DETAILED ACTION
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Information Disclosure Statement
The information disclosure statements (IDS) submitted on 06/26/2025 and 09/14/2026 were considered by the examiner.
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claims 1-20 are rejected under 35 U.S.C. 101 because the claimed invention in each of these claims is directed to a judicial exception (i.e., a law of nature, a natural phenomenon, or an abstract idea) without significantly more.
Specifically, representative Claim 1 recites:
“A method for classifying casing joint connections for casing in a wellbore, the method comprising:
obtaining a training dataset including a plurality of torque-turns datasets for a plurality of casing joint connections, each of the plurality of torque-turns datasets including a joint quality score for an associated casing joint connection of the plurality of casing joint connections; and
training, using the training dataset, a deep learning model to generate a new joint quality score for a new torque-turns dataset.”
The claim limitations considered to fall within in the abstract idea are highlighted in bold font above; the remaining features are “additional elements.”
Step 1 of the subject matter eligibility analysis entails determining whether the claimed subject matter falls within one of the four statutory categories of patentable subject matter identified by 35 U.S.C. 101: process, machine, manufacture, or composition of matter. Claim 1 recites a process and is therefore falls within a statutory category.
Step 2A, Prong One of the analysis entails determining whether the claim recites a judicial exception such as an abstract idea. Under a broadest reasonable interpretation, the highlighted portion of claim 1 comprises process steps that fall within the abstract idea judicial exception. Specifically, under the 2019 Revised Patent Subject matter Eligibility Guidance, the highlighted subject matter falls within the mental processes category.
Individually and collectively, the steps:
“obtaining a training dataset including a plurality of torque-turns datasets for a plurality of casing joint connections, each of the plurality of torque-turns datasets including a joint quality score for an associated casing joint connection of the plurality of casing joint connections;” and
“training, using the training dataset, a deep learning model to generate a new joint quality score for a new torque-turns dataset.”
may be performed as mental processes and/or mathematical concepts. Obtaining a training dataset is gathering information, which may be performed as mental processes. Training a deep learning model to generate a joint quality score is an analysis, which may be performed as mental processes. Further, the training of a deep learning model may be a mathematical concept.
The type of high-level information collecting and analyzing data recited in these elements has been found by the Federal Circuit to constitute patent ineligible matter (see Electric Power Group v. Alstom, S.A., 830 F.3d 1350, 1353-54, 119 USPQ2d 1739, 1741-42 (Fed. Cir. 2016), a claim to "collecting information, analyzing it, and displaying certain results of the collection and analysis," where the data analysis steps are recited at a high level of generality such that they could practically be performed in the human mind).
Similar limitations comprise the mental processes type and/or mathematical concepts abstract idea recited by independent claims 12 and 18. Independent claim 12 recites the steps “obtaining a torque-turn dataset”, “receiving […] a joint quality score, “scaling the torque-turns dataset”, and “training […] a deep learning model”. The obtaining a receiving are gathering information, which may be performed as mental processes. Scaling the dataset is an analysis, which may be performed as mental processes. Training a deep learning model may be performed as mental processes and/or mathematical concepts.
Step 2A, Prong Two of the analysis entails determining whether a claim includes additional elements that integrate the recited judicial exception (e.g., abstract idea) into a practical application. In view of the various considerations encompassed by the Step 2A, Prong Two analysis, claim 1 does not include additional elements that integrate the recited abstract idea into a practical application. Based on the individual and collective limitations of claim 1, applying a broadest reasonable interpretation, the most significant of such considerations appear to include: improvements to the functioning of a computer, or to any other technology or technical field (MPEP 2106.05(a)); applying the judicial exception with, or by use of, a particular machine (MPEP 2106.05(b)); and effecting a transformation or reduction of a particular article to a different state or thing (MPEP 2106.05(c)).
Regarding improvements to the functioning of a computer or other technology, none of the “additional elements” in any combination appear to integrate the abstract idea to technologically improve any aspect of a system that may be used to implement the highlighted steps such a generic computer.
Regarding application of the judicial exception with, or by use of, a particular machine, none of the additional elements are recited in a particularized manner of implementing the abstract idea process steps. Instead, independent claim 1 as a whole amounts to obtaining information and training a deep learning model.
