DETAILED ACTION
The following is a non-final Office action (“Action”) in response to U.S. patent application no. 18/264,193 (“the ‘193 application” or “this application”) filed on Aug. 3, 2023. The ‘193 application has also been published as U.S. Patent Application Publication No. 2024/0110469 (hereinafter “PgPub”) and all citations to the specification of the ‘193 application will be to the PgPub.
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 .
Cited and Applied Prior Art
The following is the cited and applied prior art in this Action.
U.S. Patent Application Publication No. 2021/0116598, to Thorne et al. (“Thorne”), which qualifies as prior art under 35 U.S.C. § 102(a)(2).
U.S. Patent Application Publication No. 2021/0123334, to Madasu et al. (“Madasu”), which qualifies as prior art under section 102(a)(2).
Information Disclosure Statement
The information disclosure statements submitted on June 9, 2025, Oct. 17, 2025, and Feb. 1, 2026 are in compliance with the provisions of 37 C.F.R. §§ 1.97 and 1.98 and the information cited has been considered.
Claim Interpretation
The claim limitations are interpreted under a broadest reasonable interpretation standard consistent with the specification. See MPEP § 2111.01. Unless noted here, the interpretations are reflected based on the mappings to the prior art in the rejections below.
The term “a proxy machine learning model” is understood to be a simplified, surrogate of a more complex machine learning model.
The term “a physics-based model” is understood to be a probabilistic model or one that uses mathematical equations and physical laws to simulate or represent natural systems.
Drawings
The drawings are objected to as failing to comply with 37 CFR 1.84(p)(5) because they do not include the following reference characters mentioned in the description and because they include the following reference characters mentioned in the description but not shown in the drawings.
Fig. 1 shows “processor(s) 134” but the reference character “134” is not mentioned in the description; however, paragraph 68 of the PgPub describes “processors 130” with respect to Fig. 1 but the reference character “130” is not shown in Fig. 1.
Fig. 1 shows “electronic storage 132” but the reference character “132” is not mentioned in the description.
The drawings are objected to as failing to comply with 37 CFR 1.84(p)(4) because reference character “1202” has been used to designate different elements of the graphs shown in Fig. 12. Similarly, reference character “1204” has also been used to designate different elements of the graphs shown in Fig. 12.
Corrected drawing sheets in compliance with 37 CFR 1.121(d) are required in reply to the Office action to avoid abandonment of the application. Any amended replacement drawing sheet should include all of the figures appearing on the immediate prior version of the sheet, even if only one figure is being amended. Each drawing sheet submitted after the filing date of an application must be labeled in the top margin as either “Replacement Sheet” or “New Sheet” pursuant to 37 CFR 1.121(d). If the changes are not accepted by the examiner, the applicant will be notified and informed of any required corrective action in the next Office action. The objection to the drawings will not be held in abeyance.
Comment on 35 U.S.C. § 101 – Patent Eligible Subject Matter
Based on the current claim language, the claims are understood to be directed to patent eligible subject matter. While at least independent claims 1, 11, and 18 are generally understood to be directed to “collecting information, analyzing it, and displaying certain results of the collection and analysis,” (see Electric Power Group v. Alstom, S.A., 830 F.3d 1350, 1353-54 (Fec. Cir. 2016)), the claims do not recite limitations that can be practically performed in the human mind, are mathematical in nature, or cover a law of nature. A “proxy machine learning model” cannot be practically trained in the human mind. Even so, the claims recite that the training has the objective of being able to “predict an output of a simulation of a physics-based model of a subsurface volume” and is “based on simulation result generated based on the physics-based model.” While “the physics-based model” is understood to be both mathematical in nature and representative of the physics of “a subsurface volume,” the claim limitations only apply the generated results of the “model” and are not directed to the mathematical relationship or underlying law of nature themselves. For example, the claim limitations are understood to use the generated “simulation results,” which is an application of the “model” and not an attempt to claim the model itself. As a result, claims 1-23, as currently written, are directed to patent eligible subject matter under section 101 based on current Office guidance.
