Friction reducer performance analysis in well systems

Abstract

Techniques for analyzing the performance of a friction reducer include receiving a pressure pulse signal associated with a well system, determining a friction coefficient corresponding to the performance of the friction reducer in the well system, and performing one or more operations based, at least in part, on the friction coefficient.

Description

2024-INV-112633-WO01FRICTION REDUCER PERFORMANCE ANALYSIS IN WELL SYSTEMSBACKGROUND

[0001] Well system operations use fluid for various purposes, including drilling fluid to help transport material out of a wellbore and hydraulic fracturing fluid to fracture gas formations. Friction reducers are added to fluids used in well systems to reduce the friction of the fluids and thereby improve their flow. Friction reducer performance can be analyzed in the laboratory but laboratory analyses may not be reflective of real-world performance.BRIEF DESCRIPTION OF THE DRAWINGS

[0002] Embodiments of the disclosure may be better understood by referencing the accompanying drawings.

[0003] Figure 1 depicts a graph of an example pressure pulse, according to some implementations.

[0004] Figure 2 is a diagrammatic illustration of an example well system with multiple pressure gauges along the wellbore, according to some implementations.

[0005] Figure 3 depicts an example system for determining friction reducer performance, according to some implementations.

[0006] Figure 4 depicts an example system for training a machine learning model to predict friction reducer performance, according to some implementations.

[0007] Figure 5 depicts an example system for using a machine learning model to predict friction reducer performance, according to some implementations.

[0008] Figure 6 is a flowchart of operations for determining friction reducer performance and modifying well system operational characteristics and conditions, according to some implementations.

[0009] Figure 7 is a flowchart of operations for training a machine learning model to predict friction reducer performance, according to some implementations.

[0010] Figure 8 is a flowchart of operations for predicting friction coefficient performance using a machine learning model, according to some implementations.2024-INV-112633-WO01

[0011] Figure 9 is a diagrammatic illustration of an example well system, according to some implementations.

[0012] Figure 10 is a block diagram depicting an example computing system, according to some implementations.DESCRIPTION

[0013] The description that follows includes example systems, methods, techniques, and program flows that embody aspects of the disclosure. However, it is understood that this disclosure may be practiced without these specific details. In some instances, well-known instruction instances, protocols, structures, and techniques have not been shown in detail in order not to obfuscate the description.

[0014] While the performance of friction reducing agents (“friction reducers’’) can be measured in the laboratory, conditions in well systems can vary significantly from the conditions in the laboratory. Thus, the real-world performance of friction reducers may not match the performance demonstrated in the laboratory. However, pressure pulse technology can be utilized to assess the performance of friction reducers during use in an actual well system, thus allowing adjustment of the friction reducer(s) in the well system for improved efficiency. The amount, ty pe, or concentration of the friction reducer can be changed to decrease costs or improve fluid flow based on the assessed performance of the friction reducer.

[0015] Pressure pulses occur when changes in well system operations result in a change in the pressure of the fluid being pumped into the well system. Pressure pulses can be triggered using a variety of means and may result from normal changes to the operation conditions, such as the ramp down process.

[0016] Ramp dow n involves a process where the pumping rate is changed from a steady state to a new, lower pumping rate in a very short duration. This operational change results in a velocity change at the wellbore inlet, generating a pressure pulse that travels along the wellbore and reflects back to the inlet. This pressure pulse may continue traveling up and down the w ellbore until pressure has equalized throughout the wellbore.

[0017] The pressure pulse can be measured at the surface using existing pressure gauges, transducers, or other sensors fluidly coupled with the fluid in the wellbore. The pressure pulse is ty pically measured by sampling the fluid pressure at periodic intervals with a frequency greater than that of pressure pulse itself. The pressure pulse readings may be taken manually (e.g.,2024-INV-112633-WO01viewed by an operator and recorded manually) or automatically (e.g., via a transducer that converts the pressure pulse into an electrical signal that is recorded to a storage device).

