DETAILED ACTION
Claims 1-20 are presented for examination.
This Office Action is in response to submission of documents on December 4, 2025.
Rejection of claims 1-20 under 35 U.S.C. 101 for being directed to unpatentable subject matter is maintained.
Rejection of claims 1-5, 8-12, and 15-19, under 35 U.S.C. 103 as being obvious over Chu in view of Hwang, Campin, Kovachki, Chattopadhyay, and Seroka is withdrawn.
Rejection of claims 6-7, 13-14,and 20 under 35 U.S.C. 103 as being obvious over Chu in view of Hwang, Campin, Kovachki, Chattopadhyay, Seroka, and Kida is withdrawn.
New rejection of claims 1-3, 5, 8-10 , 12, 15-17, and 19 under 35 U.S.C. 103 as being obvious over Jiang in view of Seroka.
New rejection of claims 6-7, 13-14, and 20 under 35 U.S.C. 103 as being obvious over Jiang in view of Seroka and Kida.
New rejection of claims 4, 11 and 18 under 35 U.S.C. 103 as being obvious over Jiang in view of Seroka, Kida, and Campin.
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Information Disclosure Statement
The information disclosure statement (IDS) submitted on Jul 11. 2025, is in compliance with the provisions of 37 CFR 1.97. Accordingly, the information disclosure statement is being considered by the examiner.
Response to Arguments
Regarding the rejection of claims 1-20 as being directed to unpatentable subject matter, Examiner is not persuaded by the present amendments and arguments. As currently presented, the claims are directed to abstract ideas and additional elements that do not integrate the judicial exceptions into a practical application nor recite an improvement in computer technology and/or another field. Further, the claims do not include elements that amount to significantly more than the judicial exceptions. Accordingly, the rejection of claims 1-20 as being directed to unpatentable subject matter are maintained for the following reasons:
Applicant asserts that “presenting a visual map of flood depth changing over time generated based on the operational forecasting model predicting a storm surge” is not an abstract idea because it does not fall into one of the judicial exceptions, as analyzed at Step 2A, Prong 1. Examiner disagrees. “Presenting a visual map of flood depth” is a method of organizing human activity because the step of “presenting a map” is an activity that can be performed by a human. See MPEP 2106.04(a)(2), Subsection II. For example, “presenting a visual map of flood depth changing over time” is a step that can be performed by a meteorologist on a broadcast by projecting a map onto a screen for viewers to observe. Thus, this portion of the limitation is judicial exception.
Further, the limitation of the “map…generated based on the operational forecasting model predicting a storm surge” is an idea of a solution. See MPEP 2106.05(f)(1). “The recitation of claim limitations that attempt to cover any solution to an identified problem with no restriction on how the result is accomplished and no description of the mechanism for accomplishing the result, does not integrate a judicial exception into a practical application or provide significantly more because this type of recitation is equivalent to the words ‘apply it’.” Id. Courts have found that ideas of a solution do not integrate a judicial exception into a practical application and do not amount to significantly more than the recited judicial exception. See Electric Power Group, LLC v. Alstom, S.A., 830 F.3d 1350, 1356, 119 USPQ2d 1739, 1743-44 (Fed. Cir. 2016); Intellectual Ventures I v. Symantec, 838 F.3d 1307, 1327, 120 USPQ2d 1353, 1366 (Fed. Cir. 2016); Internet Patents Corp. v. Active Network, Inc., 790 F.3d 1343, 1348, 115 USPQ2d 1414, 1417 (Fed. Cir. 2015).
Applicant further asserts that the claim recites an improvement to a computer or to a technological field. Response at pg. 8. To support the argument, Applicant cites the Specification at [0024], [0043], and [0045]. However, none of the asserted improvements are reflected in the claim itself. “During examination, the examiner should analyze the ‘improvements’ consideration by evaluating the specification and the claims to ensure that a technical explanation of the asserted improvement is present in the specification, and that the claim reflects the asserted improvement.” MPEP 2106.05(a). Examiner maintains that the improvements set forth in the specification are not reflected in the claims other than conclusory recitations (e.g. “predicting a storm surge in real time”).
Finally, Applicant asserts that “building a surrogate model using neural operators” is one example of an improvement in computer technology and/or the field of weather prediction. However, this recitation does not further recite how the neural operators are generated, trained, configured, and/or utilized in performing any of the asserted improvements. Thus, while the title of the application and over three pages of the specification are directed to the use of neural operators,” the claims themselves recite little detail into how neural operators are integrated into the other claimed features.