Claim Rejections - 35 USC § 112
The following is a quotation of 35 U.S.C. 112(b):
(b) CONCLUSION.—The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention.
Claims 1-23 are rejected under 35 U.S.C. 112(b) as indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor regards as the invention.
Claims 1, 11, and 18 – “Uncertainty Parameters”
The term “uncertainty parameters” in at least independent claims 1, 11, and 18 is unclear because neither the claims nor the specification offer any explanation, definition, or examples of what are “uncertainty parameters.” See PgPub, ¶¶2, 4, 5, 12, 13, 15, 16, 66, 68, 69, 71, 72, 73, 79, Abstract, Figs. 4, 6. For purposes of examination, the term appears in the prior art, thus, it is assumed to be the same thing.
Claim 6
Claim 6 recites, “the trained proxy machine learning model is used to generate suitable scenarios based on specified criteria.” Neither the claims nor the specification explain, define, or give examples of “suitable scenarios” or what type of “specified criteria” is used to generate the scenarios. The specification does not use the phrase “suitable scenario” anywhere except when reciting the claim language. See e.g., PgPub, ¶9. Moreover, the specification does not appear to describe that “the trained proxy machine learning model” generates the “suitable scenarios,” as recited in claim 6. As a result, it is not clear what “criteria” is used to generate the “suitable scenarios” let alone doing so with “the trained proxy machine learning model.”
Remaining Dependent Claims 2-6, 8, 12-14, and 19-21
The remaining dependent claims 2-6, 8, 12-14, and 19-21, which depend from respective independent claims 1, 11, and 18, are also rejected as indefinite under section 112(b) for the same reasons as presented above.
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.
This application currently names joint inventors. In considering patentability of the claims the examiner presumes that the subject matter of the various claims was commonly owned as of the effective filing date of the claimed invention(s) absent any evidence to the contrary. Applicant is advised of the obligation under 37 CFR 1.56 to point out the inventor and effective filing dates of each claim that was not commonly owned as of the effective filing date of the later invention in order for the examiner to consider the applicability of 35 U.S.C. 102(b)(2)(C) for any potential 35 U.S.C. 102(a)(2) prior art against the later invention.
Claims 1, 7, 9-11, 15-18, 22, and 23 are rejected under 35 U.S.C. 103 as being unpatentable over Thorne in view of Madasu, both of which are in the same field of modeling subsurface volumes as the claimed invention.
Independent Claim 1
Claim 1 recites and Thorne discloses:
A method of calibrating reservoir uncertainty parameters (Thorne, ¶¶72-77, Figs. 2, 3), the method comprising:
training a … machine learning model to predict an output of a simulation of a physics-based model of a subsurface volume, based on simulation results generated based on the physics-based model and historical data (Thorne, ¶¶60, 72-75, Fig. 2, steps 202-216, for a subsurface volume, an initial production model may be, for example, a probabilistic model and is understood as a “physics-based model,” similar to what is described in the PgPub, ¶107, and a production model acts as the “machine learning model” that is to be trained based on the initial production model and training data, i.e., “historical data”);
applying a respective set of uncertainty parameters to the trained proxy machine learning model to generate a solution (Thorne, ¶¶36, 50, 59-61, 72, 75, Fig. 2, steps 206, 214, the trained production model is the “solution”);
returning the generated solution as a solution (Thorne, ¶¶60, 75, Fig. 2, step 214) …; and
visualizing one or more properties of a subsurface volume using the trained proxy model (Thorne, ¶77, Fig. 3, steps 310, 312).
Thorne does not disclose or is silent with respect to the remaining claimed limitations. Even so, Madasu remedies this and teaches that the trained model can be “a proxy machine learning model,” (Madasu, ¶¶19, 40), and that a generated solution may be returned “responsive to determining that a difference between the generated solution and the historical data is less than an error tolerance,” (Madasu, ¶62, Fig. 8, steps 816, 818, where if the error tolerance is acceptable, then the hybrid model (“solution”) is used). It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the methods of Thorne to train a “proxy” model and generate a solution based on “a difference between the generated solution and the historical data is less than an error tolerance,” as in Madasu, to ensure that the generated solution meets “a predetermined convergence criterion” and is, thus, more accurate. See Madasu, ¶¶4, 81.
