DETAILED ACTION
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and the fee set forth in 37 CFR 1.17(e) has been timely paid, the finality of the previous Office action has been withdrawn pursuant to 37 CFR 1.114.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 03/17/2026 has been entered.
Claims 1 – 20 have been presented for examination. Claims 1, 10 and 20 are currently amended.
The instant office action relies on Eslinger et al. WO 2018/165352) which is cited on the IDS.
Response to Claim Objections
Applicant’s amendments overcome the claim objection. Therefore, it is withdrawn.
Response to Rejections Under 35 U.S.C. § 112
Applicant’s arguments have been fully considered and they are persuasive. Therefore, the 112(b) rejection is withdrawn. However, a new 112(b) is included which is necessitated by the amendments (see Claim Rejections - 35 U.S.C. § 112).
Response to Rejections Under 35 U.S.C. § 101
Applicant’s arguments have been fully considered, and they are persuasive. Therefore, the 101 rejection is withdrawn.
Specifically, the instant claims 1 (and similarly claim 10) is viewed as not reciting an abstract idea under step 2A(ii) since there is recited “generating a digital twin of the hydrocarbon system by: (i) instantiating the one or more ROMs at an operational point of the hydrocarbon system, (ii) calibrating the digital twin based on data from an on-site test at the hydrocarbon system, and (iii) configuring the digital twin to use real-time data obtained from the hydrocarbon system” which is not practically performed in the mind, and which does not explicitly recite mathematical concepts. Examiner notes that the digital twin is generated from the ROMs which are instantiated and calibrated and configured, which does not explicitly recite any mathematical algorithms or formulas for effectuating the claimed “generating”.
Specifically, the instant claim 20 is viewed as not reciting an abstract idea under step 2A(ii) since there is recited “generate a digital twin of the hydrocarbon system based on the plurality of quasistatic parameters and the plurality of ROMs” which is not practically performed in the mind, and which does not explicitly recite mathematical concepts. Examiner notes that the digital twin is generated from the ROMs, which does not explicitly recite any mathematical algorithms or formulas for effectuating the claimed “generating”.
Response to Rejections Under 35 U.S.C. § 103
Applicant’s arguments have been fully considered. However, the Office does not consider them to be persuasive (see Claim Rejections - 35 USC § 103 for the detailed mapping).
Applicant argues: “Such an indication typically signals that the previously-presented§ 103 amendment was overcome, such that the further search and consideration is needed. Accordingly, with such amendments now carried forward into the pending claims at this time, the previously-presented rejection of Claim 1 is believed to be moot. Claim 10 was similarly amended, such that the rejection of Claim 10 is also believed to be moot.”
Examiner notes that After-Final responses do not produce a detailed Office Actions compared to earlier actions having a full and complete search and fully considering all the claim elements. Examiner notes the Advisory Action clearly states that the scope is changed, which necessarily requires further search and consideration prior to Allowance, and does not in any way disparage the instant rejection or the references relied upon therein. The Advisory Action did not explicitly state that the 103 rejection was overcome, and no agreement was reached during any interview that the 103 rejection had been overcome.
Applicant argues: “Claim 1 was amended with a grammatical edit emphasizing that "calibrating the digital twin based on data from an on-site test at the hydrocarbon system" is part of "generating a digital twin of the hydrocarbon system" along with the "instantiating" and "configuring" steps (rather than "generating" and "calibrating" being sequential steps as interpreted by the Examiner). Claim 1 recites:
Because the Examiner's rejection relies instead on interpreting the "calibrating the digital twin based on data from an on-site test at the hydrocarbon system" as corresponding to routine operations during online use after generating the digital twin ("where the calibration of the digital twin is after the generation" OA at 9 as quoted above), as allegedly taught in Filippov, the rejection of Claim 1 is moot and should be withdrawn.’ (emphasis added)
Examiner notes that claims are interpreted under their broadest reasonable interpretation during prosecution (see MPEP 2111). Therefore, the previous claims were interpreted as the generating and calibrating being sequential steps which is broader than the instant claims (i.e., instantiating and calibrating further limit the “generating”). Examiner further notes that Applicant explicitly argues that Filipov is merely related to routine operations during online use after generating the digital, and that the Office never argued and/or limiting Filipov to such as interpretation (See Applicant’s remarks, dated 07/14/2025, Page 14 “Instead, it appears to relate to routine operations during online use of the digital twin for correction over time.”). Instead, the Office argued that the previous claim scope covered the steps being performed sequentially. Therefore, Applicant’s arguments are not persuasive.
Applicant argues: “The amendment emphasizes that the on-site test (as in "calibrating the digital twin based on data from an on-site test at the hydrocarbon system") is "executed at the hydrocarbon system independent of the digital twin." As observed above, these steps are provided as part of "generating" the digital twin.
In Filipov, "Output predictions of the digital twin are used to modify operational parameters of an oil or gas recovery process." Abstract. See, also, all independent claims, [0076]-[0078]. Because the outputs of the digital twin are being used to modify operations of the oil or gas recovery process, the physical system (e.g., well site) in Filipov is affected by the output of the digital twin, such that the measured data 140 in FIG. 1 of Filipov is a measurement of physical conditions affected by the output of the digital twin. As such, one of skill in the art would understand the data fed back in for reinforcement learning of Filipov's model to be dependent on prior outputs of the digital twin. Thus, Filipov does not explicitly teach that a calibration of the digital twin would be based on from an on-site test at the hydrocarbon site executed independent of the digital twin.” (emphasis added)
Applicant argues that the digital twin is used to modify a physical oil or gas recovery process, therefore, the measured data is not independent of the digital twin. However, Filipov merely teaches that the digital twin may be used to execute an oil or gas recovery process after receiving measurements (see Paragraph 57 and Figure 7 “For example, the corrections to the process parameters may be used as input parameters for the digital twin to adapt subsequent executions of the digital twin to the physical process parameter measurements. The process parameter predictions generated by the digital twin may be used to execute the oil or gas recovery process.”). Therefore, Applicant’s arguments are not persuasive.
