System and method for detecting abnormal flowback

Abstract

A method of analyzing flowback of a downhole system includes generating active flowback data by monitoring active flowback from a wellbore. Historical flowback data from historical flowback of a wellbore is used to determine a flowback cluster. A flowback cluster is selected based on comparing the active flowback data to the historical flowback data and determining one or more data instances of the historical flowback data having features similar to those of the active flowback data. Based on the flowback cluster, one or more thresholds may be determined to generate an alert when the active flowback data exceeds the threshold.

Description

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application claims priority to and the benefit of U.S. Provisional Application No. 63 / 367,346, filed on June 30, 2022, which is hereby incorporated by reference in its entirety. Background Art

[0003] Wellbores can be drilled into surface locations or seabeds for various exploration or production purposes. For example, wellbores can be drilled to obtain fluids (such as liquid and gaseous hydrocarbons) stored in underground formations and to extract fluids from the formations. Wellbores for producing or extracting fluids can be formed in formations using earth-boring tools (such as a drill bit for drilling the wellbore and an expander for expanding the diameter of the wellbore).

[0004] During downhole drilling activities, flowback may occur in the downhole system, or the return of fluids from the wellbore to the surface. Monitoring flowback may be critical in order to assess the productivity of the well, identify potential problems, and ensure safe and efficient drilling operations. In some cases, flowback monitoring may be difficult, inaccurate, time-consuming, delayed, and / or prone to human error and subjectivity. Therefore, an automated system that can provide real-time monitoring and analysis of active flowback and accurate characterization and interpretation of flowback patterns may be advantageous. Summary of the invention

[0005] In some embodiments, a method for analyzing flowback of a downhole system includes generating active flowback data by monitoring active flowback from a wellbore and identifying historical flowback data of historical flowback from the wellbore. The method includes determining a flowback cluster based on comparing the active flowback data with the historical flowback data. The method includes generating a flowback threshold based on the flowback cluster. The method includes generating an alarm based on the active flowback data exceeding the flowback threshold.

[0006] In some embodiments, a method of analyzing flowback of a downhole system includes generating active flowback data by monitoring active flowback from a wellbore and identifying historical flowback data of historical flowback from the wellbore. The method includes determining a flowback cluster based on comparing the active flowback data with the historical flowback data. The method includes predicting the active flowback based on the flowback cluster.

[0007] In some embodiments, a system includes: at least one processor; a memory in electronic communication with the at least one processor; and instructions stored in the memory, the instructions executable by the at least one processor to generate active flowback data by monitoring active flowback from a wellbore and identifying historical flowback data of historical flowback from the wellbore. The memory includes instructions for determining a flowback cluster based on comparing the active flowback data to the historical flowback data. The memory includes instructions for generating a flowback threshold based on the flowback cluster. The memory includes instructions for generating an alarm based on the active flowback data exceeding the flowback threshold.

[0008] This summary is provided to introduce a series of concepts that will be further described in the detailed description. This summary is not intended to identify key or essential features of the claimed subject matter, nor is it intended to be used as an aid to limit the scope of the claimed subject matter. Additional features and aspects of the embodiments of the present disclosure will be set forth herein and will be apparent in part from the description or may be learned through the practice of such embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] In order to describe the manner in which the above and other features of the present disclosure can be obtained, a more specific description will be presented by reference to the specific embodiments of the present disclosure shown in the accompanying drawings. For better understanding, the same elements have been represented by the same reference numerals throughout the various drawings. Although some of the drawings may be schematic or exaggerated representations of concepts, at least some of the drawings may be drawn to scale. It should be understood that the drawings depict some exemplary embodiments, and the embodiments will be described and explained in more detail and detail through the use of the drawings, in which:

[0010] Figure 1 is a schematic diagram of a downhole system according to at least one embodiment of the present disclosure;

[0011] Figure 2 An exemplary computing device having a flowback monitoring system implemented thereon is shown in accordance with at least one embodiment of the present disclosure;

[0012] Figure 3 It shows that there is implemented Figure 2 A computing device for a flowback monitoring system;

[0013] Figure 4 An exemplary implementation of determining flowback data clusters according to at least one embodiment of the present disclosure is shown;

[0014] Figure 5 shows a best fit curve and one or more threshold values ​​according to at least one embodiment of the present disclosure;

[0015] Figure 6 An example of monitoring active flowback in combination with a best fit curve and one or more thresholds according to at least one embodiment of the present disclosure is shown;

[0016] Figure 7 An exemplary implementation of an alert associated with active flowback in accordance with at least one embodiment of the present disclosure is shown;

[0017] Figure 8 A flow chart showing a method or series of actions for analyzing flowback of a downhole system described herein, in accordance with at least one embodiment of the present disclosure;

[0018] Fig. 9 A flow chart illustrating a method or series of actions for analyzing flowback of a downhole system described herein, in accordance with at least one embodiment of the present disclosure; and

[0019] Fig.10 Some components that may be included in a computer system are shown. DETAILED DESCRIPTION

[0020] Embodiments of the present disclosure relate to systems, methods, and computer-readable media for detecting abnormal flowback of a wellbore. Specifically, the present disclosure relates to monitoring several downhole parameters associated with a pump shut-down period or a period when drilling fluid is not actively pumped into the wellbore. For example, one or more prior conditions, such as the flow rate of drilling fluid into the wellbore, the depth of the downhole tool, the pressure of the drilling fluid, the flow pattern of the drilling fluid, etc., can be monitored (e.g., periodically or continuously). When the drilling fluid pump is shut down, the flowback of drilling fluid upward and / or out of the wellbore, as well as the prior conditions before or before the pump shut-down period, can be measured, monitored, and recorded. Based on the prior conditions (or any other relevant measurements), the active flowback can be compared with one or more previous flowback instances to characterize and explain the flowback. For example, one or more thresholds can be established based on historical instances of flowback to determine whether the active flowback is higher or lower than expected. In this way, the technology of the present disclosure can be useful for determining abnormal flowback.

[0021] As will be discussed in further detail below, the present disclosure includes multiple practical applications having the features described herein that provide benefits and / or solve problems associated with monitoring and detecting abnormal flowback in a downhole system. Some exemplary benefits are discussed herein in conjunction with various features and functions provided by a flowback monitoring system implemented on one or more computing devices. It should be understood that the benefits explicitly discussed in conjunction with one or more embodiments described herein are provided by way of example and are not intended to be an exhaustive list of all possible benefits of a flowback monitoring system.

[0022] For example, flowback in downhole systems is often monitored and interpreted by skilled operators through a combination of manual observation and data analysis. In some cases, flowback monitoring often includes personnel visually inspecting the flowback process to look for visual clues in the discharged drilling fluid. In addition, field measurements and data logging can measure and record downhole parameters associated with flowback, and operators can retrospectively manually review the data to identify trends and / or compare the data to historical data and thresholds to detect irregularities. Therefore, traditional flowback monitoring technologies can be delayed, time-consuming, prone to human error, and may lack the ability to detect abnormal flowback in real time to make informed, timely decisions.

[0023] In contrast, the flowback monitoring system described in this article provides significant improvements over traditional methods. For example, by leveraging real-time data analysis and continuous monitoring capabilities, the flowback monitoring system can automatically collect and process data from multiple sensors to provide a more comprehensive and accurate assessment of wellbore conditions associated with flowback, compared to manual observation and periodic sampling. In addition, by monitoring real-time data, the system can promptly detect abnormal flowback events in near real time, rather than delayed or after-the-fact detection as with traditional technologies. This early detection enables operators to take corrective actions in a timely manner, which helps minimize risks, reduce downtime and improve operational efficiency. In addition, the flowback monitoring system eliminates the limitations of human error, fatigue, and subjectivity associated with traditional operator-based monitoring, ensuring consistent and reliable flowback interpretation and decision-making.

[0024] In addition to general automatic monitoring of flowback, the flowback monitoring system detects abnormal flowback and provides alerts related to abnormal flowback based on thresholds established from historical flowback data. Analysis and comparison of historical flowback data can provide valuable insights and comparisons into patterns and trends in the wellbore. By comparing real-time data to automatically updated thresholds and / or patterns determined based on historical data, the flowback monitoring system can distinguish between expected changes and abnormal flowback events. By utilizing real-time monitoring with thresholds based on historical analysis of wellbore measurements, the flowback monitoring system facilitates more informed decisions for effective and timely flowback management.

[0025] In addition, the flowback monitoring system improves the accuracy of abnormal flowback detection by selectively comparing historical flowback data with the flowback pattern of active observation. For example, traditional methods can use data logging and trend analysis to implement predefined thresholds and compare the flowback data with the established historical patterns in the flowback. However, these technologies do not take into account the specific wellbore conditions associated with the measurement results of historical and active flowback. By monitoring one or more previous conditions of the wellbore and associating those previous conditions with the flowback measurement results, the flowback monitoring system compares the active flowback with the flowback cluster of the historical flowback with similar previous wellbore conditions. For example, the flowback monitoring system can select one or more instances of the historical flowback of the flowback cluster based on instances with previous flow velocity measurements and / or depth measurements (these instances are within the threshold range of the measurement results of the active flowback). The flowback monitoring system can generate a best fit curve that approximates the behavior of the flowback cluster. In this way, the flowback monitoring system can more accurately predict the expected behavior of the current active flowback. In addition, the flowback monitoring system calculates and / or generates a probability distribution based on the flowback clusters to better characterize and explain the deviation of the active flowback from the predicted flowback behavior of the best fit curve. In this way, the flowback monitoring system can determine abnormal flowback with higher confidence.

[0026] Additional details regarding the systems described herein will now be provided with respect to illustrative drawings that depict exemplary embodiments. For example, Figure 1 An example of a downhole system 100 for drilling into a surface formation 101 to form a wellbore 102 is shown. The downhole system 100 includes a drilling rig 103 for rotating a drill assembly 104 extending downwardly into the wellbore 102. The drill assembly 104 may include a drill string 105, a bottom hole assembly (BHA) 106, and a drill bit 110 attached to the downhole end of the drill string 105.

[0027] The drill string 105 may include several joints of drill pipe 108 connected end to end by tool joints 109. The drill string 105 transmits drilling fluid through a center hole and transmits rotational power from the drilling rig 103 to the BHA 106. In some embodiments, the drill string 105 also includes additional components, such as pup joints, pup joints, etc. The drill pipe 108 provides hydraulic passages through which drilling fluid is pumped from the surface. The drilling fluid is discharged through nozzles, jets, or other orifices in the drill bit 110 for cooling the drill bit 110 and the cutting structure thereon, and for conveying drill cuttings out of the wellbore 102 as it is drilled.

[0028] The BHA 106 may include a drill bit 110 or other components. An exemplary BHA 106 may include additional or other components (e.g., coupled between the drill string 105 and the drill bit 110). Examples of additional BHA components include drill collars, stabilizers, measurement while drilling (MWD) tools, logging while drilling (LWD) tools, downhole motors, underreamers, slicers, hydraulic breakers, cans, vibration or shock absorbing tools, other components, or combinations of the foregoing components. The BHA 106 may also include a rotary steerable system (RSS). The RSS may include a directional drilling tool that changes the direction of the drill bit 110, thereby changing the trajectory of the wellbore 102. In some cases, at least a portion of the RSS maintains a geostationary position relative to an absolute reference system (such as gravity, magnetic north, or true north). Using measurements obtained through the geostationary position, the RSS can position the drill bit 110, change the course of the drill bit 110, and guide the directional drilling tool on a predicted trajectory.

[0029] In general, the downhole system 100 may include additional or other drilling components and accessories, such as specialized valves (e.g., kelly cocks, blowout preventers, and safety valves). The additional components included in the downhole system 100 may be considered part of the drill string 104, the drill string 105, or part of the BHA 106, depending on their location in the downhole system 100.

[0030] In some embodiments, a downhole motor in the BHA 106 generates power for the downhole system and / or provides rotational energy to downhole components (e.g., to rotate the drill bit 110). The downhole motor can be any type of downhole motor, including a positive displacement pump (such as a progressive cavity motor) or a turbine. In some embodiments, the downhole motor is powered by drilling fluid. In other words, drilling fluid pumped downhole from the surface can provide energy to rotate the rotor in the downhole motor. The downhole motor can operate at an optimal pressure differential or range of pressure differentials. The optimal pressure differential can be a pressure differential at which the downhole motor does not stall, burn out, over-rotate, or suffer other damage. In some cases, the downhole motor drives the rotation of the drill bit. In some embodiments, the rotation of the drill bit is driven by components at the surface of the wellbore 102.

[0031] The drill bit 110 in the BHA 106 can be any type of drill bit suitable for degrading downhole materials. For example, the drill bit 110 can be a drill bit suitable for drilling the surface formation 101. Exemplary types of drill bits for drilling surface formations are fixed cutter or drag bits and roller cone drill bits. In other embodiments, the drill bit 110 can be a milling shoe for removing metals, composites, elastomers, other downhole materials, or combinations thereof. For example, the drill bit 110 can be used with a whipstock to grind into a casing 107 sleeved on the wellbore 102. The drill bit 110 can also be a flat-head milling shoe for milling out tools, plugs, cement, other materials, or combinations thereof within the wellbore 102. The cuttings or other drill cuttings formed by using the milling shoe can be transported to the surface, or can be allowed to fall downhole. In other embodiments, the drill bit 110 can include a reamer. For example, a reamer can be used in conjunction with the drill bit, and the drill bit can drill into the formation, while the reamer can expand the size of the borehole.

[0032] As described above, the downhole system 100 may include or may implement a drilling fluid or drilling mud. In some embodiments, the downhole system 100 includes a mud tank 111. The mud tank 111 may be a reservoir for holding a certain amount of drilling fluid. The drilling fluid may be pumped from the mud tank 111, through one or more fluid lines, and down the drill tool assembly 104 to the drill bit 110. As described above, the drilling fluid may flow out of the drill bit, for example at the bottom of the wellbore 102, in order to cool the drill bit 110 and / or remove and carry away drill cuttings.

[0033] In some embodiments, the drilling fluid is implemented in this manner and at least partially fills the wellbore 102. For example, the drilling fluid can be pumped through the drill assembly and out the drill bit 110 so that the wellbore can be filled, and the drilling fluid can flow out of the annular region of the wellbore around the drill string 105. In this manner, the drilling fluid can flow upward through the annular region of the wellbore 102 to carry formation cuttings and heat away from the drill bit 110. In this manner, the drilling fluid can be circulated from the mud tank 111, downward through the drill assembly 104, back to the wellbore 102, and back to the mud tank 111.