Regarding effectuation of a transformation or reduction of a particular article to a different state or thing, the claim includes no such transformation or reduction. The training of a deep learning model is not the transformation of a particular article, but instead a process of analysis and/or mathematical concepts.
Similar analysis applies to the additional elements of independent claim 12, which does not integrate the judicial exception into a practical application.
Under similar analysis, independent claim 18 recites additional elements including “a torque sensor”, “a rotation sensor”, “a processor and memory”, which are not recited in a particularized manner. The “processor and memory” amount to a generic computer, and do not amount to the improvement of a computer or technology. Therefore claim 18 does not integrate the judicial exception into a practical application.
The above additional elements, considered individually and in combination with the claim elements reciting an abstract idea do not reflect an improvement to other technology or technical field, and, therefore, do not integrate the judicial exception into a practical application. Therefore, the claims are directed to a judicial exception and require further analysis under Step 2B.
Regarding Step 2B, independent claims 1, 12 and 18, do not include additional elements that are sufficient to amount to significantly more than the judicial exception because they are generically recited and are well-understood/conventional in the relevant art as evidenced by the prior art of record as indicated in the rejections under 35 U.S.C. §103.
Independent claims 1, 12 and 18 are therefore not patent eligible.
Dependent claims 2-11, 13-17 and 19-20 provide additional features/steps which are part of an expanded algorithm that includes the abstract idea of the independent claims (Step 2A, Prong One). None of dependent claims 2-11, 13-17 and 19-20 recite additional elements that integrate the abstract idea into practical application (Step 2A, Prong Two), and all fail the “significantly more” test under the step 2B for the same reasons as discussed with regards to the independent claims. Claims 2, 13, 14, and 19 recite further details of the training the deep learning model, which may be performed as mental processes. Claims 3, 5, 15, and 20 further detail the obtaining the training dataset. Claims 4, 9, and 10 further detail the torque-turns data set. Claims 6 and 16 recite a step of inputting the new torque-turns dataset to the deep learning model to generate the new joint quality score, which is an analysis, which may be performed as mental processes. Claims 7, 8, and 17 detail conditionally disconnecting a casing segment, which is an extra-solution for the method for classifying casing joint connections (MPEP 2106.05(g)). Claim 11 details the deep learning model to include a convoluted neural network, which may be a mathematical concept.
The dependent claims 2-11, 13-17 and 19-20 therefore are also ineligible subject matter.
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claim(s) 1-4, 9-15, and 18-20 is/are rejected under 35 U.S.C. 103 as being unpatentable over Ruehmann (WO2014071056A2, provided by applicant) in view of Bisht et al. (US 20240062119 A1).
Regarding claim 1, Ruehmann teaches A method for classifying casing joint connections for casing (Abstract; [0016] “Figure 1A illustrates a connection 1 between premium grade tubulars 2, 4. The tubulars 2, 4 may be any oil country tubular good, such as production tubing, casing, liner, or drill pipe.”) in a wellbore ([0004] “A drilling rig is constructed on the earth's surface or floated on water to facilitate the insertion and removal of tubular strings (e.g., drill pipe, casing, sucker rod, riser, or production tubing) into a wellbore. The drilling rig includes a platform and power tools, such as an elevator and slips, to engage, assemble, and lower the tubulars into the wellbore.”), the method comprising:
obtaining a (turns counter 208, torque cell 212; [0032] “Illustrative predetermined values 223-230 which may be input, by an technician or otherwise, include minimum and maximum delta gradient values 223, a shoulder threshold gradient 224, a dump torque value 226, minimum and maximum delta turns values 228, minimum and maximum torque values 230, and reference curve data 231 . The minimum and maximum torque values 230 may include a set for the shoulder position and a set for the final position. The torque values 230 may be derived theoretically, such as by finite element analysis, or empirically, such as by laboratory testing and/or analysis of historical data for a particular connection.”) for a plurality of casing joint connections ([0040] “The database 242 may include torque-turns curves of previously assembled good connections and the reference curves may be the maximum (reference curve 1 ) and minimum (reference curve 2) curves of the previously assembled good connections. The database may include curves from laboratory assembled connections and/or field assembled connections.”), each of the plurality of torque-turns datasets including a joint quality score ([0037] “a determination may be made as to whether the turns value and/or torque value are within specified limits. The specified limits may be predetermined, or based off of a value measured during makeup. If the connection evaluator 252 determines a bad connection, rotation may be terminated. Otherwise, rotation continues and the shoulder detection algorithm 248 monitors for the shoulder position. This may be accomplished by comparing the calculated torque gradient with the shoulder threshold gradient 224.”) for an associated casing joint connection of the plurality of casing joint connections ([0050] “the graphical evaluator 250 may include a comments field for allowing the technician to enter notes for each individual connection and the notes may be recorded on the storage device 221 for inclusion in the report.”). The determination that the torque/turns are within specific limits, which may be labeled a “bad connection”, is a joint quality score. The evaluations for each individual connection teaches that the casing joint connection is one of a plurality of casing joint connections.