[0018] Figure 1 depicts a graph of an example pressure pulse, according to some implementations. In particular, Figure 1 depicts an example graph 102 showing the pressure changes in the fluid over time. At Po, the pressure is at the original steady state flowrate QO. At Pl, the pressure pulse is triggered (e g., at the start of the ramp down procedure), dropping the flowrate to Q such that At? = Qo— Q. As the pressure pulse travels down the w ellbore, the flowrate decreases further to Q — e. where e is the loss due to friction.

[0019] The surface wellbore pressure P(t) for this period can be written as the summation of the bottom hole pressure PBH. the friction pressure Pf, and the hydrostatic pressure Ph as shown in Equation 1, where a is the wave speed, t is the time, L is the length of wellbore up to current treatment stage and e is the change in the flowrate due to frictional effect: / at\ fat\P(t) = PBH+ Pf(Q + A(?) (1 - y) + PfCQ ~ e) [-) ~ PhEquation 1

[0020] Assuming the bottom hole pressure remains constant and no change in fluid density occurs, the change in the pressure over time can be written as Equation 2:dP az.a7 = t(W-0-W + i<?))Equation 2

[0021] The frictional pressure loss can be written as Equation 3, where Q is the flow rate, / is the friction coefficient, and k is the constant shown in Equation 4.Pf= PfQ2Equation 3k =^Ln2D5Equation 4

[0022] Thus, the change in the pressure over time can be written as Equation 5:2024-INV-112633-WO01dP akf >-^ = -^-(e2- W2- 2Q(e + AQy)Equation 5

[0023] The pressure loss due to friction is typically significantly less than the original drop in the flowrate (AQ), so Equation 5 can be simplified to Equation 6:dP akfLQ(2Q + A(?)dt LEquation 6

[0024] The friction coefficient can be found by solving Equation 6 for f.L dPf~ ~ ak& Q(2Q + bQ^dtEquation 7

[0025] Further, in some scenarios, AQ is significantly less than 2Q, allowing Equation 7 to be rewritten as Equation 8:L dPf = - ’ 2akQ Q dtEquation 8

[0026] Updating Equation 8 to include the flowrate at P3, which includes both the effect of friction and the boundary resistance, yields Equation 9:L dP ( 1 12ak& Q ~dt K2Q +& Q+2Q - AQEquation 9

[0027] However, in scenarios where AQ is significantly less than 2Q, Equation 9 simplifies back to Equation 8. If sufficiently precise measurement of the flowrate is available, the boundary resistance can be calculated.

[0028] In scenarios where estimations of the time derivative of the pressure is difficult (e.g., due to insufficient pressure measurement resolution), the pressure values at Pi and P2 can be used. In particular, Pi and P2can be represented by Equation 10 and Equation 11, respectively:2024-INV-112633-WO01Pl = PBH + Pf(Q + A(?) - PhEquation 10P2= PBH + Pf(Q ~ PhEquation 11

[0029] Subtracting P2 from Pi yields Equation 12:Pi ~ P2= Pf(. Q + A<?) - PfQ)

[0030] Using Taylor series approximation yields Equation 12 and using the definition of friction pressure yields Equation 13:dPfPf(Q+AQ) = Pf(Q) + AQ-^- + - Equation 12Pf(Q + AQ) = + 2kfQAQ + - Equation 13

[0031] Thus, the friction coefficient can be determined by Equation 14:Pi - P2' 2kfQAQEquation 14

[0032] Other friction pressure models can be utilized in addition to, or in place of, the model described above.

[0033] In addition to the model described above, the friction coefficient can be computed directly if the wellbore has an additional pressure gauge located along the length of the wellbore.

[0034] Figure 2 is a diagrammatic illustration of an example well system with multiple pressure gauges along the wellbore, according to some implementations. In particular, Figure 2 depicts an example well system 200, including a wellbore 202, wellbore perforations 204, a surface-level pressure gauge 206, and a bottom hole pressure gauge 208.2024-INV-112633-WO01

[0035] The frictional pressure loss can be measured by measuring the pressure at the surface using the surface-level pressure gauge 206 and measuring the pressure at the surface using the bottom hole pressure gauge 208 and using Equation 15:Ps ~ PBH + Ph ~ PfEquation 15

[0036] Applying the definition of Pf from Equation 3 and solving for the coefficient of friction yields Equation 16:„ _ Ps ~ PBH + Phr~ kQ2Equation 16

[0037] The pressure pulse can then be used to calculate the coefficient of friction directly.Machine Learning-based Prediction of Friction Reducer Performance

[0038] Machine learning can also be used to estimate the performance of friction reducers based on well system operational characteristics and conditions.

