AUTOMATED HEAVE COMPENSATION

Description

CROSS-REFERENCE TO RELATED APPLICATIONS

This application claims the benefit of U.S. Provisional Appl. No. 63/670,587 filed Jul. 12, 2024, which is incorporated herein by reference in its entirety.

FIELD OF THE DISCLOSURE

The subject matter of the present disclosure generally relates to deepwater Managed Pressure Drilling (MPD) operations. More particularly, the subject matter of the present disclosure relates to an apparatus and a method of transmitting data, monitoring data, and implementing a method to compensate for bottom hole pressure fluctuations caused by Heave effects on a drilling vessel (e.g., a drillship, a semisubmersible,, a mobile offshore drilling unit (MODU), and other movable offshore facility) used in deepwater MPD operations.

BACKGROUND OF THE DISCLOSURE

Drilling a wellbore for hydrocarbons requires significant expenditures of manpower and equipment. Therefore, constant advances are being sought to improve operational safety, reduce any downtime of equipment, and expedite any repairs that become necessary. Bottom hole pressure fluctuations that exceed limitations of the formation can be the main cause of downtime.

In a typical MPD operation, a drill bit is attached to a drill pipe. Thereafter, a drive unit rotates the drill pipe using a drive member as the drill pipe and the drill bit are urged downward to form the wellbore. Several components are used to control the gas and fluid pressure. Typically, a blowout preventer (BOP) is used to seal the mouth of the wellbore. In many instances, a conventional rotating control device is mounted above the BOP stack so an internal portion of the rotating control device can seal and rotate with the drill pipe. The seal around the drillpipe provides a closed loop circulating system.

Surge and swab effects can occur during pipe movements when performing the MPD operation. During various points of the MPD operation, tripping of the drillstring may be performed where the drillstring is pulled out of hole (POOH) or run in hole (RIH). For example, a tripping operation may pull the drillstring out of hole to replace a downhole component (e.g., a damaged drillpipe, a worn drill bit, a malfunctioning mud motor, etc.) or to add a downhole component so the drillstring can then be run in back in hole to continue drilling. A trip (movement of the drillstring) may also be done for logging, coming off bottom, reaming the wellbore between connections, etc.

When pulling the drillstring out of the wellbore, the drillstring is lifted at the drilling rig's derrick, and stands (two or more drill pipe joints) are disconnected from the drillstring and stacked in the derrick in consecutive steps. Any replacements or additions to downhole components can be performed, and the drillstring can be run in hole by reconnecting stands to continue with drilling operations. Pulling the drillstring out of the hole can decrease the bottom hole pressure due to a swabbing effect. By contrast, running the drillstring in hole can increase the bottom hole pressure due to a surging effect.

Additionally, deepwater MPD operations drilled from a floating drilling vessel (e.g., Mobile Offshore Drilling Unit (MODU)) are exposed to met-ocean conditions. In general, the met-ocean conditions (i.e., “meteorological” and “oceanographic” conditions) refer to the various weather and sea conditions that may be encountered during deepwater drilling operations. As will be appreciated, met-ocean conditions can vary significantly over short periods and can dramatically impact drilling operations. The met-ocean conditions generate undesired fluctuations in bottom hole pressure caused by vessel motion, with Heave, Sway and Roll being the prominent effects. Current systems and methods may not effectively compensate for bottom hole pressure fluctuations during deepwater MPD when the drilling vessel is influenced by met-ocean conditions.

The subject matter of the present disclosure is directed to overcoming, or at least reducing the effects of, one or more of the problems set forth above.

SUMMARY OF THE DISCLOSURE

A computer-implemented method disclosed herein is used with a drilling system on a drilling vessel in an ocean environment to drill a wellbore in a formation. The drilling system is configured to circulate fluid in a closed loop between a drillstring and the wellbore and is configured to regulate a downhole pressure in the wellbore using pressure regulation of the circulated fluid. The method comprises: identifying, with a control system, between a non-drilling state and a drilling state of the drilling system, the non-drilling state being indicative of the drillstring being held at the drilling vessel at least during vessel motion, the drillstring held during the vessel motion being expected to produce a first piston effect that changes the downhole pressure in the wellbore, the drilling state being indicative of the drillstring being tripped in the wellbore; determining, with the control system in response to an identification of the non-drilling state, a first adjustment to the pressure regulation provided by the drilling system, the first adjustment being configured to maintain the downhole pressure in the wellbore; and counteracting the change to the downhole pressure produced by the first piston effect by automatically adjusting, with the control system, the pressure regulation provided by the drilling system according to the first adjustment.

To determine the first adjustment, the method can comprise: obtaining one or more environmental measurements indicative of an influence from the ocean environment on the drilling vessel; predicting predicted movement of the drilling vessel by motion modelling how the drilling vessel responds to the one or more environmental measurements; and determining the first adjustment based on the predicted movement.

For example, predicting the predicted movement of the drilling vessel by motion modelling can involve using transfer function calculations that define how one or more vessel characteristics for the drilling vessel respond to one or more environmental characteristics for the one or more environmental measurements.

In another example, predicting the predicted movement of the drilling vessel by motion modelling can further include: obtaining one or more vessel measurements indicative of a measured movement of the drilling vessel in the ocean environment; and supplementing the predicted movement of the drilling vessel by using the measured movement of the drilling vessel in the motion modelling.

In the disclosed method, the method can use a machine learning model to at least one of: identify between the non-drilling state and the drilling state, predict the predicted movement, and determine the first adjustment.

Identifying the non-drilling state indicative of the drillstring being held at the drilling vessel can comprise: identifying a first instance of the drillstring being moved out of the wellbore during first of the vessel motion expected to produce swabbing as the first piston effect decreasing the downhole pressure in the wellbore; and identifying a second instance of the drillstring being moved into the wellbore during second of the vessel motion expected to produce surging as the first piston effect increasing the downhole pressure in the wellbore. Accordingly, automatically adjusting, with the control system, the pressure regulation provided by the drilling system according to the first adjustment can comprise: automatically adjusting, with the control system, the pressure regulation provided by the drilling system according to a downhole pressure increase for the first adjustment in response to the first instance to counteract the change to the downhole pressure produced by the swabbing; and automatically adjusting, with the control system, the pressure regulation provided by the drilling system according to a downhole pressure decrease for the first adjustment in response to the second instance to counteract the change to the downhole pressure produced by the surging.

In the method, identifying the drilling state can comprise identifying a trip of the drillstring being expected to produce a second piston effect that changes the downhole pressure in the wellbore.

In one example for the drilling state, the method can further comprise: determining, with the control system in response to an identification of the drilling state, a second adjustment to a surface backpressure for the pressure regulation provided by the drilling system to maintain the downhole pressure in the wellbore; and counteracting the change in the downhole pressure produced by the second piston effect by automatically adjusting, with the control system, the surface backpressure according to the second adjustment. In this example, identifying the trip expected to produce the second piston effect can comprise: identifying a first instance of pulling the drillstring out of the wellbore that produces swabbing as the second piston effect decreasing the downhole pressure in the wellbore; and identifying a second instance of running the drillstring into the wellbore that produces surging as the second piston effect increasing the downhole pressure in the wellbore.

In another example for the drilling state, the method can further comprise mitigating a change in the downhole pressure produced by the second piston effect by operating a crown block heave compensator mounted on the drilling system.

To determine the first adjustment configured to maintain the downhole pressure in the wellbore, the method can comprise obtaining, with the control system, a value of the downhole pressure of the fluid in the wellbore. For example, the steps can include obtaining one or more pressure measurements of the drilling system; and calculating a bottom hole pressure (BHP) in the wellbore based at least on the one or more pressure measurements.

To determine the first adjustment, the steps can include defining a target for the downhole pressure at a depth in the wellbore as being at least less than one of: (i) a fracture pressure gradient of the formation, and (ii) a pore pressure gradient of the formation; and setting the first adjustment to effectuate the target for the downhole pressure.

As disclosed herein, a programmable storage device has program instructions stored thereon for causing a programmable control device to perform a method of drilling a wellbore with drilling fluid using a drilling system as described above.

As disclosed herein, a computerized system is used with a drilling system on a drilling vessel in an ocean environment to drill a wellbore in a formation. The computerized system comprises a programmable control device being configured to perform a method of drilling a wellbore with drilling fluid using a drilling system as described above.

The foregoing summary is not intended to summarize each potential embodiment or every aspect of the present disclosure.

BRIEF DESCRIPTION OF DRAWINGS

FIG. 1 illustrates a controlled pressure drilling system having a control system according to the present disclosure.

FIG. 2 schematically illustrates the control system of the present disclosure.

FIG. 3 illustrates a process of performing automated heave compensation during non-drilling states and drilling states.

FIG. 4A illustrates a graph of heaving of a drilling vessel.

FIG. 4B illustrates another graph of response amplitude operators in response to the heaving of the drilling vessel.

FIG. 5 illustrates a process to train a machine learning model according to the present disclosure.

FIG. 6 illustrates a framework to train and deploy a machine learning model according to the present disclosure.

FIG. 7 schematically illustrates a deep reinforcement learning-based model for use by the control system of the present disclosure.

DETAILED DESCRIPTION OF THE DISCLOSURE

Systems and methods disclosed herein automatically compensate for surge and swab effects during movements of a drillstring in a wellbore to maintain constant bottom hole pressure (BHP) when performing a Managed Pressure Drilling (MPD) operation on a drilling vessel. Using a control system and various techniques discussed below, the influence of these surge and swab effects on the bottom hole pressure can be mitigated during a drilling state and a non-drilling state of the drilling system.

FIG. 1 illustrates a controlled pressure drilling system 10 having a control system 200 according to the present disclosure, and FIG. 2 schematically illustrates the control system 200 of the present disclosure.

As shown in FIG. 1, this drilling system 10 can be a managed pressure drilling (MPD) system and, more particularly, a Constant Bottom-hole Pressure (CBHP) form of MPD system. Although discussed in this context, the teachings of the present disclosure can apply equally to other types of controlled pressure drilling systems, such as other MPD systems (Returns-Flow-Control Drilling, Dual Gradient Drilling, etc.) as well as to Under-Balanced Drilling (UBD) systems, as will be appreciated by one skilled in the art having the benefit of the present disclosure.

The drilling system 10 (“drilling system or closed loop drilling system”) is used on a drilling vessel 100, which has a drilling rig 110 and components for fluid handling. The drilling vessel 100 shown in FIG. 1 can be a Mobile Offshore Drilling Unit (MODU), a drillship, a semisubmersible, and other movable offshore facility. During drilling operations while a drillstring 14 is connected to the vessel's rig 110, the drilling vessel 100 may be subject to environmental effects. Heaving motions of the drilling vessel 100 produced from the environmental effects may change the bottom hole pressure in the wellbore 12. Mechanical compensation can account for at least some of these heaving motions during drilling opertions. For example, a crown block heave compensator 117 mounted on the rig 110 can compensate for the heaving motions when the drillstring 14 is connected to the compensator 117 so the bottom hole pressure can be maintained.

When drilling, pulling the drillstring 14 out of the hole in a trip can decrease the bottom hole pressure due to a swabbing effect. For example, the swabbing effect between the mud and the drillstring 14 being pulled can create changes in pressure in the wellbore 12. The tools (drill bit, stabilizer, drill collar, etc.), which are typically full gauge of the wellbore 12, on the bottom hole assembly (BHA) 16 being pulled out of hole can lift mud in the annulus and produce lower pressures in the formation (F). An influx of formation fluids can also enter the wellbore 12.

Likewise, when drilling, running the drillstring 14 in hole in a trip can increase the bottom hole pressure due to a surging effect. Should the run-in speed be too fast, the increasing bottom hole pressure may result in mud losses due to the increasing bottomhole pressure being greater than the fracture pressure of the formation F.

Accordingly, the control system 200 disclosed herein can identify an instance when a trip (POOH, RIH) is needed for the drillstring 14 in the wellbore 12. The trip may be needed for any particular reason, such as reaming the wellbore 12 between connections, replacing components of the bottom hole assembly 16, etc. The trip is expected to produce a piston effect (i.e., swabbing effect for POOH and surging effect for RIH) that can change the downhole pressure (e.g., bottom hole pressure) in the wellbore 12. In response to the identified trip, the control system 200 can automatically adjust the surface backpressure or other form of pressure regulation performed by the drilling system 10 to compensate for the swabbing/surging effects so a more constant bottomhole pressure can be maintained.