Accordingly, rejection of claims 1-20 under 35 U.S.C. 103 as being directed to unpatentable subject matter is maintained.
Regarding the rejection of claims 1-20 under 35 U.S.C. 103, Examiner is persuaded by the arguments and amendments presented with the Response. Accordingly, rejections of the claims under 35 U.S.C. 103 as being obvious over Chu in view of Hwang, Campin, Kovachki, Chattopadhyay, Seroka, and/or Kida are withdrawn. However, in light of the amendments to the claims, additional searching was required and the claims are rejected herein under 35 U.S.C. 103 as being obvious over Jiang in view of Seroka, Kida, and/or Campin.
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claims 1-20 are rejected under 35 U.S.C. 101 because the claimed invention is directed to judicial exceptions without significantly more. The claims recite mathematical calculations and additional elements. This judicial exception is not integrated into a practical application because the additional elements that are recited in the claims are extra-solution activities that do not integrate the judicial exceptions into a practical application. The claims do not include additional elements that are sufficient to amount to significantly more than the judicial exception because courts have found that the step of data acquisition and reciting the limitations of an idea of a solution and generic computer components are not significantly more than a judicial exception.
Claim 1
Step 1: The claim is directed to a process, falling under one of the four statutory categories of invention.
Step 2A, Prong 1: The claim 1 limitations include (bolded for abstract idea identification):
Claim 1
Mapping Under Step 2A Prong 1
A computer-implemented method comprising:
receiving initial and boundary conditions, and parameters associated with geophysical modeling;
based on the received initial and boundary conditions and parameters, running a multiscale model for data generation to produce first resolution simulation data and second resolution simulation data for a surrogate machine learning model training,
wherein the second resolution simulation data has higher resolution than the first resolution simulation data,
wherein the running of the multiscale model includes running a neural-network based model that generates the second resolution simulation data and
running a physics-based model that generates the first resolution simulation data,
wherein the neural-network based model receive low resolution information from the physics-based model at time t and
the physics-based model receives high resolution information from the neural-network based model at time t+1 until the data generation ends;
creating a surrogate model using neural operators, wherein the surrogate model is trained using the first resolution simulation data and second resolution simulation data;
generating an operational forecasting model using the surrogate model; and
presenting a visual map of flood depth changing over time generated based on the operational forecasting model predicting a storm surge.
Abstract Idea: Mathematical Calculations
A model is comprised on mathematical functions that are calculated to generate output. See MPEP § 2106.04(a)(2), Subsection I.
Abstract Idea: Mathematical Calculations
Running a model includes executing one or more mathematical functions to generate output that is indicative of real-world conditions. See MPEP § 2106.04(a)(2), Subsection I.
Abstract Idea: Mathematical Calculations
Running a model includes executing one or more mathematical functions to generate output that is indicative of real-world conditions. See MPEP § 2106.04(a)(2), Subsection I.
Abstract Idea: Mathematical Calculations
A surrogate model is a substitute for the previously executed model. The surrogate model, when executed and/or trained, requires multiple calculations to be performed. See MPEP § 2106.04(a)(2), Subsection I.
Abstract Idea: Mathematical Calculations
A model is comprised on mathematical functions that are calculated to generate output. See MPEP § 2106.04(a)(2), Subsection I.
Abstract Idea: Organizing Human Activity
The limitation is directed to “managing personal behavior and relationships or interactions between people” because presenting, as claimed herein, can be performed by a human and can include interacting with other humans to provide and/or otherwise make known a map to others. See MPEP 2106.04(a)(2), Subsection II.
Step 2A, Prong 2: The claim 1 limitations recite (bolded for additional element identification):
Claim 1
Mapping Under Step 2A Prong 2
A computer-implemented method comprising:
receiving initial and boundary conditions, and parameters associated with geophysical modeling;
based on the received initial and boundary conditions and parameters, running a multiscale model for data generation to produce first resolution simulation data and second resolution simulation data for a surrogate machine learning model training,
wherein the second resolution simulation data has higher resolution than the first resolution simulation data,
wherein the running of the multiscale model includes running a neural-network based model that generates the second resolution simulation data and
running a physics-based model that generates the first resolution simulation data,
wherein the neural-network based model receive low resolution information from the physics-based model at time t and
the physics-based model receives high resolution information from the neural-network based model at time t+1 until the data generation ends;
creating a surrogate model using neural operators, wherein the surrogate model is trained using the first resolution simulation data and second resolution simulation data;
generating an operational forecasting model using the surrogate model; and
presenting a visual map of flood depth changing over time generated based on the operational forecasting model predicting a storm surge.