[0039] To develop a machine learning model for estimating the performance of friction reducers, a machine learning model is trained on training data. The training data consists of data samples, with each data sample consisting of a number of features and a target. The features correspond to well system operational characteristics and conditions, such as the pressure pulse signature, fluid conditions (e.g., pump rate, friction reducer concentration, proppant concentration, etc.), geometric conditions (e.g., length of the wellbore, diameter of the wellbore, etc.), and any other relevant details. The target is the friction coefficient (or a representation thereof) measured under the operational characteristics and conditions corresponding to the features.

[0040] Once trained on the data samples, the machine learning model can generate predictions of the friction coefficient based on the operational characteristics and conditions of other well systems.2024-INV-112633-WO01Application to Well System Operations

[0041] Knowledge about friction reducer performance can be applied to well system operations in a number of ways. First, for existing well systems, a determination that the friction reducer is not performing as expected can result in a change to the friction reducer amount, type, concentration, etc. Second, for well systems being designed, a prediction (e.g., from a machine learning model) about how the friction reducer will perform under the expected operational characteristics and conditions may result in changes to the design (including the initial friction reducer amount, type, concentration, etc ).

[0042] Further, predictions about how the friction reducer will perform can be applied to existing well systems as well. For example, the operational characteristics and conditions of the next stage of a well system may be provided as input to a machine learning model and the friction reducer amount, type, concentration, etc. may be adjusted in anticipation of the next stage.

[0043] Thus, in some implementations, a system operation or attribute in the wellbore or other system component may be modified or updated based on a determination of the friction reducer performance. For example, an operation (at the surface, downhole, within a pipeline, etc.) may be performed and / or directed to be performed to change a system operation or attribute based on the measured or predicted performance of the friction reducer. For example, attributes of an actual drilling or extraction operation in the wellbore may be set based on the measured or predicted performance of the friction reducer. Examples of such attributes of the drilling / extraction operation may include drilling fluid / mud density, hydraulic pressure, friction reducer amount, friction reducer concentration, friction reducer type, etc. For further example, operations in a wellbore or other system component that may be performed or modified in response to the measured or predicted performance of the friction reducer, such as increasing or reducing the amount of friction reducer added to the fluidExample System for Determining Friction Reducer Performance

[0044] Figure 3 depicts an example system for determining friction reducer performance, according to some implementations. In particular, Figure 3 depicts a friction reducer performance analysis system 300, including a friction reducer analysis module 304, and an operations analysis module 308.2024-INV-112633-WO01

[0045] In operation, the friction reducer analysis module 304 receives a pressure pulse signal 302 corresponding to a pressure pulse generated within a well system. The friction reducer analysis module 304 can determine the friction reducer performance by utilizing the techniques described herein. For example, if the sampling frequency of the pressure pulse signal 302 is sufficiently high, the friction reducer analysis module 304 may use the pressure pulse signal 302 and Equation 8 to determine the coefficient of friction resulting from the friction reducers in use. If the sampling frequency of the pressure pulse signal 302 is not high enough, the friction reducer analysis module 304 may use the pressure pulse signal 302 and Equation 14 to determine the coefficient of friction resulting from the friction reducers in use.

[0046] After the coefficient of friction is determined, the operations analysis module 308 can identify one or more operational changes 310 to be made. In particular, the operations analysis module 308 can determine whether the friction reducer is performing the same, better, or worse than expected and can determine whether one or more operational changes should be made in response. For example, if the operations analysis module 308 determines that the friction reducer is performing better than expected, the operations analysis module 308 may determine the amount of friction reducer being used can be reduced.Example System for Predicting Friction Reducer Performance

[0047] Figure 4 depicts an example system for training a machine learning model to predict friction reducer performance, according to some implementations. In particular, Figure 4 depicts a machine learning model training system 400, including a machine learning training module 404.