[0034] During drilling operations, it may be necessary to periodically stop or pause pumping drilling fluid into the wellbore 102. For example, additional lengths of drill pipe 108 may be periodically added to the drill tool assembly 104 as the drill bit 110 is advanced downward through the surface formation 101, and the pump supplying drilling fluid to the drill tool assembly 104 may be shut down. In many cases, during the pump shut-off period, a certain amount of fluid (e.g., drilling fluid and / or formation fluid) may flow back and / or out of the wellbore 102 (e.g., the annular region). For example, pumping drilling fluid into the wellbore 102 may create a pressure imbalance in the wellbore, which may offset the formation pressure and the inflow of formation fluid (such as oil, gas, or water). Shutting down the pump may release the pressure imbalance, and the formation pressure may cause the drilling fluid and / or formation fluid to flow upward and out of the wellbore 102. Additionally, in many cases, the natural gas may become dissolved in the drilling fluid and when the pressure of the drilling fluid is released, the natural gas may expand, causing the volume to increase and the drilling fluid to flow up and out of the wellbore 102. The flow of fluid up and out of the wellbore may also be caused by a variety of other reasons.

[0035] After the pump is shut off, the flow of fluid from the wellbore 102 is called flowback. Measures can usually be taken to control, predict and / or anticipate a certain volume of flowback and to safely recover and dispose of the fluid. In some embodiments, an excess of flowback (e.g., higher than expected) indicates a well kick or formation kick, or formation fluid flows from the formation into or into the wellbore. Detecting and responding to formation kicks is essential for maintaining well control and preventing more serious accidents such as blowouts, thereby ensuring the safe operation of the downhole system 100. In some embodiments, a reduction in flowback (e.g., lower than expected) indicates formation loss, or drilling fluid is lost (e.g., accidentally) into the formation. Detecting and responding to formation loss may be essential for maintaining the integrity of the well, preventing damage to the surrounding formations, and maintaining sufficient hydrostatic pressure in the wellbore 102 to control the well and prevent formation inflow. Therefore, the interpretation and analysis of the flowback pattern (referred to as flowback fingerprint comparison) can provide favorable information for the effective operation of the downhole system 100.

[0036] However, in some cases, it may be difficult to adequately and / or accurately characterize and / or explain the flowback patterns observed in the downhole system 100. For example, many factors may affect the amount and extent of flowback after a pump shut-off. Therefore, it may be difficult to accurately characterize and / or predict the expected flowback volume to determine whether the flowback volume is excessive or reduced during any given pump shut-off period. Therefore, improved systems and methods for detecting abnormal flowback may have advantages over conventional techniques.

[0037] In some embodiments, the downhole system 100 includes a computing device 112. The computing device 112 may refer to a variety of computing devices, including mobile devices, such as mobile phones, smart phones, laptop computers, or other mobile devices. The computing device 112 may refer to a non-mobile device, such as a desktop computer, a server device, or other non-portable devices. The computing device 112 may be a surface computing device, a mobile computing device, a remote computing device, a downhole computing device, and combinations thereof. The computing device 112 may be in electronic communication with one or more components of the downhole system 100, such as sensors, electronic components, and other computing devices.

[0038] In some embodiments, the computing device includes a flowback monitoring system 120 implemented thereon. As will be discussed in further detail below, the flowback monitoring system 120 can facilitate monitoring, analyzing, and / or interpreting flowback experienced by the downhole system 100, for example, during one or more pump shut-off periods. In this way, the flowback monitoring system 120 can facilitate more accurately understanding and characterizing one or more downhole conditions based on observed flowback.

[0039] Figure 2 An exemplary computing device 212 having a flowback monitoring system 220 implemented thereon is shown in accordance with at least one embodiment of the present disclosure. The flowback monitoring system 220 may include a sensor manager 222, a data manager 224, a data clustering engine 226, a flowback threshold engine 228, and a flowback alarm manager 230. Additionally, the flowback monitoring system 220 may include a data storage device 232 that includes active flowback data 234, historical flowback data 236, and threshold data 238. While one or more embodiments described herein describe features and functions performed by specific components 222-230 of the flowback monitoring system 220, it should be understood that specific features described in conjunction with one or more components of the flowback monitoring system 220 may be performed by one or more other components of the flowback monitoring system 220 in some examples.

[0040] By way of example, as will be discussed below, one or more features of the data manager 224 may be delegated to other components of the flowback monitoring system 220. As another example, the data clustering engine 226 may select and / or process one or more instances of data, and in some instances, some or all of these features may be performed by the flowback threshold engine 228 (or other components of the flowback monitoring system 220). In fact, it should be understood that some or all of the specific components may be combined into other components, and that specific functions may be performed by one or more components 222-230 of the flowback monitoring system 220.

[0041] As just mentioned, the flowback monitoring system 220 includes a sensor manager 222. The sensor manager 222 can communicate data with one or more sensors and can monitor and / or manage measurements and / or signals received from the sensors. In this way, the sensor manager 222 can measure and / or monitor one or more parameters of the downhole system.

[0042] like Figure 3 As shown, the backflow monitoring system 220 may be connected to the sensor 240 and / or may communicate with the sensor. The sensor 240 may include one or more sensors implemented in the downhole system or implemented in conjunction with the downhole system. In some embodiments, the sensor 240 includes a pump on / off sensor 242. As discussed above, the downhole system can pump drilling fluid into the wellbore via the drill tool assembly. In some cases, the pump can be shut down, such as to add or replace one or more sections of drill pipe, or to stop, adjust or change the operation of the downhole system. The pump on / off sensor 242 can monitor the state of the pump and can transmit the state or condition of the pump to the sensor manager 222. In this way, the backflow monitoring system 220 can monitor the current operating state of the pump, for example, in order to detect a pump shutdown period.

[0043] In some embodiments, sensor 240 includes flow rate sensor 244. Flow rate sensor 244 can measure or detect the flow rate of drilling fluid supplied to the wellbore. For example, flow rate sensor 244 can measure the flow rate of drilling fluid entering the drill tool assembly. Flow rate sensor 244 can measure or detect the flow rate of drilling fluid, such as the flow rate at the surface of the wellbore. Flow rate sensor 244 can measure the flow rate of drilling fluid at the pump. Flow rate sensor 244 can measure the flow rate of drilling fluid at one or more locations in the drill tool assembly (such as at the drill bit). As will be discussed in further detail below, the flow rate before the pump shutdown period can be a previous condition associated with determining abnormal backflow during the pump shutdown period. Therefore, measuring the flow rate in this way can be beneficial to characterize observed or active backflow. To this end, in some embodiments, flow rate sensor 244 constantly monitors the flow rate so as to determine the previous flow rate when the pump shutdown period is identified. In some embodiments, flow rate sensor 244 regularly measures the flow rate, or is activated to measure the flow rate before the pump is shut down. In this manner, the flowback monitoring system 220 may monitor and / or measure the flow rate of drilling fluid supplied to the wellbore prior to the pump shut-off period.

[0044] In some embodiments, the sensor 240 includes a flow rate sensor 246. The flowback sensor 246 can measure the flow rate (e.g., the amount of flowback from the wellbore) of the fluid (e.g., drilling fluid and / or formation fluid) flowing or flowing out of the wellbore. The flowback sensor 246 can measure the pattern of the flowback, such as the rate and / or total amount of the fluid flowing out of the wellbore. The flowback sensor 246 can measure the rate and / or total amount of the fluid flowing into the mud tank and / or can measure the level or amount of the fluid in the mud tank (e.g., effective volume). In some embodiments, the flowback sensor 246 continuously monitors the flowback, such as during part or all of the drilling operation. In some embodiments, the flowback sensor 246 periodically monitors the flowback during part or all of the drilling operation. In some embodiments, the flowback sensor 246 is activated during the pump off period so that the flowback sensor 246 monitors and / or measures the flowback only when the pump is off. In this way, the flowback monitoring system 220 can monitor and / or measure the flowback of the fluid flowing or flowing out of the wellbore.

[0045] In some embodiments, one or more of the flow rate sensor 244 and the flowback sensor 246, alone or in combination, monitor and / or measure (or facilitate monitoring and / or measuring) a flow pattern of the drilling fluid. For example, the movement and distribution of the drilling fluid within the wellbore may vary based on factors such as drilling technology, wellbore conditions, and drilling fluid properties. Therefore, the sensor manager 222 can monitor and / or measure the flow pattern to facilitate characterizing the flowback, as described herein.

[0046] In some embodiments, the sensor 240 includes a drill bit depth sensor 248. The drill bit depth sensor 248 can monitor and / or measure the depth of one or more components of the downhole system, such as one or more downhole tools. For example, the drill bit depth sensor 248 can measure the depth of a drill bit used to extend and / or widen a wellbore. The measurements of the drill bit depth sensor 248 can be performed in association with a pump shut-in period. For example, the depth of the downhole tool may be related to interpreting the associated backflow (e.g., measured at or near the surface), and the depth of the downhole tool can be measured before and / or during the associated pump shut-in period. In this way, the backflow monitoring system 220 can monitor one or more depths associated with downhole operations.

[0047] In some embodiments, the sensor 240 includes a pressure sensor 250. The pressure sensor 250 can measure the fluid pressure of the drilling fluid at one or more locations in the downhole system. For example, the pressure sensor 250 can measure the drilling fluid pressure at the pump, at the riser, at one or more locations in the drill tool assembly, at the drill bit, at one or more locations in the wellbore (e.g., an annular area), and combinations thereof. The pressure sensor 250 can measure the drilling fluid pressure continuously, periodically, and / or can be triggered or activated in response to a pump shut-off period (e.g., measuring the pressure before the pump shut-off period). In this way, the backflow monitoring system 220 can monitor the drilling fluid pressure at one or more locations in the downhole system.

[0048] It should be understood that the downhole system and the flowback monitoring system 220 are not limited to including a combination of Figure 3 Those sensors shown and discussed, but the sensor may include any other sensor (or one or more sensors may be omitted) and / or the sensor manager 222 may also be connected to one or more sensors other than the sensors discussed herein and / or communicate with the one or more sensors. In fact, the sensor 240 may include any sensor for monitoring and / or measuring one or more parameters (e.g., previous conditions) related to the monitoring, measuring and / or characterizing the backflow described herein. In addition, the various sensors described herein may be implemented as one or more physical sensing devices, one or more calculations based on signals from the physical sensing devices, and combinations thereof. For example, a pump on / off sensor may be a physical sensor that detects the active state of the pump, or may be a calculation based on another signal that the pump is turned on or off, such as a flow rate sensor. In this way, the term "sensor" should be understood to cover devices for physically sensing conditions or measurements, as well as software implementations for calculations associated with conditions or measurements (and combinations thereof).

[0049] As mentioned above, and as Figure 2As shown, the flowback monitoring system 220 may include a data manager 224. The data manager 224 may receive data or information corresponding to one or more sensor measurements (e.g., from the sensor manager 222). The data manager 224 may save and / or store the sensor measurements to a data storage device 232 (e.g., as active flowback data 234 and / or historical flowback data 236 discussed below). In some embodiments, the data manager 224 associates one or more types and / or instances of sensor measurements with each other. For example, the data manager 224 may receive flowback measurements and may map or associate the flowback measurements with flow rate measurements, flow pattern measurements, depth measurements, pressure measurements, any other measurements acquired and / or accessible by the data manager 224, and combinations thereof. According to at least one embodiment of the present disclosure, the data manager 224 associates the flowback measurements with (e.g., previous) flow rate measurements.

[0050] In some embodiments, the data manager 224 receives pump on / off measurements corresponding to pump on or off periods from the sensor manager 222. The data manager 224 can save and / or store one or more of the sensor measurements based on receiving the indication of the pump off period. The data manager 224 can associate or map two or more measurements and / or instances of sensor measurements based on receiving the indication of the pump off period.

[0051] In some embodiments, at least some of the sensor measurements are associated with a prior condition of the downhole system and / or the wellbore. For example, the sensor manager 222 may monitor and / or measure one or more aspects of the downhole system prior to a pump shut-off period. Upon receiving an indication of a pump shut-off period, the data manager 224 may save and / or store one or more measurements and / or instances of sensor measurements prior to the pump shut-off period as prior conditions of the wellbore associated with the pump shut-off period. For example, the data manager 224 may receive a pump shut-off indication and may call flow rate, pressure, depth, flow pattern, or any other measurement (and combination thereof) and may save these measurements as prior conditions for the pump shut-off period. The prior condition may be one or more measurements taken within 10 seconds, 30 seconds, 1 minute, 2 minutes, 4 minutes, 5 minutes, or 10 minutes (or any value therebetween) after the start of the pump shut-off period. In some embodiments, it is critical that each prior condition is measured within 2 minutes of the start of the pump off period so that the prior condition measurements accurately reflect the state and / or behavior of the well associated with the flowback measurements taken during the pump off period (e.g., taken after the prior condition). Each prior condition can be calculated or associated with the average, minimum, maximum, mean, median, mode, range, or any other value or range of values ​​of the prior condition measurements. In this way, flowback measurements taken during the pump off period can be associated with one or more prior conditions that indicate observable well conditions when the pump is shut off.

[0052] In some embodiments, saving the prior conditions is associated with monitoring and / or measuring and / or storing (e.g., active) flowback measurements after an indication of a pump-off period initiation. In this way, flowback measurements can be captured (e.g., during a pump-off period) and one or more prior conditions of the downhole system and / or wellbore can be saved and / or associated with measurements of active (e.g., pump-off) flowback. As will be discussed below, mapping and / or associating flowback to prior conditions of the wellbore can be beneficial in characterizing and / or interpreting flowback measurements.

[0053] As described above, the data storage device 232 may include active flowback data 234 and historical flowback data 236. In some embodiments, the data manager 224 receives sensor measurements and stores them as active flowback data 234. For example, the sensor manager 222 may receive a pump shut-off indication and may begin recording and / or saving flowback measurements as active flowback data 234, corresponding to active flowback of the downhole system. The data manager 224 may call and save prior conditions associated with active flowback and / or active pump shut-off periods as (e.g., a portion of) the active flowback data 234. In this way, the active flowback data 234 may include flowback measurements and prior conditions associated with and / or based on active pump shut-off periods of the downhole system.

[0054] In some embodiments, the data manager 224 stores and / or caches some or all of the active flowback data 234 as historical flowback data 236. The data manager 224 may cache the active flowback data 234 based on one or more trigger conditions, such as a period of time that has passed since the pump was turned off, the pump was turned back on, and / or the flowback stopped or became identified as stable (e.g., non-abnormal). For example, the data manager 224 may record and / or store the active flowback data 234 during part or all of the active pump off period. Based on receiving an indication that the pump was turned back on (e.g., the active pump off period ended), the data manager 224 may migrate the active flowback data 234 for the (now past) pump off period and save the data as historical flowback data 236. In some embodiments, the data manager 224 may clear the active flowback data 234 in preparation for measuring and / or storing (e.g., new) active flowback data corresponding to the next pump off period. The data manager 224 may begin recording and / or storing additional (e.g., new) sensor measurements as active flowback data 234. In this manner, the active flowback data 234 may include and / or may be associated with measurements corresponding only to an active (or upcoming) pump-off period, and the historical flowback data 236 may include measurements associated with one or more previous pump-off periods (e.g., previous instances of the active flowback data 234). In this manner, as discussed below, the active flowback data 234 may be compared to the historical flowback data 236 to facilitate accurate characterization and interpretation of active flowback of a downhole system.

