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 .
Information Disclosure Statement
The information disclosure statement (IDS) submitted on 02/24/2026 is in compliance with the provisions of 37 CFR 1.97. Accordingly, the information disclosure statement is being considered by the examiner.
Status of the Claims
3. This Office Action is in response to the Applicant’s filing on 02/24/2026. Claims 13 - 29 were previously pending, of which claims 13, 17, & 29 have been amended, no claims have been cancelled, and no new claims have been newly added. Accordingly, claims 13 - 29 are currently pending and are being examined below.
Response to Arguments
4. With respect to the Applicant’s remarks, see pages 9 - 11, filed on 02/24/2026; Applicant’s “Amendment and Remarks” have been fully considered. Applicant’s remarks will be addressed in sequential order as they were presented.
5. With respect to the rejection under 35 U.S.C. 103, applicant’s “Amendment and Remarks” have been fully considered and are persuasive. The prior art of record does not appear to disclose the limitations “…and after the first layer…” & “…and is selected based on the amount of importance to be given to the condition indicator data,…” as amended in claim 1. However, due to the nature of the applicant’s amendments, the scope of the applicant’s invention has changed and thus requires new analysis and new application of prior art and further search found that Kursun did disclose this limitation as mapped in the final office action below.
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.
Claim(s) 13 – 14, 16, & 29 are rejected under 35 U.S.C. 103 as being unpatentable over US20190279443A1 (hereinafter, “Bharadwaj”), and further in view of US20210173905A1 (hereinafter, “Kursun”).
Regarding claim 13, Bharadwaj discloses a health monitoring system comprising [0006]: This system is configured to provide prognostic indicators for use in aircraft maintenance and also provide aircraft health data.
a computer system [0006]; This is a computer-implemented system which implies a computer system of some sort is implemented.
a neural network in the computer system, wherein the neural network comprises ([0050] Fig. 4): A neural network is implemented into this computer system. Each encoder (408) in this computer system has a feedforward neural network which consists of an encoder network (410) and a decoder network (412).
layers ([0050], [0052] Fig. 4); The encoder network (410) has an input layer and a hidden layer. The decoder network (412) has a hidden layer and an output layer.
a first layer in the layers in the neural network, wherein the first layer is configured to receive multivariate time series sensor data for a set of variables for a component, wherein the set of variables is associated with at least one of a voltage, vibration, or temperature associated with the component; and ([0040], [0050] Fig. 4, 6A – 6B) It is inherent that this data of healthy aircraft components is retrieved by sensors on the aircraft. Monitoring aircraft components requires the usage of sensors. Bharadwaj teaches a multivariate time series because it discloses collecting and analyzing multiple system indicators in the set of health indicator models (240) over time to predict system health and failure conditions [0040]. Each indicator represents a distinct variable, one of those variables being vibration spectrum data (vibration), and the indicators are evaluated based on their temporal behavior. Accordingly, at any given time instance, the system processes a set of indicator values corresponding to multiple variables, which forms a multivariate time series dataset. Figures 6A – 6B further show this. Data that can be a set of variables/indicators from healthy aircraft components are inputted into the input nodes (411) which we can consider a first layer due to the fact that it is the first input for the encoder (408) ([0050] Fig. 4).
9. Bharadwaj further teaches …wherein the subsequent layer is configured to receive condition indicator data for a set of condition indicators…and wherein the neural network is trained to predict a health status of the component using the multivariate time series sensor data for the set of variables and the condition indicator data for the set of condition indicators; and ([0050] – [0051], [0064] Fig. 4) Bharadwaj teaches on a hidden layer (413) which is a subsequent layer before the output layer. This hidden layer (413) generates weighting factors for the healthy aircraft components data that are inputted into the input nodes (411) to be reconstructed at the output nodes (415). Since this hidden layer (413) depends on the healthy aircraft components data, we can consider this a condition indicator because it is indicating based on the healthy aircraft components condition, a weighting factor. This weighting factor is also that of a numerical value such as a condition indicator. Therefore, we can consider this weighting factor as a condition indicator. This neural network may incorporate trend detection algorithms which may be used to predict trends that relate to the health status of the health aircraft components data [0051]. This neural network uses sensor data because the sensor data is the healthy aircraft components data considering that data is measured inherently with sensors and uses weighting factors which is considered as a condition indicator. Therefore, the trend detection algorithm which may be used to predict trends is predicting a trend of the status of the healthy aircraft components data.