Ruehmann does not teach the method, comprising:
training, using the training dataset, a deep learning model to generate a new joint quality score for a new torque-turns dataset.
Bisht teaches an analogous method of classifying (Fig. 10; [0133] lines 20-22, “a ML model or ML models may provide for identification and quality assessment, for example, via one or more of classification and prediction.”) components (Fig. 4, wellbore production equipment 464) in a wellbore (wellbore 406), comprising:
training, using the training dataset ([0160] “As an example, a training method can include various actions that can operate on a dataset to train a ML model. As an example, a dataset can be split into training data and test data where test data can provide for evaluation. A method can include cross-validation of parameters and best parameters, which can be provided for model training.”), a deep learning model ([0157] lines 4-7, “a machine learning model can be a deep learning model (e.g., deep Boltzmann machine, deep belief network, convolutional neural network (CNN), stacked auto-encoder, etc.)”) to generate a new joint quality score for a new dataset (Fig. 10, steps 1020 and 1030).
It would be obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the method of Ruehmann to include the training of a deep learning model of Bisht because the training of a deep learning model to generate new quality scores for a new dataset would yield predictable and advantageous results, such as producing similar quality scores for similar datasets, thereby increasing the accuracy of the evaluation.
Regarding claim 2, Ruehmann in view of Bisht teaches The method of claim 1, wherein training the deep learning model includes training the deep learning model based on a constraint to over-generate a poor joint quality score (Ruehmann: [0036] “The values (torque, rotation, and torque gradient) may then be compared by the connection evaluator 252, either continuously or at selected events, with predetermined values, such as the values 223-230.”; Bisht: [0188] “As to training and building a ML model, training can be perform on appropriate split data where one or more of various technique can be utilized to fine tune performance, for example, if a given threshold is not met.”). One of ordinary skill in the art would recognize that generating the dataset at selected events (e.g. known poor quality casing joint connection) and training the model on appropriate split data for a threshold to be met, would include the constraint on the dataset and training resulting in poor quality scores being over-generated.
Regarding claim 3, Ruehmann in view of Bisht teaches The method of claim 1, wherein obtaining the training dataset includes receiving the plurality of torque-turns datasets and processing the plurality of torque-turns datasets (Ruehmann: [0040] “he graphical evaluator may retrieve the reference curves from the database 242. The database 242 may include torque-turns curves of previously assembled good connections and the reference curves may be the maximum (reference curve 1 ) and minimum (reference curve 2) curves of the previously assembled good connections. The database may include curves from laboratory assembled connections and/or field assembled connections”).
Regarding claim 4, Ruehmann in view of Bisht teaches The method of claim 3, wherein processing the plurality of torque-turns datasets includes scaling the plurality of torque-turns datasets to a pre-determined scale (Bisht: [0181] lines 4-5, “pre-processing can include analysis of the outliers, null treatment, standardization, normalization of data, etc.”). The standardization and normalization of data is scaling datasets to a pre-determinized scale
Regarding claim 9, Ruehmann in view of Bisht teaches The method of claim 1, wherein the plurality of torque-turns datasets include a plurality of graphs illustrating torque with respect to number of rotations. (Ruehmann: Figs. 4A, 5A, 6A, 7A, 8A)
Regarding claim 10, Ruehmann in view of Bisht teaches The method of claim 1, wherein the plurality of torque-turns datasets each include a plurality of torque measurements at a plurality of rotational positions. (Ruehmann: [0035] “The frequency with which torque and rotation are measured may be specified by the sampler 240. The sampler 240 may be configurable, so that an technician may input a desired sampling frequency.”)