Independent Claim 11
Claim 11 recites and Thorne discloses:
A computing system for calibrating reservoir uncertainty parameters to obtain production data as close as possible to historical data (Thorne, ¶¶5, 53, 72-77, Figs. 1-3), the computing system comprising:
at least one processor (Thorne, ¶5, Fig. 1, processor(s) 134); and
a memory connected with the at least one processor, the memory including instructions for the at least one processor to perform operations (Thorne, ¶54, Fig. 1, at least machine-readable instructions 106), the operations comprising:
training a … machine learning model to predict an output of a simulation of a physics-based model of a subsurface volume, based on simulation results generated based on the physics-based model and historical data (Thorne, ¶¶60, 72-75, Fig. 2, steps 202-216, for a subsurface volume, an initial production model may be, for example, a probabilistic model and is understood as a “physics-based model,” similar to what is described in the PgPub, ¶107, and a production model acts as the “machine learning model” that is to be trained based on the initial production model and training data, i.e., “historical data”);
applying a respective set of uncertainty parameters to the trained proxy machine learning model to generate a solution (Thorne, ¶¶36, 50, 59-61, 72, 75, Fig. 2, steps 206, 214, the trained production model is the “solution”);
returning the generated solution as a solution (Thorne, ¶¶60, 75, Fig. 2, step 214) …; and
visualizing one or more properties of a subsurface volume using the trained proxy model (Thorne, ¶77, Fig. 3, steps 310, 312).
Thorne does not disclose or is silent with respect to the remaining claimed limitations. Even so, Madasu remedies this and teaches that the trained model can be “a proxy machine learning model,” (Madasu, ¶¶19, 40), and that a generated solution may be returned “responsive to determining that a difference between the generated solution and the historical data is less than an error tolerance,” (Madasu, ¶62, Fig. 8, steps 816, 818, where if the error tolerance is acceptable, then the hybrid model (“solution”) is used). It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the methods of Thorne to train a “proxy” model and generate a solution based on “a difference between the generated solution and the historical data is less than an error tolerance,” as in Madasu, to ensure that the generated solution meets “a predetermined convergence criterion” and is, thus, more accurate. See Madasu, ¶¶4, 81.
Independent Claim 18
Claim 18 recites and Thorne discloses:
A non-transitory computer-readable storage medium having instructions stored thereon for a computer to perform a plurality of operations (Thorne, ¶¶5, 54, Fig. 1, at least machine-readable instructions 106 are executed on processor(s) 134), the plurality of operations comprising:
training a … machine learning model to predict an output of a simulation of a physics-based model of a subsurface volume, based on simulation results generated based on the physics-based model and historical data (Thorne, ¶¶60, 72-75, Fig. 2, steps 202-216, for a subsurface volume, an initial production model may be, for example, a probabilistic model and is understood as a “physics-based model,” similar to what is described in the PgPub, ¶107, and a production model acts as the “machine learning model” that is to be trained based on the initial production model and training data, i.e., “historical data”);
applying a respective set of uncertainty parameters to the trained proxy machine learning model to generate a solution (Thorne, ¶¶36, 50, 59-61, 72, 75, Fig. 2, steps 206, 214, the trained production model is the “solution”);
returning the generated solution as a solution (Thorne, ¶¶60, 75, Fig. 2, step 214) …; and
visualizing one or more properties of a subsurface volume using the trained proxy model (Thorne, ¶77, Fig. 3, steps 310, 312).