Applicant argues: “Additionally, Applicant notes that Filippov is not cited for any particular "on-site test," but rather simply for measuring data during ongoing operations. Applicant respectfully submits the cited passages of Filippov therefore do not disclose "wherein the on-site test is executed at the hydrocarbon system independent of the digital twin." There does not appear to be any affirmative disclosure of an "on-site test executed at the hydrocarbon site."” (bold emphasis added) (italicized emphasis in original)
Applicant argues that Filipov does not teach a “particular” on-site test. Examiner notes that the claim is not limited to any particular test, and merely requires that it be “on-site”. Further, an on-site test is necessarily performed at the same hydrocarbon system to which the digital twin corresponds. Therefore, Filipov teaches “wherein the on-site test is executed at the hydrocarbon system“ since the measured sensor data can come from the site to monitor desired parameters (see Paragraph 22). Therefore, Applicant’s arguments are not persuasive.
Applicant argues: “Independent Claim 20 is amended to recite "wherein the on-site test is executed at the hydrocarbon system independent of the plurality of ROMs." For similar reasons as discussed above, the cited art does not appear to affirmatively disclose that an "on-site test is executed at the hydrocarbon system independent of the plurality of ROMs." Claim 20 is therefore patentable over the cited art.
Additionally, with respect to Claim 20, the Advisory Action partially responds to Applicant's prior remarks, but also states that "Regarding the remaining arguments over Zagayevsky and Eslinger, further search and consideration is required." Such a note also signifies that the rejection of Claim 20 is moot and may be replaced following such further search and consideration. The following passages repeat are based on those previously submitted remarks.”
Applicant’s arguments are not persuasive based on the preceding remarks.
Applicant argues: “Page 57 of the Office Action cites Zagayevsky's "predicted field production" with respect to the claimed "plurality of quasi-static parameters" which "converge to stable values," in particular with paragraph 22 of Zagayevsky cited for a model which is updated so that its predictions match observed values. However, reducing error in predicting a parameter does not suggest that the parameter being predicted converged to a stable value; rather, it suggests that the performance of the predictive model has improved. In particular, if the observed value being predicted is changing over time (which one of skill in the art might expect given that there might otherwise be no reason to predict it), that the prediction being output is not converging to a stable value, but rather become more accurate at predicting changes in a value” (emphasis added)
Applicant argues that predictions matching observed values does not read on "the quasi-static parameters configured to converge to sable values". Examiner notes that Zagayevsky explicitly teaches a changing value (i.e., the predictions as iteratively improved) are adjusted to match a fixed value (i.e., the observed values). Applicant presents a hypothetical arguments that the observed values change over time which “one of skill in the art might be expect given that there might otherwise be no reason to predict it”. Such a hypothetical arguments is not directly linked to the teachings of Zagayesky, nor does Applicant argue that Zagaeysky reasonably requires such an interpretation. Therefore, Applicant’s arguments are not persuasive.
Applicant argues: “Next, as claimed, a digital twin of the hydrocarbon system is determined based on the quasi-static parameters. Page 58 of the Office Action cites the preceding paragraph of Zagayevsky (paragraph 21) for disclosure of a digital representation of the reservoir, i.e., for the model that predicts the field production cited as the "plurality of quasi-static parameters." It is unclear how the model that output the predicted field production could be interpreted as being based on the predicted field production. Thus, the cited disclosure does not seem to align with the combination of claimed features.”
Applicant appears to argue that a digital twin generated based on its output is not within the broadest reasonable interpretation of the recited limitation “generate a digital twin of the hydrocarbon system based on the plurality of quasistatic parameters and the plurality of ROMs”. Examiner that the digital representation of the reservoir is iteratively generated until the predicted values converge. Therefore, the final digital representation is based on its output which is not in any way precluded by the recited limitation since it is merely “based on the plurality of quasi-static parameters” which does not require them to be inputs to the digital representation. Therefore, Applicant’s arguments are not persuasive.
Applicant argues: “Page 58 of the Office Action also cites Eslinger's production parameters as the claimed "quasi-static parameters." However, for the same reasons above, predicted production is not disclosed as being "quasi-static parameters configured to converge to stable values." If predicted production were found to be stable values in the cited art, it would not be useful to have the tools of those references for predicting production (as the stable value could simply be observed and then known).”
Applicant’s arguments are not persuasive based on the preceding remarks.
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.
The following is a quotation of 35 U.S.C. 112 (pre-AIA ), second paragraph:
The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.
Claim 20 rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), second paragraph, as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor (or for applications subject to pre-AIA 35 U.S.C. 112, the applicant), regards as the invention.
With regard to claim 20, it recites “the on-site test” in “wherein the on-site test is executed at the hydrocarbon system independent of the plurality of ROMs” which has insufficient antecedent basis for this limitation in the claim since there is no previously-recited “one-site test”. Examiner notes that independent claims 1 and 10 recite the same limitation, however, they also recite “calibrating the digital twin based on data from an on-site test at the hydrocarbon system”. The limitation is interpreted for examination purposes as any “on-site test” performed which does not depend on the plurality of ROMs
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 set forth in Graham v. John Deere Co., 383 U.S. 1, 148 USPQ 459 (1966), that are applied for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
Determining the scope and contents of the prior art.
Ascertaining the differences between the prior art and the claims at issue.
Resolving the level of ordinary skill in the pertinent art.
Considering objective evidence present in the application indicating obviousness or nonobviousness.
Claims 1, 7 and 10 - 13 are rejected under 35 U.S.C. 103 as being unpatentable over Zagayevsky et al. (US 2021/0133375) (henceforth “Zagayevsky (375)”) in view of Filippov et al. (US 2021/0404314) (henceforth “Filippov (314)”). Zagayevsky (375) and Filippov (314) are analogous art because they solve the same problem of modeling a hydrocarbon system, and because they are from the same field of endeavor of oil and gas exploration.