[0055] In some embodiments, the data manager 224 purges or deletes one or more data instances from the data storage device 232. For example, in some cases, when the historical flowback data 236 does not occur and / or is measured within a particular time frame, it may be determined that the historical flowback data is less relevant and / or less valuable for comparison with the active flowback data 234. Thus, the data manager 224 may purge or delete one or more instances of the historical flowback data 236 based on instances having a threshold age or occurring before a threshold time interval. In some embodiments, the data manager 224 purges data instances older than 1 hour, 2 hours, 3 hours, 4 hours, 5 hours, 6 hours, 7 hours, 8 hours, 9 hours, or 10 hours (or any value therebetween) from the historical flowback data 236 (e.g., occurred more than that long before the active flowback data 234 and / or the active pump off period). In another example, in some cases, when the historical flowback data 236 does not occur during the same downhole operation, the same type of downhole operation, or the same downhole tool operation, it can be determined that the historical flowback data is less relevant and / or less valuable for comparison with the active flowback data 234. Therefore, the data manager 224 can clear or delete some or all of the historical flowback data 236 based on the downhole tool being tripped from the wellbore and / or based on a change in the downhole operation or type of downhole operation. In this way, the data manager 224 can maintain the data in the data storage device 232 so that the historical flowback data 236 is the most relevant, useful, and / or valuable for facilitating the characterization and / or interpretation of active flowback.

[0056] As mentioned above, and as Figure 2 As shown, the flowback monitoring system 220 may include a data clustering engine 226. The data clustering engine 226 may sort, index, filter, and / or select one or more instances of the historical flowback data 236 in the data storage device 232. For example, the data clustering engine 226 may select a flowback cluster, or one or more instances of the historical flowback data 236 (e.g., corresponding to one or more previous pump shut-in periods), and may save or store the flowback cluster to the data storage device 232 as threshold data 238. The flowback cluster and threshold data 238 may facilitate comparison of one or more previously observed wellbores for flowback to be compared with active flowback (e.g., compared with active flowback data 234) in order to characterize and / or interpret the flowback.

[0057] In some embodiments, the data clustering engine 226 automatically selects a flowback cluster based on a pump on / off period. For example, the data clustering engine 226 may receive an indication that a pump has been turned off and may automatically sort, filter, and / or select one or more instances of the historical flowback data 236 as a flowback cluster and may store the flowback cluster as threshold data 238 to the data storage device 232. The data clustering engine 226 may select a flowback cluster at any other time or based on any other triggering factor.

[0058] In some embodiments, the data clustering engine 226 filters and / or sorts the historical flowback data 236, and can clear, delete, mark and / or isolate one or more instances of the historical flowback data 236 (and / or active flowback data) from the data storage device 232. The data clustering engine 226 can mark, filter or remove data instances based on the data quality of the data. For example, in some cases, one or more measurement results and / or data instances may or may become damaged, incomplete, disjointed, unclear or of poor quality in other aspects. In some cases, the data clustering engine identifies one or more measurement results and / or data instances as inaccurate, uncalibrated, unexpected or outliers. These sensors and / or data and / or measurement errors may be the result of sensor failure, disconnection or damage or any other reason. The data clustering engine can mark, separate, filter and / or delete one or more measurement results and / or data instances of the historical flowback data 236 and / or active flowback data 234 accordingly. In this way, the data clustering engine 226 can maintain the quality of the measurement results and / or data of the flowback monitoring system 220.

[0059] The data clustering engine 226 can select a flowback cluster based on data instances that are well suited for evaluating active flowback and / or more relevant to characterization of active flowback. For example, one or more instances of the historical flowback data 236 can be relevant and / or useful for comparison with the active flowback data 234 when the data instances have one or more characteristics that are similar to characteristics of the active flowback data 234.

[0060] In some embodiments, a flowback cluster is automatically selected based on one or more of the prior conditions associated with the active flowback data 234 and the historical flowback data 236. For example, Figure 4As shown in the diagram of , the data clustering engine 226 can sort, rank, classify, list, or otherwise organize the historical flowback data 236 based on one or more of the previous conditions associated with the historical flowback data 236. According to at least one embodiment of the present disclosure, the data clustering engine 226 organizes the historical flowback data 236 based on previous flow rate measurements associated with the historical flowback data 236 (e.g., the flow rate before each pump shut-off period associated with each data instance). The data clustering engine 226 can organize the historical flowback data 236 based on pressure measurements, depth measurements, flow pattern measurements, any other measurements, and combinations thereof.

[0061] After sorting the historical flowback data 236, the data clustering engine 226 may select one or more data instances of the historical flowback data 236 as a flowback cluster 254. For example, the data clustering engine 226 may select 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10 (or any value therebetween) data instances of the historical flowback data 236 to form a flowback cluster 254. In some embodiments, it is critical that the flowback cluster 254 include three data instances of the historical flowback data 236 so that the flowback threshold and / or envelope (as discussed below) is not too narrow or too wide. This can be beneficial for correctly and accurately characterizing active flowback and preventing false positive and false negative alarms due to incorrect threshold calculation and / or size.

[0062] The data clustering engine 226 can select the flowback cluster 254 based on the similarity of the previous conditions of the historical flowback data 236 to the previous conditions of the active flowback data 234. For example, the flowback cluster 254 can be selected based on the data instance having a previous flow rate within a predefined range compared to the active flowback data 234. The flowback cluster 254 can be selected based on the previous flow rate of the data instance, which is within 10, 20, 40, 50, 100, 150 or 200 gallons per minute (or any value therebetween) of the previous flow rate of the active flowback data 234. For example, the data clustering engine 226 can select all data instances that fall within a given range, or can select a specific number of data instances that are closest to the flow rate of the active flowback data 234 (e.g., while also within a given range). In some embodiments, it is critical that the flow rate of the flowback cluster 254 is within 50 gallons per minute of the active flowback data 234 to ensure that the flowback cluster 254 is sufficiently representative of the wellbore conditions of the active flowback data 234.

[0063] In another example, the flowback cluster 254 may be selected based on data instances having depths similar or close to the active flowback data 234 (e.g., the depth of the downhole tool before and / or during the corresponding pump shut-in period). The flowback cluster 254 may be selected based on the depth of the data instance being within a range of 25m, 50m, 100m, 150m, 200m, 250m, 300m, 400m, or 500m (or any value therebetween) of the depth of the active flowback data 234. For example, the data clustering engine 226 may select all data instances that fall within a given range, or may select a specific number of data instances that are closest to the depth of the active flowback data 234. In some embodiments, it is critical that the depth of the flowback cluster 254 is within 200 meters of the active flowback data 234 to ensure that the flowback cluster 254 is sufficiently representative of the wellbore conditions at or near the depth of the downhole tool during the active pump shut-in period of the active flowback data 234. The data clustering engine 226 can select the flowback cluster 254 based on similarity of any other measurements of previous conditions of the historical flowback data 236 to the measurements of the active flowback data 234 .

[0064] In some embodiments, the data clustering engine 226 selects the flowback cluster 254 based on two or more prior conditions. For example, the flowback cluster 254 may be selected based on one or more instances of the historical flowback data 236 having a flow rate and depth that is similar (e.g., close) to the flow rate and depth of the active flowback data 234. In another example, the flowback cluster 254 may be selected based on one or more instances of the historical flowback data 236 having a pressure and depth that is similar (e.g., close) to the pressure and depth of the active flowback data 234. The flowback cluster 254 may be selected based on one or more instances of the historical flowback data 236 having any other combination of similar prior conditions and / or measurements.

[0065] In some embodiments, the data clustering engine 226 determines that one or more (or all) data instances of the historical flowback data 236 do not have a prior condition (a prior condition among various prior conditions) that is sufficiently similar or close to the condition of the active flowback data 234. Therefore, the data clustering engine 226 may not select one or more data instances for the flowback cluster 254 based on the prior condition, and may continue to compare and / or select one or more data instances for the flowback cluster 254 based on another prior condition or another measurement result. In some embodiments, this selective process is based on a hierarchy or order of relevance of the prior conditions and / or other measurements to the characterization of the active flowback.

[0066] For example, the data clustering engine 226 may sort the historical flowback data 236 based on a first prior condition, such as flow velocity, and may determine that none (or not enough) of the historical flowback data 236 have a previous flow velocity within a threshold of the previous flow velocity of the active flowback data 234. Therefore, the data clustering engine 226 may not select one or more data instances for the flowback cluster 254 based on flow velocity, and may continue to sort the historical flowback data 236 based on a second prior condition, such as depth. The data clustering engine 226 may select one or more data instances for the flowback cluster 254 based on depth measurements of a downhole tool acquired or measured at a depth similar to the active flowback data 234. This process may proceed through any number of prior conditions and / or measurements until the data clustering engine 226 selects the flowback cluster 254 based on sufficient similarity to the active flowback data 234.

[0067] In some embodiments, the data clustering engine 226 selects one or more data instances for the flowback cluster 254 based on a first prior condition or measurement result, and may further refine and / or filter the selected data instances by adding and / or removing one or more data instances to the flowback cluster 254 based on a second prior condition or measurement result.

[0068] In this manner, the data clustering engine 226 can determine the flowback clusters 254 by selecting one or more data instances of the historical flowback data 236 based on data instances having one or more characteristics similar to the characteristics of the active flowback data 234. In this manner, defining the flowback clusters 254 of the historical flowback data 236 based on similarity to the active flowback data 234 can help to more accurately characterize and / or interpret the active flowback data 234.

[0069] In some embodiments, the data clustering engine 226 determines, based on one or more prior conditions, other measurements, or any other basis, that an insufficient number (or no) of data instances of the historical flowback data 236 have sufficient similarity to the active flowback data 234. Therefore, the data clustering engine 226 may not select and / or define a flowback cluster 254. In some embodiments, the data clustering engine 226 indicates to the flowback monitoring system 220 that the flowback cluster 254 is not defined, for example, so that thresholding and / or alerting may not be performed.

[0070] As mentioned above, and as Figure 2 As shown, the flowback monitoring system 220 may include a flowback threshold engine 228. The flowback threshold engine 228 may access and / or process some or all of the data in the data storage device 232 to predict the value, rate, and / or behavior of active flowback.

[0071] In some embodiments, the flowback threshold engine 228 generates or fits a curve for one or more instances of the historical flowback data 236. For example, based on the flowback cluster, the flowback threshold engine 228 can generate a best fit curve representing the flowback cluster. In some embodiments, the flowback threshold engine 228 generates a best fit curve for each data instance of the data cluster. In some embodiments, the flowback threshold engine 228 generates a best fit curve for all data instances of the flowback cluster. The best fit curve can be generated using any curve fitting technique, such as linear, polynomial, logarithmic, trigonometric, power, exponential, moving average, or any other curve fitting technique (and combination thereof). The flowback threshold engine 228 can save or store the best fit curve as the threshold data 238 (or as a portion thereof) to the data storage device 232.

[0072] According to at least one embodiment of the present disclosure, a best fit curve can be generated based on a hyperbolic tangent function. For example, the best fit curve can be fitted based on the following formula:

[0073]

[0074] One or more embodiments may change, enhance, and / or simplify the exemplary formula or one or more portions of the exemplary formula. In some embodiments, a best fit curve based on a hyperbolic tangent function (such as the hyperbolic tangent function shown above) approximates the general shape of the flowback measurements typically observed for flowback from the wellbore. For example, adjusting variable A may adjust the slope of the curve, and adjusting variable B may adjust the asymptote of the curve. In this way, the best fit curve may fit one or more data instances of the flowback cluster so that the best fit curve approximates the trend of the flowback cluster.

[0075] According to at least one embodiment of the present disclosure, Figure 5 As shown, the flowback threshold engine 228 can generate a best fit curve 252 for the flowback cluster 254 (e.g., a curve that is the best fit for all data instances of the flowback cluster). The best fit curve 252 can approximate the behavior of all data instances of the flowback cluster 254. As discussed above, the flowback cluster can be determined or selected based on historical data instances having similar characteristics (e.g., previous conditions) to the active flowback being actively observed. In this way, the best fit curve 252 can approximate and / or predict the behavior of the active flowback and / or the best fit curve 252 can represent a curve or set of values ​​that the active flowback is expected to follow. For example, when a pump off period is detected, the data clustering engine 226 can (e.g., automatically) select the flowback cluster 254, and the flowback threshold engine 228 can (e.g., automatically) generate the best fit curve 252 for the flowback cluster 254. In this way, once a pump off period is detected, the behavior of the active flowback can be (e.g., automatically) predicted.

[0076] In some embodiments, the flowback threshold engine 228 determines and / or generates a probability distribution of the flowback clusters 254. The flowback threshold engine 228 can determine the probability distribution of the flowback clusters 254 relative to, and / or in contrast to, and / or fit to the best fit curve 252. For example, for one or more (or all) points or increments in a domain (e.g., time), the flowback threshold engine 228 can determine the probability distribution of the flowback clusters 254 (e.g., relative to the best fit curve 252) at the corresponding domain points or (e.g., time) increments. In some embodiments, the probability distribution is a Gaussian or normal probability distribution. In some embodiments, the probability distribution is a Bayesian probability distribution. The probability distribution can be any probability distribution and / or can be based on any form or statistical theory, such as a discrete distribution, a continuous distribution, a binomial distribution, a Poisson distribution, a Bernoulli distribution, a hypergeometric distribution, a normal distribution, a uniform distribution, an exponential distribution, a chi-square distribution, a logistic distribution, a Student's T distribution, any other distribution, and combinations thereof. The flowback threshold engine 228 may save or store the probability distribution as threshold data 238 to the data storage device 232 .

[0077] In some embodiments, the probability distribution includes and / or can be based on or relative to the best fit curve 252. For example, the probability distribution can be generated so that the best fit curve 252 is and / or represents one or more of the mean, median, mode, average, quartiles, variance, standard deviation, any other probability or statistical value, and combinations thereof. In this way, the best fit curve 252 can predict the expected, ideal, or smooth behavior of active flowback, and the probability distribution can be useful in defining and / or characterizing any deviation or divergence of the observed active flowback from the prediction of the best fit curve 252.