[0048] In operation, the machine learning model training system 400 receives the training data 402. The training data 402 comprises a plurality of samples, wherein each sample includes a set of one or more features and a target. The set of one or more features for each sample includes operational characteristics and conditions associated with a particular well system and the target is the friction coefficient of that well system.

[0049] The machine learning model training system 400 then trains a machine learning model based, at least in part, on the sample data, producing a machine learning model 406.2024-INV-112633-WO01

[0050] The operations described in relation to Figure 4 can be performed periodically, e.g., when new sample data has been collected. Such a pattern produces an updated machine learning model that takes into account more recent data.

[0051] Figure 5 depicts an example system for using a machine learning model to predict friction reducer performance, according to some implementations. In particular, Figure 5 depicts a machine learning model prediction system 500, including a machine learning analysis module 506.

[0052] In operation, the machine learning analysis module receives operational characteristics and conditions 502 associated with a particular well system and a trained machine learning model 504. The operational characteristics and conditions 502 may be for an operational well system or for a well system being designed or developed and ty pically mirror the operational characteristics and conditions used to train the machine learning model 504.

[0053] The machine learning analysis module 506 provides the operational characteristics and conditions 502 as input to the machine learning model 504 and uses the output of the machine learning model 504 to generate the predicted friction coefficient 508. The machine learning analysis module 506 may use the output from the machine learning model 504 directly or may perform additional operations to transform the output from the machine learning model 504 into the predicted friction coefficient.Example Operations for Determining and Using Friction Reducer Performance

[0054] Figure 6 is a flowchart of operations for determining friction reducer performance and modifying well system operational characteristics and conditions, according to some implementations. In particular, Figure 6 depicts a flowchart 600 of operations that begin at block 602. The operations depicted in Figure 6 may be performed by the friction reducer performance analysis system 300 of Figure 3 or any suitable system.

[0055] At block 602, it is determined whether friction reducer performance should be analyzed. There are a variety of reasons that the friction reducer performance should be analyzed or re-analyzed. For example, a change in the components of the fluid being pumped into the well system, such as a change to the friction reducer, sand, etc., might trigger an analysis of the friction reducer performance.2024-INV-112633-WO01

[0056] As another example, an analysis of the friction reducer performance may be triggered if it has been determined that the friction coefficient may be incorrect. In particular, if the friction coefficient is known, the bottom hole pressure can be estimated using Equation 17:PBH = Ps + Pf ) ~ PhEquation 17

[0057] The surface pressure can then be estimated using the bottom hole pressure, using Equation 18:Ps,e=PBH ~ Pf(J) + PhEquation 18

[0058] Any difference in the measured surface pressure and the estimated surface pressure may be indicative of a change in the friction coefficient and thus may trigger reanalysis of the friction reducer performance.

[0059] If it is determined that the friction reducer performance should be analyzed, control then flows to block 604. If it is determined that the friction reducer performance should not be analyzed, control then flows to block 616.

[0060] At block 604, a pressure pulse in the well system is triggered. The pressure pulse may be triggered using any appropriate means, including lowering the pumping rate, closing one or more valves, etc.

[0061] At block 606, the pressure pulse signal is generated. The pressure pulse signal can be generated using any appropriate means, such as recording the pressure displayed on a pressure gauge, using a transducer to turn the pressure pulse into an electrical signal, etc.

[0062] At block 608, the friction coefficient is determined based on the appropriate model. For example, if the sampling frequency of the pressure pulse signal is sufficiently high, Equation 8 may be used to determine the coefficient of friction resulting from the friction reducers in use. If the sampling frequency of the pressure pulse signal is not high enough, Equation 14 may be used to determine the coefficient of friction resulting from the friction reducers in use.

[0063] At block 610. it is determined whether operational conditions or characteristics should be changed. For example, if the friction reducer is not performing as expected (e.g., better or w orse), the amount, ty pe, concentration, etc. of the friction reducer may be modified such that2024-INV-112633-WO01the friction reducer performance meets expectations. If it is determined that operational conditions or characteristics should be changed, control then flows to block 612. If it is determined that operational conditions or characteristics should not be changed, control then flows to block 614.