In addition to the mechanical compensation performed by the crown block heave compensator 117 and in addition to the controlled compensation performed by the drilling system 10 when tripping (RIH, POOH) the drillstring 14 during drilling operations noted above, the control system 200 can perform adjustments when the drillstring 14 is held at the drilling vessel 100 in a “non-drilling state.” Such a non-drilling state can occur when the drillstring 14 is held in slips, collar system, or another device 111 on the rig 110 during an extended period. For example, drillpipe slips 111 fit into a rotary table, spider, or similar device on the rig 100 to grip and hold the drillstring 14 in place. Other devices can be used. For example, an elevator or other clamp-like device attaching to the top of the drill string can be used. In another example, a collar system includes an application-specific collar (not shown), a sliding collar table at the rig floor, and a hydraulically operated automated side-door (ASD) elevator (not shown). An example of such a collar system is disclosed in U.S. Pat. No. 10,337,263, which is incorporated herein by reference. (For simplicity, the discussion uses “slips” herein to refer to these and other types devices to hold the drillstring 14 in place at the rig 110.)

When the drillstring 14 is held, the benefits of the crown block heave compensator 117 cannot be used. During the non-drilling state, pumped fluid may or may not be provided to the drillstring 14 through the standpipe supply line 165a and/or through the bypass line 165b and continuous flow apparatus 170 (i.e., flow subcomponent 178 and clamp 176). In any event, pumped fluid may be provided to the riser 22 through any suitable riser lines, such as the riser booster line 25.

When the drillstring 14 is held in such a non-drilling state, for example, the movements (heaving) of the drilling vessel 100 on the ocean surface can move the drillstring 14 within the wellbore 12, essentially pulling the drillstring 14 out of the wellbore 12 and running the drillstring 14 into the wellbore 12 to varying extents. Because the slips 111 are devices used to grip the drillstring 14 and hold it in place in the rotary table on the rig 110, the movements of the drillstring 14 caused by the vessel movements (heaving) can produce piston effects that can change the bottom hole pressure as the bottom hole assembly (BHA) 16 is moved within the wellbore 12. For instance, vessel movements (heaving) can produce a swabbing effect that can decrease the bottom hole pressure as the bottom hole assembly (BHA) 16 being pulled out of hole lifts mud in the annulus, which can produce lower pressures in the formation F. This swabbing effect can be similar to tripping (POOH) of the drillstring 14 during drilling operations. Likewise, as with the tripping (RIH), the movements of the drillstring 14 into the wellbore 12 caused by the vessel movements (heaving) can produce a surging effect that can increase bottom hole pressure, resulting in mud losses.

The drillstring 14 may be held in a non-drilling state for any number of reasons. For example, when a connection is made at the rig 110 by adding a new section of drillpipe to the drillstring 14, the drillstring 14 is placed in the slips 111 while the new drillpipe section is threaded into the mill end of the held drillstring 14. A similar situation occurs when the drillstring 14 is held in the slips 111, and a drillpipe section is broken out during tripping operations.

Alternatively, the drillstring 14 may be held in a non-drilling state in the slips 111 during routine inspections or maintenance, which can include checking the integrity of the drillpipe, tool joints, or other downhole tools. More commonly, the drillstring 14 is secured in a non-drilling state in the slips 111 when changing or adjusting downhole tools, such as drill bits 18, logging tools, or measurement-while-drilling (MWD) tools.

Longer periods of holding the drillstring 14 in a non-drilling state may occur due to unexpected issues and pauses in drilling. The unexpected issues can include equipment malfunctions, and prolonged pauses in drilling may be due to various issues that require the drillstring 14 to be secured in the slips 111. In other situations, complications, such as stuck pipe scenarios, or when dealing with well control issues may require the drillstring 14 to be placed in a non-drilling state in the slips 111 so operators can stabilize the situation and make necessary adjustments. Furthermore, spurts of severe weather or any emergencies can necessitate stopping of drilling operations, requiring the drillstring 14 to be secured in a non-drilling state in the slips 111.

The control system 200 performs automated heave compensation according to the present disclosure in these situations when the drillstring 14 is held in such a non-drilling state, especially when met-ocean conditions on the drilling vessel 100 are significant. The disclosed automated heave compensation adjusts the pressure regulation (e.g., surface backpressure (SBP) or other form) needed to compensate for the piston (surge and swab) effects while the drillstring 14 is held in the non-drilling state. The pressure regulation controlled by the control system 200 and performed by components of the drilling system 10 can take different forms, as described much further below. One such form of pressure regulation involves controlling the surface backpressure (SBP) of flow out of the wellbore 12 using the choke manifold 120.

The particular adjustments depend on a number of factors. For instance, the pressures produced by the piston (surge and swab) effects depend on the rheological properties of the fluid, the dimension of the annulus, the speed of the pipe movement, the length of drillstring 14 in the wellbore 12, the annular clearance between the wellbore 12 and the drillstring's BHA 16, the mud cake in the wellbore 12, cuttings in the wellbore 12, etc. In fact, these values change as drilling continues into an open hole section of the wellbore 12 and as different depths are reached in the formation F.

For these reasons, the setpoints (222; FIG. 2) to which the pressure regulation (e.g., surface backpressure) is adjusted by the disclosed automated heave compensation of the control system 200 can be calculated using a hydraulics model (224; FIG. 2) as discussed below. In this way, when the drillstring 14 is held in a non-drilling state, the automated heave compensation disclosed herein adjusts the pressure regulation (e.g., surface backpressure) automatically to maintain a target bottom hole pressure in response to the movements of the drillstring 14 into and out of the wellbore 12 as the vessel 100 experiences heaving movements from the environmental conditions encountered.

Having an overview of the disclosed automated heave compensation, discussion turns to additional details in FIG. 1, which shows the closed-loop drilling system 10 according to the present disclosure for controlled pressure drilling. As shown, the drilling rig 110 includes a derrick having a traveling block 114 supporting a top drive 116, which couples to a flow subcomponent 178. A top of the drillstring 14 connects to the flow subcomponent 178, such as by a threaded connection, or by a gripper (not shown), such as a torque head or spear. The top drive 116 is operable to rotate the drillstring 14 extending from the derrick and includes an inlet coupled to a Kelly hose to provide fluid communication between the Kelly hose and the flow subcomponent 178 and drillstring 14 extending therefrom.

The drillstring 14 extending from the rig 110 includes a bottom hole assembly (BHA) 16 at the end of the connected joints of drillpipe. The BHA 16 can typically include a drill bit 18, drill collars, stabilizers, a drilling motor, a measurement while drilling sub, a logging while drilling sub, and the like for drilling a wellbore 12.

The drilling system 10 further includes an upper marine riser package (UMRP) 30, a riser 22, auxiliary lines (kill and choke lines 24, booster line 25, etc.), and other components. As is customary, the riser 22 extends from the rig 110 to a wellhead 20 located on the sea floor. The riser 22 typically connects to the wellhead 20 with a wellhead adapter, and the wellhead 20 typically has blow-out preventers (BOPs). The drilling system 10 connects to the wellhead 20 via riser lines, such the choke and kill lines 24.

The drilling system 10 can also include connections to the riser 22 via riser lines, such as with the booster line 25. As shown, the booster line 25 is typically tied into the lower end of the riser 22. The booster line 25 can communicate drilling fluid into the riser 22 so fluids can be circulated in the riser 22.

The riser package 30 includes a diverter 70, a flex joint 72, a telescopic joint 74, a tensioner 76, a tensioner ring 78, and a rotating control device (RCD) 60. For example, the telescopic joint 74 includes an outer barrel connected to an upper end of the RCD 60 and includes an inner barrel connected to the flex joint 72. The outer barrel may also be connected to the tensioner 76 by the tensioner ring 78.

The RCD 60 can include any suitable pressure containment device that keeps the wellbore 12 in a closed-loop at all times while the wellbore 12 is being drilled. (As will be appreciated, the wellbore 12 includes the wellbore in the formation F and includes the riser 22 which constitutes an extension of the wellbore). In this way, the RCD 60 can contain and divert annular drilling returns via a flow line 62 to complete the circulating system to create the closed-loop of incompressible drilling fluid. The RCD 60 can include any typical construction.

The RCD 60 may be submerged adjacent the waterline. The RCD interface may be in fluid communication with an auxiliary hydraulic power unit (HPU) (not shown) of a control system 200 via control lines 202. An active seal can be used for the RCD 60. Alternatively, the RCD 60 may be located above the waterline and/or along the UMRP 30 at any other location besides a lower end thereof. Alternatively, the RCD 60 may be assembled as part of the riser 22 at any location therealong.

The RCD 60 may be connected to other flow control devices, such as an annular seal device 50, a flow spool 40 having controllable valves, and the like, as used in MPD operation. The annular seal device 50 can be used to sealingly engage (i.e., seal against) the drillstring 14 or to fully close off the riser 22 when the drillstring 14 is removed so fluid flow up through the riser 22 can be prevented. Typically, the annular seal device 50 can use a sealing element that is closed radially inward by hydraulically actuated pistons. The control lines 202 from hydraulic components on the rig 110 can be used to deliver controls to the annular seal device 50.

The flow spool 40 can include a number of controllable valves (not shown) that connect to flow connections 42 to communicate the internal passage of the riser 22 with rig components on the rig 110. Flow lines 32 from the riser package 30 may be used to communicate flow, and the control lines 202 on the riser 22 may also be used to deliver controls to open and close the controllable valves.

In addition to the riser package 30, the drilling system 10 also includes a choke manifold 120, a shaker 140, mud tanks 142, mud pumps 150. In addition to these, the drilling system 10 includes flow equipment 160 to deliver flow to the drillstring 14 through the Kelly hose connected to a supply line 165a or through a clamp 174 connected to a bypass line 165b and couplable to the flow subcomponent 178. The clamp 174 and flow subcomponent 178 are part of a continuous flow apparatus that allows flow to be maintained while pipe connections are being made.

One or more return flow lines 32 connect from the riser package 30 to the choke manifold 120. A return pressure sensor 240a, return choke 122, and return flowmeter 124 communicate with the flow from the return flow line 32. After the choke manifold 120, the flow eventually communicates with the mud gas separator 130 and the shaker 140.

A transfer line 144 connects an outlet of the mud tanks 142 to the mud pumps 150. A standpipe 152 connects from the mud pumps 150 to the drilling rig 110 to conduct drilling fluid from the mud pumps 150 to the Kelly hose and other flow connections. The standpipe 152 can include a pressure sensor 240d near the pumps 150 or elsewhere in the flow after the pumps 150.

Here, the standpipe 152 also includes the flow equipment 160, which is connected between the mud pumps 150 and the rig 110 for directing drilling flow into the drillstring 14 via the Kelly hose or via the clamp 174. The flow equipment 160 includes the supply line 165a connected from the mud pumps 150 to the top drive inlet 114. A supply pressure sensor 240b, a supply flow meter (not shown), and a supply shutoff valve (not shown) may be assembled as part of the supply line 165a.

Additionally, the flow equipment 160 includes the bypass line 165b connecting the standpipe 152 from the mud pump 150 to the clamp 176 of the continuous flow apparatus 170. An HPU 172 connects by hydraulic lines and manifold 174 to the clamp 176 to control its operation. For example, when the top drive 116 runs the drillstring 14 into the wellbore 12, the clamp 176 can engage the flow subcomponent 178, and the pumped flow of the drilling fluid can be bypassed to the bypass line 165b. In this way, continuous flow into the drillstring 14 can be maintained while making up new stands 13 of drillpipe to the drillstring 14. A bypass pressure sensor 240c, bypass flowmeter (not shown), and bypass shutoff valve (not shown) can be assembled as part of the bypass line 165b.

Finally, the flow equipment 160 can further include a drain line 161 connecting the transfer line 144 to the supply and bypass lines 165a-b. Drain prongs of the drain line 161 can have drain valves, pressure chokes (not shown), and the like connected to an outlet of the mud pump 150.

The pressure sensor 240a-d can use any suitable sensor for measuring pressure, such as a pressure transducer, a pressure gauge, a diaphragm-based pressure transducer, a strain gauge-based pressure transducer, an analog device, an electronic device, or the like.