[0078] In some embodiments, the backflow threshold engine 228 determines and / or generates a backflow threshold 256. The backflow threshold engine 228 can generate the backflow threshold 256 based at least in part on the best fit curve 252 and / or the probability distribution. For example, the probability distribution can define the probability that the backflow measurement result takes its measured value within all possible values ​​within a given time interval. The backflow threshold 256 can be generated based on the percentage of all possible values ​​of each measurement result that are included, captured and / or limited within a threshold probability or percentage (e.g., centered on the prediction) within each time interval. The backflow threshold engine 228 can save or store the backflow threshold 256 as threshold data 238 to the data storage device 232.

[0079] In this way, the flowback threshold engine 228 may generate one or more flowback thresholds 256, which may indicate a value or range of values ​​of the flowback measurement results at a threshold probability or percentage of deviation from the expected value of the flowback measurement results (e.g., the best fit curve). For example, the flowback threshold 256 may represent the amount or degree of acceptable and / or safe deviation of the measured active flowback relative to the best fit curve 252, and the active flowback measurement results exceeding or exceeding one or more flowback thresholds 256 may indicate excessive and / or undesirable active flowback. According to at least one embodiment of the present disclosure, the flowback threshold 256 is based on one or more standard deviations of the probability distribution. For example, the flowback threshold 256 may capture or include all possible flowback measurement results within 1, 2, or 3 standard deviations from the best fit curve 252 or any value (and combination thereof) therebetween. In some embodiments, it is critical that the flowback threshold 256 includes three standard deviations above and / or below the best fit curve 252 so as to facilitate the characterization of the flowback measurement outside the flowback threshold 256 as abnormal with a sufficiently high confidence. For example, the flowback threshold 256 may include an upper limit 256-1 that is three standard deviations above the best fit curve 252, and a lower limit 256-2 that is three standard deviations below the best fit curve 252. In this manner, the best fit curve 252 and / or the probability distribution may be useful for characterizing and / or explaining deviations or divergences of measured (e.g., active) flowback relative to an expected or predicted range of values ​​based on the best fit curve 252.

[0080] As discussed above, in some embodiments, the flowback cluster 254 is not generated and / or selected, such as due to a lack of suitable data instances of historical flowback data. In some embodiments, the flowback threshold engine 228 does not generate one or more of the best fit curve 252, the probability distribution, and the one or more flowback thresholds 256. Not generating one or more of these features can help eliminate false positive and / or false negative indications of abnormal and / or undesirable flowback (as discussed below) that may occur by generating predictions and / or thresholds based on insufficient and / or inappropriate historical data.

[0081] In some embodiments, the backflow threshold engine 228 generates one or more of the best fit curve 252, the probability distribution, and the one or more backflow thresholds 256 based on data other than the backflow cluster 254. (For example, in the case where the backflow cluster 254 is not selected). For example, the backflow threshold engine 228 can generate one or more of these features based on all historical data. In another example, the backflow threshold engine 228 can generate one or more of these features based on one or more data instances of historical data having features that are closest or most similar to the features of the active backflow data. For example, there may not be enough (or any) data instances sufficient to be within the predefined threshold for selecting the backflow cluster 254, and the backflow threshold engine 228 can select and use one or more of the immediately closest data instances (based on one or more previous condition measurements) to generate one or more of these features. In this way, the backflow threshold engine 228 can provide at least some indication of the expectation or prediction of active backflow, such as in the case where it may not be feasible to make a backflow prediction according to the more specific and / or improved techniques described herein.

[0082] As mentioned above, and as Figure 2 As shown, the flowback monitoring system 220 may include a flowback alarm manager 230. The flowback alarm manager 230 may access and / or process some or all of the data in the data storage device 232 (e.g., threshold data 238) to detect abnormal active flowback and generate one or more warnings or alarms.

[0083] In some embodiments, the flowback alarm manager 230 receives flowback measurements of active flowback from the sensor manager 222 (eg, or accesses active flowback data 234). Figure 6 As shown, the flowback alarm manager 230 can monitor active flowback 258, for example, against a predicted or expected active flowback represented by the best fit curve 252. For example, the flowback alarm manager 230 can generate and / or monitor a volume flowback 260 that varies over time for both the best fit curve 252 and the active flowback 258. The flowback alarm manager 230 can monitor the active flowback relative to one or more of the flowback thresholds 256 determined by the flowback threshold engine 228. For example, the volume flowback 260 that varies over time can include an upper limit 256-1 and / or a lower limit 256-2. In some embodiments, the flowback alarm manager 230 plots, graphs, or otherwise displays the volume flowback 260 that varies over time.

[0084] In another example, the flowback alarm manager 230 can generate and / or monitor a differential volume flowback 262 over time. For example, the flowback alarm manager 230 can generate and monitor a differential active flowback 264, or a difference in the active flowback 258 compared to the best fit curve 252. The flowback alarm manager 230 can monitor the differential active flowback 264 relative to one or more of the flowback thresholds 256 determined by the flowback threshold engine 228. For example, the differential volume flowback 262 over time can include an upper limit 256-1 and / or a lower limit 256-2. In some embodiments, the flowback alarm manager 230 plots, graphs, or otherwise displays the differential volume flowback 262 over time.

[0085] Based on monitoring active flowback (e.g., via the time-varying volume flowback 260 and / or the time-varying differential volume flowback 262), the flowback alarm manager 230 may implement and / or generate one or more alarms. The alarm may be any type or form of alarm, warning, indication, or flag associated with the detection of abnormal flowback. For example, the alarm may be a visual or audible alarm or any other communication used to alert a user of abnormal flowback. The alarm may be a flag, annotation, record, file, or data entry (and combinations thereof) saved to the data storage device 232. The alarm may be implemented in conjunction with the hardware and / or software of the computing device 212 and / or the flowback monitoring system 220. In this way, the flowback alarm manager 230 may generate one or more forms of alarms to warn and / or record instances of abnormal active flowback.

[0086] As described above, the volume flowback 260 over time and / or the differential volume flowback 262 over time can include one or more of the flowback thresholds 256 (e.g., including the upper limit 256-1 and / or the lower limit 256-2) determined by the flowback threshold engine 228. The flowback threshold 256 can correspond to an alarm, or the flowback alarm manager 230 can trigger an alarm based on the flowback threshold 256.

[0087] For example, Figure 7 An exemplary portion of differential volume flowback 262 over time is shown. In some cases, differential active flowback 264 may exceed or exceed one of the flowback thresholds 256, such as upper limit 256-1, at a high alarm point 270. The flowback alarm manager 230 may generate an alarm (e.g., a high alarm) based on the high alarm point 270 accordingly. In some embodiments, the high alarm corresponds to a formation kick associated with the wellbore. In this way, the flowback alarm manager can generate an alarm and / or warn a user that the active flowback volume from the wellbore is excessive (e.g., a value that exceeds an expected value range).

[0088] Similarly, in some cases, the differential active flowback 264 may exceed or exceed one or more other thresholds, such as the lower limit 256-2. The flowback alarm manager 230 can generate an alarm (e.g., a low alarm) based on the low alarm point accordingly. In some embodiments, the low alarm corresponds to a formation loss associated with the wellbore. In this way, the flowback alarm manager can generate an alarm and / or warn a user of a decrease in active flowback from the wellbore (e.g., a value below an expected value range).

[0089] Furthermore, it should be appreciated that the flowback alarm manager 230 may implement and / or generate one or more alarms associated with the volumetric flowback 260 over time (e.g., associated with monitoring active flowback 258), such that the flowback alarm manager 230 is not limited to Figure 7 An embodiment of differential volume flowback 262 over time is shown. For example, flowback alarm manager 230 can monitor active flowback 258 for volume flowback 260 over time and can generate one or more alarms based on active flowback 258 exceeding or exceeding one or more of flowback thresholds 256 .

[0090] In this way, the flowback alarm manager 230 may generate one or more alarms to indicate an abnormal level, amount or rate of active flowback from the wellbore, such as an excess or reduction in flowback compared to an expected or predicted flowback. Such indications may be useful, for example, for detecting abnormal wellbore conditions, such as abnormal or unexpected inflow of formation fluids into the wellbore (formation kick) or abnormal loss of drilling fluids to the formation (formation loss). In some embodiments, the flowback alarm manager 230 takes one or more mitigation measures in response to the alarm. For example, one or more drilling parameters may be adjusted, modified or changed (e.g., by a user or automatically by the flowback monitoring system 220) to mitigate wellbore conditions associated with active flowback and / or alarms. This functionality may help reduce the risk of severe, troublesome and / or catastrophic wellbore events, such as blowouts, formation / wellbore instability, valve damage or valve control problems, fluid circulation problems or other wellbore problems (and combinations thereof). In this way, the alarms generated by the flowback alarm manager 230 may be useful for downhole systems to operate efficiently, effectively and safely.

[0091] In some embodiments, the flowback alarm manager 230 implements an alarm delay. The alarm delay can indicate a time period during which the flowback alarm manager 230 will not or has not generated one or more alarms. For example, the differential active flowback 264 (and / or active flowback 258) may exceed one or more thresholds during the time period of the alarm delay, and the flowback alarm manager 230 may not generate a corresponding alarm. The alarm delay can be a period of time after the start of the pump off period, or a period of time after receiving an indication of the pump off period. For example, the alarm delay can be 10 seconds, 20 seconds, 30 seconds, 40 seconds, 50 seconds, 60 seconds, 90 seconds, 120 seconds, 240 seconds, 360 seconds, or any value therebetween. In some embodiments, it is important that the alarm delay is at least 120 seconds after the start of the pump off period to allow active flowback to initialize and / or normalize so that the flowback can be accurately measured and / or characterized.

[0092] In some embodiments, the flowback alarm manager 230 cancels, deletes, or otherwise turns off one or more alarms. For example, the flowback alarm manager 230 may cancel an alarm based on the differential active flowback 264 weakening or returning to one or more of the flowback thresholds 256. For example, the flowback alarm manager 230 may observe that the differential active flowback 264 exceeds the upper limit 256-1 (e.g., at the high alarm point 270), and the flowback alarm manager 230 may generate a high alarm accordingly. In some cases, the differential active flowback 264 may weaken and / or return to within the upper limit 256-1. The flowback alarm manager 230 may cancel or turn off the high alarm accordingly.

[0093] In another example, the flowback alarm manager 230 may not cancel the alarm based on the differential active flowback 264 returning to exceed the same flowback threshold 256 that triggered the alarm, but may maintain the alarm until the observed flowback exceeds the additional flowback threshold 256. For example, a high alarm may be triggered at the high alarm point 270, and the differential active flowback 264 may weaken or return to within the upper limit 256-1 at the return point 274. The flowback alarm manager 230 may not cancel the high alarm at the return point 274, but may maintain the alarm until the differential active flowback 264 returns to within the alarm closure threshold 256-3 at the alarm closure point 272. In some embodiments, the alarm closure threshold 256-3 is a narrower or tighter threshold (e.g., closer to the best fit curve 252) than the upper limit 256-1 that triggered the alarm. For example, the upper limit 256-1 may be three standard deviations from the best fit curve 252, and the alarm closure threshold may be 1 standard deviation from the best fit curve 252. In this manner, the flowback alarm manager 230 may implement the alarm shutoff threshold 256 - 3 to prevent alarms indicating abnormal active flowback from being regularly triggered and canceled and / or to ensure that abnormal active flowback abates to a sufficient degree so that the alarm can be confidently canceled.

[0094] Figure 8 800, or a flow chart of a series of actions for analyzing flowback of a downhole system as described herein, in accordance with at least one embodiment of the present disclosure. Figure 8 Actions are shown according to one embodiment, but alternative embodiments may add, omit, reorder, and / or modify Figure 8 Any action shown.

[0095] In some embodiments, method 800 includes an act 810 of generating active flowback data by monitoring active flowback from the wellbore. For example, a flowback monitoring system may identify an active pump-off period of the downhole system and may generate active flowback data corresponding to the active pump-off period.

[0096] In some embodiments, method 800 includes action 820, i.e., identifying historical flowback data of historical flowback from the wellbore. In some embodiments, the flowback monitoring system generates historical flowback data by monitoring historical flowback from the wellbore and caching the measurement results as historical blowback data. The historical flowback data can correspond to one or more previous pump shutdown periods. For example, the historical data can be one or more instances of active flowback data cached after the corresponding active pump shutdown period ends. In some embodiments, the active flowback data is generated during the same operation of the downhole tool corresponding to the historical flowback data. For example, the cache of historical flowback data can be cleared based on the downhole tool being pulled out to the ground. In some embodiments, one or more data instances of historical flowback data are cleared from the cache based on a time interval or the age of the data instance.

[0097] In some embodiments, method 800 includes action 830, i.e., determining a flowback cluster based on comparing the active flowback data with the historical flowback data. For example, the flowback monitoring system can identify one or more instances of historical flowback data having previous conditions similar to the conditions of the active flowback data. The similar previous conditions can include one or more of flow rate, depth of downhole tool, standpipe pressure, and flow pattern. In some embodiments, the flowback monitoring system can filter, clear, or delete one or more instances of historical flowback data based on the data quality of the one or more instances.

[0098] In some embodiments, method 800 includes action 840, i.e., generating a flowback threshold based on the flowback cluster. For example, the flowback monitoring system can generate a flowback threshold with an upper limit and / or a lower limit. In some embodiments, the flowback monitoring system automatically generates the threshold in real time when an active pump off period is identified.

[0099] In some embodiments, method 800 includes an action 850 of generating an alarm based on the active flowback data exceeding a flowback threshold. For example, the flowback monitoring system may generate an alarm based on the active flowback data exceeding an upper limit. In some embodiments, a drilling parameter of a downhole system may be adjusted based on the alarm. In some embodiments, the flowback monitoring system may cancel an alarm based on the active flowback data exceeding an alarm off threshold. The alarm off threshold may be different from the flowback threshold (e.g., different from the upper limit).

[0100] Fig. 9 A flowchart showing a method 900, or series of actions, for analyzing flowback of a downhole system as described herein is shown in accordance with at least one embodiment of the present disclosure. Fig. 9 Actions are shown according to one embodiment, but alternative embodiments may add, omit, reorder, and / or modify Fig. 9 any action.

[0101] In some embodiments, method 900 includes an act 910 of generating active flowback data by monitoring active flowback from a wellbore. In some embodiments, method 900 includes an act 920 of identifying historical flowback data of historical flowback from a wellbore. In some embodiments, method 900 includes an act 930 of determining a flowback cluster based on comparing the active flowback data to the historical flowback data.

[0102] In some embodiments, method 900 includes action 940, i.e. predicting active flowback based on flowback clusters. For example, the flowback monitoring system can generate a best fit curve based on the flowback clusters. The flowback monitoring system can calculate the probability distribution of the flowback clusters based on the best fit curve. The flowback monitoring system can generate a threshold value based on the probability distribution. The threshold value can be based on the standard deviation of the probability distribution. In some embodiments, the probability distribution is a Gaussian distribution.