With regard to claim 1, Zagayevsky (375) teaches a method for generating and using a digital twin of a hydrocarbon system, the method comprising: (Paragraph 21 representation of a reservoir is obtained and can be analyzed)
performing a plurality of simulations to create a hyperdimensional space mapping simulation inputs, outputs and attributes; using the hyperdimensional space from the plurality of simulations to generate one or more reduced order models (ROMs) using a regression technique or a machine learning technique; (Figure 2A design of experiments is used to generated proxy models using machine learning
PNG
media_image1.png
597
875
media_image1.png
Greyscale
)
generating a digital twin of the hydrocarbon system by: (i) instantiating the one or more ROMs, and (Paragraph 21 a digital representation of the reservoir is obtained based on the model order reduction and assisted history matching)
estimating values of one or more variables of the hydrocarbon system using the digital twin; and (Paragraph 28 best field operational settings are determined)
Zagayevsky (375) does not appear to explicitly disclose: that the generating is by: (i) instantiating at an operational point of the hydrocarbon system, (ii) calibrating the digital-twin based on data from an on-site test at the hydrocarbon system, and (iii) configuring the digital twin to use real-time data obtained from the hydrocarbon system; that the estimated values are in real-time; controlling the hydrocarbon system based on the estimated values of the one or more variables.
However, Filippov (314) teaches:
generating a digital twin of a hydrocarbon system by: (i) instantiating a model at an operational point of the hydrocarbon system, (ii) calibrating the digital twin based on data from an on-site hydrocarbon system, and (iii) configuring the digital twin to use real-time data obtained from the hydrocarbon system (Figure 1 digital twin tracks the evolution of a hydrocarbon system based on initial best guess (generated by instantiating at operational point)
PNG
media_image2.png
558
599
media_image2.png
Greyscale
, and Figure 1 digital twin is continuously updated based on measured data (generating by calibrating the digital twin), and Paragraph 22 the sensor data can come from the site to monitor desired parameters (calibrating based on data from on-site test at the hydrocarbon system, and generating by configuring to use real-time data) “Additional sensors (not shown) may be disposed on the drilling arrangement (e.g., on the wellhead) to monitor process parameters, for example, but not limited to, production fluid viscosity, density, etc.”)
wherein the on-site test is executed at the hydrocarbon system independent of the digital twin; (Paragraph 57 and Figure 7 the digital twin may be used to execute an oil or gas recovery process after receiving measurements “For example, the corrections to the process parameters may be used as input parameters for the digital twin to adapt subsequent executions of the digital twin to the physical process parameter measurements. The process parameter predictions generated by the digital twin may be used to execute the oil or gas recovery process.”).
configuring the digital twin to use real-time data obtained from the hydrocarbon system; (Figure 1 digital twin is continuously updated based on the measured data)
estimating values of one or more variables of the hydrocarbon system in real-time using the digital twin; and (Paragraph 16 “using streaming measurable data from multiple sensors can produce a self-adapting digital twin that can be used in real-time applications.”)
controlling or optimizing the hydrocarbon system based on the estimated values of the one or more variables. (Paragraph 57 predicted process parameters are used to execute a recovery process)
It would have been obvious to one of ordinary skill in the art to combine the proxy model generation for reservoir modeling disclosed by Zagayevsky (375) with the digital twin updating of a hydrocarbon system disclosed by Filippov (314). One of ordinary skill in the art would have been motivated to make this modification in order to automatically update a digital twin to reflect new measured data (Filippov (314) Figure 1).
With regard to claim 10, Zagayevsky (375) teaches a system for generating and using a digital twin of a hydrocarbon system, the system comprising: a processor configured to: (Paragraph 21 representation of a reservoir is obtained and can be analyzed, and Figure 7 using a processor)
perform a plurality of simulations in a hyperdimensional space to generate outputs; use the outputs of the plurality of simulations to generate one or more reduced order models (ROMs) using a regression technique or a machine learning technique; (Figure 2A design of experiments is used to generated proxy models using machine learning
PNG
media_image1.png
597
875
media_image1.png
Greyscale
)
generate a digital twin of the hydrocarbon system by: (i) instantiating the one or more ROMs, and (Paragraph 21 a digital representation of the reservoir is obtained based on the model order reduction and assisted history matching)
operate the hydrocarbon system based on outputs of the digital twin. (Paragraph 28 reservoir equipment is optimally set according the optimal reservoir model)
Zagayevsky (375) does not appear to explicitly disclose: that the generating is by: (i) instantiating at an operational point of the hydrocarbon system, (ii) calibrating the digital-twin based on data from an on-site test at the hydrocarbon system, and (iii) configuring the digital twin to use real-time data obtained from the hydrocarbon system.
However, Filippov (314) teaches:
generating a digital twin of a hydrocarbon system by: (i) instantiating a model at an operational point of the hydrocarbon system, (ii) calibrating the digital twin based on data from an on-site hydrocarbon system, and (iii) configuring the digital twin to use real-time data obtained from the hydrocarbon system (Figure 1 digital twin tracks the evolution of a hydrocarbon system based on initial best guess (generated by instantiating at operational point)
PNG
media_image2.png
558
599
media_image2.png
Greyscale
, and Figure 1 digital twin is continuously updated based on measured data (generating by calibrating the digital twin), and Paragraph 22 the sensor data can come from the site to monitor desired parameters (calibrating based on data from on-site test at the hydrocarbon system, and generating by configuring to use real-time data) “Additional sensors (not shown) may be disposed on the drilling arrangement (e.g., on the wellhead) to monitor process parameters, for example, but not limited to, production fluid viscosity, density, etc.”)
wherein the on-site test is executed at the hydrocarbon system independent of the digital twin; (Paragraph 57 and Figure 7 the digital twin may be used to execute an oil or gas recovery process after receiving measurements “For example, the corrections to the process parameters may be used as input parameters for the digital twin to adapt subsequent executions of the digital twin to the physical process parameter measurements. The process parameter predictions generated by the digital twin may be used to execute the oil or gas recovery process.”).