Each choke 122 may include a hydraulic or electric actuator operated by the control system 200 via an auxiliary HPU (not shown). The return choke 122 receiving flow returns diverted from the riser package 30 is operated by the control system 200 to adjust surface backpressure in the riser 22 and the wellbore 12 for well control.

The control system 200 of the drilling system 10 integrates hardware, software, and applications across the drilling system 10 and is used for monitoring, measuring, and controlling parameters in the drilling system 10. In this contained environment of the closed-loop drilling system 10, for example, minute wellbore influxes or losses are detectable at the surface, and the control system 200 can further analyze pressure and flow data to detect kicks, losses, and other events. In turn, at least some operations of the drilling system 10 can be automatically handled by the control system 200.

To monitor operations, the control system 200 uses data from a number of the sensors and devices in the drilling system 10. In particular, the control system 200 uses the one or more pressure sensors 240a uphole of the choke manifold 120 to measure pressure in the flow returns from the riser 22 and the wellbore 12. As the choke 122 in the manifold 120 is adjusted, the one or more pressure sensors 240a measure the surface backpressure SBP applied to the riser 22 and the wellbore 12.

In addition, the control system 200 can use the one or more sensors 240b-d downstream of the mud pumps 150 to measure pressure in the standpipe 152 (i.e., the standpipe pressure SPP). One or more other sensors (i.e., stroke counters) can measure the speed of the mud pumps 150 for deriving the flow rate of drilling fluid into the drillstring 14. In this way, flow into the drillstring 14 may be determined from strokes-per-minute and/or standpipe pressure SPP. Flowmeters (not shown) after the pumps 150 can also be used to measure flow-in to the wellbore 12. Coriolis flowmeters can be used, and a high-pressure Coriolis flowmeter can be used on a standpipe manifold for the standpipe 152.

One or more sensors (not shown) can measure the volume of fluid in the mud tanks 142 and can measure the rate of flow into and out of mud tanks 142. In turn, because a change in mud tank level can indicate a change in drilling fluid volume, flow-out of the wellbore 12 may be determined from the volume entering the mud tanks 142.

Rather than relying on conventional pit level measurements, paddle movements, and the like, the drilling system 10 can use mud logging equipment and flowmeters to improve the accuracy of detection. For example, the drilling system 10 preferably uses the flowmeter 124, such as a Coriolis mass flowmeter, on the choke manifold 120 to capture fluid data-including mass and volume flow, mud weight (i.e., density), and temperature-from the returning annular fluids in real-time, at a sample rate of several times per second. Because the Coriolis flowmeter 124 gives a direct mass rate measurement, the flowmeter 124 can measure gas, liquid, or slurry. Other sensors can be used, such as ultrasonic Doppler flowmeters, SONAR flowmeters, magnetic flowmeter, rolling flowmeter, paddle meters, etc.

Each pressure sensor 240a-d may be in data communication with the control system 200. The return pressure sensor 240a measures surface backpressure (SBP) exerted by the returns choke 122. The pressure sensor 240d and/or the supply pressure sensor 240b measures standpipe pressure (SPP) to the Kelly hose, whereas the pressure sensor 240d and/or the bypass pressure sensor 240c measures the standpipe pressure SPP to the clamp 174 during connection of a stand of pipe.

As noted above, the return flowmeter 124 may be a mass flow meter, such as a Coriolis flowmeter, and is in data communication with the control system 200. The return flowmeter 124 connected in the flow line 62 downstream of the returns choke 122 measures a flow rate of the returns. A supply flowmeter (not shown) can measure a flow rate of drilling fluid supplied by the mud pump 150 to the drillstring 14 via the top drive 116. Additional sensors can measure mud gas, flow line temperature, mud density, and other parameters.

With this overview of an example drilling system 10 provided above, discussion turns to operation of the drilling system 10 in drilling a wellbore 12. During drilling operations (i.e., MPD operations), the mud pumps 150 pump drilling fluid from the transfer line 144 (or fluid tank connected thereto), through the standpipe 152 and the Kelly hose to the top drive 116. The drilling fluid may include a base liquid, such as oil, water, brine, or a water/oil emulsion. The base oil may be diesel, kerosene, naphtha, mineral oil, or synthetic oil. The drilling fluid may further include solids dissolved or suspended in the base liquid, such as organophilic clay, lignite, and/or asphalt, thereby forming a mud.

The drilling fluid at the inlet flows into the drillstring 14 via the top drive 116 and flow subcomponent 178. The drilling fluid flows down through the drillstring 14 and exits the drill bit 18 of the BHA 16, where the fluid circulates the cuttings away from the drill bit 18 and returns the cuttings up an annulus formed between the casing or wellbore 12 and the drillstring 14. The returns (drilling fluid plus cuttings) flow through the annulus to the wellhead 20 and then continue into the annulus of the riser 22 up to the RCD 60.

At the RCD 60, the drilling system 10 uses the RCD 60 to keep the well closed to atmospheric conditions. The returns are diverted into the return flow line 32 and continue through the returns choke 122 and the flowmeter 124. Therefore, fluid leaving the wellbore 12 flows through the automated choke manifold 120, which measures return flow (e.g., flow-out) and density using the flowmeter 124 installed in line with the chokes 122. The returns then flow into the shaker 140, which removes the cuttings. As the drilling fluid and returns circulate, the drillstring 14 may be rotated by the top drive 116 and lowered by the traveling block 114, thereby extending the wellbore 12 into the lower formation F.

Throughout the drilling operation, the fluid data and other measurements noted herein are transmitted to the control system 200, which in turn operates drilling functions. In particular, the control system 200 operates the automated choke manifold 120 to manage surface backpressure and flow during drilling. This can be achieved using an automated choke response in the closed and pressurized circulating system 10 made possible by the RCD 60.

To do this, the control system 200 controls the chokes 122 with an automated response by monitoring the flow-in and the flow-out of the well, and software algorithms in the control system 200 seek to maintain a mass flow balance. If a deviation from mass flow balance is identified, the control system 200 initiates an automated choke response that changes the well's annular pressure profile and thereby changes the wellbore's equivalent mud weight. This automated capability of the control system 200 allows the control system 200 to perform dynamic well control or CBHP techniques.

Software components of the control system 200 then compare the flow rate in and flow rate out of the wellbore 12, the injection or standpipe pressure SPP (measured by the one or more sensors 240b-d), the surface backpressure SBP (measured by the one or more pressure sensors 240a upstream from the drilling chokes 122), the position of the chokes 122, and the mud density, among other possible variables. Comparing these variables, the control system 200 then identifies minute downhole influxes and losses on a real-time basis to manage the annular pressure (AP) during drilling by apply adjustments to the surface backpressure (SBP) with the choke manifold 120.

By identifying the downhole influxes and losses during drilling, for example, the control system 200 monitors circulation to maintain balanced flow for CBHP under operating conditions and to detect kicks and lost circulation events that jeopardize that balance. The drilling fluid is continuously circulated through the drilling system 10, choke manifold 120, and the Coriolis flowmeter 124. As will be appreciated, the flow values may fluctuate during normal operations due to noise, sensor errors, etc. so that the control system 200 can be calibrated to accommodate for such fluctuations. In any event, the control system 200 measures the flow-in and flow-out of the well and detects variations. In general, if the flow-out is higher than the flow-in, then fluid is being gained in the drilling system 10, indicating a kick. By contrast, if the flow-out is lower than the flow-in, then drilling fluid is being lost to the formation, indicating lost circulation.

To then control pressure, the control system 200 introduces pressure and flow changes to the incompressible circuit of fluid at the surface to change the annular pressure profile in the wellbore 12. In particular, using the choke manifold 120 to apply surface backpressure SBP within the closed loop, the control system 200 can produce a reciprocal change in BHP. In this way, the control system 200 uses real-time flow and pressure data and manipulates the surface backpressure to manage wellbore influxes and losses.

To do this, the control system 200 uses internal algorithms to identify what event is occurring downhole and can react automatically. For example, the control system 200 monitors for any deviations in values during drilling operations, and alerts the operators of any problems that might be caused by a fluid influx into the wellbore 12 from the formation F or a loss of drilling mud into the formation F. In addition, the control system 200 can automatically detect, control, and circulate out such influxes and losses by operating the chokes 122 on the choke manifold 120 and performing other automated operations.

A change between the flow-in and the flow-out can involve various types of differences, relationships, decreases, increases, etc. between the flow-in and the flow-out. For example, flow-out may increase/decrease while flow-in is maintained; flow-in may increase/decrease while flow-out is maintained, or both flow-in and flow-out may increase/decrease.

During drilling operations, the control system 200 operates the return choke 122 so that a target bottom hole pressure (BHP) is maintained in the annulus during the drilling operation. The target BHP may be selected within a drilling window defined as greater than or equal to a minimum threshold pressure, such as pore pressure (PP), of the lower formation F and less than or equal to a maximum threshold pressure, such as fracture pressure (FP), of the lower formation, such as an average of the pore and fracture BHPs. Alternatively, the minimum threshold may be stability pressure and/or the maximum threshold may be leak off pressure. Alternatively, threshold pressure gradients may be used instead of pressures and the gradients may be at other depths along the lower formation F besides bottom hole, such as the depth of the maximum pore gradient and the depth of the minimum fracture gradient. Alternatively, the control system 200 may be free to vary the BHP within the window during the drilling operation. A static density of the drilling fluid (typically assumed equal to returns; effect of cuttings typically assumed to be negligible) may correspond to a threshold pressure gradient of the lower formation F, such as being greater than or equal to a pore pressure gradient.

During the drilling operation, the control system 200 can execute a real-time simulation of the drilling operation to predict the actual BHP from measured data, such as from the standpipe pressure SPP measured from the sensor 240b-d, mud pump flowrate measured from the supply flowmeter 166a, wellhead pressure from any of the sensors, and return fluid flowrate measured from the return flowmeter 124. The control system 200 then compares the predicted BHP to the target BHP and adjusts the return choke 122 accordingly.

During the drilling operation, the control system 200 also performs a mass balance to monitor for instability of the lower formation F, such as a kick even or lost circulation event. As the drilling fluid is being pumped into the wellbore 12 by the mud pump 150 and the returns are being received from the return flow line 32, the control system 200 may compare the mass flow rates (i.e., drilling fluid flow rate minus returns flow rate) using the respective flowmeters 124, 166a. The control system 200 may use the mass balance to monitor for formation fluid (not shown) entering the annulus and contaminating the returns or returns entering the formation F.

Upon detection of instability (e.g., kick), the control system 200 takes remedial action, such as diverting the flow of returns from an outlet of the return flowmeter 124 to the mud gas separator 130. A gas detector of the separator 130 can use a probe having a membrane for sampling gas from the returns, a gas chromatograph, and a carrier system for delivering the gas sample to the chromatograph. The control system 200 may also adjust the returns choke 122 accordingly, such as closing the choke 122 in response to a kick and opening the choke 122 in response to loss of the returns.

Alternatively, the control system 200 may include other factors in the mass balance, such as displacement of the drillstring and/or cuttings removal. The control system 200 may calculate a rate of penetration (ROP) of the drill bit 18 by being in communication with the drawworks and/or from a pipe tally. A mass flowmeter may be added to the cuttings chute of the shaker 140. and the control system 200 may directly measure the cuttings mass rate.

Understanding the examples described above for the drilling system 10 and the control system 200, discussion now turns to some additional details of the components of the control system 200. In particular, FIG. 2 schematically illustrates some details of the control system 200 of the present disclosure.

The control system 200 includes a processing unit 210, which can be part of or can include one or more of a computer system, a server, a programmable control device, a programmable logic controller, etc. Using input/output interfaces 230, the processing unit 210 can communicate with the rig 110, the choke manifold 120, and other system components to obtain and send communication signals, sensor signals, actuator signals, and control signals 232 for the various system components as the case may be. In terms of the current controls discussed, the signals 232 can include, but are not limited to, the choke position signals, position of the traveling block 114, speed of the drawworks 115, and the like, among other signals, such as pressure signals, flow signals, temperature signals, fluid density signals, etc.

As shown, the choke manifold 120 includes the chokes 122a-b, the flowmeter 124, and pressure sensors 240a, among other elements, such as a local controller (not shown) to control operation of the manifold 120. Of course, the choke manifold 120 can also include a hydraulic power unit (HPU) and/or electric motor to actuate the chokes 122. The control system 200 is communicatively coupled to the manifold 120 and has a choke controller 214, which can have a control panel and a user interface and can include processing capabilities to monitor and control the manifold 120.

