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 .
Responsive to the communication dated 4/20/2026
Claims 20, 21, 22 are amended.
Claim 7 is cancelled.
Claims 1 – 6, 8 – 19 are presented for examination.
Final Action
THIS ACTION IS MADE FINAL. Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Response to Arguments
Citing the instant specification, figures, and paragraph 36, the Applicant asserts that the art of record does not make obvious the claimed “misfit.”
In response the argument is not persuasive because Kolbjornsen teaches comparing the seismic response extracted in the neighborhood to seismic features in FIG. 3, where a bad match decreases probability, [page 433 col 1 paragraph 3 lines 1-6]; see FIG. 3 O-O, [page 433]).
The Applicant argues that Kolbjornsen and Miotti and Guldbrandsoy do not specifically teach hydrocarbon flood results in brine flood results, nor generating a model where zones are labeled based on relative performance.
In response the argument is not persuasive because a reading of claim 1 shows that claim 1 does not recites zones or labels based on relative performance.
The applicant argues that the motivation to combine the references is deficient because the Examiner describes the benefit of having an ensemble model with high predictive quality as a motivation for combining the references but the combination does not address why one of ordinary skill in the art would label relative inversion performance zones of a model and accordingly there is no motivation to combine Kacewicz_2011 with McCormack_2017.
In response the argument is not persuasive. MacCormack_2017, in the abstract states: “… it is important to ensure that the modeling techniques and procedures are able to accurately delineate and characterize the heterogeneity of the various geological environments included within the regional model domain…” The teaches that it is important to delineate and characterize heterogeneity. Labeling of zones according to their performance provides delineation and characterization of heterogeneity. Therefore, an explicit reason as to why one of ordinary skill in the art would label different zone is because its important to provide demarcations and characterizations (i.e., labels) to provide informative value. Further, the Applicant is arguing about labels and zones, but claim 1 does not recite the elements label or zone.
Regarding claim 5, the Applicant citing the specification states that the specification indicates that “some techniques utilize an artificial rock type to model implausible sets of petrophysical parameters” and the cited prior art does not make these element obvious.
In response the argument is not persuasive because The Office action pointed towards Kolbjornsen for teaching an artificial rock type. At page 433 COL 2 , Kolbjornsen states: “… In synthetic examples, it is always possible to list all alternative scenario; in real situations, the case is more complex as it is possible that neither of the listed scenarios provides an adequate fit. To investigate this, it is common to introduce an additional scenario, which is denoted as an undefined event. The undefined event represents a base model, which is parsimonious and has less spatial structure than what is imposed in the Bayesian model. lf this model is assigned a high probability in the posterior; this should be a warning sign that There are problems with the statistical model. Such an undefined rock can be modeled with independent elastic parameters with wide spread…” This teaches to use a “synthetic” rock model (i.e., artificial rock type) and that doing so is common to introduce additional scenarios.
Additionally, Kacewicz_2011; however, makes obvious identifying a first set of misfits “Used to model an implausible set of petrophysical parameters indicative of prior misfits or data misfits” ((Figure 2 illustrates “good fit” yes or no and if there is not a good fit (i.e., mismatch/misfit) then to update the model; par 1: “the present invention relates generally to modeling of hydrocarbon reservoirs and more particularly to integrating mechanical earth models, earth models and basin models”; par 4: “… a method of modeling properties in a subsurface region of interest. The method includes using data representative of geological, geophysical, and/or petrophysical attributes in the subsurface region, building an earth model for the subsurface region… calibrated against data relating to the subsurface region of interest. The earth model… are iteratively evaluated to converge modeled properties with measured attributes in the subsurface region of interest.”; par 15: “… the models are iteratively evaluated, with the goal of converging modeled attributes of the subsurface region towards measured attributes… these predicted modeled attributes can be compared with the measured data, where there is disagreement between prediction and measurement, the models 110, 112, 114 are adjusted and new predictions are generated.” The above citations teaches that predictions can be implausible because they do not agree with the measured/observed physical attributes at the region of interest. Additionally, the above citations teach an iterative process which means that the current disagreement between predicted and measured data is indicative of the prior iteration according to the prior disagreement between the prior prediction and the prior measured data.
The Applicant asserts that the Examiner implicitly acknowledges deficiencies in the art by breaking claim elements into sentence fragments for purposes of analysis.
In response, the very nature of an obviousness rejection under 35 USC 103 is one in which claim elements are not explicitly taught by a single reference and accordingly the elements are mapped to different prior art.
The Applicant argues that MacCormack_2017 teaches away because MacCormack_2017 teaches that some models are more predictive and that this means one of ordinary skill in the art would simply pick the most predictive model rather than creating a trial saturation model that incorporates locations of artificial rock types.
In response this interpretation of the MacCormack_2017 is misguided. MAcCormack-2017 does not indicate that a reservoir engineer or anyone else has perfect foreknowledge of which model is a best fit for rock types and therefore negates the need to testing models. Rather a teaching that some models are better suited for certain areas of a rock formation than other model supports the notion of evaluating models against various rock types including those that are synthetic/artificial types. In other words, to test a model’s predictivity with regard to various rock parameters by experimentally artificially changing rock parameters.
The Applicant also argues that one of ordinary skill in the art would not combine Kolbjornsen with Miotti with Kacewicz_2011 because Kolbjornsen has a stated goal to reduce computational cost by avod large-scale iterations and accordingly the iterative complexity of Miotti and attribute-matching of Kacewicz_2011 are not compatible with Kolbjornsen because the goal of Kobjernsen is to avoid large-scale iterations.
In response, the argument is not persuasive. Kolbjernsen does not use the word “iteration” (or any other synonym e.g., loop, etc.) and accordingly, they cannot have a stated goal of avoiding large-iterations.
The Applicant appears to be asserting that because Kolbjernsen teaches to evaluate a local-neighborhood that this means the method cannot be combined to derive properties on a larger scale, however, page 435 explicitly states: “… an analysis on a local scale gives interesting results that in turn can be further combined to derive properties on a larger scale…”. Therefore, Kolbjernsen explicitly states that you can combine the method to derive properties on a larger scale.
A review finds that what Kolbjernsen on page 435 clarifies how they address the computational cost by stating that: “the local-neighborhood approach makes it possible to evaluate posterior uncertainty in a spatial model without resorting to time-consuming McMC algorithm (i.e., Markov chains). Miotti and Kacewicz_2011 does not use Markov chains for posterior uncertainty. Accordingly, the Applicants assertion that Miotti and Kacewicz_2011 is incompatible with Kolbjernsen is not persuasive.
With regard to claim 13 the Applicant asserts that the prior art does not make obvious to have an artificial rock type as a prior facie of an inversion process.
In response the argument is not persuasive. In geological inversion, a prior is the pre-existing knowledge, data, or set of assumptions about the subsurface introduced before processing any new geophysical measurements. Because geological inversions are mathematically non-unique, the prior narrows down the infinite possibilities to produce a reasonable output. Performing a rock type inversion using a hypothetical rock type (i.e., artificial or synthetic rock type) allows for testing limits and calibrate algorithms. Kacewicz_2011 explicitly teaches to “calibrate against data relating to the subsurface region of interest” (par 4).
