatively, the systems may provide prompts/guidance that assist human operators.
In embodiments, the entire completion process, from liner hanging to cementing and clean up, may be optimized: from the simple timing of dropping darts that pressure-activate downhole tools and switching pumps off/on to complicated rotational speed (aka revolutions per minute, RPM)/axial movement variations along the process to accommodate varying geological or geometrical environments along the well path. Illustrative systems may compare simulation/modeling data with real-time sensor and offset data and adjust the procedures on the fly. It will be appreciated that such systems not only enhance the quality of the outcome, but also help to evaluate the final completion in view of certain quality goals such as cement channeling or completeness of a cleanup. Embodiments of the present disclosure may be used in conventional cementing/liner hanging operations, or in conjunction with a “one-trip” liner hanging and cementing tools.
In
A controller 50 may be used in connection with the liner hanging and/or cementing activity. In some embodiments, the controller 50 uses preprogrammed algorithms, historical data, and real-time information in order to advise personnel on action(s) to take and/or automatically send commands to take such actions. These action may include, but are not limited to, varying RPM, pump rates, timing of dart insertion, and spacer sizes. The controller 50 may include conventional components such as microprocessors, a memory controller, a main memory, a network interface, a transceiver, etc. Near real-time or real-time information may be obtained using sensors that are distributed at surface and downhole locations. Surface sensors are labeled with numeral 52 and downhole sensors are labeled with numeral 54. Non-limiting examples of sensors 52, 54 include pressure sensors, temperature sensors, flow rate sensors, pump rate sensors, RPM sensors, vibration sensors, load sensors, e.g. hook load sensors, torque sensors, weight sensors, etc. A suitable communication system 56 may be used to transfer data and command signals. The communication system 56 may utilize wired pipe, optical fibers, EM signals, RF signals, acoustic signals, pressure pulses (e.g., mud pulses), etc.
Referring now to
At step 74, real-time or near real-time information relating to one or more operating parameters is acquired. By “real-time” or “near real-time,” it is meant that such information is collected while liner hanging or cementing operations are ongoing. The operating parameters may be for surface and/or downhole equipment. Illustrative surface parameters include, but are not limited to, pump flow rates (e.g., drilling mud, spacer fluid, cement, etc.), pressure (e.g., bore pressure, annulus pressure, etc.), pressure, temperature, flow rate sensors, pump rate, RPM, vibration, load, e.g. hook load sensors, torque sensors, weight sensors, etc. and properties of returning fluids and/or entrained material such as mud cake or cement. Illustrative downhole parameters include, but are not limited to, pressure, temperature, flow rate, pump rate, RPM, vibration, load, e.g. hook load, torque, weight, etc. The measurements of such parameters sensors 52, 54 may provide absolute values and also the basis for estimating fluctuations or rates of change (e.g., for torque). The fluid returning to the surface may be analyzed to determine the quality/quantity of the cement slurry and the composition may be analyzed to determine the nature of the returning fluid (e.g., spacer fluid vs. drilling mud).
At step 76, the information from the database and acquired real-time information is used to build one or more models that may utilize algorithms to determine one or more subsequent process steps such as identifying a parameter or process step for adjustment. In one embodiment, the models are configured to estimate the condition or behavior of the completion equipment and/or wellbore environment and determine a course of action to optimize subsequent activity. The output of this set may be advice, command signals, and/or alarms. As used herein, the term “advice” is information communicated to a human operator, a “command signal” is a message that can be understood by a machine to perform or stop performing a given task or to adjust or otherwise amend a parameter that is used to perform a task, and an “alarm” is a special category of “advice” that indicates that an “out of norm” condition may be present. The model may comprise formation models, geo-mechanical models, hydraulics models, cement setting speed models, and models estimating the risk of channeling and stand-off annulus distribution. Such model generally are mathematically based algorithms and/or data maps providing one, two, or three dimensional distributions of a particular parameter in space or time that is either based on calculated data, measured data, or both. The environment and dynamics being simulated is shown in
Two non-limiting examples of relationships that may be incorporated into a model are shown in
Generally, when pumping cement, the string/liner is rotated at low speeds. The selection of rotation speed is a balance between (i) keeping ECD low enough for the formation not to fracture (by keeping pump rate and RPM below certain thresholds), (ii) keeping cement channeling low by keeping pump rate at optimum rate which is usually high and RPM low or vice versa, (iii) and minimizing time spent by keeping pump rate high, especially given the cement setting speed. While a higher pump rate generally helps with channeling, by manipulating RPM and flow, one can influence the channeling effect by reducing or increasing these parameters independently. Other considerations may include fatigue of liner connection when experiencing too many total revolutions at a certain local hole curvature. The appropriate RPM, and other operating parameters, can change gradually over the course of the cement pumping procedure.