[0064] At block 612. operational conditions or characteristics are modified according to the determination at block 610.

[0065] At block 614, it is determined whether to continue analyzing the performance of the friction reducer. Analysis of the friction reducer performance may be halted based on a variety of conditions. For example, if well operations have completed, additional analysis may not be needed. Similarly, if the friction reducer performance has maintained a consistent level for a sufficient period of time, additional analysis may not be needed. If it is determined that analysis of the friction reducer performance should be continued, control then flows back to block 602. If it is determined that analysis of the friction reducer performance should not be continued, the process ends.Example Operations for Predicting Friction Reducer Performance

[0052] Figure 7 is a flowchart of operations for training a machine learning model to predict friction reducer performance, according to some implementations. In particular. Figure 7 depicts a flowchart 700 of operations that begin at block 702. The operations depicted in Figure 7 may be performed by the machine learning model training system 400 of Figure 4 or any suitable system.

[0066] At block 702, training data is received. The training data comprises a plurality of samples, wherein each sample includes a set of one or more features and a target. The set of one or more features for each sample includes operational characteristics and conditions associated with a particular well system and the target is the friction coefficient of that well system.

[0067] At block 704, a machine learning model is trained based on the training data. The particular operations performed to train the machine learning model may vary depending on the training data, the type of machine learning model, etc.2024-INV-112633-WO01

[0068] At block 706. the machine learning model is persisted. The particular operations performed to persist the machine learning model may vary. The machine learning model may be persisted on any suitable storage medium, including cloud-based storage.

[0069] Figure 8 is a flowchart of operations for predicting friction coefficient performance using a machine learning model, according to some implementations. In particular. Figure 8 depicts a flowchart 800 of operations that begin at block 802. The operations depicted in Figure 8 may be performed by the machine learning model prediction system 500 of Figure 5 or any suitable system.

[0070] At block 802, well system operational characteristics and conditions are received. The operational characteristics and conditions includes operational characteristics and conditions associated with a particular well system and may mirror the operational characteristics and conditions used to train the machine learning model.

[0071] At block 804, a machine learning model is executed using the well system operational characteristics and conditions as input. The machine learning model produces an output based on the well system operational characteristics and conditions.

[0072] At block 806, a predicted friction coefficient is generated. The predicted friction coefficient may be the output from the machine learning model produced at block 804 or may be a transformation thereof.Example Well System

[0073] Figure 9 is a diagrammatic illustration of an example well system, according to some implementations. In particular, Figure 9 depicts a well system 900, including a wellbore 902 located within a formation 901, a tubular string 906 within the wellbore 902, a wellhead 910, a fluid source 912, and a friction reducer source 914. The well system 900 further includes a pressure sensor 916 and a pressure gauge 918.

[0074] In operation, the fluid from the fluid source 912 is pumped into the wellbore 902 via the wellhead 910 and the tubular string 906. The pumps used to pump the fluid may be located in the wellhead 910 or elsewhere. Friction reducer from the friction reducer source 914 is mixed with the fluid from the fluid source 912 prior to being pumped into the wellbore 902.2024-INV-112633-WO01

[0075] When a pressure pulse is generated, the pressure sensor 916 detects the pressure pulse and displays the pressure on the pressure gauge 918. In some implementations, the pressure sensor 916 is a transducer and the pressure gauge 918 samples the pressure sensor 916 readings and generates a pressure pulse signals.Example Computing Systems