Additionally, The Office action pointed towards Kolbjornsen for teaching an artificial rock type. At page 433, Kolbjornsen states: “… In synthetic examples, it is always possible to list all alternative scenario; in real situations, the case is more complex as it is possible that neither of the listed scenarios provides an adequate fit. To investigate this, it is common to introduce an additional scenario, which is denoted as an undefined event. The undefined event represents a base model, which is parsimonious and has less spatial structure than what is imposed in the Bayesian model. lf this model is assigned a high probability in the posterior; this should be a warning sign that There are problems with the statistical model. Such an undefined rock can be modeled with independent elastic parameters with wide spread…” This teaches to use a “synthetic” rock model (i.e., artificial rock type) and that doing so is common to introduce additional scenarios.
The Applicant asserts that the combination of prior art is incompatible because Kolbjornsen has a purpose of reducing cost by using a local neighborhood approach.
In response the argument is not persuasive. As outlined previously, Kolbjernsen does not use the word “iteration” (or any other synonym e.g., loop, etc.) and accordingly, they cannot have a stated goal of avoiding large-iterations.
The Applicant appears to be asserting that because Kolbjernsen teaches to evaluate a local-neighborhood that this means the method cannot be combined to derive properties on a larger scale, however, page 435 explicitly states: “… an analysis on a local scale gives interesting results that in turn can be further combined to derive properties on a larger scale…”. Therefore, Kolbjernsen explicitly states that you can combine the method to derive properties on a larger scale.
A review finds that what Kolbjernsen on page 435 clarifies how they address the computational cost by stating that: “the local-neighborhood approach makes it possible to evaluate posterior uncertainty in a spatial model without resorting to time-consuming McMC algorithm (i.e., Markov chains). Miotti and Kacewicz_2011 does not use Markov chains for posterior uncertainty. Accordingly, the Applicants assertion that Miotti and Kacewicz_2011 is incompatible with Kolbjernsen is not persuasive.
With regard to claim 12, the Applicant asserts that Grana does not teach artificial rock type.
In response the argument is not persuasive. The Office action pointed towards Kolbjornsen for teaching an artificial rock type. At page 433, Kolbjornsen states: “… In synthetic examples, it is always possible to list all alternative scenario; in real situations, the case is more complex as it is possible that neither of the listed scenarios provides an adequate fit. To investigate this, it is common to introduce an additional scenario, which is denoted as an undefined event. The undefined event represents a base model, which is parsimonious and has less spatial structure than what is imposed in the Bayesian model. lf this model is assigned a high probability in the posterior; this should be a warning sign that There are problems with the statistical model. Such an undefined rock can be modeled with independent elastic parameters with wide spread…” This teaches to use a “synthetic” rock model (i.e., artificial rock type) and that doing so is common to introduce additional scenarios.
Potentially Allowable Subject Matter
Claim 20, 21, 22 are objected to as being dependent upon a rejected base claim, but would be allowable if rewritten in independent form including all of the limitations of the base claim and any intervening claims.
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claim(s) 1-6, 8-11 are rejected under 35 U.S.C. 103 as being unpatentable over "Bayesian AVO inversion to rock properties using a local neighborhood in a spatial prior model" (Kolbjornsen) in view of U.S. 2015/0362623 (Miotti) in further view of “Seismic signatures of siliciclastic reservoirs as function of burial history – a Barents Sea example” (Guldbrandsoy) in view of Kacewicz_2011 (US 2011/0141851 A1) in view of MacCormack_2017 (Using a multiple variogram approach to improve the accuracy of subsurface geological models, Can. J. Earth Sci. 00: 1 – 16 (0000), Sep 16, 2017).
claim 1, Kolbjornsen teaches A method for generating a fluid saturation model for a subsurface region comprising (section labeled "The local-neighborhood approach for integrating spatial structure", [page 432 col 1 paragraph 3]-[page 433 col 2 paragraph 5]): performing a first petrophysical inversion for the subsurface region with hydrocarbon flood parameters to generate hydrocarbon flood results (seismic response of local lithology-fluid combination, [page 433 col 1 paragraph 2 lines 1-2]; where the combination is O-O from FIG. 3, [page 433], and inversion is the swapped point of view, [page 431 col 2 paragraph 2 line 3], i.e. seismic response to rock properties rather than vice versa), the hydrocarbon flood parameters representing an assumption of hydrocarbon flooding (the seismic response of O-O is one of eight possible lithology fluid scenarios, [page 433 col 1 paragraph 2 lines 3-5], where a scenario in two neighboring cells assumes a lithology-fluid classification in order to perform the inverse problem on a local neighborhood, [page 432 col 1 paragraph 2 lines 4-14]); performing a second petrophysical inversion for the subsurface region with brine flood parameters to generate brine flood results (seismic response of local lithology-fluid combination, [page 433 col 1 paragraph 2 lines 1-2]; where the combination is B-B from FIG. 3, [page 433], and inversion is the swapped point of view, [page 431 col 2 paragraph 2 line 3], i.e. seismic response to rock properties rather than vice versa), the brine flood parameters representing an assumption of brine flooding (the seismic response of B-B is one of eight possible lithology fluid scenarios, [page 433 col 1 paragraph 2 lines 3-5], where a scenario in two neighboring cells assumes a lithology-fluid classification in order to perform the inverse problem on a local neighborhood, [page 432 col 1 paragraph 2 lines 4-14]); identifying a first set of misfits in the hydrocarbon flood results (comparing the seismic response extracted in the neighborhood to seismic features in FIG. 3, where a bad match decreases probability, [page 433 col 1 paragraph 3 lines 1-6]; see FIG. 3 O-O, [page 433]); identifying a second set of misfits in the brine flood results (comparing the seismic response extracted in the neighborhood to seismic features in FIG. 3, where a bad match decreases probability, [page 433 col 1 paragraph 3 lines 1-6]; see FIG. 3 B-B, [page 433]); generating a trial fluid saturation model based on at least one of the first set of misfits and the second set of misfits (define posterior transition probabilities, which can be used to build spatial model, where the spatial model is not the true posterior distribution, but a first order approximation, [page 433 col 2 paragraph 1 lines 1-4]; note the spatial model is not a local window, but instead the entire grid, [page 432 col 1 paragraph 4 lines 3-4]; see also FIG. 4(b) (right) panel, [page 434]), ; identifying the potential hydrocarbon-bearing formations in the subsurface region based on the fluid saturation model (calculate probability of oil sand by assigning most probable lithology fluid class, [page 434 col 2 paragraph 1 lines 1-5]; see also FIG. 4(c) right panel, [page 434]); wherein each one of the first, second, and petrophysical inversion is carried out using a computer, and each one of the trial fluid saturation model is generated using a computer (method is introduced to reduce computation cost, [Abstract] line 6).
Kolbjornsen does not teach in substantially all available pore space in the subsurface region; in substantially all available pore space in the subsurface region; performing a third petrophysical inversion for the subsurface region with the trial fluid saturation model to generate final results; and generating the fluid saturation model for the subsurface region based on the final results; wherein each one of the third petrophysical inversion is carried out using a computer, and the fluid saturation model is generated using a computer.
Kolbjornsen does not explicitly teach identifying a first set of misfits “of a predicted property as compared to a measured property at the subsurface region of interest” nor identifying a second set of misfits “of a predicted property as compared to a measured property at the subsurface region of interest” nor generating a model “wherein zones in the trial fluid saturation model are labeled based on relative performance under the two inversions.”
However, Miotti teaches performing a third petrophysical inversion for the subsurface region with the trial fluid saturation model to generate final results (inverse solver 1150 operates from input block 1110 including input data and uncertainty and input block 1136, which includes synthetic computed data from the forward modelling using the Gassmann and Archie equations to output confidence information, [0178] lines 5-7); and generating the fluid saturation model for the subsurface region based on the final results (output estimated model 1160, [0178] lines 7-9; inverse formula may be solved in the form d=g(m) where vector m includes water saturation, [0181] lines 1-9); wherein each one of the third petrophysical inversion is carried out using a computer, and the fluid saturation model is generated using a computer (example system includes processor, and memory operatively coupled to the process with one or more instructions to implement the inverse problem, [0184] lines 1-12).