Implementing a judicial exception using generic computer components is an extra-solution activity that does not integrate the exception into a practical application because the limitation is mere instructions to apply it using a computer as a tool. See MPEP 2106.05(f)(2)
Receiving data is the extra-solution activity of data gathering because the step is necessary to train and/or construct any mathematical model. See MPEP 2106.05(g)(3).
Receiving data is the extra-solution activity of data gathering and transmission which is necessarily a step required to generate all models. See MPEP 2106.05(g)(3).
Receiving data is the extra-solution activity of data gathering and transmission which is necessarily a step required to generate all models. See MPEP 2106.05(g)(3).
Step 2B: Regarding Step 2B, the inquiry is whether any of the additional elements (i.e., the elements that are not the judicial exception) amount to significantly more than the recited judicial exception. Reciting generic computer components to apply a judicial exception is an additional element that courts have found to be insignificantly more than the judicial exception. See MPEP 2106.05(f)(2). See also See Affinity Labs v. DirecTV, 838 F.3d 1253, 1262, 120 USPQ2d 1201, 1207 (Fed. Cir. 2016) (cellular telephone); TLI Communications LLC v. AV Auto, LLC, 823 F.3d 607, 613, 118 USPQ2d 1744, 1748 (Fed. Cir. 2016) (computer server and telephone unit); Alice Corp. Pty. Ltd. V. CLS Bank Int’l, 573 U.S. 208, 223, 110 USPQ2d 1976, 1983 (2014); Gottschalk v. Benson, 409 U.S. 63, 64, 175 USPQ 673, 674 (1972); Versata Dev. Group, Inc. v. SAP Am., Inc., 793 F.3d 1306, 1334, 115 USPQ2d 1681, 1701 (Fed. Cir. 2015). Further, receiving data is an extra-solution activity that courts have found to be insignificantly more than the recited judicial exception. See 2106.05(g). See also In re Grams, 888 F.2d 835, 839-40; 12 USPQ2d 1824, 1827-28 (Fed. Cir. 1989); In re Meyers, 688 F.2d 789, 794; 215 USPQ 193, 196-97 (CCPA 1982); OIP Technologies, 788 F.3d at 1363, 115 USPQ2d at 1092-93; CyberSource v. Retail Decisions, Inc., 654 F.3d 1366, 1375, 99 USPQ2d 1690, 1694 (Fed. Cir. 2011).
Accordingly, claim 1 is rejected for being directed to unpatentable subject matter.
Claim 2
Claim 2 recites wherein the operational forecasting model functions as a partial differential equation solver that can work on a plurality of different multiscale modeling formulations. A partial differentiation solver is a mathematical framework to solve a mathematical problem. Thus, the claim is directed to a specific type of model, which has been rejected for being a mathematical concept or calculation. Claim 2 does not add any additional elements that would integrate the judicial exception into a practical application. Accordingly, claim 2 is rejected for being directed to unpatentable subject matter.
Claim 3
Claim 3 recites wherein a partial differential equations family is learned over all parameters using the neural operators. The claim merely further specifies the behavior of the surrogate model and does not include any additional elements that would integrate the surrogate model into a practical application. Instead, the claim specifies the type of functions that are approximated using the model. Accordingly, claim 3 is rejected for being directed to unpatentable subject matter.
Claim 4
Claim 4 recites wherein an amount of data needed for neural operator training is reduced by using super-parametrization. The claim merely specifies the type of data that is utilized by the system to train the neural network. Using super-parametrization in multi-scale modeling is an extra-solution activity that is insignificantly more than the judicial exception because it is well-understood, routine, and conventional. See, e.g., Campin, et al., “Super-parametrization in Ocean Modeling: Application to Deep Convention” at Abstract (“We explore the efficacy of ‘super parametrization’ (SP) in ocean modeling…”). Accordingly, claim 4 is rejected for being directed to unpatentable subject matter.
Claim 5
Claim 5 recites wherein the surrogate model captures features at a resolution higher than the first resolution simulation data. The claim includes limitations that further specify the type of output from the surrogate model, which has been previously rejected for being directed to the judicial exception of a mathematical concept or calculation. The claim does not include additional elements that could integrate the judicial exception into a practical application. Accordingly, claim 5 is rejected for being directed to unpatentable subject matter.