Regarding claim 11, Ruehmann in view of Bisht teaches The method of claim 1, wherein the deep learning model includes a convoluted neural network.
(Bisht: [0157] lines 4-7, “a machine learning model can be a deep learning model (e.g., deep Boltzmann machine, deep belief network, convolutional neural network (CNN), stacked auto-encoder, etc.)”
Regarding claim 12, Ruehmann teaches A method for classifying casing joint connections for casing (Abstract; [0016] “Figure 1A illustrates a connection 1 between premium grade tubulars 2, 4. The tubulars 2, 4 may be any oil country tubular good, such as production tubing, casing, liner, or drill pipe.”) in a wellbore ([0004] “A drilling rig is constructed on the earth's surface or floated on water to facilitate the insertion and removal of tubular strings (e.g., drill pipe, casing, sucker rod, riser, or production tubing) into a wellbore. The drilling rig includes a platform and power tools, such as an elevator and slips, to engage, assemble, and lower the tubulars into the wellbore.”), the method comprising:
obtaining a torque-turns dataset (turns counter 208, torque cell 212; [0032] “Illustrative predetermined values 223-230 which may be input, by an technician or otherwise, include minimum and maximum delta gradient values 223, a shoulder threshold gradient 224, a dump torque value 226, minimum and maximum delta turns values 228, minimum and maximum torque values 230, and reference curve data 231 . The minimum and maximum torque values 230 may include a set for the shoulder position and a set for the final position. The torque values 230 may be derived theoretically, such as by finite element analysis, or empirically, such as by laboratory testing and/or analysis of historical data for a particular connection.”), the torque-turns dataset including a plurality of rotation-series torque measurements for a casing joint connection ([0035] “The applied torque and rotation are measured at regular intervals throughout the makeup. The frequency with which torque and rotation are measured may be specified by the sampler 240. The sampler 240 may be configurable, so that an technician may input a desired sampling frequency.”);
receiving, based on a pre-determined standard for the casing joint connection, a joint quality score ([0037] “a determination may be made as to whether the turns value and/or torque value are within specified limits. The specified limits may be predetermined, or based off of a value measured during makeup. If the connection evaluator 252 determines a bad connection, rotation may be terminated. Otherwise, rotation continues and the shoulder detection algorithm 248 monitors for the shoulder position. This may be accomplished by comparing the calculated torque gradient with the shoulder threshold gradient 224.”).
Ruehmann does not teach the method, comprising:
scaling the torque-turns dataset to a pre-determined scale based on the pre-determined standard; and
training, using the torque-turns dataset scaled to the pre-determined scale, a deep learning model to generate a new joint quality score for a new torque-turns dataset.
Bisht teaches an analogous method of classifying (Fig. 10; [0133] lines 20-22, “a ML model or ML models may provide for identification and quality assessment, for example, via one or more of classification and prediction.”) components (Fig. 4, wellbore production equipment 464) in a wellbore (wellbore 406), comprising:
scaling the torque-turns dataset to a pre-determined scale based on the pre-determined standard ([0181] lines 4-5, “pre-processing can include analysis of the outliers, null treatment, standardization, normalization of data, etc.”). The standardization and normalization of data is scaling datasets to a pre-determinized scale; and
training, using the dataset scaled to the pre-determined scale ([0160] “As an example, a training method can include various actions that can operate on a dataset to train a ML model. As an example, a dataset can be split into training data and test data where test data can provide for evaluation. A method can include cross-validation of parameters and best parameters, which can be provided for model training.”), a deep learning model ([0157] lines 4-7, “a machine learning model can be a deep learning model (e.g., deep Boltzmann machine, deep belief network, convolutional neural network (CNN), stacked auto-encoder, etc.)”) to generate a new joint quality score for a new dataset (Fig. 10, steps 1020 and 1030).
