DETAILED ACTION
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Continued Examination Under 37 CFR 1.114
A request for continued examination under 37 CFR 1.114, including the fee set forth in 37 CFR 1.17(e), was filed in this application after final rejection. Since this application is eligible for continued examination under 37 CFR 1.114, and the fee set forth in 37 CFR 1.17(e) has been timely paid, the finality of the previous Office action has been withdrawn pursuant to 37 CFR 1.114. Applicant's submission filed on 04/10/2026 has been entered.
Response to Amendment
This Office Action is in response to communication filed on 04/10/2026, wherein Claims 1-18 and 20-21 are pending, Claims 1 and 8 are amended, Claim 19 is cancelled, and Claims 13-18 and 20 are withdrawn. Claim 21 is new.
Response to Arguments
Regarding 35 USC 101 rejection: Applicant’s arguments with respect to claims 1-12 filed on 04/10/2026, were fully considered but found not persuasive.
In pages 1 and 2, Applicant states:
“Even assuming that the claims could be construed as being drawn towards an abstract
idea or mental process, the additional elements of the claims do, in fact, integrate an abstract idea into a practical application, and also impose meaningful limits to the claims. Transforming the prediction via mathematically parametrized health domain transfer functions on the computing system, estimating remaining useful life of the engineering system based on the transformed prediction data, and initiating a remedial action based on the estimated remaining useful life are clearly practical applications”.
Examiner respectfully disagrees.
Applicant’s claims do not specify the parameters used (can be any), what the engineering system is (can be any), what the remedial action is (can be any). Claim 1 is the example of field of use attached to the algorithm and elements/steps recited at a high level of generality, and the algorithm itself is not specified.
There is no technological advancement in the claim, no improvement for the process, only routine and well-known operations steps. Inputting the data into a machine learning model and initiating a remedial action does not show the advancement in technology, it is the standard and routine operation.
Regarding new Claim 21: scheduling maintenance is organizing human activity, and replacing and/or retiring a component are generic operations, well-known routing and conventional specified at a high level of generality.
Practical application must be shown using meaningful additional elements. The additional elements, cited by the Applicant, are recited in generality and do not recite particular machines applying or being used by the abstract idea (see MPEP 2106.05, specifically about the particular machine: see part I, The particularity or generality of the elements of the machine or apparatus; Part II, Whether the machine or apparatus implements the steps of the method, and Part III, Whether its involvement is extra-solution activity of a field-of use). Prior art cited in the rejection, shows claimed additional elements as well-known in the art, routine and conventional.
Regarding the Allowable Subject Matter: The Allowable Subject Matter is revoked. The prior art previously cited by the applicant, was found relevant in combinations with already cited prior art.
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 1-12 and 21 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more. The claims recite an abstract idea as discussed below. This abstract idea is not integrated into a practical application for the reasons discussed below. The claims do not include additional elements that are sufficient to amount to significantly more than the judicial exception for the reasons discussed below.
Step 1 of the 2019 Guidance requires the examiner to determine if the claims are to one of the statutory categories of invention. Applied to the present application, the claims belong to one of the statutory classes of a process. The below claim is considered to be a statutory category (process).
Step 2A of the 2019 Guidance is divided into two Prongs. Prong 1 requires the examiner to determine if the claims recite an abstract idea, and further requires that the abstract idea belongs to one of three enumerated groupings: mathematical concepts, mental processes, and certain methods of organizing human activity.
Independent Claim 1 is copied below, with the limitations belonging to an abstract idea highlighted in bold; the remaining limitations are ‘’additional elements’’.
A method, comprising:
receiving condition-monitoring data of an engineering system at a computing system;
inputting the condition-monitoring data to a hybrid model that includes a machine learning model empowered with physics-informed transfer functions on the computing system, the machine learning model outputting a prediction of health variables of the engineering system as intermediate variables;
transforming the prediction via mathematically parametrized health domain transfer functions on the computing system to create transformed prediction data;
estimating remaining useful life of the engineering system based on the transformed prediction data
initiating a remedial action on the engineering system based on the estimated remaining useful life.