Claim 6
Claim 6 recites further including combining the trained surrogate model with an impact model for risk assessment. The claim includes limitations that incorporate a model into a larger framework that utilizes simulated weather patterns to determine risk (e.g., flood risk). The claim is directed to an idea of a solution, which is an extra-solution activity that is well-understood, routine, and conventional. For example, the surrogate model is a stand-in for a higher resolution model (e.g., a multi-scale model) that is commonly used in conjunction with an impact model. See, e.g., Chattopadhyay, et al., “Data-driven Surrogate Models for Climate Change modeling: Application of Echo State Networks, RNN-LSTM and ANN to Multi-scale Lorenz System as a Test Case” at Abstract, disclosing “These surrogate models train on observational data.…In this work, we have built a data-driven model based on echo state networks (ESN) aimed, specifically at climate modeling.”; Kida (U.S. Patent Pub. No.), disclosing “The runoff analysis unit 23 has a runoff analysis model M2, and predicts a flood region (point) based on the input future predicted rainfall, water level information, and flow rate information. Note that the runoff analysis model M2 itself, which predicts the flood region based on the predicted rainfall is a known technique….” Kida at [0049].
Accordingly, claim 6 is rejected for being directed to unpatentable subject matter.
Claim 7
Claim 7 recites:
wherein the impact model includes coastal flood prediction model, the first resolution simulation data and the second resolution simulation data can include at least data associated sea surface height, and the method further includes triggering a physical barrier to open or close.
This limitation limits the judicial exception to a particular field of use by merely specifying the types of data that are utilized for the multi-scale and surrogate model. “As explained by the Supreme Court, a claim directed to a judicial exception cannot be made eligible “simply by having the applicant acquiesce to limiting the reach of the patent for the formula to a particular technological use.” Diamond v. Diehr, 450 U.S. 175, 192 n.14, 209 USPQ 1, 10 n. 14 (1981). Thus, limitations that amount to merely indicating a field of use or technological environment in which to apply a judicial exception do not amount to significantly more than the exception itself, and cannot integrate a judicial exception into a practical application.” MPEP 2106.05(h).
Accordingly, claim 7 is rejected for being directed to unpatentable subject matter. “When determining whether a claim simply recites a judicial exception with the words “apply it” (or an equivalent), such as mere instructions to implement an abstract idea on a computer, examiners may consider the following: (1) Whether the claim recites only the idea of a solution or outcome i.e., the claim fails to recite details of how a solution to a problem is accomplished.” MPEP 2106.05(f)(1). The limitation is an attempt to cover all solutions to the identified problem without restriction because it merely recites “triggering a physical barrier” and does not tie the step into the rest of the claims. Further, courts have found that ideas of a solution are not significantly more than the judicial exception. See Electric Power Group, LLC v. Alstom, S.A., 830 F.3d 1350, 1356, 119 USPQ2d 1739, 1743-44 (Fed. Cir. 2016); Intellectual Ventures I v. Symantec, 838 F.3d 1307, 1327, 120 USPQ2d 1353, 1366 (Fed. Cir. 2016); Internet Patents Corp. v. Active Network, Inc., 790 F.3d 1343, 1348, 115 USPQ2d 1414, 1417 (Fed. Cir. 2015).
Claims 8-14
Claims 8-14 and 20 are rejected under 35 U.S.C. 101 because the claimed invention is directed to non-statutory subject matter. The claims do not fall within at least one of the four categories of patent eligible subject matter because they are directed to software per se. However, assuming that the claim is amended to remedy the software per se issue, the claims are further analyzed to determine whether the amended claims are otherwise patent eligible. See MPEP 2106.03, Subsection II: “If a claim is clearly not within one of the four categories (Step 1: NO), then a rejection under 35 U.S.C. 101 must be made indicating that the claim is directed to non-statutory subject matter.…see MPEP § 2106.07(a)(1). However, as shown in the flowchart in MPEP § 2106 subsection III, when a claim fails under Step 1 (Step 1: NO), but it appears from applicant’s disclosure that the claim could be amended to fall within a statutory category (Step 1: YES), the analysis should proceed to determine whether such an amended claim would qualify as eligible…”
Claim 8
Claim 8 recites:
A computer program product comprising a computer readable storage medium having program instructions embodied therewith, the program instructions readable by a device to cause the device to: perform a method that is substantially the same as the method recited in claim 1. A computer program is not a statutory category of patentable subject matter. Even if amended to remedy the software per se rejection of the claim, for at least the same reasons as asserted regarding claim 1, claim 8 is rejected under 35 U.S.C. 101 for being directed to unpatentable subject matter.
Claim 9-14
Claims 9-14 recite the computer program of claim 8 and substantially the same limitations as recited in claims 2-7, with all of the limitations of claim 14 recited in claim 7. Even if amended to remedy the software per se rejection of the claim, for at least the same reasons as asserted regarding claims 2-7, claims 9-14 are directed to unpatentable subject matter.