Kolbjornsen and Miotti do not teach in substantially all available pore space in the subsurface region; in substantially all available pore space in the subsurface region.
Kolbjornsen and Miotti does not explicitly teach identifying a first set of misfits “of a predicted property as compared to a measured property at the subsurface region of interest” nor identifying a second set of misfits “of a predicted property as compared to a measured property at the subsurface region of interest” nor generating a model “wherein zones in the trial fluid saturation model are labeled based on relative performance under the two inversions.”
However, Guldbrandsoy teaches in substantially all available pore space in the subsurface region; in substantially all available pore space in the subsurface region (“The rock physics modeling has been done using 100% brine saturation and 100% gas saturation, to give a better understanding of the extremes of fluid substitution, [page 43 paragraph 2 lines 7-9]; for a description of how pore fluids are used in the Gassmann equation, see [pages 27-29]; and for the application to velocity models in each specific layer Period A-Top Layer, see [pages 58-66], example on [page 65] describes how misfits can be used to update the model).
It would have been obvious to one skilled in the art before the effective filing date to combine Kolbjornsen with Miotti because a teaching, suggestion, or motivation in the prior art would have led one skilled in the art to combine prior art teaching to arrive at the claimed invention. Kolbjornsen discloses a system that teaches all of the claimed features except for performing a third petrophysical inversion and generating the fluid saturation mode from the results of that inversion. Kolbjornsen suggests “An analysis on a local scale gives interesting results that in turn can be further combined to derive properties on a larger scale” (Kolbjornsen [page 435 col 2 paragraph 4 lines 3-5]). Miotti teaches an iterative approach, where in an iteration the inverse solver block 1150 receives additional information to solve for the estimated model, (Miotti [0178] lines 7-18). A person having skill in the art would have a reasonable expectation of successfully increasing the confidence of the results in the system and method of Kolbjornsen by modifying Kolbjornsen with the iterative forward modeling (Miotti 1134, [0178] line 3) and inverse solver of Miotti (Miotti [0178] lines 6-10) by including the statistical methods of Kolbjornsen (Kolbjornsen [page 433 col 1 paragraph 1 lines 1-16]) as the additional information in Miotti (Miotti [0178] lines 10-18). Therefore, it would have been obvious to combine Kolbjornsen with Miotti to a person having ordinary skill in the art.
It would have been obvious to one skilled in the art before the effective filing date to combine Kolbjornsen in view of Miotti with Guldbrandsoy because a teaching, suggestion, or motivation in the prior art would have led one skilled in the art to combine prior art teaching to arrive at the claimed invention. Kolbjornsen in view of Miotti discloses a system and method that teaches all of the claimed features except for representing an assumption of hydrocarbon/brine flooding in substantially all available pore space. Kolbjornsen provides lithology-fluid classes, but does not specifically define the percentage of brine in brine sand or percentage of oil in oil sand (Kolbjornsen [page 432 col 1 paragraph 4]). Miotti teaches how the Gassmann equation can be used to perform fluid substitution in order to forward model, (Miotti see step 1134 in FIG. 11, [0099], and [0178] line 3). Finally, Guldbrandsoy provides the same Gassmann equation, (Guldbrandsoy [pages 27-29]), but further explains that modeling is done as 100% brine and 100% gas to give a better understanding of the extremes of fluid substitution (Guldbrandsoy [page 43 paragraph 2 lines 7-9]). In an example, when a layer does not fit well with a given substitution, the modeling can be redone as a mix between the two extremes (Guldbrandsoy [page 65 paragraph 2 lines 1-6]). A person having skill in the art would have a reasonable expectation of understanding the extremes in the system and method of Kolbjornsen in view of Miotti by modifying Kolbjornsen in view of Miotti with the 100% substitutions of Guldbrandsoy. Therefore, it would have been obvious to combine Kolbjornsen in view of Miotti with Guldbrandsoy to a person having ordinary skill in the art, and this claim is rejected under 35 U.S.C. 103.
Kolbjornsen and Miotti and Guldbrandsoy do not explicitly teach identifying a first set of misfits “of a predicted property as compared to a measured property at the subsurface region of interest” nor identifying a second set of misfits “of a predicted property as compared to a measured property at the subsurface region of interest” nor generating a model “wherein zones in the trial fluid saturation model are labeled based on relative performance under the two inversions.”
Kacewicz_2011; however, makes obvious identifying a first set of misfits “of a predicted property as compared to a measured property at the subsurface region of interest” and identifying a second set of misfits “of a predicted property as compared to a measured property at the subsurface region of interest” (Figure 2 illustrates “good fit” yes or no and if there is not a good fit (i.e., mismatch) then to update the model; par 1: “the present invention relates generally to modeling of hydrocarbon reservoirs and more particularly to integrating mechanical earth models, earth models and basin models”; par 4: “… a method of modeling properties in a subsurface region of interest. The method includes using data representative of geological, geophysical, and/or petrophysical attributes in the subsurface region, building an earth model for the subsurface region… calibrated against data relating to the subsurface region of interest. The earth model… are iteratively evaluated to converge modeled properties with measured attributes in the subsurface region of interest.”; par 15: “… the models are iteratively evaluated, with the goal of converging modeled attributes of the subsurface region towards measured attributes… these predicted modeled attributes can be compared with the measured data, where there is disagreement between prediction and measurement, the models 110, 112, 114 are adjusted and new predictions are generated…” EXAMINER NOTE: the above teaches to have an ensemble of earth models and to compare a predicted attribute (i.e. property) with a measured attribute in the subsurface region of interest. The above citation teaches to do this for all models in the ensemble of models. The claim recites a first predictive model for hydrocarbon flood and second predictive model for brine flood. Accordingly, the claim recites an ensemble of earth models. The citation teaches to compared the predicted attributes with the measured attribute for each model. Therefore, it would have been obvious to one of ordinary skill in the art to do such a comparison to find disagreements (i.e., misfits) for each model in order to have all the models have high predictive quality by producing predictions that match (i.e., agree with) measured values so that the models predict what is actually observed in the ground.).
Kolbjornsen and Miotti and Guldbrandsoy and Kacewicz_2011 are analogous art because they are from the same field of endeavor called subterranean models. Before the effective filing date, it would have been obvious to a person of ordinary skill in the art to combine Kolbjornsen and Miotti and Guldbrandsoy and Kacewicz_2011. The rationale for doing so would have been that, as outlined above, Kolbjornsen and Miotti and Guldbrandsoy make obvious to perform a first petrophysical inversion generating hydrocarbon flood results and to perform a second petrophysical inversion generating brine flood results. Kacewicz_2011 teaches what when there is an ensemble of models, including earth models, that predict petrophysical attributes (see par 4: petrophysical attributes) to iteratively evaluate the petrophysical attributes by comparing with measured attributes in the subsurface region of interest and if there is a disagreement (i.e., mismatch) to update the ensemble of models until their predictive quality converges to match the attributes of the subsurface region of interest. Therefore, it would have been obvious to combine Kolbjornsen and Miotti and Guldbrandsoy and Kacewicz_2011 for the benefit of having an ensemble of models that has high predictive quality that agrees with real world observations/measurements to obtain the invention as specified in the claims.
