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 . Claims 1-20 are pending.
Continued Examination Under 37 CFR 1.114
A request for continued examination under 37 CFR 1.114, including the fee set forth in 37 CFR 1.17(e), was filed in this application after final rejection. Since this application is eligible for continued examination under 37 CFR 1.114, and the fee set forth in 37 CFR 1.17(e) has been timely paid, the finality of the previous Office action has been withdrawn pursuant to 37 CFR 1.114. Applicant's submission filed on 7/23/26 has been entered.
Claim Rejections - 35 USC § 101
Claims 1-20 are rejected under 35 U.S.C. 101 because the claimed invention is directed to a judicial exception without significantly more.
Step 1 (The Statutory Categories): Is the claim to a process, machine, manufacture, or composition of matter? MPEP 2106.03.
Per Step 1, claims 1-7 are to a method (i.e., a process), claims 8-14 to a system (i.e., a machine), and claims 15-20 to non-transitory machine-readable media (i.e., an article), which are statutory categories of invention. However, the claims are rejected under 35 U.S.C. 101 because they are directed to an abstract idea, a judicial exception, without reciting additional elements that integrate the judicial exception into a practical application.
The analysis proceeds to Step 2A Prong One.
Step 2A Prong One: Does the claim recite an abstract idea, law of nature, or natural phenomenon? MPEP 2106.04.
The abstract idea of claims 1, 8, and 15 is (claim 1 being representative):
generating an expanded dataset of simulated pulsed neutron logging (PNL) data including simulated spectra corresponding to neutron interactions with formation materials based, at least in part, on an original dataset of empirical PNL data;
converting, using one or more calibration coefficients determined from calibration curves generated by fitting the simulated PNL data and the empirical PNL data, the simulated PNL data into lab-equivalent synthetic data.
The abstract idea steps italicized above are those which could be performed mentally, including with pen and paper. The steps describe, at a high level, 1) generating an expanded data set of simulated data including simulated spectra, based on an original empirical data; 2) converting the simulated data using one or more calibration coefficients. These are steps an administrator could accomplish these tasks with pen and paper. If a claim limitation, under its broadest reasonable interpretation, covers performance of the limitation in the mind, including observations, evaluations, judgements, and/or opinions, then it falls within the Mental Processes – Concepts Performed in the Human Mind grouping of abstract ideas. Accordingly, the claim recites an abstract idea. (Examiner notes that the converting step is rudimentary and merely involves the application of previously determined calibration coefficients.)
Additionally and alternatively, the abstract idea steps italicized above describe converting simulated data using one or more calibration coefficients, which constitutes a process that, under its broadest reasonable interpretation, covers mathematical concepts. If a claim limitation, under its broadest reasonable interpretation, covers mathematical concepts, including mathematical relationships, mathematical formulas or equations, mathematical calculations, then it falls within the Mathematical Concepts grouping of abstract ideas. Accordingly, the claim recites an abstract idea.
Step 2A Prong Two: Does the claim recite additional elements that integrate the judicial exception into a practical application? MPEP 2106.04.
This judicial exception is not integrated into a practical application because the additional elements are merely instructions to apply the abstract idea to a computer, as described in MPEP 2106.05(f).
Claim 1 recites the following additional elements: controlling a learning machine [to predict fluid holdup in a borehole]; training an ensemble of machine learning models based on the lab-equivalent synthetic data.
Claim 8 recites the following additional elements: a processor; a learning machine [configured to predict fluid holdup in a borehole drilled into a subsurface formation]; instructions executable on the processor; instructions to train an ensemble of machine learning models based on the lab-equivalent synthetic data.
Claim 15: one or more non-transitory machine-readable media; instructions executable on one or more processors; instructions to train an ensemble of machine learning models based on the lab-equivalent synthetic data.
These elements are merely instructions to apply the abstract idea to a computer, per MPEP 2106.05(f). Applicant has only described generic computing elements in their specification, as seen in [0012] of applicant’s specification as filed, for example.
Examiner interprets the machine learning ensemble training features as additional elements. MPEP 2106.05(f) is explicit that simply using other machinery as a tool also amounts to no more than merely applying the abstract idea to a computer, especially when claimed in a solution-oriented manner:
(1) Whether the claim recites only the idea of a solution or outcome i.e., the claim fails to recite details of how a solution to a problem is accomplished. The recitation of claim limitations that attempt to cover any solution to an identified problem with no restriction on how the result is accomplished and no description of the mechanism for accomplishing the result, does not integrate a judicial exception into a practical application or provide significantly more because this type of recitation is equivalent to the words "apply it". See Electric Power Group, LLC v. Alstom, S.A., 830 F.3d 1350, 1356, 119 USPQ2d 1739, 1743-44 (Fed. Cir. 2016); Intellectual Ventures I v. Symantec, 838 F.3d 1307, 1327, 120 USPQ2d 1353, 1366 (Fed. Cir. 2016); Internet Patents Corp. v. Active Network, Inc., 790 F.3d 1343, 1348, 115 USPQ2d 1414, 1417 (Fed. Cir. 2015). In contrast, claiming a particular solution to a problem or a particular way to achieve a desired outcome may integrate the judicial exception into a practical application or provide significantly more. See Electric Power, 830 F.3d at 1356, 119 USPQ2d at 1743.
