DETAILED ACTION
Claims 1-20 have been presented for examination.
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 statements (IDS) submitted on November 11, 2024, and June 07, 2023 are in compliance with the provisions of 37 CFR 1.97. Accordingly, the Examiner has considered the IDS as to the merits.
Drawings
The drawings submitted on June 07, 2023, have been accepted.
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.
Regarding claims 1-3 and 5-20 are rejected under 35 U.S.C. 101 because the claimed invention is directed to a judicial exception (i.e. abstract idea) without anything significantly more.
Step 1: Claims 1-11 are directed to a method, which is a process, which is a statutory category of invention. Claims 12-20 are directed to a system, which is a machine, which is a statutory category of invention. Therefore, claims 1-21 are directed to patent eligible categories of invention.
Step 2A, Prong 1: Claims 1 and 12 recite the abstract idea of optimizing the assembly of rotating hardware in gas turbine engines, constituting an abstract idea based on Mathematical Concepts including mathematical formulas or equations as well as calculations or alternatively Mental Processes based on concepts performed in the human mind, or with the aid of pencil and paper. The limitation of “wherein each of the one or more first neural networks has been trained with: training data comprising the one or more contributors to unbalance; and one or more rotor dynamics models that use the training data and the set of clock angles from the one or more first neural networks to predict vibration at one or more locations of interest in the gas turbine engine”, in claims 1 and 12, covers mathematical concepts including defining the data that will be used as inputs into the neural networks and the models that will use the input parameters. Alternatively, this limitation covers mental processes including defining the data that will be used as input into the neural networks and the models that will use the input parameters, which can be performed with the use of a pencil and paper. Thus, the claims recite the abstract idea of a mental process performed in the human mind, or with the aid of pencil and paper.
Dependent claims 2-3, 5-11, and 13-20 further narrow the abstract ideas, identified in the independent claims.
Step 2A, Prong 2: The judicial exception is not integrated into a practical application. In Claims 1 and 12, the additional element of “one or more first neural networks”, and “processing circuitry operatively connected to memory” in claim 12, merely uses a computer device as a tool to perform the abstract idea. (MPEP 2106.05(f)) The limitations of “obtaining for each of one or more modules of a gas turbine engine, a respective input data set indicative of one or more contributors to unbalance for one or more of a plurality of stages of the module” and “obtain, based on the respective input data set for the module, a set of optimized clock angles for arranging the stages of the module relative to each other to mitigate vibration of the gas turbine engine” in claim 1, and “obtain for each of one or more modules of a gas turbine engine, a respective input data set indicative of one or more contributors to unbalance for one or more of a plurality of stages of the module” and “obtain, based on the respective input data set for the module, a set of optimized clock angles for arrangement of the stages of the module relative to each other to mitigate vibration of the gas turbine engine” in claim 12 can be viewed as is insignificant extra-solution activity, specifically pertaining to mere data gathering/output necessary to perform the abstract idea (MPEP 2106.05(g)) and is not sufficient to integrate the judicial exception into a practical application. This is akin to selecting information, based on types of information and availability of information in a manufacturing and industrial engineering environment, for collection, analysis and display, which has been identified as extra solution activity. The additional limitation of “for each of the one or more modules, utilizing one or more first neural networks associated with the module” in claim 1 and “for each of the one or more modules, utilize one or more first neural networks associated with the module” in claim 12 are mere instructions to implement an abstract idea using a computer in its ordinary capacity, or merely uses the computer as a tool to perform the identified abstract idea. See MPEP (2106.05(f)) Use of a computer or other machinery in its ordinary capacity for economic or other tasks (e.g., to receive, store, or transmit data) or simply adding a general purpose computer or computer components after the fact to an abstract idea (e.g., a mental process) does not integrate a judicial exception into a practical application. (MPEP 2106.05(f)(2)) Therefore, the judicial exception is not integrated into a practical application.
Step 2B: Claims 1 and 12 do not include additional elements that are sufficient to amount to significantly more than the judicial exception. In Claims 1 and 12, the additional element of “one or more first neural networks”, and “processing circuitry operatively connected to memory” in claim 12, merely uses a computer device as a tool to perform the abstract idea. (MPEP 2106.05(f)) The limitations of “obtaining for each of one or more modules of a gas turbine engine, a respective input data set indicative of one or more contributors to unbalance for one or more of a plurality of stages of the module” and “obtain, based on the respective input data set for the module, a set of optimized clock angles for arranging the stages of the module relative to each other to mitigate vibration of the gas turbine engine” in claim 1, and “obtain for each of one or more modules of a gas turbine engine, a respective input data set indicative of one or more contributors to unbalance for one or more of a plurality of stages of the module” and “obtain, based on the respective input data set for the module, a set of optimized clock angles for arrangement of the stages of the module relative to each other to mitigate vibration of the gas turbine engine” in claim 12 can be viewed as is insignificant extra-solution activity, specifically pertaining to mere data gathering/output necessary to perform the abstract idea (MPEP 2106.05(g)) and is not sufficient to integrate the judicial exception into a practical application. This is akin to selecting information, based on types of information and availability of information in a manufacturing and industrial engineering environment, for collection, analysis and display, which has been identified as extra solution activity. The additional limitation of “for each of the one or more modules, utilizing one or more first neural networks associated with the module” in claim 1 and “for each of the one or more modules, utilize one or more first neural networks associated with the module” in claim 12 are mere instructions to implement an abstract idea using a computer in its ordinary capacity, or merely uses the computer as a tool to perform the identified abstract idea. See MPEP (2106.05(f)) Use of a computer or other machinery in its ordinary capacity for economic or other tasks (e.g., to receive, store, or transmit data) or simply adding a general purpose computer or computer components after the fact to an abstract idea (e.g., a mental process) does not integrate a judicial exception into a practical application. (MPEP 2106.05(f)(2)) Therefore, the claim as a whole does not include additional elements that are sufficient to amount to significantly more than the judicial exception because the additional elements, when considered alone or in combination, do not amount to significantly more than the judicial exception. As stated in Section I.B. of the December 16, 2014 101 Examination Guidelines, “[t]o be patent-eligible, a claim that is directed to a judicial exception must include additional features to ensure that the claim describes a process or product that applies the exception in a meaningful way, such that it is more than a drafting effort designed to monopolize the exception.”
The dependent claims include the same abstract ideas recited as recited in the independent claims, and merely incorporate additional details that narrow the abstract ideas and fail to add significantly more to the claims.
Dependent claims 2 and 13 are directed to further defining the one or more contributors to unbalance, which further narrows the abstract idea identified in the independent claim, which is directed to “Mental Processes” or alternatively “Mathematical Concepts.”
Dependent claims 3 and 14 are directed to using a neural network to find the optimal clock angles for a module relative to a second module, and defining the input parameters the neural network is trained with, which further narrows the abstract idea identified in the independent claim, which is directed to “Mere Instructions to Apply an Exception” (MPEP 2106.05(f)) and “Mathematical Concepts,” or alternatively “Mental Processes.”
Dependent claims 5 and 15 are directed to using a second neural network to determine at least one trim weight angle and a third neural network to determine at least one trim weight magnitude, defining the arranging of the first and second module relative to each other, and further defining the training of the two neural networks, which further narrows the abstract idea identified in the independent claim, which is directed to “Mere Instructions to Apply an Exception” (MPEP 2106.05(f)) and “Mathematical Concepts,” or alternatively “Mental Processes.”
Dependent claims 6 and 16 are directed to using one of the rotor dynamics models to perform a simulation for the training data to determine predictions related to the vibration of the gas turbine engine, using a reward function and performance metric for the training data, and using an optimization algorithm to update weights in at least one neural networks to improve the performance of the neural networks, which further narrows the abstract idea identified in the independent claim, which is directed to “Mere Instructions to Apply an Exception” (MPEP 2106.05(f)) and “Mathematical Concepts,” or alternatively “Mental Processes.”
Dependent claims 7-8 and 17-18 are directed to further defining the relation to the neural networks and the specific modules, which further narrows the abstract idea identified in the independent claim, which is directed to “Mere Instructions to Apply an Exception” (MPEP 2106.05(f)) and “Mathematical Concepts,” or alternatively “Mental Processes.”
Dependent claims 9-11 and 19-20 are directed to further defining the first and second module, which further narrows the abstract idea identified in the independent claim, which is directed to “Mental Processes.”
Accordingly, claims 1-3 and 5-20 are rejected under 35 U.S.C. 101 because the claimed invention is directed to a judicial exception (i.e. an abstract idea) without anything significantly more.
Claim Rejections - 35 USC § 112
The following is a quotation of 35 U.S.C. 112(b):
(b) CONCLUSION.—The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention.
The following is a quotation of 35 U.S.C. 112 (pre-AIA ), second paragraph:
The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.
Claims 9-11 and 19-20 are rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), second paragraph, as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor (or for applications subject to pre-AIA 35 U.S.C. 112, the applicant), regards as the invention.
The terms “high pressure” in claims 9, 11 and 19 and “low pressure” in claims 10, 11 and 20 are relative terms which render the claim indefinite. The terms “high pressure” and “low pressure” are not defined by the claim, the specification does not provide a standard for ascertaining the requisite degree, and one of ordinary skill in the art would not be reasonably apprised of the scope of the invention. The terms “high pressure” and “low pressure” lack any numerical values or specific numerical ranges in the specification to establish a clear boundary. Appropriate correction is required.
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
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.
Claims 1-5, 7-15, and 17-20 are rejected under 35 U.S.C 103 as being unpatentable over U.S. Patent Publication 2021/0102465 A1, hereafter R, in view of U.S. Patent Publication 2023/0012799 A1, hereafter A.
Regarding Claim 1: R discloses a method of optimizing the assembly of rotating hardware of a gas turbine engine to mitigate vibration, comprising:
obtaining for each of one or more modules of a gas turbine engine, a respective input data set indicative of one or more contributors to unbalance for one or more of a plurality of stages of the module;
R [0025] “This type of configuration of centers of masses along the stack or sequence of rotor components 130a, 130b, 130c, 130d, 130e will be referred to herein as a “bow” shape, and is characterized by the fact that in the plane of the main amplitude of center of mass offset (the Y-Z plane illustrated in FIGS. 4A and 4B again in this example and the top view of FIG. 5A in the corresponding example), the amplitude of the offset of the centers of mass, or runout, increases, and then decreases along the axial sequence, reaching its maximum amplitude in an intermediate one of the rotor components of the stack, such as the midshaft in the specific examples illustrated.”
and for each of the one or more modules, based on the respective input data set for the module, a set of optimized clock angles for arranging the stages of the module relative to each other to mitigate vibration of the gas turbine engine;
R [0018] “Indeed, the wedge angle α and the wedge angle β of a receiving rotor component can be additive or subtractive, depending on the relative circumferential orientations. That is, if the receiving rotor component has a wedge angle β which tends to offset the center of mass of the received rotor component upwardly, this offset can be amplified, or to the contrary, partially, fully, or over-compensated, depending on the amplitude of the wedge angle α of the received component, and the relative circumferential orientations between the two assembled components. The wedge angles α, β and the offset h can, be accurately measured, and the orientation of the wedge angles of the received rotor component can be changed, in the final assembly, by rotating, or “clocking”, the received rotor component 30 around its geometrical axis 42, to a circumferential position determined to achieve this compensating effect, relative to the circumferential orientation of the receiving component and its mating wedge angle β, before assembling it to the receiving component. In any event, it will be understood that even if it is aimed to fully cancel out the offset between the center of mass 40 of the received rotor component 30 and the virtual rotation axis 11 of the overall rotor assembly (e.g. 20 or 22), this cancelling out will, in practice, be imperfect and there will remain an offset, or runout, between the received rotor component 30 and the rotation axis 11 of the rotor assembly.”