Under the Step 2A, Prong One, we consider whether the claim recites a judicial exception (abstract idea). In the above claim, the highlighted portion constitutes an abstract idea because, under a broadest reasonable interpretation in light of the specification, it recites limitations that fall into abstract idea exceptions. Specifically, under the 2019 Revised Patent Subject Matter Eligibility Guidance, it falls into the grouping of subject matter that when recited as such in a claim limitation covers mathematical processes (mathematical relationships, mathematical formulas or equations, mathematical calculations).
Steps of “inputting the condition-monitoring data to a hybrid model that includes a machine learning model empowered with physics-informed transfer functions on the computing system, the machine learning model outputting a prediction of health variables of the engineering system as intermediate variables”; “transforming the prediction via mathematically parametrized health domain transfer functions on the computing system to create transformed prediction data”; and “estimating remaining useful life of the engineering system based on the transformed prediction data” are treated by the Examiner as belonging to mathematical process grouping.
Limitations of “receiving condition-monitoring data of an engineering system at a computing system”, “inputting the condition-monitoring data to …” and “initiating a remedial action on the engineering system based on the estimated remaining useful life” are treated as an extra solution activity recited in generality (e.g., mere data gathering, inputting data into a model). The “computing system” is a generic computer used to facilitate the application of the judicial exception. The preamble of Claim 1: “A method, comprising” is a generically recited preamble.
Similar limitations comprise the abstract ideas of Claims 2-11.
Prong 2 of Step 2A of the 2019 Guidance requires the examiner to determine if the claims recite additional elements or a combination of additional elements which integrate the abstract idea into a practical application. This requires additional elements in the claim to apply, rely on, or use the abstract idea in a manner that imposes a meaningful limit on the abstract idea, such that the claim is more than a drafting effort designed to monopolize the abstract idea.
In this step, we evaluate whether the claim recites additional elements that integrate the exception into a practical application of that exception.
Additional elements: “engineering system”, “computing system”, “receiving condition-monitoring data of an engineering system at a computing system”, and “inputting the condition-monitoring data to …” add extra-solution activities (i.e., mere data gathering/inputting, source/type of data to be manipulated) using elements recited at a high level of generality (see MPEP 2106.05(g)); generally link the use of the judicial exception to a particular technological environment or field of use (see MPEP 2106.05(h)); add the words “apply it” (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely use a computer as a tool to perform an abstract idea (see MPEP 2106.05(f)); and append steps at a high level of generality such that substantially all practical applications of the judicial exception(s) are covered (see MPEP 2106.05(c)).
Regarding Claim 12: the limitation of “comprising a hardware processor and non-transitory memory that are operable vis instructions to perform the method of Claim 1” recites the additional elements of “a hardware processor” and “non-transitory memory”, which are recited at a high level of generality (see MPEP 2106.05(g)); generally link the use of the judicial exception to a particular technological environment or field of use (see MPEP 2106.05(h)).
Various considerations are used to determine whether the additional elements are sufficient to integrate the abstract idea into a practical application. In this particular case, the claim does not recite a particular machine applying or being used by the abstract idea. The claim does not effect a real-world transformation or reduction of any particular article to a different state or thing. (Manipulating data from one form to another or obtaining a mathematical answer using input data does not qualify as a transformation in the sense of Prong 2.) The claim does not contain additional elements which describe the functioning of a computer, or which describe a particular technology or technical field, being improved by the use of the abstract idea. (This is understood in the sense of the claimed invention from Diamond v Diehr, in which the claim as a whole recited a complete rubber-curing process including a rubber-molding press, a timer, a temperature sensor adjacent the mold cavity, and the steps of closing and opening the press, in which the recited use of a mathematical calculation served to improve that particular technology by providing a better estimate of the time when curing was complete. Here, the claim does not recite carrying out any comparable particular technological process).