[…]
(2) Whether the claim invokes computers or other machinery merely as a tool to perform an existing process. Use of a computer or other machinery in its ordinary capacity for economic or other tasks (e.g., to receive, store, or transmit data) or simply adding a general purpose computer or computer components after the fact to an abstract idea (e.g., a fundamental economic practice or mathematical equation) does not integrate a judicial exception into a practical application or provide significantly more. See Affinity Labs v. DirecTV, 838 F.3d 1253, 1262, 120 USPQ2d 1201, 1207 (Fed. Cir. 2016) (cellular telephone); TLI Communications LLC v. AV Auto, LLC, 823 F.3d 607, 613, 118 USPQ2d 1744, 1748 (Fed. Cir. 2016) (computer server and telephone unit). Similarly, "claiming the improved speed or efficiency inherent with applying the abstract idea on a computer" does not integrate a judicial exception into a practical application or provide an inventive concept. Intellectual Ventures I LLC v. Capital One Bank (USA), 792 F.3d 1363, 1367, 115 USPQ2d 1636, 1639 (Fed. Cir. 2015). In contrast, a claim that purports to improve computer capabilities or to improve an existing technology may integrate a judicial exception into a practical application or provide significantly more. McRO, Inc. v. Bandai Namco Games Am. Inc., 837 F.3d 1299, 1314-15, 120 USPQ2d 1091, 1101-02 (Fed. Cir. 2016); Enfish, LLC v. Microsoft Corp., 822 F.3d 1327, 1335-36, 118 USPQ2d 1684, 1688-89 (Fed. Cir. 2016). See MPEP §§ 2106.04(d)(1) and 2106.05(a) for a discussion of improvements to the functioning of a computer or to another technology or technical field.
In this case, the machine learning ensemble training features are merely being used to facilitate the tasks of the abstract idea, which provides nothing more than a results-oriented solution that lacks detail of the mechanism for accomplishing the result and is equivalent to the words “apply it,” per MPEP 2106.05(f).
Further, the combination of these elements is nothing more than a generic computing system applied to the tasks of the abstract idea. Because the additional elements are merely instructions to apply the abstract idea to a generic computing system, they do not integrate the abstract idea into a practical application, when viewed in combination. See MPEP 2106.05(f).
Therefore, per Step 2A Prong Two, the additional elements, alone and in combination, do not integrate the judicial exception into a practical application. The claim is directed to an abstract idea.
Step 2B (The Inventive Concept): Does the claim recite additional elements that amount to significantly more than the judicial exception? MPEP 2106.05.
Step 2B involves evaluating the additional elements to determine whether they amount to significantly more than the judicial exception itself.
The examination process involves carrying over identification of the additional element(s) in the claim from Step 2A Prong Two and carrying over conclusions from Step 2A Prong Two pertaining to MPEP 2106.05(f).
The additional elements and their analysis are therefore carried over: applicant has merely recited elements that facilitate the tasks of the abstract idea, as described in MPEP 2106.05(f).
Further, the combination of these elements is nothing more than a generic computing system applied to the tasks of the abstract idea. When the claim elements above are considered, alone and in combination, they do not amount to significantly more.
Therefore, per Step 2B, the additional elements, alone and in combination, are not significantly more. The claims are not patent eligible.
The analysis takes into consideration all dependent claims as well:
Claims 3-7, 10-14, and 17-20 further narrow the abstract idea by adding additional mental and/or mathematical steps. This narrowing of the abstract idea does not result in integration into practical application and/or being significantly more.
Claims 2, 9, and 16, in addition to narrowing the abstract idea with mental and/or mathematical steps, also recite further additional elements: using a selected machine learning model/using a selected machine learning. Similar to above, these additional elements are recited at a high level of generality and in a results-oriented manner and do no more than facilitate the tasks of the abstract idea. Whether viewed alone or in combination, this does not integrate the abstract idea into practical application and/or add significantly more. See MPEP 2106.05(f).
Accordingly, claims 1-20 are rejected under 35 USC § 101 as being directed to non-statutory 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.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
Claims 1, 3, 8, 10, 15, and 17 are rejected under 35 U.S.C. 103 as being unpatentable over Al Madani (US 20210304060) in view of Whetton (US 20210003737) and Chen (US 20050246297) and Boualleg (US 20220170359).