R [0027] “Indeed, using dynamic response computer model analysis, the optimal unbalance alignment can be to align the compressor rotor and the turbine rotors residual unbalance in-phase and to stack the rotor parts by aligning the individual component such as the unbalance is additive, creating a ‘bowed-shape” spool and resulting in a “inphase unbalance” for the rotor spool end-to-end. (FIG. 5)”
R [0025] “Indeed, it was found that at least in the case of some aircraft engines, at full operating speeds, this couple unbalance can result in very high vibrations resulting in the engine not passing the vibration acceptance tests.” Examiner notes that if the vibration acceptance test fails, the engine must be disassembled, rebalanced, and reassembled (See R [0002]).
R [0037] “Moreover, the rotor assembly can be rotated at operating speeds on a test bed, and vibrations can be measured and checked against design tolerances or threshold levels, for instance.”
and the set of clock angles;
R [0018] “Indeed, the wedge angle α and the wedge angle β of a receiving rotor component can be additive or subtractive, depending on the relative circumferential orientations. That is, if the receiving rotor component has a wedge angle β which tends to offset the center of mass of the received rotor component upwardly, this offset can be amplified, or to the contrary, partially, fully, or over-compensated, depending on the amplitude of the wedge angle α of the received component, and the relative circumferential orientations between the two assembled components. The wedge angles α, β and the offset h can, be accurately measured, and the orientation of the wedge angles of the received rotor component can be changed, in the final assembly, by rotating, or “clocking”, the received rotor component 30 around its geometrical axis 42, to a circumferential position determined to achieve this compensating effect, relative to the circumferential orientation of the receiving component and its mating wedge angle β, before assembling it to the receiving component. In any event, it will be understood that even if it is aimed to fully cancel out the offset between the center of mass 40 of the received rotor component 30 and the virtual rotation axis 11 of the overall rotor assembly (e.g. 20 or 22), this cancelling out will, in practice, be imperfect and there will remain an offset, or runout, between the received rotor component 30 and the rotation axis 11 of the rotor assembly.”
R does not disclose utilizing one or more first neural networks associated with the module, wherein each of the one or more first neural networks has been trained with: training data comprising the one or more contributors to unbalance, and one or more rotor dynamics models that use the training data … from the one or more first neural networks to predict vibration at one or more locations of interest in the gas turbine engine.
However, A discloses utilizing one or more first neural networks associated with the module,
A [0045] “With reference now to FIGS. 2, 3, 4, and 5, FIG. 5 provides an example block diagram of the one or more machine-learned models 132. As depicted best in FIG. 5, the data 140 is input into the one or more machine-learned models 132. The one or more machine-learned models 132 can be structured as any suitable type of machine-learned model. By way of example, the one or more machine-learned models 132 can be or can include a machine or statistical learning model structured as one of a non-linear least squares optimization model, a linear discriminant analysis model, a partial least squares discriminant analysis model, a support vector machine model, a random tree model, a logistic regression model, a naïve Bayes model, a K-nearest neighbor model, a quadratic discriminant analysis model, an anomaly detection model, a boosted and bagged decision tree model, an artificial neural network model, a C4.5 model, a k-means model, or a combination of one or more of the foregoing.
wherein each of the one or more first neural networks has been trained with: training data comprising the one or more contributors to unbalance,
A [0043-0044] “The assembly parameters 142 can include the residual unbalance of the rotating system 112, rotor concentricity of the rotating system 112, etc. … Stated another way, by applying the one or more machine-learned models 132 to the parameter values specifically associated with the gas turbine engine 110 or rotating system 112, a customized transfer function specifically tailored to the rotating system 112 can be generated. The self-learning transfer function is generated so that an optimal balance shot 150 can be generated to reduce the vibration response of the engine.”
and one or more rotor dynamics models that use the training data … from the one or more first neural networks to predict vibration at one or more locations of interest in the gas turbine engine.
A [0044] “Stated another way, by applying the one or more machine-learned models 132 to the parameter values specifically associated with the gas turbine engine 110 or rotating system 112, a customized transfer function specifically tailored to the rotating system 112 can be generated. The self-learning transfer function is generated so that an optimal balance shot 150 can be generated to reduce the vibration response of the engine.”
A [0033] “The engine operating data of the data 140 can include parameter values for one or more engine operating parameters 141. Example engine operating parameters 141 for which parameter values can be sensed other otherwise calculated based on such sensed parameter values can include, without limitation … parameters associated with vibration of the engine, including a vibration response of the engine, phase angle, mode position of one or more rotors or spools, and/or vibration variations”
R and A are analogous to the claimed invention because they both pertain to balancing rotor components to mitigate vibrations in a gas turbine engine.
It would have been obvious to one with ordinary skill in the art before the effective filing date to combine the teachings of A with R because the system and method of R generates a balance shot that is optimized specifically for a rotating system, which reduces the vibration response of the rotating system, with the use of machine learning models. (See A [Abstract]).
Regarding Claim 2: R in view of A discloses the method of claim 1, wherein for each stage of the one or more modules, the one or more contributors to unbalance include at least one of: a radial offset for at least one of the plurality of stages; a squareness error for at least one of the plurality of stages; or a residual unbalance due to an inherent mass offset for at least one of the plurality of stages.
R [0025] “This type of configuration of centers of masses along the stack or sequence of rotor components 130a, 130b, 130c, 130d, 130e will be referred to herein as a “bow” shape, and is characterized by the fact that in the plane of the main amplitude of center of mass offset (the Y-Z plane illustrated in FIGS. 4A and 4B again in this example and the top view of FIG. 5A in the corresponding example), the amplitude of the offset of the centers of mass, or runout, increases, and then decreases along the axial sequence, reaching its maximum amplitude in an intermediate one of the rotor components of the stack, such as the midshaft in the specific examples illustrated.”
Regarding Claim 3: R in view of A discloses the method of claim 1, wherein: the one or more modules includes a first module and a second module; and the method comprises:
… determine an optimized inter-module clock angle for arranging the second module relative to the first module to mitigate vibration of the gas turbine engine… the sets of clock angles, and the inter-module clock angle.
R [0023] “In the example shown in FIGS. 3A, 3B and 3C, the rotor assembly 120 consists of a stack including a #1 bearing 130a, an impeller 130b, a midshaft 130c, a two-stage turbine 130d, and a #2 bearing 130e. The circumferential orientation of the impeller 130b relative to the #1 bearing 130a is selected in a manner for the center of mass 140b of the impeller 130b to be offset by 0.002 inches along the Y axis relative to the rotor assembly's rotation axis 111. The circumferential orientation of the successive components is selected in a manner for the center of mass 140c of the midshaft 130c to coincide with the rotation axis 111 of the assembly 120, within measuring tolerances, and for the center of mass 140d of the turbine 130d to have a negative, balancing offset, in this case of −0.002 inches along the Y axis, essentially compensating for the offset of the center of mass 140b of the impeller 130b. The resulting rotor assembly 120 has a total center of mass which is aligned with the axis 111 of the assembly 120, within measuring tolerances. This type of configuration of centers of masses 140b, 140c, 140d, along the stack or sequence of rotor components 130a, 130b, 130c, 130d, 130e will be referred to herein as a zig-zag or corkscrew configuration, and is characterized by the fact that in the plane of the main amplitude of center of mass offset (which is the illustrated Y-Z plane in this example, assuming that the centers of mass are much more closely located relative to the rotation axis in the X-Y plane), the direction of the offset of the centers of mass, or runout, alternates from one side of the rotation axis 111 to the other along the stack, when plotted in a graph such as FIG. 3B, leading to a relatively small runout when considering the rotor assembly 120 in its entirety.” Examiner notes that the inter-module clock angle is represented through the positioning method used to construct the zigzag or corkscrew configurations, and the circumferential orientation was used to align individual module mass offsets.
R [0018] “Indeed, the wedge angle α and the wedge angle β of a receiving rotor component can be additive or subtractive, depending on the relative circumferential orientations. That is, if the receiving rotor component has a wedge angle β which tends to offset the center of mass of the received rotor component upwardly, this offset can be amplified, or to the contrary, partially, fully, or over-compensated, depending on the amplitude of the wedge angle α of the received component, and the relative circumferential orientations between the two assembled components. The wedge angles α, β and the offset h can, be accurately measured, and the orientation of the wedge angles of the received rotor component can be changed, in the final assembly, by rotating, or “clocking”, the received rotor component 30 around its geometrical axis 42, to a circumferential position determined to achieve this compensating effect, relative to the circumferential orientation of the receiving component and its mating wedge angle β, before assembling it to the receiving component. In any event, it will be understood that even if it is aimed to fully cancel out the offset between the center of mass 40 of the received rotor component 30 and the virtual rotation axis 11 of the overall rotor assembly (e.g. 20 or 22), this cancelling out will, in practice, be imperfect and there will remain an offset, or runout, between the received rotor component 30 and the rotation axis 11 of the rotor assembly.”
R [0025] “Indeed, it was found that at least in the case of some aircraft engines, at full operating speeds, this couple unbalance can result in very high vibrations resulting in the engine not passing the vibration acceptance tests.” Examiner notes that if the vibration acceptance test fails, the engine must be disassembled, rebalanced, and reassembled (See R [0002]).
R [0037] “Moreover, the rotor assembly can be rotated at operating speeds on a test bed, and vibrations can be measured and checked against design tolerances or threshold levels, for instance.”
R does not disclose utilizing a second neural network, or wherein the second neural network has also been trained with one of the one or more rotor dynamics models, which uses the training data, … to predict vibration at one or more locations of interest in the gas turbine engine.
However, A discloses utilizing a second neural network, or wherein the second neural network has also been trained with one of the one or more rotor dynamics models, which uses the training data, … to predict vibration at one or more locations of interest in the gas turbine engine.
A [0045] “With reference now to FIGS. 2, 3, 4, and 5, FIG. 5 provides an example block diagram of the one or more machine-learned models 132. As depicted best in FIG. 5, the data 140 is input into the one or more machine-learned models 132. The one or more machine-learned models 132 can be structured as any suitable type of machine-learned model. By way of example, the one or more machine-learned models 132 can be or can include a machine or statistical learning model structured as one of a non-linear least squares optimization model, a linear discriminant analysis model, a partial least squares discriminant analysis model, a support vector machine model, a random tree model, a logistic regression model, a naïve Bayes model, a K-nearest neighbor model, a quadratic discriminant analysis model, an anomaly detection model, a boosted and bagged decision tree model, an artificial neural network model, a C4.5 model, a k-means model, or a combination of one or more of the foregoing.
