DETAILED ACTION
This action is responsive to the Application/amendment filed on 01/26/2026. Claims 1-20 are pending in the case. Claims 1 and 16 are independent claims. Claim 1 and 16 are amended.
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 .
Response to Arguments
Applicant's arguments filed 01/26/2026 have been fully considered but they are not persuasive.
With respect to the 101 rejection:
Applicant argues the claim goes beyond retrieving and combing data using a computer, specifically citing the amended limitations. Applicant notes that the claimed pre-processing and classification can not be performed in the mind and ties the recited limitation to a processors ability to generate a complex model. Further, Applicant notes that these limitations add significantly more and compares the claims to Example 3 of the USPTO guidance.
Examiner disagrees.
Firstly, preprocessing and classification of operations as claimed recites an abstract idea alone, and as such can not itself be indicative of a practical application nor significantly more. As noted in the rejection the human mind is capable of identifying quality issues in data as well as making classifications about said data or operations, these limitations do not invoke any particular processor functions. The rejection does not make the suggest that the claim merely recites retrieving and combing data using a computer. Rather, the claims recite several abstract idea limitations, and no additional elements providing a practical application nor significantly more than the recited exception. The additional elements provided, at least in claim 1, do nothing more than describe collecting/obtaining data, and reciting a generic computer, and/or computer process, for performing the recited abstract ideas.
Unlike Example 3, which describes particular limitations such as “converting the binary image array into a halftoned image” which can not be considered abstract ideas and further ties the recited math to the processors ability to process digital images, the instant claim does not describe any such particular limitations but for using the computer technology to perform the recited abstract ideas. Using a computer technology to perform and abstract idea is outlined expressly in MPEP 2106.05(f), as noted in the rejection, as not indicative of a practical application nor significantly more.
Therefore, the rejection is maintained.
With respect to the 102 rejection:
Applicant argues Mohammed fails to disclose the amended limitation.
Examiner notes that Mohammed does not teach the entirety of the amended claim. As noted by Applicant, Mohammed does not teach the pre-processing limitation. The updated 103 rejection in view of Mohammed/Wu addresses these deficiencies.
With respect to the 103 rejection:
Applicant notes the remaining cited art does not address the deficiencies already pointed out in Mohammed. Applicant does not provide any particular additional substantive arguments about the teaching of the cited art.
Examiner notes that the amendments necessitated an updated rejection in view of Mohammed/Wu, the updated rejection is provided below.
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 claims are directed to an abstract idea without significantly more.
Regarding Claim 1
Under step 1, the claim is directed to one or more processors…for constructing a machine learning model correlating fatigue crack growth with operational data, which is directed to a machine, one of the statutory categories.
Under Step 2A Prong 1, the claim recites the following limitations which are considered mental evaluations “ predicting fatigue crack growth for the rotatable structure …, wherein the machine learning model is constructed by…classifying the historical data and historical operational data based on at least two regions of fatigue crack size…to predict fatigue crack growth based on the historical data, the operational data, and the classifications of the historical data based on the at least two regions of fatigue crack size…wherein training the machine learning model comprises: pre-processing the historical operational data to identify quality issue within the historical operational data and classifying operations of the historical operational data based on a growth region associated with the fatigue crack growth.”
Predictions and classification based on data is a process performed in the human mind. Identification of quality issues as well as operation classification is an activity which can be performed in the mind as well. For example, one may evaluate a set of data and based on attributes of the data form a judgement which is a prediction or classification or quality identification of the data.
Step 2A Prong Two Analysis: The judicial exception in not integrated into a practical application. In particular, the claims recite the additional element(s) the limitations “and one or more memory devices, the one or more memory devices storing computer-readable instructions that when executed by the one or more processors cause the one or more processors to perform operations…using a machine learning model…training the machine learning model” amounts to mere instructions to apply a computer technology to an abstract idea. No details about the technical functioning of the machine learning model or its training are described such that these limitations are any more that instructions to apply a computer technology to the performance of the recited abstract ideas. see MPEP 2106.05(f) consideration (2). In addition, the claim recites additional element(s) “obtaining data indicative of fatigue crack size and operational data for a rotatable structure…obtaining historical data indicative of fatigue crack size for a plurality of rotatable structures” that amounts to adding insignificant extra-solution activity to the judicial exception, in particular these are mere data gathering limitations, See MPEP 2106.05(g). Accordingly, the additional elements do not integrate the abstract idea into a practical application because they do not impose any meaningful limits on practicing the abstract idea. The claim is directed to an abstract idea.