The processing unit 210 also communicatively couples to a database or storage 220 having setpoints 222, a hydraulics model 224, a motion model 226, and other stored information. The setpoints 222 define values and targets for various system variables, such as surface backpressure and the like. The hydraulics model 224 characterizes the well pressure system. The information for the hydraulics model 224 can be stored in any suitable form, such as lookup tables, curves, functions, equations, data sets, etc. Additionally, multiple hydraulics models 224 or the like can be stored and can characterize the drilling system (10) in terms of different system arrangements, different drilling fluids, different operating conditions, and other scenarios.

As will be appreciated, the hydraulics model 224 of the control system 200 can be built based on the various components, elements, and the like in drilling system 10. The hydraulics model 224 can be built with any complexity desired to model the drilling system 10, which as noted above with reference to FIG. 1 can have a great deal of complexity and information associated with it and which can change over time depending on drilling parameters.

The motion model 226 characterizes the vessel 100 and its motion in response to environmental conditions. The information for the motion model 226 can be stored in any suitable form, such as lookup tables, curves, functions, equations, data sets, etc. Further details related to the motion model 226 are discussed much further below.

The processing unit 210 operates a pressure controller 212 according to the present disclosure, which uses the hydraulics model 224. In particular, the processing unit 210 uses the current pressure profile from the pressure controller 212 to operate the choke controller 214 for monitoring and controlling the choke(s) 122a-b. For example, the processing unit 210 can transmit signals to one or more of the chokes 122a-b of the drilling system 10 using any suitable communication. In general, the signals are indicative of a choke position or position adjustment to be applied to the chokes 122a-b. Typically, the chokes 122a-b are controlled by hydraulic power so that the signals 232 transmitted by the processing unit 210 may be electronic signals that operate solenoids, valves, or the like of an HPU for operating the chokes 122a-b.

As shown here in FIG. 2, two or more chokes 122a-b may be used. The same choke controller 214 can apply adjustments to both chokes 122a-b, or separate choke controllers 214 can be used for each choke 122a-b. In fact, the two chokes 122a-b may have differences that can be accounted for in the two choke controllers 214 used.

As discussed herein, the control system 200 uses the choke controller 214 tuned in real-time to manage the pressure regulation (e.g., surface backpressure), and the control system 200 uses pressure measurements from pressure sensors 240a associated with the choke(s) 122a-b to determine the surface backpressure of the disclosed system (10) so certain targeted setpoints 222 can be reached and maintained within acceptable tolerances.

At times during operation, the drillstring 14 may need to be POOH and then RIH. For example, the drillstring 14 may need to be removed from the wellbore (12) stand-by-stand to replace or change components of the BHA (16). The drillstring 14 may then be reinserted stand-by-stand into the wellbore (12) to continue drilling into the formation (F). Also, when operators make a connection of a new stand of drillpipe at the rig 110 during drilling, the drillstring 14 may be pulled in the wellbore (12) by the traveling block 114 and then run in the wellbore (12) by the traveling block 114 to ream the previously drilled section of the wellbore (12) before continuing with drilling. Once the reaming is done, a new stand of drillpipe can be connected to the drillstring 14 so further drilling of the formation (F) can be continued.

As discussed herein, the movement of the drillstring 14 in the wellbore (12) may produce a piston effect (swabbing/surging) that changes a downhole pressure in the wellbore (12). To handle swab and surge effects during POOH and RIH respectively during the drilling state of the drilling operation, the processing unit 210 uses a swab/surge controller 216, which operates in conjunction with the pressure controller 212 and the choke controller 214 to maintain the bottom hole pressure (within tolerances) as the processing unit 210 moves the traveling block 114 with the drawworks 115 during POOH and RIH in the drilling states of the drilling operation. For surge/swab control during tripping, the swab/surge controller 216 determines that the drillstring 14 is to be run out of (or into) the hole at a given speed and determines the “end of pipe” condition (i.e., open, closed, or auto-fill) of the drillstring 14. In addition, an optimum pipe velocity profile versus depth that maintains the drilling margin can be calculated.

For example, the traveling block 114 of the rig 110 may be supported by wire rope connected at its upper end to the crown block 112. The wire rope may be woven through sheaves of the blocks 112, 114 and extend to the drawworks 115 for reeling thereof, thereby raising or lowering the traveling block 114 relative to the derrick of the rig 110.

To handle swab effects when POOH, the swab/surge controller 216 can perform automatic adjustments to the choke(s) 122a-b in reactive or proactive ways. In one arrangement to handle swab effects when POOH, the swab/surge controller 216 can use the hydraulics model 224 and can determine an optimal speed for moving the drillstring 14. The swab/surge controller 216 determines choke and SBP setpoints 222 associated with that determined speed and sends commands to the drawworks 115 to move the traveling block 114 and connected drillstring 14 at that determined speed. As the drillstring 14 is moved, the swab/surge controller 216 then automatically adjusts the choke(s) 122a-b to maintain the SBP so the BHP stays within tolerances and can prevent formation fluid from entering the wellbore due to swab effects. Additional details for these automatic adjustments are described in U.S. Pat. No. 11,047,224, which is incorporated herein by reference.

The swab/surge controller 216 can likewise perform automatic adjustments to the choke(s) 122a-b in comparable reactive or proactive ways to handle surge effects when RIH during drilling operations. In one arrangement to handle swab effects when POOH, the swab/surge controller 216 uses the hydraulics model 224 and determines an optimal speed for moving the drillstring 14. The swab/surge controller 216 determines choke and SBP setpoints 222 associated with that determined speed and sends commands to the drawworks 115 to move the traveling block 114 and connected drillstring 14 at that determined speed. As the drillstring 14 is moved, the swab/surge controller 216 then automatically adjusts the choke(s) 122a-b to maintain the SBP so the BHP stays within tolerances and can prevent wellbore fluid from entering the formation F due to surge effects. Additional details for these automatic adjustments are also described in incorporated U.S. Pat. No. 11,047,224.

Heaving motions caused by the environmental influences on the drilling vessel (100) may be encountered during the drilling state. This heaving can be particularly challenging during drilling operations, as it can affect the stability and positioning of the drillstring 14 that extends from the rig 110 to the seabed. During the drilling operations in the drilling state, certain amounts of these heaving motions can be handled by the crown block heave compensator 117, which is mounted above the rig's drive unit and can counteract the vertical motion (heave) of the vessel (100) due to ocean swells.

The crown block heave compensator 117 on the rig 110 is connected to the top of the drillstring 14, often through the top drive system, so a consistent weight on the drill bit (18) can be maintained irrespective of the rig's vertical motion. The crown block heave compensator 117 can include a counterbalancing system and a control unit. The counterbalancing system can have an arrangement of counterweights or springs that absorb and compensate for the up-and-down movement of the rig 110 to reduce the transfer of the heave motion to the drillstring 14. For its part, the control unit of the crown block heave compensator 117 can adjust the compensating force in real-time based on the heave movements detected using integrated sensors. In turn, the compensating force can be adjusted using integrated hydraulic or pneumatic components of the crown block heave compensator 117.

During non-drilling states when the drillstring 14 is held in the slips 111 at the rig 110, however, the crown block heave compensator 117 cannot provide its benefits for heave compensation. When a current section of drillpipe has been drilled down, for example, a new drillpipe section needs to be added to the drive unit so drilling can resume. At this point, a connection is performed, in which the drillstring 14 is lowered from the drive unit to the rotary table and is held in place to the rig floor by the slips 111. Once the drillstring 14 is put in the slips 111, the drive unit is disconnected, and the crown block heave compensator 117 is no longer operational. At that point, the bottom hole pressure in the drilled wellbore (12) may be prone to fluctuations caused by the drillstring's motion mimicking the vessel motion in response to the met-ocean conditions.

The non-drilling state discussed herein may be longer than a typical amount of time needed to connect/disconnect a drillpipe stand for the drillstring 14. All the same, although most connections can be made rather quickly (e.g., usually within a couple of minutes), the disclosed techniques can be used for such a typical amount of time depending on the needs.

In addition to the connections and break out of drillpipe, there may be other non-drilling states in which the drillstring 14 may be held for periods at the rig 110. For example, the drillstring 14 may be held in the slips 111 during repairs to the top drive system, the drawworks 115, and the like. As noted, the drillstring 14 may be held in the slips 111 for prolonged periods for any number of reasons. Of course, once the slips 111 are removed and the drillstring 14 is reconnected to the crown block heave compensator 117, the BHP fluctuations can then be mitigated again using the crown block heave compensator 117. Moreover, when the drillstring 14 is tripped in the wellbore 12, the swab/surge controller 216 can be used as noted above. Nevertheless, fluctuations in the bottom hole pressure that happen during the non-drilling states can be significant and can lead to wellbore collapse, fluid losses to the formation, or fluid gains from influxes.

Thus, during the non-drilling state as the drillstring 14 is held in slips 111 at the rig 110, the control system 200 uses a heave compensation controller 250, which operates in conjunction other parts of the control system 200 to maintain the bottom hole pressure (within tolerances) as the drillstring 14 is held at the rig 110 and the drilling vessel (100) is subjected to the influences from the ocean environment. To do this, the heave compensation controller 250 determines that the drilling system 10 is in a non-drilling state, such as when the drillstring 14 is supported in the slips 111 at the rig 110, being concurrently held in the wellbore (12). The heave compensation controller 250 can also determine the “end of pipe” condition (i.e., open, closed, or auto-fill) of the drillstring 14.

To handle the piston (swab) effects when the drilling vessel (100) is heaved upward as the drillstring 14 is held in the non-drilling state, the heave compensation controller 250 estimates heave motions, calculated piston effects, and determines adjustments so the heave compensation controller 250 can perform automatic adjustments to the pressure regulation (e.g., the choke(s) 122a-b for surface backpressure) in reactive or proactive ways. To handle the piston (swab) effects when the continuous flow apparatus 170 is used as the drillstring 14 is held in the slips 111, for example, the heave compensation controller 250 uses the hydraulics model 224 to determine choke and SBP setpoints 222 so the BHP can stay within tolerances to prevent formation fluid from entering the wellbore (12) due to swabbing piston effect from the heaving motion.

The heave compensation controller 250 can likewise perform automatic adjustments to the pressure regulation (e.g., the choke(s) 122a-b for surface backpressure) in comparable reactive or proactive ways to handle the piston (surge) effects when the drilling vessel 100 is heaved downward as the drillstring 14 is held in the non-drilling state at the rig 110. To handle the piston (swab) effects when the continuous flow apparatus 170 is used as the drillstring 14 is held in the slips 111, the heave compensation controller 250 uses the hydraulics model 224 to determine choke and SBP setpoints 222 so the BHP can stay within tolerances to prevent wellbore fluid from entering the formation F due to surge effects from the heaving motion.

The goal of the automatic controllers 212, 214, 216, 250 during drilling states and non-drilling states is to satisfy downhole criteria, such as keeping the annular pressure greater than the pore pressure (AP>PP), greater than wellbore strengthening pressures (AP>WBS), less than leak off test pressure (AP<LOT), less than the fracture pressure (AP<FP), and less than formation integrity test pressure (AP<FIT). During drilling, for example, the bottom hole pressure (BHP) in the drilling system 10 is calculated so a narrow drilling window can be maintained to prevent typical drilling hazards like kicks or lost circulation. The calculations of the bottom hole pressure are based on both the drilling system 10 and the hydrostatic and dynamic pressures (hydraulic model 224) in the drilling operation. As is known, the hydrostatic pressure is the pressure exerted by the column of drilling fluid (mud) and is calculated using known techniques.

The surface backpressure (SBP) is the additional pressure applied at the surface by the drilling system 10 to precisely manage the pressure in the well. This surface backpressure can be determined directly from the drilling system 10. The drilling fluid circulated through the drillstring 14 and the annulus encounters resistance that leads to friction pressure losses, which depend on the mud properties, flow rate, and wellbore geometry. The bottom hole pressure is calculated by (i) the hydrostatic pressure (i.e., calculated as the density of the drilling fluid multiplied by the depth of the well and the gravitational constant) added to (ii) the surface backpressure (determined from the drilling system 10) and subtracting (iii) the frictional pressure losses (calculated using drilling hydraulics models to estimate the pressure loss due to the mud circulation). These calculations are performed by software of the control system 200 using real-time monitoring of the dynamic conditions.