Claim 15
Claim 15 recites A system comprising: a processor; and a memory device coupled with the processor, the processor configured to at least: perform the steps of the method recited in claim 1. Accordingly, for at least the reasons asserted regarding claim 1, claim 10 is directed to unpatentable subject matter.
Claims 16-19
Claims 16-19 recite the system of claim 8 and substantially the same limitations as recited in claims 2-5. Accordingly, for at least the same reasons as asserted regarding claims 2-5, claims 16-19 are directed to unpatentable subject matter.
Claim 20
Claim 20 is identical to claim 13 and therefore is directed to patentable subject matter for at least the same reasons as claim 6. Further, even if amended, as previously suggested in the rejection of claim 20 under 35 U.S.C. 112(b), for at least the same reasons as asserted regarding claim 6, claim 20 is directed to unpatentable subject matter.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
This application currently names joint inventors. In considering patentability of the claims the examiner presumes that the subject matter of the various claims was commonly owned as of the effective filing date of the claimed invention(s) absent any evidence to the contrary. Applicant is advised of the obligation under 37 CFR 1.56 to point out the inventor and effective filing dates of each claim that was not commonly owned as of the effective filing date of the later invention in order for the examiner to consider the applicability of 35 U.S.C. 102(b)(2)(C) for any potential 35 U.S.C. 102(a)(2) prior art against the later invention.
Claims 1-3, 5, 8-10 , 12, 15-17, and 19 are rejected under 35 U.S.C. 103 as being obvious over Jiang, et al. (“Digital Twin Earth - Coasts: Developing a fast and physics-informed surrogate model for coastal floods via neural operators,” hereinafter “Jiang”) in view of Seroka, et al. (“NOAA’s Global Extratropical Surge and Tide Operational Forecast System (Global ESTOFS),” hereinafter “Seroka”).
Claim 1
Jiang discloses:
A computer-implemented method comprising:
We trained each model using MSE as the loss function over 50 epochs and batch size 32, on one Tesla A100 Graphics Processing Unit (GPU). Jiang at pg. 3.
receiving initial and boundary conditions, and parameters associated with geophysical modeling;
Here, we simulate a coupled system of nonlinear equations including 2D momentum balance for water velocity, mass balance, and boundary conditions between ocean/sea floor/sea surface. Jiang at pg. 3.
“Water velocity” and “mass balance” include initial conditions required to initialize the model.
based on the received initial and boundary conditions and parameters, running a multiscale model for data generation to produce first resolution simulation data and second resolution simulation data for a surrogate machine learning model training, wherein the second resolution simulation data has higher resolution than the first resolution simulation data,
Here, we propose the first “coastal digital twin”, an emulator built on state-of-art physics-informed ML techniques to produce computationally lightweight surrogate models that provide fast and accurate predictions of sea surface heights in coastal regions. As a proof-of-concept experiment, we developed a digital twin for the NEMO simulations in northwestern Europe using an improved version of FNO. Our results show: (1) the extension of FNO to learn multivariate dynamics (note that FNO was used for univariate cases in its original development [14]); (2) the overall superior performance of FNO over the baseline model UNet [15] in emulating sea surface height; (3) the adverse impact of masked land boundaries in training FNO; and (4) a 45x acceleration achieved by FNO compared with NEMO simulation. We deliver the code and data to reproduce these results with our open-source platform CoastalTwin, including tools to extend our initial experiments. Jiang at pg. 2.
The “NEMO” simulation is a physics-based model that generates lower resolution data than the “FNO” (Fourier Neural Operator), which generates higher resolution data.
wherein the running of the multiscale model includes running a neural-network based model that generates the second resolution simulation data and
For implementing FNO and other ML-based surrogates with NEMO, we developed CoastalTwin, a modular and extensible platform to integrate simulations from physical models, such as NEMO, with ML models, to produce reliable, accelerated emulation of coastal dynamics. Jiang at pg. 3.
running a physics-based model that generates the first resolution simulation data,
NEMO was set up at 7-km regular grids in northwestern Europe, composing overall 520x292 grids. The atmospheric forcings of NEMO include mean sea level pressure (MSLP), U-direction wind speed (U10), and V-direction wind speed (V10) averaged on the top 10m above the sea surface from the downscaled product of ECMWF Reanalysis 5th Generation [16]. The bathymetry profile was from the General Bathymetric Chart of the Oceans product [17]. The simulation of two-dimensional (2D) sea surface height (SSH) was performed for all of 2020 at every 5min. Jiang at pg. 2.