It would be obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the method of Ruehmann to include the training of a deep learning model of Bisht because the training of a deep learning model to generate new quality scores for a new dataset would yield predictable and advantageous results, such as producing similar quality scores for similar datasets, thereby increasing the accuracy of the evaluation.
Regarding claim 13, Ruehmann in view of Bisht teaches The method of claim 12, wherein training the deep learning model includes training the deep learning model based on a constraint to over-generate a poor joint quality score (Ruehmann: [0036] “The values (torque, rotation, and torque gradient) may then be compared by the connection evaluator 252, either continuously or at selected events, with predetermined values, such as the values 223-230.”; Bisht: [0188] “As to training and building a ML model, training can be perform on appropriate split data where one or more of various technique can be utilized to fine tune performance, for example, if a given threshold is not met.”). One of ordinary skill in the art would recognize that generating the dataset at selected events (e.g. known poor quality casing joint connection) and training the model on appropriate split data for a threshold to be met, would include the constraint on the dataset and training resulting in poor quality scores being over-generated.
Regarding claim 14, Ruehmann in view of Bisht teaches The method of claim 12, wherein training the deep learning model includes the deep learning model identifying parameters used to determine the joint quality score (Bisht: [0200] lines 14-16, “the system includes one or more ML models that can identify data types and/or assess data quality in an automated manner.”).
Regarding claim 15, Ruehmann in view of Bisht teaches The method of claim 12, wherein receiving the torque-turns dataset includes receiving a graph of the torque-turns dataset (Ruehmann: Figs. 4A, 5A, 6A, 7A, 8A).
Regarding claim 18, Ruehmann teaches A casing installation manager (Fig. 3B), comprising:
a torque sensor (torque cell 212);
a rotation sensor (turn counter 208); and
a processor (control system 206, process monitor234) and memory (storage device 221), the memory including instructions that causes the processor to:
obtain a (turns counter 208, torque cell 212; [0032] “Illustrative predetermined values 223-230 which may be input, by an technician or otherwise, include minimum and maximum delta gradient values 223, a shoulder threshold gradient 224, a dump torque value 226, minimum and maximum delta turns values 228, minimum and maximum torque values 230, and reference curve data 231 . The minimum and maximum torque values 230 may include a set for the shoulder position and a set for the final position. The torque values 230 may be derived theoretically, such as by finite element analysis, or empirically, such as by laboratory testing and/or analysis of historical data for a particular connection.”) for a plurality of casing joint connections ([0040] “The database 242 may include torque-turns curves of previously assembled good connections and the reference curves may be the maximum (reference curve 1 ) and minimum (reference curve 2) curves of the previously assembled good connections. The database may include curves from laboratory assembled connections and/or field assembled connections.”), each of the plurality of torque-turns datasets including a joint quality score ([0037] “a determination may be made as to whether the turns value and/or torque value are within specified limits. The specified limits may be predetermined, or based off of a value measured during makeup. If the connection evaluator 252 determines a bad connection, rotation may be terminated. Otherwise, rotation continues and the shoulder detection algorithm 248 monitors for the shoulder position. This may be accomplished by comparing the calculated torque gradient with the shoulder threshold gradient 224.”) for an associated casing joint connection of the plurality of casing joint connections ([0050] “the graphical evaluator 250 may include a comments field for allowing the technician to enter notes for each individual connection and the notes may be recorded on the storage device 221 for inclusion in the report.”). The determination that the torque/turns are within specific limits, which may be labeled a “bad connection”, is a joint quality score. The evaluations for each individual connection teaches that the casing joint connection is one of a plurality of casing joint connections.
Ruehmann does not teach the manager, comprising: train, using the training dataset, a deep learning model to generate a new joint quality score for a new torque-turns dataset.