A [0033] “The engine operating data of the data 140 can include parameter values for one or more engine operating parameters 141. Example engine operating parameters 141 for which parameter values can be sensed other otherwise calculated based on such sensed parameter values can include, without limitation … parameters associated with vibration of the engine, including a vibration response of the engine, phase angle, mode position of one or more rotors or spools, and/or vibration variations”
A [0038] “As noted above, the computing system 120 can generate a balance shot 150 that provides an optimized balancing solution to minimize and/or reduce the vibration response of the rotating system 112. The balance shot 150 can indicate one or more physical locations at which one or more balancing weights are to be added or removed from the rotating system 112. As will be explained further below, the balance shot 150 is generated using an optimized engine specific non-linear self-learning transfer function that is generated by applying one or more machine-learned models on the received data 140. The self-learning transfer function accounts for a variety of parameters, such as engine operating parameters 141, including a rotor vibration response and mode, oil properties, assembly parameters 142, usage parameters 143, environmental parameters 144, life cycle parameters 145, and/or other parameters associated with the gas turbine engine 110 and/or the rotating system 112.”
R and A are analogous to the claimed invention because they both pertain to balancing rotor components to mitigate vibrations in a gas turbine engine.
It would have been obvious to one with ordinary skill in the art before the effective filing date to combine the teachings of A with R because the system and method of R generates a balance shot that is optimized specifically for a rotating system, which reduces the vibration response of the rotating system, with the use of machine learning models. (See A [Abstract]).
Regarding Claim 4: R in view of A disclose the method of claim 3, comprising: assembling the
R [0028] “Accordingly, and more generally, for assembling a plurality of rotor components (typically at least three) into a rotor assembly, one can begin by obtaining geometrical data about the individual rotor components. This can be achieved by measuring the wedge angles at the mating members and determining intrinsic runout (center of mass offset) for each component of the stack. Then, a computer can simulate the different possible combinations of circumferential orientations for the different rotor components in the stack, taking into consideration the geometrical data obtained about the specific set of rotor components to assemble. This simulation can produce a number of different solutions, in which the solutions are each characterized by a given runout configuration. Some of these runout configurations can be corkscrew shaped, for instance, whereas others can be bow shaped. A combination of relative circumferential orientations leading to a bow shape can be selected, and more specifically, if more than one bow shape solution is available, one of these which is considered as producing the lowest overall runout can be selected, and the rotor components can then be assembled to one another according to this selected solution, by “clocking” the individual rotor components to corresponding circumferential orientations, in a manner for the rotor assembly formed by the resulting axially-sequenced stack to correspond to the selected solution.
R does not explicitly disclose assembling and arranging the full modules.
However, A discloses assembling and arranging the full modules.
A [0019] “The turbofan 10 includes a fan section 14 and a core engine 16 disposed downstream of the fan section 14. The core engine 16 includes a substantially tubular engine cowl 18 that defines an annular core inlet 20. As schematically shown in FIG. 1, the engine cowl 18 encases, in serial flow relationship, a compressor section including a booster or low pressure (LP) compressor 22 followed downstream by a high pressure (HP) compressor 24; a combustion section 26; a turbine section including an HP turbine 28 followed downstream by an LP turbine 30; and a jet exhaust nozzle section 32. The compressor section, combustion section 26, turbine section, and nozzle section 32 together define a core air flowpath. An HP shaft 34 drivingly connects the HP turbine 28 to the HP compressor 24 to rotate them in unison concentrically with respect to the longitudinal centerline 12. An LP shaft 36 drivingly connects the LP turbine 30 to the LP compressor 22 to rotate them in unison concentrically with respect to the longitudinal centerline 12. The HP and LP shafts 34, 36 are each rotating components, rotating about the axial direction A during operation of the turbofan 10. The turbofan 10 can include a plurality of bearings to support such rotating components.”
A [0020] “The HP shaft 34, the rotating components of the HP compressor 24, and the rotating components of the HP turbine 28 collectively form a first rotating system or HP spool. The LP shaft 36, the rotating components of the LP compressor 22, and the rotating components of the LP turbine 30 collectively form a second rotating system or LP spool. As will be explained herein, rotating systems, such as the LP and/or HP spools, can undergo one or more balancing operations to reduce the vibration response thereof.” Examiner notes that the full modules listed here comprise the high pressure and low pressure compressor, turbine and rotating spool modules, the combustion module, and the exhaust nozzle.
R and A are analogous to the claimed invention because they both pertain to balancing rotor components to mitigate vibrations in a gas turbine engine.
It would have been obvious to one with ordinary skill in the art before the effective filing date to combine the teachings of A with R because the system and method of R generates a balance shot that is optimized specifically for a rotating system, which reduces the vibration response of the rotating system, with the use of machine learning models. (See A [Abstract]).
Regarding Claim 5: R in view of A disclose the method of claim 3, wherein: the method includes, for each of the one or more modules: the set of clock angles, and the inter-module clock angles to predict vibration at one or more locations of interest in the gas turbine engine.
R [0028] “Accordingly, and more generally, for assembling a plurality of rotor components (typically at least three) into a rotor assembly, one can begin by obtaining geometrical data about the individual rotor components. This can be achieved by measuring the wedge angles at the mating members and determining intrinsic runout (center of mass offset) for each component of the stack. Then, a computer can simulate the different possible combinations of circumferential orientations for the different rotor components in the stack, taking into consideration the geometrical data obtained about the specific set of rotor components to assemble. This simulation can produce a number of different solutions, in which the solutions are each characterized by a given runout configuration. Some of these runout configurations can be corkscrew shaped, for instance, whereas others can be bow shaped. A combination of relative circumferential orientations leading to a bow shape can be selected, and more specifically, if more than one bow shape solution is available, one of these which is considered as producing the lowest overall runout can be selected, and the rotor components can then be assembled to one another according to this selected solution, by “clocking” the individual rotor components to corresponding circumferential orientations, in a manner for the rotor assembly formed by the resulting axially-sequenced stack to correspond to the selected solution.
[0037] Subsequently to the assembly of the rotor assembly with the clocked rotor components, the runout at different planes corresponding to individual ones of the rotor components, for instance, can be measured. Moreover, the rotor assembly can be rotated at operating speeds on a test bed, and vibrations can be measured and checked against design tolerances or threshold levels, for instance.
R does not disclose utilizing the second neural network to determine at least one trim weight angle, utilizing a third neural network to determine at least one trim weight magnitude, wherein said arranging the first module and second module relative to each other includes adding one or more trim weights to the first module or second module that use one of the determined trim weight magnitudes and one of the determined trim weight angles; and wherein the third neural network has also been trained with one of the one or more rotor dynamics models, which uses the training data, to predict vibration at one or more locations of interest in the gas turbine engine or wherein the second and third neural networks have also been trained with the at least one trim weight angle and the at least one trim weight magnitude.
However, A discloses utilizing the second neural network to determine at least one trim weight angle,
A [0045] “With reference now to FIGS. 2, 3, 4, and 5, FIG. 5 provides an example block diagram of the one or more machine-learned models 132. As depicted best in FIG. 5, the data 140 is input into the one or more machine-learned models 132. The one or more machine-learned models 132 can be structured as any suitable type of machine-learned model. By way of example, the one or more machine-learned models 132 can be or can include a machine or statistical learning model structured as one of a non-linear least squares optimization model, a linear discriminant analysis model, a partial least squares discriminant analysis model, a support vector machine model, a random tree model, a logistic regression model, a naïve Bayes model, a K-nearest neighbor model, a quadratic discriminant analysis model, an anomaly detection model, a boosted and bagged decision tree model, an artificial neural network model, a C4.5 model, a k-means model, or a combination of one or more of the foregoing.
A [0038] “As noted above, the computing system 120 can generate a balance shot 150 that provides an optimized balancing solution to minimize and/or reduce the vibration response of the rotating system 112. The balance shot 150 can indicate one or more physical locations at which one or more balancing weights are to be added or removed from the rotating system 112. As will be explained further below, the balance shot 150 is generated using an optimized engine specific non-linear self-learning transfer function that is generated by applying one or more machine-learned models on the received data 140. The self-learning transfer function accounts for a variety of parameters, such as engine operating parameters 141, including a rotor vibration response and mode, oil properties, assembly parameters 142, usage parameters 143, environmental parameters 144, life cycle parameters 145, and/or other parameters associated with the gas turbine engine 110 and/or the rotating system 112.” Examiner notes that one with ordinary skill in the art would understand that in the context of balancing a rotating system, a ‘physical location’ on a rotor where a balancing/trim weight is applied is defined by its angular position relative to a reference point on the rotor. The citation below discloses the phase angle, which is the angular position relative to a reference point on the rotor of the location of unbalance. Therefore, one with ordinary skill in the art could apply the same reference point to the trim weight angle.
A [0033] “The engine operating data of the data 140 can include parameter values for one or more engine operating parameters 141. Example engine operating parameters 141 for which parameter values can be sensed other otherwise calculated based on such sensed parameter values can include, without limitation ... parameters associated with vibration of the engine, including a vibration response of the engine, phase angle, mode position of one or more rotors or spools, and/or vibration variations.
utilizing a third neural network to determine at least one trim weight magnitude,
A [0027] “A balance shot, also referred herein as a balance solution, is a balancing plan that indicates one or more physical locations at which one or more balancing weights are to be added or removed from a rotating system so as to minimize and/or reduce the vibration response of the rotating system. The balance shot can indicate the mass and position of each of the balancing weights.”
wherein said arranging the first module and second module relative to each other includes adding one or more trim weights to the first module or second module that use one of the determined trim weight magnitudes and one of the determined trim weight angles;
A [0027] “A balance shot, also referred herein as a balance solution, is a balancing plan that indicates one or more physical locations at which one or more balancing weights are to be added or removed from a rotating system so as to minimize and/or reduce the vibration response of the rotating system. The balance shot can indicate the mass and position of each of the balancing weights.”
A [0038] “As noted above, the computing system 120 can generate a balance shot 150 that provides an optimized balancing solution to minimize and/or reduce the vibration response of the rotating system 112. The balance shot 150 can indicate one or more physical locations at which one or more balancing weights are to be added or removed from the rotating system 112. As will be explained further below, the balance shot 150 is generated using an optimized engine specific non-linear self-learning transfer function that is generated by applying one or more machine-learned models on the received data 140. The self-learning transfer function accounts for a variety of parameters, such as engine operating parameters 141, including a rotor vibration response and mode, oil properties, assembly parameters 142, usage parameters 143, environmental parameters 144, life cycle parameters 145, and/or other parameters associated with the gas turbine engine 110 and/or the rotating system 112.”