Step 2B Analysis: The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception. Further, the additional elements, “obtaining data indicative of fatigue crack size and historical operational data for a rotatable structure…obtaining historical data indicative of fatigue crack size for a plurality of rotatable structures” are insignificant extra-solution activities that are considered well-understood, routine, conventional activities, for the following reasons. Examiner notes that these limitations amount to receiving or transmitting data over a network (MPEP 2106.05(d)(II)(i). According to MPEP 2106.05(d)(II)(i), “The courts have recognized the following computer functions as well‐understood, routine, and conventional functions when they are claimed in a merely generic manner”. As such, the insignificant extra-solution activities are considered well-understood, routine, conventional activities. Therefore, the claim is not patent eligible.
Regarding Claim 2-15
The claim is directed to a machine. Each of the limitations described in these claims, under Step 2A Prong 1, only serve to describe the abstract idea addressed in the independent claim. In particular, these limitations at most describe additional contextual details about the data which is evaluated by the recited abstract idea and/or subsequent determinations and classifications which are similarly evaluations capable of being performed in the mind.
Furthermore, under step 2A Prong 2 and 2B, the claim(s) do not recite additional elements to consider other than those considered in the independent claim. Accordingly, the recited additional elements do not integrate the abstract idea into a practical application because they do not impose any meaningful limits on practicing the abstract idea, nor do they amount to significantly more than the judicial exception because they do not impose any meaningful limits on practicing the abstract idea.
Regarding Claim 16
Under step 1, the claim is directed to A method of constructing a machine learning model, the method comprising, which is directed to a process, one of the statutory categories.
Under Step 2A Prong 1, the claim recites the following limitations which are considered mental evaluations “classifying the historical data and the historical operation data based on at least two regions of fatigue crack size… to predict fatigue crack growth based on the historical data, the historical operational data, and the classifications of the historical data based on the at least two regions of fatigue crack size, wherein training the machine learning model comprises: pre-processing the historical operational data to identify quality issue within the historical operational data and classifying operations of the historical operational data based on a growth region associated with the fatigue crack growth”
Predictions and classification based on data is a process performed in the human mind. For example, one may evaluate a set of data and based on attributes of the data form a judgement which is a prediction or classification of the data.
Step 2A Prong Two Analysis: The judicial exception in not integrated into a practical application. In particular, the claims recite the additional element(s) the limitations “training the machine learning model” amounts to mere instructions to apply a computer technology to an abstract idea. No details about the technical functioning of the machine learning model or its training are described such that these limitations are any more that instructions to apply a computer technology to the performance of the recited abstract ideas. see MPEP 2106.05(f) consideration (2). In addition, the claim recites additional element(s) “obtaining historical data indicative of fatigue crack size and historical operational data for a plurality of rotatable structures” that amounts to adding insignificant extra-solution activity to the judicial exception, in particular these are mere data gathering limitations, See MPEP 2106.05(g). Accordingly, the additional elements do not integrate the abstract idea into a practical application because they do not impose any meaningful limits on practicing the abstract idea. The claim is directed to an abstract idea.
Step 2B Analysis: The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception. Further, the additional elements, “obtaining historical data indicative of fatigue crack size and historical operational data for a plurality of rotatable structures” are insignificant extra-solution activities that are considered well-understood, routine, conventional activities, for the following reasons. Examiner notes that these limitations amount to receiving or transmitting data over a network (MPEP 2106.05(d)(II)(i). According to MPEP 2106.05(d)(II)(i), “The courts have recognized the following computer functions as well‐understood, routine, and conventional functions when they are claimed in a merely generic manner”. As such, the insignificant extra-solution activities are considered well-understood, routine, conventional activities. Therefore, the claim is not patent eligible.
Regarding Claim 17-20
The claim is directed to a process. Each of the limitations described in these claims, under Step 2A Prong 1, only serve to describe the abstract idea addressed in the independent claim. In particular, these limitations at most describe additional contextual details about the data which is evaluated via the recited abstract idea and/or subsequent determinations and classifications which are similarly evaluations capable of being performed in the mind.