During the non-drilling states, the surface backpressure (SBP) can be applied to manage the pressure in the well in a number of ways. For example, pumping down the drillstring 14 may be available. During certain non-drilling states (e.g., when drillpipe is being connected to the drillstring 14 held at the rig 110), for example, the drilling system 10 can use the continuous flow apparatus 170 to pump fluid through the flow subcomponent 178 connected to the clamp 176 and the bypass line 165b in one configuration to adjust the surface backpressure. In this configuration, fluid circulation occurs in the closed looped drilling system 10 so the choke manifold 120 can provide the surface backpressure as the form of pressure regulation to control the bottom hole pressure of the wellbore 12 in the non-drilling state.

During certain non-drilling states, the drilling system 10 can use the booster line 25 connected to the bottom of the riser 22 where the riser 22 meets the wellhead 20. The fluid provided by the riser booster line 25 can be used, especially in deepwater installation, to supply pressure regulation (e.g., surface backpressure) during drillpipe connections or other situations. When making connections, for example, the booster line 25 can be used to compensate for the loss of friction pressure in the well should pumping down the drillstring 14 (e.g., using flow subcomponent 178, clamp 176, bypass line 165b) be unavailable, interrupted, or changed. In some instances, the column of fluid in the booster line 25 can compensate for changes to the annular well pressure.

In one configuration, circulation can be provided through the MPD manifold 120 using the riser booster line 25 and the rig's mud pump 150 to provide pressure regulation in the wellbore 12, which can be used to compensate for surge and swab effect due to rig heave during connections or other non-drilling state. Additionally, although not shown in FIG. 1, a subsea pump and a subsea choke valve can be used subsea in conjunction with the booster line 25 to provide pressure regulation in the wellbore annulus.

Continuing with the discussion of the automatic heave compensation performed by the control system 200, FIG. 3 illustrates a process 300 used in drilling a wellbore 12 in a formation F with a drilling system 10. (Reference numerals for components in FIG. 1-2 are provided for better understanding.) As noted above, the drilling system 10 can be used on the drilling vessel 100 in an ocean environment, and the drilling system 10 can circulate fluid in a closed loop between the drillstring 14 and the wellbore 12. According to present example, the automated heave compensation controller 250 of the control system 200 disclosed above can perform one or more process blocks of FIG. 3.

In the process 300, the control system 200 stores a motion model 226 for the drilling vessel 100 (Block 302). The motion model 226 is based on one or more vessel characteristics of the drilling vessel 100 and models predicted movement of the drilling vessel 100 in response to influence from one or more environmental characteristics in the ocean environment. To model the predicted movement, the motion model 226 can include transfer function calculations that define how one or more environmental conditions (measurements) indicative of the influence from the ocean environment can cause the predicted movement of the drilling vessel 100.

Motions of the drilling vessel 100 exposed to ocean waves can be defined by response amplitude operators (RAOs). The RAOs are transfer functions that define responses of the drilling vessel 100 in its available degrees of freedom in response to wave directions and periods. As will be appreciated, the vessel 100 can have as many as six degrees of freedom, which include translations of surge, sway, and heave and include rotations of roll, pitch, yaw. The RAOs can be calculated in a number of ways, such as those known in the art.

During operation of the drilling system 10, the heave compensation controller 250 determines the motion of the drilling vessel 100 (i.e., “vessel motion”) produced in response to the ocean environment's influence on the drilling vessel 100 (Block 304). To do this, the heave compensation controller 250 obtains one or more environmental measurements indicative of the influence from the ocean environment on the drilling vessel 100. The heave compensation controller 250 then predicts the predicted movement of the drilling vessel 100 by motion modelling the movement of the drilling vessel 100 in response to influence from the one or more environmental measurements as the one or more environmental characteristics acting on the one or more vessel characteristics of the drilling vessel 100. The heave compensation controller 250 can use an artificial intelligence (AI) function, such as a machine learning model disclosed below to predict the predicted movement of the drilling vessel 100.

When predicting the predicted movement of the drilling vessel 100, the heave compensation controller 250 can supplement the predicted movement by obtaining one or more vessel measurements indicative of a measured movement of the drilling vessel 100 in the ocean environment. The measured movement can then be used in the motion modeling to improve the predicted movement of the drilling vessel 100 as it experiences current conditions in the ocean environment.

The heave compensation controller 250 also obtains a downhole pressure (e.g., a bottom hole pressure) in the wellbore 12 (Block 306). For example, the heave compensation controller 250 may determine a bottom hole pressure of the fluid in the wellbore using the techniques as described above.

The heave compensation controller 250 then identifies between a drilling state and a non-drilling state of the drilling system 10 during operation (Block 308). In other words, the heave compensation controller 250 identifies when drillstring 14 is being used for drilling (i.e., active or drilling state) or when the drillstring 14 is currently being held in the slips 111 (e.g., inactive or non-drilling state). The non-drilling state is indicative of the drillstring 14 being held at the drilling vessel 100 at least during vessel motion. The held drillstring 14 during the vessel motion is expected to produce a piston effect that changes the downhole pressure in the wellbore 12. For its part, the drilling state is indicative of the drillstring 14 being tripped in the wellbore 12 (i.e., RIH or POOH).

As noted, the heave compensation controller 250 uses MPD controls and equipment to perform a first (nondrilling) heave compensation technique or mode for heave effects during nondrilling times (e.g., when the drillstring 14 is held in the slips 111 on the rig floor during connections, repairs, delays, etc.) and to perform a second (drilling) heave compensation technique or mode for heave effects during drilling. The heave compensation controller 250 can be configured to automatically switch between the two modes using a machine learning model disclosed below. If fact, the heave compensation controller 250 can use such a machine learning model to identify the state either in addition to or in the alternative to predicting the predicted movement of the drilling vessel 100.

To identify the drilling state and the non-drilling state in Block 308, the heave compensation controller 250 can detect the state directly using one or more sensors of the drilling system 10. Alternatively, the heave compensation controller 250 can interrogate components of the drilling system 10 with a request for the state, or the heave compensation controller 250 may receive an indication of the state automatically from components of the drilling system 10 during the standard course of operation. Moreover, the heave compensation controller 250 can receive a manual input of the current state of the drilling system 10.

As noted above, when the drillstring 14 is in the rotary table slips 111 especially in deepwater operations, BHP fluctuations occur due the inability to compensate for the vessel motion caused by heave effects. Those BHP fluctuations can lead to adverse effects, such as fluid losses, fluid gains, and wellbore instability. By applying the techniques of the present disclosure, the disclosed control system 200 can detect and compensate for the vessel's heave motion, attempting to maintain a more constant bottom hole pressure.

In response to identification of the non-drilling state, the heave compensation controller 250 determines a first adjustment to the pressure regulation (e.g., surface backpressure or other form of available pressure regulation) of the drilling system 10 (Block 310). The change to the bottom hole pressure produced by the first piston effect is then counteracted by automatically adjusting the pressure regulation (e.g., surface backpressure) according to the first adjustment (Block 316). The heave compensation controller 250 can use a machine learning model as discussed below to determine the adjustment, and this determination can be made either in addition to or in the alternative to identifying the state and predicting the predicted movement of the drilling vessel 100 noted above. Preferably, the heave compensation controller 250 uses the machine learning model to identify the state, to predict the predicted movement, and to determine the adjustment.

In response to identification of the drilling state, however, at least one compensation due to heave motion of the vessel 100 can be handled using the crown block heave compensator 117 on the rig 110 (Block 312). Yet, during tripping in the drilling state, compensation due to surging/swabbing when POOH, RIH of the drillstring 14 can be handled using the functionalities of the swab/surge controller 216 of the disclosed control system 200 (Block 314). The change to the bottom hole pressure produced by the second piston effect is then counteracted by automatically adjusting the pressure regulation (e.g., surface backpressure) according to the second adjustment (Block 316).

Although FIG. 3 shows example blocks of the process 300, the process 300 in some implementations may include additional blocks, fewer blocks, different blocks, or differently arranged blocks than those depicted in FIG. 3. Additionally, or alternatively, two or more of the blocks of the process 300 may be performed in parallel. Some additional details of the steps in the process 300 of FIG. 3 are discussed below.

In Block 304 of FIG. 3, the heave compensation controller 250 programmed with the motion model 226 can use various sensors and measurements to obtain data to determine the vessel motion. For example, the heave compensation controller 250 can obtain one or more environmental measurements, which can be indicative of the influence from the ocean environment on the drilling vessel 100. In particular, the heave compensation controller 250 can detect the one or more environmental measurements directly using one or more environmental sensors 262 of an environmental system 260 as in FIG. 2 on the drilling vessel 100.

The environmental sensors 262 can include any suitable sensors to measure environmental conditions. Typical sensos can include an anemometer, an ultrasonic wind sensor, a sonic wave level sensor, a wave gauge having a temperature compensated pressure sensor, a wave staff for measuring water level, a wave logger, and the like. Additionally, the environmental sensors 262 can use Micro-electromechanical system (MEMS) sensors, such as accelerometers, gyroscopes, inclinometers, etc. In general, the environmental sensors 262 can measure environmental conditions, including, but not limited to, wave height, wave direction, wind speed, wind direction, velocity of ocean currents, and the like. Measuring these condition, adjusting for these conditions effectively, and implementing predictive technology as disclosed herein can lead to effective heave compensation.

Alternatively, the heave compensation controller 250 can receive one or more environmental measurements from one or more external sources, such as receiving data of the Wave Observation (IOOS) from a governmental source. The heave compensation controller 250 can interrogate the environmental system 260 of the drilling vessel 100 with a request for the environmental measurements, or the heave compensation controller 250 can receive the environmental measurements automatically from the environmental system 260 during the standard course of operation. Moreover, the heave compensation controller 250 can receive environmental measurements as manual inputs. For example, met-ocean data measurements can be input to the disclosed control system 200 from various data sources, such as the Dynamic Position Officer (DPO). This data can include the raw amplitude and wavelength data for ocean waves.

As noted, data associated with met-ocean conditions producing heave can be obtained from the environmental system 260. For example, the data used to determine the met-ocean heave can be obtained using a Wellsite Information Transfer Specification Markup Language (WITSML) signal from the rig's DPO Kongsberg system 260. In turn, the heave compensation controller 250 utilizes the nominal heave to calculate a surge/swab effect on the bottom hole pressure. A reference table or formula can be used to determine the adjustments to the pressure regulation (e.g., surface backpressure) needed to meet the different rig accelerations encountered during a heave cycle. Based on the live heave signal or a heave curve, the heave compensation controller 250 can then automatically adjust the pressure regulation (e.g., the chokes 122a-b for surface backpressure) to maintain a constant bottom hole pressure. In basic terms, upward heave of the rig 110 would require increasing amounts of the pressure regulation (e.g., surface backpressure), whereas downward heave of the rig 110 would require decreasing amounts of the pressure regulation (e.g., surface backpressure).

In the context of offshore drilling operations, the meteorological aspects of met-ocean conditions that can cause vessel motion relate to atmospheric aspects and can include: (i) the speed and direction of wind that can affect the stability of the drilling vessel (100) and the operation of cranes and other equipment; (ii) temperature and humidity that can impact both equipment performance (iii) precipitation (rain, snow, and hail) that can affect sea state and equipment functionality; and other atmospheric aspects. The oceanographic aspects of met-ocean conditions that can cause vessel motion relate to physical and chemical aspects of the ocean environment and can include: (i) wave height and period that can be relevant to stability and operation of equipment; (ii) surface and subsurface ocean currents that can affect the positioning of the drillship or platform and the behavior of the drilling riser (22); (iii) general sea state conditions of the sea surface, influenced by wind, wave, and current conditions; and other aspects of the ocean environment. In general, the environmental measurements from the ocean environment can include one or more of: met-ocean data, a meteorological condition, an atmospheric aspect, a speed of wind, a direction of wind, a temperature, a humidity, precipitation, an oceanographic condition, a wave height, a wave period, a surface ocean current, a subsurface ocean current, and a sea surface condition.

Additionally, data measurements for vessel motion can be obtained from existing data and feeds available on the vessel 100. For example, the data for vessel motion can be obtained from the Dynamic Position Officer (DPO) or can be obtained from a data feed from a dynamic positioning system 270, such as shown in FIG. 2. For example, a data transfer can be made using WITSML from the dynamic positioning system 270 to the disclosed control system 200. The vessel motion data can include actual vessel motion that includes adjustments of the RAOs.