wherein the neural-network based model receive low resolution information from the physics-based model at time t and the physics-based model receives high resolution information from the neural-network based model at time t+1 until the data generation ends,
Using CoastalTwin, we developed the surrogate models of NEMO to predict SSH at time tM based on both atmospheric forcings (i.e., U10, V10, and MSLP) at preceding times tN , ..., t0 and the bathymetry, where t0 is the present time, N ∈ I the history and M ∈ I the lookahead, and ∆t = 5min the FNO time step. Jiang at pg. 3.
wherein multiples of the neural-network based model that generates the second resolution simulation data are embedded at discrete locations of the physics-based model to simulate local processes at higher resolution;
Each FNO was developed by sequentially stacking a linear layer outputting 20 channels, 5 Fourier layers, and a final linear layer outputting 1 channel. Each Fourier layer contains 20 channels and a maximum of 40 frequency modes in both spatial dimensions, followed by a batch normalization and ReLU activation. Each UNet adopted three blocks of convolution in both contracting and expansive paths with the remaining architecture equivalent to [15]. We used the Adam optimization and a step-wise decreased learning scheduler with an initial rate 0.01, step size 20 epochs, and decay rate 0.1. We trained each model using MSE as the loss function over 50 epochs and batch size 32, on one Tesla A100 Graphics Processing Unit (GPU). We masked the land simulation in the loss to alleviate the adverse impact of land, where SSH is supposed to be zero. In addition to MSE as a performance metric, we computed the Structural Similarity Index (SSIM) [20] and the correlation (CORR) between times series of prediction and true at each grid point. Jiang at pg. 3.
Physics-informed ML methods integrate mathematical physics models with data-driven learning, namely with neural networks (NNs) [10]. A promising direction in spatiotemporal use-cases is neural operator learning: using NNs to learn mesh-independent, resolution-invariant solution operators for PDEs [18, 19]. To achieve this, Li et al. [14] use a Fourier layer that implements a Fourier transform, then a linear transform, and an inverse Fourier transform for a convolution-like operation in a NN. Jiang at pg. 3.
The FNOs are embedded in NEMO to simulate the partial differential equations, thus each FNO is embedded at a “discrete location” in the physics-based model.
creating a surrogate model using neural operators, wherein the surrogate model is trained using the first resolution simulation data and second resolution simulation data; and
Here, we propose the first “coastal digital twin”, an emulator built on state-of-art physics-informed ML techniques to produce computationally lightweight surrogate models that provide fast and accurate predictions of sea surface heights in coastal regions. As a proof-of-concept experiment, we developed a digital twin for the NEMO simulations in northwestern Europe using an improved version of FNO. Jiang at pg. 2.
Physics-informed ML methods integrate mathematical physics models with data-driven learning, namely with neural networks (NNs) [10]. A promising direction in spatiotemporal use-cases is neural operator learning: using NNs to learn mesh-independent, resolution-invariant solution operators for PDEs [18, 19]. To achieve this, Li et al. [14] use a Fourier layer that implements a Fourier transform, then a linear transform, and an inverse Fourier transform for a convolution-like operation in a NN.
generating an operational forecasting model using the surrogate model; and
For this reason, our coastal digital twin will be helpful for research and operation activities that require fast simulations, such as uncertainty quantification and real-time forecasting. Jiang at Abstract.
Jiang does not appear to disclose:
presenting a visual map of flood depth changing over time generated based on the operational forecasting model predicting a storm surge.
Seroka, which is analogous art, discloses:
presenting a visual map of flood depth changing over time generated based on the operational forecasting model predicting a storm surge.
NOAA/NOS’ model is, to our knowledge, the highest resolution operational global surge model available today. Coastline resolution is at least 1.5 km, up to 80 m globally. Seroka at Slide 4.
4 cycles a day: 00, 06, 12, and 18 UTC. Seroka at Slide 4.
Seroka is analogous art because both are directed to utilizing a multiscale model to generate a display indicating water levels from storm surges over time. It would have been obvious to a person having ordinary skill in the art, before the effective filing date of the claimed invention, to combine the Jiang (teaching a multiscale model for weather data) with Seroka to result in a system that provides a visual representation of the operational forecast model. Motivation to combine includes improving widespread information regarding storm dangers via a user-friendly interface, thus improving safety and reliability of model data.
Claim 2
Jiang discloses:
wherein the operational forecasting model functions as a partial differential equation solver that can work on a plurality of different multiscale modeling formulations.