A relatively simple model may be used to control torque applied to the rotating liner. As cement sets, the rotating torque increases as the viscosity increases. The model may set a torque limit at which the running string must be pulled out to prevent the running string from being stuck in the well. In this particular case, the model may comprise not more than a single torque threshold value.
More complex models may be used to optimize and evaluate the actual cleanup effectiveness. Such a model would comprise parameters such as RPM, time, and/or the axial movement speed of the running string/liner, depth range, and number of repetitions of movement of a running string/liner. This model would make particular use of measurements of pressure and torque, and possibly require downhole sensing in the annulus. For instance, a communication system 56 at the top of the liner 20 may be utilized to communicate pressure, temperature, flow rate, pump rate, RPM, vibration, load, e.g. hook load, torque, weight, etc. Additionally, surface returns can be evaluated. The evaluation would primarily be used to characterize the nature of the returns and the distribution over time of the returns of cement slurry. Illustrative tools and sensor for such evaluation include a cuttings catcher, a measurement system for density, chemical or mineralogical composition, or similar devices (not shown).
Exemplary models may also be configured for cement channeling prediction. Rotating at low RPM may result in the cement not fully encircling the liner (i.e., no full circumference), which is due to low friction/low torque. Rotating at increasing RPM while evaluating torque changes can help predict gaps in cement column and when they are filled. For channeling detection, the torque when rotating at different rpms with the cement in the annulus is a function of the amount of circumference of the liner in contact with cement. In a horizontal application, the narrow annulus on the lowside frequently is not filled with the viscous cement, but needs rotation of a certain speed to squeeze the cement fully around the narrow lowside. This would show as an increase in torque that can be observed downhole or on surface. A non-limiting test may involve initiating a sequence of rotations at 10, 20, 30, and 40 RPM for a few minutes each, recording the torque increase and comparing the increases to modeled or offset torque curves of offset wells or earlier times with different location of the cement slurry. The visualized curves in comparison then may indicate the area of liner surface covered with cement.
Exemplary models may also be configured for mud cake removal prediction. The efficiency of spacer fluids in removing mud cake is influential for a cementing operation. Thus, a model may provide real time analysis of spacer fluid mud cake content volume/quality. This can be evaluated in real-time by checking the mud cake weight over time in the returns; e.g. by using a cuttings catcher. The evaluation may be inputted into the model to identify the depths at which residual mud cake exists. When combined with formation models, geo-mechanical models, hydraulics models, cement setting speed models, and models estimating the risk of channeling and stand-off annulus distribution, etc., this may then trigger the decision to extend or abbreviate the spacer pumping or to change quality or quantity of the cement pumped.
Depending on the model, illustrative input parameters include: open hole volume, formation integrity, cement volume, cement rheology, pump efficiencies, acceptable pump rates, acceptable rotation rates, and pressures (e.g., stand pipe pressure (SPP)), downhole pressure. These inputs may be used to automate critical operational parameters for cementing liners and casing downhole based on real-time simulations (RPM, SPP, flow rate). The models may simulate or predict relationships between RPM and flow rate, predict cement/spacer interface, predict cement column height (e.g., using torque and pressure), and/or estimate formation strength to manage flow rate (e.g., using equivalent circulating density). Illustrative actions may include starting/stopping rotation of the liner, when to apply weight or pull up the liner, when to actuate downhole mechanisms (e.g., dropping a dart).
It should be understood that a variety of control schemes may be used; e.g., complete machine automation, human control with machine generated advice, a hybrid of machine control with selective human intervention, etc. Thus, the modes discussed below are merely illustrative.
In a principally automated operating mode, the controller 50 sends command signals to surface and/or downhole equipment that adjust one or more operating set points. For instance, command signals may be sent to actuators for devices such as pumps, mixing equipment, valves, heaters, packers, top drive (or other surface rotary power source), etc. Command signal may include activation commands to activate actuators for such devices. For instance command signals may include an activation command to activate an actuator that is configured to allow pressurization of a packer or packer element in order to set the packer. Of course, human operators may be given the opportunity to stop or modify the action(s) to be taken. In a principally “advice” operating mode, the controller 50 presents a proposed course of action(s) to human operators, who then decide whether or not implement such action(s).
It should be appreciated that the methods and systems according to the present disclosure provide several advantages over conventional liner hanging and cementing techniques. These advantages include operational time savings due to faster procedures on average, repeatability of the process that yields higher quality of cementing, reduced need for training/building experience for operators, and the ability to build a database of historical friction/hydraulics curves useful to characterized operational issues, which can yield a better understanding for future applications.
The foregoing description is directed to particular embodiments of the present disclosure for the purpose of illustration and explanation. It will be apparent, however, to one skilled in the art that many modifications and changes to the embodiment set forth above are possible without departing from the scope of the disclosure. It is intended that the following claims be interpreted to embrace all such modifications and changes.