Instead the additional elements in the claim appear to be merely insignificant extra-solution activity – merely receiving and manipulating data, steps recited at a high level of generality such that substantially all practical applications of the judicial exception(s) are covered, generic computer components used to facilitate the application of the judicial exception and/or a field of use.
Therefore, the claim is directed to a judicial exception and require further analysis under the Step 2B.
Step 2B of the 2019 Guidance requires the examiner to determine whether the additional elements cause the claim to amount to significantly more than the abstract idea itself. The considerations for this particular claim are essentially the same as the considerations for Prong 2 of Step 2A, and the same analysis leads to the conclusion that the claim does not amount to significantly more than the abstract idea.
Essentially, the above claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception (Step 2B analysis) because they are well-understood and conventional in the relevant art of US20220004182 to Natsumeda et al. (hereinafter Natsumeda) and US20080140352 to Goebel et al. (hereinafter Goebel).
Therefore, claim 1 is rejected under 35 U.S.C. 101 as directed to an abstract idea without significantly more and is not patent eligible.
Dependent claims 2-12 and 21 merely add limitations which further detail the abstract idea, namely further mathematical steps detailing how the data processing algorithm is implemented, i.e. additional limitations corresponding to mathematical relationship grouping. These limitations do not help to integrate the claim into a practical application or make it significantly more than the abstract idea (which is recited in slightly more detail, but not in enough detail to be considered to narrow the claim to a particular practical application itself).
The dependent claims are, therefore, also ineligible.
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 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 1-10, 12, and 21 are rejected under 35 U.S.C. 103 as being unpatentable over US20220004182 to Natsumeda et al. (hereinafter Natsumeda) in view of NPL “Remaining useful life prediction of induction motors using nonlinear degradation of health index” to Yang et al. (hereinafter Yang).
Regarding Claim 1: Natsumeda discloses:
“A method, comprising: receiving condition-monitoring data of an engineering system at a computing system” (para 0023 – “Predictive Maintenance can be used to monitor an equipment's and/or component's condition and identify maintenance program(s)… Data-driven approaches can utilize sensor and operational data to estimate RUL, where the data can be collected as a time series. This approach can involve large amounts of data that may be generated by monitoring many equipment and performance characteristics”; para 0006 – “a computer system is provided for determining a remaining useful life (RUL) of a system.”);
“the machine learning model outputting a prediction of health variables of the engineering system as intermediate variables” (para 0021 – “a neural network can be trained to predict the RUL of a monitored system given partial temporal observation samples of a time-series sequence.”; para 0046 – “where H is a variable holding a value of HI (i.e. health index, added by examiner), and ‘min’ returns the minimum value in the given list of arguments”; para 0056 – “At inference, the trained neural network model 170 can receive a multivariate time series x 601 as the input for RUL estimation, then the trained neural network model outputs H, which is HI at the end of the time series data x.”; see also paras 0021, 0022, 0029, and para 0078); and
“estimating remaining useful life of the engineering system based on the transformed prediction data, the remaining useful life used to perform a remedial action on the engineering system” (para 0029 – “the time series data 140 can be used by a neural network training system 160 to train a neural network model 170, such that the neural network model can predict the remaining useful life of the system 110, and a health index (HI) for the system 110, for scheduling predictive maintenance (i.e. remedial action, added by examiner). ”).
Natsumeda does not specifically disclose:
“inputting the condition-monitoring data to a hybrid model that includes a machine learning model empowered with physics-informed transfer functions on the computing system; transforming the prediction via mathematically parametrized health domain transfer functions on the computing system to create transformed prediction data; and initiating a remedial action on the engineering system based on the estimated remaining useful life”.