Claims 1, 8, and 15
Al Madani discloses:
[A method for controlling a learning machine to predict fluid holdup in a borehole {method described in [0002]; controlling a neural network or learning machine to predict fluid holdup in a borehole described in [0015], [0022]}, the method comprising:]
[A holdup prediction system {system described in [0003]} comprising: a processor {processor described in [0004]}; a learning machine configured to predict fluid holdup in a borehole drilled into a subsurface formation {See previous citations to [0002], [0015], and [0022].}, the learning machine including instructions executable on the processor {instructions described in [0077]}, the instructions comprising:]
[One or more non-transitory machine-readable media instructions to predict fluid holdup in a borehole drilled into a subsurface formation, the instructions executable on one or more processors {See previous citations to [0002], [0004], [0015], [0022], and [0077]}, the instructions comprising:]
converting, using one or more calibration coefficients, the simulated data into lab-equivalent synthetic data {converting, using one or more calibration coefficients, the simulated data into lab-equivalent synthetic data described in [0042]: In some embodiments, well data is augmented to account for the lifetime and calibration of a logging tool being used to acquire well data measurements. For example, depending on the length of time and/or physical conditions of a logging tool in a well, the logging tool may need to be recalibrated in order to provide accurate sensor measurements. Without calibration, the well data may be offset from the actual well properties. Thus, data augmentation may generate augmented well data similar to well data produced by a logging tool in need of calibration.}.
Al Madani doesn’t explicitly disclose, however, Whetton, in a similar field of endeavor directed to determining component weight and/or volume fractions of subterranean rock, teaches:
generating an expanded dataset of simulated pulsed neutron logging (PNL) data including simulated spectra corresponding to neutron interactions with formation materials based, at least in part, on an original dataset of empirical PNL data {described in [0102] and [0103], where expanded dataset of simulated pulsed neutron logging (PNL) data represented by simulated (forward modeled) burst and capture spectra, i.e., simulated spectra corresponding to neutron interactions with formation materials, which is used to augment an original dataset of empirical PNL data, described in [0088]. pulsed neutron logging (PNL) further described in [0085]: [0085] FIG. 6 shows an example of the total number (“counts”) of detected gamma radiation photons 152 (of various different energy levels) that are observed during the course of a single neutron pulse cycle as well as the number of those total photons that are emitted as a result of a neutron capture interaction. In the illustrated example, the neutron source 130 is pulsed at 10 kHz, so each detection cycle lasts 100 microseconds. When the neutron source 130 pulse begins at t0, the one or more detectors 200 of the tool 100 continue to observe gamma radiation photons that are emitted as a result of neutron capture interactions associated with neutrons that were emitted during the previous neutron pulse. [0088] The digitized magnitudes 404 for the pulses 402 detected within the intervals 502 and 504 are provided from memory 254 to a controller 256 (e.g., a microprocessor, a microcontroller, a FPGA, or other logic circuitry). From the data corresponding to the intervals 502 and 504, the controller 256 generates a raw total spectrum and a raw capture spectrum, respectively. Although a single neutron pulse cycle is illustrated in FIG. 5, the data utilized to generate the raw spectra is typically collected across multiple neutron pulse cycles. The number of cycles included in the spectra is dependent upon the rate at which the tool 100 is conveyed through the wellbore 104 and the desired spectral resolution increment (i.e., the depth interval over which a spectrum will be generated). [0102] The inversion process begins with initial (first) estimates of the mineral concentration and (when present) fluid concentration and the generation of signals 603 indicative thereof. Such signals represent a first iterative assessment of the minerals and fluids making up the rock in the form of e.g. a mineral and fluid weight and/or volume fraction list, matrix or array. The method proceeds to process this in a manner comparing 606 signals representing the burst and capture spectrums with simulated (forward modeled) burst and capture spectra while also comparing the one or more constraining log with one or more simulated constraining log calculated from the mineral and fluid model. [0103] This stage involves calculating from the one or more first mineral and fluid concentration data set signal one or more first elemental concentration signal representing a first simulated log of elemental concentrations in the rock; and forward modeling from the one or more elemental concentration signal one or more simulated energy spectrum signal.}
It would have been obvious to one of ordinary skill in the art before the effective filing date of the invention to modify Al Madani to include the features of Whetton. Given that Al Madani is directed to determining subsurface formations based on collected and generated data, one of ordinary skill in the art would have been motivated to look to Whetton, in order to facilitate avoiding the ambiguities and errors apparent in conventional nuclear spectroscopy logging {[0119] of Whetton}.