Kolbjornsen and Miotti and Guldbrandsoy and Kacewicz_2011 does not explicitly recite generating a model “wherein zones in the trial fluid saturation model are labeled based on relative performance under the two inversions.”
MacCormack_2017, however, makes obvious generating a model “wherein zones in the trial fluid saturation model are labeled based on relative performance under the two inversions” (abstract: “… it is important to ensure that the modeling techniques and procedures are able to accurately delineate and characterize the heterogeneity of the various geological environments included within the regional model domain… improved by splitting the regional model into multiple submodels based on the degree of variability observed between surrounding data points and informed by expert geological knowledge of the geological depositional framework… resulting in a more geologically realistic interpolation of the entire model domain as demonstrated…”;page 2 introduction: “… many three-dimensional (3D) subsurface geological models were created for various regions… the overall aim of these models is to characterize the nature and distribution of subsurface materials… the objective of this study is to determine if a model can be improved by splitting the regional model into multiple submodels based on the degree of variability observed between surrounding data points and our understanding of the depositional environment… thus optimizing the measured data values as inputs to the geostatistical model. The hypothesis is that by separating the regional model domain into submodels based on the data variability and our understanding of the geologic context, a variogram can be produced for each submodel region that is more representative of the true variability within the data for each area…”; page 12: “… a variogram model must be fit to the data to provide a prediction of the unknown values at specific location. The variogram model is essentially a generalized mathematical equation that provides the best fit to the observed data…”; page 14 conclusion: “the objective of this study was to determine if subsurface geological model accuracy could be improved by splitting the regional model into multiple submodels based on the degree of variability observed between surrounding data points… the hypothesis is that by separating the regional model domain into submodels based on the data variability and our understanding of the geological context, a variogram can be produced for each submodel region that is more representative of variability within each area, thus resulting in a more geostatistically appropriate and geologically realistic model… the proposed submodel approach produces a more geologically realistic model of the study area when compared with the regional model approach as evidenced by comparing the model results against the regional surficial geology map…”; FIG. 3, 4, 5, 6
EXAMINER NOTE: The above citations teach to break the geologic region of interest up into zones and to associate predictive models with the zone where they are most accurately predictive. This makes obvious to those of ordinary skill in the art to have: “wherein zones in the trial fluid saturation model are labeled based on relative performance under the two inversions” because labeling the zones based on the relative performance under the two inversions identifies which models are most accurately predictive in that zone of the region of interest. This provides a more geologically realistic model of the region because the most appropriate inversion model can be used for predictions in the zones where it is most accurately predictive.)
Kolbjornsen and Miotti and Guldbrandsoy and Kacewicz_2011 and MacCormack_2017 are analogous art because they are from the same field of endeavor called geologic models. Before the effective filing date, it would have been obvious to a person of ordinary skill in the art to combine Kolbjornsen and Miotti and Guldbrandsoy and Kacewicz_2011 and MacCormack_2017. The rationale for doing so would have been that Kolbjornsen and Miotti and Guldbrandsoy and Kacewicz_2011 make obvious to have an ensemble of models that predict attributes for a region of interest. MacCormack_2017 teaches that among an ensemble of models, some models are more accurately predictive than others in certain zones of the region of interest and because of this the region of interest should be subdivided into zones and the model most predictive for each zone should be identified for each of those zone in order the have a more geologically realistic model for the whole geologic region. Therefore, it would have been obvious to combine Kolbjornsen and Miotti and Guldbrandsoy and Kacewicz_2011 and MacCormack_2017 for the benefit of having a more geologically realistic model to obtain the invention as specified in the claims.
With respect to claim 2, Kolbjornsen in view of Miotti and Guldbrandsoy teaches all of the limitations of claim 1, as noted above. Kolbjornsen further teaches wherein at least one of the first, second, and third petrophysical inversions comprises a facies-based inversion (methodology referred to as Bayesian AVO inversion, [Title]; which includes a discrete lithology-fluid prediction, [Abstract] line 11; specifically, the methodology lithology-fluid classes of share, brine-sand, and oil-sand, [page 432 col 1 paragraph 4 lines 1-3]).
With respect to claim 3, Kolbjornsen in view of Miotti and Guldbrandsoy teaches all of the limitations of claim 1, as noted above. Kolbjornsen further teaches at least one of: identifying potential hydrocarbon-bearing formations in the subsurface region based on the fluid saturation model (FIG. 4 bottom right results showing probability of oil-sand in an image, [page 433]); generating an image of the subsurface region based on the fluid saturation model (FIG. 4 bottom right results showing probability of oil-sand in an image, [page 433]); and managing hydrocarbons in the subsurface region based on the fluid saturation model (inversion results give shale thickness, well B is drilled in one section or two sections based on thickness of shale, [page 434 col 2 paragraph 3 lines 12-16]).
With respect to claim 4, Kolbjornsen in view of Miotti and Guldbrandsoy teaches all of the limitations of claim 1, as noted above. Kolbjornsen further teaches wherein at least one of the first set of misfits and the second set of misfits comprises at least one of: porosity misfits; volume of clay misfits; seismic data misfits (comparing the seismic response extracted in the neighborhood to seismic features in FIG. 3, [page 433 col 1 paragraph 3 lines 2-3]; see FIG. 3 O-O and S-O, [page 433]); and 30P-wave velocity misfits.
With respect to claim 5, Kolbjornsen teaches method for generating a fluid saturation model for a subsurface region comprising (section labeled "The local-neighborhood approach for integrating spatial structure", [page 432 col 1 paragraph 3]-[page 433 col 2 paragraph 5]): performing a first petrophysical inversion for the subsurface region with brine flood parameters to generate brine flood results (seismic response of local lithology-fluid combination, [page 433 col 1 paragraph 2 lines 1-2]; where the combination is B-B from FIG. 3, [page 433], and inversion is the swapped point of view, [page 431 col 2 paragraph 2 line 3], i.e. seismic response to rock properties rather than vice versa), the brine flood parameters representing an assumption of brine flooding (the seismic response of B-B is one of eight possible lithology fluid scenarios, [page 433 col 1 paragraph 2 lines 3-5], where a scenario in two neighboring cells assumes a lithology-fluid classification in order to perform the inverse problem on a local neighborhood, [page 432 col 1 paragraph 2 lines 4-14]); classifying rock types in the subsurface region based on the brine flood results (obtain probability of eight different scenarios and sum events of top cell to get posterior probability, [page 433 col 1 paragraph 3 lines 7-12]), wherein the rock types comprise at least one artificial rock type (introduce additional scenario of undefined event, which can model an undefined rock [page 433 col 2 paragraph 2 lines 1-11])
; generating a trial fluid saturation model based on the classified rock types (define posterior transition probabilities, which can be used to build spatial model, where the spatial model is not the true posterior distribution, but a first order approximation, [page 433 col 2 paragraph 1 lines 1-4]; note the spatial model is not a local window, but instead the entire grid, [page 432 col 1 paragraph 4 lines 3-4]; see also FIG. 4(b) (right) panel, [page 434])
; identifying the potential hydrocarbon-bearing formations in the subsurface region based on the fluid saturation model (calculate probability of oil sand by assigning most probable lithology fluid class, [page 434 col 2 paragraph 1 lines 1-5]; see also FIG. 4(c) right panel, [page 434]); wherein each one of the first petrophysical inversion is carried out using a computer, and each one of the trial fluid saturation model is generated using a computer (method is introduced to reduce computation cost, [Abstract] line 6).