However, Yang discloses:
“inputting the condition-monitoring data to a hybrid model that includes a machine learning model empowered with physics-informed transfer functions on the computing system” (page 3 – “The developed prognostics framework for RUL prediction is illustrated in Fig. 1, where four modules are included, namely, data & pre-processing, feature engineering, two-stage modelling and the final performance evaluation… In data & pre-processing, the raw signals are time-series data (i.e. condition-monitoring data, added by examiner) collected from a Degradation Process (DP) with sensors”; page 4 – Fig.1; “In two-stage modelling, the relationship between input and output, i.e. feature vector (or instance) to RUL, is built. As shown in Fig. 1, this is done through HI modelling (to first predict HI values and then smooth them) and RUL modelling (to map from predicted HI to RUL) (i.e. hybrid modelling, added by examiner). In HI modelling, different types of HI degradations can be assumed and adopted (HI being the physics-informed transfer function (see Specification para 0041), added by examiner ), making this prognostics framework generic and adaptable to various applications with specific underline degradation patterns”);
“transforming the prediction via mathematically parametrized health domain transfer functions on the computing system to create transformed prediction data” (page 4 – “The performance evaluation module is included in the framework primarily for assessing the performance of the RUL prediction process (i.e. from raw signals to RUL). Here, the validation method refers to how the data is split for training and testing, e.g. using k-fold cross-validation or leave-one-out validation. For the metrics, there are many different types of formulations that can be used for evaluating RUL predictions, such as Root Mean Square Error (RMSE) and Confidence Interval (CI) (i.e. mathematically parametrized health domain transfer function (i.e. HI value), added by examiner)… In the realization depicted in Fig. 2, with n sets of DP training data, n individual HI prediction models are first trained. Then, in the prediction stage, the testing data is input to each of the n trained HI prediction models to obtain the corresponding
predicted HI values. Upon applying the same HI dynamic smoothing procedure, the n smoothed HI values are then mapped to the corresponding RUL values, from which the final RUL prediction is produced by ensembling them (”); and
“initiating a remedial action on the engineering system based on the estimated remaining useful life” (page 7 – Table 1; “Gradualness _ Description: The health of a motor will decrease gradually and will not experience a sudden big drop._ Implementation: For nonlinear HI degradation, this is realized by restricting the decreasing speed of HI. In fulfilling this rule, a
maximum decreasing speed is set. (i.e. initiating a remedial action, added by examiner)”).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the method, disclosed Natsumeda, as taught by Yang, in order to estimate the remaining useful life of the system with higher accuracy utilizing the physics-informed hybrid machine-learning model.
Regarding Claim 2: Natsumeda/Yang combination discloses the method of Claim 1.
Natsumeda does not specifically disclose:
“wherein the intermediate variables of the hybrid model comprise a health indicator HI and a degradation progression rate c”.
However, Yang discloses:
“wherein the intermediate variables of the hybrid model comprise a health indicator HI and a degradation progression rate c” (page 8, Fig. 6 – “Fig. 6 illustrates with an example the advantage of the nonlinear HI dynamic smoothing, where HI prediction curves of originally predicted (see ‘Original’) and dynamically smoothed (see ‘Smoothed’) HI values are compared with the true values (see ‘True’). HI values from the intermediate variable, HI dynamic, are also included in the figure (see ‘Dynamic’)”; page 5 – “the multi-model ensemble is simple and easy to implement and it makes the realization generic for any number of DP training data, whilst keeping into consideration each training data’s individual (different) degradation pattern (i.e. degradation progression rate, added by examiner)”).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the method, disclosed by Natsumeda/Yang combination, as taught by Yang, in order to estimate the remaining useful life of the system with higher accuracy utilizing the physics-informed hybrid machine-learning model.
Regarding Claim 3: Natsumeda/Yang combination discloses the method of Claim 2.
Natsumeda does not specifically disclose:
“the mathematically parametrized health domain transfer functions comprise an exponential function”.