The combination of Al Madani and Whetton, while disclosing lab-equivalent synthetic data {see [0042] of Al Madani}, doesn’t explicitly teach, however, Chen, in a similar field of endeavor directed to processing well logging data, teaches:
pulsed neutron logging (PNL); PNL {pulsed neutron logging (PNL) described in [0011]: Systems using a single neural network trained in this way are capable of providing good synthetic or artificial triple combo open hole logs from real data taken by pulsed neutron logging tools, at least for wells near, or in the same geological area as, the well or wells used for training.}
training an ensemble of machine learning models based on the data {training an ensemble of machine learning models based on the data described in [0027]: At block 16, the data sets 12 and 14 are used to train a set of neural networks. That is, a plurality of neural networks are generated by inputting the data 14 and adjusting network coefficients until the network outputs are close approximations to the actual open hole data 12 from the same well. At 18, a genetic algorithm is used to select a subset of neural networks to form a neural network ensemble 20.}.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the invention to modify the combination of Al Madani and Whetton to include the features of Chen. Given that Al Madani is directed to determining subsurface formations based on collected and generated data, one of ordinary skill in the art would have been motivated to look to Chen, in order to reduce superfluous data collection, thereby saving time and money {[0006] of Chen}.
The combination of Al Madani, Whetton, and Chen, while disclosing one or more calibration coefficients {see [0042] of Al Madani}, doesn’t explicitly teach, however, Boualleg, in a similar field of endeavor directed to acquiring drilling performance data for a downhole tool, teaches:
one or more calibration coefficients determined from calibration curves generated by fitting the simulated PNL data and the empirical PNL data {one or more calibration coefficients determined from calibration curves generated by fitting the simulated PNL data and the empirical PNL data described in [0251]: FIG. 8 shows an example of a method 800 that includes an acquisition block 810 for acquiring offset well data (e.g., steering commands, drilling parameters, drilling dynamics and mechanics data, LWD data, caliper and D&I responses), a determination block 820 for determining estimates of formation tops and signatures of LWD measurements versus MD/TVD for subject well, a re-calibration block 830 for re-calibrating hole-propagation model parameters (data-based/analytical/semi-analytical model) using the offset data (using steering commands; drilling parameters (surface and/or downhole); LWD data and D&I responses versus MD/TVD), a reception block 840 for receiving target information and testing the behavior of closed-loop trajectory controller(s) for ranges of drilling parameters (WOB, ROP, RPM, flow rate) and formation characteristics of the offset wells for the given planned trajectory, an identification block 850 for identifying risks regions with regard to steerability performance and, as appropriate, redefining a plan (e.g., loading a risk matrix in the tool, loading re-calibrated model parameters as well as the LWD patterns versus MD/TVD, etc.), a correlation block 860 for, during real-time drilling, at the surface (or downhole if the estimates of the LWD and the steering model are loaded in the tool and ROP is available) correlating the LWD patterns from the offset wells with real-time recordings from the LWD (e.g. gamma ray), an adjustment block 870 for adjusting MD of formation tops using the correlation at the surface or adjust downhole using the estimated MD, and a utilization block 880 for using the MD-adjusted LWD patterns and the steerability model (e.g., closed-loop) in predicting to the bit or ahead of the bit to hit the target(s) including uncertainty in steerability model parameters as well as including maximum ability of tool to drill a curve and uncertainties in future MD-formation top changes. As an example, one or more actions of the method 800 may be utilized in the method 700. Also see [0257].}.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the invention to modify the combination of Al Madani, Whetton, and Chen to include the features of Boualleg. Given that Al Madani is directed to determining subsurface formations based on collected and generated data, one of ordinary skill in the art would have been motivated to look to Boualleg, in order to dynamically adjust models upon receiving real-time well data, thereby ensuring that or more outputs can be revised {[0257] of Boualleg}.
Claims 3, 10, and 17
Chen further teaches: selecting a machine learning model of least error from the ensemble {selecting a machine learning model from the ensemble described in [0027]: At 18, a genetic algorithm is used to select a subset of neural networks to form a neural network ensemble 20. of least error described in [0069]: n the above described methods, the weighting functions, k.sub.1, k.sub.2, and k.sub.3 may be selected based on various factors. FIGS. 9 and 10 illustrate a process by which the weighting functions may be estimated using a separate genetic algorithm driven inverse process if additional data from a test well is available. This process may be used to determine what weighting factors applied to the primary validation data from the training well or wells would lead to the finding of a set of ensembles that could minimize the prediction error on other application wells similar to the test well.}; and validating the selected machine learning model against the original dataset of empirical PNL data {error minimization and validating described in [0069]: In the above described methods, the weighting functions, k.sub.1, k.sub.2, and k.sub.3 may be selected based on various factors. FIGS. 9 and 10 illustrate a process by which the weighting functions may be estimated using a separate genetic algorithm driven inverse process if additional data from a test well is available. This process may be used to determine what weighting factors applied to the primary validation data from the training well or wells would lead to the finding of a set of ensembles that could minimize the prediction error on other application wells similar to the test well.}.
The motivation and rationale to include the additional features of Chen is the same as above.
Claims 2, 9, and 16 are rejected under 35 U.S.C. 103 as being unpatentable over the combination of Al Madani, Whetton, Chen, and Boualleg, further in view of Inanc (US 20160320523).