[0103] Now go to Fig.10 , which illustrates certain components that may be included within a computer system 1000. One or more computer systems 1000 may be used to implement the various devices, components, and systems described herein.

[0104] The computer system 1000 includes a processor 1001. The processor 1001 may be a general-purpose single-chip or multi-chip microprocessor (e.g., an Advanced RISC (Reduced Instruction Set Computer) Machine (ARM)), a special-purpose microprocessor (e.g., a Digital Signal Processor (DSP)), a microcontroller, a programmable gate array, etc. The processor 1001 may be referred to as a central processing unit (CPU). Although Fig.10 Only a single processor 1001 is shown in the computer system 1000 , but in an alternative configuration, a combination of processors (eg, an ARM and DSP) could be used.

[0105] The computer system 1000 also includes a memory 1003 in electronic communication with the processor 1001. The memory 1003 may be any electronic component capable of storing electronic information. For example, the memory 1003 may be embodied as a random access memory (RAM), a read-only memory (ROM), a disk storage medium, an optical storage medium, a flash memory device in a RAM, an onboard memory included in a processor, an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM) memory, a register, etc., including combinations thereof.

[0106] Instructions 1005 and data 1007 may be stored in memory 1003. Instructions 1005 may be executed by processor 1001 to implement some or all of the functionality disclosed herein. Execution of instructions 1005 may involve the use of data 1007 stored in memory 1003. Any of the various examples of modules and components described herein may be implemented in part or in whole as instructions 1005 stored in memory 1003 and executed by processor 1001. Any of the various examples of data described herein may be data 1007 stored in memory 1003 and used during execution of instructions 1005 by processor 1001.

[0107] The computer system 1000 may also include one or more communication interfaces 1009 for communicating with other electronic devices. The communication interface 1009 may be based on wired communication technology, wireless communication technology, or both. Some examples of the communication interface 1009 include a universal serial bus (USB), an Ethernet adapter, a wireless adapter operating according to the Institute of Electrical and Electronics Engineers (IEEE) 802.11 wireless communication protocol, Wireless communication adapter and infrared (IR) communication port.

[0108] The computer system 1000 may also include one or more input devices 1011 and one or more output devices 1013. Some examples of input devices 1011 include keyboards, mice, microphones, remote controls, buttons, joysticks, trackballs, touch pads, and light pens. Some examples of output devices 1013 include speakers and printers. One specific type of output device typically included in the computer system 1000 is a display device 1015. The display device 1015 used with the embodiments disclosed herein may utilize any suitable image projection technology, such as a liquid crystal display (LCD), a light emitting diode (LED), gas plasma, electroluminescence, and the like. A display controller 1017 may also be provided for converting data 1007 stored in the memory 1003 into text, graphics, and / or moving images (as appropriate) shown on the display device 1015.

[0109] The various components of the computer system 1000 may be connected together via one or more buses, which may include a power bus, a control signal bus, a status signal bus, a data bus, etc. For clarity, Fig.10 The individual buses are shown as bus system 1019.

[0110] The techniques described herein may be implemented in hardware, software, firmware, or any combination thereof, unless specifically described as being implemented in a particular manner. Any features described as modules, components, or the like may also be implemented together in an integrated logic device or separately as discrete but interoperable logic devices. If implemented in software, the techniques may be implemented at least in part by a non-transitory processor-readable storage medium comprising instructions that, when executed by at least one processor, perform one or more of the methods described herein. Instructions may be organized into routines, programs, objects, components, data structures, and the like that may perform specific tasks and / or implement specific data types, and may be combined or distributed as desired in various embodiments.

[0111] Computer-readable media can be any available media that can be accessed by a general or special purpose computer system. A computer-readable medium that stores computer-executable instructions is a non-transitory computer-readable storage medium (device). A computer-readable medium that carries computer-executable instructions is a transmission medium. Therefore, by way of example and not limitation, embodiments of the present disclosure may include at least two distinct computer-readable media: a non-transitory computer-readable storage medium (device) and a transmission medium.

[0112] Industrial Applicability

[0113] In some embodiments, the downhole system includes a drilling rig for rotating a drilling assembly extending downwardly into the wellbore. The drilling assembly may include a drill string, a bottom hole assembly (BHA), and a drill bit attached to the downhole end of the drill string.

[0114] The drill string may include several joints of drill pipe connected end to end by tool joints. The drill string transmits drilling fluid through a center hole and transmits rotational power from the drilling rig to the BHA. In some embodiments, the drill string also includes additional components such as short sections, short drill pipes, etc. The drill pipe provides hydraulic channels through which drilling fluid is pumped from the surface. The drilling fluid is discharged through nozzles, orifices or other orifices in the drill bit for cooling the drill bit and the cutting structure thereon, and for transporting drill cuttings out of the wellbore when the wellbore is drilled.

[0115] The BHA may include a drill bit or other components. An exemplary BHA may include additional or other components (e.g., coupled between the drill string and the drill bit). Examples of additional BHA components include drill collars, stabilizers, measurement while drilling (MWD) tools, logging while drilling (LWD) tools, downhole motors, under-pipe reamer, slicer mills, hydraulic circuit breakers, cans, vibration or shock absorbing tools, other components, or combinations of the foregoing components. The BHA may also include a rotary steerable system (RSS). The RSS may include a directional drilling tool that changes the direction of the drill bit, thereby changing the trajectory of the wellbore. In some cases, at least a portion of the RSS maintains a geostationary position relative to an absolute reference system (such as gravity, magnetic north, or true north). Using the measurements obtained through the geostationary position, the RSS can position the drill bit, change the route of the drill bit 110, and guide the directional drilling tool on the expected trajectory.

[0116] Generally speaking, a downhole system may include additional or other drilling components and accessories, such as specialized valves (e.g., kelly cocks, blowout preventers, and safety valves). Additional components included in a downhole system may be considered part of a drill string, a drill string, or a part of a BHA, depending on their location in the downhole system.

[0117] In some embodiments, the downhole motor in the BHA generates power for the downhole system and / or provides rotational energy to the downhole components (e.g., to rotate the drill bit). The downhole motor can be any type of downhole motor, including a positive displacement pump (such as a progressive cavity motor) or a turbine. In some embodiments, the downhole motor is powered by drilling fluid. In other words, the drilling fluid pumped downhole from the surface can provide energy to rotate the rotor in the downhole motor. The downhole motor can operate at an optimal pressure differential or a range of pressure differentials. The optimal pressure differential can be a pressure differential at which the downhole motor does not stall, burn out, over-rotate, or suffer other damage. In some cases, the downhole motor drives the rotation of the drill bit. In some embodiments, the rotation of the drill bit is driven by a component at the surface of the wellbore.

[0118] The drill bit in the BHA can be any type of drill bit suitable for degrading downhole materials. For example, the drill bit can be a drill bit suitable for drilling surface formations. Exemplary types of drill bits for drilling surface formations are fixed cutter or scraper bits and roller cone drill bits. In other embodiments, the drill bit can be a milling shoe for removing metals, composites, elastomers, other downhole materials, or combinations thereof. For example, the drill bit can be used with a whipstock to grind into a casing sleeved on a wellbore. The drill bit can also be a flat-head milling shoe for milling out tools, plugs, cement, other materials, or combinations thereof within the wellbore. The cuttings or other drill cuttings formed by the use of the milling shoe can be transported to the surface, or can be allowed to fall downhole. In other embodiments, the drill bit can include a reamer. For example, a reamer can be used in conjunction with the drill bit, and the drill bit can drill into the formation, while the reamer can expand the size of the borehole.

[0119] As described above, the downhole system may include or may implement drilling fluid or drilling mud. In some embodiments, the downhole system includes a mud tank. The mud tank may be a reservoir for holding a certain amount of drilling fluid. The drilling fluid may be pumped from the mud tank, through one or more fluid lines, and down the drill bit along the drill tool assembly. As described above, the drilling fluid may flow out of the drill bit, such as at the bottom of the wellbore, in order to cool the drill bit and / or remove and carry away drill cuttings.

[0120] In some embodiments, the drilling fluid is implemented in this manner and at least partially fills the wellbore. For example, the drilling fluid can be pumped through the drill tool assembly and out of the drill bit so that the wellbore can be filled, and the drilling fluid can flow out of the wellbore annular area around the drill string. In this manner, the drilling fluid can flow upward through the annular area of ​​the wellbore to carry formation cuttings and heat away from the drill bit. In this manner, the drilling fluid can be circulated from the mud tank, down through the drill tool assembly, back to the wellbore, and back to the mud tank.

[0121] During the drilling operation, it may be necessary to periodically stop or pause the pumping of drilling fluid into the wellbore. For example, an extra length of drill pipe may be added to the drill tool assembly periodically as the drill bit is pushed downward through the surface formation, and the pump supplying drilling fluid to the drill tool assembly may be shut down. In many cases, during the pump shut-down period, a certain amount of fluid (e.g., drilling fluid and / or formation fluid) may be backflowed and / or flow out of the wellbore (e.g., an annular region). For example, pumping drilling fluid into the wellbore may create a pressure imbalance in the wellbore, which may offset the inflow of formation pressure and formation fluid (such as oil, gas or water). Shutting down the pump may release the pressure imbalance, and the formation pressure may cause the drilling fluid and / or formation fluid to flow upward and flow out of the wellbore. In addition, in many cases, natural gas may become dissolved in the drilling fluid, and when the pressure of the drilling fluid is released, the natural gas may expand, causing the volume to increase and the drilling fluid to flow upward and flow out of the wellbore. Fluid flowing upward and flowing out of the wellbore may also be caused by a variety of other reasons.

[0122] After the pump is shut off, the outflow of fluid from the wellbore is called flowback. Measures can usually be taken to control, predict and / or anticipate a certain volume of flowback and to safely recover and process the fluid. In some embodiments, an excessive amount of flowback (e.g., higher than expected) indicates a well kick or formation kick, or formation fluid flows from the formation into or into the wellbore. Detecting and responding to formation kicks is essential for maintaining well control and preventing more serious accidents such as blowouts, thereby ensuring the safe operation of the downhole system. In some embodiments, a reduction in flowback (e.g., lower than expected) indicates formation loss, or drilling fluid (e.g., accidentally) is lost into the formation. Detecting and responding to formation loss is essential for maintaining the integrity of the well, preventing damage to the surrounding formations, and maintaining sufficient hydrostatic pressure in the wellbore to control the well and prevent formation inflow. Therefore, the interpretation and analysis of the flowback pattern (referred to as flowback fingerprint comparison) can provide favorable information for the effective operation of the downhole system.

[0123] However, in some cases, it may be difficult to adequately and / or accurately characterize and / or explain the flowback patterns observed in a downhole system. For example, many factors may affect the amount and extent of flowback after a pump shut-off. Thus, it may be difficult to accurately characterize and / or predict the expected flowback volume to determine excess or underflow during any given pump shut-off period. Therefore, improved systems and methods for detecting abnormal flowback may provide advantages over conventional techniques.

[0124] In some embodiments, the downhole system includes a computing device. The computing device may refer to various computing devices, including mobile devices, such as mobile phones, smart phones, laptop computers, or other mobile devices. The computing device may refer to a non-mobile device, such as a desktop computer, a server device, or other non-portable devices. The computing device may be a surface computing device, a mobile computing device, a remote computing device, a downhole computing device, and combinations thereof. The computing device may electronically communicate with one or more components of the downhole system, such as sensors, electronic components, and other computing devices.

[0125] In some embodiments, the computing device includes a flowback monitoring system implemented thereon. As will be discussed in further detail below, the flowback monitoring system can facilitate monitoring, analyzing, and / or interpreting flowback experienced by a downhole system, such as during one or more pump shut-off periods. In this way, the flowback monitoring system can facilitate more accurately understanding and characterizing one or more downhole conditions based on observed flowback.

[0126] In some embodiments, the computing device has a backflow monitoring system implemented thereon. The backflow monitoring system may include a sensor manager, a data manager, a data cluster engine, a backflow threshold engine, and a backflow alarm manager. In addition, the backflow monitoring system may include a data storage device, which includes active backflow data, historical backflow data, and threshold data. Although one or more embodiments described herein describe the features and functions performed by specific components of the backflow monitoring system, it should be understood that the specific features described in conjunction with one or more components of the backflow monitoring system can be performed by one or more other components of the backflow monitoring system in some examples.

[0127] By way of example, as will be discussed below, one or more features of the data manager may be delegated to other components of the flowback monitoring system. As another example, the data clustering engine may select and / or process one or more instances of data, and in some instances, some or all of these features may be performed by the flowback threshold engine (or other components of the flowback monitoring system). In fact, it should be understood that some or all of a particular component may be combined into other components, and that a particular function may be performed by one or more components of the flowback monitoring system.

[0128] As just mentioned, the flowback monitoring system includes a sensor manager. The sensor manager can communicate data with one or more sensors and can monitor and / or manage measurements and / or signals received from the sensors. In this way, the sensor manager can measure and / or monitor one or more parameters of the downhole system.

[0129] The flowback monitoring system may be connected to the sensor and / or may communicate with the sensor. The sensor may include one or more sensors implemented in or in conjunction with the downhole system. In some embodiments, the sensor includes a pump on / off sensor. As discussed above, the downhole system may pump drilling fluid into the wellbore via a drill bit assembly. In some cases, the pump may be shut down, such as to add or replace one or more sections of drill pipe, or to stop, adjust or change the operation of the downhole system. The pump on / off sensor may monitor the state of the pump and may transmit the state or condition of the pump to the sensor manager. In this way, the flowback monitoring system may monitor the current operating state of the pump, for example, in order to detect a pump shutdown period.

[0130] In some embodiments, the sensor includes a flow rate sensor. The flow rate sensor can measure or detect the flow rate of the drilling fluid supplied to the wellbore. For example, the flow rate sensor can measure the flow rate of the drilling fluid entering the drill tool assembly. The flow rate sensor can measure or detect the flow rate of the drilling fluid, such as the flow rate at the surface of the wellbore. The flow rate sensor can measure the flow rate of the drilling fluid at the pump. The flow rate sensor can measure the flow rate of the drilling fluid at one or more locations in the drill tool assembly (such as at the drill bit). As will be discussed in further detail below, the flow rate before the pump shutdown period can be a previous condition associated with determining the abnormal backflow during the pump shutdown period. Therefore, measuring the flow rate in this way can be beneficial to characterize the observed or active backflow. To this end, in some embodiments, the flow rate sensor constantly monitors the flow rate so as to determine the previous flow rate when the pump shutdown period is identified. In some embodiments, the flow rate sensor regularly measures the flow rate, or is activated to measure the flow rate before the pump is shut down. In this way, the backflow monitoring system can monitor and / or measure the flow rate of the drilling fluid supplied to the wellbore before the pump shutdown period.