[0076] Figure 10 is a block diagram depicting an example computing system, according to some implementations. Figure 10 depicts a computing system 1000 for analyzing the performance of friction reducer. The computing system 1000 includes a processor 1001 (possibly including multiple processors, multiple cores, multiple nodes, and / or implementing multithreading, etc.). The computing system 1000 also includes a friction reducer performance analysis module 1015. The friction reducer performance analysis module 1015 may perform the operations described herein. For example, the friction reducer performance analysis module 1015 may analyze a pressure pulse signal to determine a friction coefficient representing the performance of the friction reducer. Any one of the previously described functionalities may be partially (or entirely) implemented in hardware and / or on the friction reducer performance analysis module 1015. For example, the functionality may be implemented with an application specific integrated circuit, in logic implemented in the friction reducer performance analysis module 1015, in a co-processor on a peripheral device or card, etc. Further, realizations may include fewer or additional components not illustrated in Figure 10 (e.g., video cards, audio cards, additional network interfaces, peripheral devices, etc.). The processor 1001 and the network interface 1005 are coupled to the bus 1003. Although illustrated as being coupled to the bus 1003, the memory 1007 may be coupled to the processor 1001. The computing system 1000 includes memory 1007. The memory 1007 may be system memory or any one or more possible realizations of machine-readable media. The computing system 1000 can communicate via transmissions to and / or from remote devices via the network interface 1005 in accordance with a network protocol corresponding to the type of network interface, whether wired or wireless and depending upon the carrying medium. In addition, a communication or transmission can involve other layers of a communication protocol and or communication protocol suites (e.g., transmission control protocol, Internet Protocol, user datagram protocol, virtual private network protocols, etc.).2024-INV-112633-WO01

[0077] While the aspects of the disclosure are described with reference to various implementations and exploitations, it will be understood that these aspects are illustrative and that the scope of the claims is not limited to them. In general, techniques for analyzing friction reducer performance as described herein may be implemented with facilities consistent with any hardware system or hardware systems. Many variations, modifications, additions, and improvements are possible.

[0078] Plural instances may be provided for components, operations or structures described herein as a single instance. Further, boundaries between various components, operations and data stores are somewhat arbitrary, and particular operations are illustrated in the context of specific illustrative configurations. Other allocations of functionality are envisioned and may fall within the scope of the disclosure. In general, structures and functionality presented as separate components in the example configurations may be implemented as a combined structure or component. Similarly, structures and functionality presented as a single component may be implemented as separate components. These and other variations, modifications, additions, and improvements may fall within the scope of the disclosure.

[0079] The flowcharts are provided to aid in understanding the illustrations and are not to be used to limit the scope of the claims. The flowcharts depict example operations that can vary within the scope of the claims. Additional operations may be performed; fewer operations may be performed; the operations may be performed in parallel; and the operations may be performed in a different order. It will be understood that each block of the flow chart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by program code. The program code may be provided to a processor of a general-purpose computer, special purpose computer, or other programmable machine or apparatus.

[0080] Use of the phrase “at least one of’ preceding a list with the conjunction “and” should not be treated as an exclusive list and should not be construed as a list of categories with one item from each category, unless specifically stated otherwise. A clause that recites “at least one of A, B, and C” can be infringed with only one of the listed items, multiple of the listed items, and one or more of the items in the list and another item not listed.

[0081] As used herein, the term “or” is inclusive unless otherwise explicitly noted. Thus, the phrase “at least one of A, B, or C” is satisfied by any element from the set {A, B. C} or any combination thereof, including multiples of any element.2024-INV-112633-WO01Example Implementations

[0082] Implementation 1: A method for analyzing performance of a friction reducer, the method comprising: receiving a pressure pulse signal associated with a well system; determining a friction coefficient corresponding to the performance of the friction reducer in the well system; and performing one or more operations based, at least in part, on the friction coefficient.

[0083] Implementation 2: The method according to Implementation 1, wherein at least one of a downhole operation or a downhole attribute in a wellbore of the well system is modified based, at least in part, on the friction coefficient.

[0084] Implementation 3: The method according to any of the preceding implementations, further comprising directing an operation to modify at least one of a downhole operation or a downhole attribute in a wellbore of the well system based, at least in part, on the friction coefficient.

[0085] Implementation 4: The method according to any of the preceding implementations, further comprising modifying at least one of a downhole operation or a downhole attribute in a wellbore of the well system based, at least in part, on the friction coefficient.

[0086] Implementation 5: The method according to any of the preceding implementations, wherein said performing the one or more operations based, at least in part, on the friction coefficient comprises modifying one or more operational characteristics or conditions associated with the well system.