Furthermore, under step 2A Prong 2 and 2B, the claim(s) do not recite additional elements to consider other than those considered in the independent claim. Accordingly, the recited additional elements do not integrate the abstract idea into a practical application because they do not impose any meaningful limits on practicing the abstract idea, nor do they amount to significantly more than the judicial exception because they do not impose any meaningful limits on practicing the abstract idea.
Claim Rejections - 35 U.S.C. § 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 of this title, 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) 1-9, 11-12, 16-18 are rejected under 35 U.S.C. § 103 as being unpatentable over Mohammed “Crack detection in a rotating shaft using artificial neural networks and PSD characterization” further in view of Wu “Machine Learning Approach for Shaft Crack Detection through Acoustical Emission Signals”
Regarding Claim 1
Mohammed teaches, one or more processors; and one or more memory devices, the one or more memory devices storing computer-readable instructions that when executed by the one or more processors cause the one or more processors to perform operations for constructing a machine learning model correlating fatigue crack growth with operational data, the operations comprising: (abstract “The system uses vertical vibration of the system measured over time and characterises its behaviour using elements… The PSDs were used as an input into an artificial neural network (ANN) to detect the presence of cracks using changes in the spectral content of the vibration of the system.” Pg 10 “For the training of the networks, the sum-squared goal was 10−10 and the number of epochs was limited to a maximum of 1000” on of ordinary skill would understand that such training is performed on a computer with memory via instructions as claimed.) obtaining data indicative of fatigue crack size and operational data for a rotatable structure…predicting fatigue crack growth for the rotatable structure using a machine learning model, wherein the machine learning model is constructed by:… classifying the historical data and the historical operational data based on at least two regions of fatigue crack size; …training the machine learning model to predict fatigue crack growth based on the historical data, the operational data, and the classifications of the historical data based on the at least two regions of fatigue crack size. … wherein training the machine learning model comprises:… and classifying operations of the historical operational data based on a growth region associated with the fatigue crack growth. (pg 2 “Vibration data is collected as digitally sampled time domain signals and the development of transforms, such as the fast Fourier transform (FFT)” pg 8 “In this study the data was divided into 2 groups for each fault condition all of which were processed according to the procedure documented in Sect. 3.2 to provide the arrays ANda. The first group of data was used to train the ANN. The second group was used for testing using “unseen” data. The network was trained using the fault condition data as inputs along with the corresponding fault conditions as the known outputs” pg 4 “The cracked shaft was tested with different crack depths ranging between 0 % and 60 % of the shaft diameter as shown in Table 1” the data is obtained and used to predict fault condition outputs, which in the art are described as cracks of varying depths, thus of at least varying regions in the shaft. The predicting involves classifying, training and predicting cracks. This obtained data is from the historical operating of the shafts, thus considered historical operational data and historical data. The training on different crack depths amounts to classification of different working operations of the shaft based on growth regions associated with crack growth) obtaining historical data indicative of fatigue crack size and historical operational data for a plurality of rotatable structures; (pg 3 section 2.2 “A series of shaft samples were tested with the experiments being carried out for increasing depth of cut…. The experiments were carried out on two shafts; the first shaft without any defects was used as a reference. The second shaft…” data, which is considered historical and operational data, is extracted from at least two shafts.)
Mohammed does not explicitly teach, [wherein training the machine learning model comprises:]…pre-processing the historical operational data to identify quality issue within the historical operational data
Wu however teaches, [wherein training the machine learning model comprises:]…pre-processing the historical operational data to identify quality issue within the historical operational data ( pg 4 Section B and C “As shown in Fig 3, the shaft is inserted with rotor and MFS is assembled for simulation. There are four accelerometers in total which are used to collect the AE data…. Fig.5 shows the data acquisition process. All data were collected by KISTLER K-Shear Accelerometer Type 8152B111….In our experiment, the signal data was not used for data analysis when the speed is changing because of unstable signals. It is postulated that setting up a speed to a certain value takes 5 second and depreciating it back to 0 rpm consumes 5 seconds. Thus, data is reduced to 10 seconds region. Fig.7 shows the way of data reduction…” pg 5 “Meanwhile, from the figures of RMS and Kurtosis, the result curves in 2300rpm are smoother. In order to build the better prediction model, the feature RMS and Kurtosis are selected” pg 6 “For the backpropagation method, there were still two data selection plan to build the models. For the selected data, 80% was randomly chosen as training set,… Number of hidden layers was set as 10 neurons, step size was set to 0.1, and training goal was set” the training data is pre-processed in a variety of ways including removal of specific data and different features.)