Table 1 summarizes the experiment results. Our FNO approach outperforms UNet for all the four cases with respect to MSE and 1-SSIM. This illustrates that FNO can better capture the PDE-based simulations than the baseline model, particularly in this multivariate scenario. Jiang at pg. 4.
Claim 3
Jiang discloses:
wherein a partial differential equations family is learned over all parameters using the neural operators.
On the other hand, the recently proposed Fourier Neural Operator (FNO) [14] shows a promising alternative by learning the dynamics in the frequency domain. Jiang at pg. 2.
Claim 5
Jiang discloses:
wherein the surrogate model captures features at a resolution higher than the first resolution simulation data.
For implementing FNO and other ML-based surrogates with NEMO, we developed CoastalTwin, a modular and extensible platform to integrate simulations from physical models, such as NEMO, with ML models, to produce reliable, accelerated emulation of coastal dynamics. Jiang at pg. 3.
Claim 8
Claims 8 recites:
A computer program product comprising a computer readable storage medium having program instructions embodied therewith, the program instructions readable by a device to cause the device to:
Physics-based numerical models, such as Nucleus for European Modelling of the Ocean (NEMO) [4], have been developed to simulate and predict coastal and ocean dynamics. Jiang at pg. 1.
perform a method that is substantially the same as the method disclosed in claim 1.
Accordingly, for at least the same reasons and based on the same prior art as claim 1, claim 8 is rejected under 35 U.S.C. 103 for being obvious over Jiang in view of Seroka.
Claims 9-10 and 12
Claims 9-10 and 12 recite limitations that are substantially the same as the limitations recited in claims 2-3 and 5. Accordingly, for at least the same reasons and based on the same prior art as claims 2-3 and 5, claims 9-10 and 12 are rejected under 35 U.S.C. 103 for being obvious over Jiang in view of Seroka.
Claim 15
Claim 15 recites:
A system comprising: a processor; and a memory device coupled with the processor, the processor configured to at least:
Physics-based numerical models, such as Nucleus for European Modelling of the Ocean (NEMO) [4], have been developed to simulate and predict coastal and ocean dynamics. Jiang at pg. 1.
cause a device to perform a method substantially the same as the method recited in claim 1.
Accordingly, for at least the same reasons and based on the same prior art as claim 1, claim 15 is rejected under 35 U.S.C. 103 for being obvious over Jiang in view of Seroka.
Claims 16-17 and 19
Claims 16-17 and 19 recite limitations that are substantially the same as the limitations recited in claims 2-3 and 5. Accordingly, for at least the same reasons and based on the same prior art as claims 2-3 and 5, claims 16-17 and 19 are rejected under 35 U.S.C. 103 for being obvious over Jiang in view of Seroka.
Claim 6-7, 13-14,and 20 are rejected under 35 U.S.C. 103 for being obvious over Jiang in view of Seroka and Kida, et al., (U.S. Patent Pub. No. 2017/0145648), hereinafter “Kida.”
Claim 6
Jiang and Seroka do not appear to disclose:
further including combining the trained surrogate model with an impact model for risk assessment.
Kida, which is analogous art, discloses:
further including combining the trained surrogate model with an impact model for risk assessment.
The weather prediction unit 22 has a short-time rainfall prediction function (up to 1 hr) of predicting an imminent rainfall by analyzing data of the weather radar, and a long-time prediction function (1 hr to) of predicting a future weather state by inputting weather observation information to a weather model representing a weather phenomenon by an equation and performing large-scale calculations. Kida at [0044].
Predicting rainfall to determine flood possibilities in analogous to an “impact model for risk assessment.”
Kida is analogous art to the claimed invention because both are directed to forecast models. It would have been obvious to a person of ordinary skill in the art, before the effective filing date of the application, to combine Kida with Jiang and Seroka to predict impact of geophysical forces one or more entities. Motivation to combine includes early identification of risk to lives and property such that humans can take appropriate action, thus reducing potential damage.
Claim 7
Jiang discloses:
wherein the impact model includes coastal flood prediction model, the first resolution simulation data and the second resolution simulation data can include at least data associated sea surface height, and the method further includes
As a proof-of-concept study, we built Fourier Neural Operator (FNO) surrogates on the simulations of an industry-standard flood and ocean model (NEMO). The resulting FNO surrogate accurately predicts the sea surface height in most regions while achieving upwards of 45x acceleration of NEMO. Jiang at Abstract.
Jiang and Seroka do not appear to disclose:
triggering a physical barrier to open or close.