[0131] In some embodiments, the sensor includes a flowback sensor. The flowback sensor can measure the flow rate (e.g., the amount of flowback from the wellbore) of a fluid (e.g., drilling fluid and / or formation fluid) flowing or flowing out of the wellbore. The flowback sensor can measure the pattern of the flowback, such as the rate and / or total amount of the fluid flowing out of the wellbore. The flowback sensor can measure the rate and / or total amount of the fluid flowing into the mud tank and / or can measure the level or amount (e.g., effective volume) of the fluid in the mud tank. In some embodiments, the flowback sensor continuously monitors the flowback, such as during part or all of the drilling operation. In some embodiments, the flowback sensor periodically monitors the flowback during part or all of the drilling operation. In some embodiments, the flowback sensor is activated during the pump shut-down period so that the flowback sensor monitors and / or measures the flowback only when the pump is shut down. In this way, the flowback monitoring system can monitor and / or measure the flowback of the fluid flowing or flowing out of the wellbore.

[0132] In some embodiments, one or more of the flow rate sensor and the flowback sensor, alone or in combination, monitors and / or measures (or facilitates monitoring and / or measuring) a flow pattern of the drilling fluid. For example, the movement and distribution of the drilling fluid within the wellbore may vary based on factors such as drilling technology, wellbore conditions, and drilling fluid properties. Therefore, the sensor manager can monitor and / or measure the flow pattern to facilitate characterizing the flowback, as described herein.

[0133] In some embodiments, the sensor includes a drill bit depth sensor. The drill bit depth sensor can monitor and / or measure the depth of one or more components of a downhole system, such as one or more downhole tools. For example, the drill bit depth sensor can measure the depth of a drill bit used to extend and / or widen a wellbore. The measurements of the drill bit depth sensor can be performed in association with a pump shut-off period. For example, the depth of the downhole tool may be related to interpreting the associated backflow (e.g., measured at or near the surface), and the depth of the downhole tool can be measured before and / or during the associated pump shut-off period. In this way, the backflow monitoring system can monitor one or more depths associated with downhole operations.

[0134] In some embodiments, the sensor includes a pressure sensor. The pressure sensor can measure the fluid pressure of the drilling fluid at one or more locations in the downhole system. For example, the pressure sensor can measure the drilling fluid pressure at the pump, the riser, one or more locations in the drill tool assembly, the drill bit, one or more locations in the wellbore (e.g., an annular area), and combinations thereof. The pressure sensor can continuously, periodically measure the drilling fluid pressure, and / or can be triggered or activated corresponding to a pump shut-off period (e.g., measuring the pressure before the pump shut-off period). In this way, the backflow monitoring system can monitor the drilling fluid pressure at one or more locations in the downhole system.

[0135] It should be understood that the downhole system and the backflow monitoring system are not limited to including those sensors discussed above, but the sensors may include any other sensors (or one or more sensors may be omitted) and / or the sensor manager may also be connected to one or more sensors other than the sensors discussed herein and / or communicate with the one or more sensors. In fact, the sensor may include any sensor for monitoring and / or measuring one or more parameters (e.g., previous conditions) related to the monitoring, measuring and / or characterizing backflow described herein. In addition, the various sensors described herein may be implemented as one or more physical sensing devices, one or more calculations based on signals from physical sensing devices, and combinations thereof. For example, a pump on / off sensor may be a physical sensor that detects the active state of a pump, or may be a calculation based on another signal that the pump is turned on or off, such as a flow rate sensor. In this way, the term "sensor" should be understood to cover devices for physically sensing conditions or measurements, as well as software implementations for calculations associated with conditions or measurements (and combinations thereof).

[0136] As described above, the flowback monitoring system may include a data manager. The data manager may receive data or information corresponding to one or more sensor measurements (e.g., from a sensor manager). The data manager may save and / or store the sensor measurements to a data storage device (e.g., as active flowback data and / or historical flowback data as discussed below). In some embodiments, the data manager associates one or more types and / or instances of sensor measurements with each other. For example, the data manager may receive flowback measurements and may map or associate the flowback measurements with flow rate measurements, flow pattern measurements, depth measurements, pressure measurements, any other measurements acquired and / or accessible by the data manager, and combinations thereof. According to at least one embodiment of the present disclosure, the data manager associates the flowback measurements with (e.g., previous) flow rate measurements.

[0137] In some embodiments, the data manager receives pump on / off measurements corresponding to pump on or off periods from the sensor manager. The data manager can save and / or store one or more of the sensor measurements based on receiving an indication of a pump off period. The data manager can associate or map two or more measurements and / or instances of sensor measurements based on receiving an indication of a pump off period.

[0138] In some embodiments, at least some of the sensor measurements are associated with a prior condition of the downhole system and / or wellbore. For example, the sensor manager may monitor and / or measure one or more aspects of the downhole system prior to the pump shut-off period. Upon receiving an indication of a pump shut-off period, the data manager may save and / or store one or more measurements and / or instances of sensor measurements prior to the pump shut-off period as prior conditions of the wellbore associated with the pump shut-off period. For example, the data manager may receive a pump shut-off indication and may call flow rate, pressure, depth, flow pattern, or any other measurement (and combination thereof) and may save these measurements as prior conditions for the pump shut-off period. The prior condition may be one or more measurements taken within 10 seconds, 30 seconds, 1 minute, 2 minutes, 4 minutes, 5 minutes, or 10 minutes (or any value therebetween) after the start of the pump shut-off period. In some embodiments, it is critical that each prior condition is a measurement taken within 2 minutes after the start of the pump shut-off period so that the measurements of the prior condition accurately reflect the state and / or behavior of the well associated with the backflow measurement taken during the pump shut-off period (e.g., after the prior condition). Each prior condition can be calculated or associated with an average, minimum, maximum, mean, median, mode, range, or any other value or range of values ​​of the prior condition measurements. In this way, flowback measurements taken during a pump shut-in period can be associated with one or more prior conditions that are indicative of observable well conditions when the pump is shut-in.

[0139] In some embodiments, saving prior conditions is associated with monitoring and / or measuring and / or storing (e.g., active) flowback measurements after indicating that a pump-off period has been initiated. In this way, flowback can be measured (e.g., during a pump-off period), and one or more prior conditions of the downhole system and / or wellbore can be saved and / or associated with measurements of active (e.g., pump-off) flowback. As will be discussed below, mapping and / or associating flowback to prior conditions of the wellbore can be beneficial in characterizing and / or interpreting flowback measurements.

[0140] As described above, the data storage device may include active flowback data and historical flowback data. In some embodiments, the data manager receives sensor measurements and stores them as active flowback data. For example, the sensor manager may receive a pump shut-off indication and may begin recording and / or saving flowback measurements as active flowback data, corresponding to active flowback of the downhole system. The data manager may call and save prior conditions associated with active flowback and / or active pump shut-off periods as active flowback data (e.g., a portion). In this way, the active flowback data may include flowback measurements and prior conditions associated with and / or based on active pump shut-off periods of the downhole system.

[0141] In some embodiments, the data manager stores and / or caches some or all of the active flowback data as historical flowback data. The data manager can cache the active flowback data based on one or more trigger conditions, such as a period of time after the pump is turned off, the pump is turned back on, and / or the flowback stops or becomes identified as stable (e.g., non-abnormal). For example, the data manager can record and / or store active flowback data during part or all of the active pump off period. Based on receiving an indication that the pump is turned back on (e.g., the end of the active pump off period), the data manager can migrate the active flowback data of the (now past) pump off period and save the data as historical flowback data. In some embodiments, the data manager can clear the active flowback data in preparation for measuring and / or storing (e.g., new) active flowback data corresponding to the next pump off period. The data manager can begin recording and / or storing additional (e.g., new) sensor measurements as active flowback data. In this manner, active flowback data may include and / or may be associated with measurements corresponding only to an active (or upcoming) pump-off period, and historical flowback data may include measurements associated with one or more previous pump-off periods (e.g., previous instances of active flowback data). In this manner, as discussed below, active flowback data may be compared to historical flowback data to facilitate accurate characterization and interpretation of active flowback of a downhole system.

[0142] In some embodiments, the data manager clears or deletes one or more data instances from the data storage device. For example, in some cases, when the historical flowback data does not occur and / or measure within a specific time range, it can be determined that the historical flowback data is less relevant and / or less valuable for comparison with the active flowback data. Therefore, the data manager can clear or delete one or more instances of historical flowback data based on instances with a threshold age or occurring before a threshold time interval. In some embodiments, the data manager clears data instances that are more than 1 hour, 2 hours, 3 hours, 4 hours, 5 hours, 6 hours, 7 hours, 8 hours, 9 hours or 10 hours (or any value therebetween) from the historical flowback data (e.g., occurring more than that long before the active flowback data and / or active pump shut-off period). In another example, in some cases, when the historical flowback data does not occur during the same downhole operation, the same downhole operation type, or the same downhole tool operation, it can be determined that the historical flowback data is less relevant and / or less valuable for comparison with the active flowback data. Thus, the data manager can purge or delete some or all of the historical flowback data based on the downhole tool being pulled out of the wellbore and / or based on a change in downhole operation or type of downhole operation. In this way, the data manager can maintain data in the data storage device so that the historical flowback data is most relevant, useful, and / or valuable for characterization and / or interpretation that facilitates active flowback.

[0143] As described above, the flowback monitoring system may include a data clustering engine. The data clustering engine may sort, index, filter, and / or select one or more instances of historical flowback data in a data storage device. For example, the data clustering engine may select a flowback cluster, or one or more instances of historical flowback data (e.g., corresponding to one or more previous pump shut-down periods), and may save or store the flowback cluster to the data storage device as threshold data. The flowback cluster and threshold data may facilitate comparison of one or more previously observed wellbores' flowbacks for comparison with active flowbacks (e.g., comparison with active flowback data) in order to characterize and / or interpret the flowbacks.

[0144] In some embodiments, the data clustering engine automatically selects a flowback cluster based on a pump on / off period. For example, the data clustering engine may receive an indication that a pump has been turned off, and may automatically sort, filter, and / or select one or more instances of historical flowback data as a flowback cluster, and may store the flowback cluster as threshold data to a data storage device. The data clustering engine may select a flowback cluster at any other time or based on any other triggering factor.

[0145] In some embodiments, the data cluster engine filters and / or sorts the historical flowback data, and can remove, delete, mark and / or isolate one or more instances of historical flowback data (and / or active flowback data) from the data storage device. The data cluster engine can mark, filter or remove data instances based on the data quality of the data. For example, in some cases, one or more measurement results and / or data instances may or may become damaged, incomplete, disjointed, unclear or poor quality in other aspects. In some cases, the data cluster engine identifies one or more measurement results and / or data instances as inaccurate, uncalibrated, unexpected or outliers. These sensors and / or data and / or measurement errors may be the result of sensor failure, disconnection or damage or any other reason. The data cluster engine can mark, separate, filter and / or delete one or more measurement results and / or data instances of historical flowback data and / or active flowback data accordingly. In this way, the data cluster engine can maintain the quality of the measurement results and / or data of the flowback monitoring system.

[0146] The data clustering engine can select flowback clusters based on data instances that are well suited for evaluating active flowback and / or more relevant to characterization of active flowback. For example, one or more instances of historical flowback data can be relevant and / or useful for comparison with active flowback data when the data instances have one or more characteristics that are similar to characteristics of the active flowback data.

[0147] In some embodiments, a flowback cluster is automatically selected based on one or more of the prior conditions associated with the active flowback data and the historical flowback data. For example, the data clustering engine can sort, arrange, classify, list or otherwise organize the historical flowback data based on one or more of the prior conditions associated with the historical flowback data. According to at least one embodiment of the present disclosure, the data clustering engine organizes the historical flowback data based on the prior flow rate measurements associated with the historical flowback data (e.g., the flow rate before each pump shut-down period associated with each data instance). The data clustering engine can organize the historical flowback data based on pressure measurements, depth measurements, flow pattern measurements, any other measurements, and combinations thereof.

[0148] After the historical flowback data is sorted, the data clustering engine can select one or more data instances of the historical flowback data as a flowback cluster. For example, the data clustering engine can select 1, 2, 3, 4, 5, 6, 7, 8, 9 or 10 (or any value therebetween) data instances of the historical flowback data to form a flowback cluster. In some embodiments, it is crucial that the flowback cluster includes three data instances of the historical flowback data so that the flowback threshold and / or envelope (as discussed below) will not be too narrow or too wide. This can be beneficial to correctly and accurately characterize active flowback and prevent false alarms and missed alarms caused by incorrect threshold calculations and / or sizes.

[0149] The data clustering engine can select a flowback cluster based on the similarity of the previous conditions of the historical flowback data and the previous conditions of the active flowback data. For example, a flowback cluster can be selected based on a data instance with a previous flow rate within a predefined range compared to the active flowback data. A flowback cluster can be selected based on the previous flow rate of the data instance, and the previous flow rate is within the range of 10, 20, 40, 50, 100, 150 or 200 gallons per minute (or any value therebetween) of the previous flow rate of the active flowback data. For example, the data clustering engine can select all data instances that fall within a given range, or can select a specific number of data instances (for example, also within a given range) that are closest to the flow rate of the active flowback data. In some embodiments, it is crucial that the flow rate of the flowback cluster is within the range of 50 gallons per minute of the active flowback data to ensure that the flowback cluster is sufficient to represent the wellbore conditions of the active flowback data.

[0150] In another example, a flowback cluster can be selected based on a data instance with a depth similar or close to the active flowback data (e.g., the depth of the downhole tool before and / or during the corresponding pump shut-down period). A flowback cluster can be selected based on the depth of the data instance within the range of 25m, 50m, 100m, 150m, 200m, 250m, 300m, 400m or 500m of the depth of the active flowback data (or any value therebetween). For example, the data clustering engine can select all data instances that fall within a given range, or can select a specific number of data instances that are closest to the depth of the active flowback data. In some embodiments, it is critical that the depth of the flowback cluster is within the range of 200 meters of the active flowback data to ensure that the flowback cluster is sufficient to represent the wellbore conditions at or near the depth of the downhole tool during the active pump shut-down period of the active flowback data. The data clustering engine can select a flowback cluster based on the similarity of any other measurement results of the previous conditions of the historical flowback data with the measurement results of the active flowback data.

[0151] In some embodiments, the data clustering engine selects a flowback cluster based on two or more prior conditions. For example, a flowback cluster may be selected based on one or more instances of historical flowback data having a flow rate and depth similar to (e.g., close to) the flow rate and depth of the active flowback data. In another example, a flowback cluster may be selected based on one or more instances of historical flowback data having a pressure and depth similar to (e.g., close to) the pressure and depth of the active flowback data. A flowback cluster may be selected based on one or more instances of historical flowback data having similar prior conditions and / or any other combination of measurement results.