Accordingly, it would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to modify the crack detection method using ANNs of Mohammed to comprise data preprocessing methods described by Wu. One would have been motivated to make such a combination because as noted by Wu “In addition, the AE signal often contains points of extreme value because of noise, signal distortion, sensor failure and many other reasons, such spurious data bring troubles in data analyzing and processing… Hence, a basic assignment before data analyzing is to cut down the data into suitable size and de-noising it” (Wu pg 4)
Regarding Claim 2
Mohammed/Wu teaches claim 1
Mohammed teaches, wherein each of the at least two regions is defined by a range of crack sizes (pg 4 Section 2.3 “The cracked shaft was tested with different crack depths ranging between 0 % and 60 % of the shaft diameter as shown in Table 1”)
Regarding Claim 3
Mohammed/Wu teaches claim 1
Mohammed teaches, wherein a machine learning model is determined for each of the at least two regions of fatigue crack size. (pg 9 “Half of these files were used to train the neural networks; the other half of the data set was used to test the networks performance with each case using 500 training patterns for each case of the condition. Those cases are… 1. Normal shaft 2. Faulty shaft with 40 % cut 3. Faulty shaft with 50 % cut 4. Faulty shaft with 60 % cut.” A machine learning model is determined for each of the plurality of crack size regions.)
Regarding Claim 4
Mohammed/Wu teaches claim 1
Mohammed teaches, wherein the at least two regions of fatigue crack size are determined according to different patterns of crack growth in the at least two regions. (pg 4 Section 2.3 “The cracked shaft was tested with different crack depths ranging between 0 % and 60 % of the shaft diameter as shown in Table 1” cracks of different sizes are considered different patterns of crack growth of at least two regions. For example, a crack of 10% only covers a region which is 10% of the diameter, while a 60% crack cover a different region including the remaining diameter of the shaft.)
Regarding Claim 5
Mohammed/Wu teaches claim 1
Mohammed teaches, wherein the historical data is further classified based on a dwell time and the machine learning model is trained to predict fatigue crack growth based on the dwell time. (pg 4 “To ensure that the vibration data was taken once the system had reached a stable state and steady state running, a prolonged test was carried out where the temperature was measured on the motor bearing and generator bearing. The results from the temperature test indicated that the test rig should be run for more than 4 hours for every defect condition to ensure a stable system. This ensured that all measurements were taken at the same running conditions and any changes in the vibration spectra were due to the fault condition and not to changes resulting from starting from cold after a shutdown… The shaft was run at 1550 rpm with a load of 20 kW for 4 hours until a steady state was achieved and then vibration measurements were taken at 30 minutes… The cracked shaft was tested with different crack depths ranging between 0 % and 60 % of the shaft diameter as shown in Table 1” the is classified into cracks of varying conditions based in part on the dwell time. In this case the dwell time is 4 hours running at a steady load and speed. The data collected is then used for training as previously noted.)
Regarding Claim 6
Mohammed/Wu teaches claim 5
Mohammed teaches, the dwell time includes duration of a movement event while at least one engine parameter remains within a range specified by an upper bound and a lower bound (pg 4 “The shaft was run at 1550 rpm with a load of 20 kW for 4 hours until a steady state was achieved and then vibration measurements were taken at 30 minutes” the shaft was run during a rotational movement event. The machine ran at a fixed speed thus specified to run at a speed less than 1551 RPM and greater than 1549 RPM)
Regarding Claim 7
Mohammed/Wu teaches claim 6
Mohammed teaches, wherein a movement event includes a flight, a power generation process, or a drive. (pg 4 “The shaft was run at 1550 rpm with a load of 20 kW for 4 hours until a steady state was achieved and then vibration measurements were taken at 30 minutes” driving the shaft at a particular speed is a drive event.)
Regarding Claim 8
Mohammed/Wu teaches claim 6
Mohammed teaches, the at least one engine parameter includes at least one of temperature, core engine speed, or acceleration (pg 4 “The shaft was run at 1550 rpm with a load of 20 kW for 4 hours until a steady state was achieved and then vibration measurements were taken at 30 minutes” the shaft rotational speed is an engine core speed.)