It would have been obvious to one of ordinary skill in the art to combine the proxy model generation for reservoir modeling disclosed by Zagayevsky (375) with the digital twin updating of a hydrocarbon system disclosed by Filippov (314). One of ordinary skill in the art would have been motivated to make this modification in order to automatically update a digital twin to reflect new measured data (Filippov (314) Figure 1).
With regard to claim 7, Zagayevsky (375) in view of Filippov (314) teaches all the elements of the parent claim 1, and further teaches:
performing the on-site test at the hydrocarbon system to obtain calibration data, the calibration data independent of outputs of the digital twin; and further comprising at least one: re-generating the one or more ROMs and re-generating the digital twin using the calibration data to generate a calibrated digital twin; or providing the calibration data to a calibration ROM for use in updating one or more other ROMs of the digital twin. (Filippov (314) Figure 1 digital twin is continuously updated based on measured data (calibrating the digital twin), where the measured data is directly measured (independent of outputs of the digital twin), and Paragraph 22 the sensor data can come from the site to monitor desired parameters (an on-site test at the hydrocarbon system) “Additional sensors (not shown) may be disposed on the drilling arrangement (e.g., on the wellhead) to monitor process parameters, for example, but not limited to, production fluid viscosity, density, etc.”)
It would have been obvious to one of ordinary skill in the art to combine the proxy model generation for reservoir modeling disclosed by Zagayevsky (375) with the digital twin updating of a hydrocarbon system disclosed by Filippov (314). One of ordinary skill in the art would have been motivated to make this modification in order to automatically update a digital twin to reflect new measured data (Filippov (314) Figure 1).
With regard to claim 11, Zagayevsky (375) in view of Filippov (314) teaches all the elements of the parent claim 10, and further teaches:
estimate values of one or more variables of the hydrocarbon system in real-time using the digital twin and real-time data. (Zagayevsky (375) Paragraph 28 best field operational settings are determined)
With regard to claim 12, Zagayevsky (375) in view of Filippov (314) teaches all the elements of the parent claim 11, and further teaches:
wherein the real-time data comprises at least one of a wellhead pressure, a flow line pressure, an injection pressure, an injection rate, a pump discharge pressure, a pump intake pressure, a voltage, a current, or a motor temperature of the hydrocarbon system. (Filippov (314) Paragraph 17 measurable values comprise pressures in the reservoir or well bore “a digital twin can be characterized by a state vector that includes unknown constant and transient parameters (for example, reservoir permeability, or fracture dimensions) which can be defined based on a set of observations (e.g., predictions of measurable parameters, such as flow rates, temperatures, and pressures in reservoir or well bore).”, and Paragraph 23 a flow line exists in the borehole (flow line pressure))
It would have been obvious to one of ordinary skill in the art to combine the proxy model generation for reservoir modeling disclosed by Zagayevsky (375) with the digital twin updating of a hydrocarbon system disclosed by Filippov (314). One of ordinary skill in the art would have been motivated to make this modification in order to automatically update a digital twin to reflect new measured data (Filippov (314) Figure 1).
With regard to claim 13, Zagayevsky (375) in view of Filippov (314) teaches all the elements of the parent claim 10, and further teaches:
perform an automatic calibration check of the digital twin, wherein performing the automatic calibration check comprises: obtaining measurements of one or more predictor variables of the hydrocarbon system;
determining one or more quasi-static parameters using the measurements of the one or more predictor variables, the quasi-static parameters configured to converge to stable values; determining if the one or more quasi-static parameters have converged; in response to determining that the one or more quasi-static parameters have not converged, adjusting one or more of the predictor variables and re-determining the one or more quasi-static parameters and re-determining if the one or more quasi-static parameters have converged. (Zagayevsky (375) Paragraph 27 sensitivity analysis is used to determine input parameters to the reservoir model having a threshold dependency on the output variables (predictor variables), and field production information is used (obtaining measurement of) to compute the output predictions (determining quasi-static parameters using the measurements), until the differences the predicted and measured values are accounted for (quasi-static parameters converge to stable values, in response to determining have converged, adjusting and re-determining))
Claims 2 – 6 and 14 are rejected under 35 U.S.C. 103 as being unpatentable over Zagayevsky (375) in view of Filippov (314), and further in view of Eslinger et al. (WO 2018/165352) (henceforth “Eslinger (352)”). Zagayevsky (375) and Filippov (314) and Eslinger (352) are analogous art because they solve the same problem of modeling a hydrocarbon system, and because they are from the same field of endeavor of oil and gas exploration.
With regard to claim 2, Zagayevsky (375) in view of Filippov (314) teaches all the elements of the parent claim 1, and further teaches wherein the one or more ROMs comprise:
a forward ROM; (Zagayevsky (375) Paragraph 22 a reduced parameter space reservoir model is used for field development optimization, and Paragraph 24 the proxy flow simulator for FDO can be used to generate reservoir model outputs use less computational resources)
an inverse ROM; or (Zagayevsky (375) Paragraph 21 the proxy flow model is used for generating the reservoir model (inverse ROM))
Zagayevsky (375) in view of Filippov (314) does not appear to explicitly disclose: a calibration ROM.
However, Eslinger (352) teaches:
a calibration ROM (Paragraph 174 measured sensor values can be improved with a model for observing state variables, and Paragraph 222 a calibrated physical model can be a hybrid based on analytics of sensors signals, and Paragraph 214 models can be trained on sensor data to provide a more accurate value compared to the value produced by an inversion process (a calibration ROM))
It would have been obvious to one of ordinary skill in the art to combine the proxy model generation and updating for reservoir modeling disclosed by Zagayevsky (375) in view of Filippov (314) with the model implemented sensor in a reduced order model disclosed by Eslinger (352). One of ordinary skill in the art would have been motivated to make this modification in order to more accurately computed a desired value in a hydrocarbon system (Eslinger (352) Paragraph 214).