Claims 2, 9, and 16
Chen further teaches: using a selected machine learning model {using a selected machine learning model described in [0027]: At 18, a genetic algorithm is used to select a subset of neural networks to form a neural network ensemble 20.}.
The motivation and rationale to include the additional features of Chen is the same as above.
The combination of Al Madani, Whetton, Chen, and Boualleg, while teaching the features above, doesn’t explicitly teach, however, Inanc, in a similar field of endeavor directed to density measurements using detectors on a pulsed neutron measurement platform, teaches: predicting, based on data collected from the borehole, a value of fluid holdup in the borehole {predicting, based on data collected from the borehole, a value of fluid holdup described in [0011]: In one embodiment, the fluid density measurements are used to estimate the holdup of one or more phases of the fluid based on the density measurements (i.e., the holdup density). The combination tool is configured to be disposed in a downhole environment, for example, in a wireline or logging-while-drilling (LWD) well logging application.}.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the invention to modify the combination of Al Madani, Whetton, Chen, and Boualleg to include the features of Inanc. Given that Al Madani is directed to a determining subsurface formations based on collected and generated data, one of ordinary skill in the art would have been motivated to look to Inanc, in order to facilitate estimating borehole fluid data including the holdup of phases of the fluid {[0012] of Inanc}.
Claims 4-5, 11-12, and 18 are rejected under 35 U.S.C. 103 as being unpatentable over the combination of Al Madani, Whetton, Chen, and Boualleg, further in view of Jacobson (US 20050067160).
Claims 4, 11, and 18
The combination of Al Madani, Whetton, Chen, and Boualleg, while teaching the features above, doesn’t explicitly teach, however, Jacobson, in a similar field of endeavor directed to pulsed-neutron formation density, teaches: calibrating, based on the empirical PNL data, one or more ratios and channels within the simulated PNL data, wherein the one or more channels comprise portions of a PNL spectrum {calibrating, based on the empirical PNL data, one or more ratios and channels, wherein the one or more channels comprise portions of a PNL spectrum described in [0052]: Thus, a Monte Carlo, or similar analysis, may be run to determine the tool response for various assumed porosities. Once the ratio of the near and far inelastic gamma ray count rates, and the ratio of the near and far thermal capture gamma ray count rates are determined for the modeled parameters, the coefficient needed to create the compensated ratio of inelastic count rate may be determined (along with the coefficients of the characteristic equation). The logging tool may then be used in an actual formation to obtain a ratio of actual near and far inelastic count rates, and also to determine a ratio of actual near and far thermal capture gamma rays. Using the actual ratios and the coefficient Z determined in the modeling process, a compensated ratio may be created, which may be applied to equation 6 (along with the coefficients of equation 6 determined from the model) to calculate a density. Determining density in this manner, the neutron transport effects may be substantially reduced, thus increasing the accuracy of the density determination using a pulse-neutron logging tool.}.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the invention to modify the combination of Al Madani, Whetton, Chen, and Boualleg to include the features of Jacobson. Given that Al Madani is directed to determining subsurface formations based on collected and generated data, one of ordinary skill in the art would have been motivated to look to Jacobson, in order to compensate for the effects on data from pulsed-neutron density logging tools {[0004] of Jacobson}.
Claims 5 and 12
Chen further teaches: wherein converting, using the one or more calibration coefficients, the simulated PNL data into the lab-equivalent synthetic data comprises: plotting, for each channel of the one or more channels, the simulated PNL data against the empirical PNL data {plotting the simulated data against empirical described in [0057]: FIG. 8 provides a comparison of actual triple combo logs of formation density 64, neutron porosity 65 and deep resistivity 66 to synthetic predictions 68, 69, 70 of the same log parameters. The synthetic versions were generated by inputting seven parameters from a cased hole pulsed neutron logging tool into a neural network ensemble created by the methods described above. The close correlation of the synthetic open hole logs to actual open hole logs indicates that ensembles produced by use of the present invention can accurately produce predictions of the open hole logging parameters.}; generating a calibration curve to fit the simulated PNL data and the empirical PNL data {As seen in Fig. 8 and [0057].}; selecting a set of calibration coefficients based on a function of the calibration curve {As seen in Fig. 8 and [0057].}; and converting the simulated PNL data into the lab-equivalent synthetic data based on the set of calibration coefficients {As seen in Fig. 8 and [0057].}.
The motivation and rationale to include the additional features of Chen is the same as above.
Claims 6 and 13 are rejected under 35 U.S.C. 103 as being unpatentable over the combination of Al Madani, Whetton, Chen, and Boualleg, further in view of Ma (US 20180058188).