Kolbjornsen does not teach in substantially all available pore space in the subsurface region; performing a second petrophysical inversion for the subsurface region with the trial fluid saturation model to generate final results; and generating the fluid saturation model for the subsurface region based on the final results; wherein each one of the second petrophysical inversion is carried out using a computer, and each one of the fluid saturation model is generated using a computer.
Kolbjornsen does not teach “Used to model an implausible set of petrophysical parameters indicative of prior misfits or data misfits” nor “and locations of the predicted artificial rock types”
However, Miotti teaches performing a second petrophysical inversion for the subsurface region with the trial fluid saturation model to generate final results (inverse solver 1150 operates from input block 1110 including input data and uncertainty and input block 1136, which includes synthetic computed data from the forward modelling using the Gassmann and Archie equations, [0178] lines 5-7); and generating the fluid saturation model for the subsurface region based on the final results (output estimated model & confidence 1160, [0178] lines 7-9; inverse formula may be solved in the form d=g(m) where vector m includes water saturation, [0181] lines 1-9); wherein each one of the second petrophysical inversion is carried out using a computer, and each one of the fluid saturation model is generated using a computer (example system includes processor, and memory operatively coupled to the process with one or more instructions to implement the inverse problem, [0184] lines 1-12).
Kolbjornsen and Miotti do not teach in substantially all available pore space in the subsurface region.
Kolbjornsen and Miotti does not teach “Used to model an implausible set of petrophysical parameters indicative of prior misfits or data misfits” nor “and locations of the predicted artificial rock types”
However, Guldbrandsoy teaches in substantially all available pore space in the subsurface region (“The rock physics modeling has been done using 100% brine saturation and 100% gas saturation, to give a better understanding of the extremes of fluid substitution, [page 43 paragraph 2 lines 7-9]; for a description of how pore fluids are used in the Gassmann equation, see [pages 27-29]; and for the application to velocity models in each specific layer Period A-Top Layer, see [pages 58-66], example on [page 65] describes how misfits can be used to update the model).
It would have been obvious to one skilled in the art before the effective filing date to combine Kolbjornsen with Miotti because a teaching, suggestion, or motivation in the prior art would have led one skilled in the art to combine prior art teaching to arrive at the claimed invention. Kolbjornsen discloses a system that teaches all of the claimed features except for performing a third petrophysical inversion and generating the fluid saturation mode from the results of that inversion. Kolbjornsen suggests “An analysis on a local scale gives interesting results that in turn can be further combined to derive properties on a larger scale” (Kolbjornsen [page 435 col 2 paragraph 4 lines 3-5]). Miotti teaches an iterative approach, where in an iteration the inverse solver block 1150 receives additional information to solve for the estimated model, (Miotti [0178] lines 7-18). A person having skill in the art would have a reasonable expectation of successfully increasing the confidence of the results in the system and method of Kolbjornsen by modifying Kolbjornsen with the iterative forward modeling (Miotti 1134, [0178] line 3) and inverse solver of Miotti (Miotti [0178] lines 6-10) by including the statistical methods of Kolbjornsen (Kolbjornsen [page 433 col 1 paragraph 1 lines 1-16]) as the additional information in Miotti (Miotti [0178] lines 10-18). Therefore, it would have been obvious to combine Kolbjornsen with Miotti to a person having ordinary skill in the art, and this claim is rejected under 35 U.S.C. 103.
It would have been obvious to one skilled in the art before the effective filing date to combine Kolbjornsen in view of Miotti with Guldbrandsoy because a teaching, suggestion, or motivation in the prior art would have led one skilled in the art to combine prior art teaching to arrive at the claimed invention. Kolbjornsen in view of Miotti discloses a system and method that teaches all of the claimed features except for representing an assumption of hydrocarbon/brine flooding in substantially all available pore space. Kolbjornsen provides lithology-fluid classes, but does not specifically define the percentage of brine in brine sand or percentage of oil in oil sand (Kolbjornsen [page 432 col 1 paragraph 4]). Miotti teaches how the Gassmann equation can be used to perform fluid substitution in order to forward model, (Miotti see step 1134 in FIG. 11, [0099], and [0178] line 3). Finally, Guldbrandsoy provides the same Gassmann equation, (Guldbrandsoy [pages 27-29]), but further explains that modeling is done as 100% brine and 100% gas to give a better understanding of the extremes of fluid substitution (Guldbrandsoy [page 43 paragraph 2 lines 7-9]). In an example, when a layer does not fit well with a given substitution, the modeling can be redone as a mix between the two extremes (Guldbrandsoy [page 65 paragraph 2 lines 1-6]). A person having skill in the art would have a reasonable expectation of understanding the extremes in the system and method of Kolbjornsen in view of Miotti by modifying Kolbjornsen in view of Miotti with the 100% substitutions of Guldbrandsoy. Therefore, it would have been obvious to combine Kolbjornsen in view of Miotti with Guldbrandsoy to a person having ordinary skill in the art, and this claim is rejected under 35 U.S.C. 103.
Kolbjornsen and Miotti and Guldbrandsoy does not teach “Used to model an implausible set of petrophysical parameters indicative of prior misfits or data misfits” nor “and locations of the predicted artificial rock types.”
Kacewicz_2011; however, makes obvious identifying a first set of misfits “Used to model an implausible set of petrophysical parameters indicative of prior misfits or data misfits” ((Figure 2 illustrates “good fit” yes or no and if there is not a good fit (i.e., mismatch) then to update the model; par 1: “the present invention relates generally to modeling of hydrocarbon reservoirs and more particularly to integrating mechanical earth models, earth models and basin models”; par 4: “… a method of modeling properties in a subsurface region of interest. The method includes using data representative of geological, geophysical, and/or petrophysical attributes in the subsurface region, building an earth model for the subsurface region… calibrated against data relating to the subsurface region of interest. The earth model… are iteratively evaluated to converge modeled properties with measured attributes in the subsurface region of interest.”; par 15: “… the models are iteratively evaluated, with the goal of converging modeled attributes of the subsurface region towards measured attributes… these predicted modeled attributes can be compared with the measured data, where there is disagreement between prediction and measurement, the models 110, 112, 114 are adjusted and new predictions are generated…” EXAMINER NOTE: The above citation teaches that predictions can be implausible because they do not agree with the measured/observed physical attributes at the region of interest. Additionally, the above citations teach an iterative process which means that the current disagreement between predicted and measured data is indicative of the prior iteration according to the prior disagreement between the prior prediction and the prior measured data.)
Kolbjornsen and Miotti and Guldbrandsoy and Kacewicz_2011 are analogous art because they are from the same field of endeavor called subterranean models. Before the effective filing date, it would have been obvious to a person of ordinary skill in the art to combine Kolbjornsen and Miotti and Guldbrandsoy and Kacewicz_2011. The rationale for doing so would have been that, as outlined above, Kolbjornsen and Miotti and Guldbrandsoy make obvious to perform a first petrophysical inversion generating hydrocarbon flood results and to perform a second petrophysical inversion generating brine flood results. Kacewicz_2011 teaches what when there is an ensemble of models, including earth models, that predict petrophysical attributes (see par 4: petrophysical attributes) to iteratively evaluate the petrophysical attributes by comparing with measured attributes in the subsurface region of interest and if there is a disagreement (i.e., mismatch) to update the ensemble of models until their predictive quality converges to match the attributes of the subsurface region of interest. Therefore, it would have been obvious to combine Kolbjornsen and Miotti and Guldbrandsoy and Kacewicz_2011 for the benefit of having an ensemble of models that has high predictive quality that agrees with real world observations/measurements to obtain the invention as specified in the claims.