With regard to claim 3, Zagayevsky (375) in view of Filippov (314), and further I view of Eslinger (352) teaches all the elements of the parent claim 2, and further teaches:
wherein the forward ROM is configured to predict values of one or more variables of the system for a hypothetical scenario based on values of one or more controllable variables, values of one or more configuration parameters, and (Zagayevsky (375) Paragraph 28 the field development optimization model is used to find a best vase operation scenario based on operations parameters (controllable variables) and well placement (configuration parameters))
values of one or more calibration variables. (Zagayevsky (375) Paragraph 27 the reservoir model is calibrated using field production data)
With regard to claim 4, Zagayevsky (375) in view of Filippov (314), and further I view of Eslinger (352) teaches all the elements of the parent claim 2, and further teaches:
wherein the inverse ROM is configured to solve an inversion problem (Eslinger (352) Paragraph 101 calibrating the reservoir model of Zagayevsky using the proxy flow simulator for assisted history matching is an inversion problem “An inverse problem can starts with results and then calculate causes, which is opposite of a forward problem that starts with causes and then calculates results (e.g., consider a forward simulation in time, etc.).”)
to estimate values of one or more system variables of the hydrocarbon system based on values of one or more controllable variables, values of one or more configuration parameters, values of one or more calibration variables, and values of one or more measured variables of the hydrocarbon system. (Zagayevsky (375) Figure 2A data for history matching includes rock properties (configuration parameters) and lab tests (measured variables) and field production (calibrations variables), and Paragraph 28 the reservoir modeling optimizes for operator controlled elements (controllable variables))
It would have been obvious to one of ordinary skill in the art to combine the proxy model generation and updating for reservoir modeling disclosed by Zagayevsky (375) in view of Filippov (314) with the inversion problem disclosed by Eslinger (352). One of ordinary skill in the art would have been motivated to make this modification in order to compute a desired value in a hydrocarbon system (Eslinger (352) Paragraph 214).
With regard to claim 5, Zagayevsky (375) in view of Filippov (314), and further in view of Eslinger (352) teaches all the elements of the parent claim 2, and further teaches:
wherein the calibration ROM is configured to estimate one or more calibration variables for the forward ROM and the inverse ROM based on values of one or more unmeasurable variables (Eslinger (352) Paragraph 201 and Paragraph 174 digital twin can show computed sensor values, where the digital twin can be a ROM, and Paragraph 179 – 180 computed sensor values can be outputted to a historian and are usable for forward modeling and offline simulations, where computed sensor values can be used in the forward/inverse ROM even if they are not actually measured (based on unmeasured variables), and Paragraph 27 sensor values can calibrate physical models (estimate calibration variables))
based on values of one or more controllable variables, values of one or more configuration parameters, values of one or more measured variables. (Eslinger (352) Paragraph 229 the enhanced digital twin model takes into the operating status based on operational parameters (controllable variables), and Paragraph 179 – 180 includes sensors that exist (measured variables), and Paragraph 185 the as-built data of the device is utilized (configuration parameters))
It would have been obvious to one of ordinary skill in the art to combine the proxy model generation and updating for reservoir modeling disclosed by Zagayevsky (375) in view of Filippov (314) with the computed sensor values disclosed by Eslinger (352). One of ordinary skill in the art would have been motivated to make this modification in order to more accurately compute a desired value in a hydrocarbon system (Eslinger (352) Paragraph 214).
With regard to claim 6, Zagayevsky (375) in view of Filippov (314) teaches all the elements of the parent claim 1, and further teaches:
wherein the digital twin comprises an instantiation of a plurality of ROMs,
wherein one or more of the plurality of ROMs are configured to provide an output to a different ROM of the plurality of ROMs as an input. (Eslinger (352) Paragraph 179 - 180 An output of a computed sensor from a ROM can be used forward modeling, where the forward modeling could use the forward ROM of Zagayevsky (375))
It would have been obvious to one of ordinary skill in the art to combine the proxy model generation and updating for reservoir modeling disclosed by Zagayevsky (375) in view of Filippov (314) with the computed sensor values disclosed by Eslinger (352). One of ordinary skill in the art would have been motivated to make this modification in order to more accurately compute a desired value in a hydrocarbon system (Eslinger (352) Paragraph 214).
With regard to claim 14, Zagayevsky (375) in view of Filippov (314) teaches all the elements of the parent claim 10, and further teaches:
perform an automatic calibration check of the digital twin, wherein performing the automatic calibration check comprises: determining a predicted response using one or more quasi-static parameters; the quasi-static parameters configured to converge to stable values; comparing the predicted response to an actual response of the hydrocarbon system to determine a prediction error of the response; and minimizing the prediction error by adjusting the uncertainty parameters and redetermining the predicted response, and re-comparing the predicted response to the actual response to determine the prediction error of the response.(Zagayevsky (375) Paragraph 27 uncertainty parameters are used to compute the output predictions (determining predicted responses using quasi-static parameters), which are adjusted until the differences the predicted and measured values are accounted for (quasi-static parameters converge to stable values, and comparing), and the difference can be minimized using Bayesian optimization (minimizing the prediction error))
Zagayevsky (375) in view of Filippov (314) does not appear to explicitly disclose: obtaining measurements of one or more predictor variables of the hydrocarbon system; determining one or more quasi-static parameters using the measurements of the one or more predictor variables.
However, Eslinger (352) teaches:
obtaining measurements of one or more predictor variables of the hydrocarbon system; determining one or more quasi-static parameters using the measurements of the one or more predictor variables, the quasi-static parameters configured to converge to stable values; (Paragraph 205 measured electrical signals of an ESP (obtaining measurements) can be translated to other variables related to fluid production (quasi-static parameters), and Paragraph 215 the other variables are related to fluid production and further have an uncertainty which is minimized (quasi-static parameters converge to stable values), and Paragraph 107 inputs from artificial lift systems can be used for forward modeling production rates such as those in in Zagayevsky (375))
It would have been obvious to one of ordinary skill in the art to combine the proxy model generation and updating for reservoir modeling disclosed by Zagayevsky (375) in view of Filippov (314) with the modeling unknown variables from measurements disclosed by Eslinger (352). One of ordinary skill in the art would have been motivated to make this modification in order to more accurately computed a desired value in a hydrocarbon system (Eslinger (352) Paragraph 214).