Thorne does not disclose or is silent with respect to the remaining claimed limitations. Even so, Madasu remedies this and teaches that the trained model can be “a proxy machine learning model,” (Madasu, ¶¶19, 40), and that a generated solution may be returned “responsive to determining that a difference between the generated solution and the historical data is less than an error tolerance,” (Madasu, ¶62, Fig. 8, steps 816, 818, where if the error tolerance is acceptable, then the hybrid model (“solution”) is used). It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the methods of Thorne to train a “proxy” model and generate a solution based on “a difference between the generated solution and the historical data is less than an error tolerance,” as in Madasu, to ensure that the generated solution meets “a predetermined convergence criterion” and is, thus, more accurate. See Madasu, ¶¶4, 81.
Dependent Claims 7, 15, and 22
Claim 7 recites the “method of claim 1,” claim 15 recites the “computing system of claim 11,” and claim 22 recites the “non-transitory computer-readable storage medium of claim 18.” While Thorne and Madasu make obvious claims 1, 15, and 18, as explained above, and while Thorne describes using machine learning, Thorne is silent as to the specific limitation recited in claims 7, 15, and 22. Madasu remedies this and teaches that “the proxy machine learning model includes at least one of an artificial neural network and a deep learning model.” Madasu, ¶¶18, 80, 83. It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to use “an artificial neural network” or “deep learning model,” as in Madasu, as a “proxy machine learning model,” as in Thorne, especially when combined with a physics-based model, to more accurate simulate the underlying model in cases where data is incomplete or unavailable and to mitigate reductions in system performance to do the amount of information needing to be processed. See id. at ¶4.
Dependent Claims 9, 16, and 23
Claim 9 recites the “method of claim 1,” claim 16 recites the “computing system of claim 11,” and claim 23 recites the “non-transitory computer-readable storage medium of claim 18.” Thorne further discloses, “the proxy machine learning model is trained based on simulation results using a plurality of sets of uncertainty parameters,” as recited in claims 9, 16, and 23. Thorne, ¶¶59, 72, Fig. 2, step 206.
Dependent Claims 10 and 17
Claim 10 recites the “method of claim 1” and claim 17 recites the “computing system of claim 11.” Thorne further discloses the method of claim 10 “further comprising: defining bounds constraints on each of the plurality of uncertainty parameters to limit a solutions space to feasible solutions,” and “the plurality of operations [of claim 11] further comprise: defining bounds constraints on each of the plurality of uncertainty parameters to limit a solutions space to feasible solutions.” Thorne, ¶77, Fig. 3, step 306, “target geological parameter uncertainty values” are bounds placed on the model to determine feasible solutions.
Allowable Subject Matter
Claims 2-6, 8, 12-14, and 19-21 would be allowable over the cited prior art of record if they are rewritten to overcome the rejections under 35 U.S.C. 112(b) set forth above in this Action, and to include all of the limitations of the base claim and any intervening claims.
A statement on reasons for allowance will not be made at this time, however, since no independent claims stand allowed.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure. The following prior art documents all describe various machine learning and/or modeling of subsurface volumes:
U.S. Patent Application Publication No. 2023/0409783.
U.S. Patent Application Publication No. 2023/0161061.
U.S. Patent Application Publication No. 2023/0082567.
U.S. Patent Application Publication No. 2023/0026857.
U.S. Patent Application Publication No. 2022/0170359.
U.S. Patent Application Publication No. 2021/0310345.
U.S. Patent Application Publication No. 2021/0262329.
U.S. Patent Application Publication No. 2021/0247534.
U.S. Patent Application Publication No. 2020/0226422.
U.S. Patent Application Publication No. 2020/0183035.
U.S. Patent Application Publication No. 2020/0184374.
U.S. Patent Application Publication No. 2020/0041692.
U.S. Patent Application Publication No. 2013/0124171.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to JOSHUA KADING whose telephone number is (571)270-3413. The examiner can normally be reached Monday-Friday, 8:00 AM to 5:00 PM Eastern Time.
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, Eileen Lillis can be reached at 571-272-6928. 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.
/JOSHUA KADING/ Primary Patent Examiner, Art Unit 3993
Prosecution Insights