Bharadwaj does not explicitly teach a subsequent layer before an output layer in the layers in the neural network and after the first layer,…and is selected based on the amount of importance to be given to the condition indicator data,
However, Kursun teaches a subsequent layer before an output layer in the layers in the neural network and after the first layer,…and is selected based on the amount of importance to be given to the condition indicator data,… ([0089 - [0090], [0102] Fig. 7 & 9) Kursun teaches a subsequent neural network layer that receives input data (X1 – X4, Xm) comprising multiple features (condition indicator data) and determines a relevance score (amount of importance) for each feature [0089] - [0090], [0102]. These relevance scores represent the relative importance of the features and are used to control how the features are processed within the layer. This subsequent layer labeled as Layer 1, Layer N-1, and Layer N come before the output layer (Xp) and after the first layer (Xm) (Fig. 7 & 9). Accordingly, the layer is effectively selected or configured based on the importance assigned to the input data.
Bharadwaj and Kursun are analogous art because Bharadwaj teaches a neural network that inputs health aircraft components data into the input nodes and uses trend detection algorithms to detect any trends in the aircraft components data to detect any statuses of the aircraft components while Kursun teaches a neural network that has a subsequent neural network between and output and input layer and has features that are inputted to determine a relevance score of each feature. A person of ordinary skill in the art would have the motivation to combine Bharadwaj with Kursun to allow the neural network to prioritize more significant condition indicators, thereby improving trend detection and prognostic accuracy.
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention, having the teachings of Kursun, to modify the teachings of Bharadwaj to include the teachings of Kursun to more accurately detect trends and assess aircraft component health.
10. Bharadwaj further teaches a health analyzer in the computer system, wherein the health analyzer is configured to [0028]: Bharadwaj teaches on an Aircraft health data (201) which contains condition indicator data, spectrum data, resampled data, and RTD spectrum data. Due to this, the aircraft health data (201) constitutes as a health analyzer since this aircraft health data (201) can be analyzed to determine health values and numbers.
input the multivariate time series sensor data into the first layer ([0028], [0040], [0050] Fig. 2a, 3, & 4); Bharadwaj teaches a multivariate time series because it discloses collecting and analyzing multiple system indicators in the set of health indicator models (240) over time to predict system health and failure conditions [0040]. Accordingly, at any given time instance, the system processes a set of indicator values corresponding to multiple variables, which forms a multivariate time series dataset. If we follow the aircraft health data (201) through the figures, the data from the aircraft health data (201) will eventually lead up to the input nodes (411) of the encoder (410). The aircraft health data in figure 2a carries data all the way down to raw/semi processed health data in the same figure 2a. That raw/semi processed health data (308) in figure 3 then gets transferred over into the deep auto encoders (306). In figure 4, the deep auto encoder (408) then inputs that data into the input nodes (411) of the encoder (410) (first layer). Therefore, we can say that the multivariate time series sensor data from the aircraft health data (201) will eventually be inputted into the first layer.
input the condition indicator data into the subsequent layer; and ([0028], [0050] Fig. 2a, 3, & 4) The aircraft health data (201) houses condition indicator data [0028]. Therefore, for the same reasons stated in the above limitation, eventually the hidden layer (413) will receive the condition indicator data as an input in the neural network that is housed in the aircraft health data (201) as the data makes its way through the system.
receive the health status predicted for the component from the output layer in response to inputting the multivariate time series sensor data and the condition indicator data ([0028], [0054] Fig. 4 – 6b). The aircraft health data (201) houses the important data that would need to be updated constantly due to this system being a monitoring system for aircraft maintenance. The aircraft health data (201) will inherently have to receive the predicted health status of the healthy aircraft components data to keep the status of the system updated at all times in case a failure happens. The healthy aircraft components data which is sensor data will be inputted into the input layer (411), then the hidden layer to generate weighting factors (condition indicators), which then will go through the output nodes (415) (output layer), and then eventually will provide health status information to a web server (Fig. 6b).