Bisht teaches an analogous manager (Figs. 1-3, and 10; [0133] lines 20-22, “a ML model or ML models may provide for identification and quality assessment, for example, via one or more of classification and prediction.”) for components (Fig. 4, wellbore production equipment 464) in a wellbore (wellbore 406), comprising:
train, using the training dataset ([0160] “As an example, a training method can include various actions that can operate on a dataset to train a ML model. As an example, a dataset can be split into training data and test data where test data can provide for evaluation. A method can include cross-validation of parameters and best parameters, which can be provided for model training.”), a deep learning model ([0157] lines 4-7, “a machine learning model can be a deep learning model (e.g., deep Boltzmann machine, deep belief network, convolutional neural network (CNN), stacked auto-encoder, etc.)”) to generate a new joint quality score for a new dataset (Fig. 10, steps 1020 and 1030).
It would be obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the manager of Ruehmann to include the training of a deep learning model of Bisht because the training of a deep learning model to generate new quality scores for a new dataset would yield predictable and advantageous results, such as producing similar quality scores for similar datasets, thereby increasing the accuracy of the evaluation.
Regarding claim 19, Ruehmann in view of Bisht teaches The casing installation manager of claim 18, wherein training the deep learning model includes training the deep learning model based on a constraint to over-generate a poor joint quality score (Ruehmann: [0036] “The values (torque, rotation, and torque gradient) may then be compared by the connection evaluator 252, either continuously or at selected events, with predetermined values, such as the values 223-230.”; Bisht: [0188] “As to training and building a ML model, training can be perform on appropriate split data where one or more of various technique can be utilized to fine tune performance, for example, if a given threshold is not met.”). One of ordinary skill in the art would recognize that generating the dataset at selected events (e.g. known poor quality casing joint connection) and training the model on appropriate split data for a threshold to be met, would include the constraint on the dataset and training resulting in poor quality scores being over-generated.
Regarding claim 20, teaches The casing installation manager of claim 18, wherein obtaining the training dataset includes receiving the plurality of torque-turns datasets (Ruehmann: [0040] “he graphical evaluator may retrieve the reference curves from the database 242. The database 242 may include torque-turns curves of previously assembled good connections and the reference curves may be the maximum (reference curve 1 ) and minimum (reference curve 2) curves of the previously assembled good connections. The database may include curves from laboratory assembled connections and/or field assembled connections”) and scaling the plurality of torque-turns datasets to a pre-determined scale (Bisht: [0181] lines 4-5, “pre-processing can include analysis of the outliers, null treatment, standardization, normalization of data, etc.”). The standardization and normalization of data is scaling datasets to a pre-determinized scale.
Claim(s) 5-8, and 16-17 is/are rejected under 35 U.S.C. 103 as being unpatentable over Ruehmann in view of Bisht as applied to claim 1 above, and further in view of Savini et al. (EP3608501A1, provided by applicant).
Regarding claim 5, Ruehmann in view of Bisht teaches The method of claim 1.
Ruehmann in view of Bisht does not teach the method, wherein obtaining the training dataset includes obtaining the training dataset for a casing connection type and wherein training the deep learning model includes training the deep learning model for the casing connection type.
Savini teaches an analogous method (Abstract), wherein obtaining the training dataset (Fig. 1, steps 102 and 104) includes obtaining the training dataset for a casing connection type and wherein training the deep learning model includes training the deep learning model for the casing connection type (make-up parameters of [0037]; [0017] “The make-up parameters may provide settings that enable the controller 220 to identify the conclusion of the make-up process and/or evaluate the acceptability of the connection.”). The make-up parameters indicate the casing connection type.
It would be obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the method of Ruehmann in view of Bisht to include the casing connection type of Savini in the training dataset because it would yield predictable and advantageous results, such as providing contextual information for the training dataset, thereby increasing the accuracy of determining whether the connection is correctly made and that the quality score is representative to the specific connection.