Data measurements can be obtained from any sensors of the crown block heave compensator 117. The data measurements can be used as recent and historical measurements of the vessel motion. For example, a high-fidelity sensor can be installed on the crown block heave compensator 117. As noted, such a crown block heave compensator 117 can already dampen effects of the vessel motion. The data from the crown block heave compensator 117 can be transferred to the disclosed control system 200. This vessel motion data can provide a recent history of actual vessel motions that includes adjustments of the RAO. In turn, the vessel's RAO characteristics are defined in the disclosed control system 200, which can then calculate the vessel motion based on the data inputs and the vessel's RAO characteristics.

To determine vessel motion in Block 304 of FIG. 3, the control system 200 can obtain other vessel-related measurement(s) indicative of vessel motion of the drilling vessel 100 in the ocean environment. These vessel-related measurements can be automatically obtained from a dynamic positioning system 270 as shown in FIG. 2 of the vessel 100, and the vessel-related measurements can include real-time information about the position and movement of the vessel 100. For example, the vessel 100 maintains a position over the wellbore 12 during operations, and the vessel 100 may use self-propulsion equipment and dynamic positioning or mooring equipment to maintain its position. Accordingly, the vessel 100 may use the dynamic positioning system 270 to accurately maintain position over the well.

The dynamic positioning system 270 can use multiple position/motion inputs to operate correctly and reliably. These position/motion inputs can be supplied by vessel sensors 272 disposed on the vessel 100. Additionally, these position/motion inputs can be supplied by multiple positioning reference systems 274, including one or more Global Navigation Satellite Systems (GNSS), Acoustic Positioning Reference Systems that supply X, Y and Z position coordinates, and Inertial Navigation systems (INS).

As one example, position/motion inputs can be supplied by vessel sensors 272 disposed on the vessel 100. The vessel sensors 272 can use Micro-electromechanical systems (MEMS) sensors. The vessel sensors 272 are used in the automated heave compensation to measure parameters, such as acceleration, angular rate, and displacement with high precision and reliability. Because MEMS sensors use low power, have high sensitivity, and are small in size, using the MEMS sensor for the vessel sensors 272 can be beneficial for real-time monitoring of heave motions.

Several MEMS sensors are available. To monitor heave motion, for instance, three (3) types of MEMS sensors can be used for the vessel sensors 272, and the MEMS sensor can be integrated together onto a single board. For example, the vessel sensors 272 can use accelerometers to measure acceleration forces acting on the vessel 100, which is useful for monitoring heave motions. In addition, the vessel sensors 272 can use gyroscopes to measure the rate of rotation around the vessel's axes, which can indicate pitch and roll movements. Finally, the vessel sensors 272 can use inclinometers to measure the tilt or inclination of the vessel 100, providing additional data.

The vessel sensors 272 can use these three types of MEMS sensors on a single board or device to obtain accurate readings of the vessel movements. To achieve accurate readings from the vessel sensors 272, the vessel sensors 272 can be disposed at a peripheral portion (i.e., stern, back, aft-most part, bow, or foremost part) of the vessel 100 and can be disposed at a central portion (i.e., midship or middle part) of the vessel 100 to calculate a delta or difference in response to the wave effects on the vessel 100. For their part, some environmental sensors 262 can also use these types of MEMS sensors and can be deployed around the vessel 100 on buoys to calculate actual wave characteristics. The positioning system 270 can compare the measurements of actual waves from the environmental sensors 262 to the stern and midship measurements from the vessel sensors 272 to evaluate the effects of the waves. The evaluated effects can then supplement the predicted movement of the drilling vessel during the non-drilling state so adjustments to the pressure regulation provided by the drilling system 10 can mitigate changes in the downhole pressure (e.g., BHP).

There are technical challenges to using MEMS sensors for the vessel sensors 272 and environmental sensors 262. In particular, high-frequency noise can be a primary cause of distorted data among the MEMS sensors. Accordingly, the positioning system 270 can use filtering techniques, such as Kalman filters, to reduce the effect of noise on the data acquired. In another consideration, the MEMS sensors need to be properly calibrated. However, regular calibration and validating with a known reference model can reduce any calibration issues. Finally, the control system 200 needs to be configured using known techniques to integrate data from the MEMS sensors with other sensor data along with environmental information to perform the disclosed heave compensation techniques.

As another example, the dynamic positioning system 270 can use a combination of positioning reference systems 274, such as an INS system and another positioning reference system (e.g., navigation satellite system or an acoustic system). The INS system can use position updates from the positioning reference system to make accurate position calculations. These position measurements and calculations can be correlated to vessel motion (such as heave), which can be used by the heave compensation controller 250 of the disclosed control system 200.

Of course, lacking full automation, vessel-related movements can be obtained manually in a manual input 276 from the Dynamic Position Officer on the vessel 100. Either way, the heave compensation controller 250 can then supplement the predicted movement of the drilling vessel 100 using the measured movements of the drilling vessel 100.

In the identification of Block 308 of FIG. 3, the non-drilling state as noted is indicative of a hold of the drillstring 14 on the drilling vessel 100 at least during vessel motion. For example, the drillstring 14 can be held with slips 111 or other gripping elements of a rotary table on the rig 110 of the drilling vessel 100 during the non-drilling state, and the vessel motion during the non-drilling state may be expected to produce a first-type of piston effect that changes the bottom hole pressure in the wellbore 12. As the vessel 100 heaves upward, the drillstring 14 is expected to be pulled out of the wellbore 12, producing swabbing as a piston effect that can decrease the bottom hole pressure in the wellbore 12. As the vessel 100 heaves downward, the drillstring 14 is expected to run into the wellbore 12, producing surging as a piston effect that can increase the bottom hole pressure in the wellbore 12. In general, the non-drilling state can be defined as any occurrence in which the drillstring 14 is held at the rig 110 and is not connected to the crown block heave compensator 117 on the crown block 112 of the rig 110. As expected, various occurrences during drilling operations can implicate the non-drilling state requiring the disclosed compensation performed by the control system 200.

Connections can be a non-drilling state in which the compensation performed by the disclosed control system 200 may be needed. For example, the swab and surge of the drillstring 14 can affect the BHP to an undesirable extent during a connection that takes seven to ten minutes in rough seas.

By contrast, the drilling state in Block 308 of FIG. 3 is indicative of a trip to move the drillstring 14 in the wellbore 12. The trip of the drillstring 14 is expected to produce a second-type of piston effect that changes the bottom hole pressure of the fluid in the wellbore 12. For example, the drillstring 14 in the drilling state is run in hole (RIH) or pulled out of hole (POOH). Pulling the drillstring 14 out of the wellbore 12 produces swabbing as a piston effect that can decrease the bottom hole pressure in the wellbore 12. Meanwhile, running the drillstring 14 into the wellbore 12 produces surging as a piston effect that can increase the bottom hole pressure in the wellbore 12.

In Block 310 of FIG. 3 to determine the first adjustment once the non-drilling state is initiated or identified, the heave compensation controller 250 can utilize data measurements to calculate an expected change in bottom hole pressure. The data measurements can be obtained from one or more of: vessel motion data, crown block heave compensator data, and met-ocean data. The heave compensation controller 250 then adjusts an amount of pressure regulation (e.g., surface backpressure) that the drilling system 10 applies to compensate (dampen) for the BHP fluctuations. Once the rig 110 is ready to drill again, the control system 200 can automatically stop the BHP fluctuation compensation mode.

The determination of the first adjustment can be based on the predicted movement of the drilling vessel 100 determined using the one or more environmental measurements in the motion model 226. This first adjustment is configured to maintain the bottom hole pressure at least within a tolerance. For example, a target can be defined for the bottom hole pressure at a depth in the wellbore. This target can be defined as less than one of: (i) a fracture pressure gradient of the formation, and (ii) a pore pressure gradient of the formation.

In the heave compensation and pressure regulation adjustment, the vessel motions are defined in relation to BHP fluctuations using motion variables and motion calculations. The motion variables include wave amplitude (A) and wavelength (λ). For example, FIG. 4A shows a graph 350 of the heaving (surge and swab) motion 352 of a vessel due to environmental influences, such as waves, while FIG. 4B illustrates another graph 360 of response amplitude operators 362 in response to the heaving motion 352 of the drilling vessel. Wave amplitude (A) in the heave motion 352 of the vessel (100) produces upward/downward motion of the drillstring (14) inside the wellbore (12) as well as producing acceleration of the drillstring (14). A positive (+ve) amplitude in the heave motion 352 of the vessel (100) moving up leads to swabbing of the wellbore (12) and produces a corresponding reduction in BHP. A negative (−ve) amplitude in the heave motion 352 of the vessel (100) moving down leads to a surging effect of the wellbore (12) and produces a corresponding increase in bottom hole pressure. For its part, the wavelength (λ) in the heave motion 352 of the vessel (100) dictates the number of surge/swab cycles per minute.

The motion calculations performed by the heave compensation controller 250 can use Vessel Motion Transfer Function Calculations (RAOs). The transfer functions (RAOs) are defined by the vessel's physical properties (shape, mass, buoyancy, etc.). In their simplest form, these transfer functions may be defined primarily by the vessel's metacentric height. (The vessel's metacenter lies directly above the vessel's center of buoyancy, which is the center of mass of the volume of water that the vessel's hull displaces. When the vessel is at equilibrium, the center of buoyancy is vertically in line with the vessel's center of gravity. When the vessel is upright, a vertical line passes through the original, vertical center of buoyancy and the center of gravity. When the vessel is heeled over to one side, however, the metacenter is the point at which a vertical line through the heeled center of buoyancy crosses the line through the original, vertical center of buoyancy. The metacentric height, which is a measurement of initial static stability of the vessel, is calculated as a distance between the vessel's center of gravity and its metacenter.)

The calculations define how much and how quickly the vessel (100) is expected to move based on the velocity and acceleration of the structure of the vessel (100). The effects on BHP caused by the heave motion of the vessel (100) can be dampened by the vessel's characteristics defined in the transfer functions.

The Vessel Motion Transfer Function calculations are configured to define how different waves will affect the vessel's motions. These motions can include movements like heaving (up and down), pitching (tilting forward and backward), rolling (tilting side to side), yawing (turning left or right), surging (moving forward and backward), and swaying (moving side to side). Heaving motions 352 as noted herein are of particular interest to maintaining the bottom hole pressure in the wellbore (12). To the extent that motion in the other degrees of freedom (pitching, rolling, yawing, surging, and swaying) impact the held drillstring (14) to change the bottom hole pressure, the Vessel Motion Transfer Function calculations can take these additional degrees of freedom into account. In the transfer functions, wave characteristics (like height, period, and direction) are the inputs to the functions, which output the vessel's response in terms of its six degrees of freedom. The sea states are defined using wave spectra describing the distribution of wave energy across different frequencies. Hydrodynamic modeling of the vessel's hydrodynamic characteristics calculates the vessel's response to waves, wind, and other environmental conditions, and the modeling can use computational methods, such as Computational Fluid Dynamics (CFD).

The response amplitude operators (RAOs) 362, such as shown in FIG. 4B, indicate how much the vessel (100) will move in response to a wave of a certain frequency and direction. Each RAO 362 is usually calculated for each degree of freedom of the vessel (100). Here, the RAO 362 for heaving is calculated, being of particular interest to the piston effects for the purposes of the present disclosure.

The actual motion of the vessel (100) in a specific ocean state is calculated by convoluting the wave spectrum with the RAOs 362 by integrating over all the wave frequencies to get the vessel's response. In the end, the results can be expressed statistically by calculating expected maximum response or the response's standard deviation.

In addition to the influence of waves on the vessel (100) as shown in FIGS. 4A-4B, other environmental factors, such as wind and current, can also affect the vessel's motions and can also be considered in the Vessel Motion Transfer Function calculations. These considerations can follow the teachings described above with respect to waves.

In Block 312 of FIG. 3, any changes produced by heaving of the vessel 100 during the drilling state can be mitigated by operating the crown block heave compensator 117 mounted on the drilling system 10. As noted, the crown block heave compensator 117 can mitigate the effects of heave, which is the vertical motion of the vessel 100 caused by sea waves and swells. In offshore drilling operations, maintaining a consistent weight on the drill bit 18 provides for effective drilling. The heave of the vessel 100 can lead to variations in this weight, impacting drilling operations.