Regarding claim 14, Bharadwaj discloses the health monitoring system of claim 13, wherein the health analyzer is configured to: perform a set of actions using the health status predicted for the component [0031] – [0032]. Bharadwaj teaches an aircraft display (212) that can display the health state and prognostic information. This aircraft display (212) will receive information from aircraft health data (201) (Fig. 2). The three time horizons (209, 211, 213) will each light up their respective color in accordance with component health. First time horizon (209) will light up green for good. Second time horizon (211) will light up yellow for cautious. Third time horizon (213) will light up red for not good. These color indications constitute as warnings (a set of actions) indicative of a health status of the components to the user which were predicted.
Regarding claim 16, Bharadwaj discloses the health monitoring system of claim 14, wherein the component is located in a platform selected from a group comprising a mobile platform, a stationary platform, a land-based structure, an aquatic-based structure, a space-based structure, an aircraft [0023], a commercial aircraft, a rotorcraft, a tilt-rotor aircraft, a tilt wing aircraft, a vertical takeoff and landing aircraft, an electrical vertical takeoff and landing vehicle, a personal air vehicle, a surface ship, a tank, a personnel carrier, a train, a spacecraft, a space station, a satellite, a submarine, an automobile, a power plant, a bridge, a dam, a house, a manufacturing facility, and a building. Bharadwaj discloses an aircraft maintenance that resides onboard an aircraft.
13. Regarding claim 29, Bharadwaj discloses a method for monitoring a health of a component, the method comprising:
inputting sensor data for a set of variables for the component into a first layer in layers in a neural network [0050]; It is inherent that this data of healthy aircraft components is retrieved by sensors on the aircraft. Monitoring aircraft components requires the usage of sensors. Therefore, this data can be referred to as sensor data. Sensor data that can be a set of variables from healthy aircraft components are inputted into the input nodes (411) which we can consider a first layer due to the fact that it is the first input for the encoder (408).
inputting condition indicator data for a set of condition indicators into a subsequent layer…wherein the neural network is trained to predict a health status of the component using the sensor data for the set of variables and the condition indicator data for the set of condition indicators; and ([0050] – [0051], [0064] Fig. 4) Bharadwaj teaches on a hidden layer (413) which is a subsequent layer before the output layer. This hidden layer (413) generates weighting factors for the healthy aircraft components data that are inputted into the input nodes (411) to be reconstructed at the output nodes (415). Since this hidden layer (413) depends on the healthy aircraft components data, we can consider this a condition indicator because it is indicating based on the healthy aircraft components condition, a weighting factor. This weighting factor is also that of a numerical value such as a condition indicator. Therefore, we can consider this weighting factor as a condition indicator. This neural network may incorporate trend detection algorithms which may be used to predict trends that relate to the health status of the health aircraft components data [0051]. This neural network uses sensor data because the sensor data is the healthy aircraft components data considering that data is measured inherently with sensors and uses weighting factors which is considered as a condition indicator. Therefore, the trend detection algorithm which may be used to predict trends is predicting a trend of the status of the healthy aircraft components data.
receiving the health status predicted for the component from the output layer in response to inputting the sensor data and the condition indicator data ([0028], [0054] Fig. 4 – 6b). The aircraft health data (201) houses the important data that would need to be updated constantly due to this system being a monitoring system for aircraft maintenance. The aircraft health data (201) will inherently have to receive the predicted health status of the healthy aircraft components data to keep the status of the system updated at all times in case a failure happens. The healthy aircraft components data which is sensor data will be inputted into the input layer (411), then the hidden layer to generate weighting factors (condition indicators), which then will go through the output nodes (415) (output layer), and then eventually will provide health status information to a web server (Fig. 6b).
Bharadwaj does not explicitly teach …before an output layer and after the first layer in the layers in the neural network, wherein the subsequent layer selected based on the amount of importance to be given to the condition indicator data, and…
However, Kursun teaches …before an output layer and after the first layer in the layers in the neural network, wherein the subsequent layer selected based on the amount of importance to be given to the condition indicator data, and… ([0089 - [0090], [0102] Fig. 7 & 9) Kursun teaches a subsequent neural network layer that receives input data (X1 – X4, Xm) comprising multiple features (condition indicator data) and determines a relevance score (amount of importance) for each feature [0089] - [0090], [0102]. These relevance scores represent the relative importance of the features and are used to control how the features are processed within the layer. This subsequent layer labeled as Layer 1, Layer N-1, and Layer N come before the output layer (Xp) and after the first layer (Xm) (Fig. 7 & 9). Accordingly, the layer is effectively selected or configured based on the importance assigned to the input data.