Kolbjornsen and Miotti and Guldbrandsoy and Kacewicz_2011 does not explicitly recite “and locations of the predicted artificial rock types.”
MacCormack_2017, however, makes obvious generating a model “and locations of the predicted artificial rock types” (abstract: “… it is important to ensure that the modeling techniques and procedures are able to accurately delineate and characterize the heterogeneity of the various geological environments included within the regional model domain… improved by splitting the regional model into multiple submodels based on the degree of variability observed between surrounding data points and informed by expert geological knowledge of the geological depositional framework… resulting in a more geologically realistic interpolation of the entire model domain as demonstrated…”;page 2 introduction: “… many three-dimensional (3D) subsurface geological models were created for various regions… the overall aim of these models is to characterize the nature and distribution of subsurface materials… the objective of this study is to determine if a model can be improved by splitting the regional model into multiple submodels based on the degree of variability observed between surrounding data points and our understanding of the depositional environment… thus optimizing the measured data values as inputs to the geostatistical model. The hypothesis is that by separating the regional model domain into submodels based on the data variability and our understanding of the geologic context, a variogram can be produced for each submodel region that is more representative of the true variability within the data for each area…”; page 12: “… a variogram model must be fit to the data to provide a prediction of the unknown values at specific location. The variogram model is essentially a generalized mathematical equation that provides the best fit to the observed data…”; page 14 conclusion: “the objective of this study was to determine if subsurface geological model accuracy could be improved by splitting the regional model into multiple submodels based on the degree of variability observed between surrounding data points… the hypothesis is that by separating the regional model domain into submodels based on the data variability and our understanding of the geological context, a variogram can be produced for each submodel region that is more representative of variability within each area, thus resulting in a more geostatistically appropriate and geologically realistic model… the proposed submodel approach produces a more geologically realistic model of the study area when compared with the regional model approach as evidenced by comparing the model results against the regional surficial geology map…”; FIG. 3, 4, 5, 6
EXAMINER NOTE: The above citations teach to break the geologic region of interest up into zones and to associate predictive models and the predictions (i.e., rock types) with the zone where they are most accurately predictive. Indeed, FIG. 3, 4, 5, and 6 illustrates the locations of the predicted stratigraphy on the geologic model.)
Kolbjornsen and Miotti and Guldbrandsoy and Kacewicz_2011 and MacCormack_2017 are analogous art because they are from the same field of endeavor called geologic models. Before the effective filing date, it would have been obvious to a person of ordinary skill in the art to combine Kolbjornsen and Miotti and Guldbrandsoy and Kacewicz_2011 and MacCormack_2017. The rationale for doing so would have been that Kolbjornsen and Miotti and Guldbrandsoy and Kacewicz_2011 make obvious to have an ensemble of models that predict attributes for a region of interest. MacCormack_2017 teaches that among an ensemble of models, some models are more accurately predictive than others in certain zones of the region of interest and because of this the region of interest should be subdivided into zones and the model most predictive for each zone should be identified for each of those zone in order the have a more geologically realistic model for the whole geologic region. Therefore, it would have been obvious to combine Kolbjornsen and Miotti and Guldbrandsoy and Kacewicz_2011 and MacCormack_2017 for the benefit of having a more geologically realistic model to obtain the invention as specified in the claims.
With respect to claim 6, Kolbjornsen in view of Miotti and Guldbrandsoy teaches all of the limitations of claim 5, as noted above. Kolbjornsen further teaches wherein at least one of the first and second petrophysical inversions comprises a facies-based inversion (methodology referred to as Bayesian AVO inversion, [Title]; which includes a discrete lithology-fluid prediction, [Abstract] line 11; specifically, the methodology lithology-fluid classes of shale, brine-sand, and oil-sand, [page 432 col 1 paragraph 4 lines 1-3]).
With respect to claim 8, Kolbjornsen in view of Miotti and Guldbrandsoy teaches all of the limitations of claim 5, as noted above. Kolbjornsen further teaches generating an image of the subsurface region based on the fluid saturation model (FIG. 4 bottom right results showing probability of oil-sand in an image, [page 433]).
With respect to claim 9, Kolbjornsen in view of Miotti and Guldbrandsoy teaches all of the limitations of claim 5, as noted above. Kolbjornsen further teaches managing hydrocarbons in the subsurface region based on the fluid saturation model (inversion results give shale thickness, well B is drilled in one section or two sections based on thickness of shale, [page 434 col 2 paragraph 3 lines 12-16]).
With respect to claim 10, Kolbjornsen in view of Miotti and Guldbrandsoy teaches all of the limitations of claim 5, as noted above. Kolbjornsen further teaches utilizing a machine learning system to classify the rock types (Markov chain Monte Carlo simulations are used to build initial probability models, [page 434 col 2 paragraph 3 lines 1-3; Markov referring to McMC simulations, [page 431 col 2 paragraph 2 line 16]).
With respect to claim 11, Kolbjornsen in view of Miotti and Guldbrandsoy teaches all of the limitations of claim 5, as noted above. Kolbjornsen further teaches performing one or more additional petrophysical inversions for the subsurface region with the at least one artificial rock type prior to generating the final results (methodology referred to as Bayesian AVO inversion, [Title]; which includes a discrete lithology-fluid prediction, [Abstract] line 11; specifically, the methodology lithology-fluid classes of shale, brine-sand, and oil-sand, [page 432 col 1 paragraph 4 lines 1-3]; and an additional class of undefined rock, [page 433 col 2 paragraph 2 lines 1-11]).
Claim(s) 13-19 is/are rejected under 35 U.S.C. 103 as being unpatentable over "Bayesian AVO inversion to rock properties using a local neighborhood in a spatial prior model" (Kolbjornsen) in view of U.S. 2015/0362623 (Miotti) in further view of “Seismic signatures of siliciclastic reservoirs as function of burial history – a Barents Sea example” (Guldbrandsoy) in view of Kacewicz_2011.
With respect to claim 13, Kolbjornsen teaches A method for generating a fluid saturation model for a subsurface region comprising (section labeled "The local-neighborhood approach for integrating spatial structure", [page 432 col 1 paragraph 3]-[page 433 col 2 paragraph 5]): using a computer (method is introduced to reduce computation cost, [Abstract] line 6), performing a first petrophysical inversion for the subsurface region with brine flood parameters to generate inversion results (seismic response of local lithology-fluid combination, [page 433 col 1 paragraph 2 lines 1-2]; where the combination is B-B from FIG. 3, [page 433], and inversion is the swapped point of view, [page 431 col 2 paragraph 2 line 3], i.e. seismic response to rock properties rather than vice versa), the brine flood parameters representing an assumption of brine flooding In all available pore space in the subsurface region, (the seismic response of B-B is one of eight possible lithology fluid scenarios, [page 433 col 1 paragraph 2 lines 3-5], where a scenario in two neighboring cells assumes a lithology-fluid classification in order to perform the inverse problem on a local neighborhood, [page 432 col 1 paragraph 2 lines 4-14]); classifying rock types in the subsurface region based on the inversion results (obtain probability of eight different scenarios and sum events of top cell to get posterior probability, [page 433 col 1 paragraph 3 lines 7-12]), wherein the rock types comprise at least one artificial rock type (introduce additional scenario of undefined event, which can model an undefined rock [page 433 col 2 paragraph 2 lines 1-11]); generating a trial fluid saturation model based on the classified rock types (define posterior transition probabilities, which can be used to build spatial model, where the spatial model is not the true posterior distribution, but a first order approximation, [page 433 col 2 paragraph 1 lines 1-4]; note the spatial model is not a local window, but instead the entire grid, [page 432 col 1 paragraph 4 lines 3-4]; see also FIG. 4(b) (right) panel, [page 434]); and using the computer (method is introduced to reduce computation cost, [Abstract] line 6), identifying the potential hydrocarbon-bearing formations in the subsurface region based on the fluid saturation model (calculate probability of oil sand by assigning most probable lithology fluid class, [page 434 col 2 paragraph 1 lines 1-5]; see also FIG. 4(c) right panel, [page 434]).