[0087] Implementation 6: The method according to Implementation 5, wherein the one or more operational characteristics or conditions comprises at least one of a friction reducer amount, a friction reducer type, or a friction reducer concentration.

[0088] Implementation 7: The method according to any of the preceding Implementations, wherein said determining the friction coefficient comprises determining the friction coefficient based, at least in part, on a machine learning model trained on well system operational characteristics and conditions.

[0089] Implementation 8: A system comprising a computing system comprising one or more processors and one or more non-transitory computer-readable mediums including instructions2024-INV-112633-WO01which, when executed by the one or more processors, cause the one or more processors to execute one or more operations for analyzing performance of a friction reducer, the instructions including: instructions to receive a pressure pulse signal associated with a well system; instructions to determine a friction coefficient corresponding to the performance of the friction reducer in the well system; and instructions to perform one or more operations based, at least in part, on the friction coefficient.

[0090] Implementation 9: The system according to Implementation 8, wherein at least one of a downhole operation or a downhole attribute in a wellbore of the well system is modified based, at least in part, on the friction coefficient.

[0091] Implementation 10: The system according to any of the preceding Implementations, wherein the instructions further include instructions to direct an operation to modify at least one of a downhole operation or a downhole attribute in a wellbore of the well system based, at least in part, on the friction coefficient.

[0092] Implementation 11: The system according to any of the preceding Implementations, wherein the instructions further include instructions to modify at least one of a downhole operation or a downhole attribute based, at least in part, on the friction coefficient.

[0093] Implementation 12: The system according to any of the preceding Implementations, wherein the instructions to perform one or more operations based, at least in part, on the friction coefficient comprises instructions to modify one or more operational characteristics or conditions associated with the well system.

[0094] Implementation 13: The system according to Implementation 12, wherein the one or more operational characteristics or conditions comprises at least one of a friction reducer amount, a friction reducer type, or a friction reducer concentration.

[0095] Implementation 14: The system according to any of the preceding Implementations, wherein the instructions to determine the friction coefficient comprises instructions to determine the friction coefficient based, at least in part, on a machine learning model trained on well system operational characteristics and conditions.

[0096] Implementation 15: One or more non-transitory computer-readable mediums including instructions which, when executed by a processor, cause the processor to execute one or more operations for analyzing performance of a friction reducer, the instructions comprising: instructions to receive a pressure pulse signal associated with a well system; instructions to2024-INV-112633-WO01determine a friction coefficient corresponding to the performance of the friction reducer in the well system; and instructions to perform one or more operations based, at least in part, on the friction coefficient.

[0097] Implementation 16: The one or more non-transitory computer-readable mediums of Implementation 15, wherein at least one of a downhole operation or a downhole attribute in a wellbore of the well system is modified based, at least in part, on the friction coefficient.

[0098] Implementation 17: The one or more non-transitory computer-readable mediums of any of the preceding Implementations, wherein the instructions further comprise instructions to direct an operation to modify at least one of a downhole operation or a downhole attribute in a wellbore of the well system based, at least in part, on the friction coefficient.

[0099] Implementation 18: The one or more non-transitory computer-readable mediums of any of the preceding Implementations, wherein the instructions further include instructions to modify at least one of a downhole operation or a downhole attribute based, at least in part, on the friction coefficient.

[0100] Implementation 19: The one or more non-transitory computer-readable mediums of any of the preceding Implementations, wherein the instructions to perform one or more operations based, at least in part, on the friction coefficient comprises instructions to modify one or more operational characteristics or conditions associated with the well system, wherein the one or more operational characteristics or conditions comprises at least one of a friction reducer amount, a friction reducer type, or a friction reducer concentration.

[0101] Implementation 20: The one or more non-transitory computer-readable mediums of any of the preceding Implementations, wherein the instructions to determine the friction coefficient comprises instructions to determine the friction coefficient based, at least in part, on a machine learning model trained on well system operational characteristics and conditions.

Claims

1. 2024-INV-112633-WO01CLAIMS1. A method for analyzing performance of a friction reducer, the method comprising: receiving a pressure pulse signal associated with a well system;determining a friction coefficient corresponding to the performance of the friction reducer in the well system; andperforming one or more operations based, at least in part, on the friction coefficient.