Bharadwaj and Kursun are analogous art because Bharadwaj teaches a neural network that inputs health aircraft components data into the input nodes and uses trend detection algorithms to detect any trends in the aircraft components data to detect any statuses of the aircraft components while Kursun teaches a neural network that has a subsequent neural network between and output and input layer and has features that are inputted to determine a relevance score of each feature. A person of ordinary skill in the art would have the motivation to combine Bharadwaj with Kursun to allow the neural network to prioritize more significant condition indicators, thereby improving trend detection and prognostic accuracy.
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention, having the teachings of Kursun, to modify the teachings of Bharadwaj to include the teachings of Kursun to more accurately detect trends and assess aircraft component health.
Claim(s) 15 is rejected under 35 U.S.C. 103 as being unpatentable over US20190279443A1 (hereinafter, “Bharadwaj”), and further in view of US20210173905A1 (hereinafter, “Kursun”), and further in view of US20030066352A1 (hereinafter, “Leamy”).
Regarding claim 15, Bharadwaj does not explicitly teach the health monitoring system of claim 13, wherein the component is selected from a group comprising a generator, the generator bearings, a pump, a cooling system, a heat exchanger, an auxiliary power unit, a landing gear system, a wing, an in-flight entertainment system, and a computer [0022], [0028]. Bharadwaj teaches on an aircraft maintenance system that retrieves aircraft health data for a plurality of aircraft components. One of those aircraft components is bearings.
However, Leamy in the same field of endeavor, teaches the health monitoring
system of claim 13, wherein the component is selected from a group comprising a generator, the generator bearings, a pump, a cooling system, a heat exchanger, an auxiliary power unit, a landing gear system, a wing, an in-flight entertainment system, and a computer [0026]. Leamy monitors bearings and may be used to monitor bearings mounted on gas turbine electrical generators. This constitutes as monitoring generator bearings.
One of ordinary skill in the art, before the effective filing date of the instant application with a reasonable expectation of success, would have been motivated to modify the disclosure of Bharadwaj with the teachings of Leamy, to monitor bearings on a generator to more effectively predict the health status and decaying of the bearings.
Claim(s) 17 – 20, 22 – 26, & 28 are rejected under 35 U.S.C. 103 as being unpatentable over US20030066352A1 (hereinafter, “Leamy”), and further in view of US20190279443A1 (hereinafter, “Bharadwaj”), and further in view of US20210173905A1 (hereinafter, “Kursun”).
17. Regarding claim 17, Leamy discloses a method for monitoring a health of generator bearings, the method comprising [0015] – [0016], [0026]: Leamy teaches on a method to monitor bearings that may be mounted on gas turbine electrical generators.
identifying sensor data for a set of variables from a sensor system monitoring a generator including the generator bearings ([0031] - [0033] Fig. 2 - 3); Leamy teaches on the monitoring of bearings that may be mounted on a type of generator as specified above. Leamy identifies sensor data using vibration sensors (62) or speed sensors (66 & 70) for a set of variables such as accelerometer and speed data into the sensor system.
identifying condition indicator data for a set of condition indicators for the generator bearings [0035] – [0036] Fig. 2 -3; Condition indicator data on the set of bearings that relate to generators are identified. Signals 78, 82, and 86 are considered condition data because those signals are indicating the conditions of the bearings based on sensor data that were collected from those respective sensors. The condition data from those signals get inputted into the engine vibration monitor (90) which get further analyzed and determine whether the data is “Qualified Transient” or “Qualified Steady State”.
Leamy does not explicitly teach inputting the sensor data into a first layer in layers in a neural network;
inputting the condition indicator data into a last layer before an output layer and after the first layer in the layers in the neural network, wherein the last layer selected based on the amount of importance to be given to the condition indicator data, and wherein the neural network is trained to predict a health status of the generator bearings using the sensor data for the set of variables and the condition indicator data for the set of condition indicators; and
receiving a prediction of the health status for the generator bearings from the output layer in response to inputting the sensor data and the condition indicator data.