Kida discloses:
wherein the impact model includes
The runoff analysis unit 23 has a runoff analysis model M2, and predicts a flood region (point) based on the input future predicted rainfall, water level information, and flow rate information. Note that the runoff analysis model M2 itself, which predicts the flood region based on the predicted rainfall is a known technique, Kida at [0049].
triggering a physical barrier to open or close.
Opening/closing of the water stop gate A-12 is controlled by a control computer. Kida at [0026].
As shown in FIG. 3, the cloud computer 12 includes a platform 20, a weather observation unit 21, a weather prediction unit 22, and a runoff analysis unit 23. Kida at [0041].
When notified by the cloud computer 13 that the region managed by the control computer 13 may be flooded, the control computer 13 controls to close the water stop gate A-12 in the economy concentration area A-1. Kida at [0057].
It would have been obvious to a person of ordinary skill in the art, before the effective filing date of the application, to combine Kida with the other references to result in a coastal flood prediction model because the coastal water behavior modeled in Kida can be utilized to predict flooding by coastal waters using a surrogate model, as disclosed by Jiang. The prediction model can then trigger flooding mitigation measures, such as closing a barrier wall. Motivation to combine includes generating a system that automatically mitigates flooding without human intervention, thus not requiring a technician to visually verify flooding and risk personal harm.
Claims 13-14 and 20
Claims 13 and 20 recite limitations that are substantially the same as the limitations recited in claim 6. Claim 14 recites limitations that are substantially the same as the limitations recited in claim 7. Accordingly, for at least the same reasons and based on the same prior art as claims 6 and 7, claims 13-14 and 20 are rejected under 35 U.S.C. 103 for being obvious over Jiang in view of Seroka and Kida.
Claims 4, 11, and 18 are rejected under 35 U.S.C. 103 as being obvious over Jiang in view of Seroka and Campin, et al., (“Super-parametrization in Ocean Modeling: Application to Deep Convention”, hereinafter “Campin”).
Claim 4
Jiang and Seroka do not appear to disclose:
wherein an amount of data needed for neural operator training is reduced by using super-parametrization.
Campin, which is analogous art to the claimed invention, discloses:
wherein an amount of data needed for neural operator training is reduced by using super-parametrization.
We explore the efficacy of “super parameterization” (SP) in ocean modeling in which local 2-d non-hydrostatic plume-resolving fine-grained (FG) models are embedded at each vertical column of a coarse-grained (CG) hydrostatic model...coupling the two models together. The SP model is found to be greatly superior to HYD at much less computational cost than the fully non-hydrostatic calculation. Campin at Abstract.
As illustrated in FIG. 1, the Fine Grid Model generates a Course Grid Model that has fewer data points than the fully Resolved model, thus reducing the size of the training data. See (https://www.sciencedirect.com/science/article/pii/S1463500310001496 or other source for color images to make the difference clearer) .
PNG
media_image1.png
506
667
media_image1.png
Greyscale
Campin is analogous art to the claimed invention because both are directed to multi-scale modeling. It would have been obvious to a person having ordinary skill in the art, before the effective filing date of the claimed invention, to combine the disclosure of Jiang, including the neural operators disclosed therein, with the super-parameterization of Campin to result in a system that trains neural operators in a more efficient manner. Motivation to do so includes reduction in training time and resources to generate the system by training via super-parameterization.
Claim 11 and 18
Claims 11 and 18 recite limitations that are substantially the same as the limitations recited in claim 4. Accordingly, for at least the same reasons and based on the same prior art as claim 4, claims 11 and 18 are rejected under 35 U.S.C. 103 for being obvious over Jiang in view of Seroka and Campin.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure:
Denli, et al. (U.S. Patent Pub. No. 2020/0182047): “Automated Reservoir Modeling Using Deep Generative Networks”
Raissi, et al., “Physics-informed neural networks: A deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations”
E, et al., “Integrating Machine Learning with Physics-Based Modeling”
Wang, et al., “Physics-Guided Deep Learning for Dynamical Systems: A Survey”
Stuckner, et al., “Tractable Multiscale Modeling with An Embedded Microscale Surrogate.”
Communication
Any inquiry concerning this communication or earlier communications from the examiner should be directed to JOSEPH MORRIS whose telephone number is (703)756-5735. The examiner can normally be reached M-F 8:30-5:00.
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, Ryan Pitaro can be reached at (571) 272-4071. 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.
JOSEPH MORRIS
Examiner
Art Unit 2188
/JOSEPH P MORRIS/Examiner, Art Unit 2188
/RYAN F PITARO/Supervisory Patent Examiner, Art Unit 2188