Claims 6 and 13
The combination of Al Madani, Whetton, Chen, and Boualleg, while teaching the features above, doesn’t explicitly teach, however, Ma, in a similar field of endeavor directed to determining salinity of water in a borehole of a formation, teaches: wherein converting, using the one or more calibration coefficients, the simulated PNL data into the lab-equivalent synthetic data further comprises: interpolating and extrapolating unknown data values to fill a variable space of the lab-equivalent synthetic data based, at least in part, on the one or more calibration coefficients {interpolating and extrapolating unknown data values to fill a variable space described in [0025]: The various formation and borehole conditions used in a laboratory or in a computer simulation may be selected to bracket the conditions that may be found in various field situations of formation and borehole conditions. By bracketing expected field conditions, the correlations may be used to calculate, by interpolation or extrapolation for example, results of field responses of far and near detectors. If a field situation of formation and borehole conditions lies outside a bracket, an extrapolation may be performed to determine field responses of far and near detectors.}.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the invention to modify the combination of Al Madani, Whetton, Chen, and Boualleg to include the features of Ma. Given that Al Madani is directed to determining subsurface formations based on collected and generated data, one of ordinary skill in the art would have been motivated to look to Ma, in order to facilitate determining water salinity, which is necessary to compute an accurate estimate of a water fraction in formations {[0003] of Ma}.
Claims 7, 14, and 19-20 are rejected under 35 U.S.C. 103 as being unpatentable over the combination of Al Madani, Whetton, Chen, Boualleg, and Jacobson, further in view of Ma.
Claims 7, 14, and 20
The combination of Al Madani, Whetton, Chen, Boualleg, and Jacobson, while teaching the features above, doesn’t explicitly teach, however, Ma, in a similar field of endeavor directed to determining salinity of water in a borehole of a formation, teaches: generating a set of features using a physics-based selection process, wherein the set of features maximizes a correlation between carbon, oxygen, and density measurements to a value of fluid holdup, wherein calibrating the one or more ratios and channels comprises mapping the set of features to the empirical PNL data {generating a set of features using a physics-based selection process, wherein the set of features maximizes a correlation between carbon, oxygen, and density measurements to a value of fluid holdup described in [0048]: From the inelastic spectra 204, a carbon to oxygen ratio 212 may be obtained by using a standards database of inelastic spectra for various elements, as illustrated in FIG. 3. The standards database may be used to determine coefficients for carbon and oxygen where the coefficients represent relative amounts of carbon and oxygen. The relative amounts may be determined at the near and far detectors. Carbon may be representative of the amount of oil, and oxygen may be representative of the amount of water. From the carbon to oxygen ratio 212, an oil saturation value and a water saturation value 218 may be obtained for the formation. From the capture spectra 206, a chlorine to hydrogen ratio 214 may be obtained by using a standards database of capture spectra for various elements, as illustrated in FIG. 4. The capture spectra standards database may be used to determine coefficients for chlorine and hydrogen where the coefficients represent relative amounts of the elements. The relative amounts may be determined at the near and far detectors. Chlorine may be representative of the amount of salt, and hydrogen may be representative of the amount of water and oil. Using the chlorine to hydrogen ratio 214 and a water saturation value 218, a formation water salinity 222 may be calculated. The water saturation value 218 is necessary to provide a correction to the hydrogen value for the calculation. The various data processing 220 methods may use a database for borehole and formation conditions for the PNL tool, where the PNL tool was previously characterized under various borehole and formation conditions.}.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the invention to modify the combination of Al Madani, Whetton, Chen, Boualleg, and Jacobson to include the features of Ma. Given that Al Madani is directed to determining subsurface formations based on collected and generated data, one of ordinary skill in the art would have been motivated to look to Ma, in order to facilitate determining water salinity, which is necessary to compute an accurate estimate of a water fraction in formations {[0003] of Ma}.
Claim 19
Chen further teaches: wherein the instructions to convert, using the one or more calibration coefficients, the simulated PNL data into the lab-equivalent synthetic data comprise: instructions to plot, for each channel of the one or more channels, the simulated PNL data against the empirical PNL data {plotting the simulated data against empirical described in [0057]: FIG. 8 provides a comparison of actual triple combo logs of formation density 64, neutron porosity 65 and deep resistivity 66 to synthetic predictions 68, 69, 70 of the same log parameters. The synthetic versions were generated by inputting seven parameters from a cased hole pulsed neutron logging tool into a neural network ensemble created by the methods described above. The close correlation of the synthetic open hole logs to actual open hole logs indicates that ensembles produced by use of the present invention can accurately produce predictions of the open hole logging parameters.}; instructions to generate a calibration curve to fit the simulated PNL data and the empirical PNL data {As seen in Fig. 8 and [0057].}; instructions to select a set of calibration coefficients based on a function of the calibration curve {As seen in Fig. 8 and [0057].}; instructions to convert the simulated PNL data into the lab-equivalent synthetic data based on the set of calibration coefficients {As seen in Fig. 8 and [0057].}.