However, Bharadwaj teaches inputting the sensor data into a first layer in layers in a neural network ([0035] – [0036], [0041] Fig. 2 - 3); If we follow the aircraft health data (201) through the figures, the data from the aircraft health data (201) will eventually lead up to the input nodes (411) of the encoder (410). The aircraft health data in figure 2a carries data all the way down to raw/semi processed health data in the same figure 2a. That raw/semi processed health data (308) in figure 3 then gets transferred over into the deep auto encoders (306). In figure 4, the deep auto encoder (408) then inputs that data into the input nodes (411) of the encoder (410) (first layer). Therefore, we can say that the sensor data from the aircraft health data (201) will eventually be inputted into the first layer.
inputting the condition indicator data into a last layer…wherein the neural network is trained to predict a health status of the generator bearings using the sensor data for the set of variables and the condition indicator data for the set of condition indicators; and ([0028], [0050] – [0052] Fig. 2a, 3, & 4) The aircraft health data (201) houses condition indicator data [0028]. Therefore, for the same reasons stated in the above limitation, eventually the hidden layer (413), we are treating this as a last layer before the output layer, will receive the condition indicator data as an input in the neural network that is housed in the aircraft health data (201) as the data makes its way through the system. The neural network may use a trend detection algorithm to predict trends for the healthy aircraft components data which contains data for bearings [0051] - [0052]. This data may contain condition indicators in regards to the bearings that is being used for predictions [0052].
receiving a prediction of the health status for the generator bearings from the output layer in response to inputting the sensor data and the condition indicator data ([0028], [0054] Fig. 4 – 6b). The aircraft health data (201) houses the important data that would need to be updated constantly due to the monitoring of the aircraft bearings. The aircraft health data (201) will inherently have to receive the predicted health status of the healthy aircraft components data to keep the status of the system updated at all times in case a failure happens. The healthy aircraft components data which is sensor data will be inputted into the input layer (411), then the hidden layer to generate weighting factors (condition indicators), which then will go through the output nodes (415) (output layer), and then eventually will provide health status information to a web server (Fig. 6b).
Leamy does not explicitly teach …before an output layer and after the first layer in the layers in the neural network, wherein the last layer selected based on the amount of importance to be given to the condition indicator data, and…
However, Kursun teaches …before an output layer and after the first layer in the layers in the neural network, wherein the last layer selected based on the amount of importance to be given to the condition indicator data, and… ([0089 - [0090], [0102] Fig. 7 & 9) Kursun teaches a subsequent neural network layer that receives input data (X1 – X4, Xm) comprising multiple features (condition indicator data) and determines a relevance score (amount of importance) for each feature [0089] - [0090], [0102]. These relevance scores represent the relative importance of the features and are used to control how the features are processed within the layer. This subsequent layer labeled as Layer 1, Layer N-1, and Layer N come before the output layer (Xp) and after the first layer (Xm) (Fig. 7 & 9). Accordingly, the layer is effectively selected or configured based on the importance assigned to the input data.
Leamy and Bharadwaj are analogous art because Leamy teaches on the monitoring of bearings mounted on generators while Bharadwaj teaches on the monitoring of bearings and also includes a neural network into its monitoring method while Kursun teaches a neural network that has a subsequent neural network between and output and input layer and has features that are inputted to determine a relevance score of each feature. One of ordinary skill would be motivated to add this neural network layer system that Bharadwaj teaches and Kursun’s relevance scoring into Leamy because modern neural networks improve detection/diagnosis and prognostics in noisy, multi-bearing, real-world signals, can fuse multi-sensor data, learn failure patterns that may be missed by simple methods, and are now computationally and practically feasible.
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention, having the teachings of Bharadwaj and Kursun to modify the teachings of Leamy to further improve detection, diagonosis, and prognostics of generator bearings in an aircraft.
Regarding claim 18, Leamy discloses the method of claim 17 further comprising:
generating the sensor data for the generator bearings using the sensor system [0031] - [0033] Fig. 2 – 3). Leamy teaches on the monitoring of bearings that may be mounted on a type of generator as specified above. Leamy identifies sensor data using vibration sensors (62) or speed sensors (66 & 70) for a set of variables such as accelerometer and speed data into the sensor system. Due to the identification of the data from the accelerometer and the speed sensor, this data would have to have been generated inherently by the sensors to even identify such data.