The crown block heave compensator 117 is typically a part of the drilling rig's drawworks 115. The crown block heave compensator 117 uses hydraulic or pneumatic cylinders (or accumulators) integrated with the rig's crown block 112, which is the stationary end of the block and tackle system used to move the drillstring 14. The compensator 117 adjusts the length of the drilling line between the crown block 112 and the traveling block 114 (which moves up and down with the drillstring 14) in response to the heave.

The compensator 117 can compensate for heave effects. When the vessel 100 heaves upwards due to wave action, the compensator 117 extends, increasing the length of the drilling line, to prevent the drillstring 14 from being pushed too forcefully into the seabed. Conversely, when the vessel 100 moves downwards, the compensator 117 retracts, shortening the drilling line, to maintain tension and avoid slack in the drillstring 14. This action helps maintain a consistent weight on the drill bit 18, irrespective of the vessel's vertical motion. In addition to enhancing the precision of drilling operations by ensuring consistent downward force on the drill bit 18, the crown block heave compensator 117 can reduce excessive tension or slack in the drillstring 14 and can minimize any impacts of sudden or severe sea conditions on drilling operations.

In Block 314 of FIG. 3, the swab/surge controller 216 determines a second adjustment to the pressure regulation (e.g., surface backpressure used in the drilling stage) of the drilling system 10. Like the first adjustment, this second adjustment is also configured to maintain the bottom hole pressure at least within the tolerance. Again, in Block 316, the change to the bottom hole pressure produced by the second piston effect is then counteracted by automatically adjusting the pressure regulation (e.g., surface backpressure) according to the second adjustment.

In Block 316 of FIG. 3 during either the non-drilling state or the drilling state, the control system 200 as described herein can control surface backpressure as the pressure regulation to control the bottom hole pressure. In Block 316 for the non-drilling state, for example, the heave compensation controller 250 can communicate instructions or other outputs to the choke controller 214 to perform the adjustment to the one or more chokes 122a-b of the drilling system 10 while the continuous flow apparatus 170 directs flow into the drillstring 14. Alternatively, the heave compensation controller 250 can control the chokes 122a-b to make the adjustments if the heave compensation controller 250 includes such features or is incorporated into a choke controller 214.

In any event, adjusting the surface backpressure can involve adjusting a position of at least one choke 122a-b in fluid communication with the fluid flowing out of the wellbore 12 in the closed loop. For control, a position of at least one choke 122a-b can be monitored, and measurement of the surface backpressure of the drilling system 10 upstream of the at least one choke 122a-b can be made. Additional monitoring can be made of the current depth of the drillstring 14 in the wellbore 12, a current position of a traveling block 114 connected to the drillstring 14 at the rig 110 of the drilling system 10, and a current end-of-pipe condition on the drilling system 10 in the wellbore 12.

In Block 316 of FIG. 3 for the drilling state, for example, the swab/surge controller 216 can communicate instructions or other outputs to the choke controller 214 to perform the adjustment to the one or more chokes 122a-b of the drilling system 10. Alternatively, the swab/surge controller 216 can control the chokes 122a-b to make the adjustments if the control system 200 includes such features or is incorporated into a choke controller 214. Moreover, the functionalities of the disclosed control system 200 can account for the surging/swabbing when POOH, RIH using techniques such as disclosed in incorporated U.S. Pat. No. 11,047,224, which is incorporated herein by reference in its entirety.

As noted above, the present disclosure generally relates to systems and methods to monitor and compensate for bottom hole pressure fluctuations caused by heave effects on a drilling vessel 100 used in deepwater MPD operations. In one aspect, the control system 200 can use an artificial intelligence algorithm (i.e., a machine learning model 252) to attain the met-ocean data and to assess vessel motion data with RAOs. The data is analyzed to determine the effects on bottom hole pressure. The control system 200 powered by the machine learning model 252 can then actively compensate or dampen the bottom hole pressure fluctuations based on the vessel motion characteristics.

The control system 200 power by the machine learning model 252 can receive, interpret, and monitor vessel motion based on met-ocean data and can then determine a correct correction factor to the applied surface backpressure to minimize fluctuations in the bottom hole pressure. The machine learning model 252 can also identify when the drillstring 14 is in a drilling state (e.g., being used for drilling) or is in a non-drilling state (e.g., being currently held in the slips 111). Once a non-drilling state is identified, the machine learning model 252 can utilize one or more of the vessel motion data, crown block heave compensator data, and met-ocean data to calculate a change in the bottom hole pressure. The control system 200 then adjusts the amount of surface pressure it applies to compensate (dampen) the BHP fluctuations. Once the rig 110 is ready to drill again, the control system 200 automatically stops this form of compensation for the BHP fluctuation so other forms of compensation disclosed herein can be used. Overall, the control system 200 is capable, fast, and intelligent to adjust the machine learning model 252 based on the current (drilling or non-drilling) and to apply changes dynamically where needed.

Various machine learning models can be used, and complex interactions can be built to develop an automated heave compensation model of the present disclosure. Some example machine learning models suitable for the disclosed techniques include regression models (e.g., liner and non-liner regression models); neural networks (e.g., feedforward neural networks and convolutional neural networks); long short-term memory and gated recurrent units; random forest decision trees and gradient boosting machines; Kalman filtering and particle filtering; clustering algorithms (e.g., K-Means Cluster, DBSCAN, etc.); and deep reinforcement learning and autoencoders. Based on available data, one or all these types of machine language models can be used to develop the machine learning model 252 disclosed herein.

Using an advanced Al model as above and interpreting data from MEMS sensors, crown block compensator, measurement-while-drilling (MWD) tool data, weather pattern data, and the like, the machine learning model 252 can forecast active heave compensation methods during downhole drilling operations using the MPD system 10. Data from one sensor or many sensors can be used to build the machine learning model 252 disclosed herein.

Using data (e.g., from vessel sensors 272, crown block heave compensator 117, MWD tool data, weather pattern data, etc.), for example, the analysis and control functions of the control system 200 interface with other operational components (e.g., pressure controller 212, choke controller 214, etc.) and perform automated heave compensation using a trained and deployed machine learning model 252 to perform the automated heave compensation.

FIG. 5 illustrates a process 400 to develop a machine learning model (252) of the present disclosure for used by the disclosed control system (200). In general, the control system (200) can be used to develop such a machine learning model, or a data collection and processing system or any other appropriate computer system can develop the model for used by the control system (200). As described later, FIG. 6 illustrates an example configuration for a computer system to train and deploy a deep neural network 450 for use in a machine learning model (252) of the present disclosure. Additionally, FIG. 7 illustrates an example of a machine learning model 500 utilized by a control system (200) of the present disclosure.

Turning now to FIG. 5, the process 400 to train the machine learning model uses a data collection and processing system, such as the control system (200) disclosed herein, a programmable logic controller (PLC), an embedded board, and/or some other computer system, including a personal computer (PC) based system (e.g., having a central processing unit (CPU) and/or graphics processing unit (GPU)) or a cloud-based computer system. The process 400 begins with the data collection and processing system collecting all the requisite data from vessel sensors 272, environmental sensors 262, and the like (Block 402). The collected data is divided into input data and output (target) data. Of course, the input data includes variables or conditions used to make predictions in the model. The output (target) data includes variables that model is trying to predict.

The data collection and processing system preprocesses the data into training data for the machine learning model (Block 404). Preprocessing puts the data into data structures suitable for the machine learning model. For example, preprocessing can clean the training data to account for missing values, remove outliers, and correct inconsistencies. Additionally, because the data may come from several types of sensors, the training data may need to be normalized and standardized. Encoding or conversion of the data into other formats may also be necessary.

At this point, feature engineering is performed (Block 406). In this step, the data collection and processing system extracts relevant features from the raw, pre-processed data. These relevant features can include statistical summaries, time-based details, domain specifics, and the like. Based on GPS location, time zone, annual season, storm season, and the like, the data collection and processing system can figure out what the predominant environment is and can then choose a pre-learned model for a set of the disclosed machine learning models.

The data collection and processing system now trains the machine learning model using the training data (Block 408). For example, the data collection and processing system can train a supervised learning model using time series analysis (e.g., long short-term memory (LSTM), gated recurrent units (GRU)) on the preprocessed training data to establish relationships between input variables and output variables (e.g., heave compensation outputs).

Training can involve initialization of the machine learning model by setting initial parameters or weights for the machine learning model. The training data is divided into a training subset and a testing subset. The machine learning model uses the training data to learn the relationship between the input variables and the output (target) variables. This learning involves optimizing the model's parameters so a difference between predicted values and actual target values can be minimized. A loss function, such as Mean Squared Error (MSE) and Mean Absolute Error (MAE), can be used to measure the error of the model's predictions. Additionally, an optimization algorithm, such as gradient descent, can be used to minimize the loss function by iteratively adjusting the model's parameters.

The process 400 then proceeds with testing the machine learning model to evaluate its accuracy (Block 410). For example, the data collection and processing system can compare a prediction accuracy of the trained machine learning model relative to a threshold by testing the trained machine learning model (Decision 412). The testing subset can be used to evaluate how well the model performs on unseen data. A number of evaluation metrics can be used by the regression model and can include: R-squared (R²) measuring the proportion of variance in the target variable that can be explained by the input features; Root Mean Squared Error (RMSE) as the square root of the average squared differences between predicted and actual values; and Mean Absolute Error (MAE) as the average of the absolute differences between predicted and actual values.

Further updating may be required for the machine learning model if the accuracy is not technically accepted (Block 414). Otherwise, the machine learning model can be saved to be utilized in real-time deployment (Block 416). Updating can involve adjusting those parameters not learned from the data but set before training (i.e., the model's hyperparameters), such as learning rate or the number of trees in a random forest, to improve performance. Also, techniques, such as k-fold cross-validation, can be used to ensure the model's performance is consistent across different subsets of the data.

Of course, in deployment, the trained machine learning model is integrated into drilling vessel 100 in ocean an environment to make predictions based on new data. Continuous monitoring of the machine learning model's performance can be performed by the data collection and processing system so the machine learning model is updated as needed to ensure it remains accurate over time.

FIG. 6 illustrates an example configuration for a data collection and processing system to train and deploy a deep neural network 450 for use in a machine learning model as disclosed herein. Once a given neural network has been structured for a task, the neural network is trained by a training framework 454 using a training dataset 452. To begin training the deep neural network (DNN) 450, initial weights of the untrained neural network 456 may be chosen randomly or by pre-training using a deep belief network. The training cycle can then be performed by the training framework 454 in either a supervised manner or an unsupervised manner.

Supervised learning uses the training dataset 452 to teach the machine learning model to yield a desired output. Accordingly, the training dataset 452 includes inputs and desired outputs, which allow the machine learning model to learn over time. Alternatively, when the training dataset 452 includes input having known output, the output of the neural network 456 can be manually graded.

The network processes the inputs and compares the resulting outputs against a set of expected or desired outputs. Errors are then propagated back through the training framework 454. In turn, the training framework 454 can adjust and change the weights that control the untrained neural network 456. The training framework 454 can also provide tools to monitor how well the untrained neural network 456 is converging towards a machine learning model suitable for generating correct answers based on known input data.

The training process repeatedly occurs as the network weights are adjusted to refine the output generated by the neural network 456. The training process can continue until the neural network 456 reaches a statistically desired accuracy associated with a trained neural network 460. The trained neural network 460 can then be deployed to implement any number of machine learning operations in which the trained neural network 460 process a new data 462 to output a result 464.

Supervised learning is typically separated into two types of problems—classification and regression. Classification uses an algorithm to assign the training dataset 452 accurately into specific categories. Regression is used to understand the relationship between dependent and independent variables. Numerous different algorithms and computation techniques can be used in supervised machine learning, including but not limited to, a neural network (as shown here), naïve bayes, linear regression, logistic regression, support vector machines (SVM), k-nearest neighbor, and random forest.