Regarding Claim 9
Mohammed/Wu teaches claim 6
Mohammed teaches, wherein the historical data is classified as one of at least two types of cycles based on the dwell time in a first engine speed band, with a first upper bound and a first lower bound, and the dwell time in a second engine speed band, with a second upper bound and a second lower bound. (pg 8 Section 4.1 “The reference was used for two things, firstly to check if there were any changes in the system response for reasons other than the crack, and then secondly to use it to compare with the shafts with defects. The shaft with a simulated crack was then examined with the vibration responses measured for different defect depths…. As shown in Fig. 7 Ch1 has three frequency ranges containing peaks. The first two peaks are from 300 to 400 Hz. There is then a peak between 800 to 1000 Hz, and the last peak is in the range 1000 to 1200 Hz. For Ch2 there are two ranges that are going to use for comparing between un-cracked and cracked shaft. These ranges are 325 to 365 Hz and 1100 to 1200 Hz.” The data is compared or classified as having vibrational frequency, or engine speed bands in different ranges, of different types of cycles defined by upper and lower vibration frequency. These measurements as previously noted are based on the 4 hour dwell time.)
Regarding Claim 11
Mohammed/Wu teaches claim 9
Mohammed teaches, wherein the at least two types of cycles are defined by different engine speed bands. (pg 8 Section 4.1 “The reference was used for two things, firstly to check if there were any changes in the system response for reasons other than the crack, and then secondly to use it to compare with the shafts with defects. The shaft with a simulated crack was then examined with the vibration responses measured for different defect depths…. As shown in Fig. 7 Ch1 has three frequency ranges containing peaks. The first two peaks are from 300 to 400 Hz. There is then a peak between 800 to 1000 Hz, and the last peak is in the range 1000 to 1200 Hz. For Ch2 there are two ranges that are going to use for comparing between un-cracked and cracked shaft. These ranges are 325 to 365 Hz and 1100 to 1200 Hz.” The different sensor channels measure different speed peaks in at least two different ranges.)
Regarding Claim 12
Mohammed/Wu teaches claim 1
Mohammed teaches, wherein the historical data is further classified based on at least one time-above-value feature and the machine learning model is trained to predict fatigue crack growth based on the time-above-value feature. (pg 4 “To ensure that the vibration data was taken once the system had reached a stable state and steady state running, a prolonged test was carried out where the temperature was measured on the motor bearing and generator bearing. The results from the temperature test indicated that the test rig should be run for more than 4 hours for every defect condition to ensure a stable system. This ensured that all measurements were taken at the same running conditions and any changes in the vibration spectra were due to the fault condition and not to changes resulting from starting from cold after a shutdown… The shaft was run at 1550 rpm with a load of 20 kW for 4 hours until a steady state was achieved and then vibration measurements were taken at 30 minutes… The cracked shaft was tested with different crack depths ranging between 0 % and 60 % of the shaft diameter as shown in Table 1” this data, used to predict crack growth, is based on the time running above 4 hours, thus a time above value feature.)
Regarding Claim 16
Claim 16 is rejected for the reasons set forth in the rejection of independent claim 1
Regarding Claim 17
The claim is rejected for the reasons set forth in the rejection of claim 4 in connection with claim 16
Regarding Claim 18
The claim is rejected for the reasons set forth in the rejection of claim 5 in connection with claim 16
Claim(s) 10 is rejected under 35 U.S.C. § 103 as being unpatentable over Mohammed/Wu, further in view of Gomez “Automatic condition monitoring system for crack detection in rotating machinery”
Regarding Claim 10
Mohammed teaches claim 9
Mohammed/Wu does not explicitly teach, wherein each of the at least two types of cycles includes moving from one engine speed band to another engine speed band and returning to the one engine speed band
Gomez however when detecting cracks and varying engine speeds teaches, wherein each of the at least two types of cycles includes moving from one engine speed band to another engine speed band and returning to the one engine speed band (pg 3 Section 4 “The tested element is the shaft, under different crack conditions. A first test is made at healthy state, and then nine different crack levels (a) were induced by saw cuts. All the cracks were induced without dismounting the shaft from the machine, because usually cracks appear and grow while the machine rotates and the assembly effects are constant. The values of a in Table 1 are expressed as the ratio between the crack depth d and shaft diameter D, where D=16 mm …Test are carried out at steady state at three different rotational speeds, shown in Table 2. Thus, 30 different conditions are tested; 10 different crack conditions and 3 rotation speeds” Section 5.2 pg 4 “Once the features are extracted and selected, the ANNs training is designed.” running at different speeds is running the engine at different speed bands which are different types of cycles. The same shaft is run at 3 different speeds. As the crack level increases via the saw the speed cycles are reset in order to collect the 30 test conditions. Thus, progressing through the speed bands, and returning to the one speed band with each increase in crack level. This data is used for training.)