A [0034] “As noted above, the data 140 can also include assembly data. The assembly data can include parameter values for assembly parameters 142 associated with assembly of the rotating system 112 of the gas turbine engine 110. Example assembly parameters 142 can include, without limitation, the assembly method used to assemble the rotating system and/or the gas turbine engine 110, the operators who assembled the rotating system and/or the gas turbine engine 110, the assembly line and/or assembly facility along which or at which the rotating system and/or the gas turbine engine 110 was assembled, and/or one or more parts or components included within the rotating system 112 that were provided by an entity other than the entity that manufactured the rotating system.”
and wherein the third neural network has also been trained with one of the one or more rotor dynamics models, which uses the training data, to predict vibration at one or more locations of interest in the gas turbine engine;
wherein the second and third neural networks have also been trained with the at least one trim weight angle and the at least one trim weight magnitude.
A [0038] “As noted above, the computing system 120 can generate a balance shot 150 that provides an optimized balancing solution to minimize and/or reduce the vibration response of the rotating system 112. The balance shot 150 can indicate one or more physical locations at which one or more balancing weights are to be added or removed from the rotating system 112. As will be explained further below, the balance shot 150 is generated using an optimized engine specific non-linear self-learning transfer function that is generated by applying one or more machine-learned models on the received data 140. The self-learning transfer function accounts for a variety of parameters, such as engine operating parameters 141, including a rotor vibration response and mode, oil properties, assembly parameters 142, usage parameters 143, environmental parameters 144, life cycle parameters 145, and/or other parameters associated with the gas turbine engine 110 and/or the rotating system 112.” Examiner notes that one with ordinary skill in the art would understand that in the context of balancing a rotating system, a ‘physical location’ on a rotor where a balancing/trim weight is applied is defined by its angular position relative to a reference point on the rotor. The citation below discloses the phase angle, which is the angular position relative to a reference point on the rotor of the location of unbalance. Therefore, one with ordinary skill in the art could apply the same reference point to the trim weight angle.
A [0033] “The engine operating data of the data 140 can include parameter values for one or more engine operating parameters 141. Example engine operating parameters 141 for which parameter values can be sensed other otherwise calculated based on such sensed parameter values can include, without limitation ... parameters associated with vibration of the engine, including a vibration response of the engine, phase angle, mode position of one or more rotors or spools, and/or vibration variations.
R and A are analogous to the claimed invention because they both pertain to balancing rotor components to mitigate vibrations in a gas turbine engine.
It would have been obvious to one with ordinary skill in the art before the effective filing date to combine the teachings of A with R because the system and method of R generates a balance shot that is optimized specifically for a rotating system, which reduces the vibration response of the rotating system, with the use of machine learning models. (See A [Abstract]).
Regarding Claim 7: R, in view of A discloses the method of claim 3.
R does not disclose wherein the one or more first neural networks include a neural network associated with the first module and a separate second neural network associated with the second module.
However, A discloses wherein the one or more first neural networks include a neural network associated with the first module and a separate second neural network associated with the second module.
A [0019] “ As schematically shown in FIG. 1, the engine cowl 18 encases, in serial flow relationship, a compressor section including a booster or low pressure (LP) compressor 22 followed downstream by a high pressure (HP) compressor 24; a combustion section 26; a turbine section including an HP turbine 28 followed downstream by an LP turbine 30; and a jet exhaust nozzle section 32. The compressor section, combustion section 26, turbine section, and nozzle section 32 together define a core air flowpath.”
A [0033] “The engine operating data of the data 140 can include parameter values for one or more engine operating parameters 141. Example engine operating parameters 141 for which parameter values can be sensed other otherwise calculated based on such sensed parameter values can include, without limitation, rotor or spool speed, including the low pressure spool speed N1 and/or the high pressure or core spool speed N2, the oil temperature and/or oil pressure, parameters associated with vibration of the engine, including a vibration response of the engine, phase angle, mode position of one or more rotors or spools, and/or vibration variations, the exhaust gas temperature of the engine, cold clearances between rotating and stationary components, parameters associated with acceleration and deceleration of the engine, including the number of accelerations and/or decelerations over a magnitude threshold and/or rate of acceleration and/or deceleration, the compressor inlet pressure and temperature, the compressor discharge pressure, and/or the temperature at the inlet or outlet of the combustor. The engine operating data can include parameter values for other engine operating parameters 141 as well.” Examiner notes that each parameter of the engine operating data is associated to a module of the gas turbine engine.
A [0038] “The self-learning transfer function accounts for a variety of parameters, such as engine operating parameters 141, including a rotor vibration response and mode, oil properties, assembly parameters 142, usage parameters 143, environmental parameters 144, life cycle parameters 145, and/or other parameters associated with the gas turbine engine 110 and/or the rotating system 112.”
A [0044] “At (310), the method (300) includes generating a self-learning transfer function. For instance, the one or more processors 124 of the computing system 120 can generate the self-learning transfer function by applying the one or more machine-learned models 132 to the parameter values received as part of the data 140. The self-learning transfer function can be generated such that it is specific to the rotating system 112. Stated another way, by applying the one or more machine-learned models 132 to the parameter values specifically associated with the gas turbine engine 110 or rotating system 112, a customized transfer function specifically tailored to the rotating system 112 can be generated. The self-learning transfer function is generated so that an optimal balance shot 150 can be generated to reduce the vibration response of the engine.” Examiner notes that the customizable nature of a self-learning transfer function makes it such that machine-learned models can be applied separately to individual engine modules or combined for multiple modules possible by tailoring inputs and outputs to specific rotating components or the entire system.
R and A are analogous to the claimed invention because they both pertain to balancing rotor components to mitigate vibrations in a gas turbine engine.
It would have been obvious to one with ordinary skill in the art before the effective filing date to combine the teachings of A with R because the system and method of R generates a balance shot that is optimized specifically for a rotating system, which reduces the vibration response of the rotating system, with the use of machine learning models. (See A [Abstract]).
Regarding Claim 8: R, in view of A discloses the method of claim 3.
R does not disclose wherein the one or more first neural networks include a neural network associated with both of the first module and the second module.
However, A discloses wherein the one or more first neural networks include a neural network associated with both of the first module and the second module.
A [0019] “ As schematically shown in FIG. 1, the engine cowl 18 encases, in serial flow relationship, a compressor section including a booster or low pressure (LP) compressor 22 followed downstream by a high pressure (HP) compressor 24; a combustion section 26; a turbine section including an HP turbine 28 followed downstream by an LP turbine 30; and a jet exhaust nozzle section 32. The compressor section, combustion section 26, turbine section, and nozzle section 32 together define a core air flowpath.”
A [0033] “The engine operating data of the data 140 can include parameter values for one or more engine operating parameters 141. Example engine operating parameters 141 for which parameter values can be sensed other otherwise calculated based on such sensed parameter values can include, without limitation, rotor or spool speed, including the low pressure spool speed N1 and/or the high pressure or core spool speed N2, the oil temperature and/or oil pressure, parameters associated with vibration of the engine, including a vibration response of the engine, phase angle, mode position of one or more rotors or spools, and/or vibration variations, the exhaust gas temperature of the engine, cold clearances between rotating and stationary components, parameters associated with acceleration and deceleration of the engine, including the number of accelerations and/or decelerations over a magnitude threshold and/or rate of acceleration and/or deceleration, the compressor inlet pressure and temperature, the compressor discharge pressure, and/or the temperature at the inlet or outlet of the combustor. The engine operating data can include parameter values for other engine operating parameters 141 as well.” Examiner notes that each parameter of the engine operating data is associated to a module of the gas turbine engine.
A [0038] “The self-learning transfer function accounts for a variety of parameters, such as engine operating parameters 141, including a rotor vibration response and mode, oil properties, assembly parameters 142, usage parameters 143, environmental parameters 144, life cycle parameters 145, and/or other parameters associated with the gas turbine engine 110 and/or the rotating system 112.”
A [0044] “At (310), the method (300) includes generating a self-learning transfer function. For instance, the one or more processors 124 of the computing system 120 can generate the self-learning transfer function by applying the one or more machine-learned models 132 to the parameter values received as part of the data 140. The self-learning transfer function can be generated such that it is specific to the rotating system 112. Stated another way, by applying the one or more machine-learned models 132 to the parameter values specifically associated with the gas turbine engine 110 or rotating system 112, a customized transfer function specifically tailored to the rotating system 112 can be generated. The self-learning transfer function is generated so that an optimal balance shot 150 can be generated to reduce the vibration response of the engine.” Examiner notes that the customizable nature of a self-learning transfer function makes it such that machine-learned models can be applied separately to individual engine modules or combined for multiple modules possible by tailoring inputs and outputs to specific rotating components or the entire system.
R and A are analogous to the claimed invention because they both pertain to balancing rotor components to mitigate vibrations in a gas turbine engine.
It would have been obvious to one with ordinary skill in the art before the effective filing date to combine the teachings of A with R because the system and method of R generates a balance shot that is optimized specifically for a rotating system, which reduces the vibration response of the rotating system, with the use of machine learning models. (See A [Abstract]).
Regarding Claim 9: R, in view of A discloses the method of claim 3.
R does not disclose wherein: the first module is a high pressure compressor of the gas turbine engine; and the second module is a high pressure turbine of the gas turbine engine.
However, A discloses wherein: the first module is a high pressure compressor of the gas turbine engine; and the second module is a high pressure turbine of the gas turbine engine.
A [0019] “a compressor section including … a high pressure (HP) compressor 24 … a turbine section including an HP turbine 28 … An HP shaft 34 drivingly connects the HP turbine 28 to the HP compressor 24 to rotate them in unison concentrically with respect to the longitudinal centerline 12.”
R and A are analogous to the claimed invention because they both pertain to balancing rotor components to mitigate vibrations in a gas turbine engine.
It would have been obvious to one with ordinary skill in the art before the effective filing date to combine the teachings of A with R because the system and method of R generates a balance shot that is optimized specifically for a rotating system, which reduces the vibration response of the rotating system, with the use of machine learning models. (See A [Abstract]).
Regarding Claim 10: R, in view of A discloses the method of claim 3.
R does not disclose wherein: the first module is a low pressure compressor of the gas turbine engine; and the second module is a low pressure turbine of the gas turbine engine.
However, A discloses wherein: the first module is a low pressure compressor of the gas turbine engine; and the second module is a low pressure turbine of the gas turbine engine.
A [0019] “a compressor section including a booster or low pressure (LP) compressor 22… a turbine section including … an LP turbine 30 … An LP shaft 36 drivingly connects the LP turbine 30 to the LP compressor 22 to rotate them in unison concentrically with respect to the longitudinal centerline 12.
R and A are analogous to the claimed invention because they both pertain to balancing rotor components to mitigate vibrations in a gas turbine engine.
It would have been obvious to one with ordinary skill in the art before the effective filing date to combine the teachings of A with R because the system and method of R generates a balance shot that is optimized specifically for a rotating system, which reduces the vibration response of the rotating system, with the use of machine learning models. (See A [Abstract]).