Kolbjornsen does not teach in substantially all available pore space in the subsurface region; using the computer, iteratively repeating until convergence: performing a trial petrophysical inversion for the subsurface region with the trial fluid saturation model to generate trial results; updating the inversion results with the trial results; and checking convergence; and using the computer, generating the fluid saturation model for the subsurface region based on the inversion results.
However, Miotti teaches using the computer (example system includes processor, and memory operatively coupled to the process with one or more instructions to implement the inverse problem, [0184] lines 1-12), iteratively repeating until convergence (at block 750, current model is updated iteratively, [0110] lines 1-2; until converge at stopping criteria of local optimum, [0113] lines 5-7): performing a trial petrophysical inversion for the subsurface region with the trial fluid saturation model to generate trial results (at block 750, current model is updated iteratively using solution of inverse problem to equation 11, [0110] lines 1-3); updating the inversion results with the trial results (current model m_k updated to obtain new model m_k+1, [0110] lines 6-7); and checking convergence (at block 760, compare model with stopping criteria, [0112] lines 1-2; until converge at stopping criteria of local optimum, [0113] lines 5-7); and using the computer, generating the fluid saturation model for the subsurface region based on the inversion results (at block 760, compare model with stopping criteria, [0112] lines 1-2; until converge at stopping criteria of local optimum, [0113] lines 5-7).
Kolbjornsen and Miotti do not teach in substantially all available pore space in the subsurface region.
However, Guldbrandsoy teaches in substantially all available pore space in the subsurface region (“The rock physics modeling has been done using 100% brine saturation and 100% gas saturation, to give a better understanding of the extremes of fluid substitution, [page 43 paragraph 2 lines 7-9]; for a description of how pore fluids are used in the Gassmann equation, see [pages 27-29]; and for the application to velocity models in each specific layer Period A-Top Layer, see [pages 58-66], example on [page 65] describes how misfits can be used to update the model).
It would have been obvious to one skilled in the art before the effective filing date to combine Kolbjornsen with Miotti because a teaching, suggestion, or motivation in the prior art would have led one skilled in the art to combine prior art teaching to arrive at the claimed invention. Kolbjornsen discloses a system that teaches all of the claimed features except for performing a third petrophysical inversion and generating the fluid saturation mode from the results of that inversion. Kolbjornsen suggests “An analysis on a local scale gives interesting results that in turn can be further combined to derive properties on a larger scale” (Kolbjornsen [page 435 col 2 paragraph 4 lines 3-5]). Miotti teaches an iterative approach, where in an iteration the inverse solver block 1150 receives additional information to solve for the estimated model, (Miotti [0178] lines 7-18). A person having skill in the art would have a reasonable expectation of successfully increasing the confidence of the results in the system and method of Kolbjornsen by modifying Kolbjornsen with the iterative forward modeling (Miotti 1134, [0178] line 3) and inverse solver of Miotti (Miotti [0178] lines 6-10) by including the statistical methods of Kolbjornsen (Kolbjornsen [page 433 col 1 paragraph 1 lines 1-16]) as the additional information in Miotti (Miotti [0178] lines 10-18). Therefore, it would have been obvious to combine Kolbjornsen with Miotti to a person having ordinary skill in the art, and this claim is rejected under 35 U.S.C. 103.
It would have been obvious to one skilled in the art before the effective filing date to combine Kolbjornsen in view of Miotti with Guldbrandsoy because a teaching, suggestion, or motivation in the prior art would have led one skilled in the art to combine prior art teaching to arrive at the claimed invention. Kolbjornsen in view of Miotti discloses a system and method that teaches all of the claimed features except for representing an assumption of hydrocarbon/brine flooding in substantially all available pore space. Kolbjornsen provides lithology-fluid classes, but does not specifically define the percentage of brine in brine sand or percentage of oil in oil sand (Kolbjornsen [page 432 col 1 paragraph 4]). Miotti teaches how the Gassmann equation can be used to perform fluid substitution in order to forward model, (Miotti see step 1134 in FIG. 11, [0099], and [0178] line 3). Finally, Guldbrandsoy provides the same Gassmann equation, (Guldbrandsoy [pages 27-29]), but further explains that modeling is done as 100% brine and 100% gas to give a better understanding of the extremes of fluid substitution (Guldbrandsoy [page 43 paragraph 2 lines 7-9]). In an example, when a layer does not fit well with a given substitution, the modeling can be redone as a mix between the two extremes (Guldbrandsoy [page 65 paragraph 2 lines 1-6]). A person having skill in the art would have a reasonable expectation of understanding the extremes in the system and method of Kolbjornsen in view of Miotti by modifying Kolbjornsen in view of Miotti with the 100% substitutions of Guldbrandsoy. Therefore, it would have been obvious to combine Kolbjornsen in view of Miotti with Guldbrandsoy to a person having ordinary skill in the art, and this claim is rejected under 35 U.S.C. 103.
Kacewicz_2011; however, makes obvious identifying a first set of misfits “wherein an artificial rock type used to model data misfits is added as a prior facie at the beginning of the inversion”((Figure 2 illustrates “good fit” yes or no and if there is not a good fit (i.e., mismatch) then to update the model; par 1: “the present invention relates generally to modeling of hydrocarbon reservoirs and more particularly to integrating mechanical earth models, earth models and basin models”; par 4: “… a method of modeling properties in a subsurface region of interest. The method includes using data representative of geological, geophysical, and/or petrophysical attributes in the subsurface region, building an earth model for the subsurface region… calibrated against data relating to the subsurface region of interest. The earth model… are iteratively evaluated to converge modeled properties with measured attributes in the subsurface region of interest.”; par 15: “… the models are iteratively evaluated, with the goal of converging modeled attributes of the subsurface region towards measured attributes… these predicted modeled attributes can be compared with the measured data, where there is disagreement between prediction and measurement, the models 110, 112, 114 are adjusted and new predictions are generated…” EXAMINER NOTE: Adding a geophysical attribute predicted by a previous inversion—but which differed from observations—as a prior constraint in a new iteration of the inversion, is a process known as iterative inversion, iterative re-weighting, or model-based refinement. The above citation teaches to perform iterative model-based refinement. Additionally, the above citations teach an iterative process which means that the current disagreement between predicted and measured data is indicative of the prior iteration according to the prior disagreement between the prior prediction and the prior measured data.)
Kolbjornsen and Miotti and Guldbrandsoy and Kacewicz_2011 are analogous art because they are from the same field of endeavor called subterranean models. Before the effective filing date, it would have been obvious to a person of ordinary skill in the art to combine Kolbjornsen and Miotti and Guldbrandsoy and Kacewicz_2011. The rationale for doing so would have been that, as outlined above, Kolbjornsen and Miotti and Guldbrandsoy make obvious to perform a first petrophysical inversion generating hydrocarbon flood results and to perform a second petrophysical inversion generating brine flood results. Kacewicz_2011 teaches what when there is an ensemble of models, including earth models, that predict petrophysical attributes (see par 4: petrophysical attributes) to iteratively evaluate the petrophysical attributes by comparing with measured attributes in the subsurface region of interest and if there is a disagreement (i.e., mismatch) to update the ensemble of models until their predictive quality converges to match the attributes of the subsurface region of interest. Therefore, it would have been obvious to combine Kolbjornsen and Miotti and Guldbrandsoy and Kacewicz_2011 for the benefit of having an ensemble of models that has high predictive quality that agrees with real world observations/measurements to obtain the invention as specified in the claims.