19. Regarding claim 19, Leamy discloses the method of claim 17 further comprising: performing a set of actions using the health status predicted for the generator bearings ([0039] Fig. 3). Leamy teaches providing a sufficient warning for when to take appropriate corrective action. A message will be displayed to the user or an alarm will go off to alert the user.
20. Regarding claim 20, Leamy discloses the method of claim 19, wherein the set of actions is selected from at least one of logging the health status, generating a warning [0039], scheduling maintenance for the generator bearings; or halting operation of the generator in which the generator bearings are located. Leamy teaches providing a sufficient warning for when to take appropriate corrective action. When a threshold criteria has been reached in step (109) of figure 3, a message will be displayed or an alarm will go off to alert the user.
21. Regarding claim 22, Leamy teaches the method of claim 17, wherein the health status of the generator bearings… ([0039] Fig. 3) Leamy teaches determining the health status of generator bearings. After filtering out the results (106), defects are identified (107) and determined whether they have reached a certain threshold criteria or not (109). This step-by-step process is determining the health status of the generator bearings based on a threshold criteria.
Leamy does not explicitly teach …is selected from a group comprising normal, caution, and warning.
However, Bharadwaj in the same field of endeavor, teaches …is selected from a group comprising normal, caution, and warning [0031] – [0032]. Bharadwaj teaches an aircraft display (212) that can display the health state and prognostic information. This prognostic information will include information in regards to bearings since the aircraft health data (201) contains information regarding bearings [0028]. This aircraft display (212) will receive information from aircraft health data (201) (Fig. 2A). The three time horizons (209, 211, 213) will each light up their respective color in accordance with component health. First time horizon (209) will light up green for good. Second time horizon (211) will light up yellow for cautious. Third time horizon (213) will light up red for not good. These color indications constitute as warnings for the health status of the components (bearing information is included) to the user.
Leamy and Bharadwaj are analogous art because Leamy can determine the health status of the generator bearings while Bharadwaj can give a health severity of the bearings. A person of ordinary skill would be motivated to implement a three-level health indicator with Leamy because Leamy already produces metrics that must be turned into operational maintenance. Translating amplitude and the measurements Leamy derives into a set of status bands is well within ordinary skill and is a straightforward and predictable enhancement that offers clear safety benefits.
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention, having the teachings of Bharadwaj to modify the teachings of Leamy to more accurately display to the user the health severity of the generator bearings.
22. Regarding claim 23, Leamy discloses the method of claim 17, wherein the set of condition indicators is selected from at least one of a bearing based energy, a ball energy, an inner race energy, an outer race energy, a generator frequency of vibrations for the generator [0026], [0031], [0037], hydraulic pump frequency, a hydraulic pump piston pass frequency, or side lube pump frequency. Leamy teaches on vibration sensors that monitor the generator bearings. These vibration sensors monitor vibrations of the bearings and then determine the strength of the bearing’s frequency. Defects are discovered based on the frequency.
23. Regarding claim 24, Leamy discloses the method of claim 17, wherein the set of variables is selected from at least one of a voltage, a voltage phase, an acceleration, a current, a temperature, a vibration frequency ([0037] - [0039] Fig. 4), or an acoustic emission. Leamy teaches on vibration frequencies. Figure 4 shows a graphical display of these frequencies being measured by using vibration sensors located near generator bearings.
24. Regarding claim 25, Bharadwaj discloses the method of claim 17, wherein the sensor data for the set of variables is selected from at least one of analog times series data or digital time series data [0033]. Bharadwaj teaches on using digital signal processing. Since Bharadwaj teaches on digital signal processing, data that is received will be in digital form (digital time series). Due to this, vibration sensor information that is going to be used as variables (accelerometer & speed data) as it runs through this system of analyzing the health of the generator bearings will be in digital form (digital time series).
25. Regarding claim 26, Leamy does not explicitly teach the method of claim 17, wherein the layers are fully connected layers.
However, Bharadwaj teaches the method of claim 17, wherein the layers are fully connected layers (Fig. 4). The layers in this neural network are fully connected as shown in figure 4. Input layer (411) connects to the hidden layer (413) which connects to the outer layer (415).
One of ordinary skill in the art, before the effective filing date of the instant application with a reasonable expectation of success, would have been motivated to modify the disclosure of Leamy with the teachings of Bharadwaj, to have connected layers in a neural network for smoother information transferring process.