Regarding claim 6, Ruehmann in view of Bisht teaches The method of claim 1, further comprising inputting the new torque-turns dataset to the deep learning model to generate the new joint quality score (Bisht: [0326] lines 1-12, “a method can include automatically processing field equipment data using a trained machine learning model as a first trained machine learning model for generation of a quality score and can include automatically processing the field equipment data using a second trained machine learning model to generate a prediction for the field equipment data. For example, consider a trend prediction that indicates a trend in the field equipment data, which may be a current trend that can be extrapolated to a future time or future times. As an example, a method can include generating a suggestion to improve a quality score based at least in part on a prediction.”),
Ruehmann in view of Bisht does not teach the method, wherein the new joint quality score facilitates a casing installation manager to determine a casing joint connection quality.
Savini teaches an analogous method (Abstract), wherein the new joint quality score facilitates a casing installation manager to determine a casing joint connection quality (Fig. 5; [0050] “If the predicted outcome is positive, e.g., the consequences of an anomaly are lessened and/or the anomaly is avoided, the method 500 may proceed to take such corrective action, e.g., by signaling to the tubular rotating equipment to adjust speeds, and the method 500 may loop back to obtaining measurements at 506 again, as the connection process continues. Otherwise, the controller 220 may halt the connection process, as at 519 before it is determined to be complete at 518, in order, e.g., to avoid spending time and risking the integrity of the tubulars by continuing a faulty connection process.”). The controller, able to halt or continue the connection process based on the new joint quality score, is the casing installation manager.
It would be obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the method of Ruehmann in view of Bisht to include the casing installation manager of Savini because it would yield advantageous results, such as enabling corrective action based on the joint connection quality.
Regarding claim 7, Ruehmann in view of Bisht and Savini teaches The method of claim 6, further comprising, when the new joint quality score is above a threshold (Savini: [0041] “Classification may be conducted by comparing the confidence to a threshold value. If the confidence is above the threshold, the conclusion may be that the accept/reject/anomaly identification is accurate. Thus, if the threshold is set, for example, at 0.5, if a confidence level for a particular anomaly is above 0.5, then it is selected as being apparent in the dataset. This threshold can be static and pre-determined, or may change. Further, the same or different thresholds may be employed for different accept/reject/anomaly determinations, e.g., if the machine learning algorithm has a higher propensity for false positives of a particular anomaly, the threshold may be set higher for that anomaly in the classification step at 414.”), installing a casing segment associated with the new torque-turns dataset in a wellbore (Savini: [0022] “The method 100 may, whether during or subsequent to the steps 106, 108, 110, evaluate connection acceptability based on the measurements taken at 104 and the make-up parameters, as at 112. As such, this maybe in parallel to the steps 106, 108, 110. The evaluation may be conducted in a variety of ways. In one embodiment, a set of pre-determined, static rules may be used. In another embodiment, the measurements may be fed to a machine learning analysis, trained to detect acceptable and unacceptable connections based on such measurements.”; Fig. 5; [0050] “If the predicted outcome is positive, e.g., the consequences of an anomaly are lessened and/or the anomaly is avoided, the method 500 may proceed to take such corrective action, e.g., by signaling to the tubular rotating equipment to adjust speeds, and the method 500 may loop back to obtaining measurements at 506 again, as the connection process continues. Otherwise, the controller 220 may halt the connection process, as at 519 before it is determined to be complete at 518, in order, e.g., to avoid spending time and risking the integrity of the tubulars by continuing a faulty connection process.”).
It would be obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the method of Ruehmann in view of Bisht to include the conditional installation based on the new joint quality score because it would yield advantageous results, such as advantageously installing the segment based on the determined quality score.
Regarding claim 8, Ruehmann in view of Bisht and Savini teaches The method of claim 6.
Ruehmann in view of Bisht does not teach the method further comprising, when the new joint quality score is below a threshold, disconnecting a casing segment associated with the new torque-turns dataset.