2. The method of claim 1. wherein at least one of a downhole operation or a downhole attribute in a wellbore of the well system is modified based, at least in part, on the friction coefficient.

3. The method of claim 1. further comprising directing an operation to modify at least one of a downhole operation or a downhole attribute in a wellbore of the well system based, at least in part, on the friction coefficient.

4. The method of claim 1. further comprising modifying at least one of a downhole operation or a downhole attribute in a wellbore of the well system based, at least in part, on the friction coefficient.

5. The method of claim 1, wherein said performing the one or more operations based, at least in part, on the friction coefficient comprises modifying one or more operational characteristics or conditions associated with the well system.

6. The method of claim 5, wherein the one or more operational characteristics or conditions comprises at least one of a friction reducer amount, a friction reducer type, or a friction reducer concentration.

7. The method of claim 1, wherein said determining the friction coefficient comprises determining the friction coefficient based, at least in part, on a machine learning model trained on well system operational characteristics and conditions.

8. A system comprising:a computing system comprising one or more processors and one or more non-transitory computer-readable mediums including instructions which, when executed by the2024-INV-112633-WO01one or more processors, cause the one or more processors to execute one or more operations for analyzing performance of a friction reducer, the instructions including:instructions to receive a pressure pulse signal associated with a well system; instructions to determine a friction coefficient corresponding to the performance of the friction reducer in the well system; andinstructions to perform one or more operations based, at least in part, on the friction coefficient.

9. The system of claim 8, wherein at least one of a downhole operation or a downhole attribute in a wellbore of the well system is modified based, at least in part, on the friction coefficient.

10. The system of claim 8, wherein the instructions further include instructions to direct an operation to modify at least one of a downhole operation or a downhole attribute in a wellbore of the well system based, at least in part, on the friction coefficient.

11. The system of claim 8, wherein the instructions further include instructions to modify at least one of a downhole operation or a downhole attribute based, at least in part, on the friction coefficient.

12. The system of claim 8, wherein the instructions to perform one or more operations based, at least in part, on the friction coefficient comprises instructions to modify one or more operational characteristics or conditions associated with the well system.

13. The system of claim 12, wherein the one or more operational characteristics or conditions comprises at least one of a friction reducer amount, a friction reducer type, or a friction reducer concentration.

14. The system of claim 8, wherein the instructions to determine the friction coefficient comprises instructions to determine the friction coefficient based, at least in part, on a machine learning model trained on well system operational characteristics and conditions.2024-INV-112633-WO0115. One or more non-transitory computer-readable mediums including instructions which, when executed by a processor, cause the processor to execute one or more operations for analyzing performance of a friction reducer, the instructions comprising:instructions to receive a pressure pulse signal associated with a well system; instructions to determine a friction coefficient corresponding to the performance of the friction reducer in the well system; andinstructions to perform one or more operations based, at least in part, on the friction coefficient.

16. The one or more non-transitory computer-readable mediums of claim 15, wherein at least one of a downhole operation or a downhole attribute in a wellbore of the well system is modified based, at least in part, on the friction coefficient.

17. The one or more non-transitory computer-readable mediums of claim 15, wherein the instructions further comprise instructions to direct an operation to modify at least one of a downhole operation or a downhole attribute in a wellbore of the well system based, at least in part, on the friction coefficient.

18. The one or more non-transitory computer-readable mediums of claim 15, wherein the instructions further include instructions to modify at least one of a downhole operation or a downhole attribute based, at least in part, on the friction coefficient.

19. The one or more non-transitory computer-readable mediums of claim 15, wherein the instructions to perform one or more operations based, at least in part, on the friction coefficient comprises instructions to modify one or more operational characteristics or conditions associated with the well system, wherein the one or more operational characteristics or conditions comprises at least one of a friction reducer amount, a friction reducer type, or a friction reducer concentration.

20. The one or more non-transitory computer-readable mediums of claim 15, wherein the instructions to determine the friction coefficient comprises instructions to determine the friction coefficient based, at least in part, on a machine learning model trained on well system operational characteristics and conditions.

Images