Unsupervised learning is a learning method in which the network uses algorithms to analyze and cluster unlabeled data. These algorithms discover hidden patterns or data groupings. Therefore, the training dataset 452 can include input data without any associated output data. The untrained neural network 456 can learn groupings within the unlabeled input and determine how individual inputs relate to the overall dataset. Unsupervised training can be used for three main tasks—clustering, association, and dimensionality. Clustering is a data mining technique that groups unlabeled data based on similarities and differences. This technique is often used to process raw, unclassified data objects into groups represented by structures or patterns in the information. Association is a rule-based method for finding relationships between variables in a given dataset. This method is often used for market basket analysis. Dimensionality reduction is used when a given dataset's number of features (dimensions) is too high. This technique is commonly used in the preprocessing of data.

Variations of supervised and unsupervised training may also be employed. Semi-supervised learning is a technique in which the training dataset 452 includes a mix of labeled and unlabeled data of the same distribution. Incremental learning is a variant of supervised learning in which input data is continuously used to train the model further. Incremental learning enables the trained neural network 460 to adapt to the new data 462 without forgetting the knowledge instilled within the network during initial training.

As noted above, the heave compensation controller 250 can use artificial intelligence (AI), such as a deep reinforcement learning-based (DRL-based) model. FIG. 7 schematically illustrates a deep reinforcement learning-based (DRL-based) model 500 for use by the heave compensation controller 250 of the present disclosure. The DRL-based model 500 combines a deep learning component with a reinforcement learning framework. The deep learning component can use neural networks 520 to approximate the required functions associated with the multi-dimensional input space of the vessel 100 (i.e., MODU) in the ocean. The reinforcement learning (RL) framework is an agent 510 that learns responses by interacting with an environment 530. The RL framework's agent 510 receives states 512 from the environment 530, performs actions 516, and receives positive or negative returns (feedback 532).

There are several ways to combine the deep learning component and the RL framework. For example, the neural network 520 in the DRL model 500 can be used to represent a policy of the agent (policy-based methods), a value function (value-based methods), or both (actor-critic methods). The present example shows a policy-based method in which the neural network 520 in the DRL model 500 represents a policy 514 of the agent 510. Using experiences (state 512, action 516, reward 532, new state 512, etc.), the agent 510 can learn by updating its policy 514 using gradient descent methods to adjust the weights of the neural network 520. The learning can be based on actual experiences (sampled from the environment 530) or simulated experiences (using previously encountered states). For the heave compensation disclosed herein, the policy 514 would represent the rules or guidelines that would implement the heave compensation of the present disclosure in response to environmental conditions, vessel motions, drilling and non-drilling states, etc.

In summary, after the machine learning model 252 as in FIG. 2 has been trained, the heave compensation controller 250 uses the trained learning model 252 to perform automated heave compensation so adjustments can be made to the amount of surface pressure applied during non-drilling states to compensate (dampen) BHP fluctuations caused by heave motions. When used, the heave compensation controller 250 can perform real-time data fusion. For example, Kalman filtering and particle filtering can be implemented to combine real-time data and to provide accurate state estimates.

Additionally, the heave compensation controller 250 performs predictive analysis. For example, using the trained machine learning model (252), the heave compensation controller 250 predicts future heave movements and adjusts the components of the MPD system 10 in real time. As disclosed herein, the automated heave compensation controller 250 integrates measurements (e.g., vessel motion data from MEMS sensor, positioning systems; environmental data from environmental sensors, data sources; etc.), and the machine learning model (252) to perform proactive heave compensation. Techniques, such as LSTM for time series prediction, Kalman filtering for data fusion, and deep reinforcement learning for decision-making provide the automated heave compensation controller 250 with a robust framework for managing the conditions on the offshore drilling vessel. The automated heave compensation controller 250 operates to mitigate the effects of vessel motion caused by waves and other environmental factors to ensure stability and efficiency of drilling operations by compensating for vertical movements of the vessel 100.

The teachings of the present disclosure can be characterized by the following clauses:

• • 1. A computer-implemented method used with a drilling system (10) on a drilling vessel (100) in an ocean environment to drill a wellbore (12) in a formation (F), the drilling system (10) being configured to circulate fluid in a closed loop between a drillstring (14) and the wellbore (12) and being configured to regulate a downhole pressure in the wellbore using pressure regulation of the circulated fluid, the method comprising:
  • identifying (308), with a control system (200), between a non-drilling state and a drilling state of the drilling system (10), the non-drilling state being indicative of the drillstring (14) being held at the drilling vessel (100) at least during a vessel motion, the drillstring (14) held during the vessel motion being expected to produce a first piston effect that changes the downhole pressure (BHP) in the wellbore (12), the drilling state being indicative of the drillstring (14) being tripped in the wellbore (12);
  • determining (310), with the control system (200) in response to an identification of the non-drilling state, a first adjustment to the pressure regulation provided by the drilling system (10), the first adjustment being configured to maintain the downhole pressure (BHP) in the wellbore; and
  • counteracting (314) the change to the downhole pressure (BHP) produced by the first piston effect by automatically adjusting, with the control system (200), the pressure regulation provided by the drilling system according to the first adjustment.
  • 2. The method of clause 1, wherein to determine the first adjustment, the method comprises:
  • obtaining (304) one or more environmental measurements indicative of an influence from the ocean environment on the drilling vessel (100);
  • predicting predicted movement of the drilling vessel (100) by motion modelling (302) how the drilling vessel (100) responds to the one or more environmental measurements; and
  • determining (310) the first adjustment based on the predicted movement.
  • 3. The method of clause 2, wherein predicting the predicted movement of the drilling vessel by motion modelling comprises using transfer function calculations that define how one or more vessel characteristics for the drilling vessel respond to one or more environmental characteristics for the one or more environmental measurements.
  • 4. The method of clause 2 or 3, wherein obtaining (304) the one or more environmental measurements indicative of the influence from the ocean environment comprises at least one of:
  • detecting the one or more environmental measurements using one or more environmental sensors (262) on the drilling vessel (100);
  • detecting the one or more environmental measurements using one or more second environmental sensors deployed in the ocean environment in a vicinity of the drilling vessel (100); receiving the one or more environmental measurements from one or more external sources;
  • interrogating an environmental system (260) of the drilling vessel (100) with a request for the one or more environmental measurements;
  • receiving the one or more environmental measurements automatically from the environmental system (260) of the drilling vessel (100); and
  • receiving one or more manual inputs of the one or more environmental measurements; and optionally wherein the one or more environmental measurements from the ocean environment comprises one or more of: met-ocean data, a meteorological condition, an atmospheric aspect, a speed of wind, a direction of wind, a temperature, a humidity, precipitation, an oceanographic condition, a wave height, a wave period, a surface ocean current, a subsurface ocean current, and a sea surface condition.
  • 5. The method of clause 2, 3 or 4, wherein predicting the predicted movement of the drilling vessel by motion modelling further comprises:
  • obtaining one or more vessel measurements indicative of a measured movement of the drilling vessel (100) in the ocean environment; and
  • supplementing the predicted movement of the drilling vessel (100) by using the measured movement of the drilling vessel (100) in the motion modelling.
  • 6. The method of claim 5, wherein:
  • obtaining the one or more environmental measurements comprises measuring one or more wave characteristics using one or more environmental sensors deployed in the ocean environment in a vicinity of the drilling vessel;
  • obtaining the one or more vessel measurements comprises measuring peripheral motion for the measured movement of the drilling vessel (100) using one or more first vessel sensors disposed at a peripheral portion of the drilling vessel (100), measuring central motion for the measured movement of the drilling vessel (100) using one or more second vessel sensors disposed at a central portion of the drilling vessel (100), and calculating a difference between the peripheral motion and the central motion; and
  • supplementing the predicted movement of the drilling vessel (100) in the modelling by determining, based on the calculated difference, an influence of the one or more measured wave characteristics on the predicted movement of the drilling vessel (100) in the motion modelling.
  • 7. The method of any preceding clause, wherein the method comprises using a machine learning model to at least one of: identify between the non-drilling state and the drilling state, predict the predictive movement, and to determine the first adjustment.
  • 8. The method of preceding clause, wherein identifying the non-drilling state indicative of the drillstring being held at the drilling vessel comprises: identifying a first instance of the drillstring being moved out of the wellbore during first of the vessel motion expected to produce swabbing as the first piston effect decreasing the downhole pressure in the wellbore; and identifying a second instance of the drillstring being moved into the wellbore during second of the vessel motion expected to produce surging as the first piston effect increasing the downhole pressure in the wellbore.
  • 9. The method of clause 8, wherein automatically adjusting, with the control system, the pressure regulation provided by the drilling system according to the first adjustment comprises:
  • automatically adjusting, with the control system, the pressure regulation provided by the drilling system according to a downhole pressure increase for the first adjustment in response to the first instance to counteract the change to the downhole pressure produced by the swapping; and
  • automatically adjusting, with the control system, the pressure regulation provided by the drilling system according to a downhole pressure decrease for the first adjustment in response to the second instance to counteract the change to the downhole pressure produced by the surging.
  • 10. The method of any preceding clause, wherein identifying (308) the drilling state comprises identifying a trip of the drillstring being expected to produce a second piston effect that changes the downhole pressure (BHP) of the fluid in the wellbore (12); and wherein the method further comprises:
  • determining (312), with the control system (200) in response to an identification of the drilling state, a second adjustment to a surface backpressure (SBP) for the pressure regulation provided by the drilling system (10) to maintain the downhole pressure (BHP) in the wellbore; and
  • counteracting (314) the change in the downhole pressure (BHP) produced by the second piston effect by automatically adjusting, with the control system (200), the surface backpressure (SBP) according to the second adjustment.
  • 11. The method of any preceding clause, wherein identifying (308) the drilling state comprises identifying a trip expected to produce a second piston effect that changes the downhole pressure (BHP) in the wellbore (12); wherein the method further comprises mitigating the downhole pressure (BHP) change produced by the second piston effect by operating a crown block heave compensator mounted on the drilling system (10).
  • 12. The method of any preceding clause, wherein to determine the first adjustment configured to maintain the downhole pressure (BHP) in the wellbore (12), the method comprises obtaining (306), with the control system (200), a value of the downhole pressure (BHP) of the fluid in the wellbore (12) by obtaining one or more pressure measurements of the drilling system; and calculating a bottom hole pressure (BHP) in the wellbore (12) based at least on the one or more pressure measurements.
  • 13. The method of any preceding clause, wherein determining (310) the first adjustment comprises determining a target of the downhole pressure (BHP) at a depth in the wellbore (12) as being at least less than one of: (i) a fracture pressure gradient of the formation (F), and (ii) a pore pressure gradient of the formation (F); and
  • setting the first adjustment to effectuate the target for the downhole pressure.
  • 14. The method of any preceding clause, wherein adjusting, with the control system (200), the pressure regulation of the wellbore comprises at least one of:
  • operating, with the control system, a continuous flow apparatus connected to the drillstring to communicate fluid for the pressure regulation into the wellbore through the drilling string in the non-drilling state and operating, with the control system, at least one choke in fluid communication with flow out of the wellbore in the closed loop to adjust a surface backpressure;
  • operating, with the control system, at least one choke and a mud pump of the drilling system to pump fluid for the pressure regulation into the riser via a booster line connected to the riser above the wellhead;
  • operating, with the control system, a booster line connected to the riser above the wellhead to provide a column of fluid for the pressure regulation into the riser; and
  • operating, with the control system, a subsea pump and a subsea choke to pump fluid for the pressure regulation into the riser via a booster line connected to the riser above the wellhead.
  • 15. A programmable storage device having program instructions stored thereon for causing a programmable control device (210) to perform a method of drilling a wellbore with drilling fluid using a drilling system (10) according to any one of clauses 1 to 14.
  • 16. A computerized system (200) used with a drilling system (10) on a drilling vessel (100) in an ocean environment to drill a wellbore (12) in a formation (F), the drilling system (10) circulating fluid in a closed loop between a drillstring (14) and the wellbore (12), the computerized system comprising a programmable control device (210) being configured to perform a method according to any one of clauses 1 to 14.

The foregoing description of preferred and other embodiments is not intended to limit or restrict the scope or applicability of the inventive concepts conceived of by the Applicants. It will be appreciated with the benefit of the present disclosure that features described above in accordance with any embodiment or aspect of the disclosed subject matter can be utilized, either alone or in combination, with any other described feature, in any other embodiment or aspect of the disclosed subject matter.

In exchange for disclosing the inventive concepts contained herein, the Applicants desire all patent rights afforded by the appended claims. Therefore, it is intended that the appended claims include all modifications and alterations to the full extent that they come within the scope of the following claims or the equivalents thereof.