Regarding Claim 11: R, in view of A discloses the method of claim 3.
R does not disclose wherein: the method is performed where the first module is a high pressure compressor of the gas turbine engine the second module is a high pressure turbine of the gas turbine engine; and the method is separately performed where the first module is a low pressure compressor of the gas turbine engine and the second module is a low pressure turbine of the gas turbine engine.
However, A discloses wherein: the method is performed where the first module is a high pressure compressor of the gas turbine engine the second module is a high pressure turbine of the gas turbine engine; and the method is separately performed where the first module is a low pressure compressor of the gas turbine engine and the second module is a low pressure turbine of the gas turbine engine.
A [0019] “a compressor section including a booster or low pressure (LP) compressor 22… a turbine section including … an LP turbine 30 … An LP shaft 36 drivingly connects the LP turbine 30 to the LP compressor 22 to rotate them in unison concentrically with respect to the longitudinal centerline 12.
A [0019] “a compressor section including a booster or low pressure (LP) compressor 22… a turbine section including … an LP turbine 30 … An LP shaft 36 drivingly connects the LP turbine 30 to the LP compressor 22 to rotate them in unison concentrically with respect to the longitudinal centerline 12.”
A [0038] “The self-learning transfer function accounts for a variety of parameters, such as engine operating parameters 141, including a rotor vibration response and mode, oil properties, assembly parameters 142, usage parameters 143, environmental parameters 144, life cycle parameters 145, and/or other parameters associated with the gas turbine engine 110 and/or the rotating system 112.”
A [0044] “At (310), the method (300) includes generating a self-learning transfer function. For instance, the one or more processors 124 of the computing system 120 can generate the self-learning transfer function by applying the one or more machine-learned models 132 to the parameter values received as part of the data 140. The self-learning transfer function can be generated such that it is specific to the rotating system 112. Stated another way, by applying the one or more machine-learned models 132 to the parameter values specifically associated with the gas turbine engine 110 or rotating system 112, a customized transfer function specifically tailored to the rotating system 112 can be generated. The self-learning transfer function is generated so that an optimal balance shot 150 can be generated to reduce the vibration response of the engine.” Examiner notes that the customizable nature of a self-learning transfer function makes it such that machine-learned models can be applied separately to individual engine modules or combined for multiple modules possible by tailoring inputs and outputs to specific rotating components or the entire system.
R and A are analogous to the claimed invention because they both pertain to balancing rotor components to mitigate vibrations in a gas turbine engine.
It would have been obvious to one with ordinary skill in the art before the effective filing date to combine the teachings of A with R because the system and method of R generates a balance shot that is optimized specifically for a rotating system, which reduces the vibration response of the rotating system, with the use of machine learning models. (See A [Abstract]).
Regarding Claim 12: R discloses a system for optimizing the assembly of rotating hardware of a gas turbine engine to mitigate vibration, comprising:
obtain for each of one or more modules of a gas turbine engine, a respective input data set indicative of one or more contributors to unbalance for one or more of a plurality of stages of the module;
R [0025] “This type of configuration of centers of masses along the stack or sequence of rotor components 130a, 130b, 130c, 130d, 130e will be referred to herein as a “bow” shape, and is characterized by the fact that in the plane of the main amplitude of center of mass offset (the Y-Z plane illustrated in FIGS. 4A and 4B again in this example and the top view of FIG. 5A in the corresponding example), the amplitude of the offset of the centers of mass, or runout, increases, and then decreases along the axial sequence, reaching its maximum amplitude in an intermediate one of the rotor components of the stack, such as the midshaft in the specific examples illustrated.”
and for each of the one or more modules … based on the respective input data set for the module, a set of optimized clock angles for arranging the stages of the module relative to each other to mitigate vibration of the gas turbine engine;
R [0018] “Indeed, the wedge angle α and the wedge angle β of a receiving rotor component can be additive or subtractive, depending on the relative circumferential orientations. That is, if the receiving rotor component has a wedge angle β which tends to offset the center of mass of the received rotor component upwardly, this offset can be amplified, or to the contrary, partially, fully, or over-compensated, depending on the amplitude of the wedge angle α of the received component, and the relative circumferential orientations between the two assembled components. The wedge angles α, β and the offset h can, be accurately measured, and the orientation of the wedge angles of the received rotor component can be changed, in the final assembly, by rotating, or “clocking”, the received rotor component 30 around its geometrical axis 42, to a circumferential position determined to achieve this compensating effect, relative to the circumferential orientation of the receiving component and its mating wedge angle β, before assembling it to the receiving component. In any event, it will be understood that even if it is aimed to fully cancel out the offset between the center of mass 40 of the received rotor component 30 and the virtual rotation axis 11 of the overall rotor assembly (e.g. 20 or 22), this cancelling out will, in practice, be imperfect and there will remain an offset, or runout, between the received rotor component 30 and the rotation axis 11 of the rotor assembly.”
R [0027] “Indeed, using dynamic response computer model analysis, the optimal unbalance alignment can be to align the compressor rotor and the turbine rotors residual unbalance in-phase and to stack the rotor parts by aligning the individual component such as the unbalance is additive, creating a ‘bowed-shape” spool and resulting in a “inphase unbalance” for the rotor spool end-to-end. (FIG. 5)”
R [0025] “Indeed, it was found that at least in the case of some aircraft engines, at full operating speeds, this couple unbalance can result in very high vibrations resulting in the engine not passing the vibration acceptance tests.” Examiner notes that if the vibration acceptance test fails, the engine must be disassembled, rebalanced, and reassembled (See R [0002]).
R [0037] “Moreover, the rotor assembly can be rotated at operating speeds on a test bed, and vibrations can be measured and checked against design tolerances or threshold levels, for instance.”
and the set of clock angles;
R [0018] “Indeed, the wedge angle α and the wedge angle β of a receiving rotor component can be additive or subtractive, depending on the relative circumferential orientations. That is, if the receiving rotor component has a wedge angle β which tends to offset the center of mass of the received rotor component upwardly, this offset can be amplified, or to the contrary, partially, fully, or over-compensated, depending on the amplitude of the wedge angle α of the received component, and the relative circumferential orientations between the two assembled components. The wedge angles α, β and the offset h can, be accurately measured, and the orientation of the wedge angles of the received rotor component can be changed, in the final assembly, by rotating, or “clocking”, the received rotor component 30 around its geometrical axis 42, to a circumferential position determined to achieve this compensating effect, relative to the circumferential orientation of the receiving component and its mating wedge angle β, before assembling it to the receiving component. In any event, it will be understood that even if it is aimed to fully cancel out the offset between the center of mass 40 of the received rotor component 30 and the virtual rotation axis 11 of the overall rotor assembly (e.g. 20 or 22), this cancelling out will, in practice, be imperfect and there will remain an offset, or runout, between the received rotor component 30 and the rotation axis 11 of the rotor assembly.”
R does not disclose processing circuitry operatively connected to memory, utilizing one or more first neural networks associated with the module, wherein each of the one or more first neural networks has been trained with: training data comprising the one or more contributors to unbalance, and one or more rotor dynamics models that use the training data … from the one or more first neural networks to predict vibration at one or more locations of interest in the gas turbine engine.
However, A discloses processing circuitry operatively connected to memory, the processing circuitry configured to:
A [0028] “As shown in FIG. 2, the balancing system 100 includes a computing system 120. The computing system 120 is one example computing system suitable for implementing the computing elements of the balancing system 100. The computing system includes one or more processors 124 and one or more memory devices 126. The one or more processors 124 and one or more memory devices 126 can be embodied in one or more computing devices 122. The one or more processors 124 can include any suitable processing device, such as a microprocessor, microcontroller, integrated circuit, logic device, or other suitable processing device. The one or more memory devices 126 can include one or more computer-readable medium, including, but not limited to, non-transitory computer-readable medium or media, RAM, ROM, hard drives, flash drives, and other memory devices, such as one or more buffer devices.
utilize one or more first neural networks associated with the module,
A [0045] “With reference now to FIGS. 2, 3, 4, and 5, FIG. 5 provides an example block diagram of the one or more machine-learned models 132. As depicted best in FIG. 5, the data 140 is input into the one or more machine-learned models 132. The one or more machine-learned models 132 can be structured as any suitable type of machine-learned model. By way of example, the one or more machine-learned models 132 can be or can include a machine or statistical learning model structured as one of a non-linear least squares optimization model, a linear discriminant analysis model, a partial least squares discriminant analysis model, a support vector machine model, a random tree model, a logistic regression model, a naïve Bayes model, a K-nearest neighbor model, a quadratic discriminant analysis model, an anomaly detection model, a boosted and bagged decision tree model, an artificial neural network model, a C4.5 model, a k-means model, or a combination of one or more of the foregoing.
wherein each of the one or more first neural networks has been trained with: training data comprising the one or more contributors to unbalance,
A [0043-0044] “The assembly parameters 142 can include the residual unbalance of the rotating system 112, rotor concentricity of the rotating system 112, etc. … Stated another way, by applying the one or more machine-learned models 132 to the parameter values specifically associated with the gas turbine engine 110 or rotating system 112, a customized transfer function specifically tailored to the rotating system 112 can be generated. The self-learning transfer function is generated so that an optimal balance shot 150 can be generated to reduce the vibration response of the engine.”
and one or more rotor dynamics models that use the training data … from the one or more first neural networks to predict vibration at one or more locations of interest in the gas turbine engine.
A [0044] “Stated another way, by applying the one or more machine-learned models 132 to the parameter values specifically associated with the gas turbine engine 110 or rotating system 112, a customized transfer function specifically tailored to the rotating system 112 can be generated. The self-learning transfer function is generated so that an optimal balance shot 150 can be generated to reduce the vibration response of the engine.”
A [0033] “The engine operating data of the data 140 can include parameter values for one or more engine operating parameters 141. Example engine operating parameters 141 for which parameter values can be sensed other otherwise calculated based on such sensed parameter values can include, without limitation … parameters associated with vibration of the engine, including a vibration response of the engine, phase angle, mode position of one or more rotors or spools, and/or vibration variations”
R and A are analogous to the claimed invention because they both pertain to balancing rotor components to mitigate vibrations in a gas turbine engine.
It would have been obvious to one with ordinary skill in the art before the effective filing date to combine the teachings of A with R because the system and method of R generates a balance shot that is optimized specifically for a rotating system, which reduces the vibration response of the rotating system, with the use of machine learning models. (See A [Abstract]).
Regarding Claim 13: R in view of A discloses the system of claim 12, wherein for each stage of the one or more modules, the one or more contributors to unbalance include at least one of: a radial offset for at least one of the plurality of stages; a squareness error for at least one of the plurality of stages; or a residual unbalance due to an inherent mass offset for at least one of the plurality of stages.