However, Yang discloses:
“the mathematically parametrized health domain transfer functions comprise an exponential function” (page 5 – “In this paper, Paris law, or specifically the exponential degradation form from (1), will be employed as the case implementation. As shown and will be discussed in (6) and (7), the exponential HI degradation will result in a simple analytical mapping from HI to the RUL”).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the method, disclosed by Natsumeda/Yang combination, as taught by Yang, in order to estimate the remaining useful life of the system with higher accuracy utilizing the physics-informed hybrid machine-learning model.
Regarding Claim 4: Natsumeda/Yang combination discloses the method of Claim 3.
Natsumeda does not specifically disclose:
“imposing a constraint that the degradation progression rate c is larger than 1 to impose a convex degradation progression in a normalized health domain where health index varies in a range with upper and lower bounds of 1 and 0, respectively”.
However, Yang discloses:
“imposing a constraint that the degradation progression rate c is larger than 1 to impose a convex degradation progression in a normalized health domain where health index varies in a range with upper and lower bounds of 1 and 0, respectively” (page 6 - “Upon calculating the first and second order derivatives of HIðtÞ, it can be discovered that HIðtÞ is convex when c > HIbegin (i.e. c > 1:0), and it is concave when c < HIend (i.e. c < 0:1). In illustrating and validating this, Fig. 4 plots the HI curves with variations in the parameter c for both convex and concave degradations (here for convex: c changes from 1.00001 to 1000, and for convex: c changes from 0.099999 to 0.0001). It can also be seen from the convex curves that when c!1, the degradation approaches to linear HI.”).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the method, disclosed by Natsumeda/Yang combination, as taught by Yang, in order to estimate the remaining useful life of the system with higher accuracy utilizing the physics-informed hybrid machine-learning model.
Regarding Claim 5: Natsumeda/Yang combination discloses the method of Claim 2.
Natsumeda does not specifically disclose:
“applying a first transfer function that transfers the health indicator HI and the degradation progression rate c to a life domain using a physics-based relationship to obtain a first estimate of the remaining useful life; applying a second transfer function that transfers the health indicator HI and the degradation progression rate c to the life domain using a linear relationship to obtain a second estimate of the remaining useful life; and combining the first estimate and the second estimate using a weight factor to obtain a combined estimate of the remaining useful life”.
However, Yang discloses:
“applying a first transfer function that transfers the health indicator HI and the degradation progression rate c to a life domain using a physics-based relationship to obtain a first estimate of the remaining useful life; applying a second transfer function that transfers the health indicator HI and the degradation progression rate c to the life domain using a linear relationship to obtain a second estimate of the remaining useful life; and combining the first estimate and the second estimate using a weight factor to obtain a combined estimate of the remaining useful life” (page 10 – “one linear and one nonlinear regression algorithms were employed and investigated. For linear regression (i.e. first transfer function, added by examiner), the Linear SVR with libsvm implementation and the parameter C = 1 was used [60]. For nonlinear regression (i.e. second transfer function, added by examiner), the feedforward Neural Networks (NN) was utilized with one layer of 10 hidden neurons and the Levenberg–Marquardt optimization for updating the weights and bias states. These two regression algorithms were selected because of their popularity and general effectiveness. With the predicted HI values at each cycle, the proposed HI dynamic smoothing Algorithm 1 was then used to generate a smoothed HI. In the specific case of exponential HI degradation, the analytical way derived in Section 4.2 was used to calculate RUL from HI. The multi-model ensemble was also employed to obtain the final predicted RUL when multiple training data were available.”)
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the method, disclosed by Natsumeda/Yang combination, as taught by Yang, in order to estimate the remaining useful life of the system with higher accuracy utilizing the physics-informed hybrid machine-learning model.
Regarding Claim 6: Natsumeda/Yang combination discloses the method of Claim 1.