Claims 8 - 9 are rejected under 35 U.S.C. 103 as being unpatentable over Zagayevsky (375) in view of Filippov (314), and further in view of Formentin et al. “Systematic Uncertainty Reduction for Petroleum Reservoirs Combining Reservoir Simulation and Bayesian Emulation Techniques” (henceforth “Formentin”). Zagayevsky (375) and Filippov (314) and Formentin are analogous art because they solve the same problem of modeling a hydrocarbon system, and because they are from the same field of endeavor of oil and gas exploration.
With regard to claim 8, Zagayevsky (375) in view of Filippov (314) teaches all the elements of the parent claim 7, and does not appear to explicitly disclose: where the re-generation comprises an automated selection of an optimal combination of both a sampling scheme, and ROM model and parameters, simultaneously in the presence of calibration measurements and a calibration match criteria.
However, Formentin teaches:
where the re-generation comprises an automated selection of an optimal combination of both (Page 4, Top “An iterative process is implemented in order to narrow the searching space sequentially … This data set is used to construct additional and more accurate emulators leading to further space reduction.”)
a sampling scheme, and (Page 3, Bottom sampling is based on rules which are automatically generated using Bayesian history matching “We rule out regions of the search space identified as implausible, and what remains is only the proportion of the original space currently judged as non-implausible.”)
ROM model and parameters, (Figure 14 specific emulators can be desirably selected
PNG
media_image3.png
256
600
media_image3.png
Greyscale
)
simultaneously in the presence of calibration measurements and a calibration match criteria. (Figure 4 iterations end with criteria met
PNG
media_image4.png
631
676
media_image4.png
Greyscale
)
It would have been obvious to one of ordinary skill in the art to combine the proxy model generation and updating for reservoir modeling disclosed by Zagayevsky (375) in view of Filippov (314) with the Bayesian emulation techniques disclosed by Formentin. One of ordinary skill in the art would have been motivated to make this modification in order to reduce uncertainty in reservoir prediction (Formentin Abstract).
With regard to claim 9, Zagayevsky (375) in view of Filippov (314), and further in view of Formentin teaches all the elements of the parent claim 8, and further teaches:
wherein the re-generation comprises an iterative process and a Bayesian regularization technique. (Formentin Page 3, Bottom sampling is based on rules which are automatically generated using Bayesian history matching “We rule out regions of the search space identified as implausible, and what remains is only the proportion of the original space currently judged as non-implausible.”)
It would have been obvious to one of ordinary skill in the art to combine the proxy model generation and updating for reservoir modeling disclosed by Zagayevsky (375) in view of Filippov (314) with the Bayesian emulation techniques disclosed by Formentin. One of ordinary skill in the art would have been motivated to make this modification in order to reduce uncertainty in reservoir prediction (Formentin Abstract).
Claims 15 – 17 are rejected under 35 U.S.C. 103 as being unpatentable over Zagayevsky (375) in view of Filippov (314), and further in view of Eslinger (352), and further in view of Silvia et al. “Review, Analysis and Comparison of Intelligent Well Monitoring Systems” (henceforth “Silvia”). Zagayevsky (375) and Filippov (314) and Eslinger (352) and Silvia are analogous art because they solve the same problem of modeling a hydrocarbon system, and because they are from the same field of endeavor of oil and gas exploration.
With regard to claim 15, Zagayevsky (375) in view of Filippov (314) teaches all the elements of the parent claim 10, and does not appear to explicitly disclose: wherein at least one of the ROMs of the digital twin comprises a transient ROM, wherein the transient ROM comprises a filter function, the parameters of the filter function being one or more quasi-static parameters output by a calibration ROM.
However, Eslinger (352) teaches:
wherein at least one of the ROMs of the digital twin comprises a transient ROM, (Paragraph 204 can use tROMs)
wherein the transient ROM comprises a filter function, (Paragraph 180 data inputted to the tROM can be filtered)
one or more quasi-static parameters output by a calibration ROM (Paragraph 205 measured electrical signals of an ESP (obtaining measurements) can be translated to other variables such as water-cut (quasi-static parameters))
It would have been obvious to one of ordinary skill in the art to combine the proxy model generation and updating for reservoir modeling disclosed by Zagayevsky (375) in view of Filippov (314) with the computed sensor values disclosed by Eslinger (352). One of ordinary skill in the art would have been motivated to make this modification in order to more accurately compute a desired value in a hydrocarbon system (Eslinger (352) Paragraph 214).
Zagayevsky (375) in view of Filippov (314), and further in view of Eslinger (352) does not appear to explicitly disclose: the parameters of the filter function being one or more quasi-static parameters output by a calibration ROM, the quasi-static parameters configured to converge to stable values, wherein the filter function is configured to use deconvolution to determine a predicted response given an input of one or more measurements of the hydrocarbon system.
However, Silvia teaches:
the parameters of a filter function being one or more quasi-static parameters based on calibration, the quasi-static parameters configured to converge to stable values, wherein the filter function is configured to use deconvolution to determine a predicted response given an input of one or more measurements of the hydrocarbon system. (Silvia Page 6 - 7 volumetric flow rate is determined by deconvolution of acoustic signals (filter function configured to use deconvolution) and based on a calibration procedure requiring knowledge of water cut (quasi-static parameter configured to converge to stable values)
PNG
media_image5.png
224
750
media_image5.png
Greyscale
PNG
media_image6.png
121
498
media_image6.png
Greyscale
)
It would have been obvious to one of ordinary skill in the art to combine the proxy model generation and updating for reservoir modeling disclosed by Zagayevsky (375) in view of Filippov (314), and further in view of Eslinger (352) with the sensor filtering disclosed by Silvia. One of ordinary skill in the art would have been motivated to make this modification in order to use a sensor over a wider operating range (Silvia Page 6, Bottom).