R [0025] “This type of configuration of centers of masses along the stack or sequence of rotor components 130a, 130b, 130c, 130d, 130e will be referred to herein as a “bow” shape, and is characterized by the fact that in the plane of the main amplitude of center of mass offset (the Y-Z plane illustrated in FIGS. 4A and 4B again in this example and the top view of FIG. 5A in the corresponding example), the amplitude of the offset of the centers of mass, or runout, increases, and then decreases along the axial sequence, reaching its maximum amplitude in an intermediate one of the rotor components of the stack, such as the midshaft in the specific examples illustrated.”
Regarding Claim 14: R in view of A discloses the system of claim 12, wherein: the one or more modules includes a first module and a second module; and the method comprises:
… determine an optimized inter-module clock angle for arranging the second module relative to the first module to mitigate vibration of the gas turbine engine and the sets of clock angles;
R [0023] “In the example shown in FIGS. 3A, 3B and 3C, the rotor assembly 120 consists of a stack including a #1 bearing 130a, an impeller 130b, a midshaft 130c, a two-stage turbine 130d, and a #2 bearing 130e. The circumferential orientation of the impeller 130b relative to the #1 bearing 130a is selected in a manner for the center of mass 140b of the impeller 130b to be offset by 0.002 inches along the Y axis relative to the rotor assembly's rotation axis 111. The circumferential orientation of the successive components is selected in a manner for the center of mass 140c of the midshaft 130c to coincide with the rotation axis 111 of the assembly 120, within measuring tolerances, and for the center of mass 140d of the turbine 130d to have a negative, balancing offset, in this case of −0.002 inches along the Y axis, essentially compensating for the offset of the center of mass 140b of the impeller 130b. The resulting rotor assembly 120 has a total center of mass which is aligned with the axis 111 of the assembly 120, within measuring tolerances. This type of configuration of centers of masses 140b, 140c, 140d, along the stack or sequence of rotor components 130a, 130b, 130c, 130d, 130e will be referred to herein as a zig-zag or corkscrew configuration, and is characterized by the fact that in the plane of the main amplitude of center of mass offset (which is the illustrated Y-Z plane in this example, assuming that the centers of mass are much more closely located relative to the rotation axis in the X-Y plane), the direction of the offset of the centers of mass, or runout, alternates from one side of the rotation axis 111 to the other along the stack, when plotted in a graph such as FIG. 3B, leading to a relatively small runout when considering the rotor assembly 120 in its entirety.” Examiner notes that the inter-module clock angle is represented through the positioning method used to construct the zigzag or corkscrew configurations, and the circumferential orientation was used to align individual module mass offsets.
R [0018] “Indeed, the wedge angle α and the wedge angle β of a receiving rotor component can be additive or subtractive, depending on the relative circumferential orientations. That is, if the receiving rotor component has a wedge angle β which tends to offset the center of mass of the received rotor component upwardly, this offset can be amplified, or to the contrary, partially, fully, or over-compensated, depending on the amplitude of the wedge angle α of the received component, and the relative circumferential orientations between the two assembled components. The wedge angles α, β and the offset h can, be accurately measured, and the orientation of the wedge angles of the received rotor component can be changed, in the final assembly, by rotating, or “clocking”, the received rotor component 30 around its geometrical axis 42, to a circumferential position determined to achieve this compensating effect, relative to the circumferential orientation of the receiving component and its mating wedge angle β, before assembling it to the receiving component. In any event, it will be understood that even if it is aimed to fully cancel out the offset between the center of mass 40 of the received rotor component 30 and the virtual rotation axis 11 of the overall rotor assembly (e.g. 20 or 22), this cancelling out will, in practice, be imperfect and there will remain an offset, or runout, between the received rotor component 30 and the rotation axis 11 of the rotor assembly.”
R [0025] “Indeed, it was found that at least in the case of some aircraft engines, at full operating speeds, this couple unbalance can result in very high vibrations resulting in the engine not passing the vibration acceptance tests.” Examiner notes that if the vibration acceptance test fails, the engine must be disassembled, rebalanced, and reassembled (See R [0002]).
R [0037] “Moreover, the rotor assembly can be rotated at operating speeds on a test bed, and vibrations can be measured and checked against design tolerances or threshold levels, for instance.”
R does not disclose utilizing a second neural network, or wherein the second neural network has also been trained with one of the one or more rotor dynamics models, which uses the training data, … to predict vibration at one or more locations of interest in the gas turbine engine.
However, A discloses utilizing a second neural network, or wherein the second neural network has also been trained with one of the one or more rotor dynamics models, which uses the training data, … to predict vibration at one or more locations of interest in the gas turbine engine.
A [0045] “With reference now to FIGS. 2, 3, 4, and 5, FIG. 5 provides an example block diagram of the one or more machine-learned models 132. As depicted best in FIG. 5, the data 140 is input into the one or more machine-learned models 132. The one or more machine-learned models 132 can be structured as any suitable type of machine-learned model. By way of example, the one or more machine-learned models 132 can be or can include a machine or statistical learning model structured as one of a non-linear least squares optimization model, a linear discriminant analysis model, a partial least squares discriminant analysis model, a support vector machine model, a random tree model, a logistic regression model, a naïve Bayes model, a K-nearest neighbor model, a quadratic discriminant analysis model, an anomaly detection model, a boosted and bagged decision tree model, an artificial neural network model, a C4.5 model, a k-means model, or a combination of one or more of the foregoing.
A [0033] “The engine operating data of the data 140 can include parameter values for one or more engine operating parameters 141. Example engine operating parameters 141 for which parameter values can be sensed other otherwise calculated based on such sensed parameter values can include, without limitation … parameters associated with vibration of the engine, including a vibration response of the engine, phase angle, mode position of one or more rotors or spools, and/or vibration variations”
A [0038] “As noted above, the computing system 120 can generate a balance shot 150 that provides an optimized balancing solution to minimize and/or reduce the vibration response of the rotating system 112. The balance shot 150 can indicate one or more physical locations at which one or more balancing weights are to be added or removed from the rotating system 112. As will be explained further below, the balance shot 150 is generated using an optimized engine specific non-linear self-learning transfer function that is generated by applying one or more machine-learned models on the received data 140. The self-learning transfer function accounts for a variety of parameters, such as engine operating parameters 141, including a rotor vibration response and mode, oil properties, assembly parameters 142, usage parameters 143, environmental parameters 144, life cycle parameters 145, and/or other parameters associated with the gas turbine engine 110 and/or the rotating system 112.”
R and A are analogous to the claimed invention because they both pertain to balancing rotor components to mitigate vibrations in a gas turbine engine.
It would have been obvious to one with ordinary skill in the art before the effective filing date to combine the teachings of A with R because the system and method of R generates a balance shot that is optimized specifically for a rotating system, which reduces the vibration response of the rotating system, with the use of machine learning models. (See A [Abstract]).
Regarding Claim 15: R in view of A disclose the system of claim 14, wherein the forcuessing circuitry is configured to, for each of the one or more modules: the set of clock angles, and the inter-module clock angles to predict vibration at one or more locations of interest in the gas turbine engine.
R [0028] “Accordingly, and more generally, for assembling a plurality of rotor components (typically at least three) into a rotor assembly, one can begin by obtaining geometrical data about the individual rotor components. This can be achieved by measuring the wedge angles at the mating members and determining intrinsic runout (center of mass offset) for each component of the stack. Then, a computer can simulate the different possible combinations of circumferential orientations for the different rotor components in the stack, taking into consideration the geometrical data obtained about the specific set of rotor components to assemble. This simulation can produce a number of different solutions, in which the solutions are each characterized by a given runout configuration. Some of these runout configurations can be corkscrew shaped, for instance, whereas others can be bow shaped. A combination of relative circumferential orientations leading to a bow shape can be selected, and more specifically, if more than one bow shape solution is available, one of these which is considered as producing the lowest overall runout can be selected, and the rotor components can then be assembled to one another according to this selected solution, by “clocking” the individual rotor components to corresponding circumferential orientations, in a manner for the rotor assembly formed by the resulting axially-sequenced stack to correspond to the selected solution.
[0037] Subsequently to the assembly of the rotor assembly with the clocked rotor components, the runout at different planes corresponding to individual ones of the rotor components, for instance, can be measured. Moreover, the rotor assembly can be rotated at operating speeds on a test bed, and vibrations can be measured and checked against design tolerances or threshold levels, for instance.
R does not disclose utilize the second neural network to determine at least one trim weight angle, utilize a third neural network to determine at least one trim weight magnitude, wherein said arranging the first module and second module relative to each other includes adding one or more trim weights to the first module or second module that use one of the determined trim weight magnitudes and one of the determined trim weight angles; and wherein the third neural network has also been trained with one of the one or more rotor dynamics models, which uses the training data, to predict vibration at one or more locations of interest in the gas turbine engine or wherein the second and third neural networks have also been trained with the at least one trim weight angle and the at least one trim weight magnitude.
However, A discloses utilize the second neural network to determine at least one trim weight angle,
A [0045] “With reference now to FIGS. 2, 3, 4, and 5, FIG. 5 provides an example block diagram of the one or more machine-learned models 132. As depicted best in FIG. 5, the data 140 is input into the one or more machine-learned models 132. The one or more machine-learned models 132 can be structured as any suitable type of machine-learned model. By way of example, the one or more machine-learned models 132 can be or can include a machine or statistical learning model structured as one of a non-linear least squares optimization model, a linear discriminant analysis model, a partial least squares discriminant analysis model, a support vector machine model, a random tree model, a logistic regression model, a naïve Bayes model, a K-nearest neighbor model, a quadratic discriminant analysis model, an anomaly detection model, a boosted and bagged decision tree model, an artificial neural network model, a C4.5 model, a k-means model, or a combination of one or more of the foregoing.
A [0038] “As noted above, the computing system 120 can generate a balance shot 150 that provides an optimized balancing solution to minimize and/or reduce the vibration response of the rotating system 112. The balance shot 150 can indicate one or more physical locations at which one or more balancing weights are to be added or removed from the rotating system 112. As will be explained further below, the balance shot 150 is generated using an optimized engine specific non-linear self-learning transfer function that is generated by applying one or more machine-learned models on the received data 140. The self-learning transfer function accounts for a variety of parameters, such as engine operating parameters 141, including a rotor vibration response and mode, oil properties, assembly parameters 142, usage parameters 143, environmental parameters 144, life cycle parameters 145, and/or other parameters associated with the gas turbine engine 110 and/or the rotating system 112.” Examiner notes that the location at which one or more balancing weights are to be added or removed is the trim weight angle.
utilize a third neural network to determine at least one trim weight magnitude,
A [0027] “A balance shot, also referred herein as a balance solution, is a balancing plan that indicates one or more physical locations at which one or more balancing weights are to be added or removed from a rotating system so as to minimize and/or reduce the vibration response of the rotating system. The balance shot can indicate the mass and position of each of the balancing weights.”
wherein said arranging the first module and second module relative to each other includes adding one or more trim weights to the first module or second module that use one of the determined trim weight magnitudes and one of the determined trim weight angles;
A [0027] “A balance shot, also referred herein as a balance solution, is a balancing plan that indicates one or more physical locations at which one or more balancing weights are to be added or removed from a rotating system so as to minimize and/or reduce the vibration response of the rotating system. The balance shot can indicate the mass and position of each of the balancing weights.”