Savini teaches an analogous method (Abstract) further comprising, when the new joint quality score is below a threshold ([0041] “Classification may be conducted by comparing the confidence to a threshold value. If the confidence is above the threshold, the conclusion may be that the accept/reject/anomaly identification is accurate. Thus, if the threshold is set, for example, at 0.5, if a confidence level for a particular anomaly is above 0.5, then it is selected as being apparent in the dataset. This threshold can be static and pre-determined, or may change. Further, the same or different thresholds may be employed for different accept/reject/anomaly determinations, e.g., if the machine learning algorithm has a higher propensity for false positives of a particular anomaly, the threshold may be set higher for that anomaly in the classification step at 414.”), disconnecting a casing segment associated with the new torque-turns dataset (“[0023] In at least some embodiments, the connection process may be adjusted, as at 114, e.g., on-the-fly, in real-time, in response to the controller 220 evaluating the connection acceptability at 112, prior to the connection being fully made-up. For example, the controller 220 may signal to the tubular rotating equipment 202 to change speeds, or even stop and reverse, in order to avoid a potentially damaging consequence of an unacceptable connection (e.g., to avoid thread damage).”).
It would be obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the method of Ruehmann in view of Bisht to include the conditional disconnection based on the new joint quality score because it would yield advantageous results, such as advantageously preventing damage based on the determined quality score.
Regarding claim 16, Ruehmann in view of Bisht teaches The method of claim 12, further comprising inputting the new torque-turns dataset to the deep learning model to generate the new joint quality score (Bisht: [0326] lines 1-12, “a method can include automatically processing field equipment data using a trained machine learning model as a first trained machine learning model for generation of a quality score and can include automatically processing the field equipment data using a second trained machine learning model to generate a prediction for the field equipment data. For example, consider a trend prediction that indicates a trend in the field equipment data, which may be a current trend that can be extrapolated to a future time or future times. As an example, a method can include generating a suggestion to improve a quality score based at least in part on a prediction.”).
Ruehmann in view of Bisht does not teach the method, wherein the new joint quality score facilitates a casing installation manager to determine a casing joint connection quality.
Savini teaches an analogous method (Abstract), wherein the new joint quality score facilitates a casing installation manager to determine a casing joint connection quality (Fig. 5; [0050] “If the predicted outcome is positive, e.g., the consequences of an anomaly are lessened and/or the anomaly is avoided, the method 500 may proceed to take such corrective action, e.g., by signaling to the tubular rotating equipment to adjust speeds, and the method 500 may loop back to obtaining measurements at 506 again, as the connection process continues. Otherwise, the controller 220 may halt the connection process, as at 519 before it is determined to be complete at 518, in order, e.g., to avoid spending time and risking the integrity of the tubulars by continuing a faulty connection process.”). The controller, able to halt or continue the connection process based on the new joint quality score, is the casing installation manager.
It would be obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the method of Ruehmann in view of Bisht to include the casing installation manager of Savini because it would yield advantageous results, such as enabling corrective action based on the joint connection quality.
Regarding claim 17, Ruehmann in view of Bisht and Savini teaches The method of claim 16 further comprising, when the new joint quality score is above a threshold (Savini: [0041] “Classification may be conducted by comparing the confidence to a threshold value. If the confidence is above the threshold, the conclusion may be that the accept/reject/anomaly identification is accurate. Thus, if the threshold is set, for example, at 0.5, if a confidence level for a particular anomaly is above 0.5, then it is selected as being apparent in the dataset. This threshold can be static and pre-determined, or may change. Further, the same or different thresholds may be employed for different accept/reject/anomaly determinations, e.g., if the machine learning algorithm has a higher propensity for false positives of a particular anomaly, the threshold may be set higher for that anomaly in the classification step at 414.”), installing a casing segment associated with the new torque-turns dataset in a wellbore (Savini: [0022] “The method 100 may, whether during or subsequent to the steps 106, 108, 110, evaluate connection acceptability based on the measurements taken at 104 and the make-up parameters, as at 112. As such, this maybe in parallel to the steps 106, 108, 110. The evaluation may be conducted in a variety of ways. In one embodiment, a set of pre-determined, static rules may be used. In another embodiment, the measurements may be fed to a machine learning analysis, trained to detect acceptable and unacceptable connections based on such measurements.”)
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to BRIAN GEISS whose telephone number is (571)270-1248. The examiner can normally be reached Monday - Friday 7:30 am - 4:30 pm.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Catherine Rastovski can be reached at (571) 270-0349. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/B.B.G./Examiner, Art Unit 2857
/Catherine T. Rastovski/Supervisory Primary Examiner, Art Unit 2857
Prosecution Insights