Natsumeda does not specifically disclose:
“further comprising: measuring a health stage indicator based on the condition-monitoring data to monitor the engineering system during a quasi-linear degradation stage; and using the health stage indicator, estimating a transition of the engineering system from the quasi-linear degradation stage to an accelerated degradation phase, wherein the remaining useful life is determined during the accelerated degradation phase”.
However, Yang discloses:
“further comprising: measuring a health stage indicator based on the condition-monitoring data to monitor the engineering system during a quasi-linear degradation stage; and using the health stage indicator, estimating a transition of the engineering system from the quasi-linear degradation stage to an accelerated degradation phase, wherein the remaining useful life is determined during the accelerated degradation phase” (Fig,3; page 5 – “Although originally developed for describing materials crack growth, the Paris law and its derivatives have been adapted to other applications such as modelling the accelerated degradation of electronic devices [54] and the degradation of fuel cells [47]. In this paper, Paris law, or specifically the exponential degradation form from (1), will be employed as the case implementation. As shown and will be discussed in (6) and (7), the exponential HI degradation will result in a simple analytical mapping from HI to the RUL.”).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the method, disclosed by Natsumeda/Yang combination, as taught by Yang, in order to estimate the remaining useful life of the system with higher accuracy utilizing the physics-informed hybrid machine-learning model.
Regarding Claim 7: Natsumeda/Yang combination discloses the method of Claim 6.
Natsumeda does not explicitly disclose:
“wherein determining the remaining useful life comprises extracting features from the condition-monitoring data and inputting the features to the hybrid model, the features comprising any combination of time domain features, frequency domain features, and time-frequency domain features”.
However, Yang discloses:
“wherein determining the remaining useful life comprises extracting features from the condition-monitoring data and inputting the features to the hybrid model, the features comprising any combination of time domain features, frequency domain features, and time-frequency domain features” (page 9 – “Using segmentation, feature extraction was conducted on every 0.5 s of signals in producing a set of new features. In these studies, 3 feature extraction techniques were investigated, namely, time domain using statistics [56], frequency domain using Fourier Transform and time–frequency domain based on Wavelet Transform [57]. In the wavelet transform for feature extraction, wavelet=’db4’ and level = 3 were configured”; page 11 – “Table 4 shows the best RMSE performances in predicting the RULs of each motor using NN, with time domain, frequency domain and time–frequency domain features. Other than the results generated with nonlinear HI degradation (convex shape of exponential HI, see ‘Nonlinear’), the RMSE values with linear HI degradation (see ‘Linear’) are also included” ).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the method, disclosed by Natsumeda/Yang combination, as taught by Yang, in order to estimate the remaining useful life of the system with higher accuracy utilizing the physics-informed hybrid machine-learning model.
Regarding Claim 8: Natsumeda/Yang combination discloses the method of Claim 6.
Natsumeda further discloses:
“estimating a first transition of the engineering system from a healthy stage to a quasi- linear degradation stage; and estimating a second transition of the engineering system from the quasi-linear degradation stage to an accelerated degradation phase, wherein the remaining useful life estimation starts when the accelerated degradation phase is determined” (Fig. 4; para 0044 – “RUL linearly decreases over time (i.e. part 410 of Fig. 4, quasi-linear degradation stage, Fig. 4 shows the first transition from normal stage, which is not shown in the drawing, to the quasi-linear stage 410, added by examiner), but the health index (HI) does not always due to nonlinearity in degradation process. It can be severe especially when the degradation process has multi-stages. The whole lifetime of a system can be divided into two or more different health stages. In various embodiments, a healthy stage 410 can be followed by degradation stage 420 (i.e. the transition from 410 stage to 420 stage is the second transition, added by examiner). At the beginning of use, for example, of a piece of equipment, the degradation in the system can be negligible and the health index (HI) does not change appreciably or at all. However, on approaching an end of life, the degradation can become severe and the HI starts to change noticeably. A point in time where the monitored system changes from a healthy stage to a degrading stage (i.e. RUL starts when the accelerated degradation phase is determined, added by examiner) may be determined.”).