Accordingly, it would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to modify the crack detection method using ANNs of Mohammed/Wu to comprise crack detection based on varying speeds described by Gomez. One would have been motivated to make such a combination because as noted by Gomez “Tests were carried out at different speeds that show that diagnosis results are improved with the speed. This can be assigned to the fact that when the speed increases, as the tests were performed at the same conditions, the signal-to-noise ratio is higher. Then, crack effects are more clearly distinguished when the speed increases.” (pg 8 Discussion) As such, discovery of improved signal to noise ratio is a result of running tests at different speeds.
Claim(s) 13-15 and 19-20 are rejected under 35 U.S.C. § 103 as being unpatentable over Mohammed/Wu, further in view of Cui “Multi-Scale Convolutional Neural Networks for Time Series Classification”
Regarding Claim 13
Mohammed/Wu teaches claim 1
Mohammed teaches, and the machine learning model is trained to predict fatigue crack growth (pg 1 abstract “The PSDs were used as an input into an artificial neural network (ANN) to detect the presence of cracks”)
Mohammed/Wu does not explicitly teach, wherein the historical data is further classified based on at least one rolling window feature [the machine learning model is] based on the at least rolling window feature.
Cui however when addressing filtering of historical data for machine learning classification teaches, wherein the historical data is further classified based on at least one rolling window feature [the machine learning model is] based on the at least rolling window feature. (pg 3 “A low frequency filter can reduce the variance of time series. In particular, we employ moving average to achieve this goal. Given an input time series, we generate multiple new time series with varying degrees of smoothness using moving average with different window sizes. This way, newly generated time series represent general low frequency information, which make the trend of time series more clear. Suppose the original time series is T = {t1, t2, ..., tn}, the moving average works by converting this original time series into a new time series.”)
Accordingly, it would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to modify the crack detection method using ANNs of Mohammed/Wu d to comprise a low pass filter for processing input data for machine learning described by Cui. One would have been motivated to make such a combination because as noted by Cui “In real-world applications, high-frequency perturbations and random noises widely exist in the time series data due to many reasons, which poses another challenge to achieving high prediction accuracy… It is often hard to extract useful information on raw time series data with the presence of these noises… Given an input time series, we generate multiple new time series with varying degrees of smoothness using moving average with different window sizes… which make the trend of time series more clear” (Cui pg 3)
Regarding Claim 14
Mohammed/Wu/Cui teaches claim 13
Further Cui teaches, wherein the at least one rolling window feature includes statistical aggregated values of at least one parameter. (pg 3 “In particular, we employ moving average to achieve this goal. Given an input time series, we generate multiple new time series with varying degrees of smoothness using moving average with different window sizes. …”)
Regarding Claim 15
Mohammed/Wu/Cui teaches claim 14
Further Cui teaches, wherein the statistical aggregated values include at least one of mean, median, maximum, minimum, standard deviation, interquartile range, sum, product, count, cumulative values, logarithmic transformation. (pg 3 “In particular, we employ moving average to achieve this goal. Given an input time series, we generate multiple new time series with varying degrees of smoothness using moving average with different window sizes. …” a moving average is a mean)
Regarding Claim 19-20
The claims are rejected for the reasons set forth in the rejections of claim 13-14, respectively, in connection with claim 16
Conclusion
Prior art not relied on:
He et al “Application of artificial neural networks in engine modelling” makes predictions about engine parameters using certain other engine parameters in a defined band including engine speed. However, the neural network is not used to predict cracks.
Saarimaki et al. “Time- and cycle-dependent crack propagation in Haynes 282” measures crack growth in materials based on dwell time and time cycles, but not in rotating machinery nor using neural network methods.
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 extension fee 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.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to JOHNATHAN R GERMICK whose telephone number is (571)272-8363. The examiner can normally be reached M-F 9:30-4:30.
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, Kakali Chaki can be reached on 571-272-3719. 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.
/J.R.G./
Examiner, Art Unit 2122
/KAKALI CHAKI/ Supervisory Patent Examiner, Art Unit 2122