The motivation and rationale to include the additional features of Chen is the same as above.
The combination of Al Madani, Whetton, Chen, Boualleg, and Jacobson, while teaching the features above, doesn’t explicitly teach, however, Ma, in a similar field of endeavor directed to determining salinity of water in a borehole of a formation, teaches: instructions to interpolate and extrapolate unknown data values to fill a variable space of the lab-equivalent synthetic data based, at least in part, on the set of calibration coefficients {interpolating and extrapolating unknown data values to fill a variable space described in [0025]: The various formation and borehole conditions used in a laboratory or in a computer simulation may be selected to bracket the conditions that may be found in various field situations of formation and borehole conditions. By bracketing expected field conditions, the correlations may be used to calculate, by interpolation or extrapolation for example, results of field responses of far and near detectors. If a field situation of formation and borehole conditions lies outside a bracket, an extrapolation may be performed to determine field responses of far and near detectors.}.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the invention to modify the combination of Al Madani, Whetton, Chen, Boualleg, and Jacobson to include the features of Ma. Given that Al Madani is directed to determining subsurface formations based on collected and generated data, one of ordinary skill in the art would have been motivated to look to Ma, in order to facilitate determining water salinity, which is necessary to compute an accurate estimate of a water fraction in formations {[0003] of Ma}.
Response to Arguments
Applicant’s response filed 7/23/26 has been fully considered. Examiner’s response follows.
Claim Rejections - 35 USC § 101 (Software Per Se)
Applicant is thanked for their amendments, which overcome the previous Software Per Se rejections of claims 15-20. These rejections are withdrawn.
Claim Rejections - 35 USC § 101 (Abstract Idea)
Regarding the Abstract Idea rejections under 35 USC § 101, applicant offers:
Recent guidance from the United States Patent and Trademark Office further confirms that claims directed to machine-learning technologies should not be evaluated at an unduly generalized level of abstraction. In Ex parte Desjardins (Appeal No. 2024-000567, Sept. 26, 2025), the USPTO Director vacated a § 101 rejection of claims directed to training a machine- learning model. The Director explained that the Board had improperly evaluated the claims at too high a level of abstraction and emphasized that artificial-intelligence and machine-learning innovations should not be categorically treated as abstract ideas. The Director further explained that eligibility analysis must consider the technological context and the specific implementation of the claimed machine-learning techniques.
The present claims, like those at issue in Desjardins, recite a specific technological implementation of machine learning rather than a generalized mathematical concept. The claims recite generating simulated pulsed neutron logging (PNL) data, converting the simulated PNL data into lab-equivalent synthetic data using calibration coefficients determined from calibration curves generated by fitting simulated PNL data and empirical PNL data, and training an ensemble of machine-learning models based on the lab-equivalent synthetic data. Consistent with the Director's guidance in Desjardins, these claims should be evaluated based on their specific technological implementation rather than at a generalized level of abstraction.
The rejected claims are directed to a specific technological method for predicting fluid holdup in a borehole using PNL data. Claim 1 recites generating an expanded dataset of simulated PNL data based, at least in part, on an original dataset of empirical PNL data, converting the simulated PNL data into lab-equivalent synthetic data using calibration coefficients determined from calibration curves generated by fitting the simulated PNL data and the empirical PNL data, and training an ensemble of machine learning models based on the lab- equivalent synthetic data. The claims therefore recite a specific technical process for improving the training of machine learning models used to predict fluid holdup in a borehole environment.
The Office characterizes the claims as reciting mental processes or mathematical concepts. Applicant disagrees. The claims are directed to processing and using PNL data to predict fluid holdup in a borehole. The claims recite generating simulated PNL data, converting the simulated PNL data into lab-equivalent synthetic data using calibration coefficients determined from calibration curves generated by fitting simulated PNL data and empirical PNL data, and training an ensemble of machine learning models based on the lab-equivalent synthetic data. These limitations are directed to improving the interpretation of PNL measurements used in borehole evaluation and fluid-holdup prediction. Accordingly, the claims are directed to a technological improvement in well-logging analysis rather than a mental process performed in the abstract.
While well taken, examiner maintain that an administrator could accomplish the following tasks with pen and paper (examiner notes that the converting step is rudimentary and merely involves the application of previously determined calibration coefficients):
generating an expanded dataset of simulated pulsed neutron logging (PNL) data including simulated spectra corresponding to neutron interactions with formation materials based, at least in part, on an original dataset of empirical PNL data;
converting, using one or more calibration coefficients determined from calibration curves generated by fitting the simulated PNL data and the empirical PNL data, the simulated PNL data into lab-equivalent synthetic data.
The recited steps are broadly claimed and not necessarily technical.