[0152] In some embodiments, the data clustering engine determines that one or more (or all) data instances of the historical flowback data do not have a prior condition (a prior condition among various prior conditions) that is sufficiently similar or close to the condition of the active flowback data. Therefore, the data clustering engine may not select one or more data instances for the flowback cluster based on the prior condition, and may continue to compare and / or select one or more data instances for the flowback cluster based on another prior condition or another measurement result. In some embodiments, the selective process is based on a hierarchy or order of relevance of the prior conditions and / or other measurements to the characterization of the active flowback.

[0153] For example, the data clustering engine can sort the historical flowback data based on a first prior condition (such as flow velocity), and can determine that there are no (or not enough) prior flow velocities in the historical flowback data that are within a threshold value of the prior flow velocity of the active flowback data. Therefore, the data clustering engine can select one or more data instances for the flowback cluster not based on flow velocity, and can continue to sort the historical flowback data based on a second prior condition (such as depth). The data clustering engine can select one or more data instances for the flowback cluster based on depth measurements of downhole tools acquired or measured at a depth similar to the active flowback data. This process can be performed through any number of prior conditions and / or measurements until the data clustering engine selects a flowback cluster based on sufficient similarity to the active flowback data.

[0154] In some embodiments, the data clustering engine selects one or more data instances for a flowback cluster based on a first prior condition or measurement result, and may further refine and / or filter the selected data instances by adding and / or removing one or more data instances to the flowback cluster based on a second prior condition or measurement result.

[0155] In this way, the data clustering engine can determine a flowback cluster by selecting one or more data instances of historical flowback data based on data instances having one or more characteristics similar to characteristics of the active flowback data. In this way, defining flowback clusters for historical flowback data based on similarity to the active flowback data can help to more accurately characterize and / or interpret the active flowback data.

[0156] In some embodiments, the data clustering engine determines that an insufficient number (or no) of data instances of historical flowback data have sufficient similarity to the active flowback data based on one or more prior conditions, other measurements, or any other basis. Therefore, the data clustering engine may not select and / or define a flowback cluster. In some embodiments, the data clustering engine indicates to the flowback monitoring system that the flowback cluster is not defined, for example, so that threshold processing and / or alarms may not be performed.

[0157] As described above, the flowback monitoring system may include a flowback threshold engine. The flowback threshold engine may access and / or process some or all of the data in the data storage device to predict the value, rate, and / or behavior of active flowback.

[0158] In some embodiments, the flowback threshold engine generates or fits a curve of one or more instances of historical flowback data. For example, based on the flowback cluster, the flowback threshold engine can generate a best fit curve representing the flowback cluster. In some embodiments, the flowback threshold engine generates a best fit curve for each data instance of the data cluster. In some embodiments, the flowback threshold engine generates a best fit curve for all data instances of the flowback cluster. The best fit curve can be generated using any curve fitting technique, such as linear, polynomial, logarithmic, trigonometric, power, exponential, moving average or any other curve fitting technique (and combination thereof). The flowback threshold engine can save or store the best fit curve as threshold data (or as a part thereof) to a data storage device.

[0159] According to at least one embodiment of the present disclosure, a best fit curve can be generated based on a hyperbolic tangent function. For example, the best fit curve can be fitted based on the following formula:

[0160]

[0161] One or more embodiments may change, enhance, and / or simplify the exemplary formula or one or more portions of the exemplary formula. In some embodiments, a best fit curve based on a hyperbolic tangent function (such as the hyperbolic tangent function shown above) approximates the general shape of the flowback measurements typically observed for flowback from the wellbore. For example, adjusting variable A may adjust the slope of the curve, and adjusting variable B may adjust the asymptote of the curve. In this way, the best fit curve may fit one or more data instances of the flowback cluster so that the best fit curve approximates the trend of the flowback cluster.

[0162] According to at least one embodiment of the present disclosure, a backflow threshold engine can generate a best fit curve for a backflow cluster (e.g., a curve that is the best fit for all data instances of the backflow cluster). The best fit curve can approximate the behavior of all data instances of the backflow cluster. As discussed above, a backflow cluster can be determined or selected based on historical data instances having similar characteristics (e.g., previous conditions) to the active backflow being actively observed. In this way, the best fit curve can approximate and / or predict the behavior of the active backflow and / or the best fit curve can represent a curve or a set of values ​​that the active backflow is expected to follow. For example, when a pump off period is detected, the data clustering engine can (e.g., automatically) select a backflow cluster, and the backflow threshold engine can (e.g., automatically) generate a best fit curve for the backflow cluster. In this way, once a pump off period is detected, the behavior of the active backflow can be (e.g., automatically) predicted.

[0163] In some embodiments, the backflow threshold engine determines and / or generates a probability distribution of a backflow cluster. The backflow threshold engine can determine the probability distribution of the backflow cluster relative to, and / or in contrast to, and / or fit to a best fit curve. For example, for one or more (or all) points or increments in a domain (e.g., time), the backflow threshold engine can determine the probability distribution of the backflow cluster (e.g., relative to the best fit curve) at the corresponding domain point or (e.g., time) increment. In some embodiments, the probability distribution is a Gaussian or normal probability distribution. In some embodiments, the probability distribution is a Bayesian probability distribution. The probability distribution can be any probability distribution and / or can be based on any form or statistical theory, such as a discrete distribution, a continuous distribution, a binomial distribution, a Poisson distribution, a Bernoulli distribution, a hypergeometric distribution, a normal distribution, a uniform distribution, an exponential distribution, a chi-square distribution, a logistic distribution, a student's T distribution, any other distribution and combinations thereof. The backflow threshold engine can save or store the probability distribution as threshold data to a data storage device.

[0164] In some embodiments, the probability distribution includes and / or can be based on or relative to a best fit curve. For example, a probability distribution can be generated such that the best fit curve is and / or represents one or more of a mean, median, mode, average, quartile, variance, standard deviation, any other probability or statistical value, and combinations thereof. In this way, the best fit curve can predict the expected, ideal, or smooth behavior of active flowback, and the probability distribution can be useful in defining and / or characterizing any deviation or divergence of the observed active flowback from the prediction of the best fit curve.

[0165] In some embodiments, the flowback threshold engine determines and / or generates the flowback threshold. The flowback threshold engine can generate the flowback threshold 26 based at least in part on a best fit curve and / or a probability distribution. For example, a probability distribution can define the probability that a flowback measurement result takes its measured value within a range of all possible values ​​within a given time interval. The flowback threshold can be generated based on a percentage of all possible values ​​of each measurement result that are contained, captured, and / or limited within a threshold probability or percentage (e.g., centered on a prediction) within each time interval. The flowback threshold engine can save or store the flowback threshold as threshold data to a data storage device.

[0166] In this way, the backflow threshold engine can generate one or more backflow thresholds, which can indicate the value or value range of the backflow measurement result under the threshold probability or percentage of the expected value (e.g., the best fit curve) of the backflow measurement result. For example, the backflow threshold can represent the amount or degree of acceptable and / or safe deviation of the measured active backflow relative to the best fit curve, and the active backflow measurement result exceeding or exceeding one or more backflow thresholds can indicate excessive and / or undesirable active backflow. According to at least one embodiment of the present disclosure, the backflow threshold is based on one or more standard deviations of the probability distribution. For example, the backflow threshold can capture or include all possible backflow measurement results within 1, 2 or 3 standard deviations or any value (and combination thereof) from the best fit curve. In some embodiments, it is crucial that the backflow threshold is included in three standard deviations above and / or below the best fit curve, so as to facilitate the backflow measurement outside the backflow threshold to be characterized as abnormal with a sufficiently high confidence. For example, the backflow threshold can include an upper limit above three standard deviations of the best fit curve, and a lower limit below three standard deviations of the best fit curve. In this manner, the best fit curve and / or probability distribution may be useful for characterizing and / or explaining deviations or divergences of measured (eg, active) flowback relative to an expected or predicted range of values ​​based on the best fit curve.

[0167] As discussed above, in some embodiments, a flowback cluster is not generated and / or selected, such as due to a lack of suitable data instances for historical flowback data. In some embodiments, the flowback threshold engine does not generate one or more of a best fit curve, a probability distribution, and one or more flowback thresholds. Not generating one or more of these features can help eliminate false positive and / or false negative indications of abnormal and / or undesirable flowback (as discussed below) that may occur by generating predictions and / or thresholds based on insufficient and / or inappropriate historical data.

[0168] In some embodiments, the backflow threshold engine generates one or more of the best fit curve, probability distribution, and one or more backflow thresholds based on data other than the backflow cluster. (For example, when the backflow cluster is not selected). For example, the backflow threshold engine can generate one or more of these features based on all historical data. In another example, the backflow threshold engine can generate one or more of these features based on one or more data instances of historical data with features that are closest or most similar to the features of the active backflow data. For example, there may not be enough (or any) data instances to be within the predefined threshold for selecting the backflow cluster, and the backflow threshold engine can select and use one or more of the immediately adjacent data instances (based on one or more previous conditional measurements) to generate one or more of these features. In this way, the backflow threshold engine can provide at least some indication of the expectation or prediction of active backflow, such as when it may not be feasible to perform backflow prediction according to the more specific and / or improved techniques described herein.

[0169] As described above, the flowback monitoring system may include a flowback alarm manager. The flowback alarm manager may access and / or process some or all of the data in the data storage device (eg, threshold data) to detect abnormal active flowback and generate one or more warnings or alarms.

[0170] In some embodiments, the flowback alarm manager receives flowback measurements of active flowback from the sensor manager (e.g., accesses active flowback data). The flowback alarm manager can monitor active flowback, for example, against a predicted or expected active flowback represented by a best fit curve. For example, the flowback alarm manager can generate and / or monitor a volume flowback that varies over time for both the best fit curve and the active flowback. The flowback alarm manager can monitor active flowback relative to one or more of the flowback thresholds determined by the flowback threshold engine. For example, the volume flowback that varies over time can include an upper limit and / or a lower limit. In some embodiments, the flowback alarm manager plots, graphs, or otherwise displays the volume flowback that varies over time.

[0171] In another example, the flowback alarm manager can generate and / or monitor differential volume flowback over time. For example, the flowback alarm manager can generate and monitor differential active flowback, or the difference between active flowback and the best fit curve. The flowback alarm manager can monitor differential active flowback relative to one or more of the flowback thresholds determined by the flowback threshold engine. For example, the differential volume flowback over time can include an upper limit and / or a lower limit. In some embodiments, the flowback alarm manager plots, graphs, or otherwise displays the differential volume flowback over time.

[0172] Based on monitoring active backflow (e.g., via time-varying volume backflow and / or time-varying differential volume backflow), the backflow alarm manager may implement and / or generate one or more alarms. The alarm may be any type or form of alarm, warning, indication, or sign associated with the detection of abnormal backflow. For example, the alarm may be a visual or audible alarm or any other communication used to alert a user of abnormal backflow. The alarm may be a sign, annotation, record, file, or data entry (and combinations thereof) saved to a data storage device. The alarm may be implemented in conjunction with hardware and / or software of a computing device and / or backflow monitoring system. In this way, the backflow alarm manager may generate one or more forms of alarms to warn and / or record instances of abnormal active backflow.

[0173] As described above, the volume flowback over time and / or the differential volume flowback over time may include one or more of the flowback thresholds (e.g., including upper and / or lower limits) determined by the flowback threshold engine. The flowback thresholds may correspond to alarms, or the flowback alarm manager may trigger alarms based on the flowback thresholds.

[0174] In some cases, the differential active flowback may exceed or exceed one of the flowback thresholds, such as an upper limit, at a high alarm point. The flowback alarm manager may generate an alarm (e.g., a high alarm) based on the high alarm point accordingly. In some embodiments, the high alarm corresponds to a formation kick associated with the wellbore. In this way, the flowback alarm manager may generate an alarm and / or warn a user that the active flowback amount from the wellbore is excessive (e.g., a value that exceeds an expected value range).

[0175] Similarly, in some cases, the differential active flowback may exceed or exceed one or more other thresholds, such as a lower limit. The flowback alarm manager can generate an alarm (e.g., a low alarm) based on the low alarm point accordingly. In some embodiments, the low alarm corresponds to a formation loss associated with the wellbore. In this way, the flowback alarm manager can generate an alarm and / or warn a user of a decrease in active flowback from the wellbore (e.g., a value below an expected value range).

[0176] Furthermore, it should be appreciated that the flowback alarm manager may implement and / or generate one or more alarms associated with volumetric flowback that varies over time (e.g., associated with monitoring active flowback), such that the flowback alarm manager is not limited to the above-described implementation of differential volumetric flowback that varies over time. For example, the flowback alarm manager may monitor active flowback for volumetric flowback that varies over time, and may generate one or more alarms based on the active flowback exceeding or exceeding one or more of the flowback thresholds.

[0177] In this way, the flowback alarm manager can generate one or more alarms to indicate abnormal levels, amounts or rates of active flowback from the wellbore, such as excessive or reduced flowback compared to expected or predicted flowback. Such indications can be beneficial, for example, to detect abnormal wellbore conditions, such as abnormal or unexpected inflow of formation fluids into the wellbore (formation kick) or abnormal loss of drilling fluid to the formation (formation loss). In some embodiments, the flowback alarm manager responds to the alarm and takes one or more mitigation measures. For example, one or more drilling parameters can be adjusted, modified or changed (e.g., by the user or automatically by the flowback monitoring system) to alleviate the wellbore conditions associated with active flowback and / or alarms. This can help reduce the risk of serious, troublesome and / or catastrophic wellbore events, such as blowouts, formation / wellbore instability, valve damage or valve control problems, fluid circulation problems or other wellbore problems (and combinations thereof). In this way, the alarms generated by the flowback alarm manager can be beneficial to the efficient, effective and safe operation of the downhole system.

[0178] In some embodiments, the flowback alarm manager implements an alarm delay. The alarm delay can be a time period during which the flowback alarm manager will not or does not generate one or more alarms. For example, a differential active flowback (and / or active flowback) may exceed one or more thresholds during the time period of the alarm delay, and the flowback alarm manager may not generate a corresponding alarm. The alarm delay can be a period of time after the start of the pump shutdown period, or a period of time after receiving an indication of the pump shutdown period. For example, the alarm delay can be 10 seconds, 20 seconds, 30 seconds, 40 seconds, 50 seconds, 60 seconds, 90 seconds, 120 seconds, 240 seconds, 360 seconds or any value therebetween. In some embodiments, it is important that the alarm delay is at least 120 seconds after the start of the pump shutdown period to allow active flowback to be initialized and / or normalized, so that the flowback can be accurately measured and / or characterized.