26. Regarding claim 28, Leamy discloses the method of claim 17, wherein the generator bearings are located in a platform selected from a group comprising a mobile platform, a stationary platform, a land-based structure, an aquatic-based structure, a space-based structure, an aircraft [0026], a commercial aircraft, a rotorcraft, a tilt-rotor aircraft, a tilt wing aircraft, a vertical takeoff and landing aircraft, an electrical vertical takeoff and landing vehicle, a personal air vehicle, a surface ship, a tank, a personnel carrier, a train, a spacecraft, a space station, a satellite, a submarine, an automobile, a power plant, a bridge, a dam, a house, a manufacturing facility, and a building. Leamy teaches on monitoring of bearings that may be mounted on a gas turbine electrical generator that may be inside of an aircraft.
Claim(s) 21 is rejected under 35 U.S.C. 103 as being unpatentable over US20030066352A1 (hereinafter, “Leamy”), and further in view of US20190279443A1 (hereinafter, “Bharadwaj”), and further in view of US20210173905A1 (hereinafter, “Kursun”), and further in view of US20240264590A1 (hereinafter, “Mitra”).
Regarding claim 21, Leamy as modified by Bharadwaj does not explicitly teach the method of claim 17, wherein the neural network is a physics informed neural network.
However, Mitra teaches the method of claim 17, wherein the neural network is a physics informed neural network [0016], [0069]. This computer-implemented system of monitoring degradation of and determining prognosis reasoning for engineering assets, systems, sub-systems, components, and so forth uses a physics-informed neural network [0016]. Mitra states that this computer-implemented system may be used within an aircraft [0069].
One of ordinary skill in the art, before the effective filing date of the instant application with a reasonable expectation of success, would have been motivated to modify the disclosure of Leamy as modified by Bharadwaj with the teachings of Mitra, to have the neural network further detect physics-based failures or errors that may occur with specific components that operate on physics laws.
Claim(s) 27 is rejected under 35 U.S.C. 103 as being unpatentable over US20030066352A1 (hereinafter, “Leamy”), and further in view of US20190279443A1 (hereinafter, “Bharadwaj”), and further in view of US20210173905A1 (hereinafter, “Kursun”), and further in view of US20250321571A1 (hereinafter, “Wen”).
30. Regarding claim 27, Leamy as modified by Bharadwaj does not explicitly teach the method of claim 17 further comprising:
identifying sample sensor data for the set of variables and sample condition indicator data for the set of condition indicators;
generating a training dataset using the sample sensor data for the set of variables and the sample condition indicator data for the set of condition indicators; and
training the neural network using the training dataset.
However, Wen in the same field of endeavor, teaches the method of claim 17 further comprising:
identifying sample sensor data for the set of variables and sample condition indicator data for the set of condition indicators [0013], [0015] – [0016]; Wen teaches on using parameters such as a working voltage value, current value, temperature value, loading state, a clock lock mark, and a signal amplitude of avionic products. These parameters can be considered as sample sensor data since sensors are ultimately measuring this data [0013]. Wen also teaches on prediction labels which we can consider as condition indicator data at a minimum [0015] – [0016]. This is because Wen explicitly treats them as numeric prediction values. If this prediction labels have this type of prediction, then some sort of degradation factor or health indication is inputted in order to determine a health state of the avionic product.
generating a training dataset using the sample sensor data for the set of variables and the sample condition indicator data for the set of condition indicators; and [0015] – [0016] The training rule of the support vector then generates a training set composition composed of training data and labels which contains the sample sensor data parameters and labels.
training the neural network using the training dataset [0014] – [0019]. A plurality of these base models are then trained using this data. These models constitute as neural networks [0014].
One of ordinary skill in the art, before the effective filing date of the instant application with a reasonable expectation of success, would have been motivated to modify the disclosure of Leamy as modified by Bharadwaj with the teachings of Wen, to have neural networks trained for increased speed on future predictions.
Conclusion
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to DAVID MESQUITI OVALLE JR. whose telephone number is (571)272-6229. The examiner can normally be reached Monday - Friday 7:30am - 5pm EST.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Erin Piateski can be reached on (571) 270-7429. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
/DAVID MESQUITI OVALLE/ Examiner, Art Unit 3669
/Erin M Piateski/Supervisory Patent Examiner, Art Unit 3669
Prosecution Insights