Further, examiner maintains that applicant has not arrived at an improvement to a technology, as seen in Desjardins. Rather, applicant is using off-the-shelf technology, claimed in a results-oriented manner, to facilitate the tasks of an abstract idea. MPEP 2106.05(a), with reference to MPEP 2106.05(f), is clear that this does not constitute an improvement to a technology or technical field, as suggested by applicant. Applicant has described steps that pertain to the observations and evaluations concerning generating a dataset, where these observations and evaluations could also be considered mathematical processes.
Applicant continues:
Even assuming arguendo that the claims recite mathematical operations, the claims satisfy Step 2A Prong Two because they integrate any such mathematical operations into a practical technological application. The claims are not directed to mathematics in the abstract. Instead, the claims apply the recited operations in the technical field of pulsed neutron logging and borehole fluid-holdup prediction. The claimed method uses calibration coefficients determined from calibration curves generated by fitting simulated PNL data and empirical PNL data to convert simulated PNL data into lab-equivalent synthetic data. The resulting lab- equivalent synthetic data are then used to train an ensemble of machine learning models for predicting fluid holdup in a borehole. This is a practical application in well logging and subsurface formation evaluation.
The Office further asserts that the learning machine and ensemble training features merely apply the alleged abstract idea on a generic computer. Applicant disagrees. Claim 1 does not merely recite using a computer to perform a calculation. Claim 1 recites a specific technical process for producing lab-equivalent synthetic PNL data and using that lab-equivalent synthetic data to train an ensemble of machine learning models for borehole fluid-holdup prediction. The claimed process addresses a technical problem in the field of well logging, namely, the limited availability of suitable training data for predicting fluid holdup from PNL data. The claimed process improves the training data used by the learning machine by converting simulated PNL data into lab-equivalent synthetic data before training the ensemble.
Accordingly, the claims do not merely recite an abstract idea implemented on a generic computer. The claims recite a specific technological application of simulated and empirical PNL data to train machine learning models for predicting fluid holdup in a borehole. For at least these reasons, Applicant respectfully submits that the claims are patent eligible under 35 U.S.C. § 101.
Examiner disagrees. Applicant has not provided any detail, in the claims or specification, that the “learning machine” and “ensemble training” features are anything but generic computing elements. These additional elements are claimed in a results-oriented manner and equivalent to “apply it,” per MPEP 2106.05(f). MPEP Whether viewed alone or in combination, this does not integrate the abstract idea into practical application or add significantly more.
For these reasons, examiner maintains the Abstract Idea rejections under 35 USC § 101.
Claim Rejections - 35 USC § 103
Applicant’s remarks concerning the rejections under 35 USC § 103 are predicated on newly incorporated claim features added via amendment, which necessitated the new ground of rejection herein. Accordingly, these arguments are moot. Examiner directs applicant to the claim analysis above.
In summary, examiner has responded to all of applicant’s remarks.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure:
“A novel borehole/annulus holdup calculation method based on pulsed neutron logging” (NPL attached), which teaches: The proposed method is applied to real data measured in an offshore production well. The result confirms that the formation water saturation calculated by the proposed method agrees well with the actual well production status, further confirming that the proposed method has a promising application in residual oil dynamic monitoring.
US 20070011115, which teaches: Logging systems and methods are disclosed to reduce usage of radioisotopic sources. Some embodiments comprise collecting at least one output log of a training well bore from measurements with a radioisotopic source; collecting at least one input log of the training well bore from measurements by a non-radioisotopic logging tool; training a neural network to predict the output log from the at least one input log; collecting at least one input log of a development well bore from measurements by the non-radioisotopic logging tool; and processing the at least one input log of the development well bore to synthesize at least one output log of the development well bore. The output logs may include formation density and neutron porosity logs.
US 20220171087, which teaches: Systems and methods for determining holdup in a wellbore using a neutron-based downhole tool. In examples, the tool includes nuclear detectors that may measure gammas induced by highly energized pulsed-neutrons emitted by a generator. The characteristic energy and intensity of detected gammas indicate the elemental concentration for that interaction type. A detector response may be correlated to the borehole holdup by using the entire spectrum or the ratios of selected peaks. As a result, measurements taken by the neutron-based downhole tool may allow for a two component (oil and water) or a three component (oil, water, and gas) measurement. The two component or three component measurements may be further processed using machine learning (ML) and/or artificial intelligence (AI) with additional enhancements of semi-analytical physics algorithms performed at the employed network's nodes (or hidden layers).
Any inquiry concerning this communication or earlier communications from the examiner should be directed to JOHN SAMUEL WASAFF whose telephone number is (571)270-5091. The examiner can normally be reached Monday through Friday 8:00 am to 6:00 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, SARAH MONFELDT can be reached at (571) 270-1833. 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.
JOHN SAMUEL WASAFF
Primary Examiner
Art Unit 3629
/JOHN S. WASAFF/Primary Examiner, Art Unit 3629
Prosecution Insights