With regard to claim 16, Zagayevsky (375) in view of Filippov (314), and further in view of Eslinger (352), and further in view of Silvia teaches all the elements of the parent claim 15, and further teaches:
wherein the processor is configured to use a plurality of measurements from a first data source, and data of the operational point from a second data source to instantiate the one or more ROMs of the digital twin, (Eslinger (352) Figure 8 and Paragraph 179 data from the historian can be used to instantiate digital twin, and sensor data comes from a data filer)
wherein the plurality of measurements from the first data source and the data of the operational point are temporally synchronized with each other (Silvia Figure 2 data from sensors and operation data needs to be synchronized
PNG
media_image7.png
407
603
media_image7.png
Greyscale
)
It would have been obvious to one of ordinary skill in the art to combine the proxy model generation and updating for reservoir modeling disclosed by Zagayevsky (375) in view of Filippov (314) with the computed sensor values disclosed by Eslinger (352). One of ordinary skill in the art would have been motivated to make this modification in order to more accurately compute a desired value in a hydrocarbon system (Eslinger (352) Paragraph 214).
It would have been obvious to one of ordinary skill in the art to combine the proxy model generation and updating for reservoir modeling disclosed by Zagayevsky (375) in view of Filippov (314), and further in view of Eslinger (352) with the sensor filtering disclosed by Silvia. One of ordinary skill in the art would have been motivated to make this modification in order to use a sensor over a wider operating range (Silvia Page 6, Bottom).
With regard to claim 17, Zagayevsky (375) in view of Filippov (314), and further in view of Eslinger (352), and further in view of Silvia teaches all the elements of the parent claim 16, and further teaches:
automatically temporally synchronize the plurality of measurements from the first data source, and the data of the operational point from the second data source. (Silvia Page 2, Bottom time synchronization is addressed in the data acquisition as part of an integrated computerized system “Data acquisition and pre-processing deals with issues such as: sampling, denoising, outlier removal, compression and time synchronization”)
It would have been obvious to one of ordinary skill in the art to combine the proxy model generation and updating for reservoir modeling disclosed by Zagayevsky (375) in view of Filippov (314), and further in view of Eslinger (352) with the data integration system disclosed by Silvia. One of ordinary skill in the art would have been motivated to make this modification in order to avoid error propagation during data analysis (Silvia Page 2).
Claim 18 is rejected under 35 U.S.C. 103 as being unpatentable over Zagayevsky (375) in view of Filippov (314), and further in view of Fillacier et al. “Calculating Prediction Uncertainty using Posterior Ensembles Generated from Proxy Models” (henceforth “Fillacier”). Zagayevsky (375) and Filippov (314) and Fillacier are analogous art because they solve the same problem of modeling a hydrocarbon system, and because they are from the same field of endeavor of oil and gas exploration.
With regard to claim 18, Zagayevsky (375) in view of Filippov (314) teaches all the elements of the parent claim 10, and further teaches:
instantiate the one or more ROMs using a plurality of data of the operational point of the hydrocarbon system, (Zagayevsky (375) Paragraph 28 the reservoir model input space (instantiate the one or more ROMs) are sampled to predict various reservoir management scenarios (using a plurality of data of))
Zagayevsky (375) in view of Filippov (314) does not appear to explicitly disclose: each of the plurality of data of the operational point comprising a weighting indicating a confidence factor of each of the plurality of data of the operational point.
However, Fillacier teaches:
each of a plurality of data of a operational point comprising a weighting indicating a confidence factor of each of the plurality of data of the operational point. (Fillacier Abstract and Page 2, Bottom uncertain parameters have a prior or initial probability distribution, with each combination of uncertain parameters (plurality of data of a operational point) having an associated probability (a weighting indicating a confidence factor))
It would have been obvious to one of ordinary skill in the art to combine the proxy model generation and updating for reservoir modeling disclosed by Zagayevsky (375) in view of Filippov (314) with the sampling from uncertain parameters having a prior probability distribution disclosed by Fillacier. One of ordinary skill in the art would have been motivated to make this modification in order to compute uncertainties related to predicted variables (Fillacier Abstract).
Claim 19 is rejected under 35 U.S.C. 103 as being unpatentable over Zagayevsky (375) in view of Filippov (314), and further in view of Banerjee et al. (US 2009/0084545) (henceforth (“Banerjee (545)”). Zagayevsky (375) and Filippov (314) and Banerjee (545) are analogous art because they solve the same problem of modeling a hydrocarbon system, and because they are from the same field of endeavor of oil and gas exploration.
With regard to claim 19, Zagayevsky (375) in view of Filippov (314) teaches all the elements of the parent claim 10, and further teaches:
re-calculate parameters of the one or more ROMs over time and monitor changes of the parameters of the one or more ROMs, (Filippov (314) Figure 1 a digital twin model state is continuously updated (re-calculate parameters of the one or more ROMs), and Paragraph 13 results from the models can be used for condition-based monitoring (monitor changes of the parameters))
It would have been obvious to one of ordinary skill in the art to combine the proxy model generation for reservoir modeling disclosed by Zagayevsky (375) with the digital twin updating of a hydrocarbon system disclosed by Filippov (314). One of ordinary skill in the art would have been motivated to make this modification in order to automatically update a digital twin to reflect new measured data (Filippov (314) Figure 1).
Zagayevsky (375) in view of Filippov (314) does not appear to explicitly disclose: alert a technician in response to the parameters of the one or more ROMs changing by more than a threshold amount.
However, Banerjee (545) teaches:
alert a technician in response to the parameters of the one or more models changing by more than a threshold amount. (Paragraph 115 “Automatic alerts ( e.g., indicated in color yellow) are generated whenever the gauge pressure is within a defined variance from the limit value.”, and Paragraph 83 “The real-time simulation results may be delivered in an automatic workflow (i.e., the PDG workflow) with real-time plotting of the key parameters and alarm setting based on pre-determined criteria”, and Paragraph 116 operators act on the alerts)
It would have been obvious to one of ordinary skill in the art to combine the proxy model generation and updating for reservoir modeling disclosed by Zagayevsky (375) in view of Filippov (314) with the real-time plotting of reservoir conditions using simulation disclosed by Banerjee (545). One of ordinary skill in the art would have been motivated to make this modification in order to allow operators to act more quickly in response to adverse conditions in the reservoir (Banerjee (545) Paragraph 116).