Regarding Claim 9: Natsumeda/Yang combination discloses the method of Claim 8.
Natsumeda further discloses:
“wherein a first health stage indicator is used to monitor the engineering system during the healthy stage, and where a different, second health stage indicator is used to monitor the engineering system during the accelerated degradation phase” (para 0049 – “HI refers to health index (i.e. first health stage indicator, added by examiner), and HIch refers to health index change (i.e. second health stage indicator, added by examiner). The Dual-estimator can use all components at training and can use Tss2vec 550 and Vec2HI 560 for inference at run time”).
Regarding Claim 10: Natsumeda/Yang combination discloses the method of Claim 1.
Natsumeda does not explicitly disclose:
“wherein the machine learning model comprises a feedforward, deep neural network”.
However, Yang discloses:
“wherein the machine learning model comprises a feedforward, deep neural network” (page 10 – “For nonlinear regression, the feedforward Neural Networks (NN) was utilized with one layer of 10 hidden neurons and the Levenberg–Marquardt optimization for updating the weights and bias states”).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the method, disclosed by Natsumeda/Yang combination, as taught by Yang, in order to estimate the remaining useful life of the system with higher accuracy utilizing the physics-informed hybrid machine-learning model.
Regarding Claim 12: Natsumeda/Yang combination discloses the computing system as set forth in Claim 1.
Natsumeda further discloses:
“comprising a hardware processor and non-transitory memory that are operable via instructions to perform the method of claim 1” (para 0042 – “the neural network system 300 can include one or more processor(s) (e.g., CPUs, GPUs), one or more types of memory (e.g., DRAM, SRAM, FLASH, magnetic, optical, etc.), communications bus(es) and controllers (e.g., memory management unit), and input/output devices (e.g., display(s), keyboard, mouse, etc.).”).
Regarding Claim 21: Natsumeda/Yang combination discloses the method of Claim 1.
Natsumeda further discloses:
“wherein initiating a remedial action comprises one or more of scheduling maintenance, replacing a component, and retiring a component” (para 0029 – “the neural network model can predict the remaining useful life of the system 110, and a health index (HI) for the system 110, for scheduling predictive maintenance. An operator can use the predicted RUL and scheduled predictive maintenance to maintain the monitored system”).
Claim 11 is rejected under 35 U.S.C. 103 as being unpatentable over Natsumeda in view of Yang and in further view of NPL “Fusing physics-based and deep learning models for prognostics” to Chao et al. (hereinafter Chao).
Regarding Claim 11: Natsumeda/Yang combination discloses the method of Claim 10.
Natsumeda does not explicitly disclose:
“wherein hyperparameters of the feedforward, deep neural network are determined using a grid search”.
However, Chao discloses:
“wherein hyperparameters of the feedforward, deep neural network are determined using a grid search” (page 8 – “Feed-forward neural network (FNN). The network architecture comprises six fully connected layers. The first four hidden layers have 100 neurons each. The last hidden layer has 50 neurons. A single linear neuron was used in the output layer. ReLu activation function was used throughout the entire network. The network has 39k trainable parameters. Motivated by the main study, this final architecture is the result of conducting a grid search with the following search space: number of hidden layers [5-6], number of neurons at each hidden layer [50, 100], and activation function type [tanh, ReLu].”).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the method, disclosed Natsumeda/Yang combination, as taught by Chao, in order to estimate the remaining useful life of the system with higher accuracy utilizing the physics-informed hybrid machine-learning model.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to Lyudmila Zaykova-Feldman whose telephone number is (469)295-9269. The examiner can normally be reached 7:30am - 4:30pm, Monday through Friday.
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, Arleen M. Vazquez can be reached on 571-272-2619. 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.
/LYUDMILA ZAYKOVA-FELDMAN/Examiner, Art Unit 2857
/LINA CORDERO/Primary Examiner, Art Unit 2857
Prosecution Insights