With respect to claim 14, Kolbjornsen in view of Miotti and Guldbrandsoy teaches all of the limitations of claim 13, as noted above. Kolbjornsen further teaches wherein at least one of the first petrophysical inversion and the trial petrophysical inversions comprises a facies-based inversion (methodology referred to as Bayesian AVO inversion, [Title]; which includes a discrete lithology-fluid prediction, [Abstract] line 11; specifically, the methodology lithology-fluid classes of share, brine-sand, and oil-sand, [page 432 col 1 paragraph 4 lines 1-3]).
With respect to claim 15, Kolbjornsen in view of Miotti teaches all of the limitations of claim 13, as noted above. Kolbjornsen does not teach wherein the check for convergence comprises: comparing the inversion results from a prior iteration to the trial results to determine a remaining error estimate; and determining whether the remaining error estimate is below a selected error threshold.
However, Miotti teaches wherein the check for convergence comprises: comparing the inversion results from a prior iteration to the trial results to determine a remaining error estimate (equation 13, ||m_k+1 – m_k||, [0112] line 5); and determining whether the remaining error estimate is below a selected error threshold (equation 13, < ε, [0112] line 5).
It would have been obvious to one skilled in the art before the effective filing date to combine Kolbjornsen with Miotti because a teaching, suggestion, or motivation in the prior art would have led one skilled in the art to combine prior art teaching to arrive at the claimed invention. Kolbjornsen discloses a system that teaches all of the claimed features except for performing a third petrophysical inversion and generating the fluid saturation mode from the results of that inversion. Kolbjornsen suggests “An analysis on a local scale gives interesting results that in turn can be further combined to derive properties on a larger scale” (Kolbjornsen [page 435 col 2 paragraph 4 lines 3-5]). Miotti teaches an iterative approach, where in an iteration the inverse solver block 750 receives additional cross-properties to solve for the estimated model, (Miotti [0109] lines 4-5). Then solves until convergence using the standard convergence equation (13), (Miotti [0112] line 4). A person having skill in the art would have a reasonable expectation of successfully increasing the confidence of the results in the system and method of Kolbjornsen by modifying Kolbjornsen with the iterative solving of the inverse problem (Miotti [0110] lines 1-3) until convergence (Miotti [0112]-[0113]) by including the statistical methods of Kolbjornsen (Kolbjornsen [page 433 col 1 paragraph 1 lines 1-16]) as the additional cross-properties in Miotti (Miotti [0109] lines 4-5). Therefore, it would have been obvious to combine Kolbjornsen with Miotti to a person having ordinary skill in the art, and this claim is rejected under 35 U.S.C. 103.
With respect to claim 16, Kolbjornsen in view of Miotti and Guldbrandsoy teaches all of the limitations of claim 13, as noted above. Kolbjornsen further teaches wherein the trial petrophysical inversion inverts for water 20saturation (approach is applicable for properties including saturation, [Abstract] lines 10-12).
With respect to claim 17, Kolbjornsen in view of Miotti and Guldbrandsoy teaches all of the limitations of claim 13, as noted above. Kolbjornsen further teaches managing hydrocarbons in the subsurface region based on the fluid saturation model (inversion results give shale thickness, well B is drilled in one section or two sections based on thickness of shale, [page 434 col 2 paragraph 3 lines 12-16]).
With respect to claim 18, Kolbjornsen in view of Miotti and Guldbrandsoy teaches all of the limitations of claim 13, as noted above. Kolbjornsen further teaches utilizing a machine learning system to classify the rock types (Markov chain Monte Carlo simulations are used to build initial probability models, [page 434 col 2 paragraph 3 lines 1-3; Markov referring to McMC simulations, [page 431 col 2 paragraph 2 line 16]).
With respect to claim 19, Kolbjornsen in view of Miotti and Guldbrandsoy teaches all of the limitations of claim 13, as noted above. Kolbjornsen further teaches wherein the trial petrophysical inversion utilizes the at least one artificial rock type (methodology referred to as Bayesian AVO inversion, [Title]; which includes a discrete lithology-fluid prediction, [Abstract] line 11; specifically, the methodology lithology-fluid classes of shale, brine-sand, and oil-sand, [page 432 col 1 paragraph 4 lines 1-3]; and an additional class of undefined rock, [page 433 col 2 paragraph 2 lines 1-11]).
Claim(s) 12 is rejected under 35 U.S.C. 103 as being unpatentable over "Bayesian AVO inversion to rock properties using a local neighborhood in a spatial prior model" (Kolbjornsen) in view of U.S. 2015/0362623 (Miotti) in further view of “Seismic signatures of siliciclastic reservoirs as function of burial history – a Barents Sea example” (Guldbrandsoy) in view of Kacewicz_2011 in view of MacCormack_2017 in still further view of “Probabilistic petrophysical-properties estimation integrating statistical rock physics with seismic inversion” (Grana).
With respect to claim 12, Kolbjornsen does not teach wherein classifying the rock types comprises generating one or more cross-plots of porosity and volume of clay.
However, Grana teaches classifying the rock types comprises generating one or more cross-plots of porosity and volume of clay (FIG. 5A, [page 28]).
It would have been obvious to one skilled in the art before the effective filing date to combine Kolbjornsen in view of Miotti with Grana because this is applying a known technique (Grana) to a known method and device (Kolbjornsen in view of Miotti) ready for improvement to yield predictable results. Kolbjornsen in view of Miotti is the base reference that teaches all limitations except for classifying rock types with a cross plot of volume of clay and porosity. Kolbjornsen in view of Miotti is ready for improvement because volume of clay and porosity are useful petrophysical properties in determining lithology an underground region. Grana teaches a known technique of using Bayesian seismic inversion to estimate petrophysical properties, (Grana [Abstract] lines 1-3). Because elastic attributes are correlated with petrophysical variables (effective porosity, clay content, and water saturation) and this physical link is associated with uncertainties, the petrophysical-properties estimation from seismic data can be seen as a Bayesian inversion problem, (Grana [Abstract] lines 1-3). This method allows for not only calculating probabilities of litho-fluid classes as described in Kolbjornsen, but also, specific petrophysical properties, (Grana [Abstract] lines 14-15). This approach allows one to explore all possible ranges of water saturation, clay content, and porosity, (Grana [page 22 col 2 paragraph 1 line 1]), and a cross-plot is a known technique for visually showing the relation of these properties (FIG. 5a of Grana [page 28]). Thus, one having ordinary skill in the art would have recognized that applying the known technique in Grana of estimating petrophysical properties in addition to litho-fluid classes would yield the predictable result of providing more information regarding the lithology of the underground region determine through the method and system of Kolbjornsen in view of Miotti. Therefore, it would have been obvious to combine Kolbjornsen in view of Miotti with Grana to a person having ordinary skill in the art, and this claim is rejected under 35 U.S.C. 103.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to BRIAN S COOK whose telephone number is (571)272-4276. The examiner can normally be reached 8:00 AM - 5:00 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, 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.
/BRIAN S COOK/ Primary Examiner, Art Unit 2187
Prosecution Insights