Claim 20 is rejected under 35 U.S.C. 103 as being unpatentable over Zagayevsky (375) in view of Eslinger (352), and further in view of Banerjee (545). Zagayevsky (375) and Eslinger (352) and Banerjee (545) are analogous art because they solve the same problem of modeling a hydrocarbon system, and because they are from the same field of endeavor of oil and gas exploration.
With regard to claim 20, Zagayevsky (375) teaches a twinning tool for generating and using a digital twin of a hydrocarbon system, the twinning tool comprising processing circuitry configured to: (Paragraph 21 representation of a reservoir is obtained and can be analyzed, and Figure 7 using a processor)
perform a plurality of simulations to create a hyperdimensional space mapping simulation inputs, outputs and attributes; use the hyperdimensional space from the plurality of simulations to generate a plurality of reduced order models (ROMs) using a curve-fitting technique; (Figure 2A design of experiments is used to generated proxy models using machine learning, and Paragraph 15 machine learnings algorithms comprise simple multivariate regression (curve fitting technique)
PNG
media_image1.png
597
875
media_image1.png
Greyscale
)
determine a plurality of quasi-static parameters (Paragraph 22 an updated reservoir model based on the proxy models is used to predict field production data (determine a plurality of quasi-static parameters), and Paragraph 25 - 27 a reservoir model is calibrated from field data and used to generate the proxy models)
the quasi-static parameters configured to converge to stable values (Paragraph 22 the reservoir model is updated so that the predicted field production matches the observations (converge to stable values) “The history matching feature of module 46 uses optimization techniques to update reservoir model 44 with dynamic field production data 60 by reducing the mismatch between reservoir model predictions and observed production data 60.”)
wherein the on-site test is executed at the hydrocarbon system independent of the plurality of ROMs (see Claim Rejections - 35 U.S.C. § 112) (Figure 2A lab tests are performed prior to the reservoir modeling
PNG
media_image8.png
87
114
media_image8.png
Greyscale
)
generate a digital twin of the hydrocarbon system based on the plurality of quasi-static parameters and the plurality of ROMs; (Paragraph 21 a digital representation of the reservoir is obtained based on the model order reduction and assisted history matching)
Zagayevsky (375) does not appear to explicitly disclose: that the determined quasi-static parameters are by providing calibration data obtained from performing a test at the hydrocarbon system as inputs to the plurality of ROMs; and predict a response of the hydrocarbon system over a future time horizon using the digital twin.
However, Eslinger (352) teaches:
determine a plurality of quasi-static parameters by providing calibration data obtained from performing a test at the hydrocarbon system as inputs to a plurality of ROMs (Paragraph 216 system identification can predict production parameters (determine a plurality of quasi-static parameters) based on measured responses inputted to a digital twin (by providing calibration data as inputs to), and Paragraph 201 and 205 the digital twin can be represented as a reduced order model for computing the production parameters (to a of ROMs))
predict a response of the hydrocarbon system over a future time horizon using the digital twin (Paragraph 182 the digital twin can be used in a live environment for advisory purposes (a future time horizon))
It would have been obvious to one of ordinary skill in the art to combine the proxy model generation for reservoir modeling disclosed by Zagayevsky (375) with the model implemented sensor in a reduced order model disclosed by Eslinger (352). One of ordinary skill in the art would have been motivated to make this modification in order to more accurately computed a desired value in a hydrocarbon system (Eslinger (352) Paragraph 214).
Zagayevsky (375) in view of Eslinger (352) does not appear to explicitly disclose: display the predicted response of the hydrocarbon system over the future time period to a technician.
However, Banerjee (545) teaches:
displays predicted response of a hydrocarbon system over a future time period to a technician. (Paragraph 83 “The real-time simulation results may be delivered in an automatic workflow (i.e., the PDG workflow) with real-time plotting of the key parameters and alarm setting based on pre-determined criteria”, and Paragraph 116 operators act on the alerts)
It would have been obvious to one of ordinary skill in the art to combine the proxy model generation and updating for reservoir modeling disclosed by Zagayevsky (375) in view of Eslinger (352) with the real-time plotting of reservoir conditions using simulation disclosed by Banerjee (545). One of ordinary skill in the art would have been motivated to make this modification in order to allow operators to act more quickly in response to adverse conditions in the reservoir (Banerjee (545) Paragraph 116).
Examiner General Comments
With regard to the prior art rejection(s), any cited portion of the relied upon reference(s), either to specific areas or as direct language, is intended to be interpreted in the context of the reference(s) as a whole, as would be understood by one of ordinary skill in the art. Although the specified citations are representative of the teachings in the art and are applied to the specific limitations within the individual claim, other passages and figures may apply as well. It is respectfully requested that, in preparing responses, the applicant fully consider the references in their entirety as potentially teaching all or part of the claimed invention, as well as the context of the passage as taught by the prior art or disclosed by the examiner. The entire reference is considered to provide disclosure relating to the claimed invention. The claims & only the claims form the metes & bounds of the invention. Office personnel are to give the claims their broadest reasonable interpretation in light of the supporting disclosure. Unclaimed limitations appearing in the specification are not read into the claim. Prior art was referenced using terminology familiar to one of ordinary skill in the art. Such an approach is broad in concept and can be either explicit or implicit in meaning. Examiner's Notes are provided with the cited references to assist the applicant to better understand how the examiner interprets the applied prior art. Such comments are entirely consistent with the intent and spirit of compact prosecution.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to ALFRED H. WECHSELBERGER whose telephone number is (571)272-8988. The examiner can normally be reached M - F, 10am to 6pm.
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, Emerson Puente can be reached at 571-272-3652. 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.
/ALFRED H. WECHSELBERGER/ExaminerArt Unit 2187
/EMERSON C PUENTE/Supervisory Patent Examiner, Art Unit 2187
Prosecution Insights