A [0038] “As noted above, the computing system 120 can generate a balance shot 150 that provides an optimized balancing solution to minimize and/or reduce the vibration response of the rotating system 112. The balance shot 150 can indicate one or more physical locations at which one or more balancing weights are to be added or removed from the rotating system 112. As will be explained further below, the balance shot 150 is generated using an optimized engine specific non-linear self-learning transfer function that is generated by applying one or more machine-learned models on the received data 140. The self-learning transfer function accounts for a variety of parameters, such as engine operating parameters 141, including a rotor vibration response and mode, oil properties, assembly parameters 142, usage parameters 143, environmental parameters 144, life cycle parameters 145, and/or other parameters associated with the gas turbine engine 110 and/or the rotating system 112.”
A [0034] “As noted above, the data 140 can also include assembly data. The assembly data can include parameter values for assembly parameters 142 associated with assembly of the rotating system 112 of the gas turbine engine 110. Example assembly parameters 142 can include, without limitation, the assembly method used to assemble the rotating system and/or the gas turbine engine 110, the operators who assembled the rotating system and/or the gas turbine engine 110, the assembly line and/or assembly facility along which or at which the rotating system and/or the gas turbine engine 110 was assembled, and/or one or more parts or components included within the rotating system 112 that were provided by an entity other than the entity that manufactured the rotating system.”
and wherein the third neural network has also been trained with one of the one or more rotor dynamics models, which uses the training data, to predict vibration at one or more locations of interest in the gas turbine engine;
wherein the second and third neural networks have also been trained with the at least one trim weight angle and the at least one trim weight magnitude.
A [0038] “As noted above, the computing system 120 can generate a balance shot 150 that provides an optimized balancing solution to minimize and/or reduce the vibration response of the rotating system 112. The balance shot 150 can indicate one or more physical locations at which one or more balancing weights are to be added or removed from the rotating system 112. As will be explained further below, the balance shot 150 is generated using an optimized engine specific non-linear self-learning transfer function that is generated by applying one or more machine-learned models on the received data 140. The self-learning transfer function accounts for a variety of parameters, such as engine operating parameters 141, including a rotor vibration response and mode, oil properties, assembly parameters 142, usage parameters 143, environmental parameters 144, life cycle parameters 145, and/or other parameters associated with the gas turbine engine 110 and/or the rotating system 112.” Examiner notes that one with ordinary skill in the art would understand that in the context of balancing a rotating system, a ‘physical location’ on a rotor where a balancing/trim weight is applied is defined by its angular position relative to a reference point on the rotor. The citation below discloses the phase angle, which is the angular position relative to a reference point on the rotor of the location of unbalance. Therefore, one with ordinary skill in the art could apply the same reference point to the trim weight angle.
A [0033] “The engine operating data of the data 140 can include parameter values for one or more engine operating parameters 141. Example engine operating parameters 141 for which parameter values can be sensed other otherwise calculated based on such sensed parameter values can include, without limitation ... parameters associated with vibration of the engine, including a vibration response of the engine, phase angle, mode position of one or more rotors or spools, and/or vibration variations.
R and A are analogous to the claimed invention because they both pertain to balancing rotor components to mitigate vibrations in a gas turbine engine.
It would have been obvious to one with ordinary skill in the art before the effective filing date to combine the teachings of A with R because the system and method of R generates a balance shot that is optimized specifically for a rotating system, which reduces the vibration response of the rotating system, with the use of machine learning models. (See A [Abstract]).
Regarding Claim 17: R, in view of A discloses the system of claim 14.
R does not disclose wherein the one or more first neural networks include a neural network associated with the first module and a separate second neural network associated with the second module.
However, A discloses wherein the one or more first neural networks include a neural network associated with the first module and a separate second neural network associated with the second module.
A [0019] “ As schematically shown in FIG. 1, the engine cowl 18 encases, in serial flow relationship, a compressor section including a booster or low pressure (LP) compressor 22 followed downstream by a high pressure (HP) compressor 24; a combustion section 26; a turbine section including an HP turbine 28 followed downstream by an LP turbine 30; and a jet exhaust nozzle section 32. The compressor section, combustion section 26, turbine section, and nozzle section 32 together define a core air flowpath.”
A [0033] “The engine operating data of the data 140 can include parameter values for one or more engine operating parameters 141. Example engine operating parameters 141 for which parameter values can be sensed other otherwise calculated based on such sensed parameter values can include, without limitation, rotor or spool speed, including the low pressure spool speed N1 and/or the high pressure or core spool speed N2, the oil temperature and/or oil pressure, parameters associated with vibration of the engine, including a vibration response of the engine, phase angle, mode position of one or more rotors or spools, and/or vibration variations, the exhaust gas temperature of the engine, cold clearances between rotating and stationary components, parameters associated with acceleration and deceleration of the engine, including the number of accelerations and/or decelerations over a magnitude threshold and/or rate of acceleration and/or deceleration, the compressor inlet pressure and temperature, the compressor discharge pressure, and/or the temperature at the inlet or outlet of the combustor. The engine operating data can include parameter values for other engine operating parameters 141 as well.” Examiner notes that each parameter of the engine operating data is associated to a module of the gas turbine engine.
A [0038] “The self-learning transfer function accounts for a variety of parameters, such as engine operating parameters 141, including a rotor vibration response and mode, oil properties, assembly parameters 142, usage parameters 143, environmental parameters 144, life cycle parameters 145, and/or other parameters associated with the gas turbine engine 110 and/or the rotating system 112.”
A [0044] “At (310), the method (300) includes generating a self-learning transfer function. For instance, the one or more processors 124 of the computing system 120 can generate the self-learning transfer function by applying the one or more machine-learned models 132 to the parameter values received as part of the data 140. The self-learning transfer function can be generated such that it is specific to the rotating system 112. Stated another way, by applying the one or more machine-learned models 132 to the parameter values specifically associated with the gas turbine engine 110 or rotating system 112, a customized transfer function specifically tailored to the rotating system 112 can be generated. The self-learning transfer function is generated so that an optimal balance shot 150 can be generated to reduce the vibration response of the engine.” Examiner notes that the customizable nature of a self-learning transfer function makes it such that machine-learned models can be applied separately to individual engine modules or combined for multiple modules possible by tailoring inputs and outputs to specific rotating components or the entire system.
R and A are analogous to the claimed invention because they both pertain to balancing rotor components to mitigate vibrations in a gas turbine engine.
It would have been obvious to one with ordinary skill in the art before the effective filing date to combine the teachings of A with R because the system and method of R generates a balance shot that is optimized specifically for a rotating system, which reduces the vibration response of the rotating system, with the use of machine learning models. (See A [Abstract]).
Regarding Claim 18: R, in view of A discloses the system of claim 14.
R does not disclose wherein the one or more first neural networks include a neural network associated with both of the first module and the second module.
However, A discloses wherein the one or more first neural networks include a neural network associated with both of the first module and the second module.
A [0019] “ As schematically shown in FIG. 1, the engine cowl 18 encases, in serial flow relationship, a compressor section including a booster or low pressure (LP) compressor 22 followed downstream by a high pressure (HP) compressor 24; a combustion section 26; a turbine section including an HP turbine 28 followed downstream by an LP turbine 30; and a jet exhaust nozzle section 32. The compressor section, combustion section 26, turbine section, and nozzle section 32 together define a core air flowpath.”
A [0033] “The engine operating data of the data 140 can include parameter values for one or more engine operating parameters 141. Example engine operating parameters 141 for which parameter values can be sensed other otherwise calculated based on such sensed parameter values can include, without limitation, rotor or spool speed, including the low pressure spool speed N1 and/or the high pressure or core spool speed N2, the oil temperature and/or oil pressure, parameters associated with vibration of the engine, including a vibration response of the engine, phase angle, mode position of one or more rotors or spools, and/or vibration variations, the exhaust gas temperature of the engine, cold clearances between rotating and stationary components, parameters associated with acceleration and deceleration of the engine, including the number of accelerations and/or decelerations over a magnitude threshold and/or rate of acceleration and/or deceleration, the compressor inlet pressure and temperature, the compressor discharge pressure, and/or the temperature at the inlet or outlet of the combustor. The engine operating data can include parameter values for other engine operating parameters 141 as well.” Examiner notes that each parameter of the engine operating data is associated to a module of the gas turbine engine.
A [0038] “The self-learning transfer function accounts for a variety of parameters, such as engine operating parameters 141, including a rotor vibration response and mode, oil properties, assembly parameters 142, usage parameters 143, environmental parameters 144, life cycle parameters 145, and/or other parameters associated with the gas turbine engine 110 and/or the rotating system 112.”
A [0044] “At (310), the method (300) includes generating a self-learning transfer function. For instance, the one or more processors 124 of the computing system 120 can generate the self-learning transfer function by applying the one or more machine-learned models 132 to the parameter values received as part of the data 140. The self-learning transfer function can be generated such that it is specific to the rotating system 112. Stated another way, by applying the one or more machine-learned models 132 to the parameter values specifically associated with the gas turbine engine 110 or rotating system 112, a customized transfer function specifically tailored to the rotating system 112 can be generated. The self-learning transfer function is generated so that an optimal balance shot 150 can be generated to reduce the vibration response of the engine.” Examiner notes that the customizable nature of a self-learning transfer function makes it such that machine-learned models can be applied separately to individual engine modules or combined for multiple modules possible by tailoring inputs and outputs to specific rotating components or the entire system.
R and A are analogous to the claimed invention because they both pertain to balancing rotor components to mitigate vibrations in a gas turbine engine.
It would have been obvious to one with ordinary skill in the art before the effective filing date to combine the teachings of A with R because the system and method of R generates a balance shot that is optimized specifically for a rotating system, which reduces the vibration response of the rotating system, with the use of machine learning models. (See A [Abstract]).
Regarding Claim 19: R, in view of A discloses the system of claim 14.
R does not disclose wherein: the first module is a high pressure compressor of the gas turbine engine; and the second module is a high pressure turbine of the gas turbine engine.
However, A discloses wherein: the first module is a high pressure compressor of the gas turbine engine; and the second module is a high pressure turbine of the gas turbine engine.
A [0019] “a compressor section including … a high pressure (HP) compressor 24 … a turbine section including an HP turbine 28 … An HP shaft 34 drivingly connects the HP turbine 28 to the HP compressor 24 to rotate them in unison concentrically with respect to the longitudinal centerline 12.”
R and A are analogous to the claimed invention because they both pertain to balancing rotor components to mitigate vibrations in a gas turbine engine.
It would have been obvious to one with ordinary skill in the art before the effective filing date to combine the teachings of A with R because the system and method of R generates a balance shot that is optimized specifically for a rotating system, which reduces the vibration response of the rotating system, with the use of machine learning models. (See A [Abstract]).
Regarding Claim 20: R, in view of A discloses the system of claim 14.
R does not disclose wherein: the first module is a low pressure compressor of the gas turbine engine; and the second module is a low pressure turbine of the gas turbine engine.