[0179] In some embodiments, the flowback alarm manager cancels, deletes, or otherwise turns off one or more alarms. For example, the flowback alarm manager may cancel an alarm based on the differential active flowback weakening or returning to one or more of the flowback thresholds. For example, the flowback alarm manager may observe that the differential active flowback exceeds an upper limit (e.g., at a high alarm point), and the flowback alarm manager may therefore generate a high alarm. In some cases, the differential active flowback may weaken and / or return to within the upper limit. The flowback alarm manager may cancel or turn off the high alarm accordingly.

[0180] In another example, the flowback alarm manager may not cancel the alarm based on the differential active flowback returning to exceed the same flowback threshold that triggered the alarm, but may maintain the alarm until the observed flowback exceeds the additional flowback threshold. For example, a high alarm may be triggered at a high alarm point, and the differential active flowback may weaken or return to the upper limit at the return point. The flowback alarm manager may not cancel the high alarm at the return point, but may maintain the alarm until the differential active flowback returns to the alarm off threshold at the alarm off point. In some embodiments, the alarm off threshold is a narrower or tighter threshold (e.g., closer to the best fit curve) than the upper limit that triggers the alarm. For example, the upper limit can be three standard deviations from the best fit curve, and the alarm off threshold can be one standard deviation from the best fit curve. In this way, the flowback alarm manager can implement an alarm off threshold to prevent alarms indicating abnormal active flowback from being regularly triggered and canceled and / or ensure that the abnormal active flowback weakens to a sufficient degree so that the alarm can be canceled with confidence.

[0181] In some embodiments, the method or series of actions of analyzing the flowback of the downhole system includes an action of generating active flowback data by monitoring active flowback from the wellbore. For example, the flowback monitoring system can identify an active pump off period of the downhole system and can generate active flowback data corresponding to the active pump off period.

[0182] In some embodiments, the method includes an action of identifying historical flowback data of historical flowback from the wellbore. In some embodiments, the flowback monitoring system generates historical flowback data by monitoring historical flowback from the wellbore and caching the measurements as historical blowback data. The historical flowback data may correspond to one or more previous pump shut-down periods. For example, the historical data may be one or more instances of active flowback data cached after the corresponding active pump shut-down period ends. In some embodiments, the active flowback data is generated during the same operation of the downhole tool corresponding to the historical flowback data. For example, the cache of historical flowback data may be cleared based on the downhole tool being pulled out to the surface. In some embodiments, one or more data instances of historical flowback data are cleared from the cache based on a time interval or the age of the data instance.

[0183] In some embodiments, the method includes an action of determining a flowback cluster based on comparing the active flowback data to the historical flowback data. For example, the flowback monitoring system can identify one or more instances of historical flowback data having previous conditions similar to the conditions of the active flowback data. Similar previous conditions can include one or more of flow rate, depth of downhole tool, standpipe pressure, and flow pattern. In some embodiments, the flowback monitoring system can filter, clear, or delete one or more instances of historical flowback data based on the data quality of the one or more instances.

[0184] In some embodiments, the method includes an action of generating a flowback threshold based on the flowback cluster. For example, the flowback monitoring system can generate a flowback threshold with an upper limit and / or a lower limit. In some embodiments, the flowback monitoring system automatically generates the threshold in real time when an active pump off period is identified.

[0185] In some embodiments, the method includes an action of generating an alarm based on the active flowback data exceeding a flowback threshold. For example, the flowback monitoring system can generate an alarm based on the active flowback data exceeding an upper limit. In some embodiments, a drilling parameter of the downhole system can be adjusted based on the alarm. In some embodiments, the flowback monitoring system can cancel the alarm based on the active flowback data exceeding an alarm off threshold. The alarm off threshold can be different from the flowback threshold (e.g., different from the upper limit).

[0186] In some embodiments, a method or series of acts for analyzing flowback of a downhole system includes an act 90 of generating active flowback data by monitoring active flowback from a wellbore. In some embodiments, the method includes an act of identifying historical flowback data of historical flowback from the wellbore. In some embodiments, the method includes an act of determining a flowback cluster based on comparing the active flowback data to the historical flowback data.

[0187] In some embodiments, the method includes predicting the action of active flowback based on the flowback cluster. For example, the flowback monitoring system can generate a best fit curve based on the flowback cluster. The flowback monitoring system can calculate the probability distribution of the flowback cluster based on the best fit curve. The flowback monitoring system can generate a threshold value based on the probability distribution. The threshold value can be based on the standard deviation of the probability distribution. In some embodiments, the probability distribution is a Gaussian distribution.

[0188] In some embodiments, one or more computer systems may be used to implement the various devices, components, and systems described herein.

[0189] The computer system may include a processor respectively. The processor may be a general-purpose single-chip or multi-chip microprocessor (e.g., an advanced RISC (reduced instruction set computer) machine (ARM)), a special-purpose microprocessor (e.g., a digital signal processor (DSP)), a microcontroller, a programmable gate array, etc. The processor may be referred to as a central processing unit (CPU). Although only a single processor is described herein in conjunction with a computer system, in an alternative configuration, a combination of processors (e.g., an ARM and a DSP) may be used.

[0190] The computer system may also include a memory that is in electronic communication with the processor. The memory may be any electronic component capable of storing electronic information. For example, the memory may be embodied as a random access memory (RAM), a read-only memory (ROM), a disk storage medium, an optical storage medium, a flash memory device in a RAM, an onboard memory contained in a processor, an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM) memory, a register, etc., including combinations thereof.

[0191] Instructions and data may be stored in memory. Instructions may be executed by a processor to implement some or all of the functions disclosed herein. Executing instructions may involve using data stored in memory. Any of the various examples of modules and components described herein may be implemented in part or in whole as instructions stored in memory and executed by a processor. Any of the various examples of data described herein may be data stored in memory and used during execution of instructions by a processor.

[0192] One or more computer systems may also include one or more communication interfaces for communicating with other electronic devices. The communication interfaces may be based on wired communication technology, wireless communication technology, or both. Some examples of communication interfaces include Universal Serial Bus (USB), Ethernet adapters, wireless adapters operating according to the Institute of Electrical and Electronics Engineers (IEEE) 802.11 wireless communication protocol, Wireless communication adapter and infrared (IR) communication port.

[0193] One or more computer systems may also include one or more input devices and one or more output devices. Some examples of input devices include keyboards, mice, microphones, remote controls, buttons, joysticks, trackballs, touch pads, and light pens. Some examples of output devices include speakers and printers. A specific type of output device typically included in a computer system is a display device. The display device used with the embodiments disclosed herein may utilize any suitable image projection technology, such as a liquid crystal display (LCD), a light emitting diode (LED), a gas plasma, electroluminescence, and the like. A display controller may also be provided for converting data stored in a memory into text, graphics, and / or moving images (as appropriate) shown on a display device.

[0194] The various components of the computer system can be connected together through one or more buses, which may include a power bus, a control signal bus, a status signal bus, a data bus, etc.

[0195] The techniques described herein may be implemented in hardware, software, firmware, or any combination thereof, unless specifically described as being implemented in a particular manner. Any features described as modules, components, or the like may also be implemented together in an integrated logic device or separately as discrete but interoperable logic devices. If implemented in software, the techniques may be implemented at least in part by a non-transitory processor-readable storage medium comprising instructions that, when executed by at least one processor, perform one or more of the methods described herein. Instructions may be organized into routines, programs, objects, components, data structures, and the like that may perform specific tasks and / or implement specific data types, and may be combined or distributed as desired in various embodiments.

[0196] Computer-readable media can be any available media that can be accessed by a general or special purpose computer system. A computer-readable medium that stores computer-executable instructions is a non-transitory computer-readable storage medium (device). A computer-readable medium that carries computer-executable instructions is a transmission medium. Therefore, by way of example and not limitation, embodiments of the present disclosure may include at least two distinct computer-readable media: a non-transitory computer-readable storage medium (device) and a transmission medium.

[0197] The embodiments of the flowback monitoring system are described primarily with reference to wellbore drilling operations. The flowback monitoring system described herein can be used for applications other than drilling wellbores. In other embodiments, the flowback monitoring system according to the present disclosure can be used outside of a wellbore or other downhole environment for exploring or producing natural resources. For example, the flowback monitoring system of the present disclosure can be used in a borehole for placing a utility pipeline. Therefore, the terms "wellbore", "borehole", etc. should not be interpreted as limiting the tools, systems, assemblies or methods of the present disclosure to any particular industry, field or environment.

[0198] One or more specific embodiments of the present disclosure are described herein. These described embodiments are examples of currently disclosed technologies. In addition, in order to provide a brief description of these embodiments, all features of the actual embodiments may not be described in the specification. It should be understood that when developing any such actual implementation in any engineering or design project, numerous implementation-specific decisions will be made to achieve the developer's specific goals, such as meeting system-related and business-related constraints, which may vary between different implementations. In addition, it should be understood that such development work may be complex and time-consuming, but for ordinary technicians who benefit from the present disclosure, this will still be a routine task in design, production and manufacturing.

[0199] In addition, it should be understood that reference to "one embodiment" or "embodiment" of the present disclosure is not intended to be interpreted as excluding the existence of additional embodiments that also incorporate the features. For example, any element described with respect to the embodiments herein may be combined with any element of any other embodiment described herein. As will be understood by those of ordinary skill in the art encompassed by the embodiments of the present disclosure, the numerals, percentages, ratios, or other values ​​stated herein are intended to include the value, as well as other values ​​that are "about" or "approximately" the value. Therefore, the value should be interpreted broadly enough to cover values ​​that are at least close enough to the value to perform the desired function or achieve the desired result. The value includes at least the variation expected in a suitable manufacturing or production process, and may include values ​​within 5%, within 1%, within 0.1%, or within 0.01% of the value.

[0200] In view of the present disclosure, those skilled in the art will recognize that equivalent constructions do not depart from the spirit and scope of the present disclosure, and that various changes, substitutions, and modifications may be made to the embodiments disclosed herein without departing from the spirit and scope of the present disclosure. Equivalent constructions (including functional "device plus function" clauses) are intended to cover structures described herein as performing the functions, including structural equivalents that operate in the same manner and equivalent structures that provide the same functions. It is the clear intention of the applicant not to invoke device plus function or other functional requirements for any claim, except for those claims in which the term "device for..." appears together with the associated function. Each addition, deletion, and modification to the embodiments that falls within the meaning and scope of the claims will be included in the claims.

[0201] As used herein, the terms "approximately," "about," and "substantially" refer to an amount that is within standard manufacturing or process tolerances or that is close to a stated amount while still performing a desired function or achieving a desired result. For example, the terms "approximately," "about," and "substantially" may refer to an amount that is within a range of less than 5%, less than 1%, less than 0.1%, and less than 0.01% of the stated amount. Furthermore, it should be understood that any directions or reference frames in the foregoing descriptions are only relative directions or movements. For example, any reference to "up" and "down" or "above" or "below" only describes the relative position or movement of the relevant elements.

[0202] Various features are described herein in alternative formats in order to emphasize that features can be combined in any number of combinations. Each feature should be considered combinable with other features unless those features are mutually exclusive. The word "or" as used herein is not exclusive unless expressly stated otherwise. For example, having A or B includes only A, only B, or a combination of A and B. Conversely, having only A or B includes only A or only B, but not a combination of A or B. The specification provides support for each claim to be combined with other claims (or any combination of other claims), even if not explicitly stated in multiple independent forms.

[0203] Without departing from the spirit or characteristics of the present disclosure, the present disclosure may be embodied in other specific forms. The described embodiments should be considered to be illustrative only and not restrictive. Therefore, the scope of the present disclosure is indicated by the appended claims rather than by the foregoing description. Changes in the meaning and scope of the equivalent forms of the claims will be included in the scope of the claims.

Claims

1. A method for analyzing flowback of a downhole system, comprising: generating active flowback data by monitoring active flowback from the wellbore; historical flowback data identifying historical flowback from the wellbore; determining a flowback cluster based on comparing the active flowback data with the historical flowback data; generating a flowback threshold based on the flowback cluster; as well as An alarm is generated based on the active flowback data exceeding the flowback threshold. 2 . The method of claim 1 , further comprising identifying an active pump-off period of the downhole system, wherein the active flowback data corresponds to the active pump-off period.

3. The method of claim 2, wherein the historical flowback data corresponds to at least one instance of a previous pump shut-off period.

4. The method of claim 2, wherein the active flowback data is generated during a same operation of the downhole tool as the historical flowback data. 5 . The method of claim 1 , wherein determining the flowback cluster comprises identifying at least one instance of the historical flowback data having a similar prior condition as the active flowback data.

6. The method of claim 5, wherein the similar prior conditions include at least one of: flow rate, depth of the downhole tool, standpipe pressure, and flow pattern.

7. The method of claim 1, wherein the flowback threshold comprises an upper limit, and generating the alarm is based on the active flowback data exceeding the upper limit.

8. The method of claim 1, further comprising adjusting a parameter of the downhole system based on the alarm.

9. The method of claim 1, wherein determining the flowback cluster comprises filtering at least one instance of the historical flowback data based on a data quality of the at least one instance.

10. The method of claim 1, wherein identifying the historical flowback data comprises generating and buffering the historical flowback data by monitoring historical flowback from the wellbore.

11. The method of claim 1, further comprising clearing the cache of historical flowback data based on pulling a downhole tool from the wellbore.

12. The method of claim 1, further comprising clearing at least one instance of the historical flowback data from a data cache of the historical flowback data based on a time interval.

13. The method of claim 1, further comprising canceling the alarm based on the active flowback data exceeding an alarm off threshold that is different than the flowback threshold.

14. The method of claim 1, wherein the flowback threshold is automatically generated in real time upon identification of an active pump off period.

15. A method for analyzing flowback of a downhole system, comprising: generating active flowback data by monitoring active flowback from the wellbore; historical flowback data identifying historical flowback from the wellbore; determining a flowback cluster based on comparing the active flowback data with the historical flowback data; and The active flowback is predicted based on the flowback cluster.

16. The method of claim 15, wherein predicting the active flowback comprises generating a best fit curve based on the flowback clusters. 17 . The method of claim 16 , wherein predicting the active flowback comprises calculating a probability distribution of the flowback clusters based on the best fit curve.

18. The method of claim 17, wherein predicting the active flowback comprises generating a threshold based on a standard deviation of the probability distribution. The method of claim 17 , wherein the probability distribution is a Gaussian distribution.

20. A system comprising: at least one processor; a memory in electronic communication with the at least one processor; as well as instructions stored in the memory, the instructions executable by the at least one processor to: generating active flowback data by monitoring active flowback from the wellbore; historical flowback data identifying historical flowback from the wellbore; determining a flowback cluster based on comparing the active flowback data with the historical flowback data; generating a flowback threshold based on the flowback cluster; as well as An alarm is generated based on the active flowback data exceeding the flowback threshold.

Images