However, A discloses wherein: the first module is a low pressure compressor of the gas turbine engine; and the second module is a low pressure turbine of the gas turbine engine.
A [0019] “a compressor section including a booster or low pressure (LP) compressor 22… a turbine section including … an LP turbine 30 … An LP shaft 36 drivingly connects the LP turbine 30 to the LP compressor 22 to rotate them in unison concentrically with respect to the longitudinal centerline 12.
R and A are analogous to the claimed invention because they both pertain to balancing rotor components to mitigate vibrations in a gas turbine engine.
It would have been obvious to one with ordinary skill in the art before the effective filing date to combine the teachings of A with R because the system and method of R generates a balance shot that is optimized specifically for a rotating system, which reduces the vibration response of the rotating system, with the use of machine learning models. (See A [Abstract]).
Claims 6 and 16 are rejected under 35 U.S.C 103 as being unpatentable over U.S. Patent Publication 2021/0102465 A1, hereafter R, in view of U.S. Patent Publication 2023/0012799 A1, hereafter A, further in view of Zhang, H., Wang, M., Li, Z., Zhou, J., Zhang, K., Ma, X., & Wang, M. (2022). Semi-Physical Simulation of Fan Rotor Assembly Process Optimization for Unbalance Based on Reinforcement Learning. Aerospace, 9(7), 342, hereafter Z.
Regarding Claim 6: R in view of A discloses the method of claim 5, for each a plurality of training data sets: utilizing one of the one or more rotor dynamics models to perform at least one rotor dynamics model simulation for the training data set to determine one or more metrics related to vibration of the gas turbine engine, the one or more metrics including at least one of a predicted vibration or forces transmitted to a static structure of the gas turbine engine;
R [0027] “ Indeed, using dynamic response computer model analysis, the optimal unbalance alignment can be to align the compressor rotor and the turbine rotors residual unbalance in-phase and to stack the rotor parts by aligning the individual component such as the unbalance is additive, creating a ‘bowed-shape” spool and resulting in a “inphase unbalance” for the rotor spool end-to-end. (FIG. 5)”
R [0028] “Accordingly, and more generally, for assembling a plurality of rotor components (typically at least three) into a rotor assembly, one can begin by obtaining geometrical data about the individual rotor components. This can be achieved by measuring the wedge angles at the mating members and determining intrinsic runout (center of mass offset) for each component of the stack. Then, a computer can simulate the different possible combinations of circumferential orientations for the different rotor components in the stack, taking into consideration the geometrical data obtained about the specific set of rotor components to assemble. This simulation can produce a number of different solutions, in which the solutions are each characterized by a given runout configuration. Some of these runout configurations can be corkscrew shaped, for instance, whereas others can be bow shaped.”
R and A not disclose and utilizing at least one reward function to calculate a reward for the training data set based on the one or more metrics; calculating a performance metric for the one or more first neural networks, second neural network, and third neural network using a performance function based on the rewards calculated for the training data sets; and utilizing an optimization algorithm to update weights of the one or more first neural networks, second neural network, and third neural network to improve the performance of the one or more first neural networks, second neural network, and third neural network as calculated by the performance function.
However, Z discloses for each of a plurality of training data sets:
and utilizing at least one reward function to calculate a reward for the training data set based on the one or more metrics;
Z [Page 8: Sections 4.2.1 - 4.2.2] “The optimization objective considered in this paper is to minimize the unbalance value, so the reward and punishment function is set as:
PNG
media_image1.png
72
347
media_image1.png
Greyscale
calculating a performance metric for the one or more first neural networks, second neural network, and third neural network using a performance function based on the rewards calculated for the training data sets;
Z [Page 8: Section 5.1] “The blade sequencing optimization model established in the previous section can calculate the remaining unbalance after the blade arrangement of each level, while the overall unbalance produced by the aeroengine fan rotor is the result of the accumulation of residual unbalance after the assembly of all blades. These residual unbalances produce large unbalance forces and unbalance moments when the rotor is rotating at high speed, and the engine performance is seriously affected under the unbalance forces and moments, as shown in Figure 5.”
and utilizing an optimization algorithm to update weights of the one or more first neural networks, second neural network, and third neural network to improve the performance of the one or more first neural networks, second neural network, and third neural network as calculated by the performance function.
Z [Page 8: Section 4.2.2] “R is the immediate reward for each action of the model and considers the long-term impact of each action. The matrix Qhxh is the knowledge learned from experience, h represents the number of leaves, the rows of the matrix Q represent the current state, and the columns represent the possible actions to reach the next state; the matrix Q is initialized to 0. Every time the model is trained, the elements of the matrix Q are updated by the following equation:
PNG
media_image2.png
21
260
media_image2.png
Greyscale
In the formula, s is the current state; a is the action; R is the immediate reward; and γ is the learning variable. The value of the elements in the matrix Q is then equal to the sum of the immediate reward R under the current state s and action a and the learning variable γ multiplied by the maximum reward value for all possible actions to reach the next state. The parameter γ takes a value between 0 and 1, 0≤γ≤1. If γ is closer to 0, the model tends to consider only the immediate reward; if γ is closer to 1, the model considers the overall reward with a greater weight.”
R, A, and Z are analogous to the claimed invention because they all pertain to the optimization of a gas turbine engine through reducing unbalance.
It would have been obvious to one with ordinary skill in the art before the effective filing date to combine the teachings of Z with R and A because the method of Z uses a simulation method that is based on reinforcement learning to optimize the balance in the fan rotor assembly process backed by experimental verification of feasibility and effectiveness. (See Z[Abstract])
Regarding Claim 16: R in view of A discloses the system of claim 15, wherein the processing circuitry is configured to, for each a plurality of training data sets: utilize one of the one or more rotor dynamics models to perform at least one rotor dynamics model simulation for the training data set to determine one or more metrics related to vibration of the gas turbine engine, the one or more metrics including at least one of a predicted vibration or forces transmitted to a static structure of the gas turbine engine;
R [0027] “ Indeed, using dynamic response computer model analysis, the optimal unbalance alignment can be to align the compressor rotor and the turbine rotors residual unbalance in-phase and to stack the rotor parts by aligning the individual component such as the unbalance is additive, creating a ‘bowed-shape” spool and resulting in a “inphase unbalance” for the rotor spool end-to-end. (FIG. 5)”
R [0028] “Accordingly, and more generally, for assembling a plurality of rotor components (typically at least three) into a rotor assembly, one can begin by obtaining geometrical data about the individual rotor components. This can be achieved by measuring the wedge angles at the mating members and determining intrinsic runout (center of mass offset) for each component of the stack. Then, a computer can simulate the different possible combinations of circumferential orientations for the different rotor components in the stack, taking into consideration the geometrical data obtained about the specific set of rotor components to assemble. This simulation can produce a number of different solutions, in which the solutions are each characterized by a given runout configuration. Some of these runout configurations can be corkscrew shaped, for instance, whereas others can be bow shaped.”
R and A not disclose and utilizing at least one reward function to calculate a reward for the training data set based on the one or more metrics; calculating a performance metric for the one or more first neural networks, second neural network, and third neural network using a performance function based on the rewards calculated for the training data sets; and utilizing an optimization algorithm to update weights of the one or more first neural networks, second neural network, and third neural network to improve the performance of the one or more first neural networks, second neural network, and third neural network as calculated by the performance function.
However, Z discloses for each of a plurality of training data sets:
and utilizing at least one reward function to calculate a reward for the training data set based on the one or more metrics;
Z [Page 8: Sections 4.2.1 - 4.2.2] “The optimization objective considered in this paper is to minimize the unbalance value, so the reward and punishment function is set as:
PNG
media_image1.png
72
347
media_image1.png
Greyscale
calculating a performance metric for the one or more first neural networks, second neural network, and third neural network using a performance function based on the rewards calculated for the training data sets;
Z [Page 8: Section 5.1] “The blade sequencing optimization model established in the previous section can calculate the remaining unbalance after the blade arrangement of each level, while the overall unbalance produced by the aeroengine fan rotor is the result of the accumulation of residual unbalance after the assembly of all blades. These residual unbalances produce large unbalance forces and unbalance moments when the rotor is rotating at high speed, and the engine performance is seriously affected under the unbalance forces and moments, as shown in Figure 5.”
and utilizing an optimization algorithm to update weights of the one or more first neural networks, second neural network, and third neural network to improve the performance of the one or more first neural networks, second neural network, and third neural network as calculated by the performance function.
Z [Page 8: Section 4.2.2] “R is the immediate reward for each action of the model and considers the long-term impact of each action. The matrix Qhxh is the knowledge learned from experience, h represents the number of leaves, the rows of the matrix Q represent the current state, and the columns represent the possible actions to reach the next state; the matrix Q is initialized to 0. Every time the model is trained, the elements of the matrix Q are updated by the following equation:
PNG
media_image2.png
21
260
media_image2.png
Greyscale
In the formula, s is the current state; a is the action; R is the immediate reward; and γ is the learning variable. The value of the elements in the matrix Q is then equal to the sum of the immediate reward R under the current state s and action a and the learning variable γ multiplied by the maximum reward value for all possible actions to reach the next state. The parameter γ takes a value between 0 and 1, 0≤γ≤1. If γ is closer to 0, the model tends to consider only the immediate reward; if γ is closer to 1, the model considers the overall reward with a greater weight.”
R, A, and Z are analogous to the claimed invention because they all pertain to the optimization of a gas turbine engine through reducing unbalance.
It would have been obvious to one with ordinary skill in the art before the effective filing date to combine the teachings of Z with R and A because the method of Z uses a simulation method that is based on reinforcement learning to optimize the balance in the fan rotor assembly process backed by experimental verification of feasibility and effectiveness. (See Z[Abstract])
Conclusion
All Claims are rejected.
The prior art made record of and not relied upon is considered pertinent to the applicant’s disclosure.
U.S. Patent 7,321,809 B2
U.S. Patent Publication 2018/0354630 A1
U.S. Patent Publication 2018/0354646 A1
Any inquiry concerning this communication or earlier communications from the examiner should be directed to Scott T. Tran whose telephone number is (571) 272-8533. The examiner can normally be reached on M-Thurs, 8:00-4:00.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Renee Chavez, can be reached at (571) 270-1104. The fax phone number for the organization where this application or proceeding is assigned is (571) 273-8300. Informal or draft communication, please label PROPOSED or DRAFT, can be additionally sent to the Examiner’s fax phone number (571) 272-8533.
Information regarding the status of an application may be obtained from the Patent Application Information Retrieval (PAIR) system. Status information for published a applications may be obtained from either Private PAIR or Public PAIR. Status information for unpublished applications is available through Private PAIR only. For more information about the PAIR system, see http://pair-direct.uspto.gov. Should you have questions on access to the Private PAIR system, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free).
STT
/SCOTT THANH BINH TRAN/Examiner, Art Unit 2186
/SAIF A ALHIJA/Primary Examiner, Art Unit